android_kernel_motorola_sm6225/arch/x86/xen/enlighten.c

1629 lines
39 KiB
C
Raw Normal View History

xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
/*
* Core of Xen paravirt_ops implementation.
*
* This file contains the xen_paravirt_ops structure itself, and the
* implementations for:
* - privileged instructions
* - interrupt flags
* - segment operations
* - booting and setup
*
* Jeremy Fitzhardinge <jeremy@xensource.com>, XenSource Inc, 2007
*/
#include <linux/kernel.h>
#include <linux/init.h>
#include <linux/smp.h>
#include <linux/preempt.h>
#include <linux/hardirq.h>
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
#include <linux/percpu.h>
#include <linux/delay.h>
#include <linux/start_kernel.h>
#include <linux/sched.h>
#include <linux/bootmem.h>
#include <linux/module.h>
#include <linux/mm.h>
#include <linux/page-flags.h>
#include <linux/highmem.h>
#include <linux/console.h>
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
#include <xen/interface/xen.h>
#include <xen/interface/physdev.h>
#include <xen/interface/vcpu.h>
#include <xen/interface/sched.h>
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
#include <xen/features.h>
#include <xen/page.h>
#include <xen/hvc-console.h>
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
#include <asm/paravirt.h>
#include <asm/page.h>
#include <asm/xen/hypercall.h>
#include <asm/xen/hypervisor.h>
#include <asm/fixmap.h>
#include <asm/processor.h>
#include <asm/setup.h>
#include <asm/desc.h>
#include <asm/pgtable.h>
#include <asm/tlbflush.h>
#include <asm/reboot.h>
#include <asm/pgalloc.h>
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
#include "xen-ops.h"
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 03:37:04 +02:00
#include "mmu.h"
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
#include "multicalls.h"
EXPORT_SYMBOL_GPL(hypercall_page);
DEFINE_PER_CPU(struct vcpu_info *, xen_vcpu);
DEFINE_PER_CPU(struct vcpu_info, xen_vcpu_info);
/*
* Note about cr3 (pagetable base) values:
*
* xen_cr3 contains the current logical cr3 value; it contains the
* last set cr3. This may not be the current effective cr3, because
* its update may be being lazily deferred. However, a vcpu looking
* at its own cr3 can use this value knowing that it everything will
* be self-consistent.
*
* xen_current_cr3 contains the actual vcpu cr3; it is set once the
* hypercall to set the vcpu cr3 is complete (so it may be a little
* out of date, but it will never be set early). If one vcpu is
* looking at another vcpu's cr3 value, it should use this variable.
*/
DEFINE_PER_CPU(unsigned long, xen_cr3); /* cr3 stored as physaddr */
DEFINE_PER_CPU(unsigned long, xen_current_cr3); /* actual vcpu cr3 */
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
struct start_info *xen_start_info;
EXPORT_SYMBOL_GPL(xen_start_info);
struct shared_info xen_dummy_shared_info;
/*
* Point at some empty memory to start with. We map the real shared_info
* page as soon as fixmap is up and running.
*/
struct shared_info *HYPERVISOR_shared_info = (void *)&xen_dummy_shared_info;
/*
* Flag to determine whether vcpu info placement is available on all
* VCPUs. We assume it is to start with, and then set it to zero on
* the first failure. This is because it can succeed on some VCPUs
* and not others, since it can involve hypervisor memory allocation,
* or because the guest failed to guarantee all the appropriate
* constraints on all VCPUs (ie buffer can't cross a page boundary).
*
* Note that any particular CPU may be using a placed vcpu structure,
* but we can only optimise if the all are.
*
* 0: not available, 1: available
*/
static int have_vcpu_info_placement = 1;
static void xen_vcpu_setup(int cpu)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
{
struct vcpu_register_vcpu_info info;
int err;
struct vcpu_info *vcpup;
BUG_ON(HYPERVISOR_shared_info == &xen_dummy_shared_info);
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
per_cpu(xen_vcpu, cpu) = &HYPERVISOR_shared_info->vcpu_info[cpu];
if (!have_vcpu_info_placement)
return; /* already tested, not available */
vcpup = &per_cpu(xen_vcpu_info, cpu);
info.mfn = virt_to_mfn(vcpup);
info.offset = offset_in_page(vcpup);
printk(KERN_DEBUG "trying to map vcpu_info %d at %p, mfn %llx, offset %d\n",
cpu, vcpup, info.mfn, info.offset);
/* Check to see if the hypervisor will put the vcpu_info
structure where we want it, which allows direct access via
a percpu-variable. */
err = HYPERVISOR_vcpu_op(VCPUOP_register_vcpu_info, cpu, &info);
if (err) {
printk(KERN_DEBUG "register_vcpu_info failed: err=%d\n", err);
have_vcpu_info_placement = 0;
} else {
/* This cpu is using the registered vcpu info, even if
later ones fail to. */
per_cpu(xen_vcpu, cpu) = vcpup;
printk(KERN_DEBUG "cpu %d using vcpu_info at %p\n",
cpu, vcpup);
}
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
}
/*
* On restore, set the vcpu placement up again.
* If it fails, then we're in a bad state, since
* we can't back out from using it...
*/
void xen_vcpu_restore(void)
{
if (have_vcpu_info_placement) {
int cpu;
for_each_online_cpu(cpu) {
bool other_cpu = (cpu != smp_processor_id());
if (other_cpu &&
HYPERVISOR_vcpu_op(VCPUOP_down, cpu, NULL))
BUG();
xen_vcpu_setup(cpu);
if (other_cpu &&
HYPERVISOR_vcpu_op(VCPUOP_up, cpu, NULL))
BUG();
}
BUG_ON(!have_vcpu_info_placement);
}
}
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
static void __init xen_banner(void)
{
printk(KERN_INFO "Booting paravirtualized kernel on %s\n",
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
pv_info.name);
printk(KERN_INFO "Hypervisor signature: %s%s\n",
xen_start_info->magic,
xen_feature(XENFEAT_mmu_pt_update_preserve_ad) ? " (preserve-AD)" : "");
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
}
static void xen_cpuid(unsigned int *ax, unsigned int *bx,
unsigned int *cx, unsigned int *dx)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
{
unsigned maskedx = ~0;
/*
* Mask out inconvenient features, to try and disable as many
* unsupported kernel subsystems as possible.
