Mark inc (0xfe/0 0xff/0) and dec (0xfe/1 0xff/1) as lock prefix capable.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
'level' and 'sptep' are aliases for 'interator.level' and 'iterator.sptep', no
need for them.
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Currently, when we fetch an spte, we only verify that gptes match those that
the walker saw if we build new shadow pages for them.
However, this misses the following race:
vcpu1 vcpu2
walk
change gpte
walk
instantiate sp
fetch existing sp
Fix by validating every gpte, regardless of whether it is used for building
a new sp or not.
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Partition the function into three sections:
- fetching indirect shadow pages (host_level > guest_level)
- fetching direct shadow pages (page_level < host_level <= guest_level)
- the final spte (page_level == host_level)
Instead of the current spaghetti.
A slight change from the original code is that we call validate_direct_spte()
more often: previously we called it only for gw->level, now we also call it for
lower levels. The change should have no effect.
[xiao: fix regression caused by validate_direct_spte() called too late]
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Move the code to check whether a gpte has changed since we fetched it into
a helper.
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Add a helper to verify that a direct shadow page is valid wrt the required
access permissions; drop the page if it is not valid.
Reviewed-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
To clarify spte fetching code, move large spte handling into a helper.
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
To avoid split accesses to 64 bit sptes on i386, use __set_spte() to link
shadow pages together.
(not technically required since shadow pages are __GFP_KERNEL, so upper 32
bits are always clear)
Reviewed-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
To simplify the process of fetching an spte, add a helper that links
a shadow page to an spte.
Reviewed-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Userspace needs to reset and save/restore these MSRs.
The MCE banks are not exposed since their number varies from vcpu to vcpu.
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
When shadow pages are in use sometimes KVM try to emulate an instruction
when it accesses a shadowed page. If emulation fails KVM un-shadows the
page and reenter guest to allow vcpu to execute the instruction. If page
is not in shadow page hash KVM assumes that this was attempt to do MMIO
and reports emulation failure to userspace since there is no way to fix
the situation. This logic has a race though. If two vcpus tries to write
to the same shadowed page simultaneously both will enter emulator, but
only one of them will find the page in shadow page hash since the one who
founds it also removes it from there, so another cpu will report failure
to userspace and will abort the guest.
Fix this by checking (in addition to checking shadowed page hash) that
page that caused the emulation belongs to valid memory slot. If it is
then reenter the guest to allow vcpu to reexecute the instruction.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Currently if guest access address that belongs to memory slot but is not
backed up by page or page is read only KVM treats it like MMIO access.
Remove that capability. It was never part of the interface and should
not be relied upon.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Stanse found that there is an omitted unlock in kvm_create_pit in one fail
path. Add proper unlock there.
Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Cc: Avi Kivity <avi@redhat.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
Cc: Gleb Natapov <gleb@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Gregory Haskins <ghaskins@novell.com>
Cc: kvm@vger.kernel.org
Signed-off-by: Avi Kivity <avi@redhat.com>
Real hardware disregards permission errors when computing page fault error
code bit 0 (page present). Do the same.
Reviewed-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Bit 4 of the page fault error code is set only if EFER.NX is set.
Reviewed-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This patch change to use DstAcc for decoding 'mov AL, moffs'
and introduced SrcAcc for decoding 'mov moffs, AL'.
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
If IOPL check fail, the cli/sti emulate GP and then we should
skip writeback since the default write OP is OP_REG.
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
The source operand of 'mov rm,sreg' is segment register, not
general-purpose register, so remove SrcReg from decoding.
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
'and AL,imm8' should be mask as ByteOp, otherwise the dest operand
length will no correct and we may fill the full EAX when writeback.
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Fix the comment of out instruction, using the same style as the
other instructions.
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
__set_spte() will happily replace an spte with the accessed bit set with
one that has the accessed bit clear. Add a helper update_spte() which checks
for this condition and updates the page flag if needed.
