summaryrefslogtreecommitdiffstats
path: root/block/elevator.c
Commit message (Collapse)AuthorAgeFilesLines
* block: Initial support for data-less (or empty) barrier supportJens Axboe2007-10-161-0/+8
| | | | | | | | | This implements functionality to pass down or insert a barrier in a queue, without having data attached to it. The ->prepare_flush_fn() infrastructure from data barriers are reused to provide this functionality. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
* block: add end_queued_request() and end_dequeued_request() helpersJens Axboe2007-10-161-8/+1
| | | | | | | We can use this helper in the elevator core for BLKPREP_KILL, and it'll also be useful for the empty barrier patch. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
* kobjects: fix up improper use of the kobject name fieldGreg Kroah-Hartman2007-10-121-1/+1
| | | | | | | A number of different drivers incorrect access the kobject name field directly. This is not correct as the name might not be in the array. Use the proper accessor function instead.
* [BLOCK] Get rid of request_queue_t typedefJens Axboe2007-07-241-35/+40
| | | | | | | | | Some of the code has been gradually transitioned to using the proper struct request_queue, but there's lots left. So do a full sweet of the kernel and get rid of this typedef and replace its uses with the proper type. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
* Slab allocators: Replace explicit zeroing with __GFP_ZEROChristoph Lameter2007-07-171-2/+1
| | | | | | | | | | | | | kmalloc_node() and kmem_cache_alloc_node() were not available in a zeroing variant in the past. But with __GFP_ZERO it is possible now to do zeroing while allocating. Use __GFP_ZERO to remove the explicit clearing of memory via memset whereever we can. Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* Use list_for_each_entry() instead of list_for_each() in the block deviceMatthias Kaehlcke2007-07-101-10/+3
| | | | | | | | elevator Signed-off-by: Matthias Kaehlcke <matthias.kaehlcke@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
* [PATCH] elevator: elv_list_lock does not need irq disablingJens Axboe2007-04-301-8/+9
| | | | | | It's never grabbed from irq context, so just make it plain spin_lock(). Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
* make elv_register() output atomicThibaut VARENE2007-03-271-3/+4
| | | | | | | | | | | | | | | | Booting 2.6.21-rc3-g45592145 I noticed the following on one of my machines in the bootlog: io scheduler noop registered<6>Time: jiffies clocksource has been installed. io scheduler deadline registered (default) Looking at block/elevator.c, it appears that elv_register() uses two consecutive printks in a non-atomic way, leading to the above glitch. The attached trivial patch fixes this issue, by using a single printk. Signed-off-by: Thibaut VARENE <varenet@parisc-linux.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
* elevator: don't sort reads between writesJens Axboe2007-02-111-0/+2
| | | | | | | Don't allow elv_dispatch_sort() to mix reads and writes together, it's rarely a good idea. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
* elevator: abstract out the activate and deactivate functionsJens Axboe2007-02-111-9/+20
| | | | Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
* [PATCH] elevator: move clearing of unplug flag earlierLinas Vepstas2007-01-231-5/+6
| | | | | | | | | | | | | | | | | | | | | | | A flag was recently added to the elevator code to avoid performing an unplug when reuests are being re-queued. The goal of this flag was to avoid a deep recursion that can occur when re-queueing requests after a SCSI device/host reset. See http://lkml.org/lkml/2006/5/17/254 However, that fix added the flag near the bottom of a case statement, where an earlier break (in an if statement) could transport one out of the case, without setting the flag. This patch sets the flag earlier in the case statement. I re-discovered the deep recursion recently during testing; I was told that it was a known problem, and the fix to it was in the kernel I was testing. Indeed it was ... but it didn't fix the bug. With the patch below, I no longer see the bug. Signed-off by: Linas Vepstas <linas@austin.ibm.com> Signed-off-by: Jens Axboe <axboe@suse.de> Cc: Chris Wright <chrisw@sous-sol.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* [PATCH] elevator: fixup typo in merge logicJens Axboe2006-12-211-1/+1