*/
if (*ax == 1)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
maskedx = ~((1 << X86_FEATURE_APIC) | /* disable APIC */
(1 << X86_FEATURE_ACPI) | /* disable ACPI */
(1 << X86_FEATURE_MCE) | /* disable MCE */
(1 << X86_FEATURE_MCA) | /* disable MCA */
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
(1 << X86_FEATURE_ACC)); /* thermal monitoring */
asm(XEN_EMULATE_PREFIX "cpuid"
: "=a" (*ax),
"=b" (*bx),
"=c" (*cx),
"=d" (*dx)
: "0" (*ax), "2" (*cx));
*dx &= maskedx;
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
}
static void xen_set_debugreg(int reg, unsigned long val)
{
HYPERVISOR_set_debugreg(reg, val);
}
static unsigned long xen_get_debugreg(int reg)
{
return HYPERVISOR_get_debugreg(reg);
}
static unsigned long xen_save_fl(void)
{
struct vcpu_info *vcpu;
unsigned long flags;
vcpu = x86_read_percpu(xen_vcpu);
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
/* flag has opposite sense of mask */
flags = !vcpu->evtchn_upcall_mask;
/* convert to IF type flag
-0 -> 0x00000000
-1 -> 0xffffffff
*/
return (-flags) & X86_EFLAGS_IF;
}
static void xen_restore_fl(unsigned long flags)
{
struct vcpu_info *vcpu;
/* convert from IF type flag */
flags = !(flags & X86_EFLAGS_IF);
/* There's a one instruction preempt window here. We need to
make sure we're don't switch CPUs between getting the vcpu
pointer and updating the mask. */
preempt_disable();
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
vcpu = x86_read_percpu(xen_vcpu);
vcpu->evtchn_upcall_mask = flags;
preempt_enable_no_resched();
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
/* Doesn't matter if we get preempted here, because any
pending event will get dealt with anyway. */
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
if (flags == 0) {
preempt_check_resched();
barrier(); /* unmask then check (avoid races) */
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
if (unlikely(vcpu->evtchn_upcall_pending))
force_evtchn_callback();
}
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
}
static void xen_irq_disable(void)
{
/* There's a one instruction preempt window here. We need to
make sure we're don't switch CPUs between getting the vcpu
pointer and updating the mask. */
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
preempt_disable();
x86_read_percpu(xen_vcpu)->evtchn_upcall_mask = 1;
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
preempt_enable_no_resched();
}
static void xen_irq_enable(void)
{
struct vcpu_info *vcpu;
/* We don't need to worry about being preempted here, since
either a) interrupts are disabled, so no preemption, or b)
the caller is confused and is trying to re-enable interrupts
on an indeterminate processor. */
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
vcpu = x86_read_percpu(xen_vcpu);
vcpu->evtchn_upcall_mask = 0;
/* Doesn't matter if we get preempted here, because any
pending event will get dealt with anyway. */
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
barrier(); /* unmask then check (avoid races) */
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
if (unlikely(vcpu->evtchn_upcall_pending))
force_evtchn_callback();
}
static void xen_safe_halt(void)
{
/* Blocking includes an implicit local_irq_enable(). */
if (HYPERVISOR_sched_op(SCHEDOP_block, NULL) != 0)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
BUG();
}
static void xen_halt(void)
{
if (irqs_disabled())
HYPERVISOR_vcpu_op(VCPUOP_down, smp_processor_id(), NULL);
else
xen_safe_halt();
}
static void xen_leave_lazy(void)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
{
paravirt_leave_lazy(paravirt_get_lazy_mode());
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
xen_mc_flush();
}
static unsigned long xen_store_tr(void)
{
return 0;
}
static void xen_set_ldt(const void *addr, unsigned entries)
{
struct mmuext_op *op;
struct multicall_space mcs = xen_mc_entry(sizeof(*op));
op = mcs.args;
op->cmd = MMUEXT_SET_LDT;
op->arg1.linear_addr = (unsigned long)addr;
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
op->arg2.nr_ents = entries;
MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF);
xen_mc_issue(PARAVIRT_LAZY_CPU);
}
static void xen_load_gdt(const struct desc_ptr *dtr)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
{
unsigned long *frames;
unsigned long va = dtr->address;
unsigned int size = dtr->size + 1;
unsigned pages = (size + PAGE_SIZE - 1) / PAGE_SIZE;
int f;
struct multicall_space mcs;
/* A GDT can be up to 64k in size, which corresponds to 8192
8-byte entries, or 16 4k pages.. */
BUG_ON(size > 65536);
BUG_ON(va & ~PAGE_MASK);
mcs = xen_mc_entry(sizeof(*frames) * pages);
frames = mcs.args;
for (f = 0; va < dtr->address + size; va += PAGE_SIZE, f++) {
frames[f] = virt_to_mfn(va);
make_lowmem_page_readonly((void *)va);
}
MULTI_set_gdt(mcs.mc, frames, size / sizeof(struct desc_struct));
xen_mc_issue(PARAVIRT_LAZY_CPU);
}
static void load_TLS_descriptor(struct thread_struct *t,
unsigned int cpu, unsigned int i)
{
struct desc_struct *gdt = get_cpu_gdt_table(cpu);
xmaddr_t maddr = virt_to_machine(&gdt[GDT_ENTRY_TLS_MIN+i]);
struct multicall_space mc = __xen_mc_entry(0);
MULTI_update_descriptor(mc.mc, maddr.maddr, t->tls_array[i]);
}
static void xen_load_tls(struct thread_struct *t, unsigned int cpu)
{
/*
* XXX sleazy hack: If we're being called in a lazy-cpu zone,
* it means we're in a context switch, and %gs has just been
* saved. This means we can zero it out to prevent faults on
* exit from the hypervisor if the next process has no %gs.
* Either way, it has been saved, and the new value will get
* loaded properly. This will go away as soon as Xen has been
* modified to not save/restore %gs for normal hypercalls.
*
* On x86_64, this hack is not used for %gs, because gs points
* to KERNEL_GS_BASE (and uses it for PDA references), so we
* must not zero %gs on x86_64
*
* For x86_64, we need to zero %fs, otherwise we may get an
* exception between the new %fs descriptor being loaded and
* %fs being effectively cleared at __switch_to().