Signed-off-by: Avi Kivity <avi@redhat.com>
Currently, in the window between the check for the accessed bit, and actually
dropping the spte, a vcpu can access the page through the spte and set the bit,
which will be ignored by the mmu.
Fix by using an exchange operation to atmoically fetch the spte and drop it.
Signed-off-by: Avi Kivity <avi@redhat.com>
When we call rmap_remove(), we (almost) always immediately follow it by
an __set_spte() to a nonpresent pte. Since we need to perform the two
operations atomically, to avoid losing the dirty and accessed bits, introduce
a helper drop_spte() and convert all call sites.
The operation is still nonatomic at this point.
Signed-off-by: Avi Kivity <avi@redhat.com>
Commit 341d9b535b6c simplify reload logic while entry guest mode, it
can avoid unnecessary sync-root if KVM_REQ_MMU_RELOAD and
KVM_REQ_MMU_SYNC both set.
But, it cause a issue that when we handle 'KVM_REQ_TLB_FLUSH', the
root is invalid, it is triggered during my test:
Kernel BUG at ffffffffa00212b8 [verbose debug info unavailable]
......
Fixed by directly return if the root is not ready.
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
This patch converts unnecessary divide and modulo operations
in the KVM large page related code into logical operations.
This allows to convert gfn_t to u64 while not breaking 32
bit builds.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Cleanup this function that we are already get the direct sp's access
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
If the mapping is writable but the dirty flag is not set, we will find
the read-only direct sp and setup the mapping, then if the write #PF
occur, we will mark this mapping writable in the read-only direct sp,
now, other real read-only mapping will happily write it without #PF.
It may hurt guest's COW
Fixed by re-install the mapping when write #PF occur.
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
In no-direct mapping, we mark sp is 'direct' when we mapping the
guest's larger page, but its access is encoded form upper page-struct
entire not include the last mapping, it will cause access conflict.
For example, have this mapping:
[W]
/ PDE1 -> |---|
P[W] | | LPA
\ PDE2 -> |---|
[R]
P have two children, PDE1 and PDE2, both PDE1 and PDE2 mapping the
same lage page(LPA). The P's access is WR, PDE1's access is WR,
PDE2's access is RO(just consider read-write permissions here)
When guest access PDE1, we will create a direct sp for LPA, the sp's
access is from P, is W, then we will mark the ptes is W in this sp.
Then, guest access PDE2, we will find LPA's shadow page, is the same as
PDE's, and mark the ptes is RO.
So, if guest access PDE1, the incorrect #PF is occured.
Fixed by encode the last mapping access into direct shadow page
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
While we sync many unsync sp at one time(in mmu_sync_children()),
we may mapping the spte writable, it's dangerous, if one unsync
sp's mapping gfn is another unsync page's gfn.
For example:
SP1.pte[0] = P
SP2.gfn's pfn = P
[SP1.pte[0] = SP2.gfn's pfn]
First, we write protected SP1 and SP2, but SP1 and SP2 are still the
unsync sp.
Then, sync SP1 first, it will detect SP1.pte[0].gfn only has one unsync-sp,
that is SP2, so it will mapping it writable, but we plan to sync SP2 soon,
at this point, the SP2->unsync is not reliable since later we sync SP2 but
SP2->gfn is already writable.
So the final result is: SP2 is the sync page but SP2.gfn is writable.
This bug will corrupt guest's page table, fixed by mark read-only mapping
if the mapped gfn has shadow pages.
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Some guest device driver may leverage the "Non-Snoop" I/O, and explicitly
WBINVD or CLFLUSH to a RAM space. Since migration may occur before WBINVD or
CLFLUSH, we need to maintain data consistency either by:
1: flushing cache (wbinvd) when the guest is scheduled out if there is no
wbinvd exit, or
2: execute wbinvd on all dirty physical CPUs when guest wbinvd exits.
