AcoSail/kernel_optimize_test
8
0
Fork 0
forked from luck/tmp_suning_uos_patched
Code Pull Requests Activity
7ceeb6a40a
kernel_optimize_test/drivers/md/raid6sse2.c
Dan Williams f701d589aa md/raid6: move raid6 data processing to raid6_pq.ko 
Move the raid6 data processing routines into a standalone module
(raid6_pq) to prepare them to be called from async_tx wrappers and other
non-md drivers/modules.  This precludes a circular dependency of raid456
needing the async modules for data processing while those modules in
turn depend on raid456 for the base level synchronous raid6 routines.

To support this move:
1/ The exportable definitions in raid6.h move to include/linux/raid/pq.h
2/ The raid6_call, recovery calls, and table symbols are exported
3/ Extra #ifdef __KERNEL__ statements to enable the userspace raid6test to
   compile

Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: NeilBrown <neilb@suse.de>
2009-03-31 15:09:39 +11:00

263 lines
8.3 KiB
C
Raw Blame History

/* -*- linux-c -*- ------------------------------------------------------- *
*
* Copyright 2002 H. Peter Anvin - All Rights Reserved
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, Inc., 53 Temple Place Ste 330,
* Boston MA 02111-1307, USA; either version 2 of the License, or
* (at your option) any later version; incorporated herein by reference.
*
* ----------------------------------------------------------------------- */
/*
* raid6sse2.c
*
* SSE-2 implementation of RAID-6 syndrome functions
*
*/
#if (defined(__i386__) || defined(__x86_64__)) && !defined(__arch_um__)
#include <linux/raid/pq.h>
#include "raid6x86.h"
static const struct raid6_sse_constants {
u64 x1d[2];
} raid6_sse_constants __attribute__((aligned(16))) = {
{ 0x1d1d1d1d1d1d1d1dULL, 0x1d1d1d1d1d1d1d1dULL },
};
static int raid6_have_sse2(void)
{
/* Not really boot_cpu but "all_cpus" */
return boot_cpu_has(X86_FEATURE_MMX) &&
boot_cpu_has(X86_FEATURE_FXSR) &&
boot_cpu_has(X86_FEATURE_XMM) &&
boot_cpu_has(X86_FEATURE_XMM2);
}
/*
* Plain SSE2 implementation
*/
static void raid6_sse21_gen_syndrome(int disks, size_t bytes, void **ptrs)
{
u8 **dptr = (u8 **)ptrs;
u8 *p, *q;
int d, z, z0;
z0 = disks - 3; /* Highest data disk */
p = dptr[z0+1]; /* XOR parity */
q = dptr[z0+2]; /* RS syndrome */
kernel_fpu_begin();
asm volatile("movdqa %0,%%xmm0" : : "m" (raid6_sse_constants.x1d[0]));
asm volatile("pxor %xmm5,%xmm5"); /* Zero temp */
for ( d = 0 ; d < bytes ; d += 16 ) {
asm volatile("prefetchnta %0" : : "m" (dptr[z0][d]));
asm volatile("movdqa %0,%%xmm2" : : "m" (dptr[z0][d])); /* P[0] */
asm volatile("prefetchnta %0" : : "m" (dptr[z0-1][d]));
asm volatile("movdqa %xmm2,%xmm4"); /* Q[0] */
asm volatile("movdqa %0,%%xmm6" : : "m" (dptr[z0-1][d]));
for ( z = z0-2 ; z >= 0 ; z-- ) {
asm volatile("prefetchnta %0" : : "m" (dptr[z][d]));
asm volatile("pcmpgtb %xmm4,%xmm5");
asm volatile("paddb %xmm4,%xmm4");
asm volatile("pand %xmm0,%xmm5");
asm volatile("pxor %xmm5,%xmm4");
asm volatile("pxor %xmm5,%xmm5");
