From 4aff5e2333c9a1609662f2091f55c3f6fffdad36 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 10 Aug 2006 08:44:47 +0200 Subject: [PATCH] Split struct request ->flags into two parts 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 --- include/linux/blkdev.h | 180 +++++++++++++++++++++++++++---------------------- 1 file changed, 99 insertions(+), 81 deletions(-) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index cfde8b3ee919..b2a412cf468f 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -120,6 +120,86 @@ struct request_list { wait_queue_head_t wait[2]; }; +/* + * request command types + */ +enum rq_cmd_type_bits { + REQ_TYPE_FS = 1, /* fs request */ + REQ_TYPE_BLOCK_PC, /* scsi command */ + REQ_TYPE_SENSE, /* sense request */ + REQ_TYPE_PM_SUSPEND, /* suspend request */ + REQ_TYPE_PM_RESUME, /* resume request */ + REQ_TYPE_PM_SHUTDOWN, /* shutdown request */ + REQ_TYPE_FLUSH, /* flush request */ + REQ_TYPE_SPECIAL, /* driver defined type */ + REQ_TYPE_LINUX_BLOCK, /* generic block layer message */ + /* + * for ATA/ATAPI devices. this really doesn't belong here, ide should + * use REQ_TYPE_SPECIAL and use rq->cmd[0] with the range of driver + * private REQ_LB opcodes to differentiate what type of request this is + */ + REQ_TYPE_ATA_CMD, + REQ_TYPE_ATA_TASK, + REQ_TYPE_ATA_TASKFILE, +}; + +/* + * For request of type REQ_TYPE_LINUX_BLOCK, rq->cmd[0] is the opcode being + * sent down (similar to how REQ_TYPE_BLOCK_PC means that ->cmd[] holds a + * SCSI cdb. + * + * 0x00 -> 0x3f are driver private, to be used for whatever purpose they need, + * typically to differentiate REQ_TYPE_SPECIAL requests. + * + */ +enum { + /* + * just examples for now + */ + REQ_LB_OP_EJECT = 0x40, /* eject request */ + REQ_LB_OP_FLUSH = 0x41, /* flush device */ +}; + +/* + * request type modified bits. first three bits match BIO_RW* bits, important + */ +enum rq_flag_bits { + __REQ_RW, /* not set, read. set, write */ + __REQ_FAILFAST, /* no low level driver retries */ + __REQ_SORTED, /* elevator knows about this request */ + __REQ_SOFTBARRIER, /* may not be passed by ioscheduler */ + __REQ_HARDBARRIER, /* may not be passed by drive either */ + __REQ_FUA, /* forced unit access */ + __REQ_NOMERGE, /* don't touch this for merging */ + __REQ_STARTED, /* drive already may have started this one */ + __REQ_DONTPREP, /* don't call prep for this one */ + __REQ_QUEUED, /* uses queueing */ + __REQ_ELVPRIV, /* elevator private data attached */ + __REQ_FAILED, /* set if the request failed */ + __REQ_QUIET, /* don't worry about errors */ + __REQ_PREEMPT, /* set for "ide_preempt" requests */ + __REQ_ORDERED_COLOR, /* is before or after barrier */ + __REQ_RW_SYNC, /* request is sync (O_DIRECT) */ + __REQ_NR_BITS, /* stops here */ +}; + +#define REQ_RW (1 << __REQ_RW) +#define REQ_FAILFAST (1 << __REQ_FAILFAST) +#define REQ_SORTED (1 << __REQ_SORTED) +#define REQ_SOFTBARRIER (1 << __REQ_SOFTBARRIER) +#define REQ_HARDBARRIER (1 << __REQ_HARDBARRIER) +#define REQ_FUA (1 << __REQ_FUA) +#define REQ_NOMERGE (1 << __REQ_NOMERGE) +#define REQ_STARTED (1 << __REQ_STARTED) +#define REQ_DONTPREP (1 << __REQ_DONTPREP) +#define REQ_QUEUED (1 << __REQ_QUEUED) +#define REQ_ELVPRIV (1 << __REQ_ELVPRIV) +#define REQ_FAILED (1 << __REQ_FAILED) +#define REQ_QUIET (1 << __REQ_QUIET) +#define REQ_PREEMPT (1 << __REQ_PREEMPT) +#define REQ_ORDERED_COLOR (1 << __REQ_ORDERED_COLOR) +#define REQ_RW_SYNC (1 << __REQ_RW_SYNC) + #define BLK_MAX_CDB 16 /* @@ -129,7 +209,8 @@ struct request { struct list_head queuelist; struct list_head donelist; - unsigned long flags; /* see REQ_ bits below */ + unsigned int cmd_flags; + enum rq_cmd_type_bits cmd_type; /* Maintain bio traversal state for part by part I/O submission. * hard_* are block layer