md/raid1: split out two sub-functions from sync_request_write

sync_request_write is too big and too deep.
So split out two self-contains bits of functionality into separate
function.

Signed-off-by: NeilBrown <neilb@suse.de>
This commit is contained in:
NeilBrown 2011-05-11 14:40:44 +10:00
parent 6f8d0c77ce
commit a68e587035
1 changed files with 195 additions and 176 deletions

View File

@ -1181,101 +1181,9 @@ static void end_sync_write(struct bio *bio, int error)
} }
} }
static void sync_request_write(mddev_t *mddev, r1bio_t *r1_bio) static int fix_sync_read_error(r1bio_t *r1_bio)
{ {
conf_t *conf = mddev->private; /* Try some synchronous reads of other devices to get
int i;
int disks = conf->raid_disks;
struct bio *bio, *wbio;
bio = r1_bio->bios[r1_bio->read_disk];
if (test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery)) {
/* We have read all readable devices. If we haven't
* got the block, then there is no hope left.
* If we have, then we want to do a comparison
* and skip the write if everything is the same.
* If any blocks failed to read, then we need to
* attempt an over-write
*/
int primary;
if (!test_bit(R1BIO_Uptodate, &r1_bio->state)) {
for (i=0; i<mddev->raid_disks; i++)
if (r1_bio->bios[i]->bi_end_io == end_sync_read)
md_error(mddev, conf->mirrors[i].rdev);
md_done_sync(mddev, r1_bio->sectors, 1);
put_buf(r1_bio);
return;
}
for (primary=0; primary<mddev->raid_disks; primary++)
if (r1_bio->bios[primary]->bi_end_io == end_sync_read &&
test_bit(BIO_UPTODATE, &r1_bio->bios[primary]->bi_flags)) {
r1_bio->bios[primary]->bi_end_io = NULL;
rdev_dec_pending(conf->mirrors[primary].rdev, mddev);
break;
}
r1_bio->read_disk = primary;
for (i=0; i<mddev->raid_disks; i++)
if (r1_bio->bios[i]->bi_end_io == end_sync_read) {
int j;
int vcnt = r1_bio->sectors >> (PAGE_SHIFT- 9);
struct bio *pbio = r1_bio->bios[primary];
struct bio *sbio = r1_bio->bios[i];
if (test_bit(BIO_UPTODATE, &sbio->bi_flags)) {
for (j = vcnt; j-- ; ) {
struct page *p, *s;
p = pbio->bi_io_vec[j].bv_page;
s = sbio->bi_io_vec[j].bv_page;
if (memcmp(page_address(p),
page_address(s),
PAGE_SIZE))
break;
}
} else
j = 0;
if (j >= 0)
mddev->resync_mismatches += r1_bio->sectors;
if (j < 0 || (test_bit(MD_RECOVERY_CHECK, &mddev->recovery)
&& test_bit(BIO_UPTODATE, &sbio->bi_flags))) {
sbio->bi_end_io = NULL;
rdev_dec_pending(conf->mirrors[i].rdev, mddev);
} else {
/* fixup the bio for reuse */
int size;
sbio->bi_vcnt = vcnt;
sbio->bi_size = r1_bio->sectors << 9;
sbio->bi_idx = 0;
sbio->bi_phys_segments = 0;
sbio->bi_flags &= ~(BIO_POOL_MASK - 1);
sbio->bi_flags |= 1 << BIO_UPTODATE;
sbio->bi_next = NULL;
sbio->bi_sector = r1_bio->sector +
conf->mirrors[i].rdev->data_offset;
sbio->bi_bdev = conf->mirrors[i].rdev->bdev;
size = sbio->bi_size;
for (j = 0; j < vcnt ; j++) {
struct bio_vec *bi;
bi = &sbio->bi_io_vec[j];
bi->bv_offset = 0;
if (size > PAGE_SIZE)
bi->bv_len = PAGE_SIZE;
else
bi->bv_len = size;
size -= PAGE_SIZE;
memcpy(page_address(bi->bv_page),
page_address(pbio->bi_io_vec[j].bv_page),
PAGE_SIZE);
}
}
}
}
if (!test_bit(R1BIO_Uptodate, &r1_bio->state)) {
/* ouch - failed to read all of that.
* Try some synchronous reads of other devices to get
* good data, much like with normal read errors. Only * good data, much like with normal read errors. Only
* read into the pages we already have so we don't * read into the pages we already have so we don't
* need to re-issue the read request. * need to re-issue the read request.
@ -1283,6 +1191,9 @@ static void sync_request_write(mddev_t *mddev, r1bio_t *r1_bio)