asm volatile("pxor %xmm6,%xmm2");
asm volatile("pxor %xmm6,%xmm4");
asm volatile("movdqa %0,%%xmm6" : : "m" (dptr[z][d]));
}
asm volatile("pcmpgtb %xmm4,%xmm5");
asm volatile("paddb %xmm4,%xmm4");
asm volatile("pand %xmm0,%xmm5");
asm volatile("pxor %xmm5,%xmm4");
asm volatile("pxor %xmm5,%xmm5");
asm volatile("pxor %xmm6,%xmm2");
asm volatile("pxor %xmm6,%xmm4");
asm volatile("movntdq %%xmm2,%0" : "=m" (p[d]));
asm volatile("pxor %xmm2,%xmm2");
asm volatile("movntdq %%xmm4,%0" : "=m" (q[d]));
asm volatile("pxor %xmm4,%xmm4");
}
asm volatile("sfence" : : : "memory");
kernel_fpu_end();
}
const struct raid6_calls raid6_sse2x1 = {
raid6_sse21_gen_syndrome,
raid6_have_sse2,
"sse2x1",
1 /* Has cache hints */
};
/*
* Unrolled-by-2 SSE2 implementation
*/
static void raid6_sse22_gen_syndrome(int disks, size_t bytes, void **ptrs)
{
u8 **dptr = (u8 **)ptrs;
u8 *p, *q;
int d, z, z0;
z0 = disks - 3; /* Highest data disk */
p = dptr[z0+1]; /* XOR parity */
q = dptr[z0+2]; /* RS syndrome */
kernel_fpu_begin();
asm volatile("movdqa %0,%%xmm0" : : "m" (raid6_sse_constants.x1d[0]));
asm volatile("pxor %xmm5,%xmm5"); /* Zero temp */
asm volatile("pxor %xmm7,%xmm7"); /* Zero temp */
/* We uniformly assume a single prefetch covers at least 32 bytes */
for ( d = 0 ; d < bytes ; d += 32 ) {
asm volatile("prefetchnta %0" : : "m" (dptr[z0][d]));
asm volatile("movdqa %0,%%xmm2" : : "m" (dptr[z0][d])); /* P[0] */
asm volatile("movdqa %0,%%xmm3" : : "m" (dptr[z0][d+16])); /* P[1] */
asm volatile("movdqa %xmm2,%xmm4"); /* Q[0] */
asm volatile("movdqa %xmm3,%xmm6"); /* Q[1] */
for ( z = z0-1 ; z >= 0 ; z-- ) {
asm volatile("prefetchnta %0" : : "m" (dptr[z][d]));
asm volatile("pcmpgtb %xmm4,%xmm5");
asm volatile("pcmpgtb %xmm6,%xmm7");
asm volatile("paddb %xmm4,%xmm4");
asm volatile("paddb %xmm6,%xmm6");
asm volatile("pand %xmm0,%xmm5");
asm volatile("pand %xmm0,%xmm7");
asm volatile("pxor %xmm5,%xmm4");
asm volatile("pxor %xmm7,%xmm6");
asm volatile("movdqa %0,%%xmm5" : : "m" (dptr[z][d]));
asm volatile("movdqa %0,%%xmm7" : : "m" (dptr[z][d+16]));
asm volatile("pxor %xmm5,%xmm2");
asm volatile("pxor %xmm7,%xmm3");
asm volatile("pxor %xmm5,%xmm4");
asm volatile("pxor %xmm7,%xmm6");
asm volatile("pxor %xmm5,%xmm5");
asm volatile("pxor %xmm7,%xmm7");
}
asm volatile("movntdq %%xmm2,%0" : "=m" (p[d]));
asm volatile("movntdq %%xmm3,%0" : "=m" (p[d+16]));
asm volatile("movntdq %%xmm4,%0" : "=m" (q[d]));
asm volatile("movntdq %%xmm6,%0" : "=m" (q[d+16]));
}
asm volatile("sfence" : : : "memory");
kernel_fpu_end();
}
const struct raid6_calls raid6_sse2x2 = {
raid6_sse22_gen_syndrome,
raid6_have_sse2,
"sse2x2",
1 /* Has cache hints */
};
#endif
#if defined(__x86_64__) && !defined(__arch_um__)
/*
* Unrolled-by-4 SSE2 implementation
*/
static void raid6_sse24_gen_syndrome(int disks, size_t bytes, void **ptrs)
{
u8 **dptr = (u8 **)ptrs;
u8 *p, *q;
int d, z, z0;
z0 = disks - 3; /* Highest data disk */
p = dptr[z0+1]; /* XOR parity */
q = dptr[z0+2]; /* RS syndrome */
kernel_fpu_begin();