internals, no driver should touch them! @@ -202,73 +283,7 @@ struct request { }; /* - * first three bits match BIO_RW* bits, important - */ -enum rq_flag_bits { - __REQ_RW, /* not set, read. set, write */ - __REQ_FAILFAST, /* no low level driver retries */ - __REQ_SORTED, /* elevator knows about this request */ - __REQ_SOFTBARRIER, /* may not be passed by ioscheduler */ - __REQ_HARDBARRIER, /* may not be passed by drive either */ - __REQ_FUA, /* forced unit access */ - __REQ_CMD, /* is a regular fs rw request */ - __REQ_NOMERGE, /* don't touch this for merging */ - __REQ_STARTED, /* drive already may have started this one */ - __REQ_DONTPREP, /* don't call prep for this one */ - __REQ_QUEUED, /* uses queueing */ - __REQ_ELVPRIV, /* elevator private data attached */ - /* - * for ATA/ATAPI devices - */ - __REQ_PC, /* packet command (special) */ - __REQ_BLOCK_PC, /* queued down pc from block layer */ - __REQ_SENSE, /* sense retrival */ - - __REQ_FAILED, /* set if the request failed */ - __REQ_QUIET, /* don't worry about errors */ - __REQ_SPECIAL, /* driver suplied command */ - __REQ_DRIVE_CMD, - __REQ_DRIVE_TASK, - __REQ_DRIVE_TASKFILE, - __REQ_PREEMPT, /* set for "ide_preempt" requests */ - __REQ_PM_SUSPEND, /* suspend request */ - __REQ_PM_RESUME, /* resume request */ - __REQ_PM_SHUTDOWN, /* shutdown request */ - __REQ_ORDERED_COLOR, /* is before or after barrier */ - __REQ_RW_SYNC, /* request is sync (O_DIRECT) */ - __REQ_NR_BITS, /* stops here */ -}; - -#define REQ_RW (1 << __REQ_RW) -#define REQ_FAILFAST (1 << __REQ_FAILFAST) -#define REQ_SORTED (1 << __REQ_SORTED) -#define REQ_SOFTBARRIER (1 << __REQ_SOFTBARRIER) -#define REQ_HARDBARRIER (1 << __REQ_HARDBARRIER) -#define REQ_FUA (1 << __REQ_FUA) -#define REQ_CMD (1 << __REQ_CMD) -#define REQ_NOMERGE (1 << __REQ_NOMERGE) -#define REQ_STARTED (1 << __REQ_STARTED) -#define REQ_DONTPREP (1 << __REQ_DONTPREP) -#define REQ_QUEUED (1 << __REQ_QUEUED) -#define REQ_ELVPRIV (1 << __REQ_ELVPRIV) -#define REQ_PC (1 << __REQ_PC) -#define REQ_BLOCK_PC (1 << __REQ_BLOCK_PC) -#define REQ_SENSE (1 << __REQ_SENSE) -#define REQ_FAILED (1 << __REQ_FAILED) -#define REQ_QUIET (1 << __REQ_QUIET) -#define REQ_SPECIAL (1 << __REQ_SPECIAL) -#define REQ_DRIVE_CMD (1 << __REQ_DRIVE_CMD) -#define REQ_DRIVE_TASK (1 << __REQ_DRIVE_TASK) -#define REQ_DRIVE_TASKFILE (1 << __REQ_DRIVE_TASKFILE) -#define REQ_PREEMPT (1 << __REQ_PREEMPT) -#define REQ_PM_SUSPEND (1 << __REQ_PM_SUSPEND) -#define REQ_PM_RESUME (1 << __REQ_PM_RESUME) -#define REQ_PM_SHUTDOWN (1 << __REQ_PM_SHUTDOWN) -#define REQ_ORDERED_COLOR (1 << __REQ_ORDERED_COLOR) -#define REQ_RW_SYNC (1 << __REQ_RW_SYNC) - -/* - * State information carried for REQ_PM_SUSPEND and REQ_PM_RESUME + * State information carried for REQ_TYPE_PM_SUSPEND and REQ_TYPE_PM_RESUME * requests. Some step values could eventually be made generic. */ struct request_pm_state @@ -490,25 +505,28 @@ enum { #define blk_queue_stopped(q) test_bit(QUEUE_FLAG_STOPPED, &(q)->queue_flags) #define blk_queue_flushing(q) ((q)->ordseq) -#define blk_fs_request(rq) ((rq)->flags & REQ_CMD) -#define blk_pc_request(rq) ((rq)->flags & REQ_BLOCK_PC) -#define blk_noretry_request(rq) ((rq)->flags & REQ_FAILFAST) -#define blk_rq_started(rq) ((rq)->flags & REQ_STARTED) +#define blk_fs_request(rq) ((rq)->cmd_type == REQ_TYPE_FS) +#define blk_pc_request(rq) ((rq)->cmd_type == REQ_TYPE_BLOCK_PC) +#define blk_special_request(rq) ((rq)->cmd_type == REQ_TYPE_SPECIAL) +#define blk_sense_request(rq) ((rq)->cmd_type == REQ_TYPE_SENSE) + +#define blk_noretry_request(rq) ((rq)->cmd_flags & REQ_FAILFAST) +#define blk_rq_started(rq) ((rq)->cmd_flags & REQ_STARTED) #define blk_account_rq(rq) (blk_rq_started(rq) && blk_fs_request(rq)) -#define blk_pm_suspend_request(rq) ((rq)->flags & REQ_PM_SUSPEND) -#define blk_pm_resume_request(rq) ((rq)->flags & REQ_PM_RESUME) +#define blk_pm_suspend_request(rq) ((rq)->cmd_type == REQ_TYPE_PM_SUSPEND) +#define blk_pm_resume_request(rq) ((rq)->cmd_type == REQ_TYPE_PM_RESUME) #define blk_pm_request(rq) \ - ((rq)->flags & (REQ_PM_SUSPEND | REQ_PM_RESUME)) + (blk_pm_suspend_request(rq) || blk_pm_resume_request(rq)) -#define blk_sorted_rq(rq) ((rq)->flags & REQ_SORTED) -#define blk_barrier_rq(rq) ((rq)->flags & REQ_HARDBARRIER) -#define blk_fua_rq(rq) ((rq)->flags & REQ_FUA) +#define blk_sorted_rq(rq) ((rq)->cmd_flags & REQ_SORTED) +#define blk_barrier_rq(rq) ((rq)->cmd_flags & REQ_HARDBARRIER) +#define blk_fua_rq(rq) ((rq)->cmd_flags & REQ_FUA) #define list_entry_rq(ptr) list_entry((ptr), struct request, queuelist) -#define rq_data_dir(rq) ((rq)->flags & 1) +#define rq_data_dir(rq) ((rq)->cmd_flags & 1) static inline int blk_queue_full(struct request_queue *q, int rw) { @@ -541,7 +559,7 @@ static inline void blk_clear_queue_full(struct request_queue *q, int rw) #define RQ_NOMERGE_FLAGS \ (REQ_NOMERGE | REQ_STARTED | REQ_HARDBARRIER | REQ_SOFTBARRIER) #define rq_mergeable(rq) \ - (!((rq)->flags & RQ_NOMERGE_FLAGS) && blk_fs_request((rq))) + (!((rq)->cmd_flags & RQ_NOMERGE_FLAGS) && blk_fs_request((rq))) /* * noop, requests are automagically marked as active/inactive by I/O @@ -737,7 +755,7 @@ extern void blk_put_queue(request_queue_t *); */ #define blk_queue_tag_depth(q) ((q)->queue_tags->busy) #define blk_queue_tag_queue(q) ((q)->queue_tags->busy < (q)->queue_tags->max_depth) -#define blk_rq_tagged(rq) ((rq)->flags & REQ_QUEUED) +#define blk_rq_tagged(rq) ((rq)->cmd_flags & REQ_QUEUED) extern int blk_queue_start_tag(request_queue_t *, struct request *); extern struct request *blk_queue_find_tag(request_queue_t *, int); extern void blk_queue_end_tag(request_queue_t *, struct request *); -- cgit From 9817064b68fef7e4580c6df1ea597e106b9ff88b Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Fri, 28 Jul 2006 09:23:08 +0200 Subject: [PATCH] elevator: move the backmerging logic into the elevator core 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 --- include/linux/blkdev.h | 17 ++--------------- 1 file changed, 2 insertions(+), 15 deletions(-) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index b2a412cf468f..8f5486964671 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -229,6 +229,8 @@ struct request { struct bio *bio; struct bio *biotail; + struct hlist_node hash; /* merge hash */ + void *elevator_private; void *completion_data; @@ -696,21 +698,6 @@ static inline void blkdev_dequeue_request(struct request *req) elv_dequeue_request(req->q, req); } -/* - * This should be in elevator.h, but that requires pulling in rq and q - */ -static inline void elv_dispatch_add_tail(struct request_queue *q, - struct request *rq) -{ - if (q->last_merge == rq) - q->last_merge = NULL; - q->nr_sorted--; - - q->end_sector = rq_end_sector(rq); - q->boundary_rq = rq; - list_add_tail(&rq->queuelist, &q->queue_head); -} - /* * Access functions for manipulating queue properties */ -- cgit From 2e662b65f05d550b6799ed6bfa9963b82279e6b7 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 13 Jul 2006 11:55:04 +0200 Subject: [PATCH] elevator: abstract out the rbtree sort handling 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 --- include/linux/blkdev.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 8f5486964671..a905c4934a55 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -230,6 +230,7 @@ struct request { struct bio *biotail; struct hlist_node hash; /* merge hash */ + struct rb_node rb_node; /* sort/lookup */ void *elevator_private; void *completion_data; -- cgit From 9e2585a8a23f3a42f815b2a638725d85a921cd65 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Fri, 28 Jul 2006 09:26:13 +0200 Subject: [PATCH] as-iosched: remove arq->is_sync member We can track this in struct request. Signed-off-by: Jens Axboe Signed-off-by: Nick Piggin --- include/linux/blkdev.