| | | | | | | | | | | | The recent io scheduler allow_merge commit left the block layer with no merging, oops. This patch fixes that up. That means the CFQ change needs to be verified again, it might not fix the original bug now. But that's a seperate thing, I'll double check that tomorrow. Signed-off-by: Jens Axboe <jens.axboe@oracle.com> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* [PATCH] cfq-iosched: don't allow sync merges across queuesJens Axboe2006-12-201-4/+22
| | | | | | | | | | | | Currently we allow any merge, even if the io originates from different processes. This can cause really bad starvation and unfairness, if those ios happen to be synchronous (reads or direct writes). So add a allow_merge hook to the io scheduler ops, so an io scheduler can help decide whether a bio/process combination may be merged with an existing request. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
* [BLOCK] Cleanup unused variable passingJens Axboe2006-12-011-2/+2
| | | | | | | | - ->init_queue() does not need the elevator passed in - ->put_request() is a hot path and need not have the queue passed in - cfq_update_io_seektime() does not need cfqd passed in Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
* [PATCH] block layer: elv_iosched_show should get elv_list_lockVasily Tarasov2006-10-121-2/+2
| | | | | | | | elv_iosched_show function iterates other elv_list, hence elv_list_lock should be got. Signed-off-by: Vasily Tarasov <vtaras@openvz.org> Signed-off-by: Vasily Tarasov <jens.axboe@oracle.com>
* [PATCH] block layer: elevator_find function cleanupVasily Tarasov2006-10-121-8/+5
| | | | | | | | We can easily produce search through the elevator list without introducing additional elevator_type variable. Signed-off-by: Vasily Tarasov <vtaras@openvz.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
* [PATCH] Update axboe@suse.de email addressJens Axboe2006-09-301-1/+1
| | | | | | | As people often look for the copyright in files to see who to mail, update the link to a neutral one. Signed-off-by: Jens Axboe <axboe@kernel.dk>
* [PATCH] Make sure all block/io scheduler setups are node awareJens Axboe2006-09-301-5/+6
| | | | | | | Some were kmalloc_node(), some were still kmalloc(). Change them all to kmalloc_node(). Signed-off-by: Jens Axboe <axboe@suse.de>
* [PATCH] Drop useless bio passing in may_queue/set_request APIJens Axboe2006-09-301-5/+4
| | | | | | It's not needed for anything, so kill the bio passing. Signed-off-by: Jens Axboe <axboe@suse.de>
* [PATCH] Remove ->waiting member from struct requestJens Axboe2006-09-301-2/+1
| | | | | | | | As the comments indicates in blkdev.h, we can fold it into ->end_io_data usage as that is really what ->waiting is. Fixup the users of blk_end_sync_rq(). Signed-off-by: Jens Axboe <axboe@kernel.dk>
* [PATCH] elevator: abstract out the rbtree sort handlingJens Axboe2006-09-301-14/+109
| | | | | | | | | | | | The rbtree sort/lookup/reposition logic is mostly duplicated in cfq/deadline/as, so move it to the elevator core. The io schedulers still provide the actual rb root, as we don't want to impose any sort of specific handling on the schedulers. Introduce the helpers and rb_node in struct request to help migrate the IO schedulers. Signed-off-by: Jens Axboe <axboe@suse.de>
* [PATCH] elevator: move the backmerging logic into the elevator coreJens Axboe2006-09-301-15/+132
| | | | | | | | | | | Right now, every IO scheduler implements its own backmerging (except for noop, which does no merging). That results in duplicated code for essentially the same operation, which is never a good thing. This patch moves the backmerging out of the io schedulers and into the elevator core. We save 1.6kb of text and as a bonus get backmerging for noop as well. Win-win! Signed-off-by: Jens Axboe <axboe@suse.de>