Signed-off-by: Yaozu (Eddie) Dong <eddie.dong@intel.com>
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
No need to reload the mmu in between two different vcpu->requests checks.
kvm_mmu_reload() may trigger KVM_REQ_TRIPLE_FAULT, but that will be caught
during atomic guest entry later.
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Older versions of 32-bit linux have a "Checking 'hlt' instruction"
test where they repeatedly call the 'hlt' instruction, and then
expect a timer interrupt to kick the CPU out of halt. This happens
before any LAPIC or IOAPIC setup happens, which means that all of
the APIC's are in virtual wire mode at this point. Unfortunately,
the current implementation of virtual wire mode is hardcoded to
only kick the BSP, so if a crash+kexec occurs on a different
vcpu, it will never get kicked.
This patch makes pic_unlock() do the equivalent of
kvm_irq_delivery_to_apic() for the IOAPIC code. That is, it runs
through all of the vcpus looking for one that is in virtual wire
mode. In the normal case where LAPICs and IOAPICs are configured,
this won't be used at all. In the bootstrap phase of a modern
OS, before the LAPICs and IOAPICs are configured, this will have
exactly the same behavior as today; VCPU0 is always looked at
first, so it will always get out of the loop after the first
iteration. This will only go through the loop more than once
during a kexec/kdump, in which case it will only do it a few times
until the kexec'ed kernel programs the LAPIC and IOAPIC.
Signed-off-by: Chris Lalancette <clalance@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Enable Intel(R) Advanced Vector Extension(AVX) for guest.
The detection of AVX feature includes OSXSAVE bit testing. When OSXSAVE bit is
not set, even if AVX is supported, the AVX instruction would result in UD as
well. So we're safe to expose AVX bits to guest directly.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
If a process with a memory slot is COWed, the page will change its address
(despite having an elevated reference count). This breaks internal memory
slots which have their physical addresses loaded into vmcs registers (see
the APIC access memory slot).
Signed-off-by: Avi Kivity <avi@redhat.com>
Part of the i8259 code pretends it isn't part of kvm, but we know better.
Reduce excessive abstraction, eliminating callbacks and void pointers.
Signed-off-by: Avi Kivity <avi@redhat.com>
As advertised in feature-removal-schedule.txt. Equivalent support is provided
by overlapping memory regions.
Signed-off-by: Avi Kivity <avi@redhat.com>
Instead of three temporary variables and three free calls, have one temporary
variable (with four names) and one free call.
Signed-off-by: Avi Kivity <avi@redhat.com>
Group 3 instruction with ModRM reg field as 001 is
defined as test instruction under AMD arch, and
emulate_grp3() is ready for emulate it, so fix the
decoding.
static inline int emulate_grp3(...)
{
...
switch (c->modrm_reg) {
case 0 ... 1: /* test */
emulate_2op_SrcV("test", c->src, c->dst, ctxt->eflags);
...
}
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
If the guest wants to accept timer interrupts on a CPU other
than the BSP, we need to remove this gate.
Signed-off-by: Chris Lalancette <clalance@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
We really want to "kvm_set_irq" during the hrtimer callback,
but that is risky because that is during interrupt context.
Instead, offload the work to a workqueue, which is a bit safer
and should provide most of the same functionality.
Signed-off-by: Chris Lalancette <clalance@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
emulate pusha instruction only writeback the last
EDI register, but the other registers which need
to be writeback is ignored. This patch fixed it.
Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Fix a slight error with assertion in local APIC code.
Signed-off-by: Zachary Amsden <zamsden@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
While we mark the parent's unsync_child_bitmap, if the parent is already
unsynced, it no need walk it's parent, it can reduce some unnecessary
workload
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
In current code, some page's unsync_child_bitmap is not cleared completely
in mmu_sync_children(), for example, if two PDPEs shard one PDT, one of
PDPE's unsync_child_bitmap is not cleared.
