async_tx: add support for asynchronous RAID6 recovery operations

async_raid6_2data_recov() recovers two data disk failures

 async_raid6_datap_recov() recovers a data disk and the P disk

These routines are a port of the synchronous versions found in
drivers/md/raid6recov.c.  The primary difference is breaking out the xor
operations into separate calls to async_xor.  Two helper routines are
introduced to perform scalar multiplication where needed.
async_sum_product() multiplies two sources by scalar coefficients and
then sums (xor) the result.  async_mult() simply multiplies a single
source by a scalar.

This implemention also includes, in contrast to the original
synchronous-only code, special case handling for the 4-disk and 5-disk
array cases.  In these situations the default N-disk algorithm will
present 0-source or 1-source operations to dma devices.  To cover for
dma devices where the minimum source count is 2 we implement 4-disk and
5-disk handling in the recovery code.

[ Impact: asynchronous raid6 recovery routines for 2data and datap cases ]

Cc: Yuri Tikhonov <yur@emcraft.com>
Cc: Ilya Yanok <yanok@emcraft.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: David Woodhouse <David.Woodhouse@intel.com>
Reviewed-by: Andre Noll <maan@systemlinux.org>
Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
This commit is contained in:
Dan Williams 2009-07-14 12:20:37 -07:00
parent b2f46fd8ef
commit 0a82a6239b
5 changed files with 466 additions and 0 deletions

View file

@ -67,6 +67,10 @@ xor_val - xor a series of source buffers and set a flag if the
pq - generate the p+q (raid6 syndrome) from a series of source buffers pq - generate the p+q (raid6 syndrome) from a series of source buffers
pq_val - validate that a p and or q buffer are in sync with a given series of pq_val - validate that a p and or q buffer are in sync with a given series of
sources sources
datap - (raid6_datap_recov) recover a raid6 data block and the p block
from the given sources
2data - (raid6_2data_recov) recover 2 raid6 data blocks from the given
sources
3.3 Descriptor management: 3.3 Descriptor management:
The return value is non-NULL and points to a 'descriptor' when the operation The return value is non-NULL and points to a 'descriptor' when the operation

View file

@ -18,3 +18,8 @@ config ASYNC_PQ
tristate tristate
select ASYNC_CORE select ASYNC_CORE
config ASYNC_RAID6_RECOV
tristate
select ASYNC_CORE
select ASYNC_PQ

View file

@ -3,3 +3,4 @@ obj-$(CONFIG_ASYNC_MEMCPY) += async_memcpy.o
obj-$(CONFIG_ASYNC_MEMSET) += async_memset.o obj-$(CONFIG_ASYNC_MEMSET) += async_memset.o
obj-$(CONFIG_ASYNC_XOR) += async_xor.o obj-$(CONFIG_ASYNC_XOR) += async_xor.o
obj-$(CONFIG_ASYNC_PQ) += async_pq.o obj-$(CONFIG_ASYNC_PQ) += async_pq.o
obj-$(CONFIG_ASYNC_RAID6_RECOV) += async_raid6_recov.o