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index a905c4934a55..55ef6efe3eb5 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -531,6 +531,11 @@ enum { #define rq_data_dir(rq) ((rq)->cmd_flags & 1) +/* + * We regard a request as sync, if it's a READ or a SYNC write. + */ +#define rq_is_sync(rq) (rq_data_dir((rq)) == READ || (rq)->cmd_flags & REQ_RW_SYNC) + static inline int blk_queue_full(struct request_queue *q, int rw) { if (rw == READ) -- cgit From ff7d145fd911266ae42af7552edc32681c01addb Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Wed, 12 Jul 2006 14:04:37 +0200 Subject: [PATCH] Add one more pointer to struct request for IO scheduler usage Then we have enough room in the request to get rid of the dynamic allocations in CFQ/AS. Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 55ef6efe3eb5..d2dc17151f6c 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -232,7 +232,13 @@ struct request { struct hlist_node hash; /* merge hash */ struct rb_node rb_node; /* sort/lookup */ + /* + * two pointers are available for the IO schedulers, if they need + * more they have to dynamically allocate it. + */ void *elevator_private; + void *elevator_private2; + void *completion_data; int rq_status; /* should split this into a few status bits */ -- cgit From c00895ab2f08df7044e58ee01c38bf0a661ea0eb Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Sat, 30 Sep 2006 20:29:12 +0200 Subject: [PATCH] Remove ->waiting member from struct request 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 --- include/linux/blkdev.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index d2dc17151f6c..604f23189097 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -266,7 +266,6 @@ struct request { request_queue_t *q; struct request_list *rl; - struct completion *waiting; void *special; char *buffer; @@ -285,7 +284,7 @@ struct request { int retries; /* - * completion callback. end_io_data should be folded in with waiting + * completion callback. */ rq_end_io_fn *end_io; void *end_io_data; -- cgit From 49171e5c6f414d49a061b5c1c84967c2eb569822 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 10 Aug 2006 08:59:11 +0200 Subject: [PATCH] Remove struct request_list from struct request It is always identical to &q->rq, and we only use it for detecting whether this request came out of our mempool or not. So replace it with an additional ->flags bit flag. Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 604f23189097..d4c1dd046e27 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -180,6 +180,7 @@ enum rq_flag_bits { __REQ_PREEMPT, /* set for "ide_preempt" requests */ __REQ_ORDERED_COLOR, /* is before or after barrier */ __REQ_RW_SYNC, /* request is sync (O_DIRECT) */ + __REQ_ALLOCED, /* request came from our alloc pool */ __REQ_NR_BITS, /* stops here */ }; @@ -199,6 +200,7 @@ enum rq_flag_bits { #define REQ_PREEMPT (1 << __REQ_PREEMPT) #define REQ_ORDERED_COLOR (1 << __REQ_ORDERED_COLOR) #define REQ_RW_SYNC (1 << __REQ_RW_SYNC) +#define REQ_ALLOCED (1 << __REQ_ALLOCED) #define BLK_MAX_CDB 16 @@ -264,7 +266,6 @@ struct request { int ref_count; request_queue_t *q; - struct request_list *rl; void *special; char *buffer; -- cgit From cdd6026217c0e4cda2efce1bdc318661bef1f66f Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Fri, 28 Jul 2006 09:32:07 +0200 Subject: [PATCH] Remove ->rq_status from struct request After Christophs SCSI change, the only usage left is RQ_ACTIVE and RQ_INACTIVE. The block layer sets RQ_INACTIVE right before freeing the request, so any check for RQ_INACTIVE in a driver is a bug and indicates use-after-free. So kill/clean the remaining users, straight forward. Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index d4c1dd046e27..8a3e309e0842 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -243,8 +243,6 @@ struct request { void *completion_data; - int rq_status; /* should split this into a few status bits */ - int errors; struct gendisk *rq_disk; unsigned long start_time; @@ -262,14 +260,16 @@ struct request { unsigned short ioprio; - int tag; - - int ref_count; request_queue_t *q; void *special; char *buffer; + int tag; + int errors; + + int ref_count; + /* * when request is used as a packet command carrier */ @@ -456,9 +456,6 @@ struct request_queue struct mutex sysfs_lock; }; -#define RQ_INACTIVE (-1) -#define RQ_ACTIVE 1 - #define QUEUE_FLAG_CLUSTER 0 /* cluster several segments into 1 */ #define QUEUE_FLAG_QUEUED 1 /* uses generic tag queueing */ #define QUEUE_FLAG_STOPPED 2 /* queue is stopped */ -- cgit From e6a1c874a064e7d07f24986aba7cd537b7f4a25d Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 10 Aug 2006 09:00:21 +0200 Subject: [PATCH] struct request: shrink and optimize some more Move some members around and unionize completion_data and rb_node since they cannot ever be used at the same time. Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 20 +++++++++++++------- 1 file changed, 13 insertions(+), 7 deletions(-) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 8a3e309e0842..a1e288069e2e 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -211,6 +211,8 @@ struct request { struct list_head queuelist; struct list_head donelist; + request_queue_t *q; + unsigned int cmd_flags; enum rq_cmd_type_bits cmd_type; @@ -219,12 +221,12 @@ struct request { */ sector_t sector; /* next sector to submit */ + sector_t hard_sector; /* next sector to complete */ unsigned long nr_sectors; /* no. of sectors left to submit */ + unsigned long hard_nr_sectors; /* no. of sectors left to complete */ /* no. of sectors left to submit in the current segment */ unsigned int current_nr_sectors; - sector_t hard_sector; /* next sector to complete */ - unsigned long hard_nr_sectors; /* no. of sectors left to complete */ /* no. of sectors left to complete in the current segment */ unsigned int hard_cur_sectors; @@ -232,7 +234,15 @@ struct request { struct bio *biotail; struct hlist_node hash; /* merge hash */ - struct rb_node rb_node; /* sort/lookup */ + /* + * The rb_node is only used inside the io scheduler, requests + * are pruned when moved to the dispatch queue. So let the + * completion_data share space with the rb_node. + */ + union { + struct rb_node rb_node; /* sort/lookup */ + void *completion_data; + }; /* * two pointers are available for the IO schedulers, if they need @@ -241,8 +251,6 @@ struct request { void *elevator_private; void *elevator_private2; - void *completion_data; - struct gendisk *rq_disk; unsigned long start_time; @@ -260,8 +268,6 @@ struct request { unsigned short ioprio; - request_queue_t *q; - void *special; char *buffer; -- cgit From fc46379daf90dce57bf765c81d3b39f55150aac2 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 29 Aug 2006 09:05:44 +0200 Subject: [PATCH] cfq-iosched: kill cfq_exit_lock cfq_exit_lock is protecting two things now: - The per-ioc rbtree of cfq_io_contexts - The per-cfqd linked list of cfq_io_contexts The per-cfqd linked list can be protected by the queue lock, as it is (by definition) per cfqd as the queue lock is. The per-ioc rbtree is mainly used and updated by the process itself only. The only outside use is the io priority changing. If we move the priority changing to not browsing the rbtree, we can remove any locking from the rbtree updates and lookup completely. Let the sys_ioprio syscall just mark processes as having the iopriority changed and lazily update the private cfq io contexts the next time io is queued, and we can remove this locking as well. Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index a1e288069e2e..79cb9fa8034a 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -90,7 +90,7 @@ struct io_context { atomic_t refcount; struct task_struct *task; - int (*set_ioprio)(struct io_context *, unsigned int); + unsigned int ioprio_changed; /* * For request batching -- cgit From a3b05e8f58c95dfccbf2c824d0c68e5990571f24 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Fri, 28 Jul 2006 09:36:46 +0200 Subject: [PATCH] Kill various deprecated/unused block layer defines/functions Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 29 ----------------------------- 1 file changed, 29 deletions(-) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 79cb9fa8034a..593386162f47 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -578,12 +578,6 @@ static inline void blk_clear_queue_full(struct request_queue *q, int rw) #define rq_mergeable(rq) \ (!((rq)->cmd_flags & RQ_NOMERGE_FLAGS) && blk_fs_request((rq))) -/* - * noop, requests are automagically marked as active/inactive by I/O - * scheduler -- see elv_next_request - */ -#define blk_queue_headactive(q, head_active) - /* * q->prep_rq_fn return values */ @@ -621,11 +615,6 @@ static inline void blk_queue_bounce(request_queue_t *q, struct bio **bio) if ((rq->bio)) \ for (_bio = (rq)->bio; _bio; _bio = _bio->bi_next) -struct sec_size { - unsigned block_size; - unsigned block_size_bits; -}; - extern int blk_register_queue(struct gendisk *disk); extern void blk_unregister_queue(struct gendisk *disk); extern void register_disk(struct gendisk *dev); @@ -690,16 +679,6 @@ extern void end_that_request_last(struct request *, int); extern void end_request(struct request *req, int uptodate); extern void blk_complete_request(struct request *); -static inline int rq_all_done(struct request *rq, unsigned int nr_bytes) -{ - if (blk_fs_request(rq)) - return (nr_bytes >= (rq->hard_nr_sectors << 9)); - else if (blk_pc_request(rq)) - return nr_bytes >= rq->data_len; - - return 0; -} - /* * end_that_request_first/chunk() takes an uptodate argument. we account * any value <= as an io error. 0 means -EIO for compatability reasons, @@ -807,14 +786,6 @@ static inline int queue_dma_alignment(request_queue_t *q) return retval; } -static inline int bdev_dma_aligment(struct block_device *bdev) -{ - return queue_dma_alignment(bdev_get_queue(bdev)); -} - -#define blk_finished_io(nsects) do { } while (0) -#define blk_started_io(nsects) do { } while (0) - /* assumes size > 256 */ static inline unsigned int blksize_bits(unsigned int size) { -- cgit From b5deef901282628d88c784f4c9d2f0583ec3b355 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Wed, 19 Jul 2006 23:39:40 +0200 Subject: [PATCH] Make sure all block/io scheduler setups are node aware Some were kmalloc_node(), some were still kmalloc(). Change them all to kmalloc_node(). Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 593386162f47..6609371c303e 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -104,8 +104,7 @@ struct io_context { void put_io_context(struct io_context *ioc); void exit_io_context(void); -struct io_context *current_io_context(gfp_t gfp_flags); -struct io_context *get_io_context(gfp_t gfp_flags); +struct io_context *get_io_context(gfp_t gfp_flags, int node); void copy_io_context(struct io_context **pdst, struct io_context **psrc); void swap_io_context(struct io_context **ioc1, struct io_context **ioc2); -- cgit From dc72ef4ae35c2016fb594bcc85ce871376682174 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 20 Jul 2006 14:54:05 +0200 Subject: [PATCH] Add blk_start_queueing() helper CFQ implements this on its own now, but it's really block layer knowledge. Tells a device queue to start dispatching requests to the driver, taking care to unplug if needed. Also fixes the issue where as/cfq will invoke a stopped queue, which we really don't want. Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 6609371c303e..5d8e288db163 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -635,6 +635,7 @@ extern void blk_stop_queue(request_queue_t *q); extern void blk_sync_queue(struct request_queue *q); extern void __blk_stop_queue(request_queue_t *q); extern void blk_run_queue(request_queue_t *); +extern void blk_start_queueing(request_queue_t *); extern void blk_queue_activity_fn(request_queue_t *, activity_fn *, void *); extern int blk_rq_map_user(request_queue_t *, struct request *, void __user *, unsigned int); extern int blk_rq_unmap_user(struct bio *, unsigned int); -- cgit From 5404bc7a87b9949cf61e0174b21f80e73239ab25 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 10 Aug 2006 09:01:02 +0200 Subject: [PATCH] Allow file systems to differentiate between data and meta reads We can use this information for making more intelligent priority decisions, and it will also be useful for blktrace. Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 5d8e288db163..2c01a90998a7 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -180,6 +180,7 @@ enum rq_flag_bits { __REQ_ORDERED_COLOR, /* is before or after barrier */ __REQ_RW_SYNC, /* request is sync (O_DIRECT) */ __REQ_ALLOCED, /* request came from our alloc pool */ + __REQ_RW_META, /* metadata io request */ __REQ_NR_BITS, /* stops here */ }; @@ -200,6 +201,7 @@ enum rq_flag_bits { #define REQ_ORDERED_COLOR (1 << __REQ_ORDERED_COLOR) #define REQ_RW_SYNC (1 << __REQ_RW_SYNC) #define REQ_ALLOCED (1 << __REQ_ALLOCED) +#define REQ_RW_META (1 << __REQ_RW_META) #define BLK_MAX_CDB 16 @@ -543,6 +545,7 @@ enum { * We regard a request as sync, if it's a READ or a SYNC write. */ #define rq_is_sync(rq) (rq_data_dir((rq)) == READ || (rq)->cmd_flags & REQ_RW_SYNC) +#define rq_is_meta(rq) ((rq)->cmd_flags & REQ_RW_META) static inline int blk_queue_full(struct request_queue *q, int rw) { -- cgit From 9361401eb7619c033e2394e4f9f6d410d6719ac7 Mon Sep 17 00:00:00 2001 From: David Howells Date: Sat, 30 Sep 2006 20:45:40 +0200 Subject: [PATCH] BLOCK: Make it possible to disable the block layer [try #6] Make it possible to disable the block layer. Not all embedded devices require it, some can make do with just JFFS2, NFS, ramfs, etc - none of which require the block layer to be present. This patch does the following: (*) Introduces CONFIG_BLOCK to disable the block layer, buffering and blockdev support. (*) Adds dependencies on CONFIG_BLOCK to any configuration item that controls an item that uses the block layer. This includes: (*) Block I/O tracing. (*) Disk partition code. (*) All filesystems that are block based, eg: Ext3, ReiserFS, ISOFS. (*) The SCSI layer. As far as I can tell, even SCSI chardevs use the block layer to do scheduling. Some drivers that use SCSI facilities - such as USB storage - end up disabled indirectly from this. (*) Various block-based device drivers, such as IDE and the old CDROM drivers. (*) MTD blockdev handling and FTL. (*) JFFS - which uses set_bdev_super(), something it could avoid doing by taking a leaf out of JFFS2's book. (*) Makes most of the contents of linux/blkdev.h, linux/buffer_head.h and linux/elevator.h contingent on CONFIG_BLOCK being set. sector_div() is, however, still used in places, and so is still available. (*) Also made contingent are the contents of linux/mpage.h, linux/genhd.h and parts of linux/fs.h. (*) Makes a number of files in