Kevin VanMaren a1002a48e1 [PATCH] x86_64: When allocation of merged SG lists fails in the IOMMU don't merge ... [ AK: I redid Kevin's fix to be simpler, but the idea and original analysis of the problem is from Kevin] This avoid allocation failures on some SATA systems like Nvidia CK8 when the IOMMU gets fragmented. Modern SATA devices have quite large queues (128 entries) and the FS with ext2/3 is good enough now that it often passes whole 128 page sg lists down to the driver. These require 512K of continuous free space in the IOMMU aperture to map when merged. When the IOMMU is fragmented this could lead to spurious IO errors due to failing mappings. Short term fix is to just try to map the SG list again unmerged page by page - this way fragmentation doesn't matter anymore. The code for that was already there, but it just wasn't enabled for the merge case. According to Kevin at least the Nvidia device doesn't seem to benefit from merging much anyways, so the only slowdown is from trying to do an unnecessary merge attempt. Kevin plans to implement better fragmentation avoidance in the future, but that wouldn't be 2.6.16 material. TBD: should add some statistic counters to count how often that really happens. Signed-off-by: Andi Kleen <ak@suse.de> Signed-off-by: Linus Torvalds <torvalds@osdl.org>		2006-02-04 16:43:14 -08:00
..
acpi	[ACPI] Avoid BIOS inflicted crashes by evaluating _PDC only once	2005-12-01 01:30:35 -05:00
cpufreq	Linux-2.6.12-rc2	2005-04-16 15:20:36 -07:00
aperture.c	[PATCH] x86_64: Fix off by one in IOMMU check	2006-01-11 19:04:54 -08:00
apic.c	[PATCH] x86_64: mark two routines as __cpuinit	2006-02-04 16:43:14 -08:00
asm-offsets.c	[PATCH] x86_64: Allow nesting of int3 by default for kprobes	2006-01-16 11:27:58 -08:00
crash.c	[PATCH] kdump: x86_64 save cpu registers upon crash	2006-01-10 08:01:28 -08:00
crash_dump.c	[PATCH] kdump: read previous kernel's memory	2006-01-10 08:01:28 -08:00
e820.c	[PATCH] kdump: x86_64: add memmmap command line option	2006-01-10 08:01:27 -08:00
early_printk.c	[PATCH] i386/x86-64: make setup_early_printk() usage consistent	2006-01-11 19:05:04 -08:00
entry.S	[PATCH] x86_64: Disallow kprobes on NMI handlers	2006-02-04 16:43:12 -08:00
genapic.c	[PATCH] x86-64: Use physflat on Intel for < 8 CPUs with CPU hotplug	2005-09-12 10:50:56 -07:00
genapic_cluster.c	[PATCH] x86_64/i386: Remove preempt disable calls in lowlevel IPI	2006-01-11 19:01:57 -08:00
genapic_flat.c	[PATCH] x86_64/i386: Remove preempt disable calls in lowlevel IPI	2006-01-11 19:01:57 -08:00
head.S	[PATCH] x86_64: eliminate empty_bad_{page,{pte,pmd}_table}	2006-01-16 23:18:35 -08:00
head64.c	[PATCH] i386/x86-64: make setup_early_printk() usage consistent	2006-01-11 19:05:04 -08:00
i387.c	[PATCH] amd64: task_thread_info()	2006-01-12 09:08:51 -08:00
i8259.c	[PATCH] amd64: task_thread_info()	2006-01-12 09:08:51 -08:00
init_task.c	[PATCH] Change maxaligned_in_smp alignemnt macros to internodealigned_in_smp macros	2006-01-08 20:13:38 -08:00
io_apic.c	[PATCH] x86_64: Automatically enable apicmaintimer on ATI boards	2006-02-04 16:43:13 -08:00
ioport.c	[PATCH] capable/capability.h (arch/)	2006-01-11 18:42:14 -08:00
irq.c	[PATCH] x86_64: Node local pda take 2 -- cpu_pda preparation	2006-01-11 19:04:59 -08:00