View file

@ -0,0 +1,448 @@
/*
* Asynchronous RAID-6 recovery calculations ASYNC_TX API.
* Copyright(c) 2009 Intel Corporation
*
* based on raid6recov.c:
* Copyright 2002 H. Peter Anvin
*
* 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; either version 2 of the License, or (at your option)
* any later version.
*
* This program is distributed in the hope that it will be useful, but WITHOUT
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
* more details.
*
* You should have received a copy of the GNU General Public License along with
* this program; if not, write to the Free Software Foundation, Inc., 51
* Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
*
*/
#include <linux/kernel.h>
#include <linux/interrupt.h>
#include <linux/dma-mapping.h>
#include <linux/raid/pq.h>
#include <linux/async_tx.h>
static struct dma_async_tx_descriptor *
async_sum_product(struct page *dest, struct page **srcs, unsigned char *coef,
size_t len, struct async_submit_ctl *submit)
{
struct dma_chan *chan = async_tx_find_channel(submit, DMA_PQ,
&dest, 1, srcs, 2, len);
struct dma_device *dma = chan ? chan->device : NULL;
const u8 *amul, *bmul;
u8 ax, bx;
u8 *a, *b, *c;
if (dma) {
dma_addr_t dma_dest[2];
dma_addr_t dma_src[2];
struct device *dev = dma->dev;
struct dma_async_tx_descriptor *tx;
enum dma_ctrl_flags dma_flags = DMA_PREP_PQ_DISABLE_P;
dma_dest[1] = dma_map_page(dev, dest, 0, len, DMA_BIDIRECTIONAL);
dma_src[0] = dma_map_page(dev, srcs[0], 0, len, DMA_TO_DEVICE);
dma_src[1] = dma_map_page(dev, srcs[1], 0, len, DMA_TO_DEVICE);
tx = dma->device_prep_dma_pq(chan, dma_dest, dma_src, 2, coef,
len, dma_flags);
if (tx) {
async_tx_submit(chan, tx, submit);
return tx;
}
}
/* run the operation synchronously */
async_tx_quiesce(&submit->depend_tx);
amul = raid6_gfmul[coef[0]];
bmul = raid6_gfmul[coef[1]];
a = page_address(srcs[0]);
b = page_address(srcs[1]);
c = page_address(dest);
while (len--) {
ax = amul[*a++];
bx = bmul[*b++];
*c++ = ax ^ bx;
}
return NULL;
}
static struct dma_async_tx_descriptor *
async_mult(struct page *dest, struct page *src, u8 coef, size_t len,
struct async_submit_ctl *submit)
{
struct dma_chan *chan = async_tx_find_channel(submit, DMA_PQ,
&dest, 1, &src, 1, len);
struct dma_device *dma = chan ? chan->device : NULL;
const u8 *qmul; /* Q multiplier table */
u8 *d, *s;
if (dma) {
dma_addr_t dma_dest[2];
dma_addr_t dma_src[1];
struct device *dev = dma->dev;
struct dma_async_tx_descriptor *tx;
enum dma_ctrl_flags dma_flags = DMA_PREP_PQ_DISABLE_P;
dma_dest[1] = dma_map_page(dev, dest, 0, len, DMA_BIDIRECTIONAL);
dma_src[0] = dma_map_page(dev, src, 0, len, DMA_TO_DEVICE);
tx = dma->device_prep_dma_pq(chan, dma_dest, dma_src, 1, &coef,
len, dma_flags);
if (tx) {
async_tx_submit(chan, tx, submit);
return tx;
}
}
/* no channel available, or failed to allocate a descriptor, so
* perform the operation synchronously
*/
async_tx_quiesce(&submit->depend_tx);
qmul = raid6_gfmul[coef];
d = page_address(dest);
s = page_address(src);
while (len--)
*d++ = qmul[*s++];
return NULL;
}
static struct dma_async_tx_descriptor *
__2data_recov_4(size_t bytes, int faila, int failb, struct page **blocks,
struct async_submit_ctl *submit)
{
