linux/lib/raid6/algos.c

269 lines
6.2 KiB
C
Raw Normal View 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.
*
* ----------------------------------------------------------------------- */
/*
* raid6/algos.c
*
* Algorithm list and algorithm selection for RAID-6
*/
#include <linux/raid/pq.h>
#ifndef __KERNEL__
#include <sys/mman.h>
#include <stdio.h>
#else
#include <linux/module.h>
#include <linux/gfp.h>
#if !RAID6_USE_EMPTY_ZERO_PAGE
/* In .bss so it's zeroed */
const char raid6_empty_zero_page[PAGE_SIZE] __attribute__((aligned(256)));
EXPORT_SYMBOL(raid6_empty_zero_page);
#endif
#endif
struct raid6_calls raid6_call;
EXPORT_SYMBOL_GPL(raid6_call);
const struct raid6_calls * const raid6_algos[] = {
#if defined(__ia64__)
&raid6_intx16,
&raid6_intx32,
#endif
#if defined(__i386__) && !defined(__arch_um__)
&raid6_mmxx1,
&raid6_mmxx2,
&raid6_sse1x1,
&raid6_sse1x2,
&raid6_sse2x1,
&raid6_sse2x2,
#ifdef CONFIG_AS_AVX2
&raid6_avx2x1,
&raid6_avx2x2,
#endif
#ifdef CONFIG_AS_AVX512
&raid6_avx512x1,
&raid6_avx512x2,
#endif
#endif
#if defined(__x86_64__) && !defined(__arch_um__)
&raid6_sse2x1,
&raid6_sse2x2,
&raid6_sse2x4,
#ifdef CONFIG_AS_AVX2
&raid6_avx2x1,
&raid6_avx2x2,
&raid6_avx2x4,
#endif
#ifdef CONFIG_AS_AVX512
&raid6_avx512x1,
&raid6_avx512x2,
&raid6_avx512x4,
#endif
#endif
#ifdef CONFIG_ALTIVEC
&raid6_altivec1,
&raid6_altivec2,
&raid6_altivec4,
&raid6_altivec8,
#endif
#if defined(CONFIG_TILEGX)
&raid6_tilegx8,
#endif
#if defined(CONFIG_S390)
&raid6_s390vx8,
#endif
&raid6_intx1,
&raid6_intx2,
&raid6_intx4,
&raid6_intx8,
#ifdef CONFIG_KERNEL_MODE_NEON
&raid6_neonx1,
&raid6_neonx2,
&raid6_neonx4,
&raid6_neonx8,
#endif
NULL
};
void (*raid6_2data_recov)(int, size_t, int, int, void **);
EXPORT_SYMBOL_GPL(raid6_2data_recov);
void (*raid6_datap_recov)(int, size_t, int, void **);
EXPORT_SYMBOL_GPL(raid6_datap_recov);
const struct raid6_recov_calls *const raid6_recov_algos[] = {
#ifdef CONFIG_AS_AVX512
&raid6_recov_avx512,
#endif
#ifdef CONFIG_AS_AVX2
&raid6_recov_avx2,
#endif
#ifdef CONFIG_AS_SSSE3
&raid6_recov_ssse3,
#endif
#ifdef CONFIG_S390
&raid6_recov_s390xc,
#endif
&raid6_recov_intx1,
NULL
};
#ifdef __KERNEL__
#define RAID6_TIME_JIFFIES_LG2 4
#else
/* Need more time to be stable in userspace */
#define RAID6_TIME_JIFFIES_LG2 9
#define time_before(x, y) ((x) < (y))
#endif
static inline const struct raid6_recov_calls *raid6_choose_recov(void)
{
const struct raid6_recov_calls *const *algo;
const struct raid6_recov_calls *best;
for (best = NULL, algo = raid6_recov_algos; *algo; algo++)
if (!best || (*algo)->priority > best->priority)
if (!(*algo)->valid || (*algo)->valid())
best = *algo;
if (best) {
raid6_2data_recov = best->data2;
raid6_datap_recov = best->datap;
pr_info("raid6: using %s recovery algorithm\n", best->name);
} else
pr_err("raid6: Yikes! No recovery algorithm found!\n");
return best;
}
static inline const struct raid6_calls *raid6_choose_gen(
void *(*const dptrs)[(65536/PAGE_SIZE)+2], const int disks)
{
md/raid6 algorithms: delta syndrome functions v3: s-o-b comment, explanation of performance and descision for the start/stop implementation Implementing rmw functionality for RAID6 requires optimized syndrome calculation. Up to now we can only generate a complete syndrome. The target P/Q pages are always overwritten. With this patch we provide a framework for inplace P/Q modification. In the first place simply fill those functions with NULL values. xor_syndrome() has two additional parameters: start & stop. These will indicate the first and last page that are changing during a rmw run. That makes it possible to avoid several unneccessary loops and speed up calculation. The caller needs to implement the following logic to make the functions work. 1) xor_syndrome(disks, start, stop, ...): "Remove" all data of source blocks inside P/Q between (and including) start and end. 2) modify any block with start <= block <= stop 3) xor_syndrome(disks, start, stop, ...): "Reinsert" all data of source blocks into P/Q between (and including) start and end. Pages between start and stop that won't be changed should be filled with a pointer to the kernel zero page. The reasons for not taking NULL pages are: 1) Algorithms cross the whole source data line by line. Thus avoid additional branches. 