* [PATCH] Split struct request ->flags into two partsJens Axboe2006-09-301-13/+13
| | | | | | | | | | Right now ->flags is a bit of a mess: some are request types, and others are just modifiers. Clean this up by splitting it into ->cmd_type and ->cmd_flags. This allows introduction of generic Linux block message types, useful for sending generic Linux commands to block devices. Signed-off-by: Jens Axboe <axboe@suse.de>
* elv_unregister: fix possible crash on module unloadOleg Nesterov2006-08-221-1/+2
| | | | | | | | | | An exiting task or process which didn't do I/O yet have no io context, elv_unregister() should check it is not NULL. Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru> Acked-by: Jens Axboe <axboe@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
* Remove obsolete #include <linux/config.h>Jörn Engel2006-06-301-1/+0
| | | | | Signed-off-by: Jörn Engel <joern@wohnheim.fh-wedel.de> Signed-off-by: Adrian Bunk <bunk@stusta.de>
* [PATCH] remove dead code from elevator switchingDave Jones2006-06-231-3/+0
| | | | | | | | | | We already drop the refcount in elevator_exit(), and as we're setting 'e' to NULL, we'll never take that branch anyway. Finally, as 'e' is a local var that isn't referenced afterwards, setting it to NULL is pointless. Signed-off-by: Dave Jones <davej@redhat.com> Signed-off-by: Jens Axboe <axboe@suse.de>
* [PATCH] elevator switching raceJens Axboe2006-06-081-21/+34
| | | | | | | | | | | | | | | | | | | | There's a race between shutting down one io scheduler and firing up the next, in which a new io could enter and cause the io scheduler to be invoked with bad or NULL data. To fix this, we need to maintain the queue lock for a bit longer. Unfortunately we cannot do that, since the elevator init requires to be run without the lock held. This isn't easily fixable, without also changing the mempool API. So split the initialization into two parts, and alloc-init operation and an attach operation. Then we can preallocate the io scheduler and related structures, and run the attach inside the lock after we detach the old one. This patch has survived 30 minutes of 1 second io scheduler switching with a very busy io load. Signed-off-by: Jens Axboe <axboe@suse.de> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* [BLOCK] limit request_fn recursionJens Axboe2006-05-111-1/+7
| | | | | | | | | | | | | | | | | Don't recurse back into the driver even if the unplug threshold is met, when the driver asks for a requeue. This is both silly from a logical point of view (requeues typically happen due to driver/hardware shortage), and also dangerous since we could hit an endless request_fn -> requeue -> unplug -> request_fn loop and crash on stack overrun. Also limit blk_run_queue() to one level of recursion, similar to how blk_start_queue() works. This patch fixed a real problem with SLES10 and lpfc, and it could hit any SCSI lld that returns non-zero from it's ->queuecommand() handler. Signed-off-by: Jens Axboe <axboe@suse.de> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* [PATCH] block/elevator.c: remove unused exportsAdrian Bunk2006-04-201-2/+0
| | | | | | | | | | | This patch removes the following unused EXPORT_SYMBOL's: - elv_requeue_request - elv_completed_request They are only used by the block core, hence they need not be exported. Signed-off-by: Adrian Bunk <bunk@stusta.de> Signed-off-by: Jens Axboe <axboe@suse.de>
* [PATCH] Don't pass boot parameters to argv_init[]OGAWA Hirofumi2006-03-311-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | The boot cmdline is parsed in parse_early_param() and parse_args(,unknown_bootoption). And __setup() is used in obsolete_checksetup(). start_kernel() -> parse_args() -> unknown_bootoption() -> obsolete_checksetup() If __setup()'s callback (->setup_func()) returns 1 in obsolete_checksetup(), obsolete_checksetup() thinks a parameter was handled. If ->setup_func() returns 0, obsolete_checksetup() tries other ->setup_func(). If all ->setup_func() that matched a parameter returns 0, a parameter is seted to argv_init[]. Then, when runing /sbin/init or init=app, argv_init[] is passed to the app. If the app doesn't ignore those arguments, it will warning and exit. This patch fixes a wrong usage of it, however fixes obvious one only. Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* BUG_ON() Conversion in block/elevator.cEric Sesterhenn2006-03-241-2/+1
