Jens Axboe 5fccbf61be [PATCH] CFQ: request <-> request merging rr_list fixup ... In very rare circumstances would we be pruning a merged request and at the same time delete the implicated cfqq from the rr_list, and not readd it when the merged request got added. This could cause io stalls until that process issued io again. Fix it up by putting the rr_list add handling into cfq_add_rq_rb(), identical to how pruning is handled in cfq_del_rq_rb(). This fixes a hang reproducible with fsx-linux. Signed-off-by: Jens Axboe <jens.axboe@oracle.com> Signed-off-by: Linus Torvalds <torvalds@osdl.org>		2006-10-31 08:12:45 -08:00
..
as-iosched.c	[PATCH] completions: lockdep annotate on stack completions	2006-10-01 00:39:24 -07:00
blktrace.c	[PATCH] Update axboe@suse.de email address	2006-09-30 20:52:34 +02:00
cfq-iosched.c	[PATCH] CFQ: request <-> request merging rr_list fixup	2006-10-31 08:12:45 -08:00
deadline-iosched.c	[PATCH] Update axboe@suse.de email address	2006-09-30 20:52:34 +02:00
elevator.c	[PATCH] block layer: elv_iosched_show should get elv_list_lock	2006-10-12 15:08:51 +02:00
genhd.c	[PATCH] block: handle subsystem_register() init errors	2006-09-29 09:18:05 -07:00
ioctl.c	[PATCH] dm: export blkdev_driver_ioctl	2006-10-03 08:04:13 -07:00
Kconfig	[PATCH] Only enable CONFIG_BLOCK option for embedded	2006-09-30 21:14:05 +02:00
Kconfig.iosched	[PATCH] BLOCK: Make it possible to disable the block layer [try #6]	2006-09-30 20:52:31 +02:00
ll_rw_blk.c	[PATCH] md: check bio address after mapping through partitions.	2006-10-31 08:07:01 -08:00
Makefile	[PATCH] BLOCK: Make it possible to disable the block layer [try #6]	2006-09-30 20:52:31 +02:00
noop-iosched.c	[PATCH] Make sure all block/io scheduler setups are node aware	2006-09-30 20:29:39 +02:00
scsi_ioctl.c	[PATCH] Split struct request ->flags into two parts	2006-09-30 20:23:37 +02:00