2) Having a NULL page avoids calculating the XOR P parity but still need calulation steps for the Q parity. Depending on the algorithm unrolling that might be only a difference of 2 instructions per loop. The benchmark numbers of the gen_syndrome() functions are displayed in the kernel log. Do the same for the xor_syndrome() functions. This will help to analyze performance problems and give an rough estimate how well the algorithm works. The choice of the fastest algorithm will still depend on the gen_syndrome() performance. With the start/stop page implementation the speed can vary a lot in real life. E.g. a change of page 0 & page 15 on a stripe will be harder to compute than the case where page 0 & page 1 are XOR candidates. To be not to enthusiatic about the expected speeds we will run a worse case test that simulates a change on the upper half of the stripe. So we do: 1) calculation of P/Q for the upper pages 2) continuation of Q for the lower (empty) pages Signed-off-by: Markus Stockhausen <stockhausen@collogia.de> Signed-off-by: NeilBrown <neilb@suse.de>
2014-12-15 02:57:04 +01:00
unsigned long perf, bestgenperf, bestxorperf, j0, j1;
int start = (disks>>1)-1, stop = disks-3; /* work on the second half of the disks */
const struct raid6_calls *const *algo;
const struct raid6_calls *best;
md/raid6 algorithms: delta syndrome functions v3: s-o-b comment, explanation of performance and descision for the start/stop implementation Implementing rmw functionality for RAID6 requires optimized syndrome calculation. Up to now we can only generate a complete syndrome. The target P/Q pages are always overwritten. With this patch we provide a framework for inplace P/Q modification. In the first place simply fill those functions with NULL values. xor_syndrome() has two additional parameters: start & stop. These will indicate the first and last page that are changing during a rmw run. That makes it possible to avoid several unneccessary loops and speed up calculation. The caller needs to implement the following logic to make the functions work. 1) xor_syndrome(disks, start, stop, ...): "Remove" all data of source blocks inside P/Q between (and including) start and end. 2) modify any block with start <= block <= stop 3) xor_syndrome(disks, start, stop, ...): "Reinsert" all data of source blocks into P/Q between (and including) start and end. Pages between start and stop that won't be changed should be filled with a pointer to the kernel zero page. The reasons for not taking NULL pages are: 1) Algorithms cross the whole source data line by line. Thus avoid additional branches. 2) Having a NULL page avoids calculating the XOR P parity but still need calulation steps for the Q parity. Depending on the algorithm unrolling that might be only a difference of 2 instructions per loop. The benchmark numbers of the gen_syndrome() functions are displayed in the kernel log. Do the same for the xor_syndrome() functions. This will help to analyze performance problems and give an rough estimate how well the algorithm works. The choice of the fastest algorithm will still depend on the gen_syndrome() performance. With the start/stop page implementation the speed can vary a lot in real life. E.g. a change of page 0 & page 15 on a stripe will be harder to compute than the case where page 0 & page 1 are XOR candidates. To be not to enthusiatic about the expected speeds we will run a worse case test that simulates a change on the upper half of the stripe. So we do: 1) calculation of P/Q for the upper pages 2) continuation of Q for the lower (empty) pages Signed-off-by: Markus Stockhausen <stockhausen@collogia.de> Signed-off-by: NeilBrown <neilb@suse.de>
2014-12-15 02:57:04 +01:00
for (bestgenperf = 0, bestxorperf = 0, best = NULL, algo = raid6_algos; *algo; algo++) {
if (!best || (*algo)->prefer >= best->prefer) {
if ((*algo)->valid && !(*algo)->valid())
continue;
perf = 0;
preempt_disable();
j0 = jiffies;
while ((j1 = jiffies) == j0)
cpu_relax();
while (time_before(jiffies,
j1 + (1<<RAID6_TIME_JIFFIES_LG2))) {
(*algo)->gen_syndrome(disks, PAGE_SIZE, *dptrs);
perf++;
}
preempt_enable();