| | | | | | | | this changes if() BUG(); constructs to BUG_ON() which is cleaner, contains unlikely() and can better optimized away. Signed-off-by: Eric Sesterhenn <snakebyte@gmx.de> Signed-off-by: Adrian Bunk <bunk@stusta.de>
* [PATCH] Block queue IO tracing support (blktrace) as of 2006-03-23Jens Axboe2006-03-231-0/+4
| | | | Signed-off-by: Jens Axboe <axboe@suse.de>
* [PATCH] fix rmmod problems with elevator attributes, clean them upAl Viro2006-03-181-3/+4
|
* [PATCH] elevator_t lifetime rules and sysfs fixesAl Viro2006-03-181-38/+108
|
* [PATCH] stop elv_unregister() from rogering other iosched's data, fix lockingAl Viro2006-03-181-15/+9
| | | | Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
* [PATCH] block: implement elv_insert and use it (fix ordcolor flipping bug)Tejun Heo2006-02-081-32/+38
| | | | | | | | | | | | | | | | | | | q->ordcolor must only be flipped on initial queueing of a hardbarrier request. Constructing ordered sequence and requeueing used to pass through __elv_add_request() which flips q->ordcolor when it sees a barrier request. This patch separates out elv_insert() from __elv_add_request() and uses elv_insert() when constructing ordered sequence and requeueing. elv_insert() inserts the given request at the specified position and does nothing else. Signed-off-by: Tejun Heo <htejun@gmail.com> Acked-by: Jens Axboe <axboe@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* [PATCH] block: request_queue->ordcolor must not be flipped on SOFTBARRIERTejun Heo2006-02-051-1/+2
| | | | | | | | | q->ordcolor must not be flipped on SOFTBARRIER. Signed-off-by: Tejun Heo <htejun@gmail.com> Acked-by: Jens Axboe <axboe@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* [BLOCK] elevator: allow default scheduler to potentially be modularNate Diller2006-01-241-4/+6
| | | | | | | | | | | | | | | Jens has decided that allowing the default scheduler to be a module is a bug, and should not be allowed under kconfig. However, I find that scenario useful for debugging, and wish for the kernel to be able to handle this situation without OOPSing, if I enable such an option in the .config directly. This patch dynamically checks for the presence of the compiled-in default, and falls back to no-op, emitting a suitable error message, when the default is not available Tested for a range of boot options on 2.6.16-rc1-mm2. Signed-off-by: Nate Diller <nate.diller@gmail.com> Signed-off-by: Jens Axboe <axboe@suse.de>
* [BLOCK] elevator: default choice selectionNate Diller2006-01-241-31/+14
| | | | | | | | | | | | | | | | | My previous default iosched patch did a poor job dealing with the 'elevator=' boot-time option. The old behavior falls back to the compiled-in default if the requested one is not registered at boot time. This patch dynamically evaluates which default to use, and emits a suitable error message when the requested scheduler is not available. It also does the 'as' -> 'anticipatory' conversion before elevator registration, which along with a modified registration function, allows it to correctly indicate which default scheduler is in use. Tested for a range of boot options on 2.6.16-rc1-mm2. Signed-off-by: Nate Diller <nate.diller@gmail.com> Signed-off-by: Jens Axboe <axboe@suse.de>
* [BLOCK] elevator: if specified scheduler is not found, fall back to defaultJens Axboe2006-01-161-2/+2
| | | | Signed-off-by: Jens Axboe <axboe@suse.de>
* [BLOCK] elevator: Make elevator=as work again for anticipatoryChuck Ebbert2006-01-161-0/+7
| | | | | Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com> Signed-off-by: Jens Axboe <axboe@suse.de>
* [PATCH] fix queue stalling while barrier sequencingTejun Heo2006-01-121-10/+10