fs/ contingent on CONFIG_BLOCK. (*) Makes mm/bounce.c (bounce buffering) contingent on CONFIG_BLOCK. (*) set_page_dirty() doesn't call __set_page_dirty_buffers() if CONFIG_BLOCK is not enabled. (*) fs/no-block.c is created to hold out-of-line stubs and things that are required when CONFIG_BLOCK is not set: (*) Default blockdev file operations (to give error ENODEV on opening). (*) Makes some /proc changes: (*) /proc/devices does not list any blockdevs. (*) /proc/diskstats and /proc/partitions are contingent on CONFIG_BLOCK. (*) Makes some compat ioctl handling contingent on CONFIG_BLOCK. (*) If CONFIG_BLOCK is not defined, makes sys_quotactl() return -ENODEV if given command other than Q_SYNC or if a special device is specified. (*) In init/do_mounts.c, no reference is made to the blockdev routines if CONFIG_BLOCK is not defined. This does not prohibit NFS roots or JFFS2. (*) The bdflush, ioprio_set and ioprio_get syscalls can now be absent (return error ENOSYS by way of cond_syscall if so). (*) The seclvl_bd_claim() and seclvl_bd_release() security calls do nothing if CONFIG_BLOCK is not set, since they can't then happen. Signed-Off-By: David Howells Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 50 ++++++++++++++++++++++++++++++++++++-------------- 1 file changed, 36 insertions(+), 14 deletions(-) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 2c01a90998a7..3e36107d342a 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -16,6 +16,22 @@ #include +#ifdef CONFIG_LBD +# include +# define sector_div(a, b) do_div(a, b) +#else +# define sector_div(n, b)( \ +{ \ + int _res; \ + _res = (n) % (b); \ + (n) /= (b); \ + _res; \ +} \ +) +#endif + +#ifdef CONFIG_BLOCK + struct scsi_ioctl_command; struct request_queue; @@ -818,24 +834,30 @@ struct work_struct; int kblockd_schedule_work(struct work_struct *work); void kblockd_flush(void); -#ifdef CONFIG_LBD -# include -# define sector_div(a, b) do_div(a, b) -#else -# define sector_div(n, b)( \ -{ \ - int _res; \ - _res = (n) % (b); \ - (n) /= (b); \ - _res; \ -} \ -) -#endif - #define MODULE_ALIAS_BLOCKDEV(major,minor) \ MODULE_ALIAS("block-major-" __stringify(major) "-" __stringify(minor)) #define MODULE_ALIAS_BLOCKDEV_MAJOR(major) \ MODULE_ALIAS("block-major-" __stringify(major) "-*") +#else /* CONFIG_BLOCK */ +/* + * stubs for when the block layer is configured out + */ +#define buffer_heads_over_limit 0 + +static inline long blk_congestion_wait(int rw, long timeout) +{ + return timeout; +} + +static inline long nr_blockdev_pages(void) +{ + return 0; +} + +static inline void exit_io_context(void) {} + +#endif /* CONFIG_BLOCK */ + #endif -- cgit From bcfd8d36151e531e1c6c731f1fbf792509a1c494 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 31 Aug 2006 12:56:06 +0200 Subject: [PATCH] CONFIG_BLOCK: blk_congestion_wait() fix Don't just do nothing: it'll cause busywaits all over writeback and page reclaim. For now, take a fixed-length nap. Will improve when NFS starts waking up throttled processes. Signed-off-by: Andrew Morton Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'include/linux/blkdev.h') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 3e36107d342a..1d79b8d4ca6d 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -1,6 +1,7 @@ #ifndef _LINUX_BLKDEV_H #define _LINUX_BLKDEV_H +#include #include #include #include @@ -848,7 +849,7 @@ void kblockd_flush(void); static inline long blk_congestion_wait(int rw, long timeout) { - return timeout; + return io_schedule_timeout(timeout); } static inline long nr_blockdev_pages(void) @@ -856,7 +857,9 @@ static inline long nr_blockdev_pages(void) return 0; } -static inline void exit_io_context(void) {} +static inline void exit_io_context(void) +{ +} #endif /* CONFIG_BLOCK */ -- cgit