mirror of
https://github.com/torvalds/linux.git
synced 2026-07-27 09:36:22 +02:00
30bf04bd13
Drop the pointless mention of the file name, and use standard formatting for the top of file comments. Link: https://lore.kernel.org/20260518051804.462141-14-hch@lst.de Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Ard Biesheuvel <ardb@kernel.org> Tested-by: Ard Biesheuvel <ardb@kernel.org> # kunit only on arm64 Cc: Albert Ou <aou@eecs.berkeley.edu> Cc: Alexander Gordeev <agordeev@linux.ibm.com> Cc: Alexandre Ghiti <alex@ghiti.fr> Cc: Arnd Bergmann <arnd@arndb.de> Cc: "Borislav Petkov (AMD)" <bp@alien8.de> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Chris Mason <clm@fb.com> Cc: Christian Borntraeger <borntraeger@linux.ibm.com> Cc: Dan Williams <dan.j.williams@intel.com> Cc: David Sterba <dsterba@suse.com> Cc: Heiko Carstens <hca@linux.ibm.com> Cc: Herbert Xu <herbert@gondor.apana.org.au> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Huacai Chen <chenhuacai@kernel.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Li Nan <linan122@huawei.com> Cc: Madhavan Srinivasan <maddy@linux.ibm.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Nicholas Piggin <npiggin@gmail.com> Cc: Palmer Dabbelt <palmer@dabbelt.com> Cc: Song Liu <song@kernel.org> Cc: Sven Schnelle <svens@linux.ibm.com> Cc: Vasily Gorbik <gor@linux.ibm.com> Cc: WANG Xuerui <kernel@xen0n.name> Cc: Will Deacon <will@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
119 lines
3.2 KiB
C
119 lines
3.2 KiB
C
// SPDX-License-Identifier: GPL-2.0-or-later
|
|
/*
|
|
* Copyright 2002 H. Peter Anvin - All Rights Reserved
|
|
*
|
|
* MMX implementation of RAID-6 syndrome functions.
|
|
*/
|
|
|
|
#include <asm/cpufeature.h>
|
|
#include <asm/fpu/api.h>
|
|
#include "algos.h"
|
|
|
|
/* Shared with raid6/sse1.c */
|
|
const struct raid6_mmx_constants {
|
|
u64 x1d;
|
|
} raid6_mmx_constants = {
|
|
0x1d1d1d1d1d1d1d1dULL,
|
|
};
|
|
|
|
/*
|
|
* Plain MMX implementation
|
|
*/
|
|
static void raid6_mmx1_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("movq %0,%%mm0" : : "m" (raid6_mmx_constants.x1d));
|
|
asm volatile("pxor %mm5,%mm5"); /* Zero temp */
|
|
|
|
for ( d = 0 ; d < bytes ; d += 8 ) {
|
|
asm volatile("movq %0,%%mm2" : : "m" (dptr[z0][d])); /* P[0] */
|
|
asm volatile("movq %mm2,%mm4"); /* Q[0] */
|
|
for ( z = z0-1 ; z >= 0 ; z-- ) {
|
|
asm volatile("movq %0,%%mm6" : : "m" (dptr[z][d]));
|
|
asm volatile("pcmpgtb %mm4,%mm5");
|
|
asm volatile("paddb %mm4,%mm4");
|
|
asm volatile("pand %mm0,%mm5");
|
|
asm volatile("pxor %mm5,%mm4");
|
|
asm volatile("pxor %mm5,%mm5");
|
|
asm volatile("pxor %mm6,%mm2");
|
|
asm volatile("pxor %mm6,%mm4");
|
|
}
|
|
asm volatile("movq %%mm2,%0" : "=m" (p[d]));
|
|
asm volatile("pxor %mm2,%mm2");
|
|
asm volatile("movq %%mm4,%0" : "=m" (q[d]));
|
|
asm volatile("pxor %mm4,%mm4");
|
|
}
|
|
|
|
kernel_fpu_end();
|
|
}
|
|
|
|
const struct raid6_calls raid6_mmxx1 = {
|
|
.gen_syndrome = raid6_mmx1_gen_syndrome,
|
|
.name = "mmxx1",
|
|
};
|
|
|
|
/*
|
|
* Unrolled-by-2 MMX implementation
|
|
*/
|
|
static void raid6_mmx2_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("movq %0,%%mm0" : : "m" (raid6_mmx_constants.x1d));
|
|
asm volatile("pxor %mm5,%mm5"); /* Zero temp */
|
|
asm volatile("pxor %mm7,%mm7"); /* Zero temp */
|
|
|
|
for ( d = 0 ; d < bytes ; d += 16 ) {
|
|
asm volatile("movq %0,%%mm2" : : "m" (dptr[z0][d])); /* P[0] */
|
|
asm volatile("movq %0,%%mm3" : : "m" (dptr[z0][d+8]));
|
|
asm volatile("movq %mm2,%mm4"); /* Q[0] */
|
|
asm volatile("movq %mm3,%mm6"); /* Q[1] */
|
|
for ( z = z0-1 ; z >= 0 ; z-- ) {
|
|
asm volatile("pcmpgtb %mm4,%mm5");
|
|
asm volatile("pcmpgtb %mm6,%mm7");
|
|
asm volatile("paddb %mm4,%mm4");
|
|
asm volatile("paddb %mm6,%mm6");
|
|
asm volatile("pand %mm0,%mm5");
|
|
asm volatile("pand %mm0,%mm7");
|
|
asm volatile("pxor %mm5,%mm4");
|
|
asm volatile("pxor %mm7,%mm6");
|
|
asm volatile("movq %0,%%mm5" : : "m" (dptr[z][d]));
|
|
asm volatile("movq %0,%%mm7" : : "m" (dptr[z][d+8]));
|
|
asm volatile("pxor %mm5,%mm2");
|
|
asm volatile("pxor %mm7,%mm3");
|
|
asm volatile("pxor %mm5,%mm4");
|
|
asm volatile("pxor %mm7,%mm6");
|
|
asm volatile("pxor %mm5,%mm5");
|
|
asm volatile("pxor %mm7,%mm7");
|
|
}
|
|
asm volatile("movq %%mm2,%0" : "=m" (p[d]));
|
|
asm volatile("movq %%mm3,%0" : "=m" (p[d+8]));
|
|
asm volatile("movq %%mm4,%0" : "=m" (q[d]));
|
|
asm volatile("movq %%mm6,%0" : "=m" (q[d+8]));
|
|
}
|
|
|
|
kernel_fpu_end();
|
|
}
|
|
|
|
const struct raid6_calls raid6_mmxx2 = {
|
|
.gen_syndrome = raid6_mmx2_gen_syndrome,
|
|
.name = "mmxx2",
|
|
};
|