*/
if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_CPU) {
#ifdef CONFIG_X86_32
loadsegment(gs, 0);
#else
loadsegment(fs, 0);
#endif
}
xen_mc_batch();
load_TLS_descriptor(t, cpu, 0);
load_TLS_descriptor(t, cpu, 1);
load_TLS_descriptor(t, cpu, 2);
xen_mc_issue(PARAVIRT_LAZY_CPU);
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
}
#ifdef CONFIG_X86_64
static void xen_load_gs_index(unsigned int idx)
{
if (HYPERVISOR_set_segment_base(SEGBASE_GS_USER_SEL, idx))
BUG();
}
#endif
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
static void xen_write_ldt_entry(struct desc_struct *dt, int entrynum,
const void *ptr)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
{
unsigned long lp = (unsigned long)&dt[entrynum];
xmaddr_t mach_lp = virt_to_machine(lp);
u64 entry = *(u64 *)ptr;
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
preempt_disable();
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
xen_mc_flush();
if (HYPERVISOR_update_descriptor(mach_lp.maddr, entry))
BUG();
preempt_enable();
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
}
static int cvt_gate_to_trap(int vector, const gate_desc *val,
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
struct trap_info *info)
{
if (val->type != 0xf && val->type != 0xe)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
return 0;
info->vector = vector;
info->address = gate_offset(*val);
info->cs = gate_segment(*val);
info->flags = val->dpl;
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
/* interrupt gates clear IF */
if (val->type == 0xe)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
info->flags |= 4;
return 1;
}
/* Locations of each CPU's IDT */
static DEFINE_PER_CPU(struct desc_ptr, idt_desc);
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
/* Set an IDT entry. If the entry is part of the current IDT, then
also update Xen. */
static void xen_write_idt_entry(gate_desc *dt, int entrynum, const gate_desc *g)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
{
unsigned long p = (unsigned long)&dt[entrynum];
unsigned long start, end;
preempt_disable();
start = __get_cpu_var(idt_desc).address;
end = start + __get_cpu_var(idt_desc).size + 1;
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
xen_mc_flush();
native_write_idt_entry(dt, entrynum, g);
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
if (p >= start && (p + 8) <= end) {
struct trap_info info[2];
info[1].address = 0;
if (cvt_gate_to_trap(entrynum, g, &info[0]))
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
if (HYPERVISOR_set_trap_table(info))
BUG();
}
preempt_enable();
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
}
static void xen_convert_trap_info(const struct desc_ptr *desc,
struct trap_info *traps)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
{
unsigned in, out, count;
count = (desc->size+1) / sizeof(gate_desc);
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
BUG_ON(count > 256);
for (in = out = 0; in < count; in++) {
gate_desc *entry = (gate_desc*)(desc->address) + in;
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
if (cvt_gate_to_trap(in, entry, &traps[out]))
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
out++;
}
traps[out].address = 0;
}
void xen_copy_trap_info(struct trap_info *traps)
{
const struct desc_ptr *desc = &__get_cpu_var(idt_desc);
xen_convert_trap_info(desc, traps);
}
/* Load a new IDT into Xen. In principle this can be per-CPU, so we
hold a spinlock to protect the static traps[] array (static because
it avoids allocation, and saves stack space). */
static void xen_load_idt(const struct desc_ptr *desc)
{
static DEFINE_SPINLOCK(lock);
static struct trap_info traps[257];
spin_lock(&lock);
__get_cpu_var(idt_desc) = *desc;
xen_convert_trap_info(desc, traps);
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
xen_mc_flush();
if (HYPERVISOR_set_trap_table(traps))
BUG();
spin_unlock(&lock);
}
/* Write a GDT descriptor entry. Ignore LDT descriptors, since
they're handled differently. */
static void xen_write_gdt_entry(struct desc_struct *dt, int entry,
const void *desc, int type)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
{
preempt_disable();
switch (type) {
case DESC_LDT:
case DESC_TSS:
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
/* ignore */
break;
default: {
xmaddr_t maddr = virt_to_machine(&dt[entry]);
xen_mc_flush();
if (HYPERVISOR_update_descriptor(maddr.maddr, *(u64 *)desc))
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
BUG();
}
}
preempt_enable();
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
}
static void xen_load_sp0(struct tss_struct *tss,
struct thread_struct *thread)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
{
struct multicall_space mcs = xen_mc_entry(0);
MULTI_stack_switch(mcs.mc, __KERNEL_DS, thread->sp0);
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
xen_mc_issue(PARAVIRT_LAZY_CPU);
}
static void xen_set_iopl_mask(unsigned mask)
{
struct physdev_set_iopl set_iopl;
/* Force the change at ring 0. */
set_iopl.iopl = (mask == 0) ? 1 : (mask >> 12) & 3;
HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl);
}
static void xen_io_delay(void)
{
}
#ifdef CONFIG_X86_LOCAL_APIC
static u32 xen_apic_read(unsigned long reg)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
{
return 0;
}
static void xen_apic_write(unsigned long reg, u32 val)
{
/* Warn to see if there's any stray references */
WARN_ON(1);
}
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
#endif
static void xen_flush_tlb(void)
{
struct mmuext_op *op;
struct multicall_space mcs;
preempt_disable();
mcs = xen_mc_entry(sizeof(*op));
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
op = mcs.args;
op->cmd = MMUEXT_TLB_FLUSH_LOCAL;
MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF);
xen_mc_issue(PARAVIRT_LAZY_MMU);
preempt_enable();
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
}
static void xen_flush_tlb_single(unsigned long addr)
{
struct mmuext_op *op;
struct multicall_space mcs;
preempt_disable();
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
mcs = xen_mc_entry(sizeof(*op));
op = mcs.args;
op->cmd = MMUEXT_INVLPG_LOCAL;
