diff options
| author |   Jeremy Fitzhardinge <jeremy@goop.org> | 2007-02-13 13:26:20 +0100 |
|---|---|---|
| committer |   Andi Kleen <andi@basil.nowhere.org> | 2007-02-13 13:26:20 +0100 |
| commit | 464d1a78fbf8cf6c7fd970e7b3e2db50a320ce28 (patch) | |
| tree | 536d8a92976e675b484b35dec88d40c97fab8ac8 /arch/i386/math-emu | |
| parent | [PATCH] x86-64: x86_64-make-the-numa-hash-function-nodemap-allocation fix fix (diff) | |
| download | wireguard-linux-464d1a78fbf8cf6c7fd970e7b3e2db50a320ce28.tar.xz wireguard-linux-464d1a78fbf8cf6c7fd970e7b3e2db50a320ce28.zip | |
[PATCH] i386: Convert i386 PDA code to use %fs
Convert the PDA code to use %fs rather than %gs as the segment for
per-processor data. This is because some processors show a small but
measurable performance gain for reloading a NULL segment selector (as %fs
generally is in user-space) versus a non-NULL one (as %gs generally is).
On modern processors the difference is very small, perhaps undetectable.
Some old AMD "K6 3D+" processors are noticably slower when %fs is used
rather than %gs; I have no idea why this might be, but I think they're
sufficiently rare that it doesn't matter much.
This patch also fixes the math emulator, which had not been adjusted to
match the changed struct pt_regs.
[frederik.deweerdt@gmail.com: fixit with gdb]
[mingo@elte.hu: Fix KVM too]
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Ian Campbell <Ian.Campbell@XenSource.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Zachary Amsden <zach@vmware.com>
Cc: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Frederik Deweerdt <frederik.deweerdt@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Diffstat (limited to 'arch/i386/math-emu')
| -rw-r--r-- | arch/i386/math-emu/get_address.c | 14 |
1 files changed, 5 insertions, 9 deletions
diff --git a/arch/i386/math-emu/get_address.c b/arch/i386/math-emu/get_address.c index 9819b705efa4..2e2c51a8bd3a 100644 --- a/arch/i386/math-emu/get_address.c +++ b/arch/i386/math-emu/get_address.c @@ -56,15 +56,14 @@ static int reg_offset_vm86[] = { #define VM86_REG_(x) (*(unsigned short *) \ (reg_offset_vm86[((unsigned)x)]+(u_char *) FPU_info)) -/* These are dummy, fs and gs are not saved on the stack. */ -#define ___FS ___ds +/* This dummy, gs is not saved on the stack. */ #define ___GS ___ds static int reg_offset_pm[] = { offsetof(struct info,___cs), offsetof(struct info,___ds), offsetof(struct info,___es), - offsetof(struct info,___FS), + offsetof(struct info,___fs), offsetof(struct info,___GS), offsetof(struct info,___ss), offsetof(struct info,___ds) @@ -169,13 +168,10 @@ static long pm_address(u_char FPU_modrm, u_char segment, switch ( segment ) { - /* fs and gs aren't used by the kernel, so they still have their - user-space values. */ - case PREFIX_FS_-1: - /* N.B. - movl %seg, mem is a 2 byte write regardless of prefix */ - savesegment(fs, addr->selector); - break; + /* gs isn't used by the kernel, so it still has its + user-space value. */ case PREFIX_GS_-1: + /* N.B. - movl %seg, mem is a 2 byte write regardless of prefix */ savesegment(gs, addr->selector); break; default: |