asm volatile("movdqa %0,%%xmm0" :: "m" (raid6_sse_constants.x1d[0]));
asm volatile("pxor %xmm2,%xmm2"); /* P[0] */
asm volatile("pxor %xmm3,%xmm3"); /* P[1] */
asm volatile("pxor %xmm4,%xmm4"); /* Q[0] */
asm volatile("pxor %xmm5,%xmm5"); /* Zero temp */
asm volatile("pxor %xmm6,%xmm6"); /* Q[1] */
asm volatile("pxor %xmm7,%xmm7"); /* Zero temp */
asm volatile("pxor %xmm10,%xmm10"); /* P[2] */
asm volatile("pxor %xmm11,%xmm11"); /* P[3] */
asm volatile("pxor %xmm12,%xmm12"); /* Q[2] */
asm volatile("pxor %xmm13,%xmm13"); /* Zero temp */
asm volatile("pxor %xmm14,%xmm14"); /* Q[3] */
asm volatile("pxor %xmm15,%xmm15"); /* Zero temp */
for ( d = 0 ; d < bytes ; d += 64 ) {
for ( z = z0 ; z >= 0 ; z-- ) {
/* The second prefetch seems to improve performance... */
asm volatile("prefetchnta %0" :: "m" (dptr[z][d]));
asm volatile("prefetchnta %0" :: "m" (dptr[z][d+32]));
asm volatile("pcmpgtb %xmm4,%xmm5");
asm volatile("pcmpgtb %xmm6,%xmm7");
asm volatile("pcmpgtb %xmm12,%xmm13");
asm volatile("pcmpgtb %xmm14,%xmm15");
asm volatile("paddb %xmm4,%xmm4");
asm volatile("paddb %xmm6,%xmm6");
asm volatile("paddb %xmm12,%xmm12");
asm volatile("paddb %xmm14,%xmm14");
asm volatile("pand %xmm0,%xmm5");
asm volatile("pand %xmm0,%xmm7");
asm volatile("pand %xmm0,%xmm13");
asm volatile("pand %xmm0,%xmm15");
asm volatile("pxor %xmm5,%xmm4");
asm volatile("pxor %xmm7,%xmm6");
asm volatile("pxor %xmm13,%xmm12");
asm volatile("pxor %xmm15,%xmm14");
asm volatile("movdqa %0,%%xmm5" :: "m" (dptr[z][d]));
asm volatile("movdqa %0,%%xmm7" :: "m" (dptr[z][d+16]));
asm volatile("movdqa %0,%%xmm13" :: "m" (dptr[z][d+32]));
asm volatile("movdqa %0,%%xmm15" :: "m" (dptr[z][d+48]));
asm volatile("pxor %xmm5,%xmm2");
asm volatile("pxor %xmm7,%xmm3");
asm volatile("pxor %xmm13,%xmm10");
asm volatile("pxor %xmm15,%xmm11");
asm volatile("pxor %xmm5,%xmm4");
asm volatile("pxor %xmm7,%xmm6");
asm volatile("pxor %xmm13,%xmm12");
asm volatile("pxor %xmm15,%xmm14");
asm volatile("pxor %xmm5,%xmm5");
asm volatile("pxor %xmm7,%xmm7");
asm volatile("pxor %xmm13,%xmm13");
asm volatile("pxor %xmm15,%xmm15");
}
asm volatile("movntdq %%xmm2,%0" : "=m" (p[d]));
asm volatile("pxor %xmm2,%xmm2");
asm volatile("movntdq %%xmm3,%0" : "=m" (p[d+16]));
asm volatile("pxor %xmm3,%xmm3");
asm volatile("movntdq %%xmm10,%0" : "=m" (p[d+32]));
asm volatile("pxor %xmm10,%xmm10");
asm volatile("movntdq %%xmm11,%0" : "=m" (p[d+48]));
asm volatile("pxor %xmm11,%xmm11");
asm volatile("movntdq %%xmm4,%0" : "=m" (q[d]));
asm volatile("pxor %xmm4,%xmm4");
asm volatile("movntdq %%xmm6,%0" : "=m" (q[d+16]));
asm volatile("pxor %xmm6,%xmm6");
asm volatile("movntdq %%xmm12,%0" : "=m" (q[d+32]));
asm volatile("pxor %xmm12,%xmm12");
asm volatile("movntdq %%xmm14,%0" : "=m" (q[d+48]));
asm volatile("pxor %xmm14,%xmm14");
}
asm volatile("sfence" : : : "memory");
kernel_fpu_end();
}
const struct raid6_calls raid6_sse2x4 = {
raid6_sse24_gen_syndrome,
raid6_have_sse2,
"sse2x4",
1 /* Has cache hints */
};
#endif
View Git Blame Copy Permalink