md/raid6 algorithms: delta syndrome functions v3: s-o-b comment, explanation of performance and descision for the start/stop implementation Implementing rmw functionality for RAID6 requires optimized syndrome calculation. Up to now we can only generate a complete syndrome. The target P/Q pages are always overwritten. With this patch we provide a framework for inplace P/Q modification. In the first place simply fill those functions with NULL values. xor_syndrome() has two additional parameters: start & stop. These will indicate the first and last page that are changing during a rmw run. That makes it possible to avoid several unneccessary loops and speed up calculation. The caller needs to implement the following logic to make the functions work. 1) xor_syndrome(disks, start, stop, ...): "Remove" all data of source blocks inside P/Q between (and including) start and end. 2) modify any block with start <= block <= stop 3) xor_syndrome(disks, start, stop, ...): "Reinsert" all data of source blocks into P/Q between (and including) start and end. Pages between start and stop that won't be changed should be filled with a pointer to the kernel zero page. The reasons for not taking NULL pages are: 1) Algorithms cross the whole source data line by line. Thus avoid additional branches. 2) Having a NULL page avoids calculating the XOR P parity but still need calulation steps for the Q parity. Depending on the algorithm unrolling that might be only a difference of 2 instructions per loop. The benchmark numbers of the gen_syndrome() functions are displayed in the kernel log. Do the same for the xor_syndrome() functions. This will help to analyze performance problems and give an rough estimate how well the algorithm works. The choice of the fastest algorithm will still depend on the gen_syndrome() performance. With the start/stop page implementation the speed can vary a lot in real life. E.g. a change of page 0 & page 15 on a stripe will be harder to compute than the case where page 0 & page 1 are XOR candidates. To be not to enthusiatic about the expected speeds we will run a worse case test that simulates a change on the upper half of the stripe. So we do: 1) calculation of P/Q for the upper pages 2) continuation of Q for the lower (empty) pages Signed-off-by: Markus Stockhausen <stockhausen@collogia.de> Signed-off-by: NeilBrown <neilb@suse.de>
2014-12-15 02:57:04 +01:00
if (perf > bestgenperf) {
bestgenperf = perf;
best = *algo;
}
md/raid6 algorithms: delta syndrome functions v3: s-o-b comment, explanation of performance and descision for the start/stop implementation Implementing rmw functionality for RAID6 requires optimized syndrome calculation. Up to now we can only generate a complete syndrome. The target P/Q pages are always overwritten. With this patch we provide a framework for inplace P/Q modification. In the first place simply fill those functions with NULL values. xor_syndrome() has two additional parameters: start & stop. These will indicate the first and last page that are changing during a rmw run. That makes it possible to avoid several unneccessary loops and speed up calculation. The caller needs to implement the following logic to make the functions work. 1) xor_syndrome(disks, start, stop, ...): "Remove" all data of source blocks inside P/Q between (and including) start and end. 2) modify any block with start <= block <= stop 3) xor_syndrome(disks, start, stop, ...): "Reinsert" all data of source blocks into P/Q between (and including) start and end. Pages between start and stop that won't be changed should be filled with a pointer to the kernel zero page. The reasons for not taking NULL pages are: 1) Algorithms cross the whole source data line by line. Thus avoid additional branches. 2) Having a NULL page avoids calculating the XOR P parity but still need calulation steps for the Q parity. Depending on the algorithm unrolling that might be only a difference of 2 instructions per loop. The benchmark numbers of the gen_syndrome() functions are displayed in the kernel log. Do the same for the xor_syndrome() functions. This will help to analyze performance problems and give an rough estimate how well the algorithm works. The choice of the fastest algorithm will still depend on the gen_syndrome() performance. With the start/stop page implementation the speed can vary a lot in real life. E.g. a change of page 0 & page 15 on a stripe will be harder to compute than the case where page 0 & page 1 are XOR candidates. To be not to enthusiatic about the expected speeds we will run a worse case test that simulates a change on the upper half of the stripe. So we do: 1) calculation of P/Q for the upper pages 2) continuation of Q for the lower (empty) pages Signed-off-by: Markus Stockhausen <stockhausen@collogia.de> Signed-off-by: NeilBrown <neilb@suse.de>
2014-12-15 02:57:04 +01:00
pr_info("raid6: %-8s gen() %5ld MB/s\n", (*algo)->name,
(perf*HZ) >> (20-16+RAID6_TIME_JIFFIES_LG2));