struct dma_async_tx_descriptor *tx = NULL;
struct page *p, *q, *a, *b;
struct page *srcs[2];
unsigned char coef[2];
enum async_tx_flags flags = submit->flags;
dma_async_tx_callback cb_fn = submit->cb_fn;
void *cb_param = submit->cb_param;
void *scribble = submit->scribble;
p = blocks[4-2];
q = blocks[4-1];
a = blocks[faila];
b = blocks[failb];
/* in the 4 disk case P + Pxy == P and Q + Qxy == Q */
/* Dx = A*(P+Pxy) + B*(Q+Qxy) */
srcs[0] = p;
srcs[1] = q;
coef[0] = raid6_gfexi[failb-faila];
coef[1] = raid6_gfinv[raid6_gfexp[faila]^raid6_gfexp[failb]];
init_async_submit(submit, 0, tx, NULL, NULL, scribble);
tx = async_sum_product(b, srcs, coef, bytes, submit);
/* Dy = P+Pxy+Dx */
srcs[0] = p;
srcs[1] = b;
init_async_submit(submit, flags | ASYNC_TX_XOR_ZERO_DST, tx, cb_fn,
cb_param, scribble);
tx = async_xor(a, srcs, 0, 2, bytes, submit);
return tx;
}
static struct dma_async_tx_descriptor *
__2data_recov_5(size_t bytes, int faila, int failb, struct page **blocks,
struct async_submit_ctl *submit)
{
struct dma_async_tx_descriptor *tx = NULL;
struct page *p, *q, *g, *dp, *dq;
struct page *srcs[2];
unsigned char coef[2];
enum async_tx_flags flags = submit->flags;
dma_async_tx_callback cb_fn = submit->cb_fn;
void *cb_param = submit->cb_param;
void *scribble = submit->scribble;
int uninitialized_var(good);
int i;
for (i = 0; i < 3; i++) {
if (i == faila || i == failb)
continue;
else {
good = i;
break;
}
}
BUG_ON(i >= 3);
p = blocks[5-2];
q = blocks[5-1];
g = blocks[good];
/* Compute syndrome with zero for the missing data pages
* Use the dead data pages as temporary storage for delta p and
* delta q
*/
dp = blocks[faila];
dq = blocks[failb];
init_async_submit(submit, 0, tx, NULL, NULL, scribble);
tx = async_memcpy(dp, g, 0, 0, bytes, submit);
init_async_submit(submit, 0, tx, NULL, NULL, scribble);
tx = async_mult(dq, g, raid6_gfexp[good], bytes, submit);
/* compute P + Pxy */
srcs[0] = dp;
srcs[1] = p;
init_async_submit(submit, ASYNC_TX_XOR_DROP_DST, tx, NULL, NULL,
scribble);
tx = async_xor(dp, srcs, 0, 2, bytes, submit);
/* compute Q + Qxy */
srcs[0] = dq;
srcs[1] = q;
init_async_submit(submit, ASYNC_TX_XOR_DROP_DST, tx, NULL, NULL,
scribble);
tx = async_xor(dq, srcs, 0, 2, bytes, submit);
/* Dx = A*(P+Pxy) + B*(Q+Qxy) */
srcs[0] = dp;
srcs[1] = dq;
coef[0] = raid6_gfexi[failb-faila];
coef[1] = raid6_gfinv[raid6_gfexp[faila]^raid6_gfexp[failb]];
init_async_submit(submit, 0, tx, NULL, NULL, scribble);
tx = async_sum_product(dq, srcs, coef, bytes, submit);
/* Dy = P+Pxy+Dx */
srcs[0] = dp;
srcs[1] = dq;
init_async_submit(submit, flags | ASYNC_TX_XOR_DROP_DST, tx, cb_fn,
cb_param, scribble);
tx = async_xor(dp, srcs, 0, 2, bytes, submit);
return tx;
}
static struct dma_async_tx_descriptor *
__2data_recov_n(int disks, size_t bytes, int faila, int failb,
struct page **blocks, struct async_submit_ctl *submit)
{
struct dma_async_tx_descriptor *tx = NULL;
struct page *p, *q, *dp, *dq;
struct page *srcs[2];
unsigned char coef[2];
enum async_tx_flags flags = submit->flags;
dma_async_tx_callback cb_fn = submit->cb_fn;
void *cb_param = submit->cb_param;
void *scribble = submit->scribble;
p = blocks[disks-2];
q = blocks[disks-1];