kprobes.c	[PATCH] kprobes: fix race in recovery of reentrant probe	2006-01-11 18:42:12 -08:00
ldt.c	Linux-2.6.12-rc2	2005-04-16 15:20:36 -07:00
machine_kexec.c	[PATCH] x86_64 machine_kexec: Use standard pagetable helpers	2005-07-29 13:12:49 -07:00
Makefile	[PATCH] x86_64: Inclusion of ScaleMP vSMP architecture patches - vsmp_arch	2006-01-11 19:05:01 -08:00
mce.c	[PATCH] x86_64: data/functions wrongly marked as __init with cpu hotplug.	2006-02-04 16:43:14 -08:00
mce_amd.c	[PATCH] x86_64: Sparse warnings fix.	2006-01-11 19:05:00 -08:00
mce_intel.c	[PATCH] x86_64: Some housekeeping in local APIC code	2006-01-11 19:05:03 -08:00
module.c	[PATCH] uml: kludgy compilation fixes for x86-64 subarch modules support	2005-05-05 16:36:33 -07:00
mpparse.c	[ACPI] acpi_register_gsi() fix needed for ACPICA 20051021	2005-12-28 02:50:44 -05:00
nmi.c	[PATCH] x86_64: Disallow kprobes on NMI handlers	2006-02-04 16:43:12 -08:00
pci-dma.c	[PATCH] x86_64: Fix swiotlb dma_alloc_coherent fallback	2006-02-04 16:43:13 -08:00
pci-gart.c	[PATCH] x86_64: When allocation of merged SG lists fails in the IOMMU don't merge	2006-02-04 16:43:14 -08:00
pci-nommu.c	[PATCH] x86_64: Tell user to enable GART_IOMMU when needed	2006-01-11 19:04:59 -08:00
pci-swiotlb.c	[PATCH] x86_64: Use function pointers to call DMA mapping functions	2006-01-11 19:04:55 -08:00
pmtimer.c	[PATCH] x86_64: [PATCH] timer resume	2006-02-04 16:43:13 -08:00
process.c	[PATCH] amd64: task_stack_page()	2006-01-12 09:08:51 -08:00
ptrace.c	[PATCH] amd64: task_pt_regs()	2006-01-12 09:08:51 -08:00
reboot.c	[PATCH] i386/x86-64: Don't IPI to offline cpus on shutdown	2006-01-11 19:04:50 -08:00
relocate_kernel.S	[PATCH] kexec: x86_64 kexec implementation	2005-06-25 16:24:50 -07:00
setup.c	[PATCH] x86_64: Undo the earlier changes to remove unrolled copy/memset functions	2006-02-04 16:43:13 -08:00
setup64.c	[PATCH] x86_64: Fix MCE exception stack for boot CPU	2006-01-18 19:20:16 -08:00
signal.c	[PATCH] x86_64: Save/restore CS in 64bit signal handlers and force __USER_CS for CS	2005-11-14 19:55:15 -08:00
smp.c	[PATCH] x86_64: Remove useless KDB vector	2006-01-11 19:05:00 -08:00
smpboot.c	[PATCH] x86_64: Fix the node cpumask of a cpu going down	2006-02-04 16:43:13 -08:00
suspend.c	[PATCH] x86_64: Align and pad x86_64 GDT on page boundary	2006-01-11 19:04:53 -08:00
suspend_asm.S	[PATCH] x86_64: Set up safe page tables during resume	2005-10-10 08:36:46 -07:00
sys_x86_64.c	[PATCH] x86_64: Use common sys_time64	2005-11-14 19:55:17 -08:00
syscall.c	[PATCH] x86/x86_64: mark rodata section read-only: make some datastructures const	2006-01-06 08:33:36 -08:00
time.c	[PATCH] x86_64: mark two routines as __cpuinit	2006-02-04 16:43:14 -08:00
trampoline.S	[PATCH] x86_64: Fix SMP bootup with CONFIG_KDUMP enabled	2006-01-11 19:09:58 -08:00
traps.c	[PATCH] x86_64: Disallow kprobes on NMI handlers	2006-02-04 16:43:12 -08:00
vmlinux.lds.S	[PATCH] x86_64: Allow compilation on a 32bit biarch toolchain	2006-01-11 19:04:52 -08:00
vsmp.c	[PATCH] x86_64: Inclusion of ScaleMP vSMP architecture patches - vsmp_arch	2006-01-11 19:05:01 -08:00
vsyscall.c	[PATCH] x86_64: Use standard __always_inline in vsyscall.c	2006-01-11 19:04:58 -08:00
x8664_ksyms.c	[PATCH] x86_64: No need to export get_cmos_time anymore	2006-01-11 19:05:02 -08:00