* active sync request, there is no normal IO, and * active sync request, there is no normal IO, and
* no overlapping syncs. * no overlapping syncs.
*/ */
mddev_t *mddev = r1_bio->mddev;
conf_t *conf = mddev->private;
struct bio *bio = r1_bio->bios[r1_bio->read_disk];
sector_t sect = r1_bio->sector; sector_t sect = r1_bio->sector;
int sectors = r1_bio->sectors; int sectors = r1_bio->sectors;
int idx = 0; int idx = 0;
@ -1361,14 +1272,122 @@ static void sync_request_write(mddev_t *mddev, r1bio_t *r1_bio)
(unsigned long long)r1_bio->sector); (unsigned long long)r1_bio->sector);
md_done_sync(mddev, r1_bio->sectors, 0); md_done_sync(mddev, r1_bio->sectors, 0);
put_buf(r1_bio); put_buf(r1_bio);
return; return 0;
} }
sectors -= s; sectors -= s;
sect += s; sect += s;
idx ++; idx ++;
} }
return 1;
}
static int process_checks(r1bio_t *r1_bio)
{
/* We have read all readable devices. If we haven't
* got the block, then there is no hope left.
* If we have, then we want to do a comparison
* and skip the write if everything is the same.
* If any blocks failed to read, then we need to
* attempt an over-write
*/
mddev_t *mddev = r1_bio->mddev;
conf_t *conf = mddev->private;
int primary;
int i;
if (!test_bit(R1BIO_Uptodate, &r1_bio->state)) {
for (i=0; i<mddev->raid_disks; i++)
if (r1_bio->bios[i]->bi_end_io == end_sync_read)
md_error(mddev, conf->mirrors[i].rdev);
md_done_sync(mddev, r1_bio->sectors, 1);
put_buf(r1_bio);
return -1;
}
for (primary=0; primary<mddev->raid_disks; primary++)
if (r1_bio->bios[primary]->bi_end_io == end_sync_read &&
test_bit(BIO_UPTODATE, &r1_bio->bios[primary]->bi_flags)) {
r1_bio->bios[primary]->bi_end_io = NULL;
rdev_dec_pending(conf->mirrors[primary].rdev, mddev);
break;
}
r1_bio->read_disk = primary;
for (i=0; i<mddev->raid_disks; i++)
if (r1_bio->bios[i]->bi_end_io == end_sync_read) {
int j;
int vcnt = r1_bio->sectors >> (PAGE_SHIFT- 9);
struct bio *pbio = r1_bio->bios[primary];
struct bio *sbio = r1_bio->bios[i];
if (test_bit(BIO_UPTODATE, &sbio->bi_flags)) {
for (j = vcnt; j-- ; ) {
struct page *p, *s;
p = pbio->bi_io_vec[j].bv_page;
s = sbio->bi_io_vec[j].bv_page;
if (memcmp(page_address(p),
page_address(s),
PAGE_SIZE))
break;
}
} else
j = 0;
if (j >= 0)
mddev->resync_mismatches += r1_bio->sectors;
if (j < 0 || (test_bit(MD_RECOVERY_CHECK, &mddev->recovery)
&& test_bit(BIO_UPTODATE, &sbio->bi_flags))) {
sbio->bi_end_io = NULL;
rdev_dec_pending(conf->mirrors[i].rdev, mddev);
} else {
/* fixup the bio for reuse */
int size;
sbio->bi_vcnt = vcnt;
sbio->bi_size = r1_bio->sectors << 9;
sbio->bi_idx = 0;
sbio->bi_phys_segments = 0;
sbio->bi_flags &= ~(BIO_POOL_MASK - 1);
sbio->bi_flags |= 1 << BIO_UPTODATE;
sbio->bi_next = NULL;
sbio->bi_sector = r1_bio->sector +
conf->mirrors[i].rdev->data_offset;
sbio->bi_bdev = conf->mirrors[i].rdev->bdev;
size = sbio->bi_size;
for (j = 0; j < vcnt ; j++) {
struct bio_vec *bi;
bi = &sbio->bi_io_vec[j];
bi->bv_offset = 0;
if (size > PAGE_SIZE)
bi->bv_len = PAGE_SIZE;
else
bi->bv_len = size;
size -= PAGE_SIZE;
memcpy(page_address(bi->bv_page),
page_address(pbio->bi_io_vec[j].bv_page),
PAGE_SIZE);
} }
}
}
return 0;
}
static void sync_request_write(mddev_t *mddev, r1bio_t *r1_bio)
{
conf_t *conf = mddev->private;
int i;
int disks = conf->raid_disks;
struct bio *bio, *wbio;
bio = r1_bio->bios[r1_bio->read_disk];
if (test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery))
if (process_checks(r1_bio) < 0)
return;
if (!test_bit(R1BIO_Uptodate, &r1_bio->state))
/* ouch - failed to read all of that. */
if (!fix_sync_read_error(r1_bio))
return;
/* /*
* schedule writes * schedule writes
*/ */