op->arg1.linear_addr = addr & PAGE_MASK;
MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF);
xen_mc_issue(PARAVIRT_LAZY_MMU);
preempt_enable();
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
}
static void xen_flush_tlb_others(const cpumask_t *cpus, struct mm_struct *mm,
unsigned long va)
{
struct {
struct mmuext_op op;
cpumask_t mask;
} *args;
cpumask_t cpumask = *cpus;
struct multicall_space mcs;
/*
* A couple of (to be removed) sanity checks:
*
* - current CPU must not be in mask
* - mask must exist :)
*/
BUG_ON(cpus_empty(cpumask));
BUG_ON(cpu_isset(smp_processor_id(), cpumask));
BUG_ON(!mm);
/* If a CPU which we ran on has gone down, OK. */
cpus_and(cpumask, cpumask, cpu_online_map);
if (cpus_empty(cpumask))
return;
mcs = xen_mc_entry(sizeof(*args));
args = mcs.args;
args->mask = cpumask;
args->op.arg2.vcpumask = &args->mask;
if (va == TLB_FLUSH_ALL) {
args->op.cmd = MMUEXT_TLB_FLUSH_MULTI;
} else {
args->op.cmd = MMUEXT_INVLPG_MULTI;
args->op.arg1.linear_addr = va;
}
MULTI_mmuext_op(mcs.mc, &args->op, 1, NULL, DOMID_SELF);
xen_mc_issue(PARAVIRT_LAZY_MMU);
}
static void xen_clts(void)
{
struct multicall_space mcs;
mcs = xen_mc_entry(0);
MULTI_fpu_taskswitch(mcs.mc, 0);
xen_mc_issue(PARAVIRT_LAZY_CPU);
}
static void xen_write_cr0(unsigned long cr0)
{
struct multicall_space mcs;
/* Only pay attention to cr0.TS; everything else is
ignored. */
mcs = xen_mc_entry(0);
MULTI_fpu_taskswitch(mcs.mc, (cr0 & X86_CR0_TS) != 0);
xen_mc_issue(PARAVIRT_LAZY_CPU);
}
static void xen_write_cr2(unsigned long cr2)
{
x86_read_percpu(xen_vcpu)->arch.cr2 = cr2;
}
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
static unsigned long xen_read_cr2(void)
{
return x86_read_percpu(xen_vcpu)->arch.cr2;
}
static unsigned long xen_read_cr2_direct(void)
{
return x86_read_percpu(xen_vcpu_info.arch.cr2);
}
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
static void xen_write_cr4(unsigned long cr4)
{
cr4 &= ~X86_CR4_PGE;
cr4 &= ~X86_CR4_PSE;
native_write_cr4(cr4);
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
}
static unsigned long xen_read_cr3(void)
{
return x86_read_percpu(xen_cr3);
}
static void set_current_cr3(void *v)
{
x86_write_percpu(xen_current_cr3, (unsigned long)v);
}
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
static void xen_write_cr3(unsigned long cr3)
{
struct mmuext_op *op;
struct multicall_space mcs;
unsigned long mfn = pfn_to_mfn(PFN_DOWN(cr3));
BUG_ON(preemptible());
mcs = xen_mc_entry(sizeof(*op)); /* disables interrupts */
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
/* Update while interrupts are disabled, so its atomic with
respect to ipis */
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
x86_write_percpu(xen_cr3, cr3);
op = mcs.args;
op->cmd = MMUEXT_NEW_BASEPTR;
op->arg1.mfn = mfn;
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF);
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
/* Update xen_update_cr3 once the batch has actually
been submitted. */
xen_mc_callback(set_current_cr3, (void *)cr3);
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
xen_mc_issue(PARAVIRT_LAZY_CPU); /* interrupts restored */
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
}
/* Early in boot, while setting up the initial pagetable, assume
everything is pinned. */
static __init void xen_alloc_pte_init(struct mm_struct *mm, u32 pfn)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
{
#ifdef CONFIG_FLATMEM
BUG_ON(mem_map); /* should only be used early */
#endif
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
make_lowmem_page_readonly(__va(PFN_PHYS(pfn)));
}
/* Early release_pte assumes that all pts are pinned, since there's
only init_mm and anything attached to that is pinned. */
static void xen_release_pte_init(u32 pfn)
{
make_lowmem_page_readwrite(__va(PFN_PHYS(pfn)));
}
static void pin_pagetable_pfn(unsigned cmd, unsigned long pfn)
{
struct mmuext_op op;
op.cmd = cmd;
op.arg1.mfn = pfn_to_mfn(pfn);
if (HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF))
BUG();
}
/* This needs to make sure the new pte page is pinned iff its being
attached to a pinned pagetable. */
static void xen_alloc_ptpage(struct mm_struct *mm, u32 pfn, unsigned level)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
{
struct page *page = pfn_to_page(pfn);
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
if (PagePinned(virt_to_page(mm->pgd))) {
SetPagePinned(page);
if (!PageHighMem(page)) {
make_lowmem_page_readonly(__va(PFN_PHYS(pfn)));
if (level == PT_PTE)
pin_pagetable_pfn(MMUEXT_PIN_L1_TABLE, pfn);
} else
/* make sure there are no stray mappings of
this page */
kmap_flush_unused();
}
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
}
static void xen_alloc_pte(struct mm_struct *mm, u32 pfn)
{
xen_alloc_ptpage(mm, pfn, PT_PTE);
}
static void xen_alloc_pmd(struct mm_struct *mm, u32 pfn)
{
xen_alloc_ptpage(mm, pfn, PT_PMD);
}
/* This should never happen until we're OK to use struct page */
static void xen_release_ptpage(u32 pfn, unsigned level)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
{
struct page *page = pfn_to_page(pfn);
if (PagePinned(page)) {
if (!PageHighMem(page)) {
if (level == PT_PTE)
pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, pfn);
make_lowmem_page_readwrite(__va(PFN_PHYS(pfn)));
}
ClearPagePinned(page);
}
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
}
static void xen_release_pte(u32 pfn)
{
xen_release_ptpage(pfn, PT_PTE);
}
static void xen_release_pmd(u32 pfn)
{
xen_release_ptpage(pfn, PT_PMD);
}
#if PAGETABLE_LEVELS == 4
static void xen_alloc_pud(struct mm_struct *mm, u32 pfn)
{
xen_alloc_ptpage(mm, pfn, PT_PUD);
}
static void xen_release_pud(u32 pfn)
{
xen_release_ptpage(pfn, PT_PUD);
}
#endif
#ifdef CONFIG_HIGHPTE
static void *xen_kmap_atomic_pte(struct page *page, enum km_type type)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
{
pgprot_t prot = PAGE_KERNEL;
if (PagePinned(page))
prot = PAGE_KERNEL_RO;
if (0 && PageHighMem(page))
printk("mapping highpte %lx type %d prot %s\n",
page_to_pfn(page), type,
(unsigned long)pgprot_val(prot) & _PAGE_RW ? "WRITE" : "READ");
return kmap_atomic_prot(page, type, prot);