/* Compute syndrome with zero for the missing data pages
* Use the dead data pages as temporary storage for
* delta p and delta q
*/
dp = blocks[faila];
blocks[faila] = (void *)raid6_empty_zero_page;
blocks[disks-2] = dp;
dq = blocks[failb];
blocks[failb] = (void *)raid6_empty_zero_page;
blocks[disks-1] = dq;
init_async_submit(submit, 0, tx, NULL, NULL, scribble);
tx = async_gen_syndrome(blocks, 0, disks, bytes, submit);
/* Restore pointer table */
blocks[faila] = dp;
blocks[failb] = dq;
blocks[disks-2] = p;
blocks[disks-1] = q;
/* compute P + Pxy */
srcs[0] = dp;
srcs[1] = p;
init_async_submit(submit, ASYNC_TX_XOR_DROP_DST, tx, NULL, NULL,
scribble);
tx = async_xor(dp, srcs, 0, 2, bytes, submit);
/* compute Q + Qxy */
srcs[0] = dq;
srcs[1] = q;
init_async_submit(submit, ASYNC_TX_XOR_DROP_DST, tx, NULL, NULL,
scribble);
tx = async_xor(dq, srcs, 0, 2, bytes, submit);
/* Dx = A*(P+Pxy) + B*(Q+Qxy) */
srcs[0] = dp;
srcs[1] = dq;
coef[0] = raid6_gfexi[failb-faila];
coef[1] = raid6_gfinv[raid6_gfexp[faila]^raid6_gfexp[failb]];
init_async_submit(submit, 0, tx, NULL, NULL, scribble);
tx = async_sum_product(dq, srcs, coef, bytes, submit);
/* Dy = P+Pxy+Dx */
srcs[0] = dp;
srcs[1] = dq;
init_async_submit(submit, flags | ASYNC_TX_XOR_DROP_DST, tx, cb_fn,
cb_param, scribble);
tx = async_xor(dp, srcs, 0, 2, bytes, submit);
return tx;
}
/**
* async_raid6_2data_recov - asynchronously calculate two missing data blocks
* @disks: number of disks in the RAID-6 array
* @bytes: block size
* @faila: first failed drive index
* @failb: second failed drive index
* @blocks: array of source pointers where the last two entries are p and q
* @submit: submission/completion modifiers
*/
struct dma_async_tx_descriptor *
async_raid6_2data_recov(int disks, size_t bytes, int faila, int failb,
struct page **blocks, struct async_submit_ctl *submit)
{
BUG_ON(faila == failb);
if (failb < faila)
swap(faila, failb);
pr_debug("%s: disks: %d len: %zu\n", __func__, disks, bytes);
/* we need to preserve the contents of 'blocks' for the async
* case, so punt to synchronous if a scribble buffer is not available
*/
if (!submit->scribble) {
void **ptrs = (void **) blocks;
int i;
async_tx_quiesce(&submit->depend_tx);
for (i = 0; i < disks; i++)
ptrs[i] = page_address(blocks[i]);
raid6_2data_recov(disks, bytes, faila, failb, ptrs);
async_tx_sync_epilog(submit);
return NULL;
}
switch (disks) {
case 4:
/* dma devices do not uniformly understand a zero source pq
* operation (in contrast to the synchronous case), so
* explicitly handle the 4 disk special case
*/
return __2data_recov_4(bytes, faila, failb, blocks, submit);
case 5:
/* dma devices do not uniformly understand a single
* source pq operation (in contrast to the synchronous
* case), so explicitly handle the 5 disk special case
*/
return __2data_recov_5(bytes, faila, failb, blocks, submit);
default:
return __2data_recov_n(disks, bytes, faila, failb, blocks, submit);
}
}
EXPORT_SYMBOL_GPL(async_raid6_2data_recov);
/**
* async_raid6_datap_recov - asynchronously calculate a data and the 'p' block
* @disks: number of disks in the RAID-6 array
* @bytes: block size
* @faila: failed drive index
* @blocks: array of source pointers where the last two entries are p and q
