diff options
| author | Linus Torvalds <torvalds@linux-foundation.org> | 2018-08-14 19:23:25 +0200 |
|---|---|---|
| committer | Linus Torvalds <torvalds@linux-foundation.org> | 2018-08-14 19:23:25 +0200 |
| commit | 73ba2fb33c492916853dfe63e3b3163da0be661d (patch) | |
| tree | c2fda8ca1273744d2e884d24189a15ac1a7d63c2 | |
| parent | Merge branch 'l1tf-final' of git://git.kernel.org/pub/scm/linux/kernel/git/ti... (diff) | |
| parent | blkcg: Make blkg_root_lookup() work for queues in bypass mode (diff) | |
| download | linux-73ba2fb33c492916853dfe63e3b3163da0be661d.tar.xz linux-73ba2fb33c492916853dfe63e3b3163da0be661d.zip | |
Merge tag 'for-4.19/block-20180812' of git://git.kernel.dk/linux-block
Pull block updates from Jens Axboe:
"First pull request for this merge window, there will also be a
followup request with some stragglers.
This pull request contains:
- Fix for a thundering heard issue in the wbt block code (Anchal
Agarwal)
- A few NVMe pull requests:
* Improved tracepoints (Keith)
* Larger inline data support for RDMA (Steve Wise)
* RDMA setup/teardown fixes (Sagi)
* Effects log suppor for NVMe target (Chaitanya Kulkarni)
* Buffered IO suppor for NVMe target (Chaitanya Kulkarni)
* TP4004 (ANA) support (Christoph)
* Various NVMe fixes
- Block io-latency controller support. Much needed support for
properly containing block devices. (Josef)
- Series improving how we handle sense information on the stack
(Kees)
- Lightnvm fixes and updates/improvements (Mathias/Javier et al)
- Zoned device support for null_blk (Matias)
- AIX partition fixes (Mauricio Faria de Oliveira)
- DIF checksum code made generic (Max Gurtovoy)
- Add support for discard in iostats (Michael Callahan / Tejun)
- Set of updates for BFQ (Paolo)
- Removal of async write support for bsg (Christoph)
- Bio page dirtying and clone fixups (Christoph)
- Set of bcache fix/changes (via Coly)
- Series improving blk-mq queue setup/teardown speed (Ming)
- Series improving merging performance on blk-mq (Ming)
- Lots of other fixes and cleanups from a slew of folks"
* tag 'for-4.19/block-20180812' of git://git.kernel.dk/linux-block: (190 commits)
blkcg: Make blkg_root_lookup() work for queues in bypass mode
bcache: fix error setting writeback_rate through sysfs interface
null_blk: add lock drop/acquire annotation
Blk-throttle: reduce tail io latency when iops limit is enforced
block: paride: pd: mark expected switch fall-throughs
block: Ensure that a request queue is dissociated from the cgroup controller
block: Introduce blk_exit_queue()
blkcg: Introduce blkg_root_lookup()
block: Remove two superfluous #include directives
blk-mq: count the hctx as active before allocating tag
block: bvec_nr_vecs() returns value for wrong slab
bcache: trivial - remove tailing backslash in macro BTREE_FLAG
bcache: make the pr_err statement used for ENOENT only in sysfs_attatch section
bcache: set max writeback rate when I/O request is idle
bcache: add code comments for bset.c
bcache: fix mistaken comments in request.c
bcache: fix mistaken code comments in bcache.h
bcache: add a comment in super.c
bcache: avoid unncessary cache prefetch bch_btree_node_get()
bcache: display rate debug parameters to 0 when writeback is not running
...
172 files changed, 6029 insertions, 2659 deletions
diff --git a/Documentation/ABI/testing/procfs-diskstats b/Documentation/ABI/testing/procfs-diskstats index f91a973a37fe..abac31d216de 100644 --- a/Documentation/ABI/testing/procfs-diskstats +++ b/Documentation/ABI/testing/procfs-diskstats @@ -5,6 +5,7 @@ Description: The /proc/diskstats file displays the I/O statistics of block devices. Each line contains the following 14 fields: + 1 - major number 2 - minor mumber 3 - device name @@ -19,4 +20,13 @@ Description: 12 - I/Os currently in progress 13 - time spent doing I/Os (ms) 14 - weighted time spent doing I/Os (ms) + + Kernel 4.18+ appends four more fields for discard + tracking putting the total at 18: + + 15 - discards completed successfully + 16 - discards merged + 17 - sectors discarded + 18 - time spent discarding + For more details refer to Documentation/iostats.txt diff --git a/Documentation/admin-guide/cgroup-v2.rst b/Documentation/admin-guide/cgroup-v2.rst index 8a2c52d5c53b..1746131bc9cb 100644 --- a/Documentation/admin-guide/cgroup-v2.rst +++ b/Documentation/admin-guide/cgroup-v2.rst @@ -51,6 +51,9 @@ v1 is available under Documentation/cgroup-v1/. 5-3. IO 5-3-1. IO Interface Files 5-3-2. Writeback + 5-3-3. IO Latency + 5-3-3-1. How IO Latency Throttling Works + 5-3-3-2. IO Latency Interface Files 5-4. PID 5-4-1. PID Interface Files 5-5. Device @@ -1314,17 +1317,19 @@ IO Interface Files Lines are keyed by $MAJ:$MIN device numbers and not ordered. The following nested keys are defined. - ====== =================== + ====== ===================== rbytes Bytes read wbytes Bytes written rios Number of read IOs wios Number of write IOs - ====== =================== + dbytes Bytes discarded + dios Number of discard IOs + ====== ===================== An example read output follows: - 8:16 rbytes=1459200 wbytes=314773504 rios=192 wios=353 - 8:0 rbytes=90430464 wbytes=299008000 rios=8950 wios=1252 + 8:16 rbytes=1459200 wbytes=314773504 rios=192 wios=353 dbytes=0 dios=0 + 8:0 rbytes=90430464 wbytes=299008000 rios=8950 wios=1252 dbytes=50331648 dios=3021 io.weight A read-write flat-keyed file which exists on non-root cgroups. @@ -1446,6 +1451,85 @@ writeback as follows. vm.dirty[_background]_ratio. +IO Latency +~~~~~~~~~~ + +This is a cgroup v2 controller for IO workload protection. You provide a group +with a latency target, and if the average latency exceeds that target the +controller will throttle any peers that have a lower latency target than the +protected workload. + +The limits are only applied at the peer level in the hierarchy. This means that +in the diagram below, only groups A, B, and C will influence each other, and +groups D and F will influence each other. Group G will influence nobody. + + [root] + / | \ + A B C + / \ | + D F G + + +So the ideal way to configure this is to set io.latency in groups A, B, and C. +Generally you do not want to set a value lower than the latency your device +supports. Experiment to find the value that works best for your workload. +Start at higher than the expected latency for your device and watch the +avg_lat value in io.stat for your workload group to get an idea of the +latency you see during normal operation. Use the avg_lat value as a basis for +your real setting, setting at 10-15% higher than the value in io.stat. + +How IO Latency Throttling Works +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +io.latency is work conserving; so as long as everybody is meeting their latency +target the controller doesn't do anything. Once a group starts missing its +target it begins throttling any peer group that has a higher target than itself. +This throttling takes 2 forms: + +- Queue depth throttling. This is the number of outstanding IO's a group is + allowed to have. We will clamp down relatively quickly, starting at no limit + and going all the way down to 1 IO at a time. + +- Artificial delay induction. There are certain types of IO that cannot be + throttled without possibly adversely affecting higher priority groups. This + includes swapping and metadata IO. These types of IO are allowed to occur + normally, however they are "charged" to the originating group. If the + originating group is being throttled you will see the use_delay and delay + fields in io.stat increase. The delay value is how many microseconds that are + being added to any process that runs in this group. Because this number can + grow quite large if there is a lot of swapping or metadata IO occurring we + limit the individual delay events to 1 second at a time. + +Once the victimized group starts meeting its latency target again it will start +unthrottling any peer groups that were throttled previously. If the victimized +group simply stops doing IO the global counter will unthrottle appropriately. + +IO Latency Interface Files +~~~~~~~~~~~~~~~~~~~~~~~~~~ + + io.latency + This takes a similar format as the other controllers. + + "MAJOR:MINOR target=<target time in microseconds" + + io.stat + If the controller is enabled you will see extra stats in io.stat in + addition to the normal ones. + + depth + This is the current queue depth for the group. + + avg_lat + This is an exponential moving average with a decay rate of 1/exp + bound by the sampling interval. The decay rate interval can be + calculated by multiplying the win value in io.stat by the + corresponding number of samples based on the win value. + + win + The sampling window size in milliseconds. This is the minimum + duration of time between evaluation events. Windows only elapse + with IO activity. Idle periods extend the most recent window. + PID --- diff --git a/Documentation/block/null_blk.txt b/Documentation/block/null_blk.txt index 07f147381f32..ea2dafe49ae8 100644 --- a/Documentation/block/null_blk.txt +++ b/Documentation/block/null_blk.txt @@ -85,3 +85,10 @@ shared_tags=[0/1]: Default: 0 0: Tag set is not shared. 1: Tag set shared between devices for blk-mq. Only makes sense with nr_devices > 1, otherwise there's no tag set to share. + +zoned=[0/1]: Default: 0 + 0: Block device is exposed as a random-access block device. + 1: Block device is exposed as a host-managed zoned block device. + +zone_size=[MB]: Default: 256 + Per zone size when exposed as a zoned block device. Must be a power of two. diff --git a/Documentation/block/stat.txt b/Documentation/block/stat.txt index 0dbc946de2ea..0aace9cc536c 100644 --- a/Documentation/block/stat.txt +++ b/Documentation/block/stat.txt @@ -31,28 +31,32 @@ write ticks milliseconds total wait time for write requests in_flight requests number of I/Os currently in flight io_ticks milliseconds total time this block device has been active time_in_queue milliseconds total wait time for all requests +discard I/Os requests number of discard I/Os processed +discard merges requests number of discard I/Os merged with in-queue I/O +discard sectors sectors number of sectors discarded +discard ticks milliseconds total wait time for discard requests -read I/Os, write I/Os -===================== +read I/Os, write I/Os, discard I/0s +=================================== These values increment when an I/O request completes. -read merges, write merges -========================= +read merges, write merges, discard merges +========================================= These values increment when an I/O request is merged with an already-queued I/O request. -read sectors, write sectors -=========================== +read sectors, write sectors, discard_sectors +============================================ -These values count the number of sectors read from or written to this -block device. The "sectors" in question are the standard UNIX 512-byte -sectors, not any device- or filesystem-specific block size. The -counters are incremented when the I/O completes. +These values count the number of sectors read from, written to, or +discarded from this block device. The "sectors" in question are the +standard UNIX 512-byte sectors, not any device- or filesystem-specific +block size. The counters are incremented when the I/O completes. -read ticks, write ticks -======================= +read ticks, write ticks, discard ticks +====================================== These values count the number of milliseconds that I/O requests have waited on this block device. If there are multiple I/O requests waiting, diff --git a/Documentation/iostats.txt b/Documentation/iostats.txt index 04d394a2e06c..49df45f90e8a 100644 --- a/Documentation/iostats.txt +++ b/Documentation/iostats.txt @@ -31,6 +31,9 @@ Here are examples of these different formats:: 3 0 hda 446216 784926 9550688 4382310 424847 312726 5922052 19310380 0 3376340 23705160 3 1 hda1 35486 38030 38030 38030 + 4.18+ diskstats: + 3 0 hda 446216 784926 9550688 4382310 424847 312726 5922052 19310380 0 3376340 23705160 0 0 0 0 + On 2.4 you might execute ``grep 'hda ' /proc/partitions``. On 2.6+, you have a choice of ``cat /sys/block/hda/stat`` or ``grep 'hda ' /proc/diskstats``. @@ -101,6 +104,18 @@ Field 11 -- weighted # of milliseconds spent doing I/Os last update of this field. This can provide an easy measure of both I/O completion time and the backlog that may be accumulating. +Field 12 -- # of discards completed + This is the total number of discards completed successfully. + +Field 13 -- # of discards merged + See the description of field 2 + +Field 14 -- # of sectors discarded + This is the total number of sectors discarded successfully. + +Field 15 -- # of milliseconds spent discarding + This is the total number of milliseconds spent by all discards (as + measured from __make_request() to end_that_request_last()). To avoid introducing performance bottlenecks, no locks are held while modifying these counters. This implies that minor inaccuracies may be diff --git a/block/Kconfig b/block/Kconfig index eb50fd4977c2..1f2469a0123c 100644 --- a/block/Kconfig +++ b/block/Kconfig @@ -149,6 +149,18 @@ config BLK_WBT dynamically on an algorithm loosely based on CoDel, factoring in the realtime performance of the disk. +config BLK_CGROUP_IOLATENCY + bool "Enable support for latency based cgroup IO protection" + depends on BLK_CGROUP=y + default n + ---help--- + Enabling this option enables the .latency interface for IO throttling. + The IO controller will attempt to maintain average IO latencies below + the configured latency target, throttling anybody with a higher latency + target than the victimized group. + + Note, this is an experimental interface and could be changed someday. + config BLK_WBT_SQ bool "Single queue writeback throttling" default n @@ -177,6 +189,10 @@ config BLK_DEBUG_FS Unless you are building a kernel for a tiny system, you should say Y here. +config BLK_DEBUG_FS_ZONED + bool + default BLK_DEBUG_FS && BLK_DEV_ZONED + config BLK_SED_OPAL bool "Logic for interfacing with Opal enabled SEDs" ---help--- diff --git a/block/Makefile b/block/Makefile index 6a56303b9925..572b33f32c07 100644 --- a/block/Makefile +++ b/block/Makefile @@ -9,7 +9,7 @@ obj-$(CONFIG_BLOCK) := bio.o elevator.o blk-core.o blk-tag.o blk-sysfs.o \ blk-lib.o blk-mq.o blk-mq-tag.o blk-stat.o \ blk-mq-sysfs.o blk-mq-cpumap.o blk-mq-sched.o ioctl.o \ genhd.o partition-generic.o ioprio.o \ - badblocks.o partitions/ + badblocks.o partitions/ blk-rq-qos.o obj-$(CONFIG_BOUNCE) += bounce.o obj-$(CONFIG_BLK_SCSI_REQUEST) += scsi_ioctl.o @@ -17,6 +17,7 @@ obj-$(CONFIG_BLK_DEV_BSG) += bsg.o obj-$(CONFIG_BLK_DEV_BSGLIB) += bsg-lib.o obj-$(CONFIG_BLK_CGROUP) += blk-cgroup.o obj-$(CONFIG_BLK_DEV_THROTTLING) += blk-throttle.o +obj-$(CONFIG_BLK_CGROUP_IOLATENCY) += blk-iolatency.o obj-$(CONFIG_IOSCHED_NOOP) += noop-iosched.o obj-$(CONFIG_IOSCHED_DEADLINE) += deadline-iosched.o obj-$(CONFIG_IOSCHED_CFQ) += cfq-iosched.o @@ -34,4 +35,5 @@ obj-$(CONFIG_BLK_MQ_RDMA) += blk-mq-rdma.o obj-$(CONFIG_BLK_DEV_ZONED) += blk-zoned.o obj-$(CONFIG_BLK_WBT) += blk-wbt.o obj-$(CONFIG_BLK_DEBUG_FS) += blk-mq-debugfs.o +obj-$(CONFIG_BLK_DEBUG_FS_ZONED)+= blk-mq-debugfs-zoned.o obj-$(CONFIG_BLK_SED_OPAL) += sed-opal.o diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c index 495b9ddb3355..41d9036b1822 100644 --- a/block/bfq-iosched.c +++ b/block/bfq-iosched.c @@ -634,7 +634,7 @@ static bool bfq_differentiated_weights(struct bfq_data *bfqd) * The following function returns true if every queue must receive the * same share of the throughput (this condition is used when deciding * whether idling may be disabled, see the comments in the function - * bfq_bfqq_may_idle()). + * bfq_better_to_idle()). * * Such a scenario occurs when: * 1) all active queues have the same weight, @@ -742,8 +742,9 @@ inc_counter: * See the comments to the function bfq_weights_tree_add() for considerations * about overhead. */ -void bfq_weights_tree_remove(struct bfq_data *bfqd, struct bfq_entity *entity, - struct rb_root *root) +void __bfq_weights_tree_remove(struct bfq_data *bfqd, + struct bfq_entity *entity, + struct rb_root *root) { if (!entity->weight_counter) return; @@ -760,6 +761,43 @@ reset_entity_pointer: } /* + * Invoke __bfq_weights_tree_remove on bfqq and all its inactive + * parent entities. + */ +void bfq_weights_tree_remove(struct bfq_data *bfqd, + struct bfq_queue *bfqq) +{ + struct bfq_entity *entity = bfqq->entity.parent; + + __bfq_weights_tree_remove(bfqd, &bfqq->entity, + &bfqd->queue_weights_tree); + + for_each_entity(entity) { + struct bfq_sched_data *sd = entity->my_sched_data; + + if (sd->next_in_service || sd->in_service_entity) { + /* + * entity is still active, because either + * next_in_service or in_service_entity is not + * NULL (see the comments on the definition of + * next_in_service for details on why + * in_service_entity must be checked too). + * + * As a consequence, the weight of entity is + * not to be removed. In addition, if entity + * is active, then its parent entities are + * active as well, and thus their weights are + * not to be removed either. In the end, this + * loop must stop here. + */ + break; + } + __bfq_weights_tree_remove(bfqd, entity, + &bfqd->group_weights_tree); + } +} + +/* * Return expired entry, or NULL to just start from scratch in rbtree. */ static struct request *bfq_check_fifo(struct bfq_queue *bfqq, @@ -1344,18 +1382,30 @@ static bool bfq_bfqq_update_budg_for_activation(struct bfq_data *bfqd, * remain unchanged after such an expiration, and the * following statement therefore assigns to * entity->budget the remaining budget on such an - * expiration. For clarity, entity->service is not - * updated on expiration in any case, and, in normal - * operation, is reset only when bfqq is selected for - * service (see bfq_get_next_queue). + * expiration. */ entity->budget = min_t(unsigned long, bfq_bfqq_budget_left(bfqq), bfqq->max_budget); + /* + * At this point, we have used entity->service to get + * the budget left (needed for updating + * entity->budget). Thus we finally can, and have to, + * reset entity->service. The latter must be reset + * because bfqq would otherwise be charged again for + * the service it has received during its previous + * service slot(s). + */ + entity->service = 0; + return true; } + /* + * We can finally complete expiration, by setting service to 0. + */ + entity->service = 0; entity->budget = max_t(unsigned long, bfqq->max_budget, bfq_serv_to_charge(bfqq->next_rq, bfqq)); bfq_clear_bfqq_non_blocking_wait_rq(bfqq); @@ -3233,11 +3283,21 @@ void bfq_bfqq_expire(struct bfq_data *bfqd, ref = bfqq->ref; __bfq_bfqq_expire(bfqd, bfqq); + if (ref == 1) /* bfqq is gone, no more actions on it */ + return; + /* mark bfqq as waiting a request only if a bic still points to it */ - if (ref > 1 && !bfq_bfqq_busy(bfqq) && + if (!bfq_bfqq_busy(bfqq) && reason != BFQQE_BUDGET_TIMEOUT && - reason != BFQQE_BUDGET_EXHAUSTED) + reason != BFQQE_BUDGET_EXHAUSTED) { bfq_mark_bfqq_non_blocking_wait_rq(bfqq); + /* + * Not setting service to 0, because, if the next rq + * arrives in time, the queue will go on receiving + * service with this same budget (as if it never expired) + */ + } else + entity->service = 0; } /* @@ -3295,7 +3355,7 @@ static bool bfq_may_expire_for_budg_timeout(struct bfq_queue *bfqq) * issues taken into account are not trivial. We discuss these issues * individually while introducing the variables. */ -static bool bfq_bfqq_may_idle(struct bfq_queue *bfqq) +static bool bfq_better_to_idle(struct bfq_queue *bfqq) { struct bfq_data *bfqd = bfqq->bfqd; bool rot_without_queueing = @@ -3528,19 +3588,19 @@ static bool bfq_bfqq_may_idle(struct bfq_queue *bfqq) } /* - * If the in-service queue is empty but the function bfq_bfqq_may_idle + * If the in-service queue is empty but the function bfq_better_to_idle * returns true, then: * 1) the queue must remain in service and cannot be expired, and * 2) the device must be idled to wait for the possible arrival of a new * request for the queue. - * See the comments on the function bfq_bfqq_may_idle for the reasons + * See the comments on the function bfq_better_to_idle for the reasons * why performing device idling is the best choice to boost the throughput - * and preserve service guarantees when bfq_bfqq_may_idle itself + * and preserve service guarantees when bfq_better_to_idle itself * returns true. */ static bool bfq_bfqq_must_idle(struct bfq_queue *bfqq) { - return RB_EMPTY_ROOT(&bfqq->sort_list) && bfq_bfqq_may_idle(bfqq); + return RB_EMPTY_ROOT(&bfqq->sort_list) && bfq_better_to_idle(bfqq); } /* @@ -3559,8 +3619,14 @@ static struct bfq_queue *bfq_select_queue(struct bfq_data *bfqd) bfq_log_bfqq(bfqd, bfqq, "select_queue: already in-service queue"); + /* + * Do not expire bfqq for budget timeout if bfqq may be about + * to enjoy device idling. The reason why, in this case, we + * prevent bfqq from expiring is the same as in the comments + * on the case where bfq_bfqq_must_idle() returns true, in + * bfq_completed_request(). + */ if (bfq_may_expire_for_budg_timeout(bfqq) && - !bfq_bfqq_wait_request(bfqq) && !bfq_bfqq_must_idle(bfqq)) goto expire; @@ -3620,7 +3686,7 @@ check_queue: * may idle after their completion, then keep it anyway. */ if (bfq_bfqq_wait_request(bfqq) || - (bfqq->dispatched != 0 && bfq_bfqq_may_idle(bfqq))) { + (bfqq->dispatched != 0 && bfq_better_to_idle(bfqq))) { bfqq = NULL; goto keep_queue; } @@ -4582,8 +4648,7 @@ static void bfq_completed_request(struct bfq_queue *bfqq, struct bfq_data *bfqd) */ bfqq->budget_timeout = jiffies; - bfq_weights_tree_remove(bfqd, &bfqq->entity, - &bfqd->queue_weights_tree); + bfq_weights_tree_remove(bfqd, bfqq); } now_ns = ktime_get_ns(); @@ -4637,15 +4702,39 @@ static void bfq_completed_request(struct bfq_queue *bfqq, struct bfq_data *bfqd) * or if we want to idle in case it has no pending requests. */ if (bfqd->in_service_queue == bfqq) { - if (bfqq->dispatched == 0 && bfq_bfqq_must_idle(bfqq)) { - bfq_arm_slice_timer(bfqd); + if (bfq_bfqq_must_idle(bfqq)) { + if (bfqq->dispatched == 0) + bfq_arm_slice_timer(bfqd); + /* + * If we get here, we do not expire bfqq, even + * if bfqq was in budget timeout or had no + * more requests (as controlled in the next + * conditional instructions). The reason for + * not expiring bfqq is as follows. + * + * Here bfqq->dispatched > 0 holds, but + * bfq_bfqq_must_idle() returned true. This + * implies that, even if no request arrives + * for bfqq before bfqq->dispatched reaches 0, + * bfqq will, however, not be expired on the + * completion event that causes bfqq->dispatch + * to reach zero. In contrast, on this event, + * bfqq will start enjoying device idling + * (I/O-dispatch plugging). + * + * But, if we expired bfqq here, bfqq would + * not have the chance to enjoy device idling + * when bfqq->dispatched finally reaches + * zero. This would expose bfqq to violation + * of its reserved service guarantees. + */ return; } else if (bfq_may_expire_for_budg_timeout(bfqq)) bfq_bfqq_expire(bfqd, bfqq, false, BFQQE_BUDGET_TIMEOUT); else if (RB_EMPTY_ROOT(&bfqq->sort_list) && (bfqq->dispatched == 0 || - !bfq_bfqq_may_idle(bfqq))) + !bfq_better_to_idle(bfqq))) bfq_bfqq_expire(bfqd, bfqq, false, BFQQE_NO_MORE_REQUESTS); } diff --git a/block/bfq-iosched.h b/block/bfq-iosched.h index 0f712e03b035..a8a2e5aca4d4 100644 --- a/block/bfq-iosched.h +++ b/block/bfq-iosched.h @@ -827,8 +827,11 @@ struct bfq_data *bic_to_bfqd(struct bfq_io_cq *bic); void bfq_pos_tree_add_move(struct bfq_data *bfqd, struct bfq_queue *bfqq); void bfq_weights_tree_add(struct bfq_data *bfqd, struct bfq_entity *entity, struct rb_root *root); -void bfq_weights_tree_remove(struct bfq_data *bfqd, struct bfq_entity *entity, - struct rb_root *root); +void __bfq_weights_tree_remove(struct bfq_data *bfqd, + struct bfq_entity *entity, + struct rb_root *root); +void bfq_weights_tree_remove(struct bfq_data *bfqd, + struct bfq_queue *bfqq); void bfq_bfqq_expire(struct bfq_data *bfqd, struct bfq_queue *bfqq, bool compensate, enum bfqq_expiration reason); void bfq_put_queue(struct bfq_queue *bfqq); diff --git a/block/bfq-wf2q.c b/block/bfq-wf2q.c index 4498c43245e2..dbc07b456059 100644 --- a/block/bfq-wf2q.c +++ b/block/bfq-wf2q.c @@ -499,9 +499,6 @@ static void bfq_active_insert(struct bfq_service_tree *st, if (bfqq) list_add(&bfqq->bfqq_list, &bfqq->bfqd->active_list); #ifdef CONFIG_BFQ_GROUP_IOSCHED - else /* bfq_group */ - bfq_weights_tree_add(bfqd, entity, &bfqd->group_weights_tree); - if (bfqg != bfqd->root_group) bfqg->active_entities++; #endif @@ -601,10 +598,6 @@ static void bfq_active_extract(struct bfq_service_tree *st, if (bfqq) list_del(&bfqq->bfqq_list); #ifdef CONFIG_BFQ_GROUP_IOSCHED - else /* bfq_group */ - bfq_weights_tree_remove(bfqd, entity, - &bfqd->group_weights_tree); - if (bfqg != bfqd->root_group) bfqg->active_entities--; #endif @@ -799,7 +792,7 @@ __bfq_entity_update_weight_prio(struct bfq_service_tree *old_st, if (prev_weight != new_weight) { root = bfqq ? &bfqd->queue_weights_tree : &bfqd->group_weights_tree; - bfq_weights_tree_remove(bfqd, entity, root); + __bfq_weights_tree_remove(bfqd, entity, root); } entity->weight = new_weight; /* @@ -971,7 +964,7 @@ static void bfq_update_fin_time_enqueue(struct bfq_entity *entity, * one of its children receives a new request. * * Basically, this function updates the timestamps of entity and - * inserts entity into its active tree, ater possibly extracting it + * inserts entity into its active tree, after possibly extracting it * from its idle tree. */ static void __bfq_activate_entity(struct bfq_entity *entity, @@ -1015,6 +1008,16 @@ static void __bfq_activate_entity(struct bfq_entity *entity, entity->on_st = true; } +#ifdef BFQ_GROUP_IOSCHED_ENABLED + if (!bfq_entity_to_bfqq(entity)) { /* bfq_group */ + struct bfq_group *bfqg = + container_of(entity, struct bfq_group, entity); + + bfq_weights_tree_add(bfqg->bfqd, entity, + &bfqd->group_weights_tree); + } +#endif + bfq_update_fin_time_enqueue(entity, st, backshifted); } @@ -1542,12 +1545,6 @@ struct bfq_queue *bfq_get_next_queue(struct bfq_data *bfqd) sd->in_service_entity = entity; /* - * Reset the accumulator of the amount of service that - * the entity is about to receive. - */ - entity->service = 0; - - /* * If entity is no longer a candidate for next * service, then it must be extracted from its active * tree, so as to make sure that it won't be @@ -1664,8 +1661,7 @@ void bfq_del_bfqq_busy(struct bfq_data *bfqd, struct bfq_queue *bfqq, bfqd->busy_queues--; if (!bfqq->dispatched) - bfq_weights_tree_remove(bfqd, &bfqq->entity, - &bfqd->queue_weights_tree); + bfq_weights_tree_remove(bfqd, bfqq); if (bfqq->wr_coeff > 1) bfqd->wr_busy_queues--; diff --git a/block/bio-integrity.c b/block/bio-integrity.c index add7c7c85335..67b5fb861a51 100644 --- a/block/bio-integrity.c +++ b/block/bio-integrity.c @@ -160,28 +160,6 @@ int bio_integrity_add_page(struct bio *bio, struct page *page, EXPORT_SYMBOL(bio_integrity_add_page); /** - * bio_integrity_intervals - Return number of integrity intervals for a bio - * @bi: blk_integrity profile for device - * @sectors: Size of the bio in 512-byte sectors - * - * Description: The block layer calculates everything in 512 byte - * sectors but integrity metadata is done in terms of the data integrity - * interval size of the storage device. Convert the block layer sectors - * to the appropriate number of integrity intervals. - */ -static inline unsigned int bio_integrity_intervals(struct blk_integrity *bi, - unsigned int sectors) -{ - return sectors >> (bi->interval_exp - 9); -} - -static inline unsigned int bio_integrity_bytes(struct blk_integrity *bi, - unsigned int sectors) -{ - return bio_integrity_intervals(bi, sectors) * bi->tuple_size; -} - -/** * bio_integrity_process - Process integrity metadata for a bio * @bio: bio to generate/verify integrity metadata for * @proc_iter: iterator to process diff --git a/block/bio.c b/block/bio.c index 047c5dca6d90..b12966e415d3 100644 --- a/block/bio.c +++ b/block/bio.c @@ -28,9 +28,11 @@ #include <linux/mempool.h> #include <linux/workqueue.h> #include <linux/cgroup.h> +#include <linux/blk-cgroup.h> #include <trace/events/block.h> #include "blk.h" +#include "blk-rq-qos.h" /* * Test patch to inline a certain number of bi_io_vec's inside the bio @@ -156,7 +158,7 @@ out: unsigned int bvec_nr_vecs(unsigned short idx) { - return bvec_slabs[idx].nr_vecs; + return bvec_slabs[--idx].nr_vecs; } void bvec_free(mempool_t *pool, struct bio_vec *bv, unsigned int idx) @@ -645,83 +647,6 @@ struct bio *bio_clone_fast(struct bio *bio, gfp_t gfp_mask, struct bio_set *bs) EXPORT_SYMBOL(bio_clone_fast); /** - * bio_clone_bioset - clone a bio - * @bio_src: bio to clone - * @gfp_mask: allocation priority - * @bs: bio_set to allocate from - * - * Clone bio. Caller will own the returned bio, but not the actual data it - * points to. Reference count of returned bio will be one. - */ -struct bio *bio_clone_bioset(struct bio *bio_src, gfp_t gfp_mask, - struct bio_set *bs) -{ - struct bvec_iter iter; - struct bio_vec bv; - struct bio *bio; - - /* - * Pre immutable biovecs, __bio_clone() used to just do a memcpy from - * bio_src->bi_io_vec to bio->bi_io_vec. - * - * We can't do that anymore, because: - * - * - The point of cloning the biovec is to produce a bio with a biovec - * the caller can modify: bi_idx and bi_bvec_done should be 0. - * - * - The original bio could've had more than BIO_MAX_PAGES biovecs; if - * we tried to clone the whole thing bio_alloc_bioset() would fail. - * But the clone should succeed as long as the number of biovecs we - * actually need to allocate is fewer than BIO_MAX_PAGES. - * - * - Lastly, bi_vcnt should not be looked at or relied upon by code - * that does not own the bio - reason being drivers don't use it for - * iterating over the biovec anymore, so expecting it to be kept up - * to date (i.e. for clones that share the parent biovec) is just - * asking for trouble and would force extra work on - * __bio_clone_fast() anyways. - */ - - bio = bio_alloc_bioset(gfp_mask, bio_segments(bio_src), bs); - if (!bio) - return NULL; - bio->bi_disk = bio_src->bi_disk; - bio->bi_opf = bio_src->bi_opf; - bio->bi_write_hint = bio_src->bi_write_hint; - bio->bi_iter.bi_sector = bio_src->bi_iter.bi_sector; - bio->bi_iter.bi_size = bio_src->bi_iter.bi_size; - - switch (bio_op(bio)) { - case REQ_OP_DISCARD: - case REQ_OP_SECURE_ERASE: - case REQ_OP_WRITE_ZEROES: - break; - case REQ_OP_WRITE_SAME: - bio->bi_io_vec[bio->bi_vcnt++] = bio_src->bi_io_vec[0]; - break; - default: - bio_for_each_segment(bv, bio_src, iter) - bio->bi_io_vec[bio->bi_vcnt++] = bv; - break; - } - - if (bio_integrity(bio_src)) { - int ret; - - ret = bio_integrity_clone(bio, bio_src, gfp_mask); - if (ret < 0) { - bio_put(bio); - return NULL; - } - } - - bio_clone_blkcg_association(bio, bio_src); - - return bio; -} -EXPORT_SYMBOL(bio_clone_bioset); - -/** * bio_add_pc_page - attempt to add page to bio * @q: the target queue * @bio: destination bio @@ -1661,10 +1586,8 @@ void bio_set_pages_dirty(struct bio *bio) int i; bio_for_each_segment_all(bvec, bio, i) { - struct page *page = bvec->bv_page; - - if (page && !PageCompound(page)) - set_page_dirty_lock(page); + if (!PageCompound(bvec->bv_page)) + set_page_dirty_lock(bvec->bv_page); } } EXPORT_SYMBOL_GPL(bio_set_pages_dirty); @@ -1674,19 +1597,15 @@ static void bio_release_pages(struct bio *bio) struct bio_vec *bvec; int i; - bio_for_each_segment_all(bvec, bio, i) { - struct page *page = bvec->bv_page; - - if (page) - put_page(page); - } + bio_for_each_segment_all(bvec, bio, i) + put_page(bvec->bv_page); } /* * bio_check_pages_dirty() will check that all the BIO's pages are still dirty. * If they are, then fine. If, however, some pages are clean then they must * have been written out during the direct-IO read. So we take another ref on - * the BIO and the offending pages and re-dirty the pages in process context. + * the BIO and re-dirty the pages in process context. * * It is expected that bio_check_pages_dirty() will wholly own the BIO from * here on. It will run one put_page() against each page and will run one @@ -1704,78 +1623,70 @@ static struct bio *bio_dirty_list; */ static void bio_dirty_fn(struct work_struct *work) { - unsigned long flags; - struct bio *bio; + struct bio *bio, *next; - spin_lock_irqsave(&bio_dirty_lock, flags); - bio = bio_dirty_list; + spin_lock_irq(&bio_dirty_lock); + next = bio_dirty_list; bio_dirty_list = NULL; - spin_unlock_irqrestore(&bio_dirty_lock, flags); + spin_unlock_irq(&bio_dirty_lock); - while (bio) { - struct bio *next = bio->bi_private; + while ((bio = next) != NULL) { + next = bio->bi_private; bio_set_pages_dirty(bio); bio_release_pages(bio); bio_put(bio); - bio = next; } } void bio_check_pages_dirty(struct bio *bio) { struct bio_vec *bvec; - int nr_clean_pages = 0; + unsigned long flags; int i; bio_for_each_segment_all(bvec, bio, i) { - struct page *page = bvec->bv_page; - - if (PageDirty(page) || PageCompound(page)) { - put_page(page); - bvec->bv_page = NULL; - } else { - nr_clean_pages++; - } + if (!PageDirty(bvec->bv_page) && !PageCompound(bvec->bv_page)) + goto defer; } - if (nr_clean_pages) { - unsigned long flags; - - spin_lock_irqsave(&bio_dirty_lock, flags); - bio->bi_private = bio_dirty_list; - bio_dirty_list = bio; - spin_unlock_irqrestore(&bio_dirty_lock, flags); - schedule_work(&bio_dirty_work); - } else { - bio_put(bio); - } + bio_release_pages(bio); + bio_put(bio); + return; +defer: + spin_lock_irqsave(&bio_dirty_lock, flags); + bio->bi_private = bio_dirty_list; + bio_dirty_list = bio; + spin_unlock_irqrestore(&bio_dirty_lock, flags); + schedule_work(&bio_dirty_work); } EXPORT_SYMBOL_GPL(bio_check_pages_dirty); -void generic_start_io_acct(struct request_queue *q, int rw, +void generic_start_io_acct(struct request_queue *q, int op, unsigned long sectors, struct hd_struct *part) { + const int sgrp = op_stat_group(op); int cpu = part_stat_lock(); part_round_stats(q, cpu, part); - part_stat_inc(cpu, part, ios[rw]); - part_stat_add(cpu, part, sectors[rw], sectors); - part_inc_in_flight(q, part, rw); + part_stat_inc(cpu, part, ios[sgrp]); + part_stat_add(cpu, part, sectors[sgrp], sectors); + part_inc_in_flight(q, part, op_is_write(op)); part_stat_unlock(); } EXPORT_SYMBOL(generic_start_io_acct); -void generic_end_io_acct(struct request_queue *q, int rw, +void generic_end_io_acct(struct request_queue *q, int req_op, struct hd_struct *part, unsigned long start_time) { unsigned long duration = jiffies - start_time; + const int sgrp = op_stat_group(req_op); int cpu = part_stat_lock(); - part_stat_add(cpu, part, ticks[rw], duration); + part_stat_add(cpu, part, ticks[sgrp], duration); part_round_stats(q, cpu, part); - part_dec_in_flight(q, part, rw); + part_dec_in_flight(q, part, op_is_write(req_op)); part_stat_unlock(); } @@ -1834,6 +1745,9 @@ again: if (!bio_integrity_endio(bio)) return; + if (bio->bi_disk) + rq_qos_done_bio(bio->bi_disk->queue, bio); + /* * Need to have a real endio function for chained bios, otherwise * various corner cases will break (like stacking block devices that @@ -2042,6 +1956,30 @@ EXPORT_SYMBOL(bioset_init_from_src); #ifdef CONFIG_BLK_CGROUP +#ifdef CONFIG_MEMCG +/** + * bio_associate_blkcg_from_page - associate a bio with the page's blkcg + * @bio: target bio + * @page: the page to lookup the blkcg from + * + * Associate @bio with the blkcg from @page's owning memcg. This works like + * every other associate function wrt references. + */ +int bio_associate_blkcg_from_page(struct bio *bio, struct page *page) +{ + struct cgroup_subsys_state *blkcg_css; + + if (unlikely(bio->bi_css)) + return -EBUSY; + if (!page->mem_cgroup) + return 0; + blkcg_css = cgroup_get_e_css(page->mem_cgroup->css.cgroup, + &io_cgrp_subsys); + bio->bi_css = blkcg_css; + return 0; +} +#endif /* CONFIG_MEMCG */ + /** * bio_associate_blkcg - associate a bio with the specified blkcg * @bio: target bio @@ -2065,6 +2003,24 @@ int bio_associate_blkcg(struct bio *bio, struct cgroup_subsys_state *blkcg_css) EXPORT_SYMBOL_GPL(bio_associate_blkcg); /** + * bio_associate_blkg - associate a bio with the specified blkg + * @bio: target bio + * @blkg: the blkg to associate + * + * Associate @bio with the blkg specified by @blkg. This is the queue specific + * blkcg information associated with the @bio, a reference will be taken on the + * @blkg and will be freed when the bio is freed. + */ +int bio_associate_blkg(struct bio *bio, struct blkcg_gq *blkg) +{ + if (unlikely(bio->bi_blkg)) + return -EBUSY; + blkg_get(blkg); + bio->bi_blkg = blkg; + return 0; +} + +/** * bio_disassociate_task - undo bio_associate_current() * @bio: target bio */ @@ -2078,6 +2034,10 @@ void bio_disassociate_task(struct bio *bio) css_put(bio->bi_css); bio->bi_css = NULL; } + if (bio->bi_blkg) { + blkg_put(bio->bi_blkg); + bio->bi_blkg = NULL; + } } /** diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c index eb85cb87c40f..694595b29b8f 100644 --- a/block/blk-cgroup.c +++ b/block/blk-cgroup.c @@ -27,6 +27,7 @@ #include <linux/atomic.h> #include <linux/ctype.h> #include <linux/blk-cgroup.h> +#include <linux/tracehook.h> #include "blk.h" #define MAX_KEY_LEN 100 @@ -50,6 +51,8 @@ static struct blkcg_policy *blkcg_policy[BLKCG_MAX_POLS]; static LIST_HEAD(all_blkcgs); /* protected by blkcg_pol_mutex */ +static bool blkcg_debug_stats = false; + static bool blkcg_policy_enabled(struct request_queue *q, const struct blkcg_policy *pol) { @@ -564,6 +567,7 @@ u64 __blkg_prfill_rwstat(struct seq_file *sf, struct blkg_policy_data *pd, [BLKG_RWSTAT_WRITE] = "Write", [BLKG_RWSTAT_SYNC] = "Sync", [BLKG_RWSTAT_ASYNC] = "Async", + [BLKG_RWSTAT_DISCARD] = "Discard", }; const char *dname = blkg_dev_name(pd->blkg); u64 v; @@ -577,7 +581,8 @@ u64 __blkg_prfill_rwstat(struct seq_file *sf, struct blkg_policy_data *pd, (unsigned long long)atomic64_read(&rwstat->aux_cnt[i])); v = atomic64_read(&rwstat->aux_cnt[BLKG_RWSTAT_READ]) + - atomic64_read(&rwstat->aux_cnt[BLKG_RWSTAT_WRITE]); + atomic64_read(&rwstat->aux_cnt[BLKG_RWSTAT_WRITE]) + + atomic64_read(&rwstat->aux_cnt[BLKG_RWSTAT_DISCARD]); seq_printf(sf, "%s Total %llu\n", dname, (unsigned long long)v); return v; } @@ -954,30 +959,77 @@ static int blkcg_print_stat(struct seq_file *sf, void *v) hlist_for_each_entry_rcu(blkg, &blkcg->blkg_list, blkcg_node) { const char *dname; + char *buf; struct blkg_rwstat rwstat; - u64 rbytes, wbytes, rios, wios; + u64 rbytes, wbytes, rios, wios, dbytes, dios; + size_t size = seq_get_buf(sf, &buf), off = 0; + int i; + bool has_stats = false; dname = blkg_dev_name(blkg); if (!dname) continue; + /* + * Hooray string manipulation, count is the size written NOT + * INCLUDING THE \0, so size is now count+1 less than what we + * had before, but we want to start writing the next bit from + * the \0 so we only add count to buf. + */ + off += scnprintf(buf+off, size-off, "%s ", dname); + spin_lock_irq(blkg->q->queue_lock); rwstat = blkg_rwstat_recursive_sum(blkg, NULL, offsetof(struct blkcg_gq, stat_bytes)); rbytes = atomic64_read(&rwstat.aux_cnt[BLKG_RWSTAT_READ]); wbytes = atomic64_read(&rwstat.aux_cnt[BLKG_RWSTAT_WRITE]); + dbytes = atomic64_read(&rwstat.aux_cnt[BLKG_RWSTAT_DISCARD]); rwstat = blkg_rwstat_recursive_sum(blkg, NULL, offsetof(struct blkcg_gq, stat_ios)); rios = atomic64_read(&rwstat.aux_cnt[BLKG_RWSTAT_READ]); wios = atomic64_read(&rwstat.aux_cnt[BLKG_RWSTAT_WRITE]); + dios = atomic64_read(&rwstat.aux_cnt[BLKG_RWSTAT_DISCARD]); spin_unlock_irq(blkg->q->queue_lock); - if (rbytes || wbytes || rios || wios) - seq_printf(sf, "%s rbytes=%llu wbytes=%llu rios=%llu wios=%llu\n", - dname, rbytes, wbytes, rios, wios); + if (rbytes || wbytes || rios || wios) { + has_stats = true; + off += scnprintf(buf+off, size-off, + "rbytes=%llu wbytes=%llu rios=%llu wios=%llu dbytes=%llu dios=%llu", + rbytes, wbytes, rios, wios, + dbytes, dios); + } + + if (!blkcg_debug_stats) + goto next; + + if (atomic_read(&blkg->use_delay)) { + has_stats = true; + off += scnprintf(buf+off, size-off, + " use_delay=%d delay_nsec=%llu", + atomic_read(&blkg->use_delay), + (unsigned long long)atomic64_read(&blkg->delay_nsec)); + } + + for (i = 0; i < BLKCG_MAX_POLS; i++) { + struct blkcg_policy *pol = blkcg_policy[i]; + size_t written; + + if (!blkg->pd[i] || !pol->pd_stat_fn) + continue; + + written = pol->pd_stat_fn(blkg->pd[i], buf+off, size-off); + if (written) + has_stats = true; + off += written; + } +next: + if (has_stats) { + off += scnprintf(buf+off, size-off, "\n"); + seq_commit(sf, off); + } } rcu_read_unlock(); @@ -1191,6 +1243,14 @@ int blkcg_init_queue(struct request_queue *q) if (preloaded) radix_tree_preload_end(); + ret = blk_iolatency_init(q); + if (ret) { + spin_lock_irq(q->queue_lock); + blkg_destroy_all(q); + spin_unlock_irq(q->queue_lock); + return ret; + } + ret = blk_throtl_init(q); if (ret) { spin_lock_irq(q->queue_lock); @@ -1288,6 +1348,13 @@ static void blkcg_bind(struct cgroup_subsys_state *root_css) mutex_unlock(&blkcg_pol_mutex); } +static void blkcg_exit(struct task_struct *tsk) +{ + if (tsk->throttle_queue) + blk_put_queue(tsk->throttle_queue); + tsk->throttle_queue = NULL; +} + struct cgroup_subsys io_cgrp_subsys = { .css_alloc = blkcg_css_alloc, .css_offline = blkcg_css_offline, @@ -1297,6 +1364,7 @@ struct cgroup_subsys io_cgrp_subsys = { .dfl_cftypes = blkcg_files, .legacy_cftypes = blkcg_legacy_files, .legacy_name = "blkio", + .exit = blkcg_exit, #ifdef CONFIG_MEMCG /* * This ensures that, if available, memcg is automatically enabled @@ -1547,3 +1615,209 @@ out_unlock: mutex_unlock(&blkcg_pol_register_mutex); } EXPORT_SYMBOL_GPL(blkcg_policy_unregister); + +/* + * Scale the accumulated delay based on how long it has been since we updated + * the delay. We only call this when we are adding delay, in case it's been a + * while since we added delay, and when we are checking to see if we need to + * delay a task, to account for any delays that may have occurred. + */ +static void blkcg_scale_delay(struct blkcg_gq *blkg, u64 now) +{ + u64 old = atomic64_read(&blkg->delay_start); + + /* + * We only want to scale down every second. The idea here is that we + * want to delay people for min(delay_nsec, NSEC_PER_SEC) in a certain + * time window. We only want to throttle tasks for recent delay that + * has occurred, in 1 second time windows since that's the maximum + * things can be throttled. We save the current delay window in + * blkg->last_delay so we know what amount is still left to be charged + * to the blkg from this point onward. blkg->last_use keeps track of + * the use_delay counter. The idea is if we're unthrottling the blkg we + * are ok with whatever is happening now, and we can take away more of + * the accumulated delay as we've already throttled enough that + * everybody is happy with their IO latencies. + */ + if (time_before64(old + NSEC_PER_SEC, now) && + atomic64_cmpxchg(&blkg->delay_start, old, now) == old) { + u64 cur = atomic64_read(&blkg->delay_nsec); + u64 sub = min_t(u64, blkg->last_delay, now - old); + int cur_use = atomic_read(&blkg->use_delay); + + /* + * We've been unthrottled, subtract a larger chunk of our + * accumulated delay. + */ + if (cur_use < blkg->last_use) + sub = max_t(u64, sub, blkg->last_delay >> 1); + + /* + * This shouldn't happen, but handle it anyway. Our delay_nsec + * should only ever be growing except here where we subtract out + * min(last_delay, 1 second), but lord knows bugs happen and I'd + * rather not end up with negative numbers. + */ + if (unlikely(cur < sub)) { + atomic64_set(&blkg->delay_nsec, 0); + blkg->last_delay = 0; + } else { + atomic64_sub(sub, &blkg->delay_nsec); + blkg->last_delay = cur - sub; + } + blkg->last_use = cur_use; + } +} + +/* + * This is called when we want to actually walk up the hierarchy and check to + * see if we need to throttle, and then actually throttle if there is some + * accumulated delay. This should only be called upon return to user space so + * we're not holding some lock that would induce a priority inversion. + */ +static void blkcg_maybe_throttle_blkg(struct blkcg_gq *blkg, bool use_memdelay) +{ + u64 now = ktime_to_ns(ktime_get()); + u64 exp; + u64 delay_nsec = 0; + int tok; + + while (blkg->parent) { + if (atomic_read(&blkg->use_delay)) { + blkcg_scale_delay(blkg, now); + delay_nsec = max_t(u64, delay_nsec, + atomic64_read(&blkg->delay_nsec)); + } + blkg = blkg->parent; + } + + if (!delay_nsec) + return; + + /* + * Let's not sleep for all eternity if we've amassed a huge delay. + * Swapping or metadata IO can accumulate 10's of seconds worth of + * delay, and we want userspace to be able to do _something_ so cap the + * delays at 1 second. If there's 10's of seconds worth of delay then + * the tasks will be delayed for 1 second for every syscall. + */ + delay_nsec = min_t(u64, delay_nsec, 250 * NSEC_PER_MSEC); + + /* + * TODO: the use_memdelay flag is going to be for the upcoming psi stuff + * that hasn't landed upstream yet. Once that stuff is in place we need + * to do a psi_memstall_enter/leave if memdelay is set. + */ + + exp = ktime_add_ns(now, delay_nsec); + tok = io_schedule_prepare(); + do { + __set_current_state(TASK_KILLABLE); + if (!schedule_hrtimeout(&exp, HRTIMER_MODE_ABS)) + break; + } while (!fatal_signal_pending(current)); + io_schedule_finish(tok); +} + +/** + * blkcg_maybe_throttle_current - throttle the current task if it has been marked + * + * This is only called if we've been marked with set_notify_resume(). Obviously + * we can be set_notify_resume() for reasons other than blkcg throttling, so we + * check to see if current->throttle_queue is set and if not this doesn't do + * anything. This should only ever be called by the resume code, it's not meant + * to be called by people willy-nilly as it will actually do the work to + * throttle the task if it is setup for throttling. + */ +void blkcg_maybe_throttle_current(void) +{ + struct request_queue *q = current->throttle_queue; + struct cgroup_subsys_state *css; + struct blkcg *blkcg; + struct blkcg_gq *blkg; + bool use_memdelay = current->use_memdelay; + + if (!q) + return; + + current->throttle_queue = NULL; + current->use_memdelay = false; + + rcu_read_lock(); + css = kthread_blkcg(); + if (css) + blkcg = css_to_blkcg(css); + else + blkcg = css_to_blkcg(task_css(current, io_cgrp_id)); + + if (!blkcg) + goto out; + blkg = blkg_lookup(blkcg, q); + if (!blkg) + goto out; + blkg = blkg_try_get(blkg); + if (!blkg) + goto out; + rcu_read_unlock(); + + blkcg_maybe_throttle_blkg(blkg, use_memdelay); + blkg_put(blkg); + blk_put_queue(q); + return; +out: + rcu_read_unlock(); + blk_put_queue(q); +} +EXPORT_SYMBOL_GPL(blkcg_maybe_throttle_current); + +/** + * blkcg_schedule_throttle - this task needs to check for throttling + * @q - the request queue IO was submitted on + * @use_memdelay - do we charge this to memory delay for PSI + * + * This is called by the IO controller when we know there's delay accumulated + * for the blkg for this task. We do not pass the blkg because there are places + * we call this that may not have that information, the swapping code for + * instance will only have a request_queue at that point. This set's the + * notify_resume for the task to check and see if it requires throttling before + * returning to user space. + * + * We will only schedule once per syscall. You can call this over and over + * again and it will only do the check once upon return to user space, and only + * throttle once. If the task needs to be throttled again it'll need to be + * re-set at the next time we see the task. + */ +void blkcg_schedule_throttle(struct request_queue *q, bool use_memdelay) +{ + if (unlikely(current->flags & PF_KTHREAD)) + return; + + if (!blk_get_queue(q)) + return; + + if (current->throttle_queue) + blk_put_queue(current->throttle_queue); + current->throttle_queue = q; + if (use_memdelay) + current->use_memdelay = use_memdelay; + set_notify_resume(current); +} +EXPORT_SYMBOL_GPL(blkcg_schedule_throttle); + +/** + * blkcg_add_delay - add delay to this blkg + * @now - the current time in nanoseconds + * @delta - how many nanoseconds of delay to add + * + * Charge @delta to the blkg's current delay accumulation. This is used to + * throttle tasks if an IO controller thinks we need more throttling. + */ +void blkcg_add_delay(struct blkcg_gq *blkg, u64 now, u64 delta) +{ + blkcg_scale_delay(blkg, now); + atomic64_add(delta, &blkg->delay_nsec); +} +EXPORT_SYMBOL_GPL(blkcg_add_delay); + +module_param(blkcg_debug_stats, bool, 0644); +MODULE_PARM_DESC(blkcg_debug_stats, "True if you want debug stats, false if not"); diff --git a/block/blk-core.c b/block/blk-core.c index ee33590f54eb..12550340418d 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -42,7 +42,7 @@ #include "blk.h" #include "blk-mq.h" #include "blk-mq-sched.h" -#include "blk-wbt.h" +#include "blk-rq-qos.h" #ifdef CONFIG_DEBUG_FS struct dentry *blk_debugfs_root; @@ -715,6 +715,35 @@ void blk_set_queue_dying(struct request_queue *q) } EXPORT_SYMBOL_GPL(blk_set_queue_dying); +/* Unconfigure the I/O scheduler and dissociate from the cgroup controller. */ +void blk_exit_queue(struct request_queue *q) +{ + /* + * Since the I/O scheduler exit code may access cgroup information, + * perform I/O scheduler exit before disassociating from the block + * cgroup controller. + */ + if (q->elevator) { + ioc_clear_queue(q); + elevator_exit(q, q->elevator); + q->elevator = NULL; + } + + /* + * Remove all references to @q from the block cgroup controller before + * restoring @q->queue_lock to avoid that restoring this pointer causes + * e.g. blkcg_print_blkgs() to crash. + */ + blkcg_exit_queue(q); + + /* + * Since the cgroup code may dereference the @q->backing_dev_info + * pointer, only decrease its reference count after having removed the + * association with the block cgroup controller. + */ + bdi_put(q->backing_dev_info); +} + /** * blk_cleanup_queue - shutdown a request queue * @q: request queue to shutdown @@ -762,9 +791,13 @@ void blk_cleanup_queue(struct request_queue *q) * make sure all in-progress dispatch are completed because * blk_freeze_queue() can only complete all requests, and * dispatch may still be in-progress since we dispatch requests - * from more than one contexts + * from more than one contexts. + * + * No need to quiesce queue if it isn't initialized yet since + * blk_freeze_queue() should be enough for cases of passthrough + * request. */ - if (q->mq_ops) + if (q->mq_ops && blk_queue_init_done(q)) blk_mq_quiesce_queue(q); /* for synchronous bio-based driver finish in-flight integrity i/o */ @@ -780,30 +813,7 @@ void blk_cleanup_queue(struct request_queue *q) */ WARN_ON_ONCE(q->kobj.state_in_sysfs); - /* - * Since the I/O scheduler exit code may access cgroup information, - * perform I/O scheduler exit before disassociating from the block - * cgroup controller. - */ - if (q->elevator) { - ioc_clear_queue(q); - elevator_exit(q, q->elevator); - q->elevator = NULL; - } - - /* - * Remove all references to @q from the block cgroup controller before - * restoring @q->queue_lock to avoid that restoring this pointer causes - * e.g. blkcg_print_blkgs() to crash. - */ - blkcg_exit_queue(q); - - /* - * Since the cgroup code may dereference the @q->backing_dev_info - * pointer, only decrease its reference count after having removed the - * association with the block cgroup controller. - */ - bdi_put(q->backing_dev_info); + blk_exit_queue(q); if (q->mq_ops) blk_mq_free_queue(q); @@ -1180,6 +1190,7 @@ out_exit_flush_rq: q->exit_rq_fn(q, q->fq->flush_rq); out_free_flush_queue: blk_free_flush_queue(q->fq); + q->fq = NULL; return -ENOMEM; } EXPORT_SYMBOL(blk_init_allocated_queue); @@ -1641,7 +1652,7 @@ void blk_requeue_request(struct request_queue *q, struct request *rq) blk_delete_timer(rq); blk_clear_rq_complete(rq); trace_block_rq_requeue(q, rq); - wbt_requeue(q->rq_wb, rq); + rq_qos_requeue(q, rq); if (rq->rq_flags & RQF_QUEUED) blk_queue_end_tag(q, rq); @@ -1748,7 +1759,7 @@ void __blk_put_request(struct request_queue *q, struct request *req) /* this is a bio leak */ WARN_ON(req->bio != NULL); - wbt_done(q->rq_wb, req); + rq_qos_done(q, req); /* * Request may not have originated from ll_rw_blk. if not, @@ -1982,7 +1993,6 @@ static blk_qc_t blk_queue_bio(struct request_queue *q, struct bio *bio) int where = ELEVATOR_INSERT_SORT; struct request *req, *free; unsigned int request_count = 0; - unsigned int wb_acct; /* * low level driver can indicate that it wants pages above a @@ -2040,7 +2050,7 @@ static blk_qc_t blk_queue_bio(struct request_queue *q, struct bio *bio) } get_rq: - wb_acct = wbt_wait(q->rq_wb, bio, q->queue_lock); + rq_qos_throttle(q, bio, q->queue_lock); /* * Grab a free request. This is might sleep but can not fail. @@ -2050,7 +2060,7 @@ get_rq: req = get_request(q, bio->bi_opf, bio, 0, GFP_NOIO); if (IS_ERR(req)) { blk_queue_exit(q); - __wbt_done(q->rq_wb, wb_acct); + rq_qos_cleanup(q, bio); if (PTR_ERR(req) == -ENOMEM) bio->bi_status = BLK_STS_RESOURCE; else @@ -2059,7 +2069,7 @@ get_rq: goto out_unlock; } - wbt_track(req, wb_acct); + rq_qos_track(q, req, bio); /* * After dropping the lock and possibly sleeping here, our request @@ -2700,13 +2710,13 @@ EXPORT_SYMBOL_GPL(blk_rq_err_bytes); void blk_account_io_completion(struct request *req, unsigned int bytes) { if (blk_do_io_stat(req)) { - const int rw = rq_data_dir(req); + const int sgrp = op_stat_group(req_op(req)); struct hd_struct *part; int cpu; cpu = part_stat_lock(); part = req->part; - part_stat_add(cpu, part, sectors[rw], bytes >> 9); + part_stat_add(cpu, part, sectors[sgrp], bytes >> 9); part_stat_unlock(); } } @@ -2720,7 +2730,7 @@ void blk_account_io_done(struct request *req, u64 now) */ if (blk_do_io_stat(req) && !(req->rq_flags & RQF_FLUSH_SEQ)) { unsigned long duration; - const int rw = rq_data_dir(req); + const int sgrp = op_stat_group(req_op(req)); struct hd_struct *part; int cpu; @@ -2728,10 +2738,10 @@ void blk_account_io_done(struct request *req, u64 now) cpu = part_stat_lock(); part = req->part; - part_stat_inc(cpu, part, ios[rw]); - part_stat_add(cpu, part, ticks[rw], duration); + part_stat_inc(cpu, part, ios[sgrp]); + part_stat_add(cpu, part, ticks[sgrp], duration); part_round_stats(req->q, cpu, part); - part_dec_in_flight(req->q, part, rw); + part_dec_in_flight(req->q, part, rq_data_dir(req)); hd_struct_put(part); part_stat_unlock(); @@ -2751,9 +2761,9 @@ static bool blk_pm_allow_request(struct request *rq) return rq->rq_flags & RQF_PM; case RPM_SUSPENDED: return false; + default: + return true; } - - return true; } #else static bool blk_pm_allow_request(struct request *rq) @@ -2980,7 +2990,7 @@ void blk_start_request(struct request *req) req->throtl_size = blk_rq_sectors(req); #endif req->rq_flags |= RQF_STATS; - wbt_issue(req->q->rq_wb, req); + rq_qos_issue(req->q, req); } BUG_ON(blk_rq_is_complete(req)); @@ -3053,6 +3063,10 @@ EXPORT_SYMBOL_GPL(blk_steal_bios); * Passing the result of blk_rq_bytes() as @nr_bytes guarantees * %false return from this function. * + * Note: + * The RQF_SPECIAL_PAYLOAD flag is ignored on purpose in both + * blk_rq_bytes() and in blk_update_request(). + * * Return: * %false - this request doesn't have any more data * %true - this request has more data @@ -3200,7 +3214,7 @@ void blk_finish_request(struct request *req, blk_status_t error) blk_account_io_done(req, now); if (req->end_io) { - wbt_done(req->q->rq_wb, req); + rq_qos_done(q, req); req->end_io(req, error); } else { if (blk_bidi_rq(req)) @@ -3763,9 +3777,11 @@ EXPORT_SYMBOL(blk_finish_plug); */ void blk_pm_runtime_init(struct request_queue *q, struct device *dev) { - /* not support for RQF_PM and ->rpm_status in blk-mq yet */ - if (q->mq_ops) + /* Don't enable runtime PM for blk-mq until it is ready */ + if (q->mq_ops) { + pm_runtime_disable(dev); return; + } q->dev = dev; q->rpm_status = RPM_ACTIVE; diff --git a/block/blk-ioc.c b/block/blk-ioc.c index f23311e4b201..01580f88fcb3 100644 --- a/block/blk-ioc.c +++ b/block/blk-ioc.c @@ -278,7 +278,7 @@ int create_task_io_context(struct task_struct *task, gfp_t gfp_flags, int node) atomic_set(&ioc->nr_tasks, 1); atomic_set(&ioc->active_ref, 1); spin_lock_init(&ioc->lock); - INIT_RADIX_TREE(&ioc->icq_tree, GFP_ATOMIC | __GFP_HIGH); + INIT_RADIX_TREE(&ioc->icq_tree, GFP_ATOMIC); INIT_HLIST_HEAD(&ioc->icq_list); INIT_WORK(&ioc->release_work, ioc_release_fn); diff --git a/block/blk-iolatency.c b/block/blk-iolatency.c new file mode 100644 index 000000000000..19923f8a029d --- /dev/null +++ b/block/blk-iolatency.c @@ -0,0 +1,955 @@ +/* + * Block rq-qos base io controller + * + * This works similar to wbt with a few exceptions + * + * - It's bio based, so the latency covers the whole block layer in addition to + * the actual io. + * - We will throttle all IO that comes in here if we need to. + * - We use the mean latency over the 100ms window. This is because writes can + * be particularly fast, which could give us a false sense of the impact of + * other workloads on our protected workload. + * - By default there's no throttling, we set the queue_depth to UINT_MAX so + * that we can have as many outstanding bio's as we're allowed to. Only at + * throttle time do we pay attention to the actual queue depth. + * + * The hierarchy works like the cpu controller does, we track the latency at + * every configured node, and each configured node has it's own independent + * queue depth. This means that we only care about our latency targets at the + * peer level. Some group at the bottom of the hierarchy isn't going to affect + * a group at the end of some other path if we're only configred at leaf level. + * + * Consider the following + * + * root blkg + * / \ + * fast (target=5ms) slow (target=10ms) + * / \ / \ + * a b normal(15ms) unloved + * + * "a" and "b" have no target, but their combined io under "fast" cannot exceed + * an average latency of 5ms. If it does then we will throttle the "slow" + * group. In the case of "normal", if it exceeds its 15ms target, we will + * throttle "unloved", but nobody else. + * + * In this example "fast", "slow", and "normal" will be the only groups actually + * accounting their io latencies. We have to walk up the heirarchy to the root + * on every submit and complete so we can do the appropriate stat recording and + * adjust the queue depth of ourselves if needed. + * + * There are 2 ways we throttle IO. + * + * 1) Queue depth throttling. As we throttle down we will adjust the maximum + * number of IO's we're allowed to have in flight. This starts at (u64)-1 down + * to 1. If the group is only ever submitting IO for itself then this is the + * only way we throttle. + * + * 2) Induced delay throttling. This is for the case that a group is generating + * IO that has to be issued by the root cg to avoid priority inversion. So think + * REQ_META or REQ_SWAP. If we are already at qd == 1 and we're getting a lot + * of work done for us on behalf of the root cg and are being asked to scale + * down more then we induce a latency at userspace return. We accumulate the + * total amount of time we need to be punished by doing + * + * total_time += min_lat_nsec - actual_io_completion + * + * and then at throttle time will do + * + * throttle_time = min(total_time, NSEC_PER_SEC) + * + * This induced delay will throttle back the activity that is generating the + * root cg issued io's, wethere that's some metadata intensive operation or the + * group is using so much memory that it is pushing us into swap. + * + * Copyright (C) 2018 Josef Bacik + */ +#include <linux/kernel.h> +#include <linux/blk_types.h> +#include <linux/backing-dev.h> +#include <linux/module.h> +#include <linux/timer.h> +#include <linux/memcontrol.h> +#include <linux/sched/loadavg.h> +#include <linux/sched/signal.h> +#include <trace/events/block.h> +#include "blk-rq-qos.h" +#include "blk-stat.h" + +#define DEFAULT_SCALE_COOKIE 1000000U + +static struct blkcg_policy blkcg_policy_iolatency; +struct iolatency_grp; + +struct blk_iolatency { + struct rq_qos rqos; + struct timer_list timer; + atomic_t enabled; +}; + +static inline struct blk_iolatency *BLKIOLATENCY(struct rq_qos *rqos) +{ + return container_of(rqos, struct blk_iolatency, rqos); +} + +static inline bool blk_iolatency_enabled(struct blk_iolatency *blkiolat) +{ + return atomic_read(&blkiolat->enabled) > 0; +} + +struct child_latency_info { + spinlock_t lock; + + /* Last time we adjusted the scale of everybody. */ + u64 last_scale_event; + + /* The latency that we missed. */ + u64 scale_lat; + + /* Total io's from all of our children for the last summation. */ + u64 nr_samples; + + /* The guy who actually changed the latency numbers. */ + struct iolatency_grp *scale_grp; + + /* Cookie to tell if we need to scale up or down. */ + atomic_t scale_cookie; +}; + +struct iolatency_grp { + struct blkg_policy_data pd; + struct blk_rq_stat __percpu *stats; + struct blk_iolatency *blkiolat; + struct rq_depth rq_depth; + struct rq_wait rq_wait; + atomic64_t window_start; + atomic_t scale_cookie; + u64 min_lat_nsec; + u64 cur_win_nsec; + + /* total running average of our io latency. */ + u64 lat_avg; + + /* Our current number of IO's for the last summation. */ + u64 nr_samples; + + struct child_latency_info child_lat; +}; + +#define BLKIOLATENCY_MIN_WIN_SIZE (100 * NSEC_PER_MSEC) +#define BLKIOLATENCY_MAX_WIN_SIZE NSEC_PER_SEC +/* + * These are the constants used to fake the fixed-point moving average + * calculation just like load average. The call to CALC_LOAD folds + * (FIXED_1 (2048) - exp_factor) * new_sample into lat_avg. The sampling + * window size is bucketed to try to approximately calculate average + * latency such that 1/exp (decay rate) is [1 min, 2.5 min) when windows + * elapse immediately. Note, windows only elapse with IO activity. Idle + * periods extend the most recent window. + */ +#define BLKIOLATENCY_NR_EXP_FACTORS 5 +#define BLKIOLATENCY_EXP_BUCKET_SIZE (BLKIOLATENCY_MAX_WIN_SIZE / \ + (BLKIOLATENCY_NR_EXP_FACTORS - 1)) +static const u64 iolatency_exp_factors[BLKIOLATENCY_NR_EXP_FACTORS] = { + 2045, // exp(1/600) - 600 samples + 2039, // exp(1/240) - 240 samples + 2031, // exp(1/120) - 120 samples + 2023, // exp(1/80) - 80 samples + 2014, // exp(1/60) - 60 samples +}; + +static inline struct iolatency_grp *pd_to_lat(struct blkg_policy_data *pd) +{ + return pd ? container_of(pd, struct iolatency_grp, pd) : NULL; +} + +static inline struct iolatency_grp *blkg_to_lat(struct blkcg_gq *blkg) +{ + return pd_to_lat(blkg_to_pd(blkg, &blkcg_policy_iolatency)); +} + +static inline struct blkcg_gq *lat_to_blkg(struct iolatency_grp *iolat) +{ + return pd_to_blkg(&iolat->pd); +} + +static inline bool iolatency_may_queue(struct iolatency_grp *iolat, + wait_queue_entry_t *wait, + bool first_block) +{ + struct rq_wait *rqw = &iolat->rq_wait; + + if (first_block && waitqueue_active(&rqw->wait) && + rqw->wait.head.next != &wait->entry) + return false; + return rq_wait_inc_below(rqw, iolat->rq_depth.max_depth); +} + +static void __blkcg_iolatency_throttle(struct rq_qos *rqos, + struct iolatency_grp *iolat, + spinlock_t *lock, bool issue_as_root, + bool use_memdelay) + __releases(lock) + __acquires(lock) +{ + struct rq_wait *rqw = &iolat->rq_wait; + unsigned use_delay = atomic_read(&lat_to_blkg(iolat)->use_delay); + DEFINE_WAIT(wait); + bool first_block = true; + + if (use_delay) + blkcg_schedule_throttle(rqos->q, use_memdelay); + + /* + * To avoid priority inversions we want to just take a slot if we are + * issuing as root. If we're being killed off there's no point in + * delaying things, we may have been killed by OOM so throttling may + * make recovery take even longer, so just let the IO's through so the + * task can go away. + */ + if (issue_as_root || fatal_signal_pending(current)) { + atomic_inc(&rqw->inflight); + return; + } + + if (iolatency_may_queue(iolat, &wait, first_block)) + return; + + do { + prepare_to_wait_exclusive(&rqw->wait, &wait, + TASK_UNINTERRUPTIBLE); + + if (iolatency_may_queue(iolat, &wait, first_block)) + break; + first_block = false; + + if (lock) { + spin_unlock_irq(lock); + io_schedule(); + spin_lock_irq(lock); + } else { + io_schedule(); + } + } while (1); + + finish_wait(&rqw->wait, &wait); +} + +#define SCALE_DOWN_FACTOR 2 +#define SCALE_UP_FACTOR 4 + +static inline unsigned long scale_amount(unsigned long qd, bool up) +{ + return max(up ? qd >> SCALE_UP_FACTOR : qd >> SCALE_DOWN_FACTOR, 1UL); +} + +/* + * We scale the qd down faster than we scale up, so we need to use this helper + * to adjust the scale_cookie accordingly so we don't prematurely get + * scale_cookie at DEFAULT_SCALE_COOKIE and unthrottle too much. + * + * Each group has their own local copy of the last scale cookie they saw, so if + * the global scale cookie goes up or down they know which way they need to go + * based on their last knowledge of it. + */ +static void scale_cookie_change(struct blk_iolatency *blkiolat, + struct child_latency_info *lat_info, + bool up) +{ + unsigned long qd = blk_queue_depth(blkiolat->rqos.q); + unsigned long scale = scale_amount(qd, up); + unsigned long old = atomic_read(&lat_info->scale_cookie); + unsigned long max_scale = qd << 1; + unsigned long diff = 0; + + if (old < DEFAULT_SCALE_COOKIE) + diff = DEFAULT_SCALE_COOKIE - old; + + if (up) { + if (scale + old > DEFAULT_SCALE_COOKIE) + atomic_set(&lat_info->scale_cookie, + DEFAULT_SCALE_COOKIE); + else if (diff > qd) + atomic_inc(&lat_info->scale_cookie); + else + atomic_add(scale, &lat_info->scale_cookie); + } else { + /* + * We don't want to dig a hole so deep that it takes us hours to + * dig out of it. Just enough that we don't throttle/unthrottle + * with jagged workloads but can still unthrottle once pressure + * has sufficiently dissipated. + */ + if (diff > qd) { + if (diff < max_scale) + atomic_dec(&lat_info->scale_cookie); + } else { + atomic_sub(scale, &lat_info->scale_cookie); + } + } +} + +/* + * Change the queue depth of the iolatency_grp. We add/subtract 1/16th of the + * queue depth at a time so we don't get wild swings and hopefully dial in to + * fairer distribution of the overall queue depth. + */ +static void scale_change(struct iolatency_grp *iolat, bool up) +{ + unsigned long qd = blk_queue_depth(iolat->blkiolat->rqos.q); + unsigned long scale = scale_amount(qd, up); + unsigned long old = iolat->rq_depth.max_depth; + bool changed = false; + + if (old > qd) + old = qd; + + if (up) { + if (old == 1 && blkcg_unuse_delay(lat_to_blkg(iolat))) + return; + + if (old < qd) { + changed = true; + old += scale; + old = min(old, qd); + iolat->rq_depth.max_depth = old; + wake_up_all(&iolat->rq_wait.wait); + } + } else if (old > 1) { + old >>= 1; + changed = true; + iolat->rq_depth.max_depth = max(old, 1UL); + } +} + +/* Check our parent and see if the scale cookie has changed. */ +static void check_scale_change(struct iolatency_grp *iolat) +{ + struct iolatency_grp *parent; + struct child_latency_info *lat_info; + unsigned int cur_cookie; + unsigned int our_cookie = atomic_read(&iolat->scale_cookie); + u64 scale_lat; + unsigned int old; + int direction = 0; + + if (lat_to_blkg(iolat)->parent == NULL) + return; + + parent = blkg_to_lat(lat_to_blkg(iolat)->parent); + if (!parent) + return; + + lat_info = &parent->child_lat; + cur_cookie = atomic_read(&lat_info->scale_cookie); + scale_lat = READ_ONCE(lat_info->scale_lat); + + if (cur_cookie < our_cookie) + direction = -1; + else if (cur_cookie > our_cookie) + direction = 1; + else + return; + + old = atomic_cmpxchg(&iolat->scale_cookie, our_cookie, cur_cookie); + + /* Somebody beat us to the punch, just bail. */ + if (old != our_cookie) + return; + + if (direction < 0 && iolat->min_lat_nsec) { + u64 samples_thresh; + + if (!scale_lat || iolat->min_lat_nsec <= scale_lat) + return; + + /* + * Sometimes high priority groups are their own worst enemy, so + * instead of taking it out on some poor other group that did 5% + * or less of the IO's for the last summation just skip this + * scale down event. + */ + samples_thresh = lat_info->nr_samples * 5; + samples_thresh = div64_u64(samples_thresh, 100); + if (iolat->nr_samples <= samples_thresh) + return; + } + + /* We're as low as we can go. */ + if (iolat->rq_depth.max_depth == 1 && direction < 0) { + blkcg_use_delay(lat_to_blkg(iolat)); + return; + } + + /* We're back to the default cookie, unthrottle all the things. */ + if (cur_cookie == DEFAULT_SCALE_COOKIE) { + blkcg_clear_delay(lat_to_blkg(iolat)); + iolat->rq_depth.max_depth = UINT_MAX; + wake_up_all(&iolat->rq_wait.wait); + return; + } + + scale_change(iolat, direction > 0); +} + +static void blkcg_iolatency_throttle(struct rq_qos *rqos, struct bio *bio, + spinlock_t *lock) +{ + struct blk_iolatency *blkiolat = BLKIOLATENCY(rqos); + struct blkcg *blkcg; + struct blkcg_gq *blkg; + struct request_queue *q = rqos->q; + bool issue_as_root = bio_issue_as_root_blkg(bio); + + if (!blk_iolatency_enabled(blkiolat)) + return; + + rcu_read_lock(); + blkcg = bio_blkcg(bio); + bio_associate_blkcg(bio, &blkcg->css); + blkg = blkg_lookup(blkcg, q); + if (unlikely(!blkg)) { + if (!lock) + spin_lock_irq(q->queue_lock); + blkg = blkg_lookup_create(blkcg, q); + if (IS_ERR(blkg)) + blkg = NULL; + if (!lock) + spin_unlock_irq(q->queue_lock); + } + if (!blkg) + goto out; + + bio_issue_init(&bio->bi_issue, bio_sectors(bio)); + bio_associate_blkg(bio, blkg); +out: + rcu_read_unlock(); + while (blkg && blkg->parent) { + struct iolatency_grp *iolat = blkg_to_lat(blkg); + if (!iolat) { + blkg = blkg->parent; + continue; + } + + check_scale_change(iolat); + __blkcg_iolatency_throttle(rqos, iolat, lock, issue_as_root, + (bio->bi_opf & REQ_SWAP) == REQ_SWAP); + blkg = blkg->parent; + } + if (!timer_pending(&blkiolat->timer)) + mod_timer(&blkiolat->timer, jiffies + HZ); +} + +static void iolatency_record_time(struct iolatency_grp *iolat, + struct bio_issue *issue, u64 now, + bool issue_as_root) +{ + struct blk_rq_stat *rq_stat; + u64 start = bio_issue_time(issue); + u64 req_time; + + /* + * Have to do this so we are truncated to the correct time that our + * issue is truncated to. + */ + now = __bio_issue_time(now); + + if (now <= start) + return; + + req_time = now - start; + + /* + * We don't want to count issue_as_root bio's in the cgroups latency + * statistics as it could skew the numbers downwards. + */ + if (unlikely(issue_as_root && iolat->rq_depth.max_depth != UINT_MAX)) { + u64 sub = iolat->min_lat_nsec; + if (req_time < sub) + blkcg_add_delay(lat_to_blkg(iolat), now, sub - req_time); + return; + } + + rq_stat = get_cpu_ptr(iolat->stats); + blk_rq_stat_add(rq_stat, req_time); + put_cpu_ptr(rq_stat); +} + +#define BLKIOLATENCY_MIN_ADJUST_TIME (500 * NSEC_PER_MSEC) +#define BLKIOLATENCY_MIN_GOOD_SAMPLES 5 + +static void iolatency_check_latencies(struct iolatency_grp *iolat, u64 now) +{ + struct blkcg_gq *blkg = lat_to_blkg(iolat); + struct iolatency_grp *parent; + struct child_latency_info *lat_info; + struct blk_rq_stat stat; + unsigned long flags; + int cpu, exp_idx; + + blk_rq_stat_init(&stat); + preempt_disable(); + for_each_online_cpu(cpu) { + struct blk_rq_stat *s; + s = per_cpu_ptr(iolat->stats, cpu); + blk_rq_stat_sum(&stat, s); + blk_rq_stat_init(s); + } + preempt_enable(); + + parent = blkg_to_lat(blkg->parent); + if (!parent) + return; + + lat_info = &parent->child_lat; + + /* + * CALC_LOAD takes in a number stored in fixed point representation. + * Because we are using this for IO time in ns, the values stored + * are significantly larger than the FIXED_1 denominator (2048). + * Therefore, rounding errors in the calculation are negligible and + * can be ignored. + */ + exp_idx = min_t(int, BLKIOLATENCY_NR_EXP_FACTORS - 1, + div64_u64(iolat->cur_win_nsec, + BLKIOLATENCY_EXP_BUCKET_SIZE)); + CALC_LOAD(iolat->lat_avg, iolatency_exp_factors[exp_idx], stat.mean); + + /* Everything is ok and we don't need to adjust the scale. */ + if (stat.mean <= iolat->min_lat_nsec && + atomic_read(&lat_info->scale_cookie) == DEFAULT_SCALE_COOKIE) + return; + + /* Somebody beat us to the punch, just bail. */ + spin_lock_irqsave(&lat_info->lock, flags); + lat_info->nr_samples -= iolat->nr_samples; + lat_info->nr_samples += stat.nr_samples; + iolat->nr_samples = stat.nr_samples; + + if ((lat_info->last_scale_event >= now || + now - lat_info->last_scale_event < BLKIOLATENCY_MIN_ADJUST_TIME) && + lat_info->scale_lat <= iolat->min_lat_nsec) + goto out; + + if (stat.mean <= iolat->min_lat_nsec && + stat.nr_samples >= BLKIOLATENCY_MIN_GOOD_SAMPLES) { + if (lat_info->scale_grp == iolat) { + lat_info->last_scale_event = now; + scale_cookie_change(iolat->blkiolat, lat_info, true); + } + } else if (stat.mean > iolat->min_lat_nsec) { + lat_info->last_scale_event = now; + if (!lat_info->scale_grp || + lat_info->scale_lat > iolat->min_lat_nsec) { + WRITE_ONCE(lat_info->scale_lat, iolat->min_lat_nsec); + lat_info->scale_grp = iolat; + } + scale_cookie_change(iolat->blkiolat, lat_info, false); + } +out: + spin_unlock_irqrestore(&lat_info->lock, flags); +} + +static void blkcg_iolatency_done_bio(struct rq_qos *rqos, struct bio *bio) +{ + struct blkcg_gq *blkg; + struct rq_wait *rqw; + struct iolatency_grp *iolat; + u64 window_start; + u64 now = ktime_to_ns(ktime_get()); + bool issue_as_root = bio_issue_as_root_blkg(bio); + bool enabled = false; + + blkg = bio->bi_blkg; + if (!blkg) + return; + + iolat = blkg_to_lat(bio->bi_blkg); + if (!iolat) + return; + + enabled = blk_iolatency_enabled(iolat->blkiolat); + while (blkg && blkg->parent) { + iolat = blkg_to_lat(blkg); + if (!iolat) { + blkg = blkg->parent; + continue; + } + rqw = &iolat->rq_wait; + + atomic_dec(&rqw->inflight); + if (!enabled || iolat->min_lat_nsec == 0) + goto next; + iolatency_record_time(iolat, &bio->bi_issue, now, + issue_as_root); + window_start = atomic64_read(&iolat->window_start); + if (now > window_start && + (now - window_start) >= iolat->cur_win_nsec) { + if (atomic64_cmpxchg(&iolat->window_start, + window_start, now) == window_start) + iolatency_check_latencies(iolat, now); + } +next: + wake_up(&rqw->wait); + blkg = blkg->parent; + } +} + +static void blkcg_iolatency_cleanup(struct rq_qos *rqos, struct bio *bio) +{ + struct blkcg_gq *blkg; + + blkg = bio->bi_blkg; + while (blkg && blkg->parent) { + struct rq_wait *rqw; + struct iolatency_grp *iolat; + + iolat = blkg_to_lat(blkg); + if (!iolat) + goto next; + + rqw = &iolat->rq_wait; + atomic_dec(&rqw->inflight); + wake_up(&rqw->wait); +next: + blkg = blkg->parent; + } +} + +static void blkcg_iolatency_exit(struct rq_qos *rqos) +{ + struct blk_iolatency *blkiolat = BLKIOLATENCY(rqos); + + del_timer_sync(&blkiolat->timer); + blkcg_deactivate_policy(rqos->q, &blkcg_policy_iolatency); + kfree(blkiolat); +} + +static struct rq_qos_ops blkcg_iolatency_ops = { + .throttle = blkcg_iolatency_throttle, + .cleanup = blkcg_iolatency_cleanup, + .done_bio = blkcg_iolatency_done_bio, + .exit = blkcg_iolatency_exit, +}; + +static void blkiolatency_timer_fn(struct timer_list *t) +{ + struct blk_iolatency *blkiolat = from_timer(blkiolat, t, timer); + struct blkcg_gq *blkg; + struct cgroup_subsys_state *pos_css; + u64 now = ktime_to_ns(ktime_get()); + + rcu_read_lock(); + blkg_for_each_descendant_pre(blkg, pos_css, + blkiolat->rqos.q->root_blkg) { + struct iolatency_grp *iolat; + struct child_latency_info *lat_info; + unsigned long flags; + u64 cookie; + + /* + * We could be exiting, don't access the pd unless we have a + * ref on the blkg. + */ + if (!blkg_try_get(blkg)) + continue; + + iolat = blkg_to_lat(blkg); + if (!iolat) + goto next; + + lat_info = &iolat->child_lat; + cookie = atomic_read(&lat_info->scale_cookie); + + if (cookie >= DEFAULT_SCALE_COOKIE) + goto next; + + spin_lock_irqsave(&lat_info->lock, flags); + if (lat_info->last_scale_event >= now) + goto next_lock; + + /* + * We scaled down but don't have a scale_grp, scale up and carry + * on. + */ + if (lat_info->scale_grp == NULL) { + scale_cookie_change(iolat->blkiolat, lat_info, true); + goto next_lock; + } + + /* + * It's been 5 seconds since our last scale event, clear the + * scale grp in case the group that needed the scale down isn't + * doing any IO currently. + */ + if (now - lat_info->last_scale_event >= + ((u64)NSEC_PER_SEC * 5)) + lat_info->scale_grp = NULL; +next_lock: + spin_unlock_irqrestore(&lat_info->lock, flags); +next: + blkg_put(blkg); + } + rcu_read_unlock(); +} + +int blk_iolatency_init(struct request_queue *q) +{ + struct blk_iolatency *blkiolat; + struct rq_qos *rqos; + int ret; + + blkiolat = kzalloc(sizeof(*blkiolat), GFP_KERNEL); + if (!blkiolat) + return -ENOMEM; + + rqos = &blkiolat->rqos; + rqos->id = RQ_QOS_CGROUP; + rqos->ops = &blkcg_iolatency_ops; + rqos->q = q; + + rq_qos_add(q, rqos); + + ret = blkcg_activate_policy(q, &blkcg_policy_iolatency); + if (ret) { + rq_qos_del(q, rqos); + kfree(blkiolat); + return ret; + } + + timer_setup(&blkiolat->timer, blkiolatency_timer_fn, 0); + + return 0; +} + +static void iolatency_set_min_lat_nsec(struct blkcg_gq *blkg, u64 val) +{ + struct iolatency_grp *iolat = blkg_to_lat(blkg); + struct blk_iolatency *blkiolat = iolat->blkiolat; + u64 oldval = iolat->min_lat_nsec; + + iolat->min_lat_nsec = val; + iolat->cur_win_nsec = max_t(u64, val << 4, BLKIOLATENCY_MIN_WIN_SIZE); + iolat->cur_win_nsec = min_t(u64, iolat->cur_win_nsec, + BLKIOLATENCY_MAX_WIN_SIZE); + + if (!oldval && val) + atomic_inc(&blkiolat->enabled); + if (oldval && !val) + atomic_dec(&blkiolat->enabled); +} + +static void iolatency_clear_scaling(struct blkcg_gq *blkg) +{ + if (blkg->parent) { + struct iolatency_grp *iolat = blkg_to_lat(blkg->parent); + struct child_latency_info *lat_info; + if (!iolat) + return; + + lat_info = &iolat->child_lat; + spin_lock(&lat_info->lock); + atomic_set(&lat_info->scale_cookie, DEFAULT_SCALE_COOKIE); + lat_info->last_scale_event = 0; + lat_info->scale_grp = NULL; + lat_info->scale_lat = 0; + spin_unlock(&lat_info->lock); + } +} + +static ssize_t iolatency_set_limit(struct kernfs_open_file *of, char *buf, + size_t nbytes, loff_t off) +{ + struct blkcg *blkcg = css_to_blkcg(of_css(of)); + struct blkcg_gq *blkg; + struct blk_iolatency *blkiolat; + struct blkg_conf_ctx ctx; + struct iolatency_grp *iolat; + char *p, *tok; + u64 lat_val = 0; + u64 oldval; + int ret; + + ret = blkg_conf_prep(blkcg, &blkcg_policy_iolatency, buf, &ctx); + if (ret) + return ret; + + iolat = blkg_to_lat(ctx.blkg); + blkiolat = iolat->blkiolat; + p = ctx.body; + + ret = -EINVAL; + while ((tok = strsep(&p, " "))) { + char key[16]; + char val[21]; /* 18446744073709551616 */ + + if (sscanf(tok, "%15[^=]=%20s", key, val) != 2) + goto out; + + if (!strcmp(key, "target")) { + u64 v; + + if (!strcmp(val, "max")) + lat_val = 0; + else if (sscanf(val, "%llu", &v) == 1) + lat_val = v * NSEC_PER_USEC; + else + goto out; + } else { + goto out; + } + } + + /* Walk up the tree to see if our new val is lower than it should be. */ + blkg = ctx.blkg; + oldval = iolat->min_lat_nsec; + + iolatency_set_min_lat_nsec(blkg, lat_val); + if (oldval != iolat->min_lat_nsec) { + iolatency_clear_scaling(blkg); + } + + ret = 0; +out: + blkg_conf_finish(&ctx); + return ret ?: nbytes; +} + +static u64 iolatency_prfill_limit(struct seq_file *sf, + struct blkg_policy_data *pd, int off) +{ + struct iolatency_grp *iolat = pd_to_lat(pd); + const char *dname = blkg_dev_name(pd->blkg); + + if (!dname || !iolat->min_lat_nsec) + return 0; + seq_printf(sf, "%s target=%llu\n", + dname, div_u64(iolat->min_lat_nsec, NSEC_PER_USEC)); + return 0; +} + +static int iolatency_print_limit(struct seq_file *sf, void *v) +{ + blkcg_print_blkgs(sf, css_to_blkcg(seq_css(sf)), + iolatency_prfill_limit, + &blkcg_policy_iolatency, seq_cft(sf)->private, false); + return 0; +} + +static size_t iolatency_pd_stat(struct blkg_policy_data *pd, char *buf, + size_t size) +{ + struct iolatency_grp *iolat = pd_to_lat(pd); + unsigned long long avg_lat = div64_u64(iolat->lat_avg, NSEC_PER_USEC); + unsigned long long cur_win = div64_u64(iolat->cur_win_nsec, NSEC_PER_MSEC); + + if (iolat->rq_depth.max_depth == UINT_MAX) + return scnprintf(buf, size, " depth=max avg_lat=%llu win=%llu", + avg_lat, cur_win); + + return scnprintf(buf, size, " depth=%u avg_lat=%llu win=%llu", + iolat->rq_depth.max_depth, avg_lat, cur_win); +} + + +static struct blkg_policy_data *iolatency_pd_alloc(gfp_t gfp, int node) +{ + struct iolatency_grp *iolat; + + iolat = kzalloc_node(sizeof(*iolat), gfp, node); + if (!iolat) + return NULL; + iolat->stats = __alloc_percpu_gfp(sizeof(struct blk_rq_stat), + __alignof__(struct blk_rq_stat), gfp); + if (!iolat->stats) { + kfree(iolat); + return NULL; + } + return &iolat->pd; +} + +static void iolatency_pd_init(struct blkg_policy_data *pd) +{ + struct iolatency_grp *iolat = pd_to_lat(pd); + struct blkcg_gq *blkg = lat_to_blkg(iolat); + struct rq_qos *rqos = blkcg_rq_qos(blkg->q); + struct blk_iolatency *blkiolat = BLKIOLATENCY(rqos); + u64 now = ktime_to_ns(ktime_get()); + int cpu; + + for_each_possible_cpu(cpu) { + struct blk_rq_stat *stat; + stat = per_cpu_ptr(iolat->stats, cpu); + blk_rq_stat_init(stat); + } + + rq_wait_init(&iolat->rq_wait); + spin_lock_init(&iolat->child_lat.lock); + iolat->rq_depth.queue_depth = blk_queue_depth(blkg->q); + iolat->rq_depth.max_depth = UINT_MAX; + iolat->rq_depth.default_depth = iolat->rq_depth.queue_depth; + iolat->blkiolat = blkiolat; + iolat->cur_win_nsec = 100 * NSEC_PER_MSEC; + atomic64_set(&iolat->window_start, now); + + /* + * We init things in list order, so the pd for the parent may not be + * init'ed yet for whatever reason. + */ + if (blkg->parent && blkg_to_pd(blkg->parent, &blkcg_policy_iolatency)) { + struct iolatency_grp *parent = blkg_to_lat(blkg->parent); + atomic_set(&iolat->scale_cookie, + atomic_read(&parent->child_lat.scale_cookie)); + } else { + atomic_set(&iolat->scale_cookie, DEFAULT_SCALE_COOKIE); + } + + atomic_set(&iolat->child_lat.scale_cookie, DEFAULT_SCALE_COOKIE); +} + +static void iolatency_pd_offline(struct blkg_policy_data *pd) +{ + struct iolatency_grp *iolat = pd_to_lat(pd); + struct blkcg_gq *blkg = lat_to_blkg(iolat); + + iolatency_set_min_lat_nsec(blkg, 0); + iolatency_clear_scaling(blkg); +} + +static void iolatency_pd_free(struct blkg_policy_data *pd) +{ + struct iolatency_grp *iolat = pd_to_lat(pd); + free_percpu(iolat->stats); + kfree(iolat); +} + +static struct cftype iolatency_files[] = { + { + .name = "latency", + .flags = CFTYPE_NOT_ON_ROOT, + .seq_show = iolatency_print_limit, + .write = iolatency_set_limit, + }, + {} +}; + +static struct blkcg_policy blkcg_policy_iolatency = { + .dfl_cftypes = iolatency_files, + .pd_alloc_fn = iolatency_pd_alloc, + .pd_init_fn = iolatency_pd_init, + .pd_offline_fn = iolatency_pd_offline, + .pd_free_fn = iolatency_pd_free, + .pd_stat_fn = iolatency_pd_stat, +}; + +static int __init iolatency_init(void) +{ + return blkcg_policy_register(&blkcg_policy_iolatency); +} + +static void __exit iolatency_exit(void) +{ + return blkcg_policy_unregister(&blkcg_policy_iolatency); +} + +module_init(iolatency_init); +module_exit(iolatency_exit); diff --git a/block/blk-lib.c b/block/blk-lib.c index 8faa70f26fcd..d1b9dd03da25 100644 --- a/block/blk-lib.c +++ b/block/blk-lib.c @@ -68,6 +68,8 @@ int __blkdev_issue_discard(struct block_device *bdev, sector_t sector, */ req_sects = min_t(sector_t, nr_sects, q->limits.max_discard_sectors); + if (!req_sects) + goto fail; if (req_sects > UINT_MAX >> 9) req_sects = UINT_MAX >> 9; @@ -105,6 +107,14 @@ int __blkdev_issue_discard(struct block_device *bdev, sector_t sector, *biop = bio; return 0; + +fail: + if (bio) { + submit_bio_wait(bio); + bio_put(bio); + } + *biop = NULL; + return -EOPNOTSUPP; } EXPORT_SYMBOL(__blkdev_issue_discard); diff --git a/block/blk-mq-debugfs-zoned.c b/block/blk-mq-debugfs-zoned.c new file mode 100644 index 000000000000..fb2c82c351e4 --- /dev/null +++ b/block/blk-mq-debugfs-zoned.c @@ -0,0 +1,24 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2017 Western Digital Corporation or its affiliates. + * + * This file is released under the GPL. + */ + +#include <linux/blkdev.h> +#include "blk-mq-debugfs.h" + +int queue_zone_wlock_show(void *data, struct seq_file *m) +{ + struct request_queue *q = data; + unsigned int i; + + if (!q->seq_zones_wlock) + return 0; + + for (i = 0; i < q->nr_zones; i++) + if (test_bit(i, q->seq_zones_wlock)) + seq_printf(m, "%u\n", i); + + return 0; +} diff --git a/block/blk-mq-debugfs.c b/block/blk-mq-debugfs.c index 1c4532e92938..cb1e6cf7ac48 100644 --- a/block/blk-mq-debugfs.c +++ b/block/blk-mq-debugfs.c @@ -206,21 +206,6 @@ static ssize_t queue_write_hint_store(void *data, const char __user *buf, return count; } -static int queue_zone_wlock_show(void *data, struct seq_file *m) -{ - struct request_queue *q = data; - unsigned int i; - - if (!q->seq_zones_wlock) - return 0; - - for (i = 0; i < blk_queue_nr_zones(q); i++) - if (test_bit(i, q->seq_zones_wlock)) - seq_printf(m, "%u\n", i); - - return 0; -} - static const struct blk_mq_debugfs_attr blk_mq_debugfs_queue_attrs[] = { { "poll_stat", 0400, queue_poll_stat_show }, { "requeue_list", 0400, .seq_ops = &queue_requeue_list_seq_ops }, @@ -637,6 +622,14 @@ static int hctx_active_show(void *data, struct seq_file *m) return 0; } +static int hctx_dispatch_busy_show(void *data, struct seq_file *m) +{ + struct blk_mq_hw_ctx *hctx = data; + + seq_printf(m, "%u\n", hctx->dispatch_busy); + return 0; +} + static void *ctx_rq_list_start(struct seq_file *m, loff_t *pos) __acquires(&ctx->lock) { @@ -798,6 +791,7 @@ static const struct blk_mq_debugfs_attr blk_mq_debugfs_hctx_attrs[] = { {"queued", 0600, hctx_queued_show, hctx_queued_write}, {"run", 0600, hctx_run_show, hctx_run_write}, {"active", 0400, hctx_active_show}, + {"dispatch_busy", 0400, hctx_dispatch_busy_show}, {}, }; diff --git a/block/blk-mq-debugfs.h b/block/blk-mq-debugfs.h index b9d366e57097..a9160be12be0 100644 --- a/block/blk-mq-debugfs.h +++ b/block/blk-mq-debugfs.h @@ -80,4 +80,13 @@ static inline void blk_mq_debugfs_unregister_sched_hctx(struct blk_mq_hw_ctx *hc } #endif +#ifdef CONFIG_BLK_DEBUG_FS_ZONED +int queue_zone_wlock_show(void *data, struct seq_file *m); +#else +static inline int queue_zone_wlock_show(void *data, struct seq_file *m) +{ + return 0; +} +#endif + #endif diff --git a/block/blk-mq-pci.c b/block/blk-mq-pci.c index e233996bb76f..db644ec624f5 100644 --- a/block/blk-mq-pci.c +++ b/block/blk-mq-pci.c @@ -17,6 +17,8 @@ #include <linux/pci.h> #include <linux/module.h> +#include "blk-mq.h" + /** * blk_mq_pci_map_queues - provide a default queue mapping for PCI device * @set: tagset to provide the mapping for @@ -48,8 +50,7 @@ int blk_mq_pci_map_queues(struct blk_mq_tag_set *set, struct pci_dev *pdev, fallback: WARN_ON_ONCE(set->nr_hw_queues > 1); - for_each_possible_cpu(cpu) - set->mq_map[cpu] = 0; + blk_mq_clear_mq_map(set); return 0; } EXPORT_SYMBOL_GPL(blk_mq_pci_map_queues); diff --git a/block/blk-mq-sched.c b/block/blk-mq-sched.c index 56c493c6cd90..cf9c66c6d35a 100644 --- a/block/blk-mq-sched.c +++ b/block/blk-mq-sched.c @@ -59,29 +59,16 @@ static void blk_mq_sched_mark_restart_hctx(struct blk_mq_hw_ctx *hctx) if (test_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state)) return; - if (hctx->flags & BLK_MQ_F_TAG_SHARED) { - struct request_queue *q = hctx->queue; - - if (!test_and_set_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state)) - atomic_inc(&q->shared_hctx_restart); - } else - set_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state); + set_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state); } -static bool blk_mq_sched_restart_hctx(struct blk_mq_hw_ctx *hctx) +void blk_mq_sched_restart(struct blk_mq_hw_ctx *hctx) { if (!test_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state)) - return false; - - if (hctx->flags & BLK_MQ_F_TAG_SHARED) { - struct request_queue *q = hctx->queue; - - if (test_and_clear_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state)) - atomic_dec(&q->shared_hctx_restart); - } else - clear_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state); + return; + clear_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state); - return blk_mq_run_hw_queue(hctx, true); + blk_mq_run_hw_queue(hctx, true); } /* @@ -219,15 +206,8 @@ void blk_mq_sched_dispatch_requests(struct blk_mq_hw_ctx *hctx) } } else if (has_sched_dispatch) { blk_mq_do_dispatch_sched(hctx); - } else if (q->mq_ops->get_budget) { - /* - * If we need to get budget before queuing request, we - * dequeue request one by one from sw queue for avoiding - * to mess up I/O merge when dispatch runs out of resource. - * - * TODO: get more budgets, and dequeue more requests in - * one time. - */ + } else if (hctx->dispatch_busy) { + /* dequeue request one by one from sw queue if queue is busy */ blk_mq_do_dispatch_ctx(hctx); } else { blk_mq_flush_busy_ctxs(hctx, &rq_list); @@ -339,7 +319,8 @@ bool __blk_mq_sched_bio_merge(struct request_queue *q, struct bio *bio) return e->type->ops.mq.bio_merge(hctx, bio); } - if (hctx->flags & BLK_MQ_F_SHOULD_MERGE) { + if ((hctx->flags & BLK_MQ_F_SHOULD_MERGE) && + !list_empty_careful(&ctx->rq_list)) { /* default per sw-queue merge */ spin_lock(&ctx->lock); ret = blk_mq_attempt_merge(q, ctx, bio); @@ -380,68 +361,6 @@ static bool blk_mq_sched_bypass_insert(struct blk_mq_hw_ctx *hctx, return false; } -/** - * list_for_each_entry_rcu_rr - iterate in a round-robin fashion over rcu list - * @pos: loop cursor. - * @skip: the list element that will not be examined. Iteration starts at - * @skip->next. - * @head: head of the list to examine. This list must have at least one - * element, namely @skip. - * @member: name of the list_head structure within typeof(*pos). - */ -#define list_for_each_entry_rcu_rr(pos, skip, head, member) \ - for ((pos) = (skip); \ - (pos = (pos)->member.next != (head) ? list_entry_rcu( \ - (pos)->member.next, typeof(*pos), member) : \ - list_entry_rcu((pos)->member.next->next, typeof(*pos), member)), \ - (pos) != (skip); ) - -/* - * Called after a driver tag has been freed to check whether a hctx needs to - * be restarted. Restarts @hctx if its tag set is not shared. Restarts hardware - * queues in a round-robin fashion if the tag set of @hctx is shared with other - * hardware queues. - */ -void blk_mq_sched_restart(struct blk_mq_hw_ctx *const hctx) -{ - struct blk_mq_tags *const tags = hctx->tags; - struct blk_mq_tag_set *const set = hctx->queue->tag_set; - struct request_queue *const queue = hctx->queue, *q; - struct blk_mq_hw_ctx *hctx2; - unsigned int i, j; - - if (set->flags & BLK_MQ_F_TAG_SHARED) { - /* - * If this is 0, then we know that no hardware queues - * have RESTART marked. We're done. - */ - if (!atomic_read(&queue->shared_hctx_restart)) - return; - - rcu_read_lock(); - list_for_each_entry_rcu_rr(q, queue, &set->tag_list, - tag_set_list) { - queue_for_each_hw_ctx(q, hctx2, i) - if (hctx2->tags == tags && - blk_mq_sched_restart_hctx(hctx2)) - goto done; - } - j = hctx->queue_num + 1; - for (i = 0; i < queue->nr_hw_queues; i++, j++) { - if (j == queue->nr_hw_queues) - j = 0; - hctx2 = queue->queue_hw_ctx[j]; - if (hctx2->tags == tags && - blk_mq_sched_restart_hctx(hctx2)) - break; - } -done: - rcu_read_unlock(); - } else { - blk_mq_sched_restart_hctx(hctx); - } -} - void blk_mq_sched_insert_request(struct request *rq, bool at_head, bool run_queue, bool async) { @@ -486,8 +405,19 @@ void blk_mq_sched_insert_requests(struct request_queue *q, if (e && e->type->ops.mq.insert_requests) e->type->ops.mq.insert_requests(hctx, list, false); - else + else { + /* + * try to issue requests directly if the hw queue isn't + * busy in case of 'none' scheduler, and this way may save + * us one extra enqueue & dequeue to sw queue. + */ + if (!hctx->dispatch_busy && !e && !run_queue_async) { + blk_mq_try_issue_list_directly(hctx, list); + if (list_empty(list)) + return; + } blk_mq_insert_requests(hctx, ctx, list); + } blk_mq_run_hw_queue(hctx, run_queue_async); } diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c index 3de0836163c2..816923bf874d 100644 --- a/block/blk-mq-tag.c +++ b/block/blk-mq-tag.c @@ -23,6 +23,9 @@ bool blk_mq_has_free_tags(struct blk_mq_tags *tags) /* * If a previously inactive queue goes active, bump the active user count. + * We need to do this before try to allocate driver tag, then even if fail + * to get tag when first time, the other shared-tag users could reserve + * budget for it. */ bool __blk_mq_tag_busy(struct blk_mq_hw_ctx *hctx) { @@ -399,8 +402,6 @@ int blk_mq_tag_update_depth(struct blk_mq_hw_ctx *hctx, if (tdepth <= tags->nr_reserved_tags) return -EINVAL; - tdepth -= tags->nr_reserved_tags; - /* * If we are allowed to grow beyond the original size, allocate * a new set of tags before freeing the old one. @@ -420,7 +421,8 @@ int blk_mq_tag_update_depth(struct blk_mq_hw_ctx *hctx, if (tdepth > 16 * BLKDEV_MAX_RQ) return -EINVAL; - new = blk_mq_alloc_rq_map(set, hctx->queue_num, tdepth, 0); + new = blk_mq_alloc_rq_map(set, hctx->queue_num, tdepth, + tags->nr_reserved_tags); if (!new) return -ENOMEM; ret = blk_mq_alloc_rqs(set, new, hctx->queue_num, tdepth); @@ -437,7 +439,8 @@ int blk_mq_tag_update_depth(struct blk_mq_hw_ctx *hctx, * Don't need (or can't) update reserved tags here, they * remain static and should never need resizing. */ - sbitmap_queue_resize(&tags->bitmap_tags, tdepth); + sbitmap_queue_resize(&tags->bitmap_tags, + tdepth - tags->nr_reserved_tags); } return 0; diff --git a/block/blk-mq.c b/block/blk-mq.c index 654b0dc7e001..72a0033ccee9 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -34,8 +34,8 @@ #include "blk-mq-debugfs.h" #include "blk-mq-tag.h" #include "blk-stat.h" -#include "blk-wbt.h" #include "blk-mq-sched.h" +#include "blk-rq-qos.h" static bool blk_mq_poll(struct request_queue *q, blk_qc_t cookie); static void blk_mq_poll_stats_start(struct request_queue *q); @@ -285,7 +285,7 @@ static struct request *blk_mq_rq_ctx_init(struct blk_mq_alloc_data *data, rq->tag = -1; rq->internal_tag = tag; } else { - if (blk_mq_tag_busy(data->hctx)) { + if (data->hctx->flags & BLK_MQ_F_TAG_SHARED) { rq_flags = RQF_MQ_INFLIGHT; atomic_inc(&data->hctx->nr_active); } @@ -367,6 +367,8 @@ static struct request *blk_mq_get_request(struct request_queue *q, if (!op_is_flush(op) && e->type->ops.mq.limit_depth && !(data->flags & BLK_MQ_REQ_RESERVED)) e->type->ops.mq.limit_depth(op, data); + } else { + blk_mq_tag_busy(data->hctx); } tag = blk_mq_get_tag(data); @@ -504,7 +506,7 @@ void blk_mq_free_request(struct request *rq) if (unlikely(laptop_mode && !blk_rq_is_passthrough(rq))) laptop_io_completion(q->backing_dev_info); - wbt_done(q->rq_wb, rq); + rq_qos_done(q, rq); if (blk_rq_rl(rq)) blk_put_rl(blk_rq_rl(rq)); @@ -527,7 +529,7 @@ inline void __blk_mq_end_request(struct request *rq, blk_status_t error) blk_account_io_done(rq, now); if (rq->end_io) { - wbt_done(rq->q->rq_wb, rq); + rq_qos_done(rq->q, rq); rq->end_io(rq, error); } else { if (unlikely(blk_bidi_rq(rq))) @@ -639,7 +641,7 @@ void blk_mq_start_request(struct request *rq) rq->throtl_size = blk_rq_sectors(rq); #endif rq->rq_flags |= RQF_STATS; - wbt_issue(q->rq_wb, rq); + rq_qos_issue(q, rq); } WARN_ON_ONCE(blk_mq_rq_state(rq) != MQ_RQ_IDLE); @@ -665,7 +667,7 @@ static void __blk_mq_requeue_request(struct request *rq) blk_mq_put_driver_tag(rq); trace_block_rq_requeue(q, rq); - wbt_requeue(q->rq_wb, rq); + rq_qos_requeue(q, rq); if (blk_mq_request_started(rq)) { WRITE_ONCE(rq->state, MQ_RQ_IDLE); @@ -962,16 +964,14 @@ static inline unsigned int queued_to_index(unsigned int queued) return min(BLK_MQ_MAX_DISPATCH_ORDER - 1, ilog2(queued) + 1); } -bool blk_mq_get_driver_tag(struct request *rq, struct blk_mq_hw_ctx **hctx, - bool wait) +bool blk_mq_get_driver_tag(struct request *rq) { struct blk_mq_alloc_data data = { .q = rq->q, .hctx = blk_mq_map_queue(rq->q, rq->mq_ctx->cpu), - .flags = wait ? 0 : BLK_MQ_REQ_NOWAIT, + .flags = BLK_MQ_REQ_NOWAIT, }; - - might_sleep_if(wait); + bool shared; if (rq->tag != -1) goto done; @@ -979,9 +979,10 @@ bool blk_mq_get_driver_tag(struct request *rq, struct blk_mq_hw_ctx **hctx, if (blk_mq_tag_is_reserved(data.hctx->sched_tags, rq->internal_tag)) data.flags |= BLK_MQ_REQ_RESERVED; + shared = blk_mq_tag_busy(data.hctx); rq->tag = blk_mq_get_tag(&data); if (rq->tag >= 0) { - if (blk_mq_tag_busy(data.hctx)) { + if (shared) { rq->rq_flags |= RQF_MQ_INFLIGHT; atomic_inc(&data.hctx->nr_active); } @@ -989,8 +990,6 @@ bool blk_mq_get_driver_tag(struct request *rq, struct blk_mq_hw_ctx **hctx, } done: - if (hctx) - *hctx = data.hctx; return rq->tag != -1; } @@ -1001,7 +1000,10 @@ static int blk_mq_dispatch_wake(wait_queue_entry_t *wait, unsigned mode, hctx = container_of(wait, struct blk_mq_hw_ctx, dispatch_wait); + spin_lock(&hctx->dispatch_wait_lock); list_del_init(&wait->entry); + spin_unlock(&hctx->dispatch_wait_lock); + blk_mq_run_hw_queue(hctx, true); return 1; } @@ -1012,17 +1014,16 @@ static int blk_mq_dispatch_wake(wait_queue_entry_t *wait, unsigned mode, * restart. For both cases, take care to check the condition again after * marking us as waiting. */ -static bool blk_mq_mark_tag_wait(struct blk_mq_hw_ctx **hctx, +static bool blk_mq_mark_tag_wait(struct blk_mq_hw_ctx *hctx, struct request *rq) { - struct blk_mq_hw_ctx *this_hctx = *hctx; - struct sbq_wait_state *ws; + struct wait_queue_head *wq; wait_queue_entry_t *wait; bool ret; - if (!(this_hctx->flags & BLK_MQ_F_TAG_SHARED)) { - if (!test_bit(BLK_MQ_S_SCHED_RESTART, &this_hctx->state)) - set_bit(BLK_MQ_S_SCHED_RESTART, &this_hctx->state); + if (!(hctx->flags & BLK_MQ_F_TAG_SHARED)) { + if (!test_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state)) + set_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state); /* * It's possible that a tag was freed in the window between the @@ -1032,30 +1033,35 @@ static bool blk_mq_mark_tag_wait(struct blk_mq_hw_ctx **hctx, * Don't clear RESTART here, someone else could have set it. * At most this will cost an extra queue run. */ - return blk_mq_get_driver_tag(rq, hctx, false); + return blk_mq_get_driver_tag(rq); } - wait = &this_hctx->dispatch_wait; + wait = &hctx->dispatch_wait; if (!list_empty_careful(&wait->entry)) return false; - spin_lock(&this_hctx->lock); + wq = &bt_wait_ptr(&hctx->tags->bitmap_tags, hctx)->wait; + + spin_lock_irq(&wq->lock); + spin_lock(&hctx->dispatch_wait_lock); if (!list_empty(&wait->entry)) { - spin_unlock(&this_hctx->lock); + spin_unlock(&hctx->dispatch_wait_lock); + spin_unlock_irq(&wq->lock); return false; } - ws = bt_wait_ptr(&this_hctx->tags->bitmap_tags, this_hctx); - add_wait_queue(&ws->wait, wait); + wait->flags &= ~WQ_FLAG_EXCLUSIVE; + __add_wait_queue(wq, wait); /* * It's possible that a tag was freed in the window between the * allocation failure and adding the hardware queue to the wait * queue. */ - ret = blk_mq_get_driver_tag(rq, hctx, false); + ret = blk_mq_get_driver_tag(rq); if (!ret) { - spin_unlock(&this_hctx->lock); + spin_unlock(&hctx->dispatch_wait_lock); + spin_unlock_irq(&wq->lock); return false; } @@ -1063,14 +1069,42 @@ static bool blk_mq_mark_tag_wait(struct blk_mq_hw_ctx **hctx, * We got a tag, remove ourselves from the wait queue to ensure * someone else gets the wakeup. */ - spin_lock_irq(&ws->wait.lock); list_del_init(&wait->entry); - spin_unlock_irq(&ws->wait.lock); - spin_unlock(&this_hctx->lock); + spin_unlock(&hctx->dispatch_wait_lock); + spin_unlock_irq(&wq->lock); return true; } +#define BLK_MQ_DISPATCH_BUSY_EWMA_WEIGHT 8 +#define BLK_MQ_DISPATCH_BUSY_EWMA_FACTOR 4 +/* + * Update dispatch busy with the Exponential Weighted Moving Average(EWMA): + * - EWMA is one simple way to compute running average value + * - weight(7/8 and 1/8) is applied so that it can decrease exponentially + * - take 4 as factor for avoiding to get too small(0) result, and this + * factor doesn't matter because EWMA decreases exponentially + */ +static void blk_mq_update_dispatch_busy(struct blk_mq_hw_ctx *hctx, bool busy) +{ + unsigned int ewma; + + if (hctx->queue->elevator) + return; + + ewma = hctx->dispatch_busy; + + if (!ewma && !busy) + return; + + ewma *= BLK_MQ_DISPATCH_BUSY_EWMA_WEIGHT - 1; + if (busy) + ewma += 1 << BLK_MQ_DISPATCH_BUSY_EWMA_FACTOR; + ewma /= BLK_MQ_DISPATCH_BUSY_EWMA_WEIGHT; + + hctx->dispatch_busy = ewma; +} + #define BLK_MQ_RESOURCE_DELAY 3 /* ms units */ /* @@ -1103,7 +1137,7 @@ bool blk_mq_dispatch_rq_list(struct request_queue *q, struct list_head *list, if (!got_budget && !blk_mq_get_dispatch_budget(hctx)) break; - if (!blk_mq_get_driver_tag(rq, NULL, false)) { + if (!blk_mq_get_driver_tag(rq)) { /* * The initial allocation attempt failed, so we need to * rerun the hardware queue when a tag is freed. The @@ -1111,7 +1145,7 @@ bool blk_mq_dispatch_rq_list(struct request_queue *q, struct list_head *list, * before we add this entry back on the dispatch list, * we'll re-run it below. */ - if (!blk_mq_mark_tag_wait(&hctx, rq)) { + if (!blk_mq_mark_tag_wait(hctx, rq)) { blk_mq_put_dispatch_budget(hctx); /* * For non-shared tags, the RESTART check @@ -1135,7 +1169,7 @@ bool blk_mq_dispatch_rq_list(struct request_queue *q, struct list_head *list, bd.last = true; else { nxt = list_first_entry(list, struct request, queuelist); - bd.last = !blk_mq_get_driver_tag(nxt, NULL, false); + bd.last = !blk_mq_get_driver_tag(nxt); } ret = q->mq_ops->queue_rq(hctx, &bd); @@ -1207,8 +1241,10 @@ bool blk_mq_dispatch_rq_list(struct request_queue *q, struct list_head *list, else if (needs_restart && (ret == BLK_STS_RESOURCE)) blk_mq_delay_run_hw_queue(hctx, BLK_MQ_RESOURCE_DELAY); + blk_mq_update_dispatch_busy(hctx, true); return false; - } + } else + blk_mq_update_dispatch_busy(hctx, false); /* * If the host/device is unable to accept more work, inform the @@ -1542,19 +1578,19 @@ void blk_mq_insert_requests(struct blk_mq_hw_ctx *hctx, struct blk_mq_ctx *ctx, struct list_head *list) { + struct request *rq; + /* * preemption doesn't flush plug list, so it's possible ctx->cpu is * offline now */ - spin_lock(&ctx->lock); - while (!list_empty(list)) { - struct request *rq; - - rq = list_first_entry(list, struct request, queuelist); + list_for_each_entry(rq, list, queuelist) { BUG_ON(rq->mq_ctx != ctx); - list_del_init(&rq->queuelist); - __blk_mq_insert_req_list(hctx, rq, false); + trace_block_rq_insert(hctx->queue, rq); } + + spin_lock(&ctx->lock); + list_splice_tail_init(list, &ctx->rq_list); blk_mq_hctx_mark_pending(hctx, ctx); spin_unlock(&ctx->lock); } @@ -1657,13 +1693,16 @@ static blk_status_t __blk_mq_issue_directly(struct blk_mq_hw_ctx *hctx, ret = q->mq_ops->queue_rq(hctx, &bd); switch (ret) { case BLK_STS_OK: + blk_mq_update_dispatch_busy(hctx, false); *cookie = new_cookie; break; case BLK_STS_RESOURCE: case BLK_STS_DEV_RESOURCE: + blk_mq_update_dispatch_busy(hctx, true); __blk_mq_requeue_request(rq); break; default: + blk_mq_update_dispatch_busy(hctx, false); *cookie = BLK_QC_T_NONE; break; } @@ -1698,7 +1737,7 @@ static blk_status_t __blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx, if (!blk_mq_get_dispatch_budget(hctx)) goto insert; - if (!blk_mq_get_driver_tag(rq, NULL, false)) { + if (!blk_mq_get_driver_tag(rq)) { blk_mq_put_dispatch_budget(hctx); goto insert; } @@ -1746,6 +1785,27 @@ blk_status_t blk_mq_request_issue_directly(struct request *rq) return ret; } +void blk_mq_try_issue_list_directly(struct blk_mq_hw_ctx *hctx, + struct list_head *list) +{ + while (!list_empty(list)) { + blk_status_t ret; + struct request *rq = list_first_entry(list, struct request, + queuelist); + + list_del_init(&rq->queuelist); + ret = blk_mq_request_issue_directly(rq); + if (ret != BLK_STS_OK) { + if (ret == BLK_STS_RESOURCE || + ret == BLK_STS_DEV_RESOURCE) { + list_add(&rq->queuelist, list); + break; + } + blk_mq_end_request(rq, ret); + } + } +} + static blk_qc_t blk_mq_make_request(struct request_queue *q, struct bio *bio) { const int is_sync = op_is_sync(bio->bi_opf); @@ -1756,7 +1816,6 @@ static blk_qc_t blk_mq_make_request(struct request_queue *q, struct bio *bio) struct blk_plug *plug; struct request *same_queue_rq = NULL; blk_qc_t cookie; - unsigned int wb_acct; blk_queue_bounce(q, &bio); @@ -1772,19 +1831,19 @@ static blk_qc_t blk_mq_make_request(struct request_queue *q, struct bio *bio) if (blk_mq_sched_bio_merge(q, bio)) return BLK_QC_T_NONE; - wb_acct = wbt_wait(q->rq_wb, bio, NULL); + rq_qos_throttle(q, bio, NULL); trace_block_getrq(q, bio, bio->bi_opf); rq = blk_mq_get_request(q, bio, bio->bi_opf, &data); if (unlikely(!rq)) { - __wbt_done(q->rq_wb, wb_acct); + rq_qos_cleanup(q, bio); if (bio->bi_opf & REQ_NOWAIT) bio_wouldblock_error(bio); return BLK_QC_T_NONE; } - wbt_track(rq, wb_acct); + rq_qos_track(q, rq, bio); cookie = request_to_qc_t(data.hctx, rq); @@ -1847,7 +1906,8 @@ static blk_qc_t blk_mq_make_request(struct request_queue *q, struct bio *bio) blk_mq_try_issue_directly(data.hctx, same_queue_rq, &cookie); } - } else if (q->nr_hw_queues > 1 && is_sync) { + } else if ((q->nr_hw_queues > 1 && is_sync) || (!q->elevator && + !data.hctx->dispatch_busy)) { blk_mq_put_ctx(data.ctx); blk_mq_bio_to_request(rq, bio); blk_mq_try_issue_directly(data.hctx, rq, &cookie); @@ -2146,6 +2206,7 @@ static int blk_mq_init_hctx(struct request_queue *q, hctx->nr_ctx = 0; + spin_lock_init(&hctx->dispatch_wait_lock); init_waitqueue_func_entry(&hctx->dispatch_wait, blk_mq_dispatch_wake); INIT_LIST_HEAD(&hctx->dispatch_wait.entry); @@ -2331,15 +2392,10 @@ static void queue_set_hctx_shared(struct request_queue *q, bool shared) int i; queue_for_each_hw_ctx(q, hctx, i) { - if (shared) { - if (test_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state)) - atomic_inc(&q->shared_hctx_restart); + if (shared) hctx->flags |= BLK_MQ_F_TAG_SHARED; - } else { - if (test_bit(BLK_MQ_S_SCHED_RESTART, &hctx->state)) - atomic_dec(&q->shared_hctx_restart); + else hctx->flags &= ~BLK_MQ_F_TAG_SHARED; - } } } @@ -2370,7 +2426,6 @@ static void blk_mq_del_queue_tag_set(struct request_queue *q) blk_mq_update_tag_set_depth(set, false); } mutex_unlock(&set->tag_list_lock); - synchronize_rcu(); INIT_LIST_HEAD(&q->tag_set_list); } @@ -2685,7 +2740,6 @@ static int blk_mq_alloc_rq_maps(struct blk_mq_tag_set *set) static int blk_mq_update_queue_map(struct blk_mq_tag_set *set) { if (set->ops->map_queues) { - int cpu; /* * transport .map_queues is usually done in the following * way: @@ -2700,8 +2754,7 @@ static int blk_mq_update_queue_map(struct blk_mq_tag_set *set) * killing stale mapping since one CPU may not be mapped * to any hw queue. */ - for_each_possible_cpu(cpu) - set->mq_map[cpu] = 0; + blk_mq_clear_mq_map(set); return set->ops->map_queues(set); } else @@ -2711,7 +2764,7 @@ static int blk_mq_update_queue_map(struct blk_mq_tag_set *set) /* * Alloc a tag set to be associated with one or more request queues. * May fail with EINVAL for various error conditions. May adjust the - * requested depth down, if if it too large. In that case, the set + * requested depth down, if it's too large. In that case, the set * value will be stored in set->queue_depth. */ int blk_mq_alloc_tag_set(struct blk_mq_tag_set *set) diff --git a/block/blk-mq.h b/block/blk-mq.h index 89231e439b2f..9497b47e2526 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -36,8 +36,7 @@ int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr); void blk_mq_wake_waiters(struct request_queue *q); bool blk_mq_dispatch_rq_list(struct request_queue *, struct list_head *, bool); void blk_mq_flush_busy_ctxs(struct blk_mq_hw_ctx *hctx, struct list_head *list); -bool blk_mq_get_driver_tag(struct request *rq, struct blk_mq_hw_ctx **hctx, - bool wait); +bool blk_mq_get_driver_tag(struct request *rq); struct request *blk_mq_dequeue_from_ctx(struct blk_mq_hw_ctx *hctx, struct blk_mq_ctx *start); @@ -65,6 +64,8 @@ void blk_mq_insert_requests(struct blk_mq_hw_ctx *hctx, struct blk_mq_ctx *ctx, /* Used by blk_insert_cloned_request() to issue request directly */ blk_status_t blk_mq_request_issue_directly(struct request *rq); +void blk_mq_try_issue_list_directly(struct blk_mq_hw_ctx *hctx, + struct list_head *list); /* * CPU -> queue mappings @@ -203,4 +204,12 @@ static inline void blk_mq_put_driver_tag(struct request *rq) __blk_mq_put_driver_tag(hctx, rq); } +static inline void blk_mq_clear_mq_map(struct blk_mq_tag_set *set) +{ + int cpu; + + for_each_possible_cpu(cpu) + set->mq_map[cpu] = 0; +} + #endif diff --git a/block/blk-rq-qos.c b/block/blk-rq-qos.c new file mode 100644 index 000000000000..0005dfd568dd --- /dev/null +++ b/block/blk-rq-qos.c @@ -0,0 +1,194 @@ +#include "blk-rq-qos.h" + +/* + * Increment 'v', if 'v' is below 'below'. Returns true if we succeeded, + * false if 'v' + 1 would be bigger than 'below'. + */ +static bool atomic_inc_below(atomic_t *v, unsigned int below) +{ + unsigned int cur = atomic_read(v); + + for (;;) { + unsigned int old; + + if (cur >= below) + return false; + old = atomic_cmpxchg(v, cur, cur + 1); + if (old == cur) + break; + cur = old; + } + + return true; +} + +bool rq_wait_inc_below(struct rq_wait *rq_wait, unsigned int limit) +{ + return atomic_inc_below(&rq_wait->inflight, limit); +} + +void rq_qos_cleanup(struct request_queue *q, struct bio *bio) +{ + struct rq_qos *rqos; + + for (rqos = q->rq_qos; rqos; rqos = rqos->next) { + if (rqos->ops->cleanup) + rqos->ops->cleanup(rqos, bio); + } +} + +void rq_qos_done(struct request_queue *q, struct request *rq) +{ + struct rq_qos *rqos; + + for (rqos = q->rq_qos; rqos; rqos = rqos->next) { + if (rqos->ops->done) + rqos->ops->done(rqos, rq); + } +} + +void rq_qos_issue(struct request_queue *q, struct request *rq) +{ + struct rq_qos *rqos; + + for(rqos = q->rq_qos; rqos; rqos = rqos->next) { + if (rqos->ops->issue) + rqos->ops->issue(rqos, rq); + } +} + +void rq_qos_requeue(struct request_queue *q, struct request *rq) +{ + struct rq_qos *rqos; + + for(rqos = q->rq_qos; rqos; rqos = rqos->next) { + if (rqos->ops->requeue) + rqos->ops->requeue(rqos, rq); + } +} + +void rq_qos_throttle(struct request_queue *q, struct bio *bio, + spinlock_t *lock) +{ + struct rq_qos *rqos; + + for(rqos = q->rq_qos; rqos; rqos = rqos->next) { + if (rqos->ops->throttle) + rqos->ops->throttle(rqos, bio, lock); + } +} + +void rq_qos_track(struct request_queue *q, struct request *rq, struct bio *bio) +{ + struct rq_qos *rqos; + + for(rqos = q->rq_qos; rqos; rqos = rqos->next) { + if (rqos->ops->track) + rqos->ops->track(rqos, rq, bio); + } +} + +void rq_qos_done_bio(struct request_queue *q, struct bio *bio) +{ + struct rq_qos *rqos; + + for(rqos = q->rq_qos; rqos; rqos = rqos->next) { + if (rqos->ops->done_bio) + rqos->ops->done_bio(rqos, bio); + } +} + +/* + * Return true, if we can't increase the depth further by scaling + */ +bool rq_depth_calc_max_depth(struct rq_depth *rqd) +{ + unsigned int depth; + bool ret = false; + + /* + * For QD=1 devices, this is a special case. It's important for those + * to have one request ready when one completes, so force a depth of + * 2 for those devices. On the backend, it'll be a depth of 1 anyway, + * since the device can't have more than that in flight. If we're + * scaling down, then keep a setting of 1/1/1. + */ + if (rqd->queue_depth == 1) { + if (rqd->scale_step > 0) + rqd->max_depth = 1; + else { + rqd->max_depth = 2; + ret = true; + } + } else { + /* + * scale_step == 0 is our default state. If we have suffered + * latency spikes, step will be > 0, and we shrink the + * allowed write depths. If step is < 0, we're only doing + * writes, and we allow a temporarily higher depth to + * increase performance. + */ + depth = min_t(unsigned int, rqd->default_depth, + rqd->queue_depth); + if (rqd->scale_step > 0) + depth = 1 + ((depth - 1) >> min(31, rqd->scale_step)); + else if (rqd->scale_step < 0) { + unsigned int maxd = 3 * rqd->queue_depth / 4; + + depth = 1 + ((depth - 1) << -rqd->scale_step); + if (depth > maxd) { + depth = maxd; + ret = true; + } + } + + rqd->max_depth = depth; + } + + return ret; +} + +void rq_depth_scale_up(struct rq_depth *rqd) +{ + /* + * Hit max in previous round, stop here + */ + if (rqd->scaled_max) + return; + + rqd->scale_step--; + + rqd->scaled_max = rq_depth_calc_max_depth(rqd); +} + +/* + * Scale rwb down. If 'hard_throttle' is set, do it quicker, since we + * had a latency violation. + */ +void rq_depth_scale_down(struct rq_depth *rqd, bool hard_throttle) +{ + /* + * Stop scaling down when we've hit the limit. This also prevents + * ->scale_step from going to crazy values, if the device can't + * keep up. + */ + if (rqd->max_depth == 1) + return; + + if (rqd->scale_step < 0 && hard_throttle) + rqd->scale_step = 0; + else + rqd->scale_step++; + + rqd->scaled_max = false; + rq_depth_calc_max_depth(rqd); +} + +void rq_qos_exit(struct request_queue *q) +{ + while (q->rq_qos) { + struct rq_qos *rqos = q->rq_qos; + q->rq_qos = rqos->next; + rqos->ops->exit(rqos); + } +} diff --git a/block/blk-rq-qos.h b/block/blk-rq-qos.h new file mode 100644 index 000000000000..32b02efbfa66 --- /dev/null +++ b/block/blk-rq-qos.h @@ -0,0 +1,109 @@ +#ifndef RQ_QOS_H +#define RQ_QOS_H + +#include <linux/kernel.h> +#include <linux/blkdev.h> +#include <linux/blk_types.h> +#include <linux/atomic.h> +#include <linux/wait.h> + +enum rq_qos_id { + RQ_QOS_WBT, + RQ_QOS_CGROUP, +}; + +struct rq_wait { + wait_queue_head_t wait; + atomic_t inflight; +}; + +struct rq_qos { + struct rq_qos_ops *ops; + struct request_queue *q; + enum rq_qos_id id; + struct rq_qos *next; +}; + +struct rq_qos_ops { + void (*throttle)(struct rq_qos *, struct bio *, spinlock_t *); + void (*track)(struct rq_qos *, struct request *, struct bio *); + void (*issue)(struct rq_qos *, struct request *); + void (*requeue)(struct rq_qos *, struct request *); + void (*done)(struct rq_qos *, struct request *); + void (*done_bio)(struct rq_qos *, struct bio *); + void (*cleanup)(struct rq_qos *, struct bio *); + void (*exit)(struct rq_qos *); +}; + +struct rq_depth { + unsigned int max_depth; + + int scale_step; + bool scaled_max; + + unsigned int queue_depth; + unsigned int default_depth; +}; + +static inline struct rq_qos *rq_qos_id(struct request_queue *q, + enum rq_qos_id id) +{ + struct rq_qos *rqos; + for (rqos = q->rq_qos; rqos; rqos = rqos->next) { + if (rqos->id == id) + break; + } + return rqos; +} + +static inline struct rq_qos *wbt_rq_qos(struct request_queue *q) +{ + return rq_qos_id(q, RQ_QOS_WBT); +} + +static inline struct rq_qos *blkcg_rq_qos(struct request_queue *q) +{ + return rq_qos_id(q, RQ_QOS_CGROUP); +} + +static inline void rq_wait_init(struct rq_wait *rq_wait) +{ + atomic_set(&rq_wait->inflight, 0); + init_waitqueue_head(&rq_wait->wait); +} + +static inline void rq_qos_add(struct request_queue *q, struct rq_qos *rqos) +{ + rqos->next = q->rq_qos; + q->rq_qos = rqos; +} + +static inline void rq_qos_del(struct request_queue *q, struct rq_qos *rqos) +{ + struct rq_qos *cur, *prev = NULL; + for (cur = q->rq_qos; cur; cur = cur->next) { + if (cur == rqos) { + if (prev) + prev->next = rqos->next; + else + q->rq_qos = cur; + break; + } + prev = cur; + } +} + +bool rq_wait_inc_below(struct rq_wait *rq_wait, unsigned int limit); +void rq_depth_scale_up(struct rq_depth *rqd); +void rq_depth_scale_down(struct rq_depth *rqd, bool hard_throttle); +bool rq_depth_calc_max_depth(struct rq_depth *rqd); + +void rq_qos_cleanup(struct request_queue *, struct bio *); +void rq_qos_done(struct request_queue *, struct request *); +void rq_qos_issue(struct request_queue *, struct request *); +void rq_qos_requeue(struct request_queue *, struct request *); +void rq_qos_done_bio(struct request_queue *q, struct bio *bio); +void rq_qos_throttle(struct request_queue *, struct bio *, spinlock_t *); +void rq_qos_track(struct request_queue *q, struct request *, struct bio *); +void rq_qos_exit(struct request_queue *); +#endif diff --git a/block/blk-settings.c b/block/blk-settings.c index d1de71124656..ffd459969689 100644 --- a/block/blk-settings.c +++ b/block/blk-settings.c @@ -128,7 +128,7 @@ void blk_set_stacking_limits(struct queue_limits *lim) /* Inherit limits from component devices */ lim->max_segments = USHRT_MAX; - lim->max_discard_segments = 1; + lim->max_discard_segments = USHRT_MAX; lim->max_hw_sectors = UINT_MAX; lim->max_segment_size = UINT_MAX; lim->max_sectors = UINT_MAX; @@ -875,7 +875,7 @@ EXPORT_SYMBOL_GPL(blk_queue_flush_queueable); void blk_set_queue_depth(struct request_queue *q, unsigned int depth) { q->queue_depth = depth; - wbt_set_queue_depth(q->rq_wb, depth); + wbt_set_queue_depth(q, depth); } EXPORT_SYMBOL(blk_set_queue_depth); @@ -900,7 +900,7 @@ void blk_queue_write_cache(struct request_queue *q, bool wc, bool fua) queue_flag_clear(QUEUE_FLAG_FUA, q); spin_unlock_irq(q->queue_lock); - wbt_set_write_cache(q->rq_wb, test_bit(QUEUE_FLAG_WC, &q->queue_flags)); + wbt_set_write_cache(q, test_bit(QUEUE_FLAG_WC, &q->queue_flags)); } EXPORT_SYMBOL_GPL(blk_queue_write_cache); diff --git a/block/blk-stat.c b/block/blk-stat.c index 175c143ac5b9..7587b1c3caaf 100644 --- a/block/blk-stat.c +++ b/block/blk-stat.c @@ -17,7 +17,7 @@ struct blk_queue_stats { bool enable_accounting; }; -static void blk_stat_init(struct blk_rq_stat *stat) +void blk_rq_stat_init(struct blk_rq_stat *stat) { stat->min = -1ULL; stat->max = stat->nr_samples = stat->mean = 0; @@ -25,7 +25,7 @@ static void blk_stat_init(struct blk_rq_stat *stat) } /* src is a per-cpu stat, mean isn't initialized */ -static void blk_stat_sum(struct blk_rq_stat *dst, struct blk_rq_stat *src) +void blk_rq_stat_sum(struct blk_rq_stat *dst, struct blk_rq_stat *src) { if (!src->nr_samples) return; @@ -39,7 +39,7 @@ static void blk_stat_sum(struct blk_rq_stat *dst, struct blk_rq_stat *src) dst->nr_samples += src->nr_samples; } -static void __blk_stat_add(struct blk_rq_stat *stat, u64 value) +void blk_rq_stat_add(struct blk_rq_stat *stat, u64 value) { stat->min = min(stat->min, value); stat->max = max(stat->max, value); @@ -69,7 +69,7 @@ void blk_stat_add(struct request *rq, u64 now) continue; stat = &get_cpu_ptr(cb->cpu_stat)[bucket]; - __blk_stat_add(stat, value); + blk_rq_stat_add(stat, value); put_cpu_ptr(cb->cpu_stat); } rcu_read_unlock(); @@ -82,15 +82,15 @@ static void blk_stat_timer_fn(struct timer_list *t) int cpu; for (bucket = 0; bucket < cb->buckets; bucket++) - blk_stat_init(&cb->stat[bucket]); + blk_rq_stat_init(&cb->stat[bucket]); for_each_online_cpu(cpu) { struct blk_rq_stat *cpu_stat; cpu_stat = per_cpu_ptr(cb->cpu_stat, cpu); for (bucket = 0; bucket < cb->buckets; bucket++) { - blk_stat_sum(&cb->stat[bucket], &cpu_stat[bucket]); - blk_stat_init(&cpu_stat[bucket]); + blk_rq_stat_sum(&cb->stat[bucket], &cpu_stat[bucket]); + blk_rq_stat_init(&cpu_stat[bucket]); } } @@ -143,7 +143,7 @@ void blk_stat_add_callback(struct request_queue *q, cpu_stat = per_cpu_ptr(cb->cpu_stat, cpu); for (bucket = 0; bucket < cb->buckets; bucket++) - blk_stat_init(&cpu_stat[bucket]); + blk_rq_stat_init(&cpu_stat[bucket]); } spin_lock(&q->stats->lock); diff --git a/block/blk-stat.h b/block/blk-stat.h index 78399cdde9c9..f4a1568e81a4 100644 --- a/block/blk-stat.h +++ b/block/blk-stat.h @@ -159,4 +159,8 @@ static inline void blk_stat_activate_msecs(struct blk_stat_callback *cb, mod_timer(&cb->timer, jiffies + msecs_to_jiffies(msecs)); } +void blk_rq_stat_add(struct blk_rq_stat *, u64); +void blk_rq_stat_sum(struct blk_rq_stat *, struct blk_rq_stat *); +void blk_rq_stat_init(struct blk_rq_stat *); + #endif diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index 94987b1f69e1..bb109bb0a055 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -422,16 +422,16 @@ static ssize_t queue_poll_store(struct request_queue *q, const char *page, static ssize_t queue_wb_lat_show(struct request_queue *q, char *page) { - if (!q->rq_wb) + if (!wbt_rq_qos(q)) return -EINVAL; - return sprintf(page, "%llu\n", div_u64(q->rq_wb->min_lat_nsec, 1000)); + return sprintf(page, "%llu\n", div_u64(wbt_get_min_lat(q), 1000)); } static ssize_t queue_wb_lat_store(struct request_queue *q, const char *page, size_t count) { - struct rq_wb *rwb; + struct rq_qos *rqos; ssize_t ret; s64 val; @@ -441,23 +441,21 @@ static ssize_t queue_wb_lat_store(struct request_queue *q, const char *page, if (val < -1) return -EINVAL; - rwb = q->rq_wb; - if (!rwb) { + rqos = wbt_rq_qos(q); + if (!rqos) { ret = wbt_init(q); if (ret) return ret; } - rwb = q->rq_wb; if (val == -1) - rwb->min_lat_nsec = wbt_default_latency_nsec(q); + val = wbt_default_latency_nsec(q); else if (val >= 0) - rwb->min_lat_nsec = val * 1000ULL; + val *= 1000ULL; - if (rwb->enable_state == WBT_STATE_ON_DEFAULT) - rwb->enable_state = WBT_STATE_ON_MANUAL; + wbt_set_min_lat(q, val); - wbt_update_limits(rwb); + wbt_update_limits(q); return count; } @@ -804,6 +802,21 @@ static void __blk_release_queue(struct work_struct *work) blk_stat_remove_callback(q, q->poll_cb); blk_stat_free_callback(q->poll_cb); + if (!blk_queue_dead(q)) { + /* + * Last reference was dropped without having called + * blk_cleanup_queue(). + */ + WARN_ONCE(blk_queue_init_done(q), + "request queue %p has been registered but blk_cleanup_queue() has not been called for that queue\n", + q); + blk_exit_queue(q); + } + + WARN(blk_queue_root_blkg(q), + "request queue %p is being released but it has