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
}
#endif
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
static __init pte_t mask_rw_pte(pte_t *ptep, pte_t pte)
{
/* If there's an existing pte, then don't allow _PAGE_RW to be set */
if (pte_val_ma(*ptep) & _PAGE_PRESENT)
pte = __pte_ma(((pte_val_ma(*ptep) & _PAGE_RW) | ~_PAGE_RW) &
pte_val_ma(pte));
return pte;
}
/* Init-time set_pte while constructing initial pagetables, which
doesn't allow RO pagetable pages to be remapped RW */
static __init void xen_set_pte_init(pte_t *ptep, pte_t pte)
{
pte = mask_rw_pte(ptep, pte);
xen_set_pte(ptep, pte);
}
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
static __init void xen_pagetable_setup_start(pgd_t *base)
{
}
void xen_setup_shared_info(void)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
{
if (!xen_feature(XENFEAT_auto_translated_physmap)) {
set_fixmap(FIX_PARAVIRT_BOOTMAP,
xen_start_info->shared_info);
HYPERVISOR_shared_info =
(struct shared_info *)fix_to_virt(FIX_PARAVIRT_BOOTMAP);
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
} else
HYPERVISOR_shared_info =
(struct shared_info *)__va(xen_start_info->shared_info);
#ifndef CONFIG_SMP
/* In UP this is as good a place as any to set up shared info */
xen_setup_vcpu_info_placement();
#endif
xen_setup_mfn_list_list();
}
static __init void xen_pagetable_setup_done(pgd_t *base)
{
xen_setup_shared_info();
}
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
static __init void xen_post_allocator_init(void)
{
pv_mmu_ops.set_pte = xen_set_pte;
pv_mmu_ops.set_pmd = xen_set_pmd;
pv_mmu_ops.set_pud = xen_set_pud;
#if PAGETABLE_LEVELS == 4
pv_mmu_ops.set_pgd = xen_set_pgd;
#endif
/* This will work as long as patching hasn't happened yet
(which it hasn't) */
pv_mmu_ops.alloc_pte = xen_alloc_pte;
pv_mmu_ops.alloc_pmd = xen_alloc_pmd;
pv_mmu_ops.release_pte = xen_release_pte;
pv_mmu_ops.release_pmd = xen_release_pmd;
#if PAGETABLE_LEVELS == 4
pv_mmu_ops.alloc_pud = xen_alloc_pud;
pv_mmu_ops.release_pud = xen_release_pud;
#endif
xen_mark_init_mm_pinned();
}
/* This is called once we have the cpu_possible_map */
void xen_setup_vcpu_info_placement(void)
{
int cpu;
for_each_possible_cpu(cpu)
xen_vcpu_setup(cpu);
/* xen_vcpu_setup managed to place the vcpu_info within the
percpu area for all cpus, so make use of it */
#ifdef CONFIG_X86_32
if (have_vcpu_info_placement) {
printk(KERN_INFO "Xen: using vcpu_info placement\n");
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
pv_irq_ops.save_fl = xen_save_fl_direct;
pv_irq_ops.restore_fl = xen_restore_fl_direct;
pv_irq_ops.irq_disable = xen_irq_disable_direct;
pv_irq_ops.irq_enable = xen_irq_enable_direct;
pv_mmu_ops.read_cr2 = xen_read_cr2_direct;
}
#endif
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
}
static unsigned xen_patch(u8 type, u16 clobbers, void *insnbuf,
unsigned long addr, unsigned len)
{
char *start, *end, *reloc;
unsigned ret;
start = end = reloc = NULL;
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
#define SITE(op, x) \
case PARAVIRT_PATCH(op.x): \
if (have_vcpu_info_placement) { \
start = (char *)xen_##x##_direct; \
end = xen_##x##_direct_end; \
reloc = xen_##x##_direct_reloc; \
} \
goto patch_site
switch (type) {
#ifdef CONFIG_X86_32
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
SITE(pv_irq_ops, irq_enable);
SITE(pv_irq_ops, irq_disable);
SITE(pv_irq_ops, save_fl);
SITE(pv_irq_ops, restore_fl);
#endif /* CONFIG_X86_32 */
#undef SITE
patch_site:
if (start == NULL || (end-start) > len)
goto default_patch;
ret = paravirt_patch_insns(insnbuf, len, start, end);
/* Note: because reloc is assigned from something that
appears to be an array, gcc assumes it's non-null,
but doesn't know its relationship with start and
end. */
if (reloc > start && reloc < end) {
int reloc_off = reloc - start;
long *relocp = (long *)(insnbuf + reloc_off);
long delta = start - (char *)addr;
*relocp += delta;
}
break;
default_patch:
default:
ret = paravirt_patch_default(type, clobbers, insnbuf,
addr, len);
break;
}
return ret;
}
static void xen_set_fixmap(unsigned idx, unsigned long phys, pgprot_t prot)
{
pte_t pte;
phys >>= PAGE_SHIFT;
switch (idx) {
case FIX_BTMAP_END ... FIX_BTMAP_BEGIN:
#ifdef CONFIG_X86_F00F_BUG
case FIX_F00F_IDT:
#endif
#ifdef CONFIG_X86_32
case FIX_WP_TEST:
case FIX_VDSO:
case FIX_KMAP_BEGIN ... FIX_KMAP_END:
#else
case VSYSCALL_LAST_PAGE ... VSYSCALL_FIRST_PAGE:
#endif
#ifdef CONFIG_X86_LOCAL_APIC
case FIX_APIC_BASE: /* maps dummy local APIC */
#endif
pte = pfn_pte(phys, prot);
break;
default:
pte = mfn_pte(phys, prot);
break;
}
__native_set_fixmap(idx, pte);
}
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
static const struct pv_info xen_info __initdata = {
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
.paravirt_enabled = 1,
.shared_kernel_pmd = 0,
.name = "Xen",
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
};
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
static const struct pv_init_ops xen_init_ops __initdata = {
.patch = xen_patch,
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
.banner = xen_banner,
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
.memory_setup = xen_memory_setup,
.arch_setup = xen_arch_setup,
.post_allocator_init = xen_post_allocator_init,
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
};
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
static const struct pv_time_ops xen_time_ops __initdata = {
.time_init = xen_time_init,
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
.set_wallclock = xen_set_wallclock,
.get_wallclock = xen_get_wallclock,
.get_tsc_khz = xen_tsc_khz,
.sched_clock = xen_sched_clock,
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
};
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
static const struct pv_cpu_ops xen_cpu_ops __initdata = {
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
.cpuid = xen_cpuid,
.set_debugreg = xen_set_debugreg,
.get_debugreg = xen_get_debugreg,
.clts = xen_clts,
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
.read_cr0 = native_read_cr0,
.write_cr0 = xen_write_cr0,
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
.read_cr4 = native_read_cr4,