md/raid6 algorithms: delta syndrome functions v3: s-o-b comment, explanation of performance and descision for the start/stop implementation Implementing rmw functionality for RAID6 requires optimized syndrome calculation. Up to now we can only generate a complete syndrome. The target P/Q pages are always overwritten. With this patch we provide a framework for inplace P/Q modification. In the first place simply fill those functions with NULL values. xor_syndrome() has two additional parameters: start & stop. These will indicate the first and last page that are changing during a rmw run. That makes it possible to avoid several unneccessary loops and speed up calculation. The caller needs to implement the following logic to make the functions work. 1) xor_syndrome(disks, start, stop, ...): "Remove" all data of source blocks inside P/Q between (and including) start and end. 2) modify any block with start <= block <= stop 3) xor_syndrome(disks, start, stop, ...): "Reinsert" all data of source blocks into P/Q between (and including) start and end. Pages between start and stop that won't be changed should be filled with a pointer to the kernel zero page. The reasons for not taking NULL pages are: 1) Algorithms cross the whole source data line by line. Thus avoid additional branches. 2) Having a NULL page avoids calculating the XOR P parity but still need calulation steps for the Q parity. Depending on the algorithm unrolling that might be only a difference of 2 instructions per loop. The benchmark numbers of the gen_syndrome() functions are displayed in the kernel log. Do the same for the xor_syndrome() functions. This will help to analyze performance problems and give an rough estimate how well the algorithm works. The choice of the fastest algorithm will still depend on the gen_syndrome() performance. With the start/stop page implementation the speed can vary a lot in real life. E.g. a change of page 0 & page 15 on a stripe will be harder to compute than the case where page 0 & page 1 are XOR candidates. To be not to enthusiatic about the expected speeds we will run a worse case test that simulates a change on the upper half of the stripe. So we do: 1) calculation of P/Q for the upper pages 2) continuation of Q for the lower (empty) pages Signed-off-by: Markus Stockhausen <stockhausen@collogia.de> Signed-off-by: NeilBrown <neilb@suse.de>
2014-12-15 02:57:04 +01:00
if (!(*algo)->xor_syndrome)
continue;
perf = 0;
preempt_disable();
j0 = jiffies;
while ((j1 = jiffies) == j0)
cpu_relax();
while (time_before(jiffies,
j1 + (1<<RAID6_TIME_JIFFIES_LG2))) {
(*algo)->xor_syndrome(disks, start, stop,
PAGE_SIZE, *dptrs);
perf++;
}
preempt_enable();
if (best == *algo)
bestxorperf = perf;
pr_info("raid6: %-8s xor() %5ld MB/s\n", (*algo)->name,
(perf*HZ) >> (20-16+RAID6_TIME_JIFFIES_LG2+1));
}
}
if (best) {
md/raid6 algorithms: delta syndrome functions v3: s-o-b comment, explanation of performance and descision for the start/stop implementation Implementing rmw functionality for RAID6 requires optimized syndrome calculation. Up to now we can only generate a complete syndrome. The target P/Q pages are always overwritten. With this patch we provide a framework for inplace P/Q modification. In the first place simply fill those functions with NULL values. xor_syndrome() has two additional parameters: start & stop. These will indicate the first and last page that are changing during a rmw run. That makes it possible to avoid several unneccessary loops and speed up calculation. The caller needs to implement the following logic to make the functions work. 1) xor_syndrome(disks, start, stop, ...): "Remove" all data of source blocks inside P/Q between (and including) start and end. 2) modify any block with start <= block <= stop 3) xor_syndrome(disks, start, stop, ...): "Reinsert" all data of source blocks into P/Q between (and including) start and end. Pages between start and stop that won't be changed should be filled with a pointer to the kernel zero page. The reasons for not taking NULL pages are: 1) Algorithms cross the whole source data line by line. Thus avoid additional branches. 2) Having a NULL page avoids calculating the XOR P parity but still need calulation steps for the Q parity. Depending on the algorithm unrolling that might be only a difference of 2 instructions per loop. The benchmark numbers of the gen_syndrome() functions are displayed in the kernel log. Do the same for the xor_syndrome() functions. This will help to analyze performance problems and give an rough estimate how well the algorithm works. The choice of the fastest algorithm will still depend on the gen_syndrome() performance. With the start/stop page implementation the speed can vary a lot in real life. E.g. a change of page 0 & page 15 on a stripe will be harder to compute than the case where page 0 & page 1 are XOR candidates. To be not to enthusiatic about the expected speeds we will run a worse case test that simulates a change on the upper half of the stripe. So we do: 1) calculation of P/Q for the upper pages 2) continuation of Q for the lower (empty) pages Signed-off-by: Markus Stockhausen <stockhausen@collogia.de> Signed-off-by: NeilBrown <neilb@suse.de>