not yet been removed from the blkcg controller\n", + q); + blk_free_queue_stats(q->stats); blk_exit_rl(q, &q->root_rl); @@ -964,7 +977,7 @@ void blk_unregister_queue(struct gendisk *disk) kobject_del(&q->kobj); blk_trace_remove_sysfs(disk_to_dev(disk)); - wbt_exit(q); + rq_qos_exit(q); mutex_lock(&q->sysfs_lock); if (q->request_fn || (q->mq_ops && q->elevator)) diff --git a/block/blk-throttle.c b/block/blk-throttle.c index 82282e6fdcf8..a3eede00d302 100644 --- a/block/blk-throttle.c +++ b/block/blk-throttle.c @@ -579,8 +579,10 @@ static void blk_throtl_update_limit_valid(struct throtl_data *td) struct throtl_grp *tg = blkg_to_tg(blkg); if (tg->bps[READ][LIMIT_LOW] || tg->bps[WRITE][LIMIT_LOW] || - tg->iops[READ][LIMIT_LOW] || tg->iops[WRITE][LIMIT_LOW]) + tg->iops[READ][LIMIT_LOW] || tg->iops[WRITE][LIMIT_LOW]) { low_valid = true; + break; + } } rcu_read_unlock(); @@ -920,12 +922,7 @@ static bool tg_with_in_iops_limit(struct throtl_grp *tg, struct bio *bio, } /* Calc approx time to dispatch */ - jiffy_wait = ((tg->io_disp[rw] + 1) * HZ) / tg_iops_limit(tg, rw) + 1; - - if (jiffy_wait > jiffy_elapsed) - jiffy_wait = jiffy_wait - jiffy_elapsed; - else - jiffy_wait = 1; + jiffy_wait = jiffy_elapsed_rnd - jiffy_elapsed; if (wait) *wait = jiffy_wait; @@ -2132,12 +2129,8 @@ static inline void throtl_update_latency_buckets(struct throtl_data *td) static void blk_throtl_assoc_bio(struct throtl_grp *tg, struct bio *bio) { #ifdef CONFIG_BLK_DEV_THROTTLING_LOW - if (bio->bi_css) { - if (bio->bi_cg_private) - blkg_put(tg_to_blkg(bio->bi_cg_private)); - bio->bi_cg_private = tg; - blkg_get(tg_to_blkg(tg)); - } + if (bio->bi_css) + bio_associate_blkg(bio, tg_to_blkg(tg)); bio_issue_init(&bio->bi_issue, bio_sectors(bio)); #endif } @@ -2285,6 +2278,7 @@ void blk_throtl_stat_add(struct request *rq, u64 time_ns) void blk_throtl_bio_endio(struct bio *bio) { + struct blkcg_gq *blkg; struct throtl_grp *tg; u64 finish_time_ns; unsigned long finish_time; @@ -2292,20 +2286,18 @@ void blk_throtl_bio_endio(struct bio *bio) unsigned long lat; int rw = bio_data_dir(bio); - tg = bio->bi_cg_private; - if (!tg) + blkg = bio->bi_blkg; + if (!blkg) return; - bio->bi_cg_private = NULL; + tg = blkg_to_tg(blkg); finish_time_ns = ktime_get_ns(); tg->last_finish_time = finish_time_ns >> 10; start_time = bio_issue_time(&bio->bi_issue) >> 10; finish_time = __bio_issue_time(finish_time_ns) >> 10; - if (!start_time || finish_time <= start_time) { - blkg_put(tg_to_blkg(tg)); + if (!start_time || finish_time <= start_time) return; - } lat = finish_time - start_time; /* this is only for bio based driver */ @@ -2334,8 +2326,6 @@ void blk_throtl_bio_endio(struct bio *bio) tg->bio_cnt /= 2; tg->bad_bio_cnt /= 2; } - - blkg_put(tg_to_blkg(tg)); } #endif diff --git a/block/blk-wbt.c b/block/blk-wbt.c index 4f89b28fa652..1d94a20374fc 100644 --- a/block/blk-wbt.c +++ b/block/blk-wbt.c @@ -25,6 +25,7 @@ #include <linux/swap.h> #include "blk-wbt.h" +#include "blk-rq-qos.h" #define CREATE_TRACE_POINTS #include <trace/events/wbt.h> @@ -78,28 +79,6 @@ static inline bool rwb_enabled(struct rq_wb *rwb) return rwb && rwb->wb_normal != 0; } -/* - * Increment 'v', if 'v' is below 'below'. Returns true if we succeeded, - * false if 'v' + 1 would be bigger than 'below'. - */ -static bool atomic_inc_below(atomic_t *v, int below) -{ - int cur = atomic_read(v); - - for (;;) { - int old; - - if (cur >= below) - return false; - old = atomic_cmpxchg(v, cur, cur + 1); - if (old == cur) - break; - cur = old; - } - - return true; -} - static void wb_timestamp(struct rq_wb *rwb, unsigned long *var) { if (rwb_enabled(rwb)) { @@ -116,7 +95,7 @@ static void wb_timestamp(struct rq_wb *rwb, unsigned long *var) */ static bool wb_recent_wait(struct rq_wb *rwb) { - struct bdi_writeback *wb = &rwb->queue->backing_dev_info->wb; + struct bdi_writeback *wb = &rwb->rqos.q->backing_dev_info->wb; return time_before(jiffies, wb->dirty_sleep + HZ); } @@ -144,8 +123,9 @@ static void rwb_wake_all(struct rq_wb *rwb) } } -void __wbt_done(struct rq_wb *rwb, enum wbt_flags wb_acct) +static void __wbt_done(struct rq_qos *rqos, enum wbt_flags wb_acct) { + struct rq_wb *rwb = RQWB(rqos); struct rq_wait *rqw; int inflight, limit; @@ -186,7 +166,7 @@ void __wbt_done(struct rq_wb *rwb, enum wbt_flags wb_acct) int diff = limit - inflight; if (!inflight || diff >= rwb->wb_background / 2) - wake_up_all(&rqw->wait); + wake_up(&rqw->wait); } } @@ -194,10 +174,9 @@ void __wbt_done(struct rq_wb *rwb, enum wbt_flags wb_acct) * Called on completion of a request. Note that it's also called when * a request is merged, when the request gets freed. */ -void wbt_done(struct rq_wb *rwb, struct request *rq) +static void wbt_done(struct rq_qos *rqos, struct request *rq) { - if (!rwb) - return; + struct rq_wb *rwb = RQWB(rqos); if (!wbt_is_tracked(rq)) { if (rwb->sync_cookie == rq) { @@ -209,72 +188,11 @@ void wbt_done(struct rq_wb *rwb, struct request *rq) wb_timestamp(rwb, &rwb->last_comp); } else { WARN_ON_ONCE(rq == rwb->sync_cookie); - __wbt_done(rwb, wbt_flags(rq)); + __wbt_done(rqos, wbt_flags(rq)); } wbt_clear_state(rq); } -/* - * Return true, if we can't increase the depth further by scaling - */ -static bool calc_wb_limits(struct rq_wb *rwb) -{ - unsigned int depth; - bool ret = false; - - if (!rwb->min_lat_nsec) { - rwb->wb_max = rwb->wb_normal = rwb->wb_background = 0; - return false; - } - - /* - * For QD=1 devices, this is a special case. It's important for those - * to have one request ready when one completes, so force a depth of - * 2 for those devices. On the backend, it'll be a depth of 1 anyway, - * since the device can't have more than that in flight. If we're - * scaling down, then keep a setting of 1/1/1. - */ - if (rwb->queue_depth == 1) { - if (rwb->scale_step > 0) - rwb->wb_max = rwb->wb_normal = 1; - else { - rwb->wb_max = rwb->wb_normal = 2; - ret = true; - } - rwb->wb_background = 1; - } else { - /* - * scale_step == 0 is our default state. If we have suffered - * latency spikes, step will be > 0, and we shrink the - * allowed write depths. If step is < 0, we're only doing - * writes, and we allow a temporarily higher depth to - * increase performance. - */ - depth = min_t(unsigned int, RWB_DEF_DEPTH, rwb->queue_depth); - if (rwb->scale_step > 0) - depth = 1 + ((depth - 1) >> min(31, rwb->scale_step)); - else if (rwb->scale_step < 0) { - unsigned int maxd = 3 * rwb->queue_depth / 4; - - depth = 1 + ((depth - 1) << -rwb->scale_step); - if (depth > maxd) { - depth = maxd; - ret = true; - } - } - - /* - * Set our max/normal/bg queue depths based on how far - * we have scaled down (->scale_step). - */ - rwb->wb_max = depth; - rwb->wb_normal = (rwb->wb_max + 1) / 2; - rwb->wb_background = (rwb->wb_max + 3) / 4; - } - - return ret; -} - static inline bool stat_sample_valid(struct blk_rq_stat *stat) { /* @@ -307,7 +225,8 @@ enum { static int latency_exceeded(struct rq_wb *rwb, struct blk_rq_stat *stat) { - struct backing_dev_info *bdi = rwb->queue->backing_dev_info; + struct backing_dev_info *bdi = rwb->rqos.q->backing_dev_info; + struct rq_depth *rqd = &rwb->rq_depth; u64 thislat; /* @@ -351,7 +270,7 @@ static int latency_exceeded(struct rq_wb *rwb, struct blk_rq_stat *stat) return LAT_EXCEEDED; } - if (rwb->scale_step) + if (rqd->scale_step) trace_wbt_stat(bdi, stat); return LAT_OK; @@ -359,58 +278,48 @@ static int latency_exceeded(struct rq_wb *rwb, struct blk_rq_stat *stat) static void rwb_trace_step(struct rq_wb *rwb, const char *msg) { - struct backing_dev_info *bdi = rwb->queue->backing_dev_info; + struct backing_dev_info *bdi = rwb->rqos.q->backing_dev_info; + struct rq_depth *rqd = &rwb->rq_depth; - trace_wbt_step(bdi, msg, rwb->scale_step, rwb->cur_win_nsec, - rwb->wb_background, rwb->wb_normal, rwb->wb_max); + trace_wbt_step(bdi, msg, rqd->scale_step, rwb->cur_win_nsec, + rwb->wb_background, rwb->wb_normal, rqd->max_depth); } -static void scale_up(struct rq_wb *rwb) +static void calc_wb_limits(struct rq_wb *rwb) { - /* - * Hit max in previous round, stop here - */ - if (rwb->scaled_max) - return; + if (rwb->min_lat_nsec == 0) { + rwb->wb_normal = rwb->wb_background = 0; + } else if (rwb->rq_depth.max_depth <= 2) { + rwb->wb_normal = rwb->rq_depth.max_depth; + rwb->wb_background = 1; + } else { + rwb->wb_normal = (rwb->rq_depth.max_depth + 1) / 2; + rwb->wb_background = (rwb->rq_depth.max_depth + 3) / 4; + } +} - rwb->scale_step--; +static void scale_up(struct rq_wb *rwb) +{ + rq_depth_scale_up(&rwb->rq_depth); + calc_wb_limits(rwb); rwb->unknown_cnt = 0; - - rwb->scaled_max = calc_wb_limits(rwb); - - rwb_wake_all(rwb); - - rwb_trace_step(rwb, "step up"); + rwb_trace_step(rwb, "scale up"); } -/* - * Scale rwb down. If 'hard_throttle' is set, do it quicker, since we - * had a latency violation. - */ static void scale_down(struct rq_wb *rwb, bool hard_throttle) { - /* - * Stop scaling down when we've hit the limit. This also prevents - * ->scale_step from going to crazy values, if the device can't - * keep up. - */ - if (rwb->wb_max == 1) - return; - - if (rwb->scale_step < 0 && hard_throttle) - rwb->scale_step = 0; - else - rwb->scale_step++; - - rwb->scaled_max = false; - rwb->unknown_cnt = 0; + rq_depth_scale_down(&rwb->rq_depth, hard_throttle); calc_wb_limits(rwb); - rwb_trace_step(rwb, "step down"); + rwb->unknown_cnt = 0; + rwb_wake_all(rwb); + rwb_trace_step(rwb, "scale down"); } static void rwb_arm_timer(struct rq_wb *rwb) { - if (rwb->scale_step > 0) { + struct rq_depth *rqd = &rwb->rq_depth; + + if (rqd->scale_step > 0) { /* * We should speed this up, using some variant of a fast * integer inverse square root calculation. Since we only do @@ -418,7 +327,7 @@ static void rwb_arm_timer(struct rq_wb *rwb) * though. */ rwb->cur_win_nsec = div_u64(rwb->win_nsec << 4, - int_sqrt((rwb->scale_step + 1) << 8)); + int_sqrt((rqd->scale_step + 1) << 8)); } else { /* * For step < 0, we don't want to increase/decrease the @@ -433,12 +342,13 @@ static void rwb_arm_timer(struct rq_wb *rwb) static void wb_timer_fn(struct blk_stat_callback *cb) { struct rq_wb *rwb = cb->data; + struct rq_depth *rqd = &rwb->rq_depth; unsigned int inflight = wbt_inflight(rwb); int status; status = latency_exceeded(rwb, cb->stat); - trace_wbt_timer(rwb->queue->backing_dev_info, status, rwb->scale_step, + trace_wbt_timer(rwb->rqos.q->backing_dev_info, status, rqd->scale_step, inflight); /* @@ -469,9 +379,9 @@ static void wb_timer_fn(struct blk_stat_callback *cb) * currently don't have a valid read/write sample. For that * case, slowly return to center state (step == 0). */ - if (rwb->scale_step > 0) + if (rqd->scale_step > 0) scale_up(rwb); - else if (rwb->scale_step < 0) + else if (rqd->scale_step < 0) scale_down(rwb, false); break; default: @@ -481,19 +391,50 @@ static void wb_timer_fn(struct blk_stat_callback *cb) /* * Re-arm timer, if we have IO in flight */ - if (rwb->scale_step || inflight) + if (rqd->scale_step || inflight) rwb_arm_timer(rwb); } -void wbt_update_limits(struct rq_wb *rwb) +static void __wbt_update_limits(struct rq_wb *rwb) { - rwb->scale_step = 0; - rwb->scaled_max = false; + struct rq_depth *rqd = &rwb->rq_depth; + + rqd->scale_step = 0; + rqd->scaled_max = false; + + rq_depth_calc_max_depth(rqd); calc_wb_limits(rwb); rwb_wake_all(rwb); } +void wbt_update_limits(struct request_queue *q) +{ + struct rq_qos *rqos = wbt_rq_qos(q); + if (!rqos) + return; + __wbt_update_limits(RQWB(rqos)); +} + +u64 wbt_get_min_lat(struct request_queue *q) +{ + struct rq_qos *rqos = wbt_rq_qos(q); + if (!rqos) + return 0; + return RQWB(rqos)->min_lat_nsec; +} + +void wbt_set_min_lat(struct request_queue *q, u64 val) +{ + struct rq_qos *rqos = wbt_rq_qos(q); + if (!rqos) + return; + RQWB(rqos)->min_lat_nsec = val; + RQWB(rqos)->enable_state = WBT_STATE_ON_MANUAL; + __wbt_update_limits(RQWB(rqos)); +} + + static bool close_io(struct rq_wb *rwb) { const unsigned long now = jiffies; @@ -520,7 +461,7 @@ static inline unsigned int get_limit(struct rq_wb *rwb, unsigned long rw) * IO for a bit. */ if ((rw & REQ_HIPRIO) || wb_recent_wait(rwb) || current_is_kswapd()) - limit = rwb->wb_max; + limit = rwb->rq_depth.max_depth; else if ((rw & REQ_BACKGROUND) || close_io(rwb)) { /* * If less than 100ms since we completed unrelated IO, @@ -533,30 +474,6 @@ static inline unsigned int get_limit(struct rq_wb *rwb, unsigned long rw) return limit; } -static inline bool may_queue(struct rq_wb *rwb, struct rq_wait *rqw, - wait_queue_entry_t *wait, unsigned long rw) -{ - /* - * inc it here even if disabled, since we'll dec it at completion. - * this only happens if the task was sleeping in __wbt_wait(), - * and someone turned it off at the same time. - */ - if (!rwb_enabled(rwb)) { - atomic_inc(&rqw->inflight); - return true; - } - - /* - * If the waitqueue is already active and we are not the next - * in line to be woken up, wait for our turn. - */ - if (waitqueue_active(&rqw->wait) && - rqw->wait.head.next != &wait->entry) - return false; - - return atomic_inc_below(&rqw->inflight, get_limit(rwb, rw)); -} - /* * Block if we will exceed our limit, or if we are currently waiting for * the timer to kick off queuing again. @@ -567,16 +484,32 @@ static void __wbt_wait(struct rq_wb *rwb, enum wbt_flags wb_acct, __acquires(lock) { struct rq_wait *rqw = get_rq_wait(rwb, wb_acct); - DEFINE_WAIT(wait); + DECLARE_WAITQUEUE(wait, current); - if (may_queue(rwb, rqw, &wait, rw)) + /* + * inc it here even if disabled, since we'll dec it at completion. + * this only happens if the task was sleeping in __wbt_wait(), + * and someone turned it off at the same time. + */ + if (!rwb_enabled(rwb)) { + atomic_inc(&rqw->inflight); return; + } + if (!waitqueue_active(&rqw->wait) + && rq_wait_inc_below(rqw, get_limit(rwb, rw))) + return; + + add_wait_queue_exclusive(&rqw->wait, &wait); do { - prepare_to_wait_exclusive(&rqw->wait, &wait, - TASK_UNINTERRUPTIBLE); + set_current_state(TASK_UNINTERRUPTIBLE); - if (may_queue(rwb, rqw, &wait, rw)) + if (!rwb_enabled(rwb)) { + atomic_inc(&rqw->inflight); + break; + } + + if (rq_wait_inc_below(rqw, get_limit(rwb, rw))) break; if (lock) { @@ -587,7 +520,8 @@ static void __wbt_wait(struct rq_wb *rwb, enum wbt_flags wb_acct, io_schedule(); } while (1); - finish_wait(&rqw->wait, &wait); + __set_current_state(TASK_RUNNING); + remove_wait_queue(&rqw->wait, &wait); } static inline bool wbt_should_throttle(struct rq_wb *rwb, struct bio *bio) @@ -608,43 +542,72 @@ static inline bool wbt_should_throttle(struct rq_wb *rwb, struct bio *bio) } } +static enum wbt_flags bio_to_wbt_flags(struct rq_wb *rwb, struct bio *bio) +{ + enum wbt_flags flags = 0; + + if (bio_op(bio) == REQ_OP_READ) { + flags = WBT_READ; + } else if (wbt_should_throttle(rwb, bio)) { + if (current_is_kswapd()) + flags |= WBT_KSWAPD; + if (bio_op(bio) == REQ_OP_DISCARD) + flags |= WBT_DISCARD; + flags |= WBT_TRACKED; + } + return flags; +} + +static void wbt_cleanup(struct rq_qos *rqos, struct bio *bio) +{ + struct rq_wb *rwb = RQWB(rqos); + enum wbt_flags flags = bio_to_wbt_flags(rwb, bio); + __wbt_done(rqos, flags); +} + /* * Returns true if the IO request should be accounted, false if not. * May sleep, if we have exceeded the writeback limits. Caller can pass * in an irq held spinlock, if it holds one when calling this function. * If we do sleep, we'll release and re-grab it. */ -enum wbt_flags wbt_wait(struct rq_wb *rwb, struct bio *bio, spinlock_t *lock) +static void wbt_wait(struct rq_qos *rqos, struct bio *bio, spinlock_t *lock) { - enum wbt_flags ret = 0; + struct rq_wb *rwb = RQWB(rqos); + enum wbt_flags flags; if (!rwb_enabled(rwb)) - return 0; + return; - if (bio_op(bio) == REQ_OP_READ) - ret = WBT_READ; + flags = bio_to_wbt_flags(rwb, bio); if (!wbt_should_throttle(rwb, bio)) { - if (ret & WBT_READ) + if (flags & WBT_READ) wb_timestamp(rwb, &rwb->last_issue); - return ret; + return; } if (current_is_kswapd()) - ret |= WBT_KSWAPD; + flags |= WBT_KSWAPD; if (bio_op(bio) == REQ_OP_DISCARD) - ret |= WBT_DISCARD; + flags |= WBT_DISCARD; - __wbt_wait(rwb, ret, bio->bi_opf, lock); + __wbt_wait(rwb, flags, bio->bi_opf, lock); if (!blk_stat_is_active(rwb->cb)) rwb_arm_timer(rwb); +} - return ret | WBT_TRACKED; +static void wbt_track(struct rq_qos *rqos, struct request *rq, struct bio *bio) +{ + struct rq_wb *rwb = RQWB(rqos); + rq->wbt_flags |= bio_to_wbt_flags(rwb, bio); } -void wbt_issue(struct rq_wb *rwb, struct request *rq) +void wbt_issue(struct rq_qos *rqos, struct request *rq) { + struct rq_wb *rwb = RQWB(rqos); + if (!rwb_enabled(rwb)) return; @@ -661,8 +624,9 @@ void wbt_issue(struct rq_wb *rwb, struct request *rq) } } -void wbt_requeue(struct rq_wb *rwb, struct request *rq) +void wbt_requeue(struct rq_qos *rqos, struct request *rq) { + struct rq_wb *rwb = RQWB(rqos); if (!rwb_enabled(rwb)) return; if (rq == rwb->sync_cookie) { @@ -671,39 +635,30 @@ void wbt_requeue(struct rq_wb *rwb, struct request *rq) } } -void wbt_set_queue_depth(struct rq_wb *rwb, unsigned int depth) +void wbt_set_queue_depth(struct request_queue *q, unsigned int depth) { - if (rwb) { - rwb->queue_depth = depth; - wbt_update_limits(rwb); + struct rq_qos *rqos = wbt_rq_qos(q); + if (rqos) { + RQWB(rqos)->rq_depth.queue_depth = depth; + __wbt_update_limits(RQWB(rqos)); } } -void wbt_set_write_cache(struct rq_wb *rwb, bool write_cache_on) +void wbt_set_write_cache(struct request_queue *q, bool write_cache_on) { - if (rwb) - rwb->wc = write_cache_on; + struct rq_qos *rqos = wbt_rq_qos(q); + if (rqos) + RQWB(rqos)->wc = write_cache_on; } /* - * Disable wbt, if enabled by default. - */ -void wbt_disable_default(struct request_queue *q) -{ - struct rq_wb *rwb = q->rq_wb; - - if (rwb && rwb->enable_state == WBT_STATE_ON_DEFAULT) - wbt_exit(q); -} -EXPORT_SYMBOL_GPL(wbt_disable_default); - -/* * Enable wbt if defaults are configured that way */ void wbt_enable_default(struct request_queue *q) { + struct rq_qos *rqos = wbt_rq_qos(q); /* Throttling already enabled? */ - if (q->rq_wb) + if (rqos) return; /* Queue not registered? Maybe shutting down... */ @@ -741,6 +696,42 @@ static int wbt_data_dir(const struct request *rq) return -1; } +static void wbt_exit(struct rq_qos *rqos) +{ + struct rq_wb *rwb = RQWB(rqos); + struct request_queue *q = rqos->q; + + blk_stat_remove_callback(q, rwb->cb); + blk_stat_free_callback(rwb->cb); + kfree(rwb); +} + +/* + * Disable wbt, if enabled by default. + */ +void wbt_disable_default(struct request_queue *q) +{ + struct rq_qos *rqos = wbt_rq_qos(q); + struct rq_wb *rwb; + if (!rqos) + return; + rwb = RQWB(rqos); + if (rwb->enable_state == WBT_STATE_ON_DEFAULT) + rwb->wb_normal = 0; +} +EXPORT_SYMBOL_GPL(wbt_disable_default); + + +static struct rq_qos_ops wbt_rqos_ops = { + .throttle = wbt_wait, + .issue = wbt_issue, + .track = wbt_track, + .requeue = wbt_requeue, + .done = wbt_done, + .cleanup = wbt_cleanup, + .exit = wbt_exit, +}; + int wbt_init(struct request_queue *q) { struct rq_wb *rwb; @@ -756,39 +747,29 @@ int wbt_init(struct request_queue *q) return -ENOMEM; } - for (i = 0; i < WBT_NUM_RWQ; i++) { - atomic_set(&rwb->rq_wait[i].inflight, 0); - init_waitqueue_head(&rwb->rq_wait[i].wait); - } + for (i = 0; i < WBT_NUM_RWQ; i++) + rq_wait_init(&rwb->rq_wait[i]); + rwb->rqos.id = RQ_QOS_WBT; + rwb->rqos.ops = &wbt_rqos_ops; + rwb->rqos.q = q; rwb->last_comp = rwb->last_issue = jiffies; - rwb->queue = q; rwb->win_nsec = RWB_WINDOW_NSEC; rwb->enable_state = WBT_STATE_ON_DEFAULT; - wbt_update_limits(rwb); + rwb->wc = 1; + rwb->rq_depth.default_depth = RWB_DEF_DEPTH; + __wbt_update_limits(rwb); /* * Assign rwb and add the stats callback. */ - q->rq_wb = rwb; + rq_qos_add(q, &rwb->rqos); blk_stat_add_callback(q, rwb->cb); rwb->min_lat_nsec = wbt_default_latency_nsec(q); - wbt_set_queue_depth(rwb, blk_queue_depth(q)); - wbt_set_write_cache(rwb, test_bit(QUEUE_FLAG_WC, &q->queue_flags)); + wbt_set_queue_depth(q, blk_queue_depth(q)); + wbt_set_write_cache(q, test_bit(QUEUE_FLAG_WC, &q->queue_flags)); return 0; } - -void wbt_exit(struct request_queue *q) -{ - struct rq_wb *rwb = q->rq_wb; - - if (rwb) { - blk_stat_remove_callback(q, rwb->cb); - blk_stat_free_callback(rwb->cb); - q->rq_wb = NULL; - kfree(rwb); - } -} diff --git a/block/blk-wbt.h b/block/blk-wbt.h index 300df531d0a6..f47218d5b3b2 100644 --- a/block/blk-wbt.h +++ b/block/blk-wbt.h @@ -9,6 +9,7 @@ #include <linux/ktime.h> #include "blk-stat.h" +#include "blk-rq-qos.h" enum wbt_flags { WBT_TRACKED = 1, /* write, tracked for throttling */ @@ -35,20 +36,12 @@ enum { WBT_STATE_ON_MANUAL = 2, }; -struct rq_wait { - wait_queue_head_t wait; - atomic_t inflight; -}; - struct rq_wb { /* * Settings that govern how we throttle */ unsigned int wb_background; /* background writeback */ unsigned int wb_normal; /* normal writeback */ - unsigned int wb_max; /* max throughput writeback */ - int scale_step; - bool scaled_max; short enable_state; /* WBT_STATE_* */ @@ -67,15 +60,20 @@ struct rq_wb { void *sync_cookie; unsigned int wc; - unsigned int queue_depth; unsigned long last_issue; /* last non-throttled issue */ unsigned long last_comp; /* last non-throttled comp */ unsigned long min_lat_nsec; - struct request_queue *queue; + struct rq_qos rqos; struct rq_wait rq_wait[WBT_NUM_RWQ]; + struct rq_depth rq_depth; }; +static inline struct rq_wb *RQWB(struct rq_qos *rqos) +{ + return container_of(rqos, struct rq_wb, rqos); +} + static inline unsigned int wbt_inflight(struct rq_wb *rwb) { unsigned int i, ret = 0; @@ -86,26 +84,19 @@ static inline unsigned int wbt_inflight(struct rq_wb *rwb) return ret; } -#ifdef CONFIG_BLK_WBT -static inline void wbt_track(struct request *rq, enum wbt_flags flags) -{ - rq->wbt_flags |= flags; -} +#ifdef CONFIG_BLK_WBT -void __wbt_done(struct rq_wb *, enum wbt_flags); -void wbt_done(struct rq_wb *, struct request *); -enum wbt_flags wbt_wait(struct rq_wb *, struct bio *, spinlock_t *); int wbt_init(struct request_queue *); -void wbt_exit(struct request_queue *); -void wbt_update_limits(struct rq_wb *); -void wbt_requeue(struct rq_wb *, struct request *); -void wbt_issue(struct rq_wb *, struct request *); +void wbt_update_limits(struct request_queue *); void wbt_disable_default(struct request_queue *); void wbt_enable_default(struct request_queue *); -void wbt_set_queue_depth(struct rq_wb *, unsigned int); -void wbt_set_write_cache(struct rq_wb *, bool); +u64 wbt_get_min_lat(struct request_queue *q); +void wbt_set_min_lat(struct request_queue *q, u64 val); + +void wbt_set_queue_depth(struct request_queue *, unsigned int); +void wbt_set_write_cache(struct request_queue *, bool); u64 wbt_default_latency_nsec(struct request_queue *); @@ -114,43 +105,30 @@ u64 wbt_default_latency_nsec(struct request_queue *); static inline void wbt_track(struct request *rq, enum wbt_flags flags) { } -static inline void __wbt_done(struct rq_wb *rwb, enum wbt_flags flags) -{ -} -static inline void wbt_done(struct rq_wb *rwb, struct request *rq) -{ -} -static inline enum wbt_flags wbt_wait(struct rq_wb *rwb, struct bio *bio, - spinlock_t *lock) -{ - return 0; -} static inline int wbt_init(struct request_queue *q) { return -EINVAL; } -static inline void wbt_exit(struct request_queue *q) -{ -} -static inline void wbt_update_limits(struct rq_wb *rwb) +static inline void wbt_update_limits(struct request_queue *q) { } -static inline void wbt_requeue(struct rq_wb *rwb, struct request *rq) +static inline void wbt_disable_default(struct request_queue *q) { } -static inline void wbt_issue(struct rq_wb *rwb, struct request *rq) +static inline void wbt_enable_default(struct request_queue *q) { } -static inline void wbt_disable_default(struct request_queue *q) +static inline void wbt_set_queue_depth(struct request_queue *q, unsigned int depth) { } -static inline void wbt_enable_default(struct request_queue *q) +static inline void wbt_set_write_cache(struct request_queue *q, bool wc) { } -static inline void wbt_set_queue_depth(struct rq_wb *rwb, unsigned int depth) +static inline u64 wbt_get_min_lat(struct request_queue *q) { + return 0; } -static inline void wbt_set_write_cache(struct rq_wb *rwb, bool wc) +static inline void wbt_set_min_lat(struct request_queue *q, u64 val) { } static inline u64 wbt_default_latency_nsec(struct request_queue *q) diff --git a/block/blk-zoned.c b/block/blk-zoned.c index 51000914e23f..c461cf63f1f4 100644 --- a/block/blk-zoned.c +++ b/block/blk-zoned.c @@ -200,7 +200,7 @@ int blkdev_report_zones(struct block_device *bdev, /* Get header in the first page */ ofst = 0; if (!nr_rep) { - hdr = (struct blk_zone_report_hdr *) addr; + hdr = addr; nr_rep = hdr->nr_zones; ofst = sizeof(struct blk_zone_report_hdr); } diff --git a/block/blk.h b/block/blk.h index 8d23aea96ce9..d4d67e948920 100644 --- a/block/blk.h +++ b/block/blk.h @@ -130,6 +130,7 @@ void blk_free_flush_queue(struct blk_flush_queue *q); int blk_init_rl(struct request_list *rl, struct request_queue *q, gfp_t gfp_mask); void blk_exit_rl(struct request_queue *q, struct request_list *rl); +void blk_exit_queue(struct request_queue *q); void blk_rq_bio_prep(struct request_queue *q, struct request *rq, struct bio *bio); void blk_queue_bypass_start(struct request_queue *q); @@ -412,4 +413,10 @@ static inline void blk_queue_bounce(struct request_queue *q, struct bio **bio) extern void blk_drain_queue(struct request_queue *q); +#ifdef CONFIG_BLK_CGROUP_IOLATENCY +extern int blk_iolatency_init(struct request_queue *q); +#else +static inline int blk_iolatency_init(struct request_queue *q) { return 0; } +#endif + #endif /* BLK_INTERNAL_H */ diff --git a/block/bounce.c b/block/bounce.c index fd31347b7836..bc63b3a2d18c 100644 --- a/block/bounce.c +++ b/block/bounce.c @@ -195,6 +195,73 @@ static void bounce_end_io_read_isa(struct bio *bio) __bounce_end_io_read(bio, &isa_page_pool); } +static struct bio *bounce_clone_bio(struct bio *bio_src, gfp_t gfp_mask, + struct bio_set *bs) +{ + struct bvec_iter iter; + struct bio_vec bv; + struct bio *bio; + + /* + * Pre immutable biovecs, __bio_clone() used to just do a memcpy from + * bio_src->bi_io_vec to bio->bi_io_vec. + * + * We can't do that anymore, because: + * + * - The point of cloning the biovec is to produce a bio with a biovec + * the caller can modify: bi_idx and bi_bvec_done should be 0. + * + * - The original bio could've had more than BIO_MAX_PAGES biovecs; if + * we tried to clone the whole thing bio_alloc_bioset() would fail. + * But the clone should succeed as long as the number of biovecs we + * actually need to allocate is fewer than BIO_MAX_PAGES. + * + * - Lastly, bi_vcnt should not be looked at or relied upon by code + * that does not own the bio - reason being drivers don't use it for + * iterating over the biovec anymore, so expecting it to be kept up + * to date (i.e. for clones that share the parent biovec) is just + * asking for trouble and would force extra work on + * __bio_clone_fast() anyways. + */ + + bio = bio_alloc_bioset(gfp_mask, bio_segments(bio_src), bs); + if (!bio) + return NULL; + bio->bi_disk = bio_src->bi_disk; + bio->bi_opf = bio_src->bi_opf; + bio->bi_write_hint = bio_src->bi_write_hint; + bio->bi_iter.bi_sector = bio_src->bi_iter.bi_sector; + bio->bi_iter.bi_size = bio_src->bi_iter.bi_size; + + switch (bio_op(bio)) { + case REQ_OP_DISCARD: + case REQ_OP_SECURE_ERASE: + case REQ_OP_WRITE_ZEROES: + break; + case REQ_OP_WRITE_SAME: + bio->bi_io_vec[bio->bi_vcnt++] = bio_src->bi_io_vec[0]; + break; + default: + bio_for_each_segment(bv, bio_src, iter) + bio->bi_io_vec[bio->bi_vcnt++] = bv; + break; + } + + if (bio_integrity(bio_src)) { + int ret; + + ret = bio_integrity_clone(bio, bio_src, gfp_mask); + if (ret < 0) { + bio_put(bio); + return NULL; + } + } + + bio_clone_blkcg_association(bio, bio_src); + + return bio; +} + static void __blk_queue_bounce(struct request_queue *q, struct bio **bio_orig, mempool_t *pool) { @@ -222,7 +289,7 @@ static void __blk_queue_bounce(struct request_queue *q, struct bio **bio_orig, generic_make_request(*bio_orig); *bio_orig = bio; } - bio = bio_clone_bioset(*bio_orig, GFP_NOIO, passthrough ? NULL : + bio = bounce_clone_bio(*bio_orig, GFP_NOIO, passthrough ? NULL : &bounce_bio_set); bio_for_each_segment_all(to, bio, i) { diff --git a/block/bsg-lib.c b/block/bsg-lib.c index 9419def8c017..f3501cdaf1a6 100644 --- a/block/bsg-lib.c +++ b/block/bsg-lib.c @@ -48,9 +48,8 @@ static int bsg_transport_fill_hdr(struct request *rq, struct sg_io_v4 *hdr, job->request_len = hdr->request_len; job->request = memdup_user(uptr64(hdr->request), hdr->request_len); - if (IS_ERR(job->request)) - return PTR_ERR(job->request); - return 0; + + return PTR_ERR_OR_ZERO(job->request); } static int bsg_transport_complete_rq(struct request *rq, struct sg_io_v4 *hdr) diff --git a/block/bsg.c b/block/bsg.c index 3da540faf673..db588add6ba6 100644 --- a/block/bsg.c +++ b/block/bsg.c @@ -13,11 +13,9 @@ #include <linux/init.h> #include <linux/file.h> #include <linux/blkdev.h> -#include <linux/poll.h> #include <linux/cdev.h> #include <linux/jiffies.h> #include <linux/percpu.h> -#include <linux/uio.h> #include <linux/idr.h> #include <linux/bsg.h> #include <linux/slab.h> @@ -38,21 +36,10 @@ struct bsg_device { struct request_queue *queue; spinlock_t lock; - struct list_head busy_list; - struct list_head done_list; struct hlist_node dev_list; atomic_t ref_count; - int queued_cmds; - int done_cmds; - wait_queue_head_t wq_done; - wait_queue_head_t wq_free; char name[20]; int max_queue; - unsigned long flags; -}; - -enum { - BSG_F_BLOCK = 1, }; #define BSG_DEFAULT_CMDS 64 @@ -67,64 +54,6 @@ static struct hlist_head bsg_device_list[BSG_LIST_ARRAY_SIZE]; static struct class *bsg_class; static int bsg_major; -static struct kmem_cache *bsg_cmd_cachep; - -/* - * our internal command type - */ -struct bsg_command { - struct bsg_device *bd; - struct list_head list; - struct request *rq; - struct bio *bio; - struct bio *bidi_bio; - int err; - struct sg_io_v4 hdr; -}; - -static void bsg_free_command(struct bsg_command *bc) -{ - struct bsg_device *bd = bc->bd; - unsigned long flags; - - kmem_cache_free(bsg_cmd_cachep, bc); - - spin_lock_irqsave(&bd->lock, flags); - bd->queued_cmds--; - spin_unlock_irqrestore(&bd->lock, flags); - - wake_up(&bd->wq_free); -} - -static struct bsg_command *bsg_alloc_command(struct bsg_device *bd) -{ - struct bsg_command *bc = ERR_PTR(-EINVAL); - - spin_lock_irq(&bd->lock); - - if (bd->queued_cmds >= bd->max_queue) - goto out; - - bd->queued_cmds++; - spin_unlock_irq(&bd->lock); - - bc = kmem_cache_zalloc(bsg_cmd_cachep, GFP_KERNEL); - if (unlikely(!bc)) { - spin_lock_irq(&bd->lock); - bd->queued_cmds--; - bc = ERR_PTR(-ENOMEM); - goto out; - } - - bc->bd = bd; - INIT_LIST_HEAD(&bc->list); - bsg_dbg(bd, "returning free cmd %p\n", bc); - return bc; -out: - spin_unlock_irq(&bd->lock); - return bc; -} - static inline struct hlist_head *bsg_dev_idx_hash(int index) { return &bsg_device_list[index & (BSG_LIST_ARRAY_SIZE - 1)]; @@ -285,101 +214,6 @@ out: return ERR_PTR(ret); } -/* - * async completion call-back from the block layer, when scsi/ide/whatever - * calls end_that_request_last() on a request - */ -static void bsg_rq_end_io(struct request *rq, blk_status_t status) -{ - struct bsg_command *bc = rq->end_io_data; - struct bsg_device *bd = bc->bd; - unsigned long flags; - - bsg_dbg(bd, "finished rq %p bc %p, bio %p\n", - rq, bc, bc->bio); - - bc->hdr.duration = jiffies_to_msecs(jiffies - bc->hdr.duration); - - spin_lock_irqsave(&bd->lock, flags); - list_move_tail(&bc->list, &bd->done_list); - bd->done_cmds++; - spin_unlock_irqrestore(&bd->lock, flags); - - wake_up(&bd->wq_done); -} - -/* - * do final setup of a 'bc' and submit the matching 'rq' to the block - * layer for io - */ -static void bsg_add_command(struct bsg_device *bd, struct request_queue *q, - struct bsg_command *bc, struct request *rq) -{ - int at_head = (0 == (bc->hdr.flags & BSG_FLAG_Q_AT_TAIL)); - - /* - * add bc command to busy queue and submit rq for io - */ - bc->rq = rq; - bc->bio = rq->bio; - if (rq->next_rq) - bc->bidi_bio = rq->next_rq->bio; - bc->hdr.duration = jiffies; - spin_lock_irq(&bd->lock); - list_add_tail(&bc->list, &bd->busy_list); - spin_unlock_irq(&bd->lock); - - bsg_dbg(bd, "queueing rq %p, bc %p\n", rq, bc); - - rq->end_io_data = bc; - blk_execute_rq_nowait(q, NULL, rq, at_head, bsg_rq_end_io); -} - -static struct bsg_command *bsg_next_done_cmd(struct bsg_device *bd) -{ - struct bsg_command *bc = NULL; - - spin_lock_irq(&bd->lock); - if (bd->done_cmds) { - bc = list_first_entry(&bd->done_list, struct bsg_command, list); - list_del(&bc->list); - bd->done_cmds--; - } - spin_unlock_irq(&bd->lock); - - return bc; -} - -/* - * Get a finished command from the done list - */ -static struct bsg_command *bsg_get_done_cmd(struct bsg_device *bd) -{ - struct bsg_command *bc; - int ret; - - do { - bc = bsg_next_done_cmd(bd); - if (bc) - break; - - if (!test_bit(BSG_F_BLOCK, &bd->flags)) { - bc = ERR_PTR(-EAGAIN); - break; - } - - ret = wait_event_interruptible(bd->wq_done, bd->done_cmds); - if (ret) { - bc = ERR_PTR(-ERESTARTSYS); - break; - } - } while (1); - - bsg_dbg(bd, "returning done %p\n", bc); - - return bc; -} - static int blk_complete_sgv4_hdr_rq(struct request *rq, struct sg_io_v4 *hdr, struct bio *bio, struct bio *bidi_bio) { @@ -398,234 +232,6 @@ static int blk_complete_sgv4_hdr_rq(struct request *rq, struct sg_io_v4 *hdr, return ret; } -static bool bsg_complete(struct bsg_device *bd) -{ - bool ret = false; - bool spin; - - do { - spin_lock_irq(&bd->lock); - - BUG_ON(bd->done_cmds > bd->queued_cmds); - - /* - * All commands consumed. - */ - if (bd->done_cmds == bd->queued_cmds) - ret = true; - - spin = !test_bit(BSG_F_BLOCK, &bd->flags); - - spin_unlock_irq(&bd->lock); - } while (!ret && spin); - - return ret; -} - -static int bsg_complete_all_commands(struct bsg_device *bd) -{ - struct bsg_command *bc; - int ret, tret; - - bsg_dbg(bd, "entered\n"); - - /* - * wait for all commands to complete - */ - io_wait_event(bd->wq_done, bsg_complete(bd)); - - /* - * discard done commands - */ - ret = 0; - do { - spin_lock_irq(&bd->lock); - if (!bd->queued_cmds) { - spin_unlock_irq(&bd->lock); - break; - } - spin_unlock_irq(&bd->lock); - - bc = bsg_get_done_cmd(bd); - if (IS_ERR(bc)) - break; - - tret = blk_complete_sgv4_hdr_rq(bc->rq, &bc->hdr, bc->bio, - bc->bidi_bio); - if (!ret) - ret = tret; - - bsg_free_command(bc); - } while (1); - - return ret; -} - -static int -__bsg_read(char __user *buf, size_t count, struct bsg_device *bd, - const struct iovec *iov, ssize_t *bytes_read) -{ - struct bsg_command *bc; - int nr_commands, ret; - - if (count % sizeof(struct sg_io_v4)) - return -EINVAL; - - ret = 0; - nr_commands = count / sizeof(struct sg_io_v4); - while (nr_commands) { - bc = bsg_get_done_cmd(bd); - if (IS_ERR(bc)) { - ret = PTR_ERR(bc); - break; - } - - /* - * this is the only case where we need to copy data back - * after completing the request. so do that here, - * bsg_complete_work() cannot do that for us - */ - ret = blk_complete_sgv4_hdr_rq(bc->rq, &bc->hdr, bc->bio, - bc->bidi_bio); - - if (copy_to_user(buf, &bc->hdr, sizeof(bc->hdr))) - ret = -EFAULT; - - bsg_free_command(bc); - - if (ret) - break; - - buf += sizeof(struct sg_io_v4); - *bytes_read += sizeof(struct sg_io_v4); - nr_commands--; - } - - return ret; -} - -static inline void bsg_set_block(struct bsg_device *bd, struct file *file) -{ - if (file->f_flags & O_NONBLOCK) - clear_bit(BSG_F_BLOCK, &bd->flags); - else - set_bit(BSG_F_BLOCK, &bd->flags); -} - -/* - * Check if the error is a "real" error that we should return. - */ -static inline int err_block_err(int ret) -{ - if (ret && ret != -ENOSPC && ret != -ENODATA && ret != -EAGAIN) - return 1; - - return 0; -} - -static ssize_t -bsg_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) -{ - struct bsg_device *bd = file->private_data; - int ret; - ssize_t bytes_read; - - bsg_dbg(bd, "read %zd bytes\n", count); - - bsg_set_block(bd, file); - - bytes_read = 0; - ret = __bsg_read(buf, count, bd, NULL, &bytes_read); - *ppos = bytes_read; - - if (!bytes_read || err_block_err(ret)) - bytes_read = ret; - - return bytes_read; -} - -static int __bsg_write(struct bsg_device *bd, const char __user *buf, - size_t count, ssize_t *bytes_written, fmode_t mode) -{ - struct bsg_command *bc; - struct request *rq; - int ret, nr_commands; - - if (count % sizeof(struct sg_io_v4)) - return -EINVAL; - - nr_commands = count / sizeof(struct sg_io_v4); - rq = NULL; - bc = NULL; - ret = 0; - while (nr_commands) { - struct request_queue *q = bd->queue; - - bc = bsg_alloc_command(bd); - if (IS_ERR(bc)) { - ret = PTR_ERR(bc); - bc = NULL; - break; - } - - if (copy_from_user(&bc->hdr, buf, sizeof(bc->hdr))) { - ret = -EFAULT; - break; - } - - /* - * get a request, fill in the blanks, and add to request queue - */ - rq = bsg_map_hdr(bd->queue, &bc->hdr, mode); - if (IS_ERR(rq)) { - ret = PTR_ERR(rq); - rq = NULL; - break; - } - - bsg_add_command(bd, q, bc, rq); - bc = NULL; - rq = NULL; - nr_commands--; - buf += sizeof(struct sg_io_v4); - *bytes_written += sizeof(struct sg_io_v4); - } - - if (bc) - bsg_free_command(bc); - - return ret; -} - -static ssize_t -bsg_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos) -{ - struct bsg_device *bd = file->private_data; - ssize_t bytes_written; - int ret; - - bsg_dbg(bd, "write %zd bytes\n", count); - - if (unlikely(uaccess_kernel())) - return -EINVAL; - - bsg_set_block(bd, file); - - bytes_written = 0; - ret = __bsg_write(bd, buf, count, &bytes_written, file->f_mode); - - *ppos = bytes_written; - - /* - * return bytes written on non-fatal errors - */ - if (!bytes_written || err_block_err(ret)) - bytes_written = ret; - - bsg_dbg(bd, "returning %zd\n", bytes_written); - return bytes_written; -} - static struct bsg_device *bsg_alloc_device(void) { struct bsg_device *bd; @@ -635,29 +241,20 @@ static struct bsg_device *bsg_alloc_device(void) return NULL; spin_lock_init(&bd->lock); - bd->max_queue = BSG_DEFAULT_CMDS; - - INIT_LIST_HEAD(&bd->busy_list); - INIT_LIST_HEAD(&bd->done_list); INIT_HLIST_NODE(&bd->dev_list); - - init_waitqueue_head(&bd->wq_free); - init_waitqueue_head(&bd->wq_done); return bd; } static int bsg_put_device(struct bsg_device *bd) { - int ret = 0, do_free; struct request_queue *q = bd->queue; mutex_lock(&bsg_mutex); - do_free = atomic_dec_and_test(&bd->ref_count); - if (!do_free) { + if (!atomic_dec_and_test(&bd->ref_count)) { mutex_unlock(&bsg_mutex); - goto out; + return 0; } hlist_del(&bd->dev_list); @@ -668,20 +265,9 @@ static int bsg_put_device(struct bsg_device *bd) /* * close can always block */ - set_bit(BSG_F_BLOCK, &bd->flags); - - /* - * correct error detection baddies here again. it's the responsibility - * of the app to properly reap commands before close() if it wants - * fool-proof error detection - */ - ret = bsg_complete_all_commands(bd); - kfree(bd); -out: - if (do_free) - blk_put_queue(q); - return ret; + blk_put_queue(q); + return 0; } static struct bsg_device *bsg_add_device(struct inode *inode, @@ -704,8 +290,6 @@ static struct bsg_device *bsg_add_device(struct inode *inode, bd->queue = rq; - bsg_set_block(bd, file); - atomic_set(&bd->ref_count, 1); hlist_add_head(&bd->dev_list, bsg_dev_idx_hash(iminor(inode))); @@ -779,24 +363,6 @@ static int bsg_release(struct inode *inode, struct file *file) return bsg_put_device(bd); } -static __poll_t bsg_poll(struct file *file, poll_table *wait) -{ - struct bsg_device *bd = file->private_data; - __poll_t mask = 0; - - poll_wait(file, &bd->wq_done, wait); - poll_wait(file, &bd->wq_free, wait); - - spin_lock_irq(&bd->lock); - if (!list_empty(&bd->done_list)) - mask |= EPOLLIN | EPOLLRDNORM; - if (bd->queued_cmds < bd->max_queue) - mask |= EPOLLOUT; - spin_unlock_irq(&bd->lock); - - return mask; -} - static long bsg_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { struct bsg_device *bd = file->private_data; @@ -870,9 +436,6 @@ static long bsg_ioctl(struct file *file, unsigned int cmd, unsigned long arg) } static const struct file_operations bsg_fops = { - .read = bsg_read, - .write = bsg_write, - .poll = bsg_poll, .open = bsg_open, .release = bsg_release, .unlocked_ioctl = bsg_ioctl, @@ -977,21 +540,12 @@ static int __init bsg_init(void) int ret, i; dev_t devid; - bsg_cmd_cachep = kmem_cache_create("bsg_cmd", - sizeof(struct bsg_command), 0, 0, NULL); - if (!bsg_cmd_cachep) { - printk(KERN_ERR "bsg: failed creating slab cache\n"); - return -ENOMEM; - } - for (i = 0; i < BSG_LIST_ARRAY_SIZE; i++) INIT_HLIST_HEAD(&bsg_device_list[i]); bsg_class = class_create(THIS_MODULE, "bsg"); - if (IS_ERR(bsg_class)) { - ret = PTR_ERR(bsg_class); - goto destroy_kmemcache; - } + if (IS_ERR(bsg_class)) + return PTR_ERR(bsg_class); bsg_class->devnode = bsg_devnode; ret = alloc_chrdev_region(&devid, 0, BSG_MAX_DEVS, "bsg"); @@ -1012,8 +566,6 @@ unregister_chrdev: unregister_chrdev_region(MKDEV(bsg_major, 0), BSG_MAX_DEVS); destroy_bsg_class: class_destroy(bsg_class); -destroy_kmemcache: - kmem_cache_destroy(bsg_cmd_cachep); return ret; } diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 82b6c27b3245..2eb87444b157 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -3666,6 +3666,7 @@ static void cfq_init_prio_data(struct cfq_queue *cfqq, struct cfq_io_cq *cic) switch (ioprio_class) { default: printk(KERN_ERR "cfq: bad prio %x\n", ioprio_class); + /* fall through */ case IOPRIO_CLASS_NONE: /* * no prio set, inherit CPU scheduling settings @@ -4735,12 +4736,13 @@ USEC_SHOW_FUNCTION(cfq_target_latency_us_show, cfqd->cfq_target_latency); static ssize_t __FUNC(struct elevator_queue *e, const char *page, size_t count) \ { \ struct cfq_data *cfqd = e->elevator_data; \ - unsigned int __data; \ + unsigned int __data, __min = (MIN), __max = (MAX); \ + \ cfq_var_store(&__data, (page)); \ - if (__data < (MIN)) \ - __data = (MIN); \ - else if (__data > (MAX)) \ - __data = (MAX); \ + if (__data < __min) \ + __data = __min; \ + else if (__data > __max) \ + __data = __max; \ if (__CONV) \ *(__PTR) = (u64)__data * NSEC_PER_MSEC; \ else \ @@ -4769,12 +4771,13 @@ STORE_FUNCTION(cfq_target_latency_store, &cfqd->cfq_target_latency, 1, UINT_MAX, static ssize_t __FUNC(struct elevator_queue *e, const char *page, size_t count) \ { \ struct cfq_data *cfqd = e->elevator_data; \ - unsigned int __data; \ + unsigned int __data, __min = (MIN), __max = (MAX); \ + \ cfq_var_store(&__data, (page)); \ - if (__data < (MIN)) \ - __data = (MIN); \ - else if (__data > (MAX)) \ - __data = (MAX); \ + if (__data < __min) \ + __data = __min; \ + else if (__data > __max) \ + __data = __max; \ *(__PTR) = (u64)__data * NSEC_PER_USEC; \ return count; \ } diff --git a/block/genhd.c b/block/genhd.c index f1543a45e73b..8cc719a37b32 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -1333,21 +1333,28 @@ static int diskstats_show(struct seq_file *seqf, void *v) part_round_stats(gp->queue, cpu, hd); part_stat_unlock(); part_in_flight(gp->queue, hd, inflight); - seq_printf(seqf, "%4d %7d %s %lu %lu %lu " - "%u %lu %lu %lu %u %u %u %u\n", + seq_printf(seqf, "%4d %7d %s " + "%lu %lu %lu %u " + "%lu %lu %lu %u " + "%u %u %u " + "%lu %lu %lu %u\n", MAJOR(part_devt(hd)), MINOR(part_devt(hd)), disk_name(gp, hd->partno, buf), - part_stat_read(hd, ios[READ]), - part_stat_read(hd, merges[READ]), - part_stat_read(hd, sectors[READ]), - jiffies_to_msecs(part_stat_read(hd, ticks[READ])), - part_stat_read(hd, ios[WRITE]), - part_stat_read(hd, merges[WRITE]), - part_stat_read(hd, sectors[WRITE]), - jiffies_to_msecs(part_stat_read(hd, ticks[WRITE])), + part_stat_read(hd, ios[STAT_READ]), + part_stat_read(hd, merges[STAT_READ]), + part_stat_read(hd, sectors[STAT_READ]), + jiffies_to_msecs(part_stat_read(hd, ticks[STAT_READ])), + part_stat_read(hd, ios[STAT_WRITE]), + part_stat_read(hd, merges[STAT_WRITE]), + part_stat_read(hd, sectors[STAT_WRITE]), + jiffies_to_msecs(part_stat_read(hd, ticks[STAT_WRITE])), inflight[0], jiffies_to_msecs(part_stat_read(hd, io_ticks)), - jiffies_to_msecs(part_stat_read(hd, time_in_queue)) + jiffies_to_msecs(part_stat_read(hd, time_in_queue)), + part_stat_read(hd, ios[STAT_DISCARD]), + part_stat_read(hd, merges[STAT_DISCARD]), + part_stat_read(hd, sectors[STAT_DISCARD]), + jiffies_to_msecs(part_stat_read(hd, ticks[STAT_DISCARD])) ); } disk_part_iter_exit(&piter); diff --git a/block/partition-generic.c b/block/partition-generic.c index 3dcfd4ec0e11..5a8975a1201c 100644 --- a/block/partition-generic.c +++ b/block/partition-generic.c @@ -130,19 +130,24 @@ ssize_t part_stat_show(struct device *dev, return sprintf(buf, "%8lu %8lu %8llu %8u " "%8lu %8lu %8llu %8u " - "%8u %8u %8u" + "%8u %8u %8u " + "%8lu %8lu %8llu %8u" "\n", - part_stat_read(p, ios[READ]), - part_stat_read(p, merges[READ]), - (unsigned long long)part_stat_read(p, sectors[READ]), - jiffies_to_msecs(part_stat_read(p, ticks[READ])), - part_stat_read(p, ios[WRITE]), - part_stat_read(p, merges[WRITE]), - (unsigned long long)part_stat_read(p, sectors[WRITE]), - jiffies_to_msecs(part_stat_read(p, ticks[WRITE])), + part_stat_read(p, ios[STAT_READ]), + part_stat_read(p, merges[STAT_READ]), + (unsigned long long)part_stat_read(p, sectors[STAT_READ]), + jiffies_to_msecs(part_stat_read(p, ticks[STAT_READ])), + part_stat_read(p, ios[STAT_WRITE]), + part_stat_read(p, merges[STAT_WRITE]), + (unsigned long long)part_stat_read(p, sectors[STAT_WRITE]), + jiffies_to_msecs(part_stat_read(p, ticks[STAT_WRITE])), inflight[0], jiffies_to_msecs(part_stat_read(p, io_ticks)), - jiffies_to_msecs(part_stat_read(p, time_in_queue))); + jiffies_to_msecs(part_stat_read(p, time_in_queue)), + part_stat_read(p, ios[STAT_DISCARD]), + part_stat_read(p, merges[STAT_DISCARD]), + (unsigned long long)part_stat_read(p, sectors[STAT_DISCARD]), + jiffies_to_msecs(part_stat_read(p, ticks[STAT_DISCARD]))); } ssize_t part_inflight_show(struct device *dev, struct device_attribute *attr, diff --git a/block/partitions/aix.c b/block/partitions/aix.c index 007f95eea0e1..903f3ed175d0 100644 --- a/block/partitions/aix.c +++ b/block/partitions/aix.c @@ -178,7 +178,7 @@ int aix_partition(struct parsed_partitions *state) u32 vgda_sector = 0; u32 vgda_len = 0; int numlvs = 0; - struct pvd *pvd; + struct pvd *pvd = NULL; struct lv_info { unsigned short pps_per_lv; unsigned short pps_found; @@ -232,10 +232,11 @@ int aix_partition(struct parsed_partitions *state) if (lvip[i].pps_per_lv) foundlvs += 1; } + /* pvd loops depend on n[].name and lvip[].pps_per_lv */ + pvd = alloc_pvd(state, vgda_sector + 17); } put_dev_sector(sect); } - pvd = alloc_pvd(state, vgda_sector + 17); if (pvd) { int numpps = be16_to_cpu(pvd->pp_count); int psn_part1 = be32_to_cpu(pvd->psn_part1); @@ -282,10 +283,14 @@ int aix_partition(struct parsed_partitions *state) next_lp_ix += 1; } for (i = 0; i < state->limit; i += 1) - if (lvip[i].pps_found && !lvip[i].lv_is_contiguous) + if (lvip[i].pps_found && !lvip[i].lv_is_contiguous) { + char tmp[sizeof(n[i].name) + 1]; // null char + + snprintf(tmp, sizeof(tmp), "%s", n[i].name); pr_warn("partition %s (%u pp's found) is " "not contiguous\n", - n[i].name, lvip[i].pps_found); + tmp, lvip[i].pps_found); + } kfree(pvd); } kfree(n); diff --git a/block/partitions/ldm.c b/block/partitions/ldm.c index 0417937dfe99..16766f267559 100644 --- a/block/partitions/ldm.c +++ b/block/partitions/ldm.c @@ -830,7 +830,6 @@ static bool ldm_parse_dgr4 (const u8 *buffer, int buflen, struct vblk *vb) { char buf[64]; int r_objid, r_name, r_id1, r_id2, len; - struct vblk_dgrp *dgrp; BUG_ON (!buffer || !vb); @@ -853,8 +852,6 @@ static bool ldm_parse_dgr4 (const u8 *buffer, int buflen, struct vblk *vb) if (len != get_unaligned_be32(buffer + 0x14)) return false; - dgrp = &vb->vblk.dgrp; - ldm_get_vstr (buffer + 0x18 + r_objid, buf, sizeof (buf)); return true; } diff --git a/block/t10-pi.c b/block/t10-pi.c index a98db384048f..62aed77d0bb9 100644 --- a/block/t10-pi.c +++ b/block/t10-pi.c @@ -184,3 +184,113 @@ const struct blk_integrity_profile t10_pi_type3_ip = { .verify_fn = t10_pi_type3_verify_ip, }; EXPORT_SYMBOL(t10_pi_type3_ip); + +/** + * t10_pi_prepare - prepare PI prior submitting request to device + * @rq: request with PI that should be prepared + * @protection_type: PI type (Type 1/Type 2/Type 3) + * + * For Type 1/Type 2, the virtual start sector is the one that was + * originally submitted by the block layer for the ref_tag usage. Due to + * partitioning, MD/DM cloning, etc. the actual physical start sector is + * likely to be different. Remap protection information to match the + * physical LBA. + * + * Type 3 does not have a reference tag so no remapping is required. + */ +void t10_pi_prepare(struct request *rq, u8 protection_type) +{ + const int tuple_sz = rq->q->integrity.tuple_size; + u32 ref_tag = t10_pi_ref_tag(rq); + struct bio *bio; + + if (protection_type == T10_PI_TYPE3_PROTECTION) + return; + + __rq_for_each_bio(bio, rq) { + struct bio_integrity_payload *bip = bio_integrity(bio); + u32 virt = bip_get_seed(bip) & 0xffffffff; + struct bio_vec iv; + struct bvec_iter iter; + + /* Already remapped? */ + if (bip->bip_flags & BIP_MAPPED_INTEGRITY) + break; + + bip_for_each_vec(iv, bip, iter) { + void *p, *pmap; + unsigned int j; + + pmap = kmap_atomic(iv.bv_page); + p = pmap + iv.bv_offset; + for (j = 0; j < iv.bv_len; j += tuple_sz) { + struct t10_pi_tuple *pi = p; + + if (be32_to_cpu(pi->ref_tag) == virt) + pi->ref_tag = cpu_to_be32(ref_tag); + virt++; + ref_tag++; + p += tuple_sz; + } + + kunmap_atomic(pmap); + } + + bip->bip_flags |= BIP_MAPPED_INTEGRITY; + } +} +EXPORT_SYMBOL(t10_pi_prepare); + +/** + * t10_pi_complete - prepare PI prior returning request to the block layer + * @rq: request with PI that should be prepared + * @protection_type: PI type (Type 1/Type 2/Type 3) + * @intervals: total elements to prepare + * + * For Type 1/Type 2, the virtual start sector is the one that was + * originally submitted by the block layer for the ref_tag usage. Due to + * partitioning, MD/DM cloning, etc. the actual physical start sector is + * likely to be different. Since the physical start sector was submitted + * to the device, we should remap it back to virtual values expected by the + * block layer. + * + * Type 3 does not have a reference tag so no remapping is required. + */ +void t10_pi_complete(struct request *rq, u8 protection_type, + unsigned int intervals) +{ + const int tuple_sz = rq->q->integrity.tuple_size; + u32 ref_tag = t10_pi_ref_tag(rq); + struct bio *bio; + + if (protection_type == T10_PI_TYPE3_PROTECTION) + return; + + __rq_for_each_bio(bio, rq) { + struct bio_integrity_payload *bip = bio_integrity(bio); + u32 virt = bip_get_seed(bip) & 0xffffffff; + struct bio_vec iv; + struct bvec_iter iter; + + bip_for_each_vec(iv, bip, iter) { + void *p, *pmap; + unsigned int j; + + pmap = kmap_atomic(iv.bv_page); + p = pmap + iv.bv_offset; + for (j = 0; j < iv.bv_len && intervals; j += tuple_sz) { + struct t10_pi_tuple *pi = p; + + if (be32_to_cpu(pi->ref_tag) == ref_tag) + pi->ref_tag = cpu_to_be32(virt); + virt++; + ref_tag++; + intervals--; + p += tuple_sz; + } + + kunmap_atomic(pmap); + } + } +} +EXPORT_SYMBOL(t10_pi_complete); diff --git a/drivers/Makefile b/drivers/Makefile index 24cd47014657..a6abd7a856c6 100644 --- a/drivers/Makefile +++ b/drivers/Makefile @@ -76,7 +76,7 @@ obj-$(CONFIG_DMA_SHARED_BUFFER) += dma-buf/ obj-$(CONFIG_NUBUS) += nubus/ obj-y += macintosh/ obj-$(CONFIG_IDE) += ide/ -obj-$(CONFIG_SCSI) += scsi/ +obj-y += scsi/ obj-y += nvme/ obj-$(CONFIG_ATA) += ata/ obj-$(CONFIG_TARGET_CORE) += target/ diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c index aad1b01447de..8e270962b2f3 100644 --- a/drivers/ata/libata-scsi.c +++ b/drivers/ata/libata-scsi.c @@ -597,8 +597,9 @@ static int ata_get_identity(struct ata_port *ap, struct scsi_device *sdev, int ata_cmd_ioctl(struct scsi_device *scsidev, void __user *arg) { int rc = 0; + u8 sensebuf[SCSI_SENSE_BUFFERSIZE]; u8 scsi_cmd[MAX_COMMAND_SIZE]; - u8 args[4], *argbuf = NULL, *sensebuf = NULL; + u8 args[4], *argbuf = NULL; int argsize = 0; enum dma_data_direction data_dir; struct scsi_sense_hdr sshdr; @@ -610,10 +611,7 @@ int ata_cmd_ioctl(struct scsi_device *scsidev, void __user *arg) if (copy_from_user(args, arg, sizeof(args))) return -EFAULT; - sensebuf = kzalloc(SCSI_SENSE_BUFFERSIZE, GFP_NOIO); - if (!sensebuf) - return -ENOMEM; - + memset(sensebuf, 0, sizeof(sensebuf)); memset(scsi_cmd, 0, sizeof(scsi_cmd)); if (args[3]) { @@ -685,7 +683,6 @@ int ata_cmd_ioctl(struct scsi_device *scsidev, void __user *arg) && copy_to_user(arg + sizeof(args), argbuf, argsize)) rc = -EFAULT; error: - kfree(sensebuf); kfree(argbuf); return rc; } @@ -704,8 +701,9 @@ error: int ata_task_ioctl(struct scsi_device *scsidev, void __user *arg) { int rc = 0; + u8 sensebuf[SCSI_SENSE_BUFFERSIZE]; u8 scsi_cmd[MAX_COMMAND_SIZE]; - u8 args[7], *sensebuf = NULL; + u8 args[7]; struct scsi_sense_hdr sshdr; int cmd_result; @@ -715,10 +713,7 @@ int ata_task_ioctl(struct scsi_device *scsidev, void __user *arg) if (copy_from_user(args, arg, sizeof(args))) return -EFAULT; - sensebuf = kzalloc(SCSI_SENSE_BUFFERSIZE, GFP_NOIO); - if (!sensebuf) - return -ENOMEM; - + memset(sensebuf, 0, sizeof(sensebuf)); memset(scsi_cmd, 0, sizeof(scsi_cmd)); scsi_cmd[0] = ATA_16; scsi_cmd[1] = (3 << 1); /* Non-data */ @@ -769,7 +764,6 @@ int ata_task_ioctl(struct scsi_device *scsidev, void __user *arg) } error: - kfree(sensebuf); return rc; } diff --git a/drivers/block/DAC960.c b/drivers/block/DAC960.c index f6518067aa7d..f99e5c883368 100644 --- a/drivers/block/DAC960.c +++ b/drivers/block/DAC960.c @@ -21,6 +21,7 @@ #define DAC960_DriverDate "21 Aug 2007" +#include <linux/compiler.h> #include <linux/module.h> #include <linux/types.h> #include <linux/miscdevice.h> @@ -6426,7 +6427,7 @@ static bool DAC960_V2_ExecuteUserCommand(DAC960_Controller_T *Controller, return true; } -static int dac960_proc_show(struct seq_file *m, void *v) +static int __maybe_unused dac960_proc_show(struct seq_file *m, void *v) { unsigned char *StatusMessage = "OK\n"; int ControllerNumber; @@ -6446,14 +6447,16 @@ static int dac960_proc_show(struct seq_file *m, void *v) return 0; } -static int dac960_initial_status_proc_show(struct seq_file *m, void *v) +static int __maybe_unused dac960_initial_status_proc_show(struct seq_file *m, + void *v) { DAC960_Controller_T *Controller = (DAC960_Controller_T *)m->private; seq_printf(m, "%.*s", Controller->InitialStatusLength, Controller->CombinedStatusBuffer); return 0; } -static int dac960_current_status_proc_show(struct seq_file *m, void *v) +static int __maybe_unused dac960_current_status_proc_show(struct seq_file *m, + void *v) { DAC960_Controller_T *Controller = (DAC960_Controller_T *) m->private; unsigned char *StatusMessage = diff --git a/drivers/block/Kconfig b/drivers/block/Kconfig index ad9b687a236a..d4913516823f 100644 --- a/drivers/block/Kconfig +++ b/drivers/block/Kconfig @@ -74,12 +74,12 @@ config AMIGA_Z2RAM config CDROM tristate + select BLK_SCSI_REQUEST config GDROM tristate "SEGA Dreamcast GD-ROM drive" depends on SH_DREAMCAST select CDROM - select BLK_SCSI_REQUEST # only for the generic cdrom code help A standard SEGA Dreamcast comes with a modified CD ROM drive called a "GD-ROM" by SEGA to signify it is capable of reading special disks diff --git a/drivers/block/Makefile b/drivers/block/Makefile index dc061158b403..8566b188368b 100644 --- a/drivers/block/Makefile +++ b/drivers/block/Makefile @@ -36,8 +36,11 @@ obj-$(CONFIG_BLK_DEV_RBD) += rbd.o obj-$(CONFIG_BLK_DEV_PCIESSD_MTIP32XX) += mtip32xx/ obj-$(CONFIG_BLK_DEV_RSXX) += rsxx/ -obj-$(CONFIG_BLK_DEV_NULL_BLK) += null_blk.o obj-$(CONFIG_ZRAM) += zram/ +obj-$(CONFIG_BLK_DEV_NULL_BLK) += null_blk.o +null_blk-objs := null_blk_main.o +null_blk-$(CONFIG_BLK_DEV_ZONED) += null_blk_zoned.o + skd-y := skd_main.o swim_mod-y := swim.o swim_asm.o diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c index 096882e54095..136dc507d020 100644 --- a/drivers/block/aoe/aoecmd.c +++ b/drivers/block/aoe/aoecmd.c @@ -1137,6 +1137,7 @@ noskb: if (buf) break; } bvcpy(skb, f->buf->bio, f->iter, n); + /* fall through */ case ATA_CMD_PIO_WRITE: case ATA_CMD_PIO_WRITE_EXT: spin_lock_irq(&d->lock); diff --git a/drivers/block/aoe/aoedev.c b/drivers/block/aoe/aoedev.c index 697f735b07a4..41060e9cedf2 100644 --- a/drivers/block/aoe/aoedev.c +++ b/drivers/block/aoe/aoedev.c @@ -284,8 +284,8 @@ freedev(struct aoedev *d) e = t + d->ntargets; for (; t < e && *t; t++) freetgt(d, *t); - if (d->bufpool) - mempool_destroy(d->bufpool); + + mempool_destroy(d->bufpool); skbpoolfree(d); minor_free(d->sysminor); diff --git a/drivers/block/brd.c b/drivers/block/brd.c index bb976598ee43..df8103dd40ac 100644 --- a/drivers/block/brd.c +++ b/drivers/block/brd.c @@ -254,20 +254,20 @@ static void copy_from_brd(void *dst, struct brd_device *brd, * Process a single bvec of a bio. */ static int brd_do_bvec(struct brd_device *brd, struct page *page, - unsigned int len, unsigned int off, bool is_write, + unsigned int len, unsigned int off, unsigned int op, sector_t sector) { void *mem; int err = 0; - if (is_write) { + if (op_is_write(op)) { err = copy_to_brd_setup(brd, sector, len); if (err) goto out; } mem = kmap_atomic(page); - if (!is_write) { + if (!op_is_write(op)) { copy_from_brd(mem + off, brd, sector, len); flush_dcache_page(page); } else { @@ -296,7 +296,7 @@ static blk_qc_t brd_make_request(struct request_queue *q, struct bio *bio) int err; err = brd_do_bvec(brd, bvec.bv_page, len, bvec.bv_offset, - op_is_write(bio_op(bio)), sector); + bio_op(bio), sector); if (err) goto io_error; sector += len >> SECTOR_SHIFT; @@ -310,15 +310,15 @@ io_error: } static int brd_rw_page(struct block_device *bdev, sector_t sector, - struct page *page, bool is_write) + struct page *page, unsigned int op) { struct brd_device *brd = bdev->bd_disk->private_data; int err; if (PageTransHuge(page)) return -ENOTSUPP; - err = brd_do_bvec(brd, page, PAGE_SIZE, 0, is_write, sector); - page_endio(page, is_write, err); + err = brd_do_bvec(brd, page, PAGE_SIZE, 0, op, sector); + page_endio(page, op_is_write(op), err); return err; } diff --git a/drivers/block/drbd/drbd_int.h b/drivers/block/drbd/drbd_int.h index bc4ed2ed40a2..e35a234b0a8f 100644 --- a/drivers/block/drbd/drbd_int.h +++ b/drivers/block/drbd/drbd_int.h @@ -55,12 +55,10 @@ # define __protected_by(x) __attribute__((require_context(x,1,999,"rdwr"))) # define __protected_read_by(x) __attribute__((require_context(x,1,999,"read"))) # define __protected_write_by(x) __attribute__((require_context(x,1,999,"write"))) -# define __must_hold(x) __attribute__((context(x,1,1), require_context(x,1,999,"call"))) #else # define __protected_by(x) # define __protected_read_by(x) # define __protected_write_by(x) -# define __must_hold(x) #endif /* shared module parameters, defined in drbd_main.c */ diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c index a80809bd3057..ef8212a4b73e 100644 --- a/drivers/block/drbd/drbd_main.c +++ b/drivers/block/drbd/drbd_main.c @@ -2103,14 +2103,10 @@ static void drbd_destroy_mempools(void) mempool_exit(&drbd_md_io_page_pool); mempool_exit(&drbd_ee_mempool); mempool_exit(&drbd_request_mempool); - if (drbd_ee_cache) - kmem_cache_destroy(drbd_ee_cache); - if (drbd_request_cache) - kmem_cache_destroy(drbd_request_cache); - if (drbd_bm_ext_cache) - kmem_cache_destroy(drbd_bm_ext_cache); - if (drbd_al_ext_cache) - kmem_cache_destroy(drbd_al_ext_cache); + kmem_cache_destroy(drbd_ee_cache); + kmem_cache_destroy(drbd_request_cache); + kmem_cache_destroy(drbd_bm_ext_cache); + kmem_cache_destroy(drbd_al_ext_cache); drbd_ee_cache = NULL; drbd_request_cache = NULL; diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c index be9450f5ad1c..75f6b47169e6 100644 --- a/drivers/block/drbd/drbd_receiver.c +++ b/drivers/block/drbd/drbd_receiver.c @@ -2674,8 +2674,7 @@ bool drbd_rs_c_min_rate_throttle(struct drbd_device *device) if (c_min_rate == 0) return false; - curr_events = (int)part_stat_read(&disk->part0, sectors[0]) + - (int)part_stat_read(&disk->part0, sectors[1]) - + curr_events = (int)part_stat_read_accum(&disk->part0, sectors) - atomic_read(&device->rs_sect_ev); if (atomic_read(&device->ap_actlog_cnt) @@ -2790,6 +2789,7 @@ static int receive_DataRequest(struct drbd_connection *connection, struct packet then we would do something smarter here than reading the block... */ peer_req->flags |= EE_RS_THIN_REQ; + /* fall through */ case P_RS_DATA_REQUEST: peer_req->w.cb = w_e_end_rsdata_req; fault_type = DRBD_FAULT_RS_RD; @@ -2968,6 +2968,7 @@ static int drbd_asb_recover_0p(struct drbd_peer_device *peer_device) __must_hold /* Else fall through to one of the other strategies... */ drbd_warn(device, "Discard younger/older primary did not find a decision\n" "Using discard-least-changes instead\n"); + /* fall through */ case ASB_DISCARD_ZERO_CHG: if (ch_peer == 0 && ch_self == 0) { rv = test_bit(RESOLVE_CONFLICTS, &peer_device->connection->flags) @@ -2979,6 +2980,7 @@ static int drbd_asb_recover_0p(struct drbd_peer_device *peer_device) __must_hold } if (after_sb_0p == ASB_DISCARD_ZERO_CHG) break; + /* else: fall through */ case ASB_DISCARD_LEAST_CHG: if (ch_self < ch_peer) rv = -1; diff --git a/drivers/block/drbd/drbd_req.c b/drivers/block/drbd/drbd_req.c index d146fedc38bb..19cac36e9737 100644 --- a/drivers/block/drbd/drbd_req.c +++ b/drivers/block/drbd/drbd_req.c @@ -38,7 +38,7 @@ static void _drbd_start_io_acct(struct drbd_device *device, struct drbd_request { struct request_queue *q = device->rq_queue; - generic_start_io_acct(q, bio_data_dir(req->master_bio), + generic_start_io_acct(q, bio_op(req->master_bio), req->i.size >> 9, &device->vdisk->part0); } @@ -47,7 +47,7 @@ static void _drbd_end_io_acct(struct drbd_device *device, struct drbd_request *r { struct request_queue *q = device->rq_queue; - generic_end_io_acct(q, bio_data_dir(req->master_bio), + generic_end_io_acct(q, bio_op(req->master_bio), &device->vdisk->part0, req->start_jif); } diff --git a/drivers/block/drbd/drbd_worker.c b/drivers/block/drbd/drbd_worker.c index 5e793dd7adfb..b8f77e83d456 100644 --- a/drivers/block/drbd/drbd_worker.c +++ b/drivers/block/drbd/drbd_worker.c @@ -1690,9 +1690,7 @@ void drbd_rs_controller_reset(struct drbd_device *device) atomic_set(&device->rs_sect_in, 0); atomic_set(&device->rs_sect_ev, 0); device->rs_in_flight = 0; - device->rs_last_events = - (int)part_stat_read(&disk->part0, sectors[0]) + - (int)part_stat_read(&disk->part0, sectors[1]); + device->rs_last_events = (int)part_stat_read_accum(&disk->part0, sectors); /* Updating the RCU protected object in place is necessary since this function gets called from atomic context. diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c index 8871b5044d9e..48f622728ce6 100644 --- a/drivers/block/floppy.c +++ b/drivers/block/floppy.c @@ -1461,7 +1461,6 @@ static void setup_rw_floppy(void) int i; int r; int flags; - int dflags; unsigned long ready_date; void (*function)(void); @@ -1485,8 +1484,6 @@ static void setup_rw_floppy(void) if (fd_wait_for_completion(ready_date, function)) return; } - dflags = DRS->flags; - if ((flags & FD_RAW_READ) || (flags & FD_RAW_WRITE)) setup_DMA(); diff --git a/drivers/block/loop.c b/drivers/block/loop.c index 4cb1d1be3cfb..ea9debf59b22 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -690,7 +690,6 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev, unsigned int arg) { struct file *file, *old_file; - struct inode *inode; int error; error = -ENXIO; @@ -711,7 +710,6 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev, if (error) goto out_putf; - inode = file->f_mapping->host; old_file = lo->lo_backing_file; error = -EINVAL; @@ -1611,6 +1609,7 @@ static int lo_compat_ioctl(struct block_device *bdev, fmode_t mode, case LOOP_GET_STATUS64: case LOOP_SET_STATUS64: arg = (unsigned long) compat_ptr(arg); + /* fall through */ case LOOP_SET_FD: case LOOP_CHANGE_FD: case LOOP_SET_BLOCK_SIZE: diff --git a/drivers/block/mtip32xx/mtip32xx.c b/drivers/block/mtip32xx/mtip32xx.c index c73626decb46..db253cd5b32a 100644 --- a/drivers/block/mtip32xx/mtip32xx.c +++ b/drivers/block/mtip32xx/mtip32xx.c @@ -2575,8 +2575,7 @@ static int mtip_hw_debugfs_init(struct driver_data *dd) static void mtip_hw_debugfs_exit(struct driver_data *dd) { - if (dd->dfs_node) - debugfs_remove_recursive(dd->dfs_node); + debugfs_remove_recursive(dd->dfs_node); } /* diff --git a/drivers/block/null_blk.h b/drivers/block/null_blk.h new file mode 100644 index 000000000000..d81781f22dba --- /dev/null +++ b/drivers/block/null_blk.h @@ -0,0 +1,108 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __BLK_NULL_BLK_H +#define __BLK_NULL_BLK_H + +#include <linux/blkdev.h> +#include <linux/slab.h> +#include <linux/blk-mq.h> +#include <linux/hrtimer.h> +#include <linux/configfs.h> +#include <linux/badblocks.h> +#include <linux/fault-inject.h> + +struct nullb_cmd { + struct list_head list; + struct llist_node ll_list; + struct __call_single_data csd; + struct request *rq; + struct bio *bio; + unsigned int tag; + blk_status_t error; + struct nullb_queue *nq; + struct hrtimer timer; +}; + +struct nullb_queue { + unsigned long *tag_map; + wait_queue_head_t wait; + unsigned int queue_depth; + struct nullb_device *dev; + unsigned int requeue_selection; + + struct nullb_cmd *cmds; +}; + +struct nullb_device { + struct nullb *nullb; + struct config_item item; + struct radix_tree_root data; /* data stored in the disk */ + struct radix_tree_root cache; /* disk cache data */ + unsigned long flags; /* device flags */ + unsigned int curr_cache; + struct badblocks badblocks; + + unsigned int nr_zones; + struct blk_zone *zones; + sector_t zone_size_sects; + + unsigned long size; /* device size in MB */ + unsigned long completion_nsec; /* time in ns to complete a request */ + unsigned long cache_size; /* disk cache size in MB */ + unsigned long zone_size; /* zone size in MB if device is zoned */ + unsigned int submit_queues; /* number of submission queues */ + unsigned int home_node; /* home node for the device */ + unsigned int queue_mode; /* block interface */ + unsigned int blocksize; /* block size */ + unsigned int irqmode; /* IRQ completion handler */ + unsigned int hw_queue_depth; /* queue depth */ + unsigned int index; /* index of the disk, only valid with a disk */ + unsigned int mbps; /* Bandwidth throttle cap (in MB/s) */ + bool blocking; /* blocking blk-mq device */ + bool use_per_node_hctx; /* use per-node allocation for hardware context */ + bool power; /* power on/off the device */ + bool memory_backed; /* if data is stored in memory */ + bool discard; /* if support discard */ + bool zoned; /* if device is zoned */ +}; + +struct nullb { + struct nullb_device *dev; + struct list_head list; + unsigned int index; + struct request_queue *q; + struct gendisk *disk; + struct blk_mq_tag_set *tag_set; + struct blk_mq_tag_set __tag_set; + unsigned int queue_depth; + atomic_long_t cur_bytes; + struct hrtimer bw_timer; + unsigned long cache_flush_pos; + spinlock_t lock; + + struct nullb_queue *queues; + unsigned int nr_queues; + char disk_name[DISK_NAME_LEN]; +}; + +#ifdef CONFIG_BLK_DEV_ZONED +int null_zone_init(struct nullb_device *dev); +void null_zone_exit(struct nullb_device *dev); +blk_status_t null_zone_report(struct nullb *nullb, + struct nullb_cmd *cmd); +void null_zone_write(struct nullb_cmd *cmd); +void null_zone_reset(struct nullb_cmd *cmd); +#else +static inline int null_zone_init(struct nullb_device *dev) +{ + return -EINVAL; +} +static inline void null_zone_exit(struct nullb_device *dev) {} +static inline blk_status_t null_zone_report(struct nullb *nullb, + struct nullb_cmd *cmd) +{ + return BLK_STS_NOTSUPP; +} +static inline void null_zone_write(struct nullb_cmd *cmd) {} +static inline void null_zone_reset(struct nullb_cmd *cmd) {} +#endif /* CONFIG_BLK_DEV_ZONED */ +#endif /* __NULL_BLK_H */ diff --git a/drivers/block/null_blk.c b/drivers/block/null_blk_main.c index 042c778e5a4e..6127e3ff7b4b 100644 --- a/drivers/block/null_blk.c +++ b/drivers/block/null_blk_main.c @@ -7,14 +7,8 @@ #include <linux/moduleparam.h> #include <linux/sched.h> #include <linux/fs.h> -#include <linux/blkdev.h> #include <linux/init.h> -#include <linux/slab.h> -#include <linux/blk-mq.h> -#include <linux/hrtimer.h> -#include <linux/configfs.h> -#include <linux/badblocks.h> -#include <linux/fault-inject.h> +#include "null_blk.h" #define PAGE_SECTORS_SHIFT (PAGE_SHIFT - SECTOR_SHIFT) #define PAGE_SECTORS (1 << PAGE_SECTORS_SHIFT) @@ -35,28 +29,6 @@ static inline u64 mb_per_tick(int mbps) return (1 << 20) / TICKS_PER_SEC * ((u64) mbps); } -struct nullb_cmd { - struct list_head list; - struct llist_node ll_list; - struct __call_single_data csd; - struct request *rq; - struct bio *bio; - unsigned int tag; - blk_status_t error; - struct nullb_queue *nq; - struct hrtimer timer; -}; - -struct nullb_queue { - unsigned long *tag_map; - wait_queue_head_t wait; - unsigned int queue_depth; - struct nullb_device *dev; - unsigned int requeue_selection; - - struct nullb_cmd *cmds; -}; - /* * Status flags for nullb_device. * @@ -92,52 +64,6 @@ struct nullb_page { #define NULLB_PAGE_LOCK (MAP_SZ - 1) #define NULLB_PAGE_FREE (MAP_SZ - 2) -struct nullb_device { - struct nullb *nullb; - struct config_item item; - struct radix_tree_root data; /* data stored in the disk */ - struct radix_tree_root cache; /* disk cache data */ - unsigned long flags; /* device flags */ - unsigned int curr_cache; - struct badblocks badblocks; - - unsigned long size; /* device size in MB */ - unsigned long completion_nsec; /* time in ns to complete a request */ - unsigned long cache_size; /* disk cache size in MB */ - unsigned int submit_queues; /* number of submission queues */ - unsigned int home_node; /* home node for the device */ - unsigned int queue_mode; /* block interface */ - unsigned int blocksize; /* block size */ - unsigned int irqmode; /* IRQ completion handler */ - unsigned int hw_queue_depth; /* queue depth */ - unsigned int index; /* index of the disk, only valid with a disk */ - unsigned int mbps; /* Bandwidth throttle cap (in MB/s) */ - bool blocking; /* blocking blk-mq device */ - bool use_per_node_hctx; /* use per-node allocation for hardware context */ - bool power; /* power on/off the device */ - bool memory_backed; /* if data is stored in memory */ - bool discard; /* if support discard */ -}; - -struct nullb { - struct nullb_device *dev; - struct list_head list; - unsigned int index; - struct request_queue *q; - struct gendisk *disk; - struct blk_mq_tag_set *tag_set; - struct blk_mq_tag_set __tag_set; - unsigned int queue_depth; - atomic_long_t cur_bytes; - struct hrtimer bw_timer; - unsigned long cache_flush_pos; - spinlock_t lock; - - struct nullb_queue *queues; - unsigned int nr_queues; - char disk_name[DISK_NAME_LEN]; -}; - static LIST_HEAD(nullb_list); static struct mutex lock; static int null_major; @@ -254,6 +180,14 @@ static bool g_use_per_node_hctx; module_param_named(use_per_node_hctx, g_use_per_node_hctx, bool, 0444); MODULE_PARM_DESC(use_per_node_hctx, "Use per-node allocation for hardware context queues. Default: false"); +static bool g_zoned; +module_param_named(zoned, g_zoned, bool, S_IRUGO); +MODULE_PARM_DESC(zoned, "Make device as a host-managed zoned block device. Default: false"); + +static unsigned long g_zone_size = 256; +module_param_named(zone_size, g_zone_size, ulong, S_IRUGO); +MODULE_PARM_DESC(zone_size, "Zone size in MB when block device is zoned. Must be power-of-two: Default: 256"); + static struct nullb_device *null_alloc_dev(void); static void null_free_dev(struct nullb_device *dev); static void null_del_dev(struct nullb *nullb); @@ -357,6 +291,8 @@ NULLB_DEVICE_ATTR(memory_backed, bool); NULLB_DEVICE_ATTR(discard, bool); NULLB_DEVICE_ATTR(mbps, uint); NULLB_DEVICE_ATTR(cache_size, ulong); +NULLB_DEVICE_ATTR(zoned, bool); +NULLB_DEVICE_ATTR(zone_size, ulong); static ssize_t nullb_device_power_show(struct config_item *item, char *page) { @@ -390,6 +326,7 @@ static ssize_t nullb_device_power_store(struct config_item *item, null_del_dev(dev->nullb); mutex_unlock(&lock); clear_bit(NULLB_DEV_FL_UP, &dev->flags); + clear_bit(NULLB_DEV_FL_CONFIGURED, &dev->flags); } return count; @@ -468,6 +405,8 @@ static struct configfs_attribute *nullb_device_attrs[] = { &nullb_device_attr_mbps, &nullb_device_attr_cache_size, &nullb_device_attr_badblocks, + &nullb_device_attr_zoned, + &nullb_device_attr_zone_size, NULL, }; @@ -520,7 +459,7 @@ nullb_group_drop_item(struct config_group *group, struct config_item *item) static ssize_t memb_group_features_show(struct config_item *item, char *page) { - return snprintf(page, PAGE_SIZE, "memory_backed,discard,bandwidth,cache,badblocks\n"); + return snprintf(page, PAGE_SIZE, "memory_backed,discard,bandwidth,cache,badblocks,zoned,zone_size\n"); } CONFIGFS_ATTR_RO(memb_group_, features); @@ -579,6 +518,8 @@ static struct nullb_device *null_alloc_dev(void) dev->hw_queue_depth = g_hw_queue_depth; dev->blocking = g_blocking; dev->use_per_node_hctx = g_use_per_node_hctx; + dev->zoned = g_zoned; + dev->zone_size = g_zone_size; return dev; } @@ -587,6 +528,7 @@ static void null_free_dev(struct nullb_device *dev) if (!dev) return; + null_zone_exit(dev); badblocks_exit(&dev->badblocks); kfree(dev); } @@ -862,7 +804,9 @@ static struct nullb_page *null_lookup_page(struct nullb *nullb, } static struct nullb_page *null_insert_page(struct nullb *nullb, - sector_t sector, bool ignore_cache) + sector_t sector, bool ignore_cache) + __releases(&nullb->lock) + __acquires(&nullb->lock) { u64 idx; struct nullb_page *t_page; @@ -1219,6 +1163,11 @@ static blk_status_t null_handle_cmd(struct nullb_cmd *cmd) struct nullb *nullb = dev->nullb; int err = 0; + if (req_op(cmd->rq) == REQ_OP_ZONE_REPORT) { + cmd->error = null_zone_report(nullb, cmd); + goto out; + } + if (test_bit(NULLB_DEV_FL_THROTTLED, &dev->flags)) { struct request *rq = cmd->rq; @@ -1283,6 +1232,13 @@ static blk_status_t null_handle_cmd(struct nullb_cmd *cmd) } } cmd->error = errno_to_blk_status(err); + + if (!cmd->error && dev->zoned) { + if (req_op(cmd->rq) == REQ_OP_WRITE) + null_zone_write(cmd); + else if (req_op(cmd->rq) == REQ_OP_ZONE_RESET) + null_zone_reset(cmd); + } out: /* Complete IO by inline, softirq or timer */ switch (dev->irqmode) { @@ -1810,6 +1766,15 @@ static int null_add_dev(struct nullb_device *dev) blk_queue_flush_queueable(nullb->q, true); } + if (dev->zoned) { + rv = null_zone_init(dev); + if (rv) + goto out_cleanup_blk_queue; + + blk_queue_chunk_sectors(nullb->q, dev->zone_size_sects); + nullb->q->limits.zoned = BLK_ZONED_HM; + } + nullb->q->queuedata = nullb; blk_queue_flag_set(QUEUE_FLAG_NONROT, nullb->q); blk_queue_flag_clear(QUEUE_FLAG_ADD_RANDOM, nullb->q); @@ -1828,13 +1793,16 @@ static int null_add_dev(struct nullb_device *dev) rv = null_gendisk_register(nullb); if (rv) - goto out_cleanup_blk_queue; + goto out_cleanup_zone; mutex_lock(&lock); list_add_tail(&nullb->list, &nullb_list); mutex_unlock(&lock); return 0; +out_cleanup_zone: + if (dev->zoned) + null_zone_exit(dev); out_cleanup_blk_queue: blk_cleanup_queue(nullb->q); out_cleanup_tags: @@ -1861,6 +1829,11 @@ static int __init null_init(void) g_bs = PAGE_SIZE; } + if (!is_power_of_2(g_zone_size)) { + pr_err("null_blk: zone_size must be power-of-two\n"); + return -EINVAL; + } + if (g_queue_mode == NULL_Q_MQ && g_use_per_node_hctx) { if (g_submit_queues != nr_online_nodes) { pr_warn("null_blk: submit_queues param is set to %u.\n", diff --git a/drivers/block/null_blk_zoned.c b/drivers/block/null_blk_zoned.c new file mode 100644 index 000000000000..a979ca00d7be --- /dev/null +++ b/drivers/block/null_blk_zoned.c @@ -0,0 +1,149 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/vmalloc.h> +#include "null_blk.h" + +/* zone_size in MBs to sectors. */ +#define ZONE_SIZE_SHIFT 11 + +static inline unsigned int null_zone_no(struct nullb_device *dev, sector_t sect) +{ + return sect >> ilog2(dev->zone_size_sects); +} + +int null_zone_init(struct nullb_device *dev) +{ + sector_t dev_size = (sector_t)dev->size * 1024 * 1024; + sector_t sector = 0; + unsigned int i; + + if (!is_power_of_2(dev->zone_size)) { + pr_err("null_blk: zone_size must be power-of-two\n"); + return -EINVAL; + } + + dev->zone_size_sects = dev->zone_size << ZONE_SIZE_SHIFT; + dev->nr_zones = dev_size >> + (SECTOR_SHIFT + ilog2(dev->zone_size_sects)); + dev->zones = kvmalloc_array(dev->nr_zones, sizeof(struct blk_zone), + GFP_KERNEL | __GFP_ZERO); + if (!dev->zones) + return -ENOMEM; + + for (i = 0; i < dev->nr_zones; i++) { + struct blk_zone *zone = &dev->zones[i]; + + zone->start = zone->wp = sector; + zone->len = dev->zone_size_sects; + zone->type = BLK_ZONE_TYPE_SEQWRITE_REQ; + zone->cond = BLK_ZONE_COND_EMPTY; + + sector += dev->zone_size_sects; + } + + return 0; +} + +void null_zone_exit(struct nullb_device *dev) +{ + kvfree(dev->zones); +} + +static void null_zone_fill_rq(struct nullb_device *dev, struct request *rq, + unsigned int zno, unsigned int nr_zones) +{ + struct blk_zone_report_hdr *hdr = NULL; + struct bio_vec bvec; + struct bvec_iter iter; + void *addr; + unsigned int zones_to_cpy; + + bio_for_each_segment(bvec, rq->bio, iter) { + addr = kmap_atomic(bvec.bv_page); + + zones_to_cpy = bvec.bv_len / sizeof(struct blk_zone); + + if (!hdr) { + hdr = (struct blk_zone_report_hdr *)addr; + hdr->nr_zones = nr_zones; + zones_to_cpy--; + addr += sizeof(struct blk_zone_report_hdr); + } + + zones_to_cpy = min_t(unsigned int, zones_to_cpy, nr_zones); + + memcpy(addr, &dev->zones[zno], + zones_to_cpy * sizeof(struct blk_zone)); + + kunmap_atomic(addr); + + nr_zones -= zones_to_cpy; + zno += zones_to_cpy; + + if (!nr_zones) + break; + } +} + +blk_status_t null_zone_report(struct nullb *nullb, + struct nullb_cmd *cmd) +{ + struct nullb_device *dev = nullb->dev; + struct request *rq = cmd->rq; + unsigned int zno = null_zone_no(dev, blk_rq_pos(rq)); + unsigned int nr_zones = dev->nr_zones - zno; + unsigned int max_zones = (blk_rq_bytes(rq) / + sizeof(struct blk_zone)) - 1; + + nr_zones = min_t(unsigned int, nr_zones, max_zones); + + null_zone_fill_rq(nullb->dev, rq, zno, nr_zones); + + return BLK_STS_OK; +} + +void null_zone_write(struct nullb_cmd *cmd) +{ + struct nullb_device *dev = cmd->nq->dev; + struct request *rq = cmd->rq; + sector_t sector = blk_rq_pos(rq); + unsigned int rq_sectors = blk_rq_sectors(rq); + unsigned int zno = null_zone_no(dev, sector); + struct blk_zone *zone = &dev->zones[zno]; + + switch (zone->cond) { + case BLK_ZONE_COND_FULL: + /* Cannot write to a full zone */ + cmd->error = BLK_STS_IOERR; + break; + case BLK_ZONE_COND_EMPTY: + case BLK_ZONE_COND_IMP_OPEN: + /* Writes must be at the write pointer position */ + if (blk_rq_pos(rq) != zone->wp) { + cmd->error = BLK_STS_IOERR; + break; + } + + if (zone->cond == BLK_ZONE_COND_EMPTY) + zone->cond = BLK_ZONE_COND_IMP_OPEN; + + zone->wp += rq_sectors; + if (zone->wp == zone->start + zone->len) + zone->cond = BLK_ZONE_COND_FULL; + break; + default: + /* Invalid zone condition */ + cmd->error = BLK_STS_IOERR; + break; + } +} + +void null_zone_reset(struct nullb_cmd *cmd) +{ + struct nullb_device *dev = cmd->nq->dev; + struct request *rq = cmd->rq; + unsigned int zno = null_zone_no(dev, blk_rq_pos(rq)); + struct blk_zone *zone = &dev->zones[zno]; + + zone->cond = BLK_ZONE_COND_EMPTY; + zone->wp = zone->start; +} diff --git a/drivers/block/paride/bpck.c b/drivers/block/paride/bpck.c index 4f27e7392e38..f5f63ca2889d 100644 --- a/drivers/block/paride/bpck.c +++ b/drivers/block/paride/bpck.c @@ -347,7 +347,7 @@ static int bpck_test_proto( PIA *pi, char * scratch, int verbose ) static void bpck_read_eeprom ( PIA *pi, char * buf ) -{ int i,j,k,n,p,v,f, om, od; +{ int i, j, k, p, v, f, om, od; bpck_force_spp(pi); @@ -356,7 +356,6 @@ static void bpck_read_eeprom ( PIA *pi, char * buf ) bpck_connect(pi); - n = 0; WR(4,0); for (i=0;i<64;i++) { WR(6,8); diff --git a/drivers/block/paride/pd.c b/drivers/block/paride/pd.c index 8961b190e256..7cf947586fe4 100644 --- a/drivers/block/paride/pd.c +++ b/drivers/block/paride/pd.c @@ -426,6 +426,7 @@ static void run_fsm(void) pd_claimed = 1; if (!pi_schedule_claimed(pi_current, run_fsm)) return; + /* fall through */ case 1: pd_claimed = 2; pi_current->proto->connect(pi_current); @@ -445,6 +446,7 @@ static void run_fsm(void) spin_unlock_irqrestore(&pd_lock, saved_flags); if (stop) return; + /* fall through */ case Hold: schedule_fsm(); return; diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c index b3f83cd96f33..e285413d4a75 100644 --- a/drivers/block/pktcdvd.c +++ b/drivers/block/pktcdvd.c @@ -67,7 +67,7 @@ #include <scsi/scsi.h> #include <linux/debugfs.h> #include <linux/device.h> - +#include <linux/nospec.h> #include <linux/uaccess.h> #define DRIVER_NAME "pktcdvd" @@ -748,13 +748,13 @@ static const char *sense_key_string(__u8 index) static void pkt_dump_sense(struct pktcdvd_device *pd, struct packet_command *cgc) { - struct request_sense *sense = cgc->sense; + struct scsi_sense_hdr *sshdr = cgc->sshdr; - if (sense) + if (sshdr) pkt_err(pd, "%*ph - sense %02x.