The patch titled md: fix an occasional deadlock in raid5 has been removed from the -mm tree. Its filename was md-fix-an-occasional-deadlock-in-raid5.patch This patch was dropped because it was merged into mainline or a subsystem tree The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/ ------------------------------------------------------ Subject: md: fix an occasional deadlock in raid5 From: NeilBrown <neilb@xxxxxxx> raid5's 'make_request' function calls generic_make_request on underlying devices and if we run out of stripe heads, it could end up waiting for one of those requests to complete. This is bad as recursive calls to generic_make_request go on a queue and are not even attempted until make_request completes. So: don't make any generic_make_request calls in raid5 make_request until all waiting has been done. We do this by simply setting STRIPE_HANDLE instead of calling handle_stripe(). If we need more stripe_heads, raid5d will get called to process the pending stripe_heads which will call generic_make_request from a This change by itself causes a performance hit. So add a change so that raid5_activate_delayed is only called at unplug time, never in raid5. This seems to bring back the performance numbers. Calling it in raid5d was sometimes too soon... Neil said: How about we queue it for 2.6.25-rc1 and then about when -rc2 comes out, we queue it for 2.6.24.y? Cc: "Dan Williams" <dan.j.williams@xxxxxxxxx> Signed-off-by: Neil Brown <neilb@xxxxxxx> Tested-by: dean gaudet <dean@xxxxxxxxxx> Cc: <stable@xxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- drivers/md/raid5.c | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) diff -puN drivers/md/raid5.c~md-fix-an-occasional-deadlock-in-raid5 drivers/md/raid5.c --- a/drivers/md/raid5.c~md-fix-an-occasional-deadlock-in-raid5 +++ a/drivers/md/raid5.c @@ -3159,7 +3159,8 @@ static void raid5_activate_delayed(raid5 atomic_inc(&conf->preread_active_stripes); list_add_tail(&sh->lru, &conf->handle_list); } - } + } else + blk_plug_device(conf->mddev->queue); } static void activate_bit_delay(raid5_conf_t *conf) @@ -3549,7 +3550,8 @@ static int make_request(struct request_q goto retry; } finish_wait(&conf->wait_for_overlap, &w); - handle_stripe(sh, NULL); + set_bit(STRIPE_HANDLE, &sh->state); + clear_bit(STRIPE_DELAYED, &sh->state); release_stripe(sh); } else { /* cannot get stripe for read-ahead, just give-up */ @@ -3892,7 +3894,7 @@ static int retry_aligned_read(raid5_con * During the scan, completed stripes are saved for us by the interrupt * handler, so that they will not have to wait for our next wakeup. */ -static void raid5d (mddev_t *mddev) +static void raid5d(mddev_t *mddev) { struct stripe_head *sh; raid5_conf_t *conf = mddev_to_conf(mddev); @@ -3917,12 +3919,6 @@ static void raid5d (mddev_t *mddev) activate_bit_delay(conf); } - if (list_empty(&conf->handle_list) && - atomic_read(&conf->preread_active_stripes) < IO_THRESHOLD && - !blk_queue_plugged(mddev->queue) && - !list_empty(&conf->delayed_list)) - raid5_activate_delayed(conf); - while ((bio = remove_bio_from_retry(conf))) { int ok; spin_unlock_irq(&conf->device_lock); _ Patches currently in -mm which might be from neilb@xxxxxxx are origin.patch one-less-parameter-to-__d_path.patch d_path-kerneldoc-cleanup.patch d_path-use-struct-path-in-struct-avc_audit_data.patch d_path-make-proc_get_link-use-a-struct-path-argument.patch d_path-make-get_dcookie-use-a-struct-path-argument.patch use-struct-path-in-struct-svc_export.patch use-struct-path-in-struct-svc_export-checkpatch-fixes.patch use-struct-path-in-struct-svc_expkey.patch d_path-make-seq_path-use-a-struct-path-argument.patch d_path-make-d_path-use-a-struct-path.patch d_path-make-d_path-use-a-struct-path-fix.patch - To unsubscribe from this list: send the line "unsubscribe mm-commits" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html