Currently, it not harm anything just little overload, but it's the prepare
work for the later patch
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
If the sync-sp just sync transient, don't mark its pte notrap
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
The sync page is already write protected in mmu_sync_children(), don't
write protected it again
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Rename 'page' and 'shadow_page' to 'sp' to better fit the context
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
On Intel, we call skip_emulated_instruction() even if we injected a #GP,
resulting in the #GP pointing at the wrong address.
Fix by injecting the exception and skipping the instruction at the same place,
so we can do just one or the other.
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
On Intel, we call skip_emulated_instruction() even if we injected a #GP,
resulting in the #GP pointing at the wrong address.
Fix by injecting the exception and skipping the instruction at the same place,
so we can do just one or the other.
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
On Intel, we call skip_emulated_instruction() even if we injected a #GP,
resulting in the #GP pointing at the wrong address.
Fix by injecting the exception and skipping the instruction at the same place,
so we can do just one or the other.
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
This patch enable guest to use XSAVE/XRSTOR instructions.
We assume that host_xcr0 would use all possible bits that OS supported.
And we loaded xcr0 in the same way we handled fpu - do it as late as we can.
Signed-off-by: Dexuan Cui <dexuan.cui@intel.com>
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Reviewed-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
delay local tlb flush until enter guest moden, it can reduce vpid flush
frequency and reduce remote tlb flush IPI(if KVM_REQ_TLB_FLUSH bit is
already set, IPI is not sent)
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Use kvm_mmu_flush_tlb() function instead of calling
kvm_x86_ops->tlb_flush(vcpu) directly.
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This remote tlb flush is no necessary since we have synced while
sp is zapped
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
The name "pid_sync_vcpu_all" isn't appropriate since it just affect
a single vpid, so rename it to vpid_sync_vcpu_single().
Signed-off-by: Gui Jianfeng <guijianfeng@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Using kvm_mmu_prepare_zap_page() and kvm_mmu_zap_page() instead of
kvm_mmu_zap_page() that can reduce remote tlb flush IPI
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
In the later patch, we will modify sp's zapping way like below:
kvm_mmu_prepare_zap_page A
kvm_mmu_prepare_zap_page B
kvm_mmu_prepare_zap_page C
....
kvm_mmu_commit_zap_page
[ zaped multiple sps only need to call kvm_mmu_commit_zap_page once ]
In __kvm_mmu_free_some_pages() function, the free page number is
getted form 'vcpu->kvm->arch.n_free_mmu_pages' in loop, it will
hinders us to apply kvm_mmu_prepare_zap_page() and kvm_mmu_commit_zap_page()
since kvm_mmu_prepare_zap_page() not free sp.
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Using kvm_mmu_prepare_zap_page() and kvm_mmu_commit_zap_page() to
split kvm_mmu_zap_page() function, then we can:
- traverse hlist safely
- easily to gather remote tlb flush which occurs during page zapped
Those feature can be used in the later patches
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
In kvm_mmu_unprotect_page(), the invalid sp can be skipped
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
According to SDM, we need check whether single-context INVVPID type is supported
before issuing invvpid instruction.
Signed-off-by: Gui Jianfeng <guijianfeng@cn.fujitsu.com>
Reviewed-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Should use linux/uaccess.h instead of asm/uaccess.h
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
The vmexit handler returns KVM_EXIT_UNKNOWN since there is no handler
for vmentry failures. This intercepts vmentry failures and returns
KVM_FAIL_ENTRY to userspace instead.
Signed-off-by: Mohammed Gamal <m.gamal005@gmail.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
There's no need to calculate quadrant if tdp is enabled.
Signed-off-by: Gui Jianfeng <guijianfeng@cn.fujitsu.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Memory allocation may fail. Propagate such errors.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Reviewed-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Must set VCPU register to be the guest notion of EFER even if that
setting is not valid on hardware. This was masked by the set in
set_efer until 7657fd5ace88e8092f5f3a84117e093d7b893f26 broke that.