%02x.%02x (%s)\n", CDROM_PACKET_SIZE, cgc->cmd, - sense->sense_key, sense->asc, sense->ascq, - sense_key_string(sense->sense_key)); + sshdr->sense_key, sshdr->asc, sshdr->ascq, + sense_key_string(sshdr->sense_key)); else pkt_err(pd, "%*ph - no sense\n", CDROM_PACKET_SIZE, cgc->cmd); } @@ -787,18 +787,19 @@ static noinline_for_stack int pkt_set_speed(struct pktcdvd_device *pd, unsigned write_speed, unsigned read_speed) { struct packet_command cgc; - struct request_sense sense; + struct scsi_sense_hdr sshdr; int ret; init_cdrom_command(&cgc, NULL, 0, CGC_DATA_NONE); - cgc.sense = &sense; + cgc.sshdr = &sshdr; cgc.cmd[0] = GPCMD_SET_SPEED; cgc.cmd[2] = (read_speed >> 8) & 0xff; cgc.cmd[3] = read_speed & 0xff; cgc.cmd[4] = (write_speed >> 8) & 0xff; cgc.cmd[5] = write_speed & 0xff; - if ((ret = pkt_generic_packet(pd, &cgc))) + ret = pkt_generic_packet(pd, &cgc); + if (ret) pkt_dump_sense(pd, &cgc); return ret; @@ -1562,7 +1563,8 @@ static int pkt_get_disc_info(struct pktcdvd_device *pd, disc_information *di) cgc.cmd[8] = cgc.buflen = 2; cgc.quiet = 1; - if ((ret = pkt_generic_packet(pd, &cgc))) + ret = pkt_generic_packet(pd, &cgc); + if (ret) return ret; /* not all drives have the same disc_info length, so requeue @@ -1591,7 +1593,8 @@ static int pkt_get_track_info(struct pktcdvd_device *pd, __u16 track, __u8 type, cgc.cmd[8] = 8; cgc.quiet = 1; - if ((ret = pkt_generic_packet(pd, &cgc))) + ret = pkt_generic_packet(pd, &cgc); + if (ret) return ret; cgc.buflen = be16_to_cpu(ti->track_information_length) + @@ -1612,17 +1615,20 @@ static noinline_for_stack int pkt_get_last_written(struct pktcdvd_device *pd, __u32 last_track; int ret = -1; - if ((ret = pkt_get_disc_info(pd, &di))) + ret = pkt_get_disc_info(pd, &di); + if (ret) return ret; last_track = (di.last_track_msb << 8) | di.last_track_lsb; - if ((ret = pkt_get_track_info(pd, last_track, 1, &ti))) + ret = pkt_get_track_info(pd, last_track, 1, &ti); + if (ret) return ret; /* if this track is blank, try the previous. */ if (ti.blank) { last_track--; - if ((ret = pkt_get_track_info(pd, last_track, 1, &ti))) + ret = pkt_get_track_info(pd, last_track, 1, &ti); + if (ret) return ret; } @@ -1645,7 +1651,7 @@ static noinline_for_stack int pkt_get_last_written(struct pktcdvd_device *pd, static noinline_for_stack int pkt_set_write_settings(struct pktcdvd_device *pd) { struct packet_command cgc; - struct request_sense sense; + struct scsi_sense_hdr sshdr; write_param_page *wp; char buffer[128]; int ret, size; @@ -1656,8 +1662,9 @@ static noinline_for_stack int pkt_set_write_settings(struct pktcdvd_device *pd) memset(buffer, 0, sizeof(buffer)); init_cdrom_command(&cgc, buffer, sizeof(*wp), CGC_DATA_READ); - cgc.sense = &sense; - if ((ret = pkt_mode_sense(pd, &cgc, GPMODE_WRITE_PARMS_PAGE, 0))) { + cgc.sshdr = &sshdr; + ret = pkt_mode_sense(pd, &cgc, GPMODE_WRITE_PARMS_PAGE, 0); + if (ret) { pkt_dump_sense(pd, &cgc); return ret; } @@ -1671,8 +1678,9 @@ static noinline_for_stack int pkt_set_write_settings(struct pktcdvd_device *pd) * now get it all */ init_cdrom_command(&cgc, buffer, size, CGC_DATA_READ); - cgc.sense = &sense; - if ((ret = pkt_mode_sense(pd, &cgc, GPMODE_WRITE_PARMS_PAGE, 0))) { + cgc.sshdr = &sshdr; + ret = pkt_mode_sense(pd, &cgc, GPMODE_WRITE_PARMS_PAGE, 0); + if (ret) { pkt_dump_sense(pd, &cgc); return ret; } @@ -1714,7 +1722,8 @@ static noinline_for_stack int pkt_set_write_settings(struct pktcdvd_device *pd) wp->packet_size = cpu_to_be32(pd->settings.size >> 2); cgc.buflen = cgc.cmd[8] = size; - if ((ret = pkt_mode_select(pd, &cgc))) { + ret = pkt_mode_select(pd, &cgc); + if (ret) { pkt_dump_sense(pd, &cgc); return ret; } @@ -1819,7 +1828,8 @@ static noinline_for_stack int pkt_probe_settings(struct pktcdvd_device *pd) memset(&di, 0, sizeof(disc_information)); memset(&ti, 0, sizeof(track_information)); - if ((ret = pkt_get_disc_info(pd, &di))) { + ret = pkt_get_disc_info(pd, &di); + if (ret) { pkt_err(pd, "failed get_disc\n"); return ret; } @@ -1830,7 +1840,8 @@ static noinline_for_stack int pkt_probe_settings(struct pktcdvd_device *pd) pd->type = di.erasable ? PACKET_CDRW : PACKET_CDR; track = 1; /* (di.last_track_msb << 8) | di.last_track_lsb; */ - if ((ret = pkt_get_track_info(pd, track, 1, &ti))) { + ret = pkt_get_track_info(pd, track, 1, &ti); + if (ret) { pkt_err(pd, "failed get_track\n"); return ret; } @@ -1905,12 +1916,12 @@ static noinline_for_stack int pkt_write_caching(struct pktcdvd_device *pd, int set) { struct packet_command cgc; - struct request_sense sense; + struct scsi_sense_hdr sshdr; unsigned char buf[64]; int ret; init_cdrom_command(&cgc, buf, sizeof(buf), CGC_DATA_READ); - cgc.sense = &sense; + cgc.sshdr = &sshdr; cgc.buflen = pd->mode_offset + 12; /* @@ -1918,7 +1929,8 @@ static noinline_for_stack int pkt_write_caching(struct pktcdvd_device *pd, */ cgc.quiet = 1; - if ((ret = pkt_mode_sense(pd, &cgc, GPMODE_WCACHING_PAGE, 0))) + ret = pkt_mode_sense(pd, &cgc, GPMODE_WCACHING_PAGE, 0); + if (ret) return ret; buf[pd->mode_offset + 10] |= (!!set << 2); @@ -1950,14 +1962,14 @@ static noinline_for_stack int pkt_get_max_speed(struct pktcdvd_device *pd, unsigned *write_speed) { struct packet_command cgc; - struct request_sense sense; + struct scsi_sense_hdr sshdr; unsigned char buf[256+18]; unsigned char *cap_buf; int ret, offset; cap_buf = &buf[sizeof(struct mode_page_header) + pd->mode_offset]; init_cdrom_command(&cgc, buf, sizeof(buf), CGC_DATA_UNKNOWN); - cgc.sense = &sense; + cgc.sshdr = &sshdr; ret = pkt_mode_sense(pd, &cgc, GPMODE_CAPABILITIES_PAGE, 0); if (ret) { @@ -2011,13 +2023,13 @@ static noinline_for_stack int pkt_media_speed(struct pktcdvd_device *pd, unsigned *speed) { struct packet_command cgc; - struct request_sense sense; + struct scsi_sense_hdr sshdr; unsigned char buf[64]; unsigned int size, st, sp; int ret; init_cdrom_command(&cgc, buf, 2, CGC_DATA_READ); - cgc.sense = &sense; + cgc.sshdr = &sshdr; cgc.cmd[0] = GPCMD_READ_TOC_PMA_ATIP; cgc.cmd[1] = 2; cgc.cmd[2] = 4; /* READ ATIP */ @@ -2032,7 +2044,7 @@ static noinline_for_stack int pkt_media_speed(struct pktcdvd_device *pd, size = sizeof(buf); init_cdrom_command(&cgc, buf, size, CGC_DATA_READ); - cgc.sense = &sense; + cgc.sshdr = &sshdr; cgc.cmd[0] = GPCMD_READ_TOC_PMA_ATIP; cgc.cmd[1] = 2; cgc.cmd[2] = 4; @@ -2083,17 +2095,18 @@ static noinline_for_stack int pkt_media_speed(struct pktcdvd_device *pd, static noinline_for_stack int pkt_perform_opc(struct pktcdvd_device *pd) { struct packet_command cgc; - struct request_sense sense; + struct scsi_sense_hdr sshdr; int ret; pkt_dbg(2, pd, "Performing OPC\n"); init_cdrom_command(&cgc, NULL, 0, CGC_DATA_NONE); - cgc.sense = &sense; + cgc.sshdr = &sshdr; cgc.timeout = 60*HZ; cgc.cmd[0] = GPCMD_SEND_OPC; cgc.cmd[1] = 1; - if ((ret = pkt_generic_packet(pd, &cgc))) + ret = pkt_generic_packet(pd, &cgc); + if (ret) pkt_dump_sense(pd, &cgc); return ret; } @@ -2103,19 +2116,22 @@ static int pkt_open_write(struct pktcdvd_device *pd) int ret; unsigned int write_speed, media_write_speed, read_speed; - if ((ret = pkt_probe_settings(pd))) { + ret = pkt_probe_settings(pd); + if (ret) { pkt_dbg(2, pd, "failed probe\n"); return ret; } - if ((ret = pkt_set_write_settings(pd))) { + ret = pkt_set_write_settings(pd); + if (ret) { pkt_dbg(1, pd, "failed saving write settings\n"); return -EIO; } pkt_write_caching(pd, USE_WCACHING); - if ((ret = pkt_get_max_speed(pd, &write_speed))) + ret = pkt_get_max_speed(pd, &write_speed); + if (ret) write_speed = 16 * 177; switch (pd->mmc3_profile) { case 0x13: /* DVD-RW */ @@ -2124,7 +2140,8 @@ static int pkt_open_write(struct pktcdvd_device *pd) pkt_dbg(1, pd, "write speed %ukB/s\n", write_speed); break; default: - if ((ret = pkt_media_speed(pd, &media_write_speed))) + ret = pkt_media_speed(pd, &media_write_speed); + if (ret) media_write_speed = 16; write_speed = min(write_speed, media_write_speed * 177); pkt_dbg(1, pd, "write speed %ux\n", write_speed / 176); @@ -2132,14 +2149,16 @@ static int pkt_open_write(struct pktcdvd_device *pd) } read_speed = write_speed; - if ((ret = pkt_set_speed(pd, write_speed, read_speed))) { + ret = pkt_set_speed(pd, write_speed, read_speed); + if (ret) { pkt_dbg(1, pd, "couldn't set write speed\n"); return -EIO; } pd->write_speed = write_speed; pd->read_speed = read_speed; - if ((ret = pkt_perform_opc(pd))) { + ret = pkt_perform_opc(pd); + if (ret) { pkt_dbg(1, pd, "Optimum Power Calibration failed\n"); } @@ -2161,10 +2180,12 @@ static int pkt_open_dev(struct pktcdvd_device *pd, fmode_t write) * so bdget() can't fail. */ bdget(pd->bdev->bd_dev); - if ((ret = blkdev_get(pd->bdev, FMODE_READ | FMODE_EXCL, pd))) + ret = blkdev_get(pd->bdev, FMODE_READ | FMODE_EXCL, pd); + if (ret) goto out; - if ((ret = pkt_get_last_written(pd, &lba))) { + ret = pkt_get_last_written(pd, &lba); + if (ret) { pkt_err(pd, "pkt_get_last_written failed\n"); goto out_putdev; } @@ -2175,7 +2196,8 @@ static int pkt_open_dev(struct pktcdvd_device *pd, fmode_t write) q = bdev_get_queue(pd->bdev); if (write) { - if ((ret = pkt_open_write(pd))) + ret = pkt_open_write(pd); + if (ret) goto out_putdev; /* * Some CDRW drives can not handle writes larger than one packet, @@ -2190,7 +2212,8 @@ static int pkt_open_dev(struct pktcdvd_device *pd, fmode_t write) clear_bit(PACKET_WRITABLE, &pd->flags); } - if ((ret = pkt_set_segment_merging(pd, q))) + ret = pkt_set_segment_merging(pd, q); + if (ret) goto out_putdev; if (write) { @@ -2231,6 +2254,8 @@ static struct pktcdvd_device *pkt_find_dev_from_minor(unsigned int dev_minor) { if (dev_minor >= MAX_WRITERS) return NULL; + + dev_minor = array_index_nospec(dev_minor, MAX_WRITERS); return pkt_devs[dev_minor]; } diff --git a/drivers/block/rsxx/dev.c b/drivers/block/rsxx/dev.c index dddb3f2490b6..1a92f9e65937 100644 --- a/drivers/block/rsxx/dev.c +++ b/drivers/block/rsxx/dev.c @@ -112,7 +112,7 @@ static const struct block_device_operations rsxx_fops = { static void disk_stats_start(struct rsxx_cardinfo *card, struct bio *bio) { - generic_start_io_acct(card->queue, bio_data_dir(bio), bio_sectors(bio), + generic_start_io_acct(card->queue, bio_op(bio), bio_sectors(bio), &card->gendisk->part0); } @@ -120,8 +120,8 @@ static void disk_stats_complete(struct rsxx_cardinfo *card, struct bio *bio, unsigned long start_time) { - generic_end_io_acct(card->queue, bio_data_dir(bio), - &card->gendisk->part0, start_time); + generic_end_io_acct(card->queue, bio_op(bio), + &card->gendisk->part0, start_time); } static void bio_dma_done_cb(struct rsxx_cardinfo *card, diff --git a/drivers/block/skd_main.c b/drivers/block/skd_main.c index bc7aea6d7b7c..87b9e7fbf062 100644 --- a/drivers/block/skd_main.c +++ b/drivers/block/skd_main.c @@ -657,8 +657,8 @@ static bool skd_preop_sg_list(struct skd_device *skdev, if (unlikely(skdev->dbg_level > 1)) { dev_dbg(&skdev->pdev->dev, - "skreq=%x sksg_list=%p sksg_dma=%llx\n", - skreq->id, skreq->sksg_list, skreq->sksg_dma_address); + "skreq=%x sksg_list=%p sksg_dma=%pad\n", + skreq->id, skreq->sksg_list, &skreq->sksg_dma_address); for (i = 0; i < n_sg; i++) { struct fit_sg_descriptor *sgd = &skreq->sksg_list[i]; @@ -1190,8 +1190,8 @@ static void skd_send_fitmsg(struct skd_device *skdev, { u64 qcmd; - dev_dbg(&skdev->pdev->dev, "dma address 0x%llx, busy=%d\n", - skmsg->mb_dma_address, skd_in_flight(skdev)); + dev_dbg(&skdev->pdev->dev, "dma address %pad, busy=%d\n", + &skmsg->mb_dma_address, skd_in_flight(skdev)); dev_dbg(&skdev->pdev->dev, "msg_buf %p\n", skmsg->msg_buf); qcmd = skmsg->mb_dma_address; @@ -1250,9 +1250,9 @@ static void skd_send_special_fitmsg(struct skd_device *skdev, } dev_dbg(&skdev->pdev->dev, - "skspcl=%p id=%04x sksg_list=%p sksg_dma=%llx\n", + "skspcl=%p id=%04x sksg_list=%p sksg_dma=%pad\n", skspcl, skspcl->req.id, skspcl->req.sksg_list, - skspcl->req.sksg_dma_address); + &skspcl->req.sksg_dma_address); for (i = 0; i < skspcl->req.n_sg; i++) { struct fit_sg_descriptor *sgd = &skspcl->req.sksg_list[i]; @@ -2685,8 +2685,8 @@ static int skd_cons_skmsg(struct skd_device *skdev) WARN(((uintptr_t)skmsg->msg_buf | skmsg->mb_dma_address) & (FIT_QCMD_ALIGN - 1), - "not aligned: msg_buf %p mb_dma_address %#llx\n", - skmsg->msg_buf, skmsg->mb_dma_address); + "not aligned: msg_buf %p mb_dma_address %pad\n", + skmsg->msg_buf, &skmsg->mb_dma_address); memset(skmsg->msg_buf, 0, SKD_N_FITMSG_BYTES); } diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index b5cedccb5d7d..8986adab9bf5 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -251,14 +251,9 @@ static DEFINE_SPINLOCK(minor_lock); #define GRANTS_PER_INDIRECT_FRAME \ (XEN_PAGE_SIZE / sizeof(struct blkif_request_segment)) -#define PSEGS_PER_INDIRECT_FRAME \ - (GRANTS_INDIRECT_FRAME / GRANTS_PSEGS) - #define INDIRECT_GREFS(_grants) \ DIV_ROUND_UP(_grants, GRANTS_PER_INDIRECT_FRAME) -#define GREFS(_psegs) ((_psegs) * GRANTS_PER_PSEG) - static int blkfront_setup_indirect(struct blkfront_ring_info *rinfo); static void blkfront_gather_backend_features(struct blkfront_info *info); static int negotiate_mq(struct blkfront_info *info); @@ -1441,7 +1436,7 @@ static bool blkif_completion(unsigned long *id, /* Wait the second response if not yet here. */ if (s2->status == REQ_WAITING) - return 0; + return false; bret->status = blkif_get_final_status(s->status, s2->status); @@ -1542,7 +1537,7 @@ static bool blkif_completion(unsigned long *id, } } - return 1; + return true; } static irqreturn_t blkif_interrupt(int irq, void *dev_id) diff --git a/drivers/block/zram/zram_drv.c b/drivers/block/zram/zram_drv.c index a390c6d4f72d..c7acf74253a1 100644 --- a/drivers/block/zram/zram_drv.c +++ b/drivers/block/zram/zram_drv.c @@ -1287,17 +1287,16 @@ static void zram_bio_discard(struct zram *zram, u32 index, * Returns 1 if IO request was successfully submitted. */ static int zram_bvec_rw(struct zram *zram, struct bio_vec *bvec, u32 index, - int offset, bool is_write, struct bio *bio) + int offset, unsigned int op, struct bio *bio) { unsigned long start_time = jiffies; - int rw_acct = is_write ? REQ_OP_WRITE : REQ_OP_READ; struct request_queue *q = zram->disk->queue; int ret; - generic_start_io_acct(q, rw_acct, bvec->bv_len >> SECTOR_SHIFT, + generic_start_io_acct(q, op, bvec->bv_len >> SECTOR_SHIFT, &zram->disk->part0); - if (!is_write) { + if (!op_is_write(op)) { atomic64_inc(&zram->stats.num_reads); ret = zram_bvec_read(zram, bvec, index, offset, bio); flush_dcache_page(bvec->bv_page); @@ -1306,14 +1305,14 @@ static int zram_bvec_rw(struct zram *zram, struct bio_vec *bvec, u32 index, ret = zram_bvec_write(zram, bvec, index, offset, bio); } - generic_end_io_acct(q, rw_acct, &zram->disk->part0, start_time); + generic_end_io_acct(q, op, &zram->disk->part0, start_time); zram_slot_lock(zram, index); zram_accessed(zram, index); zram_slot_unlock(zram, index); if (unlikely(ret < 0)) { - if (!is_write) + if (!op_is_write(op)) atomic64_inc(&zram->stats.failed_reads); else atomic64_inc(&zram->stats.failed_writes); @@ -1351,7 +1350,7 @@ static void __zram_make_request(struct zram *zram, struct bio *bio) bv.bv_len = min_t(unsigned int, PAGE_SIZE - offset, unwritten); if (zram_bvec_rw(zram, &bv, index, offset, - op_is_write(bio_op(bio)), bio) < 0) + bio_op(bio), bio) < 0) goto out; bv.bv_offset += bv.bv_len; @@ -1403,7 +1402,7 @@ static void zram_slot_free_notify(struct block_device *bdev, } static int zram_rw_page(struct block_device *bdev, sector_t sector, - struct page *page, bool is_write) + struct page *page, unsigned int op) { int offset, ret; u32 index; @@ -1427,7 +1426,7 @@ static int zram_rw_page(struct block_device *bdev, sector_t sector, bv.bv_len = PAGE_SIZE; bv.bv_offset = 0; - ret = zram_bvec_rw(zram, &bv, index, offset, is_write, NULL); + ret = zram_bvec_rw(zram, &bv, index, offset, op, NULL); out: /* * If I/O fails, just return error(ie, non-zero) without @@ -1442,7 +1441,7 @@ out: switch (ret) { case 0: - page_endio(page, is_write, 0); + page_endio(page, op_is_write(op), 0); break; case 1: ret = 0; diff --git a/drivers/cdrom/cdrom.c b/drivers/cdrom/cdrom.c index a78b8e7085e9..113fc6edb2b0 100644 --- a/drivers/cdrom/cdrom.c +++ b/drivers/cdrom/cdrom.c @@ -282,6 +282,7 @@ #include <linux/blkdev.h> #include <linux/times.h> #include <linux/uaccess.h> +#include <scsi/scsi_common.h> #include <scsi/scsi_request.h> /* used to tell the module to turn on full debugging messages */ @@ -345,10 +346,10 @@ static LIST_HEAD(cdrom_list); int cdrom_dummy_generic_packet(struct cdrom_device_info *cdi, struct packet_command *cgc) { - if (cgc->sense) { - cgc->sense->sense_key = 0x05; - cgc->sense->asc = 0x20; - cgc->sense->ascq = 0x00; + if (cgc->sshdr) { + cgc->sshdr->sense_key = 0x05; + cgc->sshdr->asc = 0x20; + cgc->sshdr->ascq = 0x00; } cgc->stat = -EIO; @@ -2222,9 +2223,12 @@ static int cdrom_read_cdda_bpc(struct cdrom_device_info *cdi, __u8 __user *ubuf, blk_execute_rq(q, cdi->disk, rq, 0); if (scsi_req(rq)->result) { - struct request_sense *s = req->sense; + struct scsi_sense_hdr sshdr; + ret = -EIO; - cdi->last_sense = s->sense_key; + scsi_normalize_sense(req->sense, req->sense_len, + &sshdr); + cdi->last_sense = sshdr.sense_key; } if (blk_rq_unmap_user(bio)) @@ -2943,7 +2947,7 @@ static noinline int mmc_ioctl_cdrom_read_data(struct cdrom_device_info *cdi, struct packet_command *cgc, int cmd) { - struct request_sense sense; + struct scsi_sense_hdr sshdr; struct cdrom_msf msf; int blocksize = 0, format = 0, lba; int ret; @@ -2971,13 +2975,13 @@ static noinline int mmc_ioctl_cdrom_read_data(struct cdrom_device_info *cdi, if (cgc->buffer == NULL) return -ENOMEM; - memset(&sense, 0, sizeof(sense)); - cgc->sense = &sense; + memset(&sshdr, 0, sizeof(sshdr)); + cgc->sshdr = &sshdr; cgc->data_direction = CGC_DATA_READ; ret = cdrom_read_block(cdi, cgc, lba, 1, format, blocksize); - if (ret && sense.sense_key == 0x05 && - sense.asc == 0x20 && - sense.ascq == 0x00) { + if (ret && sshdr.sense_key == 0x05 && + sshdr.asc == 0x20 && + sshdr.ascq == 0x00) { /* * SCSI-II devices are not required to support * READ_CD, so let's try switching block size @@ -2986,7 +2990,7 @@ static noinline int mmc_ioctl_cdrom_read_data(struct cdrom_device_info *cdi, ret = cdrom_switch_blocksize(cdi, blocksize); if (ret) goto out; - cgc->sense = NULL; + cgc->sshdr = NULL; ret = cdrom_read_cd(cdi, cgc, lba, blocksize, 1); ret |= cdrom_switch_blocksize(cdi, blocksize); } diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c index 5f178384876f..44a7a255ef74 100644 --- a/drivers/ide/ide-cd.c +++ b/drivers/ide/ide-cd.c @@ -419,10 +419,11 @@ static void ide_cd_request_sense_fixup(ide_drive_t *drive, struct ide_cmd *cmd) int ide_cd_queue_pc(ide_drive_t *drive, const unsigned char *cmd, int write, void *buffer, unsigned *bufflen, - struct request_sense *sense, int timeout, + struct scsi_sense_hdr *sshdr, int timeout, req_flags_t rq_flags) { struct cdrom_info *info = drive->driver_data; + struct scsi_sense_hdr local_sshdr; int retries = 10; bool failed; @@ -430,6 +431,9 @@ int ide_cd_queue_pc(ide_drive_t *drive, const unsigned char *cmd, "rq_flags: 0x%x", cmd[0], write, timeout, rq_flags); + if (!sshdr) + sshdr = &local_sshdr; + /* start of retry loop */ do { struct request *rq; @@ -456,8 +460,8 @@ int ide_cd_queue_pc(ide_drive_t *drive, const unsigned char *cmd, if (buffer) *bufflen = scsi_req(rq)->resid_len; - if (sense) - memcpy(sense, scsi_req(rq)->sense, sizeof(*sense)); + scsi_normalize_sense(scsi_req(rq)->sense, + scsi_req(rq)->sense_len, sshdr); /* * FIXME: we should probably abort/retry or something in case of @@ -469,12 +473,10 @@ int ide_cd_queue_pc(ide_drive_t *drive, const unsigned char *cmd, * The request failed. Retry if it was due to a unit * attention status (usually means media was changed). */ - struct request_sense *reqbuf = scsi_req(rq)->sense; - - if (reqbuf->sense_key == UNIT_ATTENTION) + if (sshdr->sense_key == UNIT_ATTENTION) cdrom_saw_media_change(drive); - else if (reqbuf->sense_key == NOT_READY && - reqbuf->asc == 4 && reqbuf->ascq != 4) { + else if (sshdr->sense_key == NOT_READY && + sshdr->asc == 4 && sshdr->ascq != 4) { /* * The drive is in the process of loading * a disk. Retry, but wait a little to give @@ -864,7 +866,7 @@ static void msf_from_bcd(struct atapi_msf *msf) msf->frame = bcd2bin(msf->frame); } -int cdrom_check_status(ide_drive_t *drive, struct request_sense *sense) +int cdrom_check_status(ide_drive_t *drive, struct scsi_sense_hdr *sshdr) { struct cdrom_info *info = drive->driver_data; struct cdrom_device_info *cdi; @@ -886,12 +888,11 @@ int cdrom_check_status(ide_drive_t *drive, struct request_sense *sense) */ cmd[7] = cdi->sanyo_slot % 3; - return ide_cd_queue_pc(drive, cmd, 0, NULL, NULL, sense, 0, RQF_QUIET); + return ide_cd_queue_pc(drive, cmd, 0, NULL, NULL, sshdr, 0, RQF_QUIET); } static int cdrom_read_capacity(ide_drive_t *drive, unsigned long *capacity, - unsigned long *sectors_per_frame, - struct request_sense *sense) + unsigned long *sectors_per_frame) { struct { __be32 lba; @@ -908,7 +909,7 @@ static int cdrom_read_capacity(ide_drive_t *drive, unsigned long *capacity, memset(cmd, 0, BLK_MAX_CDB); cmd[0] = GPCMD_READ_CDVD_CAPACITY; - stat = ide_cd_queue_pc(drive, cmd, 0, &capbuf, &len, sense, 0, + stat = ide_cd_queue_pc(drive, cmd, 0, &capbuf, &len, NULL, 0, RQF_QUIET); if (stat) return stat; @@ -944,8 +945,7 @@ static int cdrom_read_capacity(ide_drive_t *drive, unsigned long *capacity, } static int cdrom_read_tocentry(ide_drive_t *drive, int trackno, int msf_flag, - int format, char *buf, int buflen, - struct request_sense *sense) + int format, char *buf, int buflen) { unsigned char cmd[BLK_MAX_CDB]; @@ -962,11 +962,11 @@ static int cdrom_read_tocentry(ide_drive_t *drive, int trackno, int msf_flag, if (msf_flag) cmd[1] = 2; - return ide_cd_queue_pc(drive, cmd, 0, buf, &buflen, sense, 0, RQF_QUIET); + return ide_cd_queue_pc(drive, cmd, 0, buf, &buflen, NULL, 0, RQF_QUIET); } /* Try to read the entire TOC for the disk into our internal buffer. */ -int ide_cd_read_toc(ide_drive_t *drive, struct request_sense *sense) +int ide_cd_read_toc(ide_drive_t *drive) { int stat, ntracks, i; struct cdrom_info *info = drive->driver_data; @@ -996,14 +996,13 @@ int ide_cd_read_toc(ide_drive_t *drive, struct request_sense *sense) * Check to see if the existing data is still valid. If it is, * just return. */ - (void) cdrom_check_status(drive, sense); + (void) cdrom_check_status(drive, NULL); if (drive->atapi_flags & IDE_AFLAG_TOC_VALID) return 0; /* try to get the total cdrom capacity and sector size */ - stat = cdrom_read_capacity(drive, &toc->capacity, §ors_per_frame, - sense); + stat = cdrom_read_capacity(drive, &toc->capacity, §ors_per_frame); if (stat) toc->capacity = 0x1fffff; @@ -1016,7 +1015,7 @@ int ide_cd_read_toc(ide_drive_t *drive, struct request_sense *sense) /* first read just the header, so we know how long the TOC is */ stat = cdrom_read_tocentry(drive, 0, 1, 0, (char *) &toc->hdr, - sizeof(struct atapi_toc_header), sense); + sizeof(struct atapi_toc_header)); if (stat) return stat; @@ -1036,7 +1035,7 @@ int ide_cd_read_toc(ide_drive_t *drive, struct request_sense *sense) (char *)&toc->hdr, sizeof(struct atapi_toc_header) + (ntracks + 1) * - sizeof(struct atapi_toc_entry), sense); + sizeof(struct atapi_toc_entry)); if (stat && toc->hdr.first_track > 1) { /* @@ -1056,8 +1055,7 @@ int ide_cd_read_toc(ide_drive_t *drive, struct request_sense *sense) (char *)&toc->hdr, sizeof(struct atapi_toc_header) + (ntracks + 1) * - sizeof(struct atapi_toc_entry), - sense); + sizeof(struct atapi_toc_entry)); if (stat) return stat; @@ -1094,7 +1092,7 @@ int ide_cd_read_toc(ide_drive_t *drive, struct request_sense *sense) if (toc->hdr.first_track != CDROM_LEADOUT) { /* read the multisession information */ stat = cdrom_read_tocentry(drive, 0, 0, 1, (char *)&ms_tmp, - sizeof(ms_tmp), sense); + sizeof(ms_tmp)); if (stat) return stat; @@ -1108,7 +1106,7 @@ int ide_cd_read_toc(ide_drive_t *drive, struct request_sense *sense) if (drive->atapi_flags & IDE_AFLAG_TOCADDR_AS_BCD) { /* re-read multisession information using MSF format */ stat = cdrom_read_tocentry(drive, 0, 1, 1, (char *)&ms_tmp, - sizeof(ms_tmp), sense); + sizeof(ms_tmp)); if (stat) return stat; @@ -1412,7 +1410,7 @@ static sector_t ide_cdrom_capacity(ide_drive_t *drive) { unsigned long capacity, sectors_per_frame; - if (cdrom_read_capacity(drive, &capacity, §ors_per_frame, NULL)) + if (cdrom_read_capacity(drive, &capacity, §ors_per_frame)) return 0; return capacity * sectors_per_frame; @@ -1710,9 +1708,8 @@ static unsigned int idecd_check_events(struct gendisk *disk, static int idecd_revalidate_disk(struct gendisk *disk) { struct cdrom_info *info = ide_drv_g(disk, cdrom_info); - struct request_sense sense; - ide_cd_read_toc(info->drive, &sense); + ide_cd_read_toc(info->drive); return 0; } @@ -1736,7 +1733,6 @@ static int ide_cd_probe(ide_drive_t *drive) { struct cdrom_info *info; struct gendisk *g; - struct request_sense sense; ide_debug_log(IDE_DBG_PROBE, "driver_req: %s, media: 0x%x", drive->driver_req, drive->media); @@ -1785,7 +1781,7 @@ static int ide_cd_probe(ide_drive_t *drive) goto failed; } - ide_cd_read_toc(drive, &sense); + ide_cd_read_toc(drive); g->fops = &idecd_ops; g->flags |= GENHD_FL_REMOVABLE | GENHD_FL_BLOCK_EVENTS_ON_EXCL_WRITE; device_add_disk(&drive->gendev, g); diff --git a/drivers/ide/ide-cd.h b/drivers/ide/ide-cd.h index 04f0f310a856..a69dc7f61c4d 100644 --- a/drivers/ide/ide-cd.h +++ b/drivers/ide/ide-cd.h @@ -98,11 +98,11 @@ void ide_cd_log_error(const char *, struct request *, struct request_sense *); /* ide-cd.c functions used by ide-cd_ioctl.c */ int ide_cd_queue_pc(ide_drive_t *, const unsigned char *, int, void *, - unsigned *, struct request_sense *, int, req_flags_t); -int ide_cd_read_toc(ide_drive_t *, struct request_sense *); + unsigned *, struct scsi_sense_hdr *, int, req_flags_t); +int ide_cd_read_toc(ide_drive_t *); int ide_cdrom_get_capabilities(ide_drive_t *, u8 *); void ide_cdrom_update_speed(ide_drive_t *, u8 *); -int cdrom_check_status(ide_drive_t *, struct request_sense *); +int cdrom_check_status(ide_drive_t *, struct scsi_sense_hdr *); /* ide-cd_ioctl.c */ int ide_cdrom_open_real(struct cdrom_device_info *, int); diff --git a/drivers/ide/ide-cd_ioctl.c b/drivers/ide/ide-cd_ioctl.c index b1322400887b..4a6e1a413ead 100644 --- a/drivers/ide/ide-cd_ioctl.c +++ b/drivers/ide/ide-cd_ioctl.c @@ -43,14 +43,14 @@ int ide_cdrom_drive_status(struct cdrom_device_info *cdi, int slot_nr) { ide_drive_t *drive = cdi->handle; struct media_event_desc med; - struct request_sense sense; + struct scsi_sense_hdr sshdr; int stat; if (slot_nr != CDSL_CURRENT) return -EINVAL; - stat = cdrom_check_status(drive, &sense); - if (!stat || sense.sense_key == UNIT_ATTENTION) + stat = cdrom_check_status(drive, &sshdr); + if (!stat || sshdr.sense_key == UNIT_ATTENTION) return CDS_DISC_OK; if (!cdrom_get_media_event(cdi, &med)) { @@ -62,8 +62,8 @@ int ide_cdrom_drive_status(struct cdrom_device_info *cdi, int slot_nr) return CDS_NO_DISC; } - if (sense.sense_key == NOT_READY && sense.asc == 0x04 - && sense.ascq == 0x04) + if (sshdr.sense_key == NOT_READY && sshdr.asc == 0x04 + && sshdr.ascq == 0x04) return CDS_DISC_OK; /* @@ -71,8 +71,8 @@ int ide_cdrom_drive_status(struct cdrom_device_info *cdi, int slot_nr) * just return TRAY_OPEN since ATAPI doesn't provide * any other way to detect this... */ - if (sense.sense_key == NOT_READY) { - if (sense.asc == 0x3a && sense.ascq == 1) + if (sshdr.sense_key == NOT_READY) { + if (sshdr.asc == 0x3a && sshdr.ascq == 1) return CDS_NO_DISC; else return CDS_TRAY_OPEN; @@ -105,8 +105,7 @@ unsigned int ide_cdrom_check_events_real(struct cdrom_device_info *cdi, /* Eject the disk if EJECTFLAG is 0. If EJECTFLAG is 1, try to reload the disk. */ static -int cdrom_eject(ide_drive_t *drive, int ejectflag, - struct request_sense *sense) +int cdrom_eject(ide_drive_t *drive, int ejectflag) { struct cdrom_info *cd = drive->driver_data; struct cdrom_device_info *cdi = &cd->devinfo; @@ -129,20 +128,16 @@ int cdrom_eject(ide_drive_t *drive, int ejectflag, cmd[0] = GPCMD_START_STOP_UNIT; cmd[4] = loej | (ejectflag != 0); - return ide_cd_queue_pc(drive, cmd, 0, NULL, NULL, sense, 0, 0); + return ide_cd_queue_pc(drive, cmd, 0, NULL, NULL, NULL, 0, 0); } /* Lock the door if LOCKFLAG is nonzero; unlock it otherwise. */ static -int ide_cd_lockdoor(ide_drive_t *drive, int lockflag, - struct request_sense *sense) +int ide_cd_lockdoor(ide_drive_t *drive, int lockflag) { - struct request_sense my_sense; + struct scsi_sense_hdr sshdr; int stat; - if (sense == NULL) - sense = &my_sense; - /* If the drive cannot lock the door, just pretend. */ if ((drive->dev_flags & IDE_DFLAG_DOORLOCKING) == 0) { stat = 0; @@ -155,14 +150,14 @@ int ide_cd_lockdoor(ide_drive_t *drive, int lockflag, cmd[4] = lockflag ? 1 : 0; stat = ide_cd_queue_pc(drive, cmd, 0, NULL, NULL, - sense, 0, 0); + &sshdr, 0, 0); } /* If we got an illegal field error, the drive probably cannot lock the door. */ if (stat != 0 && - sense->sense_key == ILLEGAL_REQUEST && - (sense->asc == 0x24 || sense->asc == 0x20)) { + sshdr.sense_key == ILLEGAL_REQUEST && + (sshdr.asc == 0x24 || sshdr.asc == 0x20)) { printk(KERN_ERR "%s: door locking not supported\n", drive->name); drive->dev_flags &= ~IDE_DFLAG_DOORLOCKING; @@ -170,7 +165,7 @@ int ide_cd_lockdoor(ide_drive_t *drive, int lockflag, } /* no medium, that's alright. */ - if (stat != 0 && sense->sense_key == NOT_READY && sense->asc == 0x3a) + if (stat != 0 && sshdr.sense_key == NOT_READY && sshdr.asc == 0x3a) stat = 0; if (stat == 0) { @@ -186,23 +181,22 @@ int ide_cd_lockdoor(ide_drive_t *drive, int lockflag, int ide_cdrom_tray_move(struct cdrom_device_info *cdi, int position) { ide_drive_t *drive = cdi->handle; - struct request_sense sense; if (position) { - int stat = ide_cd_lockdoor(drive, 0, &sense); + int stat = ide_cd_lockdoor(drive, 0); if (stat) return stat; } - return cdrom_eject(drive, !position, &sense); + return cdrom_eject(drive, !position); } int ide_cdrom_lock_door(struct cdrom_device_info *cdi, int lock) { ide_drive_t *drive = cdi->handle; - return ide_cd_lockdoor(drive, lock, NULL); + return ide_cd_lockdoor(drive, lock); } /* @@ -213,7 +207,6 @@ int ide_cdrom_select_speed(struct cdrom_device_info *cdi, int speed) { ide_drive_t *drive = cdi->handle; struct cdrom_info *cd = drive->driver_data; - struct request_sense sense; u8 buf[ATAPI_CAPABILITIES_PAGE_SIZE]; int stat; unsigned char cmd[BLK_MAX_CDB]; @@ -236,7 +229,7 @@ int ide_cdrom_select_speed(struct cdrom_device_info *cdi, int speed) cmd[5] = speed & 0xff; } - stat = ide_cd_queue_pc(drive, cmd, 0, NULL, NULL, &sense, 0, 0); + stat = ide_cd_queue_pc(drive, cmd, 0, NULL, NULL, NULL, 0, 0); if (!ide_cdrom_get_capabilities(drive, buf)) { ide_cdrom_update_speed(drive, buf); @@ -252,11 +245,10 @@ int ide_cdrom_get_last_session(struct cdrom_device_info *cdi, struct atapi_toc *toc; ide_drive_t *drive = cdi->handle; struct cdrom_info *info = drive->driver_data; - struct request_sense sense; int ret; if ((drive->atapi_flags & IDE_AFLAG_TOC_VALID) == 0 || !info->toc) { - ret = ide_cd_read_toc(drive, &sense); + ret = ide_cd_read_toc(drive); if (ret) return ret; } @@ -300,7 +292,6 @@ int ide_cdrom_reset(struct cdrom_device_info *cdi) { ide_drive_t *drive = cdi->handle; struct cdrom_info *cd = drive->driver_data; - struct request_sense sense; struct request *rq; int ret; @@ -315,7 +306,7 @@ int ide_cdrom_reset(struct cdrom_device_info *cdi) * lock it again. */ if (drive->atapi_flags & IDE_AFLAG_DOOR_LOCKED) - (void)ide_cd_lockdoor(drive, 1, &sense); + (void)ide_cd_lockdoor(drive, 1); return ret; } @@ -355,7 +346,6 @@ static int ide_cd_fake_play_trkind(ide_drive_t *drive, void *arg) struct atapi_toc_entry *first_toc, *last_toc; unsigned long lba_start, lba_end; int stat; - struct request_sense sense; unsigned char cmd[BLK_MAX_CDB]; stat = ide_cd_get_toc_entry(drive, ti->cdti_trk0, &first_toc); @@ -380,7 +370,7 @@ static int ide_cd_fake_play_trkind(ide_drive_t *drive, void *arg) lba_to_msf(lba_start, &cmd[3], &cmd[4], &cmd[5]); lba_to_msf(lba_end - 1, &cmd[6], &cmd[7], &cmd[8]); - return ide_cd_queue_pc(drive, cmd, 0, NULL, NULL, &sense, 0, 0); + return ide_cd_queue_pc(drive, cmd, 0, NULL, NULL, NULL, 0, 0); } static int ide_cd_read_tochdr(ide_drive_t *drive, void *arg) @@ -391,7 +381,7 @@ static int ide_cd_read_tochdr(ide_drive_t *drive, void *arg) int stat; /* Make sure our saved TOC is valid. */ - stat = ide_cd_read_toc(drive, NULL); + stat = ide_cd_read_toc(drive); if (stat) return stat; @@ -461,8 +451,8 @@ int ide_cdrom_packet(struct cdrom_device_info *cdi, layer. the packet must be complete, as we do not touch it at all. */ - if (cgc->sense) - memset(cgc->sense, 0, sizeof(struct request_sense)); + if (cgc->sshdr) + memset(cgc->sshdr, 0, sizeof(*cgc->sshdr)); if (cgc->quiet) flags |= RQF_QUIET; @@ -470,7 +460,7 @@ int ide_cdrom_packet(struct cdrom_device_info *cdi, cgc->stat = ide_cd_queue_pc(drive, cgc->cmd, cgc->data_direction == CGC_DATA_WRITE, cgc->buffer, &len, - cgc->sense, cgc->timeout, flags); + cgc->sshdr, cgc->timeout, flags); if (!cgc->stat) cgc->buflen -= len; return cgc->stat; diff --git a/drivers/infiniband/ulp/iser/iser_memory.c b/drivers/infiniband/ulp/iser/iser_memory.c index ca844a926e6a..130bf163f066 100644 --- a/drivers/infiniband/ulp/iser/iser_memory.c +++ b/drivers/infiniband/ulp/iser/iser_memory.c @@ -311,7 +311,7 @@ iser_set_dif_domain(struct scsi_cmnd *sc, struct ib_sig_attrs *sig_attrs, { domain->sig_type = IB_SIG_TYPE_T10_DIF; domain->sig.dif.pi_interval = scsi_prot_interval(sc); - domain->sig.dif.ref_tag = scsi_prot_ref_tag(sc); + domain->sig.dif.ref_tag = t10_pi_ref_tag(sc->request); /* * At the moment we hard code those, but in the future * we will take them from sc. diff --git a/drivers/lightnvm/Kconfig b/drivers/lightnvm/Kconfig index 9c03f35d9df1..439bf90d084d 100644 --- a/drivers/lightnvm/Kconfig +++ b/drivers/lightnvm/Kconfig @@ -17,23 +17,25 @@ menuconfig NVM if NVM -config NVM_DEBUG - bool "Open-Channel SSD debugging support" - default n - ---help--- - Exposes a debug management interface to create/remove targets at: +config NVM_PBLK + tristate "Physical Block Device Open-Channel SSD target" + help + Allows an open-channel SSD to be exposed as a block device to the + host. The target assumes the device exposes raw flash and must be + explicitly managed by the host. - /sys/module/lnvm/parameters/configure_debug + Please note the disk format is considered EXPERIMENTAL for now. - It is required to create/remove targets without IOCTLs. +if NVM_PBLK -config NVM_PBLK - tristate "Physical Block Device Open-Channel SSD target" - ---help--- - Allows an open-channel SSD to be exposed as a block device to the - host. The target assumes the device exposes raw flash and must be - explicitly managed by the host. +config NVM_PBLK_DEBUG + bool "PBlk Debug Support" + default n + help + Enables debug support for pblk. This includes extra checks, more + vocal error messages, and extra tracking fields in the pblk sysfs + entries. - Please note the disk format is considered EXPERIMENTAL for now. +endif # NVM_PBLK_DEBUG endif # NVM diff --git a/drivers/lightnvm/pblk-cache.c b/drivers/lightnvm/pblk-cache.c index b1c6d7eb6115..f565a56b898a 100644 --- a/drivers/lightnvm/pblk-cache.c +++ b/drivers/lightnvm/pblk-cache.c @@ -27,7 +27,8 @@ int pblk_write_to_cache(struct pblk *pblk, struct bio *bio, unsigned long flags) int nr_entries = pblk_get_secs(bio); int i, ret; - generic_start_io_acct(q, WRITE, bio_sectors(bio), &pblk->disk->part0); + generic_start_io_acct(q, REQ_OP_WRITE, bio_sectors(bio), + &pblk->disk->part0); /* Update the write buffer head (mem) with the entries that we can * write. The write in itself cannot fail, so there is no need to @@ -67,7 +68,7 @@ retry: atomic64_add(nr_entries, &pblk->user_wa); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_long_add(nr_entries, &pblk->inflight_writes); atomic_long_add(nr_entries, &pblk->req_writes); #endif @@ -75,7 +76,7 @@ retry: pblk_rl_inserted(&pblk->rl, nr_entries); out: - generic_end_io_acct(q, WRITE, &pblk->disk->part0, start_time); + generic_end_io_acct(q, REQ_OP_WRITE, &pblk->disk->part0, start_time); pblk_write_should_kick(pblk); return ret; } @@ -123,7 +124,7 @@ retry: atomic64_add(valid_entries, &pblk->gc_wa); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_long_add(valid_entries, &pblk->inflight_writes); atomic_long_add(valid_entries, &pblk->recov_gc_writes); #endif diff --git a/drivers/lightnvm/pblk-core.c b/drivers/lightnvm/pblk-core.c index ed9cc977c8b3..00984b486fea 100644 --- a/drivers/lightnvm/pblk-core.c +++ b/drivers/lightnvm/pblk-core.c @@ -35,7 +35,7 @@ static void pblk_line_mark_bb(struct work_struct *work) line = &pblk->lines[pblk_ppa_to_line(*ppa)]; pos = pblk_ppa_to_pos(&dev->geo, *ppa); - pr_err("pblk: failed to mark bb, line:%d, pos:%d\n", + pblk_err(pblk, "failed to mark bb, line:%d, pos:%d\n", line->id, pos); } @@ -51,12 +51,12 @@ static void pblk_mark_bb(struct pblk *pblk, struct pblk_line *line, struct ppa_addr *ppa; int pos = pblk_ppa_to_pos(geo, ppa_addr); - pr_debug("pblk: erase failed: line:%d, pos:%d\n", line->id, pos); + pblk_debug(pblk, "erase failed: line:%d, pos:%d\n", line->id, pos); atomic_long_inc(&pblk->erase_failed); atomic_dec(&line->blk_in_line); if (test_and_set_bit(pos, line->blk_bitmap)) - pr_err("pblk: attempted to erase bb: line:%d, pos:%d\n", + pblk_err(pblk, "attempted to erase bb: line:%d, pos:%d\n", line->id, pos); /* Not necessary to mark bad blocks on 2.0 spec. */ @@ -194,7 +194,7 @@ void pblk_map_invalidate(struct pblk *pblk, struct ppa_addr ppa) u64 paddr; int line_id; -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG /* Callers must ensure that the ppa points to a device address */ BUG_ON(pblk_addr_in_cache(ppa)); BUG_ON(pblk_ppa_empty(ppa)); @@ -264,6 +264,7 @@ void pblk_free_rqd(struct pblk *pblk, struct nvm_rq *rqd, int type) switch (type) { case PBLK_WRITE: kfree(((struct pblk_c_ctx *)nvm_rq_to_pdu(rqd))->lun_bitmap); + /* fall through */ case PBLK_WRITE_INT: pool = &pblk->w_rq_pool; break; @@ -274,7 +275,7 @@ void pblk_free_rqd(struct pblk *pblk, struct nvm_rq *rqd, int type) pool = &pblk->e_rq_pool; break; default: - pr_err("pblk: trying to free unknown rqd type\n"); + pblk_err(pblk, "trying to free unknown rqd type\n"); return; } @@ -310,7 +311,7 @@ int pblk_bio_add_pages(struct pblk *pblk, struct bio *bio, gfp_t flags, ret = bio_add_pc_page(q, bio, page, PBLK_EXPOSED_PAGE_SIZE, 0); if (ret != PBLK_EXPOSED_PAGE_SIZE) { - pr_err("pblk: could not add page to bio\n"); + pblk_err(pblk, "could not add page to bio\n"); mempool_free(page, &pblk->page_bio_pool); goto err; } @@ -410,7 +411,7 @@ struct list_head *pblk_line_gc_list(struct pblk *pblk, struct pblk_line *line) line->state = PBLK_LINESTATE_CORRUPT; line->gc_group = PBLK_LINEGC_NONE; move_list = &l_mg->corrupt_list; - pr_err("pblk: corrupted vsc for line %d, vsc:%d (%d/%d/%d)\n", + pblk_err(pblk, "corrupted vsc for line %d, vsc:%d (%d/%d/%d)\n", line->id, vsc, line->sec_in_line, lm->high_thrs, lm->mid_thrs); @@ -430,7 +431,7 @@ void pblk_discard(struct pblk *pblk, struct bio *bio) void pblk_log_write_err(struct pblk *pblk, struct nvm_rq *rqd) { atomic_long_inc(&pblk->write_failed); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG pblk_print_failed_rqd(pblk, rqd, rqd->error); #endif } @@ -452,9 +453,9 @@ void pblk_log_read_err(struct pblk *pblk, struct nvm_rq *rqd) atomic_long_inc(&pblk->read_failed); break; default: - pr_err("pblk: unknown read error:%d\n", rqd->error); + pblk_err(pblk, "unknown read error:%d\n", rqd->error); } -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG pblk_print_failed_rqd(pblk, rqd, rqd->error); #endif } @@ -470,7 +471,7 @@ int pblk_submit_io(struct pblk *pblk, struct nvm_rq *rqd) atomic_inc(&pblk->inflight_io); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG if (pblk_check_io(pblk, rqd)) return NVM_IO_ERR; #endif @@ -484,7 +485,7 @@ int pblk_submit_io_sync(struct pblk *pblk, struct nvm_rq *rqd) atomic_inc(&pblk->inflight_io); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG if (pblk_check_io(pblk, rqd)) return NVM_IO_ERR; #endif @@ -517,7 +518,7 @@ struct bio *pblk_bio_map_addr(struct pblk *pblk, void *data, for (i = 0; i < nr_secs; i++) { page = vmalloc_to_page(kaddr); if (!page) { - pr_err("pblk: could not map vmalloc bio\n"); + pblk_err(pblk, "could not map vmalloc