2014-12-15 02:57:04 +01:00
pr_info("raid6: using algorithm %s gen() %ld MB/s\n",
best->name,
md/raid6 algorithms: delta syndrome functions v3: s-o-b comment, explanation of performance and descision for the start/stop implementation Implementing rmw functionality for RAID6 requires optimized syndrome calculation. Up to now we can only generate a complete syndrome. The target P/Q pages are always overwritten. With this patch we provide a framework for inplace P/Q modification. In the first place simply fill those functions with NULL values. xor_syndrome() has two additional parameters: start & stop. These will indicate the first and last page that are changing during a rmw run. That makes it possible to avoid several unneccessary loops and speed up calculation. The caller needs to implement the following logic to make the functions work. 1) xor_syndrome(disks, start, stop, ...): "Remove" all data of source blocks inside P/Q between (and including) start and end. 2) modify any block with start <= block <= stop 3) xor_syndrome(disks, start, stop, ...): "Reinsert" all data of source blocks into P/Q between (and including) start and end. Pages between start and stop that won't be changed should be filled with a pointer to the kernel zero page. The reasons for not taking NULL pages are: 1) Algorithms cross the whole source data line by line. Thus avoid additional branches. 2) Having a NULL page avoids calculating the XOR P parity but still need calulation steps for the Q parity. Depending on the algorithm unrolling that might be only a difference of 2 instructions per loop. The benchmark numbers of the gen_syndrome() functions are displayed in the kernel log. Do the same for the xor_syndrome() functions. This will help to analyze performance problems and give an rough estimate how well the algorithm works. The choice of the fastest algorithm will still depend on the gen_syndrome() performance. With the start/stop page implementation the speed can vary a lot in real life. E.g. a change of page 0 & page 15 on a stripe will be harder to compute than the case where page 0 & page 1 are XOR candidates. To be not to enthusiatic about the expected speeds we will run a worse case test that simulates a change on the upper half of the stripe. So we do: 1) calculation of P/Q for the upper pages 2) continuation of Q for the lower (empty) pages Signed-off-by: Markus Stockhausen <stockhausen@collogia.de> Signed-off-by: NeilBrown <neilb@suse.de>
2014-12-15 02:57:04 +01:00
(bestgenperf*HZ) >> (20-16+RAID6_TIME_JIFFIES_LG2));
if (best->xor_syndrome)
pr_info("raid6: .... xor() %ld MB/s, rmw enabled\n",
(bestxorperf*HZ) >> (20-16+RAID6_TIME_JIFFIES_LG2+1));
raid6_call = *best;
} else
pr_err("raid6: Yikes! No algorithm found!\n");
return best;
}
/* Try to pick the best algorithm */
/* This code uses the gfmul table as convenient data set to abuse */
int __init raid6_select_algo(void)
{
const int disks = (65536/PAGE_SIZE)+2;
const struct raid6_calls *gen_best;
const struct raid6_recov_calls *rec_best;
char *syndromes;
void *dptrs[(65536/PAGE_SIZE)+2];
int i;
for (i = 0; i < disks-2; i++)
dptrs[i] = ((char *)raid6_gfmul) + PAGE_SIZE*i;
/* Normal code - use a 2-page allocation to avoid D$ conflict */
syndromes = (void *) __get_free_pages(GFP_KERNEL, 1);
if (!syndromes) {
pr_err("raid6: Yikes! No memory available.\n");
return -ENOMEM;
}
dptrs[disks-2] = syndromes;
dptrs[disks-1] = syndromes + PAGE_SIZE;
/* select raid gen_syndrome function */
gen_best = raid6_choose_gen(&dptrs, disks);
/* select raid recover functions */
rec_best = raid6_choose_recov();
free_pages((unsigned long)syndromes, 1);
return gen_best && rec_best ? 0 : -EINVAL;
}
static void raid6_exit(void)
{
do { } while (0);
}
subsys_initcall(raid6_select_algo);
module_exit(raid6_exit);
MODULE_LICENSE("GPL");
MODULE_DESCRIPTION("RAID6 Q-syndrome calculations");