| | | | | | | | | | | | | | | | | | | If ordered tag isn't supported, request ordering for barrier sequencing is performed by queue draining, which basically hangs the request queue until elv_completed_request() reports completion of all previous fs requests. The condition check in elv_completed_request() was only performed for fs requests. If a special request is queued between the last to-be-drained request and the barrier sequence, draining is never completed and the queue is stalled forever. This patch moves the end-of-draining condition check such that it's performed for all requests. Signed-off-by: Tejun Heo <htejun@gmail.com> Signed-off-by: Jens Axboe <axboe@suse.de> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* make elv_try_merge() static, kill the dead declaration ofCoywolf Qi Hunt2006-01-091-2/+1
| | | | | | | elv_try_last_merge(). Signed-off-by: Coywolf Qi Hunt <qiyong@fc-cn.com> Signed-off-by: Jens Axboe <axboe@suse.de>
* [BLOCK] reimplement handling of barrier requestTejun Heo2006-01-061-26/+58
| | | | | | | | | | | | Reimplement handling of barrier requests. * Flexible handling to deal with various capabilities of target devices. * Retry support for falling back. * Tagged queues which don't support ordered tag can do ordered. Signed-off-by: Tejun Heo <htejun@gmail.com> Signed-off-by: Jens Axboe <axboe@suse.de>
* [BLOCK] add @uptodate to end_that_request_last() and @error to rq_end_io_fn()Tejun Heo2006-01-061-1/+1
| | | | | | | | | | | | | | | | | add @uptodate argument to end_that_request_last() and @error to rq_end_io_fn(). there's no generic way to pass error code to request completion function, making generic error handling of non-fs request difficult (rq->errors is driver-specific and each driver uses it differently). this patch adds @uptodate to end_that_request_last() and @error to rq_end_io_fn(). for fs requests, this doesn't really matter, so just using the same uptodate argument used in the last call to end_that_request_first() should suffice. imho, this can also help the generic command-carrying request jens is working on. Signed-off-by: tejun heo <htejun@gmail.com> Signed-Off-By: Jens Axboe <axboe@suse.de>
* [BLOCK] new block/ directory comment tidyCoywolf Qi Hunt2005-11-181-2/+0
| | | | | | | | Some leftover comments referring to drivers/block that are now block/. They don't add any information we don't already have, so kill them. Signed-off-by: Coywolf Qi Hunt <qiyong@fc-cn.com> Signed-off-by: Jens Axboe <axboe@suse.de>
* [BLOCK] elevator: elv_latter/former_request updateTejun Heo2005-11-121-14/+0
| | | | | | | | | | | | With generic dispatch queue update, implicit former/latter request handling using rq->queuelist.prev/next doesn't work as expected anymore. Also, the only iosched dependent on this feature was noop-iosched and it has been reimplemented to have its own latter/former methods. This patch removes implicit former/latter handling. Signed-off-by: Tejun Heo <htejun@gmail.com> Signed-off-by: Jens Axboe <axboe@suse.de>
* [BLOCK] fix string handling in elv_iosched_storeTejun Heo2005-11-121-4/+6
| | | | | | | | | elv_iosched_store doesn't terminate string passed from userspace if it's too long. Also, if the written length is zero (probably not possible), it accesses elevator_name[-1]. This patch fixes both bugs. Signed-off-by: Tejun Heo <htejun@gmail.com> Signed-off-by: Jens Axboe <axboe@suse.de>
* [BLOCK] Implement elv_drain_elevator for improved switch error detectionTejun Heo2005-11-121-5/+20
| | | | | | | | | | | | This patch adds request_queue->nr_sorted which keeps the number of requests in the iosched and implement elv_drain_elevator which performs forced dispatching. elv_drain_elevator checks whether iosched actually dispatches all requests it has and prints error message if it doesn't. As buggy forced dispatching can result in wrong barrier operations, I think this extra check is worthwhile. Signed-off-by: Tejun Heo <htejun@gmail.com> Signed-off-by: Jens Axboe <axboe@suse.de>
* [BLOCK] elevator: run queue in elevator_switchTejun Heo2005-11-121-0/+2
| | | | | | | | elevator_dispatch needs to run queue after forced dispatching; otherwise, the queue might stall. Signed-off-by: Tejun Heo <htejun@gmail.com> Signed-off-by: Jens Axboe <axboe@suse.de>
OpenPOWER on IntegriCloud