.read_cr4_safe = native_read_cr4_safe,
.write_cr4 = xen_write_cr4,
.wbinvd = native_wbinvd,
.read_msr = native_read_msr_safe,
.write_msr = native_write_msr_safe,
.read_tsc = native_read_tsc,
.read_pmc = native_read_pmc,
.iret = xen_iret,
.irq_enable_sysexit = xen_sysexit,
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
.load_tr_desc = paravirt_nop,
.set_ldt = xen_set_ldt,
.load_gdt = xen_load_gdt,
.load_idt = xen_load_idt,
.load_tls = xen_load_tls,
#ifdef CONFIG_X86_64
.load_gs_index = xen_load_gs_index,
#endif
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
.store_gdt = native_store_gdt,
.store_idt = native_store_idt,
.store_tr = xen_store_tr,
.write_ldt_entry = xen_write_ldt_entry,
.write_gdt_entry = xen_write_gdt_entry,
.write_idt_entry = xen_write_idt_entry,
.load_sp0 = xen_load_sp0,
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
.set_iopl_mask = xen_set_iopl_mask,
.io_delay = xen_io_delay,
/* Xen takes care of %gs when switching to usermode for us */
.swapgs = paravirt_nop,
.lazy_mode = {
.enter = paravirt_enter_lazy_cpu,
.leave = xen_leave_lazy,
},
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
};
static void __init __xen_init_IRQ(void)
{
#ifdef CONFIG_X86_64
int i;
/* Create identity vector->irq map */
for(i = 0; i < NR_VECTORS; i++) {
int cpu;
for_each_possible_cpu(cpu)
per_cpu(vector_irq, cpu)[i] = i;
}
#endif /* CONFIG_X86_64 */
xen_init_IRQ();
}
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
static const struct pv_irq_ops xen_irq_ops __initdata = {
.init_IRQ = __xen_init_IRQ,
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
.save_fl = xen_save_fl,
.restore_fl = xen_restore_fl,
.irq_disable = xen_irq_disable,
.irq_enable = xen_irq_enable,
.safe_halt = xen_safe_halt,
.halt = xen_halt,
#ifdef CONFIG_X86_64
.adjust_exception_frame = xen_adjust_exception_frame,
#endif
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
};
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
static const struct pv_apic_ops xen_apic_ops __initdata = {
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
#ifdef CONFIG_X86_LOCAL_APIC
.apic_write = xen_apic_write,
.apic_write_atomic = xen_apic_write,
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
.apic_read = xen_apic_read,
.setup_boot_clock = paravirt_nop,
.setup_secondary_clock = paravirt_nop,
.startup_ipi_hook = paravirt_nop,
#endif
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
};
static const struct pv_mmu_ops xen_mmu_ops __initdata = {
.pagetable_setup_start = xen_pagetable_setup_start,
.pagetable_setup_done = xen_pagetable_setup_done,
.read_cr2 = xen_read_cr2,
.write_cr2 = xen_write_cr2,
.read_cr3 = xen_read_cr3,
.write_cr3 = xen_write_cr3,
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
.flush_tlb_user = xen_flush_tlb,
.flush_tlb_kernel = xen_flush_tlb,
.flush_tlb_single = xen_flush_tlb_single,
.flush_tlb_others = xen_flush_tlb_others,
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
.pte_update = paravirt_nop,
.pte_update_defer = paravirt_nop,
.pgd_alloc = __paravirt_pgd_alloc,
.pgd_free = paravirt_nop,
.alloc_pte = xen_alloc_pte_init,
.release_pte = xen_release_pte_init,
.alloc_pmd = xen_alloc_pte_init,
.alloc_pmd_clone = paravirt_nop,
.release_pmd = xen_release_pte_init,
#ifdef CONFIG_HIGHPTE
.kmap_atomic_pte = xen_kmap_atomic_pte,
#endif
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
#ifdef CONFIG_X86_64
.set_pte = xen_set_pte,
#else
.set_pte = xen_set_pte_init,
#endif
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 03:37:04 +02:00
.set_pte_at = xen_set_pte_at,
.set_pmd = xen_set_pmd_hyper,
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 03:37:04 +02:00
.ptep_modify_prot_start = __ptep_modify_prot_start,
.ptep_modify_prot_commit = __ptep_modify_prot_commit,
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 03:37:04 +02:00
.pte_val = xen_pte_val,
.pte_flags = native_pte_val,
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 03:37:04 +02:00
.pgd_val = xen_pgd_val,
.make_pte = xen_make_pte,
.make_pgd = xen_make_pgd,
#ifdef CONFIG_X86_PAE
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 03:37:04 +02:00
.set_pte_atomic = xen_set_pte_atomic,
.set_pte_present = xen_set_pte_at,
.pte_clear = xen_pte_clear,
.pmd_clear = xen_pmd_clear,
#endif /* CONFIG_X86_PAE */
.set_pud = xen_set_pud_hyper,
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 03:37:04 +02:00
.make_pmd = xen_make_pmd,
.pmd_val = xen_pmd_val,
#if PAGETABLE_LEVELS == 4
.pud_val = xen_pud_val,
.make_pud = xen_make_pud,
.set_pgd = xen_set_pgd_hyper,
.alloc_pud = xen_alloc_pte_init,
.release_pud = xen_release_pte_init,
#endif /* PAGETABLE_LEVELS == 4 */
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 03:37:04 +02:00
.activate_mm = xen_activate_mm,
.dup_mmap = xen_dup_mmap,
.exit_mmap = xen_exit_mmap,
.lazy_mode = {
.enter = paravirt_enter_lazy_mmu,
.leave = xen_leave_lazy,
},
.set_fixmap = xen_set_fixmap,
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
};
static void xen_reboot(int reason)
{
struct sched_shutdown r = { .reason = reason };
#ifdef CONFIG_SMP
smp_send_stop();
#endif
if (HYPERVISOR_sched_op(SCHEDOP_shutdown, &r))
BUG();
}
static void xen_restart(char *msg)
{
xen_reboot(SHUTDOWN_reboot);
}
static void xen_emergency_restart(void)
{
xen_reboot(SHUTDOWN_reboot);
}
static void xen_machine_halt(void)
{
xen_reboot(SHUTDOWN_poweroff);
}
static void xen_crash_shutdown(struct pt_regs *regs)
{
xen_reboot(SHUTDOWN_crash);
}
static const struct machine_ops __initdata xen_machine_ops = {
.restart = xen_restart,
.halt = xen_machine_halt,
.power_off = xen_machine_halt,
.shutdown = xen_machine_halt,
.crash_shutdown = xen_crash_shutdown,
.emergency_restart = xen_emergency_restart,
};
static void __init xen_reserve_top(void)
{
#ifdef CONFIG_X86_32
unsigned long top = HYPERVISOR_VIRT_START;
struct xen_platform_parameters pp;
if (HYPERVISOR_xen_version(XENVER_platform_parameters, &pp) == 0)
top = pp.virt_start;
reserve_top_address(-top + 2 * PAGE_SIZE);
#endif /* CONFIG_X86_32 */
}
/*
* Like __va(), but returns address in the kernel mapping (which is
* all we have until the physical memory mapping has been set up.
