linux-hardened/lib/raid6/tilegx.uc
Ken Steele ae77cbc1e7 RAID: add tilegx SIMD implementation of raid6
This change adds TILE-Gx SIMD instructions to the software raid
(md), modeling the Altivec implementation. This is only for Syndrome
generation; there is more that could be done to improve recovery,
as in the recent Intel SSE3 recovery implementation.

The code unrolls 8 times; this turns out to be the best on tilegx
hardware among the set 1, 2, 4, 8 or 16.  The code reads one
cache-line of data from each disk, stores P and Q then goes to the
next cache-line.

The test code in sys/linux/lib/raid6/test reports 2008 MB/s data
read rate for syndrome generation using 18 disks (16 data and 2
parity). It was 1512 MB/s before this SIMD optimizations. This is
running on 1 core with all the data in cache.

This is based on the paper The Mathematics of RAID-6.
(http://kernel.org/pub/linux/kernel/people/hpa/raid6.pdf).

Signed-off-by: Ken Steele <ken@tilera.com>
Signed-off-by: Chris Metcalf <cmetcalf@tilera.com>
Signed-off-by: NeilBrown <neilb@suse.de>
2013-08-27 16:05:50 +10:00

86 lines
2.1 KiB
Ucode

/* -*- linux-c -*- ------------------------------------------------------- *
*
* Copyright 2002 H. Peter Anvin - All Rights Reserved
* Copyright 2012 Tilera Corporation - 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.
*
* ----------------------------------------------------------------------- */
/*
* tilegx$#.c
*
* $#-way unrolled TILE-Gx SIMD for RAID-6 math.
*
* This file is postprocessed using unroll.awk.
*
*/
#include <linux/raid/pq.h>
/* Create 8 byte copies of constant byte */
# define NBYTES(x) (__insn_v1addi(0, x))
# define NSIZE 8
/*
* The SHLBYTE() operation shifts each byte left by 1, *not*
* rolling over into the next byte
*/
static inline __attribute_const__ u64 SHLBYTE(u64 v)
{
/* Vector One Byte Shift Left Immediate. */
return __insn_v1shli(v, 1);
}
/*
* The MASK() operation returns 0xFF in any byte for which the high
* bit is 1, 0x00 for any byte for which the high bit is 0.
*/
static inline __attribute_const__ u64 MASK(u64 v)
{
/* Vector One Byte Shift Right Signed Immediate. */
return __insn_v1shrsi(v, 7);
}
void raid6_tilegx$#_gen_syndrome(int disks, size_t bytes, void **ptrs)
{
u8 **dptr = (u8 **)ptrs;
u64 *p, *q;
int d, z, z0;
u64 wd$$, wq$$, wp$$, w1$$, w2$$;
u64 x1d = NBYTES(0x1d);
u64 * z0ptr;
z0 = disks - 3; /* Highest data disk */
p = (u64 *)dptr[z0+1]; /* XOR parity */
q = (u64 *)dptr[z0+2]; /* RS syndrome */
z0ptr = (u64 *)&dptr[z0][0];
for ( d = 0 ; d < bytes ; d += NSIZE*$# ) {
wq$$ = wp$$ = *z0ptr++;
for ( z = z0-1 ; z >= 0 ; z-- ) {
wd$$ = *(u64 *)&dptr[z][d+$$*NSIZE];
wp$$ = wp$$ ^ wd$$;
w2$$ = MASK(wq$$);
w1$$ = SHLBYTE(wq$$);
w2$$ = w2$$ & x1d;
w1$$ = w1$$ ^ w2$$;
wq$$ = w1$$ ^ wd$$;
}
*p++ = wp$$;
*q++ = wq$$;
}
}
const struct raid6_calls raid6_tilegx$# = {
raid6_tilegx$#_gen_syndrome,
NULL,
"tilegx$#",
0
};