Fix is simply to set the VCPU register before stripping bits.
Signed-off-by: Zachary Amsden <zamsden@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
In Documentation/kvm/mmu.txt:
gfn:
Either the guest page table containing the translations shadowed by this
page, or the base page frame for linear translations. See role.direct.
But in function FNAME(fetch)(), sp->gfn is incorrect when one of following
situations occurred:
1) guest is 32bit paging and the guest PDE maps a 4-MByte page
(backed by 4k host pages), FNAME(fetch)() miss handling the quadrant.
And if guest use pse-36, "table_gfn = gpte_to_gfn(gw->ptes[level - delta]);"
is incorrect.
2) guest is long mode paging and the guest PDPTE maps a 1-GByte page
(backed by 4k or 2M host pages).
So we fix it to suit to the document and suit to the code which
requires sp->gfn correct when sp->role.direct=1.
We use the goal mapping gfn(gw->gfn) to calculate the base page frame
for linear translations, it is simple and easy to be understood.
Reported-by: Marcelo Tosatti <mtosatti@redhat.com>
Reported-by: Gui Jianfeng <guijianfeng@cn.fujitsu.com>
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
In Document/kvm/mmu.txt:
gfn:
Either the guest page table containing the translations shadowed by this
page, or the base page frame for linear translations. See role.direct.
But in __direct_map(), the base gfn calculation is incorrect,
it does not calculate correctly when level=3 or 4.
Fix by using PT64_LVL_ADDR_MASK() which accounts for all levels correctly.
Reported-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
When sp->role.direct is set, sp->gfns does not contain any essential
information, leaf sptes reachable from this sp are for a continuous
guest physical memory range (a linear range).
So sp->gfns[i] (if it was set) equals to sp->gfn + i. (PT_PAGE_TABLE_LEVEL)
Obviously, it is not essential information, we can calculate it when need.
It means we don't need sp->gfns when sp->role.direct=1,
Thus we can save one page usage for every kvm_mmu_page.
Note:
Access to sp->gfns must be wrapped by kvm_mmu_page_get_gfn()
or kvm_mmu_page_set_gfn().
It is only exposed in FNAME(sync_page).
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Allow more page become asynchronous at getting sp time, if need create new
shadow page for gfn but it not allow unsync(level > 1), we should unsync all
gfn's unsync page
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
In current code, shadow page can become asynchronous only if one
shadow page for a gfn, this rule is too strict, in fact, we can
let all last mapping page(i.e, it's the pte page) become unsync,
and sync them at invlpg or flush tlb time.
This patch allow more page become asynchronous at gfn mapping time
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
On SVM interrupts are injected by svm_set_irq() not svm_inject_irq().
The later is used only to wait for irq window.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Only unsync pages need updated at invlpg time since other shadow
pages are write-protected
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Two cases maybe happen in kvm_mmu_get_page() function:
- one case is, the goal sp is already in cache, if the sp is unsync,
we only need update it to assure this mapping is valid, but not
mark it sync and not write-protect sp->gfn since it not broke unsync
rule(one shadow page for a gfn)
- another case is, the goal sp not existed, we need create a new sp
for gfn, i.e, gfn (may)has another shadow page, to keep unsync rule,
we should sync(mark sync and write-protect) gfn's unsync shadow page.
After enabling multiple unsync shadows, we sync those shadow pages
only when the new sp not allow to become unsync(also for the unsyc
rule, the new rule is: allow all pte page become unsync)
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Split kvm_sync_page() into kvm_sync_page() and kvm_sync_page_transient()
to clarify the code address Avi's suggestion
kvm_sync_page_transient() function only update shadow page but not mark
it sync and not write protect sp->gfn. it will be used by later patch
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
We can avoid unnecessary fpu load when userspace process
didn't use FPU frequently.