*/
static void *__ka(phys_addr_t paddr)
{
#ifdef CONFIG_X86_64
return (void *)(paddr + __START_KERNEL_map);
#else
return __va(paddr);
#endif
}
/* Convert a machine address to physical address */
static unsigned long m2p(phys_addr_t maddr)
{
phys_addr_t paddr;
maddr &= PTE_MASK;
paddr = mfn_to_pfn(maddr >> PAGE_SHIFT) << PAGE_SHIFT;
return paddr;
}
/* Convert a machine address to kernel virtual */
static void *m2v(phys_addr_t maddr)
{
return __ka(m2p(maddr));
}
#ifdef CONFIG_X86_64
static void walk(pgd_t *pgd, unsigned long addr)
{
unsigned l4idx = pgd_index(addr);
unsigned l3idx = pud_index(addr);
unsigned l2idx = pmd_index(addr);
unsigned l1idx = pte_index(addr);
pgd_t l4;
pud_t l3;
pmd_t l2;
pte_t l1;
xen_raw_printk("walk %p, %lx -> %d %d %d %d\n",
pgd, addr, l4idx, l3idx, l2idx, l1idx);
l4 = pgd[l4idx];
xen_raw_printk(" l4: %016lx\n", l4.pgd);
xen_raw_printk(" %016lx\n", pgd_val(l4));
l3 = ((pud_t *)(m2v(l4.pgd)))[l3idx];
xen_raw_printk(" l3: %016lx\n", l3.pud);
xen_raw_printk(" %016lx\n", pud_val(l3));
l2 = ((pmd_t *)(m2v(l3.pud)))[l2idx];
xen_raw_printk(" l2: %016lx\n", l2.pmd);
xen_raw_printk(" %016lx\n", pmd_val(l2));
l1 = ((pte_t *)(m2v(l2.pmd)))[l1idx];
xen_raw_printk(" l1: %016lx\n", l1.pte);
xen_raw_printk(" %016lx\n", pte_val(l1));
}
#endif
static void set_page_prot(void *addr, pgprot_t prot)
{
unsigned long pfn = __pa(addr) >> PAGE_SHIFT;
pte_t pte = pfn_pte(pfn, prot);
xen_raw_printk("addr=%p pfn=%lx mfn=%lx prot=%016llx pte=%016llx\n",
addr, pfn, get_phys_to_machine(pfn),
pgprot_val(prot), pte.pte);
if (HYPERVISOR_update_va_mapping((unsigned long)addr, pte, 0))
BUG();
}
/*
* Identity map, in addition to plain kernel map. This needs to be
* large enough to allocate page table pages to allocate the rest.
* Each page can map 2MB.
*/
static pte_t level1_ident_pgt[PTRS_PER_PTE * 4] __page_aligned_bss;
static __init void xen_map_identity_early(pmd_t *pmd, unsigned long max_pfn)
{
unsigned pmdidx, pteidx;
unsigned ident_pte;
unsigned long pfn;
ident_pte = 0;
pfn = 0;
for(pmdidx = 0; pmdidx < PTRS_PER_PMD && pfn < max_pfn; pmdidx++) {
pte_t *pte_page;
/* Reuse or allocate a page of ptes */
if (pmd_present(pmd[pmdidx]))
pte_page = m2v(pmd[pmdidx].pmd);
else {
/* Check for free pte pages */
if (ident_pte == ARRAY_SIZE(level1_ident_pgt))
break;
pte_page = &level1_ident_pgt[ident_pte];
ident_pte += PTRS_PER_PTE;
pmd[pmdidx] = __pmd(__pa(pte_page) | _PAGE_TABLE);
}
/* Install mappings */
for(pteidx = 0; pteidx < PTRS_PER_PTE; pteidx++, pfn++) {
pte_t pte;
if (pfn > max_pfn_mapped)
max_pfn_mapped = pfn;
if (!pte_none(pte_page[pteidx]))
continue;
pte = pfn_pte(pfn, PAGE_KERNEL_EXEC);
pte_page[pteidx] = pte;
}
}
for(pteidx = 0; pteidx < ident_pte; pteidx += PTRS_PER_PTE)
set_page_prot(&level1_ident_pgt[pteidx], PAGE_KERNEL_RO);
set_page_prot(pmd, PAGE_KERNEL_RO);
}
#ifdef CONFIG_X86_64
static void convert_pfn_mfn(void *v)
{
pte_t *pte = v;
int i;
/* All levels are converted the same way, so just treat them
as ptes. */
for(i = 0; i < PTRS_PER_PTE; i++)
pte[i] = xen_make_pte(pte[i].pte);
}
/*
* Set up the inital kernel pagetable.
*
* We can construct this by grafting the Xen provided pagetable into
* head_64.S's preconstructed pagetables. We copy the Xen L2's into
* level2_ident_pgt, level2_kernel_pgt and level2_fixmap_pgt. This
* means that only the kernel has a physical mapping to start with -
* but that's enough to get __va working. We need to fill in the rest
* of the physical mapping once some sort of allocator has been set
* up.
*/
static __init pgd_t *xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn)
{
pud_t *l3;
pmd_t *l2;
/* Zap identity mapping */
init_level4_pgt[0] = __pgd(0);
/* Pre-constructed entries are in pfn, so convert to mfn */
convert_pfn_mfn(init_level4_pgt);
convert_pfn_mfn(level3_ident_pgt);
convert_pfn_mfn(level3_kernel_pgt);
l3 = m2v(pgd[pgd_index(__START_KERNEL_map)].pgd);
l2 = m2v(l3[pud_index(__START_KERNEL_map)].pud);
memcpy(level2_ident_pgt, l2, sizeof(pmd_t) * PTRS_PER_PMD);
memcpy(level2_kernel_pgt, l2, sizeof(pmd_t) * PTRS_PER_PMD);
l3 = m2v(pgd[pgd_index(__START_KERNEL_map + PMD_SIZE)].pgd);
l2 = m2v(l3[pud_index(__START_KERNEL_map + PMD_SIZE)].pud);
memcpy(level2_fixmap_pgt, l2, sizeof(pmd_t) * PTRS_PER_PMD);
/* Set up identity map */
xen_map_identity_early(level2_ident_pgt, max_pfn);
/* Make pagetable pieces RO */
set_page_prot(init_level4_pgt, PAGE_KERNEL_RO);
set_page_prot(level3_ident_pgt, PAGE_KERNEL_RO);
set_page_prot(level3_kernel_pgt, PAGE_KERNEL_RO);
set_page_prot(level2_kernel_pgt, PAGE_KERNEL_RO);
set_page_prot(level2_fixmap_pgt, PAGE_KERNEL_RO);
/* Pin down new L4 */
pin_pagetable_pfn(MMUEXT_PIN_L4_TABLE,
PFN_DOWN(__pa_symbol(init_level4_pgt)));
/* Unpin Xen-provided one */
pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, PFN_DOWN(__pa(pgd)));
/* Switch over */
pgd = init_level4_pgt;
xen_write_cr3(__pa(pgd));
reserve_early(__pa(xen_start_info->pt_base),
__pa(xen_start_info->pt_base +
xen_start_info->nr_pt_frames * PAGE_SIZE),
"XEN PAGETABLES");
return pgd;
}
#else /* !CONFIG_X86_64 */
static pmd_t level2_kernel_pgt[PTRS_PER_PMD] __page_aligned_bss;
static __init pgd_t *xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn)
{
pmd_t *kernel_pmd;
init_pg_tables_start = __pa(pgd);
init_pg_tables_end = __pa(pgd) + xen_start_info->nr_pt_frames*PAGE_SIZE;
max_pfn_mapped = PFN_DOWN(init_pg_tables_end + 512*1024);
kernel_pmd = m2v(pgd[KERNEL_PGD_BOUNDARY].pgd);
memcpy(level2_kernel_pgt, kernel_pmd, sizeof(pmd_t) * PTRS_PER_PMD);
xen_map_identity_early(level2_kernel_pgt, max_pfn);
memcpy(swapper_pg_dir, pgd, sizeof(pgd_t) * PTRS_PER_PGD);
set_pgd(&swapper_pg_dir[KERNEL_PGD_BOUNDARY],
