From: NeilBrown <neilb@suse.de>
To: linux-raid@vger.kernel.org
Subject: [md PATCH 08/16] md/raid1: split out two sub-functions from sync_request_write
Date: Wed, 11 May 2011 16:30:31 +1000 [thread overview]
Message-ID: <20110511063031.21263.30829.stgit@notabene.brown> (raw)
In-Reply-To: <20110511062743.21263.72802.stgit@notabene.brown>
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>
---
drivers/md/raid1.c | 365 +++++++++++++++++++++++++++-------------------------
1 files changed, 192 insertions(+), 173 deletions(-)
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index f0b0c79..7fd7a4d 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -1181,194 +1181,213 @@ 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)
{
+ /* Try some synchronous reads of other devices to get
+ * good data, much like with normal read errors. Only
+ * read into the pages we already have so we don't
+ * need to re-issue the read request.
+ * We don't need to freeze the array, because being in an
+ * active sync request, there is no normal IO, and
+ * no overlapping syncs.
+ */
+ mddev_t *mddev = r1_bio->mddev;
conf_t *conf = mddev->private;
- int i;
- int disks = conf->raid_disks;
- struct bio *bio, *wbio;
+ struct bio *bio = r1_bio->bios[r1_bio->read_disk];
+ sector_t sect = r1_bio->sector;
+ int sectors = r1_bio->sectors;
+ int idx = 0;
- 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);
+ while(sectors) {
+ int s = sectors;
+ int d = r1_bio->read_disk;
+ int success = 0;
+ mdk_rdev_t *rdev;
- md_done_sync(mddev, r1_bio->sectors, 1);
+ if (s > (PAGE_SIZE>>9))
+ s = PAGE_SIZE >> 9;
+ do {
+ if (r1_bio->bios[d]->bi_end_io == end_sync_read) {
+ /* No rcu protection needed here devices
+ * can only be removed when no resync is
+ * active, and resync is currently active
+ */
+ rdev = conf->mirrors[d].rdev;
+ if (sync_page_io(rdev,
+ sect,
+ s<<9,
+ bio->bi_io_vec[idx].bv_page,
+ READ, false)) {
+ success = 1;
+ break;
+ }
+ }
+ d++;
+ if (d == conf->raid_disks)
+ d = 0;
+ } while (!success && d != r1_bio->read_disk);
+
+ if (success) {
+ int start = d;
+ /* write it back and re-read */
+ set_bit(R1BIO_Uptodate, &r1_bio->state);
+ while (d != r1_bio->read_disk) {
+ if (d == 0)
+ d = conf->raid_disks;
+ d--;
+ if (r1_bio->bios[d]->bi_end_io != end_sync_read)
+ continue;
+ rdev = conf->mirrors[d].rdev;
+ atomic_add(s, &rdev->corrected_errors);
+ if (sync_page_io(rdev,
+ sect,
+ s<<9,
+ bio->bi_io_vec[idx].bv_page,
+ WRITE, false) == 0)
+ md_error(mddev, rdev);
+ }
+ d = start;
+ while (d != r1_bio->read_disk) {
+ if (d == 0)
+ d = conf->raid_disks;
+ d--;
+ if (r1_bio->bios[d]->bi_end_io != end_sync_read)
+ continue;
+ rdev = conf->mirrors[d].rdev;
+ if (sync_page_io(rdev,
+ sect,
+ s<<9,
+ bio->bi_io_vec[idx].bv_page,
+ READ, false) == 0)
+ md_error(mddev, rdev);
+ }
+ } else {
+ char b[BDEVNAME_SIZE];
+ /* Cannot read from anywhere, array is toast */
+ md_error(mddev, conf->mirrors[r1_bio->read_disk].rdev);
+ printk(KERN_ALERT "md/raid1:%s: %s: unrecoverable I/O read error"
+ " for block %llu\n",
+ mdname(mddev),
+ bdevname(bio->bi_bdev, b),
+ (unsigned long long)r1_bio->sector);
+ md_done_sync(mddev, r1_bio->sectors, 0);
put_buf(r1_bio);
- return;
+ return 0;
}
- 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;
+ sectors -= s;
+ sect += s;
+ 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) {
- 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 (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;
}
- 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
- * read into the pages we already have so we don't
- * need to re-issue the read request.
- * We don't need to freeze the array, because being in an
- * active sync request, there is no normal IO, and
- * no overlapping syncs.