* @submit: submission/completion modifiers
*/
struct dma_async_tx_descriptor *
async_raid6_datap_recov(int disks, size_t bytes, int faila,
struct page **blocks, struct async_submit_ctl *submit)
{
struct dma_async_tx_descriptor *tx = NULL;
struct page *p, *q, *dq;
u8 coef;
enum async_tx_flags flags = submit->flags;
dma_async_tx_callback cb_fn = submit->cb_fn;
void *cb_param = submit->cb_param;
void *scribble = submit->scribble;
struct page *srcs[2];
pr_debug("%s: disks: %d len: %zu\n", __func__, disks, bytes);
/* we need to preserve the contents of 'blocks' for the async
* case, so punt to synchronous if a scribble buffer is not available
*/
if (!scribble) {
void **ptrs = (void **) blocks;
int i;
async_tx_quiesce(&submit->depend_tx);
for (i = 0; i < disks; i++)
ptrs[i] = page_address(blocks[i]);
raid6_datap_recov(disks, bytes, faila, ptrs);
async_tx_sync_epilog(submit);
return NULL;
}
p = blocks[disks-2];
q = blocks[disks-1];
/* Compute syndrome with zero for the missing data page
* Use the dead data page as temporary storage for delta q
*/
dq = blocks[faila];
blocks[faila] = (void *)raid6_empty_zero_page;
blocks[disks-1] = dq;
/* in the 4 disk case we only need to perform a single source
* multiplication
*/
if (disks == 4) {
int good = faila == 0 ? 1 : 0;
struct page *g = blocks[good];
init_async_submit(submit, 0, tx, NULL, NULL, scribble);
tx = async_memcpy(p, g, 0, 0, bytes, submit);
init_async_submit(submit, 0, tx, NULL, NULL, scribble);
tx = async_mult(dq, g, raid6_gfexp[good], bytes, submit);
} else {
init_async_submit(submit, 0, tx, NULL, NULL, scribble);
tx = async_gen_syndrome(blocks, 0, disks, bytes, submit);
}
/* Restore pointer table */
blocks[faila] = dq;
blocks[disks-1] = q;
/* calculate g^{-faila} */
coef = raid6_gfinv[raid6_gfexp[faila]];
srcs[0] = dq;
srcs[1] = q;
init_async_submit(submit, ASYNC_TX_XOR_DROP_DST, tx, NULL, NULL,
scribble);
tx = async_xor(dq, srcs, 0, 2, bytes, submit);
init_async_submit(submit, 0, tx, NULL, NULL, scribble);
tx = async_mult(dq, dq, coef, bytes, submit);
srcs[0] = p;
srcs[1] = dq;
init_async_submit(submit, flags | ASYNC_TX_XOR_DROP_DST, tx, cb_fn,
cb_param, scribble);
tx = async_xor(p, srcs, 0, 2, bytes, submit);
return tx;
}
EXPORT_SYMBOL_GPL(async_raid6_datap_recov);
MODULE_AUTHOR("Dan Williams <dan.j.williams@intel.com>");
MODULE_DESCRIPTION("asynchronous RAID-6 recovery api");
MODULE_LICENSE("GPL");

View file

@ -194,5 +194,13 @@ async_syndrome_val(struct page **blocks, unsigned int offset, int src_cnt,
size_t len, enum sum_check_flags *pqres, struct page *spare, size_t len, enum sum_check_flags *pqres, struct page *spare,
struct async_submit_ctl *submit); struct async_submit_ctl *submit);
struct dma_async_tx_descriptor *
async_raid6_2data_recov(int src_num, size_t bytes, int faila, int failb,
struct page **ptrs, struct async_submit_ctl *submit);
struct dma_async_tx_descriptor *
async_raid6_datap_recov(int src_num, size_t bytes, int faila,
struct page **ptrs, struct async_submit_ctl *submit);
void async_tx_quiesce(struct dma_async_tx_descriptor **tx); void async_tx_quiesce(struct dma_async_tx_descriptor **tx);
#endif /* _ASYNC_TX_H_ */ #endif /* _ASYNC_TX_H_ */