Derived from Avi's idea.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Now that all arch specific ioctls have centralized locking, it is easy to
move it to the central dispatcher.
Signed-off-by: Avi Kivity <avi@redhat.com>
All vcpu ioctls need to be locked, so instead of locking each one specifically
we lock at the generic dispatcher.
This patch only updates generic ioctls and leaves arch specific ioctls alone.
Signed-off-by: Avi Kivity <avi@redhat.com>
Remove rmap before clear spte otherwise it will trigger BUG_ON() in
some functions such as rmap_write_protect().
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Only modifying some bits of CR0/CR4 needs paging mode switch.
Modify EFER.NXE bit would result in reserved bit updates.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
mmu.free() already set root_hpa to INVALID_PAGE, no need to do it again in the
destory_kvm_mmu().
kvm_x86_ops->set_cr4() and set_efer() already assign cr4/efer to
vcpu->arch.cr4/efer, no need to do it again later.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
This adds missing decoder flags for sub instructions (opcodes 0x2c - 0x2d)
Signed-off-by: Mohammed Gamal <m.gamal005@gmail.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
This adds test acc, imm instruction to the x86 emulator
Signed-off-by: Mohammed Gamal <m.gamal005@gmail.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
SDM suggests VMXON should be called before VMPTRLD, and VMXOFF
should be called after doing VMCLEAR.
Therefore in vmm coexistence case, we should firstly call VMXON
before any VMCS operation, and then call VMXOFF after the
operation is done.
Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Originally VMCLEAR/VMPTRLD is called on vcpu migration. To
support hosted VMM coexistance, VMCLEAR is executed on vcpu
schedule out, and VMPTRLD is executed on vcpu schedule in.
This could also eliminate the IPI when doing VMCLEAR.
Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Do some preparations for vmm coexistence support.
Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Define vmcs_load() and kvm_cpu_vmxon() to avoid direct call of asm
code. Also move VMXE bit operation out of kvm_cpu_vmxoff().
Signed-off-by: Dongxiao Xu <dongxiao.xu@intel.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
We drop the mmu lock between freeing memory and allocating the roots; this
allows some other vcpu to sneak in and allocate memory.
While the race is benign (resulting only in temporary overallocation, not oom)
it is simple and easy to fix by moving the freeing close to the allocation.
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Do not kill VM when instruction emulation fails. Inject #UD and report
failure to userspace instead. Userspace may choose to reenter guest if
vcpu is in userspace (cpl == 3) in which case guest OS will kill
offending process and continue running.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Currently, kvm_mmu_zap_page() returning the number of freed children sp.
This might confuse the caller, because caller don't know the actual freed
number. Let's make kvm_mmu_zap_page() return the number of pages it actually
freed.
Signed-off-by: Gui Jianfeng <guijianfeng@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Sometime cmpxchg_gpte doesn't modify gpte, in such case, don't mark
page table page as dirty.
Signed-off-by: Gui Jianfeng <guijianfeng@cn.fujitsu.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This patch enables setting of efer bit 13 which is allowed
in all SVM capable processors. This is necessary for the
SLES11 version of Xen 4.0 to boot with nested svm.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This patch adds a function to dump the vmcb into the kernel
log and calls it after a failed vmrun to ease debugging.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
KVM_REQ_KICK poisons vcpu->requests by having a bit set during normal
operation. This causes the fast path check for a clear vcpu->requests
to fail all the time, triggering tons of atomic operations.
Fix by replacing KVM_REQ_KICK with a vcpu->guest_mode atomic.
Signed-off-by: Avi Kivity <avi@redhat.com>
Return exception as a result of instruction emulation and handle
injection in KVM code.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
emulate_sysexit() should use shadowed registers copy instead of
looking into vcpu state directly.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Return new RIP as part of instruction emulation result instead of
updating KVM's RIP from x86 emulator code.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Return error to x86 emulator instead of injection exception behind its back.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
It is not called directly outside of the file it's defined in anymore.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Currently emulator returns -1 when emulation failed or IO is needed.