- */
- sector_t sect = r1_bio->sector;
- int sectors = r1_bio->sectors;
- int idx = 0;
-
- while(sectors) {
- int s = sectors;
- int d = r1_bio->read_disk;
- int success = 0;
- mdk_rdev_t *rdev;
-
- if (s > (PAGE_SIZE>>9))
- s = PAGE_SIZE >> 9;
- do {
- if (r1_bio->bios[d]->bi_end_io == end_sync_read) {
- /* No rcu protection needed here devices
- * can only be removed when no resync is
- * active, and resync is currently active
- */
- rdev = conf->mirrors[d].rdev;
- if (sync_page_io(rdev,
- sect,
- s<<9,
- bio->bi_io_vec[idx].bv_page,
- READ, false)) {
- success = 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;
- }
- }
- d++;
- if (d == conf->raid_disks)
- d = 0;
- } while (!success && d != r1_bio->read_disk);
-
- if (success) {
- int start = d;
- /* write it back and re-read */
- set_bit(R1BIO_Uptodate, &r1_bio->state);
- while (d != r1_bio->read_disk) {
- if (d == 0)
- d = conf->raid_disks;
- d--;
- if (r1_bio->bios[d]->bi_end_io != end_sync_read)
- continue;
- rdev = conf->mirrors[d].rdev;
- atomic_add(s, &rdev->corrected_errors);
- if (sync_page_io(rdev,
- sect,
- s<<9,
- bio->bi_io_vec[idx].bv_page,
- WRITE, false) == 0)
- md_error(mddev, rdev);
- }
- d = start;
- while (d != r1_bio->read_disk) {
- if (d == 0)
- d = conf->raid_disks;
- d--;
- if (r1_bio->bios[d]->bi_end_io != end_sync_read)
- continue;
- rdev = conf->mirrors[d].rdev;
- if (sync_page_io(rdev,
- sect,
- s<<9,
- bio->bi_io_vec[idx].bv_page,
- READ, false) == 0)
- md_error(mddev, rdev);
}
+ } 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 {
- char b[BDEVNAME_SIZE];
- /* Cannot read from anywhere, array is toast */
- md_error(mddev, conf->mirrors[r1_bio->read_disk].rdev);
- printk(KERN_ALERT "md/raid1:%s: %s: unrecoverable I/O read error"
- " for block %llu\n",
- mdname(mddev),
- bdevname(bio->bi_bdev, b),
- (unsigned long long)r1_bio->sector);
- md_done_sync(mddev, r1_bio->sectors, 0);
- put_buf(r1_bio);
- return;
+ /* 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);
+ }
+
}
- sectors -= s;
- sect += s;
- idx ++;
}
- }
+ 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
*/
next prev parent reply other threads:[~2011-05-11 6:30 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-05-11 6:30 [md PATCH 00/16] md patches for 2.6.40 NeilBrown
2011-05-11 6:30 ` [md PATCH 02/16] md: reject a re-add request that cannot be honoured NeilBrown
2011-05-11 6:30 ` [md PATCH 04/16] md: simplify raid10 read_balance NeilBrown
2011-05-11 6:30 ` [md PATCH 03/16] md/bitmap: fix saving of events_cleared and other state NeilBrown
2011-05-11 6:30 ` [md PATCH 01/16] md: Fix race when creating a new md device NeilBrown
2011-05-11 6:30 ` [md PATCH 05/16] md/raid1: clean up read_balance NeilBrown
2011-05-11 6:30 ` NeilBrown [this message]
2011-05-11 6:30 ` [md PATCH 14/16] md/raid10: remove unused variable NeilBrown
2011-05-11 6:30 ` [md PATCH 13/16] md/raid10: make more use of 'slot' in raid10d NeilBrown
2011-05-11 6:30 ` [md PATCH 10/16] md/raid1: try fix_sync_read_error before process_checks NeilBrown
2011-05-11 6:30 ` [md PATCH 16/16] md: allow resync_start to be set while an array is active NeilBrown
2011-05-11 6:30 ` [md PATCH 12/16] md/raid10: some tidying up in fix_read_error NeilBrown
2011-05-11 6:30 ` [md PATCH 11/16] md/raid1: improve handling of pages allocated for write-behind NeilBrown
2011-05-11 6:30 ` [md PATCH 15/16] md/raid10: reformat some loops with less indenting NeilBrown
2011-05-11 6:30 ` [md PATCH 07/16] md: make error_handler functions more uniform and correct NeilBrown
2011-05-11 6:30 ` [md PATCH 06/16] md/multipath: discard ->working_disks in favour of ->degraded NeilBrown
2011-05-11 6:30 ` [md PATCH 09/16] md/raid1: tidy up new functions: process_checks and fix_sync_read_error NeilBrown
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20110511063031.21263.30829.stgit@notabene.brown \
--to=neilb@suse.de \
--cc=linux-raid@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).