__pgd(__pa(level2_kernel_pgt) | _PAGE_PRESENT));
set_page_prot(level2_kernel_pgt, PAGE_KERNEL_RO);
set_page_prot(swapper_pg_dir, PAGE_KERNEL_RO);
set_page_prot(empty_zero_page, PAGE_KERNEL_RO);
pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, PFN_DOWN(__pa(pgd)));
xen_write_cr3(__pa(swapper_pg_dir));
pin_pagetable_pfn(MMUEXT_PIN_L3_TABLE, PFN_DOWN(__pa(swapper_pg_dir)));
return swapper_pg_dir;
}
#endif /* CONFIG_X86_64 */
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
/* First C function to be called on Xen boot */
asmlinkage void __init xen_start_kernel(void)
{
pgd_t *pgd;
if (!xen_start_info)
return;
BUG_ON(memcmp(xen_start_info->magic, "xen-3", 5) != 0);
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
xen_setup_features();
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
/* Install Xen paravirt ops */
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
pv_info = xen_info;
pv_init_ops = xen_init_ops;
pv_time_ops = xen_time_ops;
pv_cpu_ops = xen_cpu_ops;
pv_irq_ops = xen_irq_ops;
pv_apic_ops = xen_apic_ops;
pv_mmu_ops = xen_mmu_ops;
if (xen_feature(XENFEAT_mmu_pt_update_preserve_ad)) {
pv_mmu_ops.ptep_modify_prot_start = xen_ptep_modify_prot_start;
pv_mmu_ops.ptep_modify_prot_commit = xen_ptep_modify_prot_commit;
}
machine_ops = xen_machine_ops;
#ifdef CONFIG_X86_64
/* Disable until direct per-cpu data access. */
have_vcpu_info_placement = 0;
x86_64_init_pda();
#endif
xen_smp_init();
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
/* Get mfn list */
if (!xen_feature(XENFEAT_auto_translated_physmap))
xen_build_dynamic_phys_to_machine();
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
pgd = (pgd_t *)xen_start_info->pt_base;
/* Prevent unwanted bits from being set in PTEs. */
__supported_pte_mask &= ~_PAGE_GLOBAL;
if (!is_initial_xendomain())
__supported_pte_mask &= ~(_PAGE_PWT | _PAGE_PCD);
/* Don't do the full vcpu_info placement stuff until we have a
possible map and a non-dummy shared_info. */
per_cpu(xen_vcpu, 0) = &HYPERVISOR_shared_info->vcpu_info[0];
xen_raw_console_write("mapping kernel into physical memory\n");
pgd = xen_setup_kernel_pagetable(pgd, xen_start_info->nr_pages);
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
init_mm.pgd = pgd;
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
/* keep using Xen gdt for now; no urgent need to change it */
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
pv_info.kernel_rpl = 1;
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
if (xen_feature(XENFEAT_supervisor_mode_kernel))
paravirt: refactor struct paravirt_ops into smaller pv_*_ops This patch refactors the paravirt_ops structure into groups of functionally related ops: pv_info - random info, rather than function entrypoints pv_init_ops - functions used at boot time (some for module_init too) pv_misc_ops - lazy mode, which didn't fit well anywhere else pv_time_ops - time-related functions pv_cpu_ops - various privileged instruction ops pv_irq_ops - operations for managing interrupt state pv_apic_ops - APIC operations pv_mmu_ops - operations for managing pagetables There are several motivations for this: 1. Some of these ops will be general to all x86, and some will be i386/x86-64 specific. This makes it easier to share common stuff while allowing separate implementations where needed. 2. At the moment we must export all of paravirt_ops, but modules only need selected parts of it. This allows us to export on a case by case basis (and also choose which export license we want to apply). 3. Functional groupings make things a bit more readable. Struct paravirt_ops is now only used as a template to generate patch-site identifiers, and to extract function pointers for inserting into jmp/calls when patching. It is only instantiated when needed. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <ak@suse.de> Cc: Zach Amsden <zach@vmware.com> Cc: Avi Kivity <avi@qumranet.com> Cc: Anthony Liguory <aliguori@us.ibm.com> Cc: "Glauber de Oliveira Costa" <glommer@gmail.com> Cc: Jun Nakajima <jun.nakajima@intel.com>
2007-10-16 20:51:29 +02:00
pv_info.kernel_rpl = 0;
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
/* set the limit of our address space */
xen_reserve_top();
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
#ifdef CONFIG_X86_32
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
/* set up basic CPUID stuff */
cpu_detect(&new_cpu_data);
new_cpu_data.hard_math = 1;
new_cpu_data.x86_capability[0] = cpuid_edx(1);
#endif
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
/* Poke various useful things into boot_params */
boot_params.hdr.type_of_loader = (9 << 4) | 0;
boot_params.hdr.ramdisk_image = xen_start_info->mod_start
? __pa(xen_start_info->mod_start) : 0;
boot_params.hdr.ramdisk_size = xen_start_info->mod_len;
boot_params.hdr.cmd_line_ptr = __pa(xen_start_info->cmd_line);
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
if (!is_initial_xendomain()) {
add_preferred_console("xenboot", 0, NULL);
add_preferred_console("tty", 0, NULL);
add_preferred_console("hvc", 0, NULL);
}
xen_raw_console_write("about to get started...\n");
#if 0
xen_raw_printk("&boot_params=%p __pa(&boot_params)=%lx __va(__pa(&boot_params))=%lx\n",
&boot_params, __pa_symbol(&boot_params),
__va(__pa_symbol(&boot_params)));
walk(pgd, &boot_params);
walk(pgd, __va(__pa(&boot_params)));
#endif
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
/* Start the world */
#ifdef CONFIG_X86_32
i386_start_kernel();
#else
x86_64_start_reservations((char *)__pa_symbol(&boot_params));
#endif
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 03:37:04 +02:00
}