Caller tries to guess whether emulation failed by looking at other
variables. Make it easier for caller to recognise error condition by
always returning -1 in case of failure. For this new emulator
internal return value X86EMUL_IO_NEEDED is introduced. It is used to
distinguish between error condition (which returns X86EMUL_UNHANDLEABLE)
and condition that requires IO exit to userspace to continue emulation.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Fill in run->mmio details in (read|write)_emulated function just like
pio does. There is no point in filling only vcpu fields there just to
copy them into vcpu->run a little bit later.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Make (get|set)_dr() callback return error if it fails instead of
injecting exception behind emulator's back.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Make set_cr() callback return error if it fails instead of injecting #GP
behind emulator's back.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Use callbacks from x86_emulate_ops to access segments instead of calling
into kvm directly.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
On VMX it is expensive to call get_cached_descriptor() just to get segment
base since multiple vmcs_reads are done instead of only one. Introduce
new call back get_cached_segment_base() for efficiency.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Add (set|get)_msr callbacks to x86_emulate_ops instead of calling
them directly.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Add (set|get)_dr callbacks to x86_emulate_ops instead of calling
them directly.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
ljmp/lcall instruction operand contains address and segment.
It can be 10 bytes long. Currently we decode it as two different
operands. Fix it by introducing new kind of operand that can hold
entire far address.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Dst operand is already initialized during decoding stage. No need to
reinitialize.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This instruction does not need generic decoding for its dst operand.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Introduce read cache which is needed for instruction that require more
then one exit to userspace. After returning from userspace the instruction
will be re-executed with cached read value.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
cr0.ts may change between entries, so we copy cr0 to HOST_CR0 before each
entry. That is slow, so instead, set HOST_CR0 to have TS set unconditionally
(which is a safe value), and issue a clts() just before exiting vcpu context
if the task indeed owns the fpu.
Saves ~50 cycles/exit.
Signed-off-by: Avi Kivity <avi@redhat.com>
Although we always allocate a new dirty bitmap in x86's get_dirty_log(),
it is only used as a zero-source of copy_to_user() and freed right after
that when memslot is clean. This patch uses clear_user() instead of doing
this unnecessary zero-source allocation.
Performance improvement: as we can expect easily, the time needed to
allocate a bitmap is completely reduced. In my test, the improved ioctl
was about 4 to 10 times faster than the original one for clean slots.
Furthermore, reducing memory allocations and copies will produce good
effects to caches too.
Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
Signed-off-by: Avi Kivity <avi@redhat.com>
In common cases, guest SRAO MCE will cause corresponding poisoned page
be un-mapped and SIGBUS be sent to QEMU-KVM, then QEMU-KVM will relay
the MCE to guest OS.
But it is reported that if the poisoned page is accessed in guest
after unmapping and before MCE is relayed to guest OS, userspace will
be killed.
The reason is as follows. Because poisoned page has been un-mapped,
guest access will cause guest exit and kvm_mmu_page_fault will be
called. kvm_mmu_page_fault can not get the poisoned page for fault
address, so kernel and user space MMIO processing is tried in turn. In
user MMIO processing, poisoned page is accessed again, then userspace
is killed by force_sig_info.
To fix the bug, kvm_mmu_page_fault send HWPOISON signal to QEMU-KVM
and do not try kernel and user space MMIO processing for poisoned
page.
[xiao: fix warning introduced by avi]
Reported-by: Max Asbock <masbock@linux.vnet.ibm.com>
Signed-off-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Use the AMD errata checking framework instead of open-coding the test.
Signed-off-by: Hans Rosenfeld <hans.rosenfeld@amd.com>
LKML-Reference: <1280336972-865982-3-git-send-email-hans.rosenfeld@amd.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>