bio\n"); bio_put(bio); bio = ERR_PTR(-ENOMEM); goto out; @@ -525,7 +526,7 @@ struct bio *pblk_bio_map_addr(struct pblk *pblk, void *data, ret = bio_add_pc_page(dev->q, bio, page, PAGE_SIZE, 0); if (ret != PAGE_SIZE) { - pr_err("pblk: could not add page to bio\n"); + pblk_err(pblk, "could not add page to bio\n"); bio_put(bio); bio = ERR_PTR(-ENOMEM); goto out; @@ -711,7 +712,7 @@ next_rq: while (test_bit(pos, line->blk_bitmap)) { paddr += min; if (pblk_boundary_paddr_checks(pblk, paddr)) { - pr_err("pblk: corrupt emeta line:%d\n", + pblk_err(pblk, "corrupt emeta line:%d\n", line->id); bio_put(bio); ret = -EINTR; @@ -723,7 +724,7 @@ next_rq: } if (pblk_boundary_paddr_checks(pblk, paddr + min)) { - pr_err("pblk: corrupt emeta line:%d\n", + pblk_err(pblk, "corrupt emeta line:%d\n", line->id); bio_put(bio); ret = -EINTR; @@ -738,7 +739,7 @@ next_rq: ret = pblk_submit_io_sync(pblk, &rqd); if (ret) { - pr_err("pblk: emeta I/O submission failed: %d\n", ret); + pblk_err(pblk, "emeta I/O submission failed: %d\n", ret); bio_put(bio); goto free_rqd_dma; } @@ -843,7 +844,7 @@ static int pblk_line_submit_smeta_io(struct pblk *pblk, struct pblk_line *line, */ ret = pblk_submit_io_sync(pblk, &rqd); if (ret) { - pr_err("pblk: smeta I/O submission failed: %d\n", ret); + pblk_err(pblk, "smeta I/O submission failed: %d\n", ret); bio_put(bio); goto free_ppa_list; } @@ -905,7 +906,7 @@ static int pblk_blk_erase_sync(struct pblk *pblk, struct ppa_addr ppa) struct nvm_tgt_dev *dev = pblk->dev; struct nvm_geo *geo = &dev->geo; - pr_err("pblk: could not sync erase line:%d,blk:%d\n", + pblk_err(pblk, "could not sync erase line:%d,blk:%d\n", pblk_ppa_to_line(ppa), pblk_ppa_to_pos(geo, ppa)); @@ -945,7 +946,7 @@ int pblk_line_erase(struct pblk *pblk, struct pblk_line *line) ret = pblk_blk_erase_sync(pblk, ppa); if (ret) { - pr_err("pblk: failed to erase line %d\n", line->id); + pblk_err(pblk, "failed to erase line %d\n", line->id); return ret; } } while (1); @@ -1012,7 +1013,7 @@ static int pblk_line_init_metadata(struct pblk *pblk, struct pblk_line *line, list_add_tail(&line->list, &l_mg->bad_list); spin_unlock(&l_mg->free_lock); - pr_debug("pblk: line %d is bad\n", line->id); + pblk_debug(pblk, "line %d is bad\n", line->id); return 0; } @@ -1122,7 +1123,7 @@ static int pblk_line_init_bb(struct pblk *pblk, struct pblk_line *line, line->cur_sec = off + lm->smeta_sec; if (init && pblk_line_submit_smeta_io(pblk, line, off, PBLK_WRITE)) { - pr_debug("pblk: line smeta I/O failed. Retry\n"); + pblk_debug(pblk, "line smeta I/O failed. Retry\n"); return 0; } @@ -1154,7 +1155,7 @@ static int pblk_line_init_bb(struct pblk *pblk, struct pblk_line *line, spin_unlock(&line->lock); list_add_tail(&line->list, &l_mg->bad_list); - pr_err("pblk: unexpected line %d is bad\n", line->id); + pblk_err(pblk, "unexpected line %d is bad\n", line->id); return 0; } @@ -1299,7 +1300,7 @@ struct pblk_line *pblk_line_get(struct pblk *pblk) retry: if (list_empty(&l_mg->free_list)) { - pr_err("pblk: no free lines\n"); + pblk_err(pblk, "no free lines\n"); return NULL; } @@ -1315,7 +1316,7 @@ retry: list_add_tail(&line->list, &l_mg->bad_list); - pr_debug("pblk: line %d is bad\n", line->id); + pblk_debug(pblk, "line %d is bad\n", line->id); goto retry; } @@ -1329,7 +1330,7 @@ retry: list_add(&line->list, &l_mg->corrupt_list); goto retry; default: - pr_err("pblk: failed to prepare line %d\n", line->id); + pblk_err(pblk, "failed to prepare line %d\n", line->id); list_add(&line->list, &l_mg->free_list); l_mg->nr_free_lines++; return NULL; @@ -1477,7 +1478,7 @@ static void pblk_line_close_meta_sync(struct pblk *pblk) ret = pblk_submit_meta_io(pblk, line); if (ret) { - pr_err("pblk: sync meta line %d failed (%d)\n", + pblk_err(pblk, "sync meta line %d failed (%d)\n", line->id, ret); return; } @@ -1507,7 +1508,7 @@ void __pblk_pipeline_flush(struct pblk *pblk) ret = pblk_recov_pad(pblk); if (ret) { - pr_err("pblk: could not close data on teardown(%d)\n", ret); + pblk_err(pblk, "could not close data on teardown(%d)\n", ret); return; } @@ -1687,7 +1688,7 @@ int pblk_blk_erase_async(struct pblk *pblk, struct ppa_addr ppa) struct nvm_tgt_dev *dev = pblk->dev; struct nvm_geo *geo = &dev->geo; - pr_err("pblk: could not async erase line:%d,blk:%d\n", + pblk_err(pblk, "could not async erase line:%d,blk:%d\n", pblk_ppa_to_line(ppa), pblk_ppa_to_pos(geo, ppa)); } @@ -1726,7 +1727,7 @@ void pblk_line_close(struct pblk *pblk, struct pblk_line *line) struct list_head *move_list; int i; -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG WARN(!bitmap_full(line->map_bitmap, lm->sec_per_line), "pblk: corrupt closed line %d\n", line->id); #endif @@ -1856,7 +1857,7 @@ static void __pblk_down_page(struct pblk *pblk, struct ppa_addr *ppa_list, * Only send one inflight I/O per LUN. Since we map at a page * granurality, all ppas in the I/O will map to the same LUN */ -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG int i; for (i = 1; i < nr_ppas; i++) @@ -1866,7 +1867,8 @@ static void __pblk_down_page(struct pblk *pblk, struct ppa_addr *ppa_list, ret = down_timeout(&rlun->wr_sem, msecs_to_jiffies(30000)); if (ret == -ETIME || ret == -EINTR) - pr_err("pblk: taking lun semaphore timed out: err %d\n", -ret); + pblk_err(pblk, "taking lun semaphore timed out: err %d\n", + -ret); } void pblk_down_page(struct pblk *pblk, struct ppa_addr *ppa_list, int nr_ppas) @@ -1901,7 +1903,7 @@ void pblk_up_page(struct pblk *pblk, struct ppa_addr *ppa_list, int nr_ppas) struct pblk_lun *rlun; int pos = pblk_ppa_to_pos(geo, ppa_list[0]); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG int i; for (i = 1; i < nr_ppas; i++) @@ -1951,7 +1953,7 @@ void pblk_update_map(struct pblk *pblk, sector_t lba, struct ppa_addr ppa) void pblk_update_map_cache(struct pblk *pblk, sector_t lba, struct ppa_addr ppa) { -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG /* Callers must ensure that the ppa points to a cache address */ BUG_ON(!pblk_addr_in_cache(ppa)); BUG_ON(pblk_rb_pos_oob(&pblk->rwb, pblk_addr_to_cacheline(ppa))); @@ -1966,7 +1968,7 @@ int pblk_update_map_gc(struct pblk *pblk, sector_t lba, struct ppa_addr ppa_new, struct ppa_addr ppa_l2p, ppa_gc; int ret = 1; -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG /* Callers must ensure that the ppa points to a cache address */ BUG_ON(!pblk_addr_in_cache(ppa_new)); BUG_ON(pblk_rb_pos_oob(&pblk->rwb, pblk_addr_to_cacheline(ppa_new))); @@ -2003,14 +2005,14 @@ void pblk_update_map_dev(struct pblk *pblk, sector_t lba, { struct ppa_addr ppa_l2p; -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG /* Callers must ensure that the ppa points to a device address */ BUG_ON(pblk_addr_in_cache(ppa_mapped)); #endif /* Invalidate and discard padded entries */ if (lba == ADDR_EMPTY) { atomic64_inc(&pblk->pad_wa); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_long_inc(&pblk->padded_wb); #endif if (!pblk_ppa_empty(ppa_mapped)) @@ -2036,7 +2038,7 @@ void pblk_update_map_dev(struct pblk *pblk, sector_t lba, goto out; } -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG WARN_ON(!pblk_addr_in_cache(ppa_l2p) && !pblk_ppa_empty(ppa_l2p)); #endif diff --git a/drivers/lightnvm/pblk-gc.c b/drivers/lightnvm/pblk-gc.c index 080469d90b40..157c2567c9e8 100644 --- a/drivers/lightnvm/pblk-gc.c +++ b/drivers/lightnvm/pblk-gc.c @@ -90,7 +90,7 @@ static void pblk_gc_line_ws(struct work_struct *work) gc_rq->data = vmalloc(array_size(gc_rq->nr_secs, geo->csecs)); if (!gc_rq->data) { - pr_err("pblk: could not GC line:%d (%d/%d)\n", + pblk_err(pblk, "could not GC line:%d (%d/%d)\n", line->id, *line->vsc, gc_rq->nr_secs); goto out; } @@ -98,7 +98,7 @@ static void pblk_gc_line_ws(struct work_struct *work) /* Read from GC victim block */ ret = pblk_submit_read_gc(pblk, gc_rq); if (ret) { - pr_err("pblk: failed GC read in line:%d (err:%d)\n", + pblk_err(pblk, "failed GC read in line:%d (err:%d)\n", line->id, ret); goto out; } @@ -146,7 +146,7 @@ static __le64 *get_lba_list_from_emeta(struct pblk *pblk, ret = pblk_line_read_emeta(pblk, line, emeta_buf); if (ret) { - pr_err("pblk: line %d read emeta failed (%d)\n", + pblk_err(pblk, "line %d read emeta failed (%d)\n", line->id, ret); pblk_mfree(emeta_buf, l_mg->emeta_alloc_type); return NULL; @@ -160,7 +160,7 @@ static __le64 *get_lba_list_from_emeta(struct pblk *pblk, ret = pblk_recov_check_emeta(pblk, emeta_buf); if (ret) { - pr_err("pblk: inconsistent emeta (line %d)\n", + pblk_err(pblk, "inconsistent emeta (line %d)\n", line->id); pblk_mfree(emeta_buf, l_mg->emeta_alloc_type); return NULL; @@ -201,7 +201,7 @@ static void pblk_gc_line_prepare_ws(struct work_struct *work) } else { lba_list = get_lba_list_from_emeta(pblk, line); if (!lba_list) { - pr_err("pblk: could not interpret emeta (line %d)\n", + pblk_err(pblk, "could not interpret emeta (line %d)\n", line->id); goto fail_free_invalid_bitmap; } @@ -213,7 +213,7 @@ static void pblk_gc_line_prepare_ws(struct work_struct *work) spin_unlock(&line->lock); if (sec_left < 0) { - pr_err("pblk: corrupted GC line (%d)\n", line->id); + pblk_err(pblk, "corrupted GC line (%d)\n", line->id); goto fail_free_lba_list; } @@ -289,7 +289,7 @@ fail_free_ws: kref_put(&line->ref, pblk_line_put); atomic_dec(&gc->read_inflight_gc); - pr_err("pblk: Failed to GC line %d\n", line->id); + pblk_err(pblk, "failed to GC line %d\n", line->id); } static int pblk_gc_line(struct pblk *pblk, struct pblk_line *line) @@ -297,7 +297,7 @@ static int pblk_gc_line(struct pblk *pblk, struct pblk_line *line) struct pblk_gc *gc = &pblk->gc; struct pblk_line_ws *line_ws; - pr_debug("pblk: line '%d' being reclaimed for GC\n", line->id); + pblk_debug(pblk, "line '%d' being reclaimed for GC\n", line->id); line_ws = kmalloc(sizeof(struct pblk_line_ws), GFP_KERNEL); if (!line_ws) @@ -351,7 +351,7 @@ static int pblk_gc_read(struct pblk *pblk) pblk_gc_kick(pblk); if (pblk_gc_line(pblk, line)) - pr_err("pblk: failed to GC line %d\n", line->id); + pblk_err(pblk, "failed to GC line %d\n", line->id); return 0; } @@ -522,8 +522,8 @@ static int pblk_gc_reader_ts(void *data) io_schedule(); } -#ifdef CONFIG_NVM_DEBUG - pr_info("pblk: flushing gc pipeline, %d lines left\n", +#ifdef CONFIG_NVM_PBLK_DEBUG + pblk_info(pblk, "flushing gc pipeline, %d lines left\n", atomic_read(&gc->pipeline_gc)); #endif @@ -540,7 +540,7 @@ static int pblk_gc_reader_ts(void *data) static void pblk_gc_start(struct pblk *pblk) { pblk->gc.gc_active = 1; - pr_debug("pblk: gc start\n"); + pblk_debug(pblk, "gc start\n"); } void pblk_gc_should_start(struct pblk *pblk) @@ -605,14 +605,14 @@ int pblk_gc_init(struct pblk *pblk) gc->gc_ts = kthread_create(pblk_gc_ts, pblk, "pblk-gc-ts"); if (IS_ERR(gc->gc_ts)) { - pr_err("pblk: could not allocate GC main kthread\n"); + pblk_err(pblk, "could not allocate GC main kthread\n"); return PTR_ERR(gc->gc_ts); } gc->gc_writer_ts = kthread_create(pblk_gc_writer_ts, pblk, "pblk-gc-writer-ts"); if (IS_ERR(gc->gc_writer_ts)) { - pr_err("pblk: could not allocate GC writer kthread\n"); + pblk_err(pblk, "could not allocate GC writer kthread\n"); ret = PTR_ERR(gc->gc_writer_ts); goto fail_free_main_kthread; } @@ -620,7 +620,7 @@ int pblk_gc_init(struct pblk *pblk) gc->gc_reader_ts = kthread_create(pblk_gc_reader_ts, pblk, "pblk-gc-reader-ts"); if (IS_ERR(gc->gc_reader_ts)) { - pr_err("pblk: could not allocate GC reader kthread\n"); + pblk_err(pblk, "could not allocate GC reader kthread\n"); ret = PTR_ERR(gc->gc_reader_ts); goto fail_free_writer_kthread; } @@ -641,7 +641,7 @@ int pblk_gc_init(struct pblk *pblk) gc->gc_line_reader_wq = alloc_workqueue("pblk-gc-line-reader-wq", WQ_MEM_RECLAIM | WQ_UNBOUND, PBLK_GC_MAX_READERS); if (!gc->gc_line_reader_wq) { - pr_err("pblk: could not allocate GC line reader workqueue\n"); + pblk_err(pblk, "could not allocate GC line reader workqueue\n"); ret = -ENOMEM; goto fail_free_reader_kthread; } @@ -650,7 +650,7 @@ int pblk_gc_init(struct pblk *pblk) gc->gc_reader_wq = alloc_workqueue("pblk-gc-line_wq", WQ_MEM_RECLAIM | WQ_UNBOUND, 1); if (!gc->gc_reader_wq) { - pr_err("pblk: could not allocate GC reader workqueue\n"); + pblk_err(pblk, "could not allocate GC reader workqueue\n"); ret = -ENOMEM; goto fail_free_reader_line_wq; } diff --git a/drivers/lightnvm/pblk-init.c b/drivers/lightnvm/pblk-init.c index b57f764d6a16..537e98f2b24a 100644 --- a/drivers/lightnvm/pblk-init.c +++ b/drivers/lightnvm/pblk-init.c @@ -91,7 +91,7 @@ static size_t pblk_trans_map_size(struct pblk *pblk) return entry_size * pblk->rl.nr_secs; } -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG static u32 pblk_l2p_crc(struct pblk *pblk) { size_t map_size; @@ -117,13 +117,13 @@ static int pblk_l2p_recover(struct pblk *pblk, bool factory_init) } else { line = pblk_recov_l2p(pblk); if (IS_ERR(line)) { - pr_err("pblk: could not recover l2p table\n"); + pblk_err(pblk, "could not recover l2p table\n"); return -EFAULT; } } -#ifdef CONFIG_NVM_DEBUG - pr_info("pblk init: L2P CRC: %x\n", pblk_l2p_crc(pblk)); +#ifdef CONFIG_NVM_PBLK_DEBUG + pblk_info(pblk, "init: L2P CRC: %x\n", pblk_l2p_crc(pblk)); #endif /* Free full lines directly as GC has not been started yet */ @@ -166,7 +166,7 @@ static int pblk_l2p_init(struct pblk *pblk, bool factory_init) static void pblk_rwb_free(struct pblk *pblk) { if (pblk_rb_tear_down_check(&pblk->rwb)) - pr_err("pblk: write buffer error on tear down\n"); + pblk_err(pblk, "write buffer error on tear down\n"); pblk_rb_data_free(&pblk->rwb); vfree(pblk_rb_entries_ref(&pblk->rwb)); @@ -179,11 +179,14 @@ static int pblk_rwb_init(struct pblk *pblk) struct pblk_rb_entry *entries; unsigned long nr_entries, buffer_size; unsigned int power_size, power_seg_sz; + int pgs_in_buffer; - if (write_buffer_size && (write_buffer_size > pblk->pgs_in_buffer)) + pgs_in_buffer = max(geo->mw_cunits, geo->ws_opt) * geo->all_luns; + + if (write_buffer_size && (write_buffer_size > pgs_in_buffer)) buffer_size = write_buffer_size; else - buffer_size = pblk->pgs_in_buffer; + buffer_size = pgs_in_buffer; nr_entries = pblk_rb_calculate_size(buffer_size); @@ -200,7 +203,8 @@ static int pblk_rwb_init(struct pblk *pblk) /* Minimum pages needed within a lun */ #define ADDR_POOL_SIZE 64 -static int pblk_set_addrf_12(struct nvm_geo *geo, struct nvm_addrf_12 *dst) +static int pblk_set_addrf_12(struct pblk *pblk, struct nvm_geo *geo, + struct nvm_addrf_12 *dst) { struct nvm_addrf_12 *src = (struct nvm_addrf_12 *)&geo->addrf; int power_len; @@ -208,14 +212,14 @@ static int pblk_set_addrf_12(struct nvm_geo *geo, struct nvm_addrf_12 *dst) /* Re-calculate channel and lun format to adapt to configuration */ power_len = get_count_order(geo->num_ch); if (1 << power_len != geo->num_ch) { - pr_err("pblk: supports only power-of-two channel config.\n"); + pblk_err(pblk, "supports only power-of-two channel config.\n"); return -EINVAL; } dst->ch_len = power_len; power_len = get_count_order(geo->num_lun); if (1 << power_len != geo->num_lun) { - pr_err("pblk: supports only power-of-two LUN config.\n"); + pblk_err(pblk, "supports only power-of-two LUN config.\n"); return -EINVAL; } dst->lun_len = power_len; @@ -282,18 +286,19 @@ static int pblk_set_addrf(struct pblk *pblk) case NVM_OCSSD_SPEC_12: div_u64_rem(geo->clba, pblk->min_write_pgs, &mod); if (mod) { - pr_err("pblk: bad configuration of sectors/pages\n"); + pblk_err(pblk, "bad configuration of sectors/pages\n"); return -EINVAL; } - pblk->addrf_len = pblk_set_addrf_12(geo, (void *)&pblk->addrf); + pblk->addrf_len = pblk_set_addrf_12(pblk, geo, + (void *)&pblk->addrf); break; case NVM_OCSSD_SPEC_20: pblk->addrf_len = pblk_set_addrf_20(geo, (void *)&pblk->addrf, - &pblk->uaddrf); + &pblk->uaddrf); break; default: - pr_err("pblk: OCSSD revision not supported (%d)\n", + pblk_err(pblk, "OCSSD revision not supported (%d)\n", geo->version); return -EINVAL; } @@ -366,15 +371,13 @@ static int pblk_core_init(struct pblk *pblk) atomic64_set(&pblk->nr_flush, 0); pblk->nr_flush_rst = 0; - pblk->pgs_in_buffer = geo->mw_cunits * geo->all_luns; - pblk->min_write_pgs = geo->ws_opt * (geo->csecs / PAGE_SIZE); max_write_ppas = pblk->min_write_pgs * geo->all_luns; pblk->max_write_pgs = min_t(int, max_write_ppas, NVM_MAX_VLBA); pblk_set_sec_per_write(pblk, pblk->min_write_pgs); if (pblk->max_write_pgs > PBLK_MAX_REQ_ADDRS) { - pr_err("pblk: vector list too big(%u > %u)\n", + pblk_err(pblk, "vector list too big(%u > %u)\n", pblk->max_write_pgs, PBLK_MAX_REQ_ADDRS); return -EINVAL; } @@ -607,7 +610,7 @@ static int pblk_luns_init(struct pblk *pblk) /* TODO: Implement unbalanced LUN support */ if (geo->num_lun < 0) { - pr_err("pblk: unbalanced LUN config.\n"); + pblk_err(pblk, "unbalanced LUN config.\n"); return -EINVAL; } @@ -716,10 +719,11 @@ static int pblk_setup_line_meta_12(struct pblk *pblk, struct pblk_line *line, /* * In 1.2 spec. chunk state is not persisted by the device. Thus - * some of the values are reset each time pblk is instantiated. + * some of the values are reset each time pblk is instantiated, + * so we have to assume that the block is closed. */ if (lun_bb_meta[line->id] == NVM_BLK_T_FREE) - chunk->state = NVM_CHK_ST_FREE; + chunk->state = NVM_CHK_ST_CLOSED; else chunk->state = NVM_CHK_ST_OFFLINE; @@ -1026,7 +1030,7 @@ add_emeta_page: lm->emeta_sec[0], geo->clba); if (lm->min_blk_line > lm->blk_per_line) { - pr_err("pblk: config. not supported. Min. LUN in line:%d\n", + pblk_err(pblk, "config. not supported. Min. LUN in line:%d\n", lm->blk_per_line); return -EINVAL; } @@ -1078,7 +1082,7 @@ static int pblk_lines_init(struct pblk *pblk) } if (!nr_free_chks) { - pr_err("pblk: too many bad blocks prevent for sane instance\n"); + pblk_err(pblk, "too many bad blocks prevent for sane instance\n"); return -EINTR; } @@ -1108,7 +1112,7 @@ static int pblk_writer_init(struct pblk *pblk) int err = PTR_ERR(pblk->writer_ts); if (err != -EINTR) - pr_err("pblk: could not allocate writer kthread (%d)\n", + pblk_err(pblk, "could not allocate writer kthread (%d)\n", err); return err; } @@ -1154,7 +1158,7 @@ static void pblk_tear_down(struct pblk *pblk, bool graceful) pblk_rb_sync_l2p(&pblk->rwb); pblk_rl_free(&pblk->rl); - pr_debug("pblk: consistent tear down (graceful:%d)\n", graceful); + pblk_debug(pblk, "consistent tear down (graceful:%d)\n", graceful); } static void pblk_exit(void *private, bool graceful) @@ -1165,8 +1169,8 @@ static void pblk_exit(void *private, bool graceful) pblk_gc_exit(pblk, graceful); pblk_tear_down(pblk, graceful); -#ifdef CONFIG_NVM_DEBUG - pr_info("pblk exit: L2P CRC: %x\n", pblk_l2p_crc(pblk)); +#ifdef CONFIG_NVM_PBLK_DEBUG + pblk_info(pblk, "exit: L2P CRC: %x\n", pblk_l2p_crc(pblk)); #endif pblk_free(pblk); @@ -1189,34 +1193,35 @@ static void *pblk_init(struct nvm_tgt_dev *dev, struct gendisk *tdisk, struct pblk *pblk; int ret; - /* pblk supports 1.2 and 2.0 versions */ + pblk = kzalloc(sizeof(struct pblk), GFP_KERNEL); + if (!pblk) + return ERR_PTR(-ENOMEM); + + pblk->dev = dev; + pblk->disk = tdisk; + pblk->state = PBLK_STATE_RUNNING; + pblk->gc.gc_enabled = 0; + if (!(geo->version == NVM_OCSSD_SPEC_12 || geo->version == NVM_OCSSD_SPEC_20)) { - pr_err("pblk: OCSSD version not supported (%u)\n", + pblk_err(pblk, "OCSSD version not supported (%u)\n", geo->version); + kfree(pblk); return ERR_PTR(-EINVAL); } if (geo->version == NVM_OCSSD_SPEC_12 && geo->dom & NVM_RSP_L2P) { - pr_err("pblk: host-side L2P table not supported. (%x)\n", + pblk_err(pblk, "host-side L2P table not supported. (%x)\n", geo->dom); + kfree(pblk); return ERR_PTR(-EINVAL); } - pblk = kzalloc(sizeof(struct pblk), GFP_KERNEL); - if (!pblk) - return ERR_PTR(-ENOMEM); - - pblk->dev = dev; - pblk->disk = tdisk; - pblk->state = PBLK_STATE_RUNNING; - pblk->gc.gc_enabled = 0; - spin_lock_init(&pblk->resubmit_lock); spin_lock_init(&pblk->trans_lock); spin_lock_init(&pblk->lock); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_long_set(&pblk->inflight_writes, 0); atomic_long_set(&pblk->padded_writes, 0); atomic_long_set(&pblk->padded_wb, 0); @@ -1241,38 +1246,38 @@ static void *pblk_init(struct nvm_tgt_dev *dev, struct gendisk *tdisk, ret = pblk_core_init(pblk); if (ret) { - pr_err("pblk: could not initialize core\n"); + pblk_err(pblk, "could not initialize core\n"); goto fail; } ret = pblk_lines_init(pblk); if (ret) { - pr_err("pblk: could not initialize lines\n"); + pblk_err(pblk, "could not initialize lines\n"); goto fail_free_core; } ret = pblk_rwb_init(pblk); if (ret) { - pr_err("pblk: could not initialize write buffer\n"); + pblk_err(pblk, "could not initialize write buffer\n"); goto fail_free_lines; } ret = pblk_l2p_init(pblk, flags & NVM_TARGET_FACTORY); if (ret) { - pr_err("pblk: could not initialize maps\n"); + pblk_err(pblk, "could not initialize maps\n"); goto fail_free_rwb; } ret = pblk_writer_init(pblk); if (ret) { if (ret != -EINTR) - pr_err("pblk: could not initialize write thread\n"); + pblk_err(pblk, "could not initialize write thread\n"); goto fail_free_l2p; } ret = pblk_gc_init(pblk); if (ret) { - pr_err("pblk: could not initialize gc\n"); + pblk_err(pblk, "could not initialize gc\n"); goto fail_stop_writer; } @@ -1287,8 +1292,7 @@ static void *pblk_init(struct nvm_tgt_dev *dev, struct gendisk *tdisk, blk_queue_max_discard_sectors(tqueue, UINT_MAX >> 9); blk_queue_flag_set(QUEUE_FLAG_DISCARD, tqueue); - pr_info("pblk(%s): luns:%u, lines:%d, secs:%llu, buf entries:%u\n", - tdisk->disk_name, + pblk_info(pblk, "luns:%u, lines:%d, secs:%llu, buf entries:%u\n", geo->all_luns, pblk->l_mg.nr_lines, (unsigned long long)pblk->rl.nr_secs, pblk->rwb.nr_entries); diff --git a/drivers/lightnvm/pblk-rb.c b/drivers/lightnvm/pblk-rb.c index 55e9442a99e2..f6eec0212dfc 100644 --- a/drivers/lightnvm/pblk-rb.c +++ b/drivers/lightnvm/pblk-rb.c @@ -111,7 +111,7 @@ int pblk_rb_init(struct pblk_rb *rb, struct pblk_rb_entry *rb_entry_base, } while (iter > 0); up_write(&pblk_rb_lock); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_set(&rb->inflight_flush_point, 0); #endif @@ -308,7 +308,7 @@ void pblk_rb_write_entry_user(struct pblk_rb *rb, void *data, entry = &rb->entries[ring_pos]; flags = READ_ONCE(entry->w_ctx.flags); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG /* Caller must guarantee that the entry is free */ BUG_ON(!(flags & PBLK_WRITABLE_ENTRY)); #endif @@ -332,7 +332,7 @@ void pblk_rb_write_entry_gc(struct pblk_rb *rb, void *data, entry = &rb->entries[ring_pos]; flags = READ_ONCE(entry->w_ctx.flags); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG /* Caller must guarantee that the entry is free */ BUG_ON(!(flags & PBLK_WRITABLE_ENTRY)); #endif @@ -362,7 +362,7 @@ static int pblk_rb_flush_point_set(struct pblk_rb *rb, struct bio *bio, return 0; } -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_inc(&rb->inflight_flush_point); #endif @@ -547,7 +547,7 @@ try: page = virt_to_page(entry->data); if (!page) { - pr_err("pblk: could not allocate write bio page\n"); + pblk_err(pblk, "could not allocate write bio page\n"); flags &= ~PBLK_WRITTEN_DATA; flags |= PBLK_SUBMITTED_ENTRY; /* Release flags on context. Protect from writes */ @@ -557,7 +557,7 @@ try: if (bio_add_pc_page(q, bio, page, rb->seg_size, 0) != rb->seg_size) { - pr_err("pblk: could not add page to write bio\n"); + pblk_err(pblk, "could not add page to write bio\n"); flags &= ~PBLK_WRITTEN_DATA; flags |= PBLK_SUBMITTED_ENTRY; /* Release flags on context. Protect from writes */ @@ -576,19 +576,19 @@ try: if (pad) { if (pblk_bio_add_pages(pblk, bio, GFP_KERNEL, pad)) { - pr_err("pblk: could not pad page in write bio\n"); + pblk_err(pblk, "could not pad page in write bio\n"); return NVM_IO_ERR; } if (pad < pblk->min_write_pgs) atomic64_inc(&pblk->pad_dist[pad - 1]); else - pr_warn("pblk: padding more than min. sectors\n"); + pblk_warn(pblk, "padding more than min. sectors\n"); atomic64_add(pad, &pblk->pad_wa); } -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_long_add(pad, &pblk->padded_writes); #endif @@ -613,7 +613,7 @@ int pblk_rb_copy_to_bio(struct pblk_rb *rb, struct bio *bio, sector_t lba, int ret = 1; -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG /* Caller must ensure that the access will not cause an overflow */ BUG_ON(pos >= rb->nr_entries); #endif @@ -820,7 +820,7 @@ ssize_t pblk_rb_sysfs(struct pblk_rb *rb, char *buf) rb->subm, rb->sync, rb->l2p_update, -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_read(&rb->inflight_flush_point), #else 0, @@ -838,7 +838,7 @@ ssize_t pblk_rb_sysfs(struct pblk_rb *rb, char *buf) rb->subm, rb->sync, rb->l2p_update, -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_read(&rb->inflight_flush_point), #else 0, diff --git a/drivers/lightnvm/pblk-read.c b/drivers/lightnvm/pblk-read.c index 18694694e5f0..5a46d7f9302f 100644 --- a/drivers/lightnvm/pblk-read.c +++ b/drivers/lightnvm/pblk-read.c @@ -28,7 +28,7 @@ static int pblk_read_from_cache(struct pblk *pblk, struct bio *bio, sector_t lba, struct ppa_addr ppa, int bio_iter, bool advanced_bio) { -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG /* Callers must ensure that the ppa points to a cache address */ BUG_ON(pblk_ppa_empty(ppa)); BUG_ON(!pblk_addr_in_cache(ppa)); @@ -79,7 +79,7 @@ retry: WARN_ON(test_and_set_bit(i, read_bitmap)); meta_list[i].lba = cpu_to_le64(lba); advanced_bio = true; -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_long_inc(&pblk->cache_reads); #endif } else { @@ -97,7 +97,7 @@ next: else rqd->flags = pblk_set_read_mode(pblk, PBLK_READ_RANDOM); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_long_add(nr_secs, &pblk->inflight_reads); #endif } @@ -117,13 +117,13 @@ static void pblk_read_check_seq(struct pblk *pblk, struct nvm_rq *rqd, continue; if (lba != blba + i) { -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG struct ppa_addr *p; p = (nr_lbas == 1) ? &rqd->ppa_list[i] : &rqd->ppa_addr; - print_ppa(&pblk->dev->geo, p, "seq", i); + print_ppa(pblk, p, "seq", i); #endif - pr_err("pblk: corrupted read LBA (%llu/%llu)\n", + pblk_err(pblk, "corrupted read LBA (%llu/%llu)\n", lba, (u64)blba + i); WARN_ON(1); } @@ -149,14 +149,14 @@ static void pblk_read_check_rand(struct pblk *pblk, struct nvm_rq *rqd, meta_lba = le64_to_cpu(meta_lba_list[j].lba); if (lba != meta_lba) { -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG struct ppa_addr *p; int nr_ppas = rqd->nr_ppas; p = (nr_ppas == 1) ? &rqd->ppa_list[j] : &rqd->ppa_addr; - print_ppa(&pblk->dev->geo, p, "seq", j); + print_ppa(pblk, p, "seq", j); #endif - pr_err("pblk: corrupted read LBA (%llu/%llu)\n", + pblk_err(pblk, "corrupted read LBA (%llu/%llu)\n", lba, meta_lba); WARN_ON(1); } @@ -185,7 +185,7 @@ static void pblk_read_put_rqd_kref(struct pblk *pblk, struct nvm_rq *rqd) static void pblk_end_user_read(struct bio *bio) { -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG WARN_ONCE(bio->bi_status, "pblk: corrupted read bio\n"); #endif bio_endio(bio); @@ -199,7 +199,7 @@ static void __pblk_end_io_read(struct pblk *pblk, struct nvm_rq *rqd, struct bio *int_bio = rqd->bio; unsigned long start_time = r_ctx->start_time; - generic_end_io_acct(dev->q, READ, &pblk->disk->part0, start_time); + generic_end_io_acct(dev->q, REQ_OP_READ, &pblk->disk->part0, start_time); if (rqd->error) pblk_log_read_err(pblk, rqd); @@ -212,7 +212,7 @@ static void __pblk_end_io_read(struct pblk *pblk, struct nvm_rq *rqd, if (put_line) pblk_read_put_rqd_kref(pblk, rqd); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_long_add(rqd->nr_ppas, &pblk->sync_reads); atomic_long_sub(rqd->nr_ppas, &pblk->inflight_reads); #endif @@ -231,74 +231,36 @@ static void pblk_end_io_read(struct nvm_rq *rqd) __pblk_end_io_read(pblk, rqd, true); } -static int pblk_partial_read(struct pblk *pblk, struct nvm_rq *rqd, - struct bio *orig_bio, unsigned int bio_init_idx, - unsigned long *read_bitmap) +static void pblk_end_partial_read(struct nvm_rq *rqd) { - struct pblk_sec_meta *meta_list = rqd->meta_list; - struct bio *new_bio; + struct pblk *pblk = rqd->private; + struct pblk_g_ctx *r_ctx = nvm_rq_to_pdu(rqd); + struct pblk_pr_ctx *pr_ctx = r_ctx->private; + struct bio *new_bio = rqd->bio; + struct bio *bio = pr_ctx->orig_bio; struct bio_vec src_bv, dst_bv; - void *ppa_ptr = NULL; - void *src_p, *dst_p; - dma_addr_t dma_ppa_list = 0; - __le64 *lba_list_mem, *lba_list_media; - int nr_secs = rqd->nr_ppas; + struct pblk_sec_meta *meta_list = rqd->meta_list; + int bio_init_idx = pr_ctx->bio_init_idx; + unsigned long *read_bitmap = pr_ctx->bitmap; + int nr_secs = pr_ctx->orig_nr_secs; int nr_holes = nr_secs - bitmap_weight(read_bitmap, nr_secs); - int i, ret, hole; - - /* Re-use allocated memory for intermediate lbas */ - lba_list_mem = (((void *)rqd->ppa_list) + pblk_dma_ppa_size); - lba_list_media = (((void *)rqd->ppa_list) + 2 * pblk_dma_ppa_size); - - new_bio = bio_alloc(GFP_KERNEL, nr_holes); - - if (pblk_bio_add_pages(pblk, new_bio, GFP_KERNEL, nr_holes)) - goto fail_add_pages; - - if (nr_holes != new_bio->bi_vcnt) { - pr_err("pblk: malformed bio\n"); - goto fail; - } - - for (i = 0; i < nr_secs; i++) - lba_list_mem[i] = meta_list[i].lba; - - new_bio->bi_iter.bi_sector = 0; /* internal bio */ - bio_set_op_attrs(new_bio, REQ_OP_READ, 0); - - rqd->bio = new_bio; - rqd->nr_ppas = nr_holes; - rqd->flags = pblk_set_read_mode(pblk, PBLK_READ_RANDOM); - - if (unlikely(nr_holes == 1)) { - ppa_ptr = rqd->ppa_list; - dma_ppa_list = rqd->dma_ppa_list; - rqd->ppa_addr = rqd->ppa_list[0]; - } - - ret = pblk_submit_io_sync(pblk, rqd); - if (ret) { - bio_put(rqd->bio); - pr_err("pblk: sync read IO submission failed\n"); - goto fail; - } - - if (rqd->error) { - atomic_long_inc(&pblk->read_failed); -#ifdef CONFIG_NVM_DEBUG - pblk_print_failed_rqd(pblk, rqd, rqd->error); -#endif - } + __le64 *lba_list_mem, *lba_list_media; + void *src_p, *dst_p; + int hole, i; if (unlikely(nr_holes == 1)) { struct ppa_addr ppa; ppa = rqd->ppa_addr; - rqd->ppa_list = ppa_ptr; - rqd->dma_ppa_list = dma_ppa_list; + rqd->ppa_list = pr_ctx->ppa_ptr; + rqd->dma_ppa_list = pr_ctx->dma_ppa_list; rqd->ppa_list[0] = ppa; } + /* Re-use allocated memory for intermediate lbas */ + lba_list_mem = (((void *)rqd->ppa_list) + pblk_dma_ppa_size); + lba_list_media = (((void *)rqd->ppa_list) + 2 * pblk_dma_ppa_size); + for (i = 0; i < nr_secs; i++) { lba_list_media[i] = meta_list[i].lba; meta_list[i].lba = lba_list_mem[i]; @@ -316,7 +278,7 @@ static int pblk_partial_read(struct pblk *pblk, struct nvm_rq *rqd, meta_list[hole].lba = lba_list_media[i]; src_bv = new_bio->bi_io_vec[i++]; - dst_bv = orig_bio->bi_io_vec[bio_init_idx + hole]; + dst_bv = bio->bi_io_vec[bio_init_idx + hole]; src_p = kmap_atomic(src_bv.bv_page); dst_p = kmap_atomic(dst_bv.bv_page); @@ -334,19 +296,107 @@ static int pblk_partial_read(struct pblk *pblk, struct nvm_rq *rqd, } while (hole < nr_secs); bio_put(new_bio); + kfree(pr_ctx); /* restore original request */ rqd->bio = NULL; rqd->nr_ppas = nr_secs; + bio_endio(bio); __pblk_end_io_read(pblk, rqd, false); - return NVM_IO_DONE; +} -fail: - /* Free allocated pages in new bio */ +static int pblk_setup_partial_read(struct pblk *pblk, struct nvm_rq *rqd, + unsigned int bio_init_idx, + unsigned long *read_bitmap, + int nr_holes) +{ + struct pblk_sec_meta *meta_list = rqd->meta_list; + struct pblk_g_ctx *r_ctx = nvm_rq_to_pdu(rqd); + struct pblk_pr_ctx *pr_ctx; + struct bio *new_bio, *bio = r_ctx->private; + __le64 *lba_list_mem; + int nr_secs = rqd->nr_ppas; + int i; + + /* Re-use allocated memory for intermediate lbas */ + lba_list_mem = (((void *)rqd->ppa_list) + pblk_dma_ppa_size); + + new_bio = bio_alloc(GFP_KERNEL, nr_holes); + + if (pblk_bio_add_pages(pblk, new_bio, GFP_KERNEL, nr_holes)) + goto fail_bio_put; + + if (nr_holes != new_bio->bi_vcnt) { + WARN_ONCE(1, "pblk: malformed bio\n"); + goto fail_free_pages; + } + + pr_ctx = kmalloc(sizeof(struct pblk_pr_ctx), GFP_KERNEL); + if (!pr_ctx) + goto fail_free_pages; + + for (i = 0; i < nr_secs; i++) + lba_list_mem[i] = meta_list[i].lba; + + new_bio->bi_iter.bi_sector = 0; /* internal bio */ + bio_set_op_attrs(new_bio, REQ_OP_READ, 0); + + rqd->bio = new_bio; + rqd->nr_ppas = nr_holes; + rqd->flags = pblk_set_read_mode(pblk, PBLK_READ_RANDOM); + + pr_ctx->ppa_ptr = NULL; + pr_ctx->orig_bio = bio; + bitmap_copy(pr_ctx->bitmap, read_bitmap, NVM_MAX_VLBA); + pr_ctx->bio_init_idx = bio_init_idx; + pr_ctx->orig_nr_secs = nr_secs; + r_ctx->private = pr_ctx; + + if (unlikely(nr_holes == 1)) { + pr_ctx->ppa_ptr = rqd->ppa_list; + pr_ctx->dma_ppa_list = rqd->dma_ppa_list; + rqd->ppa_addr = rqd->ppa_list[0]; + } + return 0; + +fail_free_pages: pblk_bio_free_pages(pblk, new_bio, 0, new_bio->bi_vcnt); -fail_add_pages: - pr_err("pblk: failed to perform partial read\n"); +fail_bio_put: + bio_put(new_bio); + + return -ENOMEM; +} + +static int pblk_partial_read_bio(struct pblk *pblk, struct nvm_rq *rqd, + unsigned int bio_init_idx, + unsigned long *read_bitmap, int nr_secs) +{ + int nr_holes; + int ret; + + nr_holes = nr_secs - bitmap_weight(read_bitmap, nr_secs); + + if (pblk_setup_partial_read(pblk, rqd, bio_init_idx, read_bitmap, + nr_holes)) + return NVM_IO_ERR; + + rqd->end_io = pblk_end_partial_read; + + ret = pblk_submit_io(pblk, rqd); + if (ret) { + bio_put(rqd->bio); + pblk_err(pblk, "partial read IO submission failed\n"); + goto err; + } + + return NVM_IO_OK; + +err: + pblk_err(pblk, "failed to perform partial read\n"); + + /* Free allocated pages in new bio */ + pblk_bio_free_pages(pblk, rqd->bio, 0, rqd->bio->bi_vcnt); __pblk_end_io_read(pblk, rqd, false); return NVM_IO_ERR; } @@ -359,7 +409,7 @@ static void pblk_read_rq(struct pblk *pblk, struct nvm_rq *rqd, struct bio *bio, pblk_lookup_l2p_seq(pblk, &ppa, lba, 1); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_long_inc(&pblk->inflight_reads); #endif @@ -382,7 +432,7 @@ retry: WARN_ON(test_and_set_bit(0, read_bitmap)); meta_list[0].lba = cpu_to_le64(lba); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_long_inc(&pblk->cache_reads); #endif } else { @@ -401,7 +451,7 @@ int pblk_submit_read(struct pblk *pblk, struct bio *bio) struct pblk_g_ctx *r_ctx; struct nvm_rq *rqd; unsigned int bio_init_idx; - unsigned long read_bitmap; /* Max 64 ppas per request */ + DECLARE_BITMAP(read_bitmap, NVM_MAX_VLBA); int ret = NVM_IO_ERR; /* logic error: lba out-of-bounds. Ignore read request */ @@ -411,9 +461,10 @@ int pblk_submit_read(struct pblk *pblk, struct bio *bio) return NVM_IO_ERR; } - generic_start_io_acct(q, READ, bio_sectors(bio), &pblk->disk->part0); + generic_start_io_acct(q, REQ_OP_READ, bio_sectors(bio), + &pblk->disk->part0); - bitmap_zero(&read_bitmap, nr_secs); + bitmap_zero(read_bitmap, nr_secs); rqd = pblk_alloc_rqd(pblk, PBLK_READ); @@ -436,7 +487,7 @@ int pblk_submit_read(struct pblk *pblk, struct bio *bio) rqd->meta_list = nvm_dev_dma_alloc(dev->parent, GFP_KERNEL, &rqd->dma_meta_list); if (!rqd->meta_list) { - pr_err("pblk: not able to allocate ppa list\n"); + pblk_err(pblk, "not able to allocate ppa list\n"); goto fail_rqd_free; } @@ -444,32 +495,32 @@ int pblk_submit_read(struct pblk *pblk, struct bio *bio) rqd->ppa_list = rqd->meta_list + pblk_dma_meta_size; rqd->dma_ppa_list = rqd->dma_meta_list + pblk_dma_meta_size; - pblk_read_ppalist_rq(pblk, rqd, bio, blba, &read_bitmap); + pblk_read_ppalist_rq(pblk, rqd, bio, blba, read_bitmap); } else { - pblk_read_rq(pblk, rqd, bio, blba, &read_bitmap); + pblk_read_rq(pblk, rqd, bio, blba, read_bitmap); } - if (bitmap_full(&read_bitmap, nr_secs)) { + if (bitmap_full(read_bitmap, nr_secs)) { atomic_inc(&pblk->inflight_io); __pblk_end_io_read(pblk, rqd, false); return NVM_IO_DONE; } /* All sectors are to be read from the device */ - if (bitmap_empty(&read_bitmap, rqd->nr_ppas)) { + if (bitmap_empty(read_bitmap, rqd->nr_ppas)) { struct bio *int_bio = NULL; /* Clone read bio to deal with read errors internally */ int_bio = bio_clone_fast(bio, GFP_KERNEL, &pblk_bio_set); if (!int_bio) { - pr_err("pblk: could not clone read bio\n"); + pblk_err(pblk, "could not clone read bio\n"); goto fail_end_io; } rqd->bio = int_bio; if (pblk_submit_io(pblk, rqd)) { - pr_err("pblk: read IO submission failed\n"); + pblk_err(pblk, "read IO submission failed\n"); ret = NVM_IO_ERR; goto fail_end_io; } @@ -480,8 +531,15 @@ int pblk_submit_read(struct pblk *pblk, struct bio *bio) /* The read bio request could be partially filled by the write buffer, * but there are some holes that need to be read from the drive. */ - return pblk_partial_read(pblk, rqd, bio, bio_init_idx, &read_bitmap); + ret = pblk_partial_read_bio(pblk, rqd, bio_init_idx, read_bitmap, + nr_secs); + if (ret) + goto fail_meta_free; + + return NVM_IO_OK; +fail_meta_free: + nvm_dev_dma_free(dev->parent, rqd->meta_list, rqd->dma_meta_list); fail_rqd_free: pblk_free_rqd(pblk, rqd, PBLK_READ); return ret; @@ -514,7 +572,7 @@ static int read_ppalist_rq_gc(struct pblk *pblk, struct nvm_rq *rqd, rqd->ppa_list[valid_secs++] = ppa_list_l2p[i]; } -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_long_add(valid_secs, &pblk->inflight_reads); #endif @@ -548,7 +606,7 @@ static int read_rq_gc(struct pblk *pblk, struct nvm_rq *rqd, rqd->ppa_addr = ppa_l2p; valid_secs = 1; -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_long_inc(&pblk->inflight_reads); #endif @@ -595,7 +653,8 @@ int pblk_submit_read_gc(struct pblk *pblk, struct pblk_gc_rq *gc_rq) bio = pblk_bio_map_addr(pblk, gc_rq->data, gc_rq->secs_to_gc, data_len, PBLK_VMALLOC_META, GFP_KERNEL); if (IS_ERR(bio)) { - pr_err("pblk: could not allocate GC bio (%lu)\n", PTR_ERR(bio)); + pblk_err(pblk, "could not allocate GC bio (%lu)\n", + PTR_ERR(bio)); goto err_free_dma; } @@ -609,7 +668,7 @@ int pblk_submit_read_gc(struct pblk *pblk, struct pblk_gc_rq *gc_rq) if (pblk_submit_io_sync(pblk, &rqd)) { ret = -EIO; - pr_err("pblk: GC read request failed\n"); + pblk_err(pblk, "GC read request failed\n"); goto err_free_bio; } @@ -619,12 +678,12 @@ int pblk_submit_read_gc(struct pblk *pblk, struct pblk_gc_rq *gc_rq) if (rqd.error) { atomic_long_inc(&pblk->read_failed_gc); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG pblk_print_failed_rqd(pblk, &rqd, rqd.error); #endif } -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_long_add(gc_rq->secs_to_gc, &pblk->sync_reads); atomic_long_add(gc_rq->secs_to_gc, &pblk->recov_gc_reads); atomic_long_sub(gc_rq->secs_to_gc, &pblk->inflight_reads); diff --git a/drivers/lightnvm/pblk-recovery.c b/drivers/lightnvm/pblk-recovery.c index 3a5069183859..e232e47e1353 100644 --- a/drivers/lightnvm/pblk-recovery.c +++ b/drivers/lightnvm/pblk-recovery.c @@ -77,7 +77,7 @@ static int pblk_recov_l2p_from_emeta(struct pblk *pblk, struct pblk_line *line) } if (nr_valid_lbas != nr_lbas) - pr_err("pblk: line %d - inconsistent lba list(%llu/%llu)\n", + pblk_err(pblk, "line %d - inconsistent lba list(%llu/%llu)\n", line->id, nr_valid_lbas, nr_lbas); line->left_msecs = 0; @@ -184,7 +184,7 @@ next_read_rq: /* If read fails, more padding is needed */ ret = pblk_submit_io_sync(pblk, rqd); if (ret) { - pr_err("pblk: I/O submission failed: %d\n", ret); + pblk_err(pblk, "I/O submission failed: %d\n", ret); return ret; } @@ -194,7 +194,7 @@ next_read_rq: * we cannot recover from here. Need FTL log. */ if (rqd->error && rqd->error != NVM_RSP_WARN_HIGHECC) { - pr_err("pblk: L2P recovery failed (%d)\n", rqd->error); + pblk_err(pblk, "L2P recovery failed (%d)\n", rqd->error); return -EINTR; } @@ -273,7 +273,7 @@ static int pblk_recov_pad_oob(struct pblk *pblk, struct pblk_line *line, next_pad_rq: rq_ppas = pblk_calc_secs(pblk, left_ppas, 0); if (rq_ppas < pblk->min_write_pgs) { - pr_err("pblk: corrupted pad line %d\n", line->id); + pblk_err(pblk, "corrupted pad line %d\n", line->id); goto fail_free_pad; } @@ -342,7 +342,7 @@ next_pad_rq: ret = pblk_submit_io(pblk, rqd); if (ret) { - pr_err("pblk: I/O submission failed: %d\n", ret); + pblk_err(pblk, "I/O submission failed: %d\n", ret); pblk_up_page(pblk, rqd->ppa_list, rqd->nr_ppas); goto fail_free_bio; } @@ -356,12 +356,12 @@ next_pad_rq: if (!wait_for_completion_io_timeout(&pad_rq->wait, msecs_to_jiffies(PBLK_COMMAND_TIMEOUT_MS))) { - pr_err("pblk: pad write timed out\n"); + pblk_err(pblk, "pad write timed out\n"); ret = -ETIME; } if (!pblk_line_is_full(line)) - pr_err("pblk: corrupted padded line: %d\n", line->id); + pblk_err(pblk, "corrupted padded line: %d\n", line->id); vfree(data); free_rq: @@ -461,7 +461,7 @@ next_rq: ret = pblk_submit_io_sync(pblk, rqd); if (ret) { - pr_err("pblk: I/O submission failed: %d\n", ret); + pblk_err(pblk, "I/O submission failed: %d\n", ret); return ret; } @@ -501,11 +501,11 @@ next_rq: ret = pblk_recov_pad_oob(pblk, line, pad_secs); if (ret) - pr_err("pblk: OOB padding failed (err:%d)\n", ret); + pblk_err(pblk, "OOB padding failed (err:%d)\n", ret); ret = pblk_recov_read_oob(pblk, line, p, r_ptr); if (ret) - pr_err("pblk: OOB read failed (err:%d)\n", ret); + pblk_err(pblk, "OOB read failed (err:%d)\n", ret); left_ppas = 0; } @@ -592,7 +592,7 @@ next_rq: ret = pblk_submit_io_sync(pblk, rqd); if (ret) { - pr_err("pblk: I/O submission failed: %d\n", ret); + pblk_err(pblk, "I/O submission failed: %d\n", ret); bio_put(bio); return ret; } @@ -671,14 +671,14 @@ static int pblk_recov_l2p_from_oob(struct pblk *pblk, struct pblk_line *line) ret = pblk_recov_scan_oob(pblk, line, p, &done); if (ret) { - pr_err("pblk: could not recover L2P from OOB\n"); + pblk_err(pblk, "could not recover L2P from OOB\n"); goto out; } if (!done) { ret = pblk_recov_scan_all_oob(pblk, line, p); if (ret) { - pr_err("pblk: could not recover L2P from OOB\n"); + pblk_err(pblk, "could not recover L2P from OOB\n"); goto out; } } @@ -737,14 +737,15 @@ static int pblk_recov_check_line_version(struct pblk *pblk, struct line_header *header = &emeta->header; if (header->version_major != EMETA_VERSION_MAJOR) { - pr_err("pblk: line major version mismatch: %d, expected: %d\n", - header->version_major, EMETA_VERSION_MAJOR); + pblk_err(pblk, "line major version mismatch: %d, expected: %d\n", + header->version_major, EMETA_VERSION_MAJOR); return 1; } -#ifdef NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG if (header->version_minor > EMETA_VERSION_MINOR) - pr_info("pblk: newer line minor version found: %d\n", line_v); + pblk_info(pblk, "newer line minor version found: %d\n", + header->version_minor); #endif return 0; @@ -851,7 +852,7 @@ struct pblk_line *pblk_recov_l2p(struct pblk *pblk) continue; if (smeta_buf->header.version_major != SMETA_VERSION_MAJOR) { - pr_err("pblk: found incompatible line version %u\n", + pblk_err(pblk, "found incompatible line version %u\n", smeta_buf->header.version_major); return ERR_PTR(-EINVAL); } @@ -863,7 +864,7 @@ struct pblk_line *pblk_recov_l2p(struct pblk *pblk) } if (memcmp(pblk->instance_uuid, smeta_buf->header.uuid, 16)) { - pr_debug("pblk: ignore line %u due to uuid mismatch\n", + pblk_debug(pblk, "ignore line %u due to uuid mismatch\n", i); continue; } @@ -887,7 +888,7 @@ struct pblk_line *pblk_recov_l2p(struct pblk *pblk) pblk_recov_line_add_ordered(&recov_list, line); found_lines++; - pr_debug("pblk: recovering data line %d, seq:%llu\n", + pblk_debug(pblk, "recovering data line %d, seq:%llu\n", line->id, smeta_buf->seq_nr); } @@ -947,7 +948,7 @@ next: line->emeta = NULL; } else { if (open_lines > 1) - pr_err("pblk: failed to recover L2P\n"); + pblk_err(pblk, "failed to recover L2P\n"); open_lines++; line->meta_line = meta_line; @@ -976,7 +977,7 @@ next: out: if (found_lines != recovered_lines) - pr_err("pblk: failed to recover all found lines %d/%d\n", + pblk_err(pblk, "failed to recover all found lines %d/%d\n", found_lines, recovered_lines); return data_line; @@ -999,7 +1000,7 @@ int pblk_recov_pad(struct pblk *pblk) ret = pblk_recov_pad_oob(pblk, line, left_msecs); if (ret) { - pr_err("pblk: Tear down padding failed (%d)\n", ret); + pblk_err(pblk, "tear down padding failed (%d)\n", ret); return ret; } diff --git a/drivers/lightnvm/pblk-sysfs.c b/drivers/lightnvm/pblk-sysfs.c index 88a0a7c407aa..9fc3dfa168b4 100644 --- a/drivers/lightnvm/pblk-sysfs.c +++ b/drivers/lightnvm/pblk-sysfs.c @@ -268,7 +268,7 @@ static ssize_t pblk_sysfs_lines(struct pblk *pblk, char *page) spin_unlock(&l_mg->free_lock); if (nr_free_lines != free_line_cnt) - pr_err("pblk: corrupted free line list:%d/%d\n", + pblk_err(pblk, "corrupted free line list:%d/%d\n", nr_free_lines, free_line_cnt); sz = snprintf(page, PAGE_SIZE - sz, @@ -421,7 +421,7 @@ static ssize_t pblk_sysfs_get_padding_dist(struct pblk *pblk, char *page) return sz; } -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG static ssize_t pblk_sysfs_stats_debug(struct pblk *pblk, char *page) { return snprintf(page, PAGE_SIZE, @@ -598,7 +598,7 @@ static struct attribute sys_padding_dist = { .mode = 0644, }; -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG static struct attribute sys_stats_debug_attr = { .name = "stats", .mode = 0444, @@ -619,7 +619,7 @@ static struct attribute *pblk_attrs[] = { &sys_write_amp_mileage, &sys_write_amp_trip, &sys_padding_dist, -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG &sys_stats_debug_attr, #endif NULL, @@ -654,7 +654,7 @@ static ssize_t pblk_sysfs_show(struct kobject *kobj, struct attribute *attr, return pblk_sysfs_get_write_amp_trip(pblk, buf); else if (strcmp(attr->name, "padding_dist") == 0) return pblk_sysfs_get_padding_dist(pblk, buf); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG else if (strcmp(attr->name, "stats") == 0) return pblk_sysfs_stats_debug(pblk, buf); #endif @@ -697,8 +697,7 @@ int pblk_sysfs_init(struct gendisk *tdisk) kobject_get(&parent_dev->kobj), "%s", "pblk"); if (ret) { - pr_err("pblk: could not register %s/pblk\n", - tdisk->disk_name); + pblk_err(pblk, "could not register\n"); return ret; } diff --git a/drivers/lightnvm/pblk-write.c b/drivers/lightnvm/pblk-write.c index f353e52941f5..ee774a86cf1e 100644 --- a/drivers/lightnvm/pblk-write.c +++ b/drivers/lightnvm/pblk-write.c @@ -38,7 +38,7 @@ static unsigned long pblk_end_w_bio(struct pblk *pblk, struct nvm_rq *rqd, /* Release flags on context. Protect from writes */ smp_store_release(&w_ctx->flags, flags); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_dec(&rwb->inflight_flush_point); #endif } @@ -51,7 +51,7 @@ static unsigned long pblk_end_w_bio(struct pblk *pblk, struct nvm_rq *rqd, pblk_bio_free_pages(pblk, rqd->bio, c_ctx->nr_valid, c_ctx->nr_padded); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_long_add(rqd->nr_ppas, &pblk->sync_writes); #endif @@ -78,7 +78,7 @@ static void pblk_complete_write(struct pblk *pblk, struct nvm_rq *rqd, unsigned long flags; unsigned long pos; -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_long_sub(c_ctx->nr_valid, &pblk->inflight_writes); #endif @@ -196,7 +196,7 @@ static void pblk_queue_resubmit(struct pblk *pblk, struct pblk_c_ctx *c_ctx) list_add_tail(&r_ctx->list, &pblk->resubmit_list); spin_unlock(&pblk->resubmit_lock); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_long_add(c_ctx->nr_valid, &pblk->recov_writes); #endif } @@ -238,7 +238,7 @@ static void pblk_end_w_fail(struct pblk *pblk, struct nvm_rq *rqd) recovery = mempool_alloc(&pblk->rec_pool, GFP_ATOMIC); if (!recovery) { - pr_err("pblk: could not allocate recovery work\n"); + pblk_err(pblk, "could not allocate recovery work\n"); return; } @@ -258,7 +258,7 @@ static void pblk_end_io_write(struct nvm_rq *rqd) pblk_end_w_fail(pblk, rqd); return; } -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG else WARN_ONCE(rqd->bio->bi_status, "pblk: corrupted write error\n"); #endif @@ -279,7 +279,7 @@ static void pblk_end_io_write_meta(struct nvm_rq *rqd) if (rqd->error) { pblk_log_write_err(pblk, rqd); - pr_err("pblk: metadata I/O failed. Line %d\n", line->id); + pblk_err(pblk, "metadata I/O failed. Line %d\n", line->id); line->w_err_gc->has_write_err = 1; } @@ -356,11 +356,11 @@ static int pblk_calc_secs_to_sync(struct pblk *pblk, unsigned int secs_avail, secs_to_sync = pblk_calc_secs(pblk, secs_avail, secs_to_flush); -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG if ((!secs_to_sync && secs_to_flush) || (secs_to_sync < 0) || (secs_to_sync > secs_avail && !secs_to_flush)) { - pr_err("pblk: bad sector calculation (a:%d,s:%d,f:%d)\n", + pblk_err(pblk, "bad sector calculation (a:%d,s:%d,f:%d)\n", secs_avail, secs_to_sync, secs_to_flush); } #endif @@ -397,7 +397,7 @@ int pblk_submit_meta_io(struct pblk *pblk, struct pblk_line *meta_line) bio = pblk_bio_map_addr(pblk, data, rq_ppas, rq_len, l_mg->emeta_alloc_type, GFP_KERNEL); if (IS_ERR(bio)) { - pr_err("pblk: failed to map emeta io"); + pblk_err(pblk, "failed to map emeta io"); ret = PTR_ERR(bio); goto fail_free_rqd; } @@ -428,7 +428,7 @@ int pblk_submit_meta_io(struct pblk *pblk, struct pblk_line *meta_line) ret = pblk_submit_io(pblk, rqd); if (ret) { - pr_err("pblk: emeta I/O submission failed: %d\n", ret); + pblk_err(pblk, "emeta I/O submission failed: %d\n", ret); goto fail_rollback; } @@ -518,7 +518,7 @@ static int pblk_submit_io_set(struct pblk *pblk, struct nvm_rq *rqd) /* Assign lbas to ppas and populate request structure */ err = pblk_setup_w_rq(pblk, rqd, &erase_ppa); if (err) { - pr_err("pblk: could not setup write request: %d\n", err); + pblk_err(pblk, "could not setup write request: %d\n", err); return NVM_IO_ERR; } @@ -527,7 +527,7 @@ static int pblk_submit_io_set(struct pblk *pblk, struct nvm_rq *rqd) /* Submit data write for current data line */ err = pblk_submit_io(pblk, rqd); if (err) { - pr_err("pblk: data I/O submission failed: %d\n", err); + pblk_err(pblk, "data I/O submission failed: %d\n", err); return NVM_IO_ERR; } @@ -549,7 +549,8 @@ static int pblk_submit_io_set(struct pblk *pblk, struct nvm_rq *rqd) /* Submit metadata write for previous data line */ err = pblk_submit_meta_io(pblk, meta_line); if (err) { - pr_err("pblk: metadata I/O submission failed: %d", err); + pblk_err(pblk, "metadata I/O submission failed: %d", + err); return NVM_IO_ERR; } } @@ -614,7 +615,7 @@ static int pblk_submit_write(struct pblk *pblk) secs_to_sync = pblk_calc_secs_to_sync(pblk, secs_avail, secs_to_flush); if (secs_to_sync > pblk->max_write_pgs) { - pr_err("pblk: bad buffer sync calculation\n"); + pblk_err(pblk, "bad buffer sync calculation\n"); return 1; } @@ -633,14 +634,14 @@ static int pblk_submit_write(struct pblk *pblk) if (pblk_rb_read_to_bio(&pblk->rwb, rqd, pos, secs_to_sync, secs_avail)) { - pr_err("pblk: corrupted write bio\n"); + pblk_err(pblk, "corrupted write bio\n"); goto fail_put_bio; } if (pblk_submit_io_set(pblk, rqd)) goto fail_free_bio; -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_long_add(secs_to_sync, &pblk->sub_writes); #endif diff --git a/drivers/lightnvm/pblk.h b/drivers/lightnvm/pblk.h index 34cc1d64a9d4..4760af7b6499 100644 --- a/drivers/lightnvm/pblk.h +++ b/drivers/lightnvm/pblk.h @@ -119,6 +119,16 @@ struct pblk_g_ctx { u64 lba; }; +/* partial read context */ +struct pblk_pr_ctx { + struct bio *orig_bio; + DECLARE_BITMAP(bitmap, NVM_MAX_VLBA); + unsigned int orig_nr_secs; + unsigned int bio_init_idx; + void *ppa_ptr; + dma_addr_t dma_ppa_list; +}; + /* Pad context */ struct pblk_pad_rq { struct pblk *pblk; @@ -193,7 +203,7 @@ struct pblk_rb { spinlock_t w_lock; /* Write lock */ spinlock_t s_lock; /* Sync lock */ -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG atomic_t inflight_flush_point; /* Not served REQ_FLUSH | REQ_FUA */ #endif }; @@ -608,9 +618,6 @@ struct pblk { int min_write_pgs; /* Minimum amount of pages required by controller */ int max_write_pgs; /* Maximum amount of pages supported by controller */ - int pgs_in_buffer; /* Number of pages that need to be held in buffer to - * guarantee successful reads. - */ sector_t capacity; /* Device capacity when bad blocks are subtracted */ @@ -639,7 +646,7 @@ struct pblk { u64 nr_flush_rst; /* Flushes reset value for pad dist.*/ atomic64_t nr_flush; /* Number of flush/fua I/O */ -#ifdef CONFIG_NVM_DEBUG +#ifdef CONFIG_NVM_PBLK_DEBUG /* Non-persistent debug counters, 4kb sector I/Os */ atomic_long_t inflight_writes; /* Inflight writes (user and gc) */ atomic_long_t padded_writes; /* Sectors padded due to flush/fua */ @@ -706,6 +713,15 @@ struct pblk_line_ws { #define pblk_g_rq_size (sizeof(struct nvm_rq) + sizeof(struct pblk_g_ctx)) #define pblk_w_rq_size (sizeof(struct nvm_rq) + sizeof(struct pblk_c_ctx)) +#define pblk_err(pblk, fmt, ...) \ + pr_err("pblk %s: " fmt, pblk->disk->disk_name, ##__VA_ARGS__) +#define pblk_info(pblk, fmt, ...) \ + pr_info("pblk %s: " fmt, pblk->disk->disk_name, ##__VA_ARGS__) +#define pblk_warn(pblk, fmt, ...) \ + pr_warn("pblk %s: " fmt, pblk->disk->disk_name, ##__VA_ARGS__) +#define pblk_debug(pblk, fmt, ...) \ + pr_debug("pblk %s: " fmt, pblk->disk->disk_name, ##__VA_ARGS__) + /* * pblk ring buffer operations */ @@ -1282,20 +1298,22 @@ static inline int pblk_io_aligned(struct pblk *pblk, int nr_secs) return !(nr_secs % pblk->min_write_pgs); } -#ifdef CONFIG_NVM_DEBUG -static inline void print_ppa(struct nvm_geo *geo, struct ppa_addr *p, +#ifdef CONFIG_NVM_PBLK_DEBUG +static inline void print_ppa(struct pblk *pblk, struct ppa_addr *p, char *msg, int error) { + struct nvm_geo *geo = &pblk->dev->geo; + if (p->c.is_cached) { - pr_err("ppa: (%s: %x) cache line: %llu\n", + pblk_err(pblk, "ppa: (%s: %x) cache line: %llu\n", msg, error, (u64)p->c.line); } else if (geo->version == NVM_OCSSD_SPEC_12) { - pr_err("ppa: (%s: %x):ch:%d,lun:%d,blk:%d,pg:%d,pl:%d,sec:%d\n", + pblk_err(pblk, "ppa: (%s: %x):ch:%d,lun:%d,blk:%d,pg:%d,pl:%d,sec:%d\n", msg, error, p->g.ch, p->g.lun, p->g.blk, p->g.pg, p->g.pl, p->g.sec); } else { - pr_err("ppa: (%s: %x):ch:%d,lun:%d,chk:%d,sec:%d\n", + pblk_err(pblk, "ppa: (%s: %x):ch:%d,lun:%d,chk:%d,sec:%d\n", msg, error, p->m.grp, p->m.pu, p->m.chk, p->m.sec); } @@ -1307,16 +1325,16 @@ static inline void pblk_print_failed_rqd(struct pblk *pblk, struct nvm_rq *rqd, int bit = -1; if (rqd->nr_ppas == 1) { - print_ppa(&pblk->dev->geo, &rqd->ppa_addr, "rqd", error); + print_ppa(pblk, &rqd->ppa_addr, "rqd", error); return; } while ((bit = find_next_bit((void *)&rqd->ppa_status, rqd->nr_ppas, bit + 1)) < rqd->nr_ppas) { - print_ppa(&pblk->dev->geo, &rqd->ppa_list[bit], "rqd", error); + print_ppa(pblk, &rqd->ppa_list[bit], "rqd", error); } - pr_err("error:%d, ppa_status:%llx\n", error, rqd->ppa_status); + pblk_err(pblk, "error:%d, ppa_status:%llx\n", error, rqd->ppa_status); } static inline int pblk_boundary_ppa_checks(struct nvm_tgt_dev *tgt_dev, @@ -1347,7 +1365,7 @@ static inline int pblk_boundary_ppa_checks(struct nvm_tgt_dev *tgt_dev, continue; } - print_ppa(geo, ppa, "boundary", i); + print_ppa(tgt_dev->q->queuedata, ppa, "boundary", i); return 1; } @@ -1377,7 +1395,7 @@ static inline int pblk_check_io(struct pblk *pblk, struct nvm_rq *rqd) spin_lock(&line->lock); if (line->state != PBLK_LINESTATE_OPEN) { - pr_err("pblk: bad ppa: line:%d,state:%d\n", + pblk_err(pblk, "bad ppa: line:%d,state:%d\n", line->id, line->state); WARN_ON(1); spin_unlock(&line->lock); diff --git a/drivers/md/bcache/bcache.h b/drivers/md/bcache/bcache.h index d6bf294f3907..05f82ff6f016 100644 --- a/drivers/md/bcache/bcache.h +++ b/drivers/md/bcache/bcache.h @@ -328,13 +328,6 @@ struct cached_dev { */ atomic_t has_dirty; - /* - * Set to zero by things that touch the backing volume-- except - * writeback. Incremented by writeback. Used to determine when to - * accelerate idle writeback. - */ - atomic_t backing_idle; - struct bch_ratelimit writeback_rate; struct delayed_work writeback_rate_update; @@ -423,9 +416,9 @@ struct cache { /* * When allocating new buckets, prio_write() gets first dibs - since we * may not be allocate at all without writing priorities and gens. - * prio_buckets[] contains the last buckets we wrote priorities to (so - * gc can mark them as metadata), prio_next[] contains the buckets - * allocated for the next prio write. + * prio_last_buckets[] contains the last buckets we wrote priorities to + * (so gc can mark them as metadata), prio_buckets[] contains the + * buckets allocated for the next prio write. */ uint64_t *prio_buckets; uint64_t *prio_last_buckets; @@ -474,6 +467,7 @@ struct cache { struct gc_stat { size_t nodes; + size_t nodes_pre; size_t key_bytes; size_t nkeys; @@ -514,6 +508,8 @@ struct cache_set { struct cache_accounting accounting; unsigned long flags; + atomic_t idle_counter; + atomic_t at_max_writeback_rate; struct cache_sb sb; @@ -523,8 +519,10 @@ struct cache_set { struct bcache_device **devices; unsigned devices_max_used; + atomic_t attached_dev_nr; struct list_head cached_devs; uint64_t cached_dev_sectors; + atomic_long_t flash_dev_dirty_sectors; struct closure caching; struct closure sb_write; @@ -603,6 +601,10 @@ struct cache_set { */ atomic_t rescale; /* + * used for GC, identify if any front side I/Os is inflight + */ + atomic_t search_inflight; + /* * When we invalidate buckets, we use both the priority and the amount * of good data to determine which buckets to reuse first - to weight * those together consistently we keep track of the smallest nonzero @@ -995,7 +997,7 @@ void bch_open_buckets_free(struct cache_set *); int bch_cache_allocator_start(struct cache *ca); void bch_debug_exit(void); -int bch_debug_init(struct kobject *); +void bch_debug_init(struct kobject *kobj); void bch_request_exit(void); int bch_request_init(void); diff --git a/drivers/md/bcache/bset.c b/drivers/md/bcache/bset.c index f3403b45bc28..596c93b44e9b 100644 --- a/drivers/md/bcache/bset.c +++ b/drivers/md/bcache/bset.c @@ -366,6 +366,10 @@ EXPORT_SYMBOL(bch_btree_keys_init); /* Binary tree stuff for auxiliary search trees */ +/* + * return array index next to j when does in-order traverse + * of a binary tree which is stored in a linear array + */ static unsigned inorder_next(unsigned j, unsigned size) { if (j * 2 + 1 < size) { @@ -379,6 +383,10 @@ static unsigned inorder_next(unsigned j, unsigned size) return j; } +/* + * return array index previous to j when does in-order traverse + * of a binary tree which is stored in a linear array + */ static unsigned inorder_prev(unsigned j, unsigned size) { if (j * 2 < size) { @@ -421,6 +429,10 @@ static unsigned __to_inorder(unsigned j, unsigned size, unsigned extra) return j; } +/* + * Return the cacheline index in bset_tree->data, where j is index + * from a linear array which stores the auxiliar binary tree + */ static unsigned to_inorder(unsigned j, struct bset_tree *t) { return __to_inorder(j, t->size, t->extra); @@ -441,6 +453,10 @@ static unsigned __inorder_to_tree(unsigned j, unsigned size, unsigned extra) return j; } +/* + * Return an index from a linear array which stores the auxiliar binary + * tree, j is the cacheline index of t->data. + */ static unsigned inorder_to_tree(unsigned j, struct bset_tree *t) { return __inorder_to_tree(j, t->size, t->extra); @@ -546,6 +562,20 @@ static inline uint64_t shrd128(uint64_t high, uint64_t low, uint8_t shift) return low; } +/* + * Calculate mantissa value for struct bkey_float. + * If most significant bit of f->exponent is not set, then + * - f->exponent >> 6 is 0 + * - p[0] points to bkey->low + * - p[-1] borrows bits from KEY_INODE() of bkey->high + * if most isgnificant bits of f->exponent is set, then + * - f->exponent >> 6 is 1 + * - p[0] points to bits from KEY_INODE() of bkey->high + * - p[-1] points to other bits from KEY_INODE() of + * bkey->high too. + * See make_bfloat() to check when most significant bit of f->exponent + * is set or not. + */ static inline unsigned bfloat_mantissa(const struct bkey *k, struct bkey_float *f) { @@ -570,6 +600,16 @@ static void make_bfloat(struct bset_tree *t, unsigned j) BUG_ON(m < l || m > r); BUG_ON(bkey_next(p) != m); + /* + * If l and r have different KEY_INODE values (different backing + * device), f->exponent records how many least significant bits + * are different in KEY_INODE values and sets most significant + * bits to 1 (by +64). + * If l and r have same KEY_INODE value, f->exponent records + * how many different bits in least significant bits of bkey->low. + * See bfloat_mantiss() how the most significant bit of + * f->exponent is used to calculate bfloat mantissa value. + */ if (KEY_INODE(l) != KEY_INODE(r)) f->exponent = fls64(KEY_INODE(r) ^ KEY_INODE(l)) + 64; else @@ -633,6 +673,15 @@ void bch_bset_init_next(struct btree_keys *b, struct bset *i, uint64_t magic) } EXPORT_SYMBOL(bch_bset_init_next); +/* + * Build auxiliary binary tree 'struct bset_tree *t', this tree is used to + * accelerate bkey search in a btree node (pointed by bset_tree->data in + * memory). After search in the auxiliar tree by calling bset_search_tree(), + * a struct bset_search_iter is returned which indicates range [l, r] from + * bset_tree->data where the searching bkey might be inside. Then a followed + * linear comparison does the exact search, see __bch_bset_search() for how + * the auxiliary tree is used. + */ void bch_bset_build_written_tree(struct btree_keys *b) { struct bset_tree *t = bset_tree_last(b); @@ -898,6 +947,17 @@ static struct bset_search_iter bset_search_tree(struct bset_tree *t, unsigned inorder, j, n = 1; do { + /* + * A bit trick here. + * If p < t->size, (int)(p - t->size) is a minus value and + * the most significant bit is set, right shifting 31 bits + * gets 1. If p >= t->size, the most significant bit is + * not set, right shifting 31 bits gets 0. + * So the following 2 lines equals to + * if (p >= t->size) + * p = 0; + * but a branch instruction is avoided. + */ unsigned p = n << 4; p &= ((int) (p - t->size)) >> 31; @@ -907,6 +967,9 @@ static struct bset_search_iter bset_search_tree(struct bset_tree *t, f = &t->tree[j]; /* + * Similar bit trick, use subtract operation to avoid a branch + * instruction. + * * n = (f->mantissa > bfloat_mantissa()) * ? j * 2 * : j * 2 + 1; diff --git a/drivers/md/bcache/btree.c b/drivers/md/bcache/btree.c index 547c9eedc2f4..c19f7716df88 100644 --- a/drivers/md/bcache/btree.c +++ b/drivers/md/bcache/btree.c @@ -90,6 +90,9 @@ #define MAX_NEED_GC 64 #define MAX_SAVE_PRIO 72 +#define MAX_GC_TIMES 100 +#define MIN_GC_NODES 100 +#define GC_SLEEP_MS 100 #define PTR_DIRTY_BIT (((uint64_t) 1 << 36)) @@ -1008,6 +1011,13 @@ retry: BUG_ON(b->level != level); } + if (btree_node_io_error(b)) { + rw_unlock(write, b); + return ERR_PTR(-EIO); + } + + BUG_ON(!b->written); + b->parent = parent; b->accessed = 1; @@ -1019,13 +1029,6 @@ retry: for (; i <= b->keys.nsets; i++) prefetch(b->keys.set[i].data); - if (btree_node_io_error(b)) { - rw_unlock(write, b); - return ERR_PTR(-EIO); - } - - BUG_ON(!b->written); - return b; } @@ -1520,6 +1523,32 @@ static unsigned btree_gc_count_keys(struct btree *b) return ret; } +static size_t btree_gc_min_nodes(struct cache_set *c) +{ + size_t min_nodes; + + /* + * Since incremental GC would stop 100ms when front + * side I/O comes, so when there are many btree nodes, + * if GC only processes constant (100) nodes each time, + * GC would last a long time, and the front side I/Os + * would run out of the buckets (since no new bucket + * can be allocated during GC), and be blocked again. + * So GC should not process constant nodes, but varied + * nodes according to the number of btree nodes, which + * realized by dividing GC into constant(100) times, + * so when there are many btree nodes, GC can process + * more nodes each time, otherwise, GC will process less + * nodes each time (but no less than MIN_GC_NODES) + */ + min_nodes = c->gc_stats.nodes / MAX_GC_TIMES; + if (min_nodes < MIN_GC_NODES) + min_nodes = MIN_GC_NODES; + + return min_nodes; +} + + static int btree_gc_recurse(struct btree *b, struct btree_op *op, struct closure *writes, struct gc_stat *gc) { @@ -1585,6 +1614,13 @@ static int btree_gc_recurse(struct btree *b, struct btree_op *op, memmove(r + 1, r, sizeof(r[0]) * (GC_MERGE_NODES - 1)); r->b = NULL; + if (atomic_read(&b->c->search_inflight) && + gc->nodes >= gc->nodes_pre + btree_gc_min_nodes(b->c)) { + gc->nodes_pre = gc->nodes; + ret = -EAGAIN; + break; + } + if (need_resched()) { ret = -EAGAIN; break; @@ -1753,7 +1789,10 @@ static void bch_btree_gc(struct cache_set *c) closure_sync(&writes); cond_resched(); - if (ret && ret != -EAGAIN) + if (ret == -EAGAIN) + schedule_timeout_interruptible(msecs_to_jiffies + (GC_SLEEP_MS)); + else if (ret) pr_warn("gc failed!"); } while (ret && !test_bit(CACHE_SET_IO_DISABLE, &c->flags)); @@ -1834,8 +1873,14 @@ static int bch_btree_check_recurse(struct btree *b, struct btree_op *op) do { k = bch_btree_iter_next_filter(&iter, &b->keys, bch_ptr_bad); - if (k) + if (k) { btree_node_prefetch(b, k); + /* + * initiallize c->gc_stats.nodes + * for incremental GC + */ + b->c->gc_stats.nodes++; + } if (p) ret = btree(check_recurse, p, b, op); diff --git a/drivers/md/bcache/btree.h b/drivers/md/bcache/btree.h index d211e2c25b6b..68e9d926134d 100644 --- a/drivers/md/bcache/btree.h +++ b/drivers/md/bcache/btree.h @@ -152,7 +152,7 @@ static inline bool btree_node_ ## flag(struct btree *b) \ { return test_bit(BTREE_NODE_ ## flag, &b->flags); } \ \ static inline void set_btree_node_ ## flag(struct btree *b) \ -{ set_bit(BTREE_NODE_ ## flag, &b->flags); } \ +{ set_bit(BTREE_NODE_ ## flag, &b->flags); } enum btree_flags { BTREE_NODE_io_error, diff --git a/drivers/md/bcache/closure.c b/drivers/md/bcache/closure.c index 0e14969182c6..618253683d40 100644 --- a/drivers/md/bcache/closure.c +++ b/drivers/md/bcache/closure.c @@ -199,11 +199,16 @@ static const struct file_operations debug_ops = { .release = single_release }; -int __init closure_debug_init(void) +void __init closure_debug_init(void) { - closure_debug = debugfs_create_file("closures", - 0400, bcache_debug, NULL, &debug_ops); - return IS_ERR_OR_NULL(closure_debug); + if (!IS_ERR_OR_NULL(bcache_debug)) + /* + * it is unnecessary to check return value of + * debugfs_create_file(), we should not care + * about this. + */ + closure_debug = debugfs_create_file( + "closures", 0400, bcache_debug, NULL, &debug_ops); } #endif diff --git a/drivers/md/bcache/closure.h b/drivers/md/bcache/closure.h index 71427eb5fdae..7c2c5bc7c88b 100644 --- a/drivers/md/bcache/closure.h +++ b/drivers/md/bcache/closure.h @@ -186,13 +186,13 @@ static inline void closure_sync(struct closure *cl) #ifdef CONFIG_BCACHE_CLOSURES_DEBUG -int closure_debug_init(void); +void closure_debug_init(void); void closure_debug_create(struct closure *cl); void closure_debug_destroy(struct closure *cl); #else -static inline int closure_debug_init(void) { return 0; } +static inline void closure_debug_init(void) {} static inline void closure_debug_create(struct closure *cl) {} static inline void closure_debug_destroy(struct closure *cl) {} diff --git a/drivers/md/bcache/debug.c b/drivers/md/bcache/debug.c index d030ce3025a6..12034c07257b 100644 --- a/drivers/md/bcache/debug.c +++ b/drivers/md/bcache/debug.c @@ -110,11 +110,15 @@ void bch_data_verify(struct cached_dev *dc, struct bio *bio) struct bio_vec bv, cbv; struct bvec_iter iter, citer = { 0 }; - check = bio_clone_kmalloc(bio, GFP_NOIO); + check = bio_kmalloc(GFP_NOIO, bio_segments(bio)); if (!check) return; + check->bi_disk = bio->bi_disk; check->bi_opf = REQ_OP_READ; + check->bi_iter.bi_sector = bio->bi_iter.bi_sector; + check->bi_iter.bi_size = bio->bi_iter.bi_size; + bch_bio_map(check, NULL); if (bch_bio_alloc_pages(check, GFP_NOIO)) goto out_put; @@ -248,11 +252,12 @@ void bch_debug_exit(void) debugfs_remove_recursive(bcache_debug); } -int __init bch_debug_init(struct kobject *kobj) +void __init bch_debug_init(struct kobject *kobj) { - if (!IS_ENABLED(CONFIG_DEBUG_FS)) - return 0; - + /* + * it is unnecessary to check return value of + * debugfs_create_file(), we should not care + * about this. + */ bcache_debug = debugfs_create_dir("bcache", NULL); - return IS_ERR_OR_NULL(bcache_debug); } diff --git a/drivers/md/bcache/journal.c b/drivers/md/bcache/journal.c index 18f1b5239620..10748c626a1d 100644 --- a/drivers/md/bcache/journal.c +++ b/drivers/md/bcache/journal.c @@ -828,6 +828,7 @@ void bch_journal_free(struct cache_set *c) free_pages((unsigned long) c->journal.w[1].data, JSET_BITS); free_pages((unsigned long) c->journal.w[0].data, JSET_BITS); free_fifo(&c->journal.pin); + free_heap(&c->flush_btree); } int bch_journal_alloc(struct cache_set *c) diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c index ae67f5fa8047..7dbe8b6316a0 100644 --- a/drivers/md/bcache/request.c +++ b/drivers/md/bcache/request.c @@ -107,7 +107,7 @@ static int bch_keylist_realloc(struct keylist *l, unsigned u64s, /* * The journalling code doesn't handle the case where the keys to insert * is bigger than an empty write: If we just return -ENOMEM here, - * bio_insert() and bio_invalidate() will insert the keys created so far + * bch_data_insert_keys() will insert the keys created so far * and finish the rest when the keylist is empty. */ if (newsize * sizeof(uint64_t) > block_bytes(c) - sizeof(struct jset)) @@ -667,8 +667,7 @@ static void backing_request_endio(struct bio *bio) static void bio_complete(struct search *s) { if (s->orig_bio) { - generic_end_io_acct(s->d->disk->queue, - bio_data_dir(s->orig_bio), + generic_end_io_acct(s->d->disk->queue, bio_op(s->orig_bio), &s->d->disk->part0, s->start_time); trace_bcache_request_end(s->d, s->orig_bio); @@ -702,6 +701,8 @@ static void search_free(struct closure *cl) { struct search *s = container_of(cl, struct search, cl); + atomic_dec(&s->d->c->search_inflight); + if (s->iop.bio) bio_put(s->iop.bio); @@ -719,6 +720,7 @@ static inline struct search *search_alloc(struct bio *bio, closure_init(&s->cl, NULL); do_bio_hook(s, bio, request_endio); + atomic_inc(&d->c->search_inflight); s->orig_bio = bio; s->cache_miss = NULL; @@ -1062,8 +1064,7 @@ static void detached_dev_end_io(struct bio *bio) bio->bi_end_io = ddip->bi_end_io; bio->bi_private = ddip->bi_private; - generic_end_io_acct(ddip->d->disk->queue, - bio_data_dir(bio), + generic_end_io_acct(ddip->d->disk->queue, bio_op(bio), &ddip->d->disk->part0, ddip->start_time); if (bio->bi_status) { @@ -1102,6 +1103,44 @@ static void detached_dev_do_request(struct bcache_device *d, struct bio *bio) generic_make_request(bio); } +static void quit_max_writeback_rate(struct cache_set *c, + struct cached_dev *this_dc) +{ + int i; + struct bcache_device *d; + struct cached_dev *dc; + + /* + * mutex bch_register_lock may compete with other parallel requesters, + * or attach/detach operations on other backing device. Waiting to + * the mutex lock may increase I/O request latency for seconds or more. + * To avoid such situation, if mutext_trylock() failed, only writeback + * rate of current cached device is set to 1, and __update_write_back() + * will decide writeback rate of other cached devices (remember now + * c->idle_counter is 0 already). + */ + if (mutex_trylock(&bch_register_lock)) { + for (i = 0; i < c->devices_max_used; i++) { + if (!c->devices[i]) + continue; + + if (UUID_FLASH_ONLY(&c->uuids[i])) + continue; + + d = c->devices[i]; + dc = container_of(d, struct cached_dev, disk); + /* + * set writeback rate to default minimum value, + * then let update_writeback_rate() to decide the + * upcoming rate. + */ + atomic_long_set(&dc->writeback_rate.rate, 1); + } + mutex_unlock(&bch_register_lock); + } else + atomic_long_set(&this_dc->writeback_rate.rate, 1); +} + /* Cached devices - read & write stuff */ static blk_qc_t cached_dev_make_request(struct request_queue *q, @@ -1119,8 +1158,25 @@ static blk_qc_t cached_dev_make_request(struct request_queue *q, return BLK_QC_T_NONE; } - atomic_set(&dc->backing_idle, 0); - generic_start_io_acct(q, rw, bio_sectors(bio), &d->disk->part0); + if (likely(d->c)) { + if (atomic_read(&d->c->idle_counter)) + atomic_set(&d->c->idle_counter, 0); + /* + * If at_max_writeback_rate of cache set is true and new I/O + * comes, quit max writeback rate of all cached devices + * attached to this cache set, and set at_max_writeback_rate + * to false. + */ + if (unlikely(atomic_read(&d->c->at_max_writeback_rate) == 1)) { + atomic_set(&d->c->at_max_writeback_rate, 0); + quit_max_writeback_rate(d->c, dc); + } + } + + generic_start_io_acct(q, + bio_op(bio), + bio_sectors(bio), + &d->disk->part0); bio_set_dev(bio, dc->bdev); bio->bi_iter.bi_sector += dc->sb.data_offset; @@ -1229,7 +1285,6 @@ static blk_qc_t flash_dev_make_request(struct request_queue *q, struct search *s; struct closure *cl; struct bcache_device *d = bio->bi_disk->private_data; - int rw = bio_data_dir(bio); if (unlikely(d->c && test_bit(CACHE_SET_IO_DISABLE, &d->c->flags))) { bio->bi_status = BLK_STS_IOERR; @@ -1237,7 +1292,7 @@ static blk_qc_t flash_dev_make_request(struct request_queue *q, return BLK_QC_T_NONE; } - generic_start_io_acct(q, rw, bio_sectors(bio), &d->disk->part0); + generic_start_io_acct(q, bio_op(bio), bio_sectors(bio), &d->disk->part0); s = search_alloc(bio, d); cl = &s->cl; @@ -1254,7 +1309,7 @@ static blk_qc_t flash_dev_make_request(struct request_queue *q, flash_dev_nodata, bcache_wq); return BLK_QC_T_NONE; - } else if (rw) { + } else if (bio_data_dir(bio)) { bch_keybuf_check_overlapping(&s->iop.c->moving_gc_keys, &KEY(d->id, bio->bi_iter.bi_sector, 0), &KEY(d->id, bio_end_sector(bio), 0)); diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c index fa4058e43202..55a37641aa95 100644 --- a/drivers/md/bcache/super.c +++ b/drivers/md/bcache/super.c @@ -181,7 +181,7 @@ static const char *read_super(struct cache_sb *sb, struct block_device *bdev, goto err; } - sb->last_mount = get_seconds(); + sb->last_mount = (u32)ktime_get_real_seconds(); err = NULL; get_page(bh->b_page); @@ -696,12 +696,14 @@ static void bcache_device_detach(struct bcache_device *d) { lockdep_assert_held(&bch_register_lock); + atomic_dec(&d->c->attached_dev_nr); + if (test_bit(BCACHE_DEV_DETACHING, &d->flags)) { struct uuid_entry *u = d->c->uuids + d->id; SET_UUID_FLASH_ONLY(u, 0); memcpy(u->uuid, invalid_uuid, 16); - u->invalidated = cpu_to_le32(get_seconds()); + u->invalidated = cpu_to_le32((u32)ktime_get_real_seconds()); bch_uuid_write(d->c); } @@ -796,11 +798,12 @@ static int bcache_device_init(struct bcache_device *d, unsigned block_size, return idx; if (bioset_init(&d->bio_split, 4, offsetof(struct bbio, bio), - BIOSET_NEED_BVECS|BIOSET_NEED_RESCUER) || - !(d->disk = alloc_disk(BCACHE_MINORS))) { - ida_simple_remove(&bcache_device_idx, idx); - return -ENOMEM; - } + BIOSET_NEED_BVECS|BIOSET_NEED_RESCUER)) + goto err; + + d->disk = alloc_disk(BCACHE_MINORS); + if (!d->disk) + goto err; set_capacity(d->disk, sectors); snprintf(d->disk->disk_name, DISK_NAME_LEN, "bcache%i", idx); @@ -834,6 +837,11 @@ static int bcache_device_init(struct bcache_device *d, unsigned block_size, blk_queue_write_cache(q, true, true); return 0; + +err: + ida_simple_remove(&bcache_device_idx, idx); + return -ENOMEM; + } /* Cached device */ @@ -1027,7 +1035,7 @@ void bch_cached_dev_detach(struct cached_dev *dc) int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c, uint8_t *set_uuid) { - uint32_t rtime = cpu_to_le32(get_seconds()); + uint32_t rtime = cpu_to_le32((u32)ktime_get_real_seconds()); struct uuid_entry *u; struct cached_dev *exist_dc, *t; @@ -1070,7 +1078,7 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c, (BDEV_STATE(&dc->sb) == BDEV_STATE_STALE || BDEV_STATE(&dc->sb) == BDEV_STATE_NONE)) { memcpy(u->uuid, invalid_uuid, 16); - u->invalidated = cpu_to_le32(get_seconds()); + u->invalidated = cpu_to_le32((u32)ktime_get_real_seconds()); u = NULL; } @@ -1138,6 +1146,7 @@ int bch_cached_dev_attach(struct cached_dev *dc, struct cache_set *c, bch_cached_dev_run(dc); bcache_device_link(&dc->disk, c, "bdev"); + atomic_inc(&c->attached_dev_nr); /* Allow the writeback thread to proceed */ up_write(&dc->writeback_lock); @@ -1285,6 +1294,7 @@ static void register_bdev(struct cache_sb *sb, struct page *sb_page, pr_info("registered backing device %s", dc->backing_dev_name); list_add(&dc->list, &uncached_devices); + /* attach to a matched cache set if it exists */ list_for_each_entry(c, &bch_cache_sets, list) bch_cached_dev_attach(dc, c, NULL); @@ -1311,6 +1321,8 @@ static void flash_dev_free(struct closure *cl) { struct bcache_device *d = container_of(cl, struct bcache_device, cl); mutex_lock(&bch_register_lock); + atomic_long_sub(bcache_dev_sectors_dirty(d), + &d->c->flash_dev_dirty_sectors); bcache_device_free(d); mutex_unlock(&bch_register_lock); kobject_put(&d->kobj); @@ -1390,7 +1402,7 @@ int bch_flash_dev_create(struct cache_set *c, uint64_t size) get_random_bytes(u->uuid, 16); memset(u->label, 0, 32); - u->first_reg = u->last_reg = cpu_to_le32(get_seconds()); + u->first_reg = u->last_reg = cpu_to_le32((u32)ktime_get_real_seconds()); SET_UUID_FLASH_ONLY(u, 1); u->sectors = size >> 9; @@ -1687,6 +1699,7 @@ struct cache_set *bch_cache_set_alloc(struct cache_sb *sb) c->block_bits = ilog2(sb->block_size); c->nr_uuids = bucket_bytes(c) / sizeof(struct uuid_entry); c->devices_max_used = 0; + atomic_set(&c->attached_dev_nr, 0); c->btree_pages = bucket_pages(c); if (c->btree_pages > BTREE_MAX_PAGES) c->btree_pages = max_t(int, c->btree_pages / 4, @@ -1894,7 +1907,7 @@ static void run_cache_set(struct cache_set *c) goto err; closure_sync(&cl); - c->sb.last_mount = get_seconds(); + c->sb.last_mount = (u32)ktime_get_real_seconds(); bcache_write_super(c); list_for_each_entry_safe(dc, t, &uncached_devices, list) @@ -2163,8 +2176,12 @@ static ssize_t register_bcache(struct kobject *k, struct kobj_attribute *attr, if (!try_module_get(THIS_MODULE)) return -EBUSY; - if (!(path = kstrndup(buffer, size, GFP_KERNEL)) || - !(sb = kmalloc(sizeof(struct cache_sb), GFP_KERNEL))) + path = kstrndup(buffer, size, GFP_KERNEL); + if (!path) + goto err; + + sb = kmalloc(sizeof(struct cache_sb), GFP_KERNEL); + if (!sb) goto err; err = "failed to open device"; @@ -2324,13 +2341,21 @@ static int __init bcache_init(void) return bcache_major; } - if (!(bcache_wq = alloc_workqueue("bcache", WQ_MEM_RECLAIM, 0)) || - !(bcache_kobj = kobject_create_and_add("bcache", fs_kobj)) || - bch_request_init() || - bch_debug_init(bcache_kobj) || closure_debug_init() || + bcache_wq = alloc_workqueue("bcache", WQ_MEM_RECLAIM, 0); + if (!bcache_wq) + goto err; + + bcache_kobj = kobject_create_and_add("bcache", fs_kobj); + if (!bcache_kobj) + goto err; + + if (bch_request_init() || sysfs_create_files(bcache_kobj, files)) goto err; + bch_debug_init(bcache_kobj); + closure_debug_init(); + return 0; err: bcache_exit(); diff --git a/drivers/md/bcache/sysfs.c b/drivers/md/bcache/sysfs.c index 225b15aa0340..81d3520b0702 100644 --- a/drivers/md/bcache/sysfs.c +++ b/drivers/md/bcache/sysfs.c @@ -149,6 +149,7 @@ SHOW(__bch_cached_dev) struct cached_dev *dc = container_of(kobj, struct cached_dev, disk.kobj); const char *states[] = { "no cache", "clean", "dirty", "inconsistent" }; + int wb = dc->writeback_running; #define var(stat) (dc->stat) @@ -170,7 +171,8 @@ SHOW(__bch_cached_dev) var_printf(writeback_running, "%i"); var_print(writeback_delay); var_print(writeback_percent); - sysfs_hprint(writeback_rate, dc->writeback_rate.rate << 9); + sysfs_hprint(writeback_rate, + wb ? atomic_long_read(&dc->writeback_rate.rate) << 9 : 0); sysfs_hprint(io_errors, atomic_read(&dc->io_errors)); sysfs_printf(io_error_limit, "%i", dc->error_limit); sysfs_printf(io_disable, "%i", dc->io_disable); @@ -188,15 +190,22 @@ SHOW(__bch_cached_dev) char change[20]; s64 next_io; - bch_hprint(rate, dc->writeback_rate.rate << 9); - bch_hprint(dirty, bcache_dev_sectors_dirty(&dc->disk) << 9); - bch_hprint(target, dc->writeback_rate_target << 9); - bch_hprint(proportional,dc->writeback_rate_proportional << 9); - bch_hprint(integral, dc->writeback_rate_integral_scaled << 9); - bch_hprint(change, dc->writeback_rate_change << 9); - - next_io = div64_s64(dc->writeback_rate.next - local_clock(), - NSEC_PER_MSEC); + /* + * Except for dirty and target, other values should + * be 0 if writeback is not running. + */ + bch_hprint(rate, + wb ? atomic_long_read(&dc->writeback_rate.rate) << 9 + : 0); + bch_hprint(dirty, bcache_dev_sectors_dirty(&dc->disk) << 9); + bch_hprint(target, dc->writeback_rate_target << 9); + bch_hprint(proportional, + wb ? dc->writeback_rate_proportional << 9 : 0); + bch_hprint(integral, + wb ? dc->writeback_rate_integral_scaled << 9 : 0); + bch_hprint(change, wb ? dc->writeback_rate_change << 9 : 0); + next_io = wb ? div64_s64(dc->writeback_rate.next-local_clock(), + NSEC_PER_MSEC) : 0; return sprintf(buf, "rate:\t\t%s/sec\n" @@ -255,8 +264,19 @@ STORE(__cached_dev) sysfs_strtoul_clamp(writeback_percent, dc->writeback_percent, 0, 40); - sysfs_strtoul_clamp(writeback_rate, - dc->writeback_rate.rate, 1, INT_MAX); + if (attr == &sysfs_writeback_rate) { + ssize_t ret; + long int v = atomic_long_read(&dc->writeback_rate.rate); + + ret = strtoul_safe_clamp(buf, v, 1, INT_MAX); + + if (!ret) { + atomic_long_set(&dc->writeback_rate.rate, v); + ret = size; + } + + return ret; + } sysfs_strtoul_clamp(writeback_rate_update_seconds, dc->writeback_rate_update_seconds, @@ -338,8 +358,8 @@ STORE(__cached_dev) if (!v) return size; } - - pr_err("Can't attach %s: cache set not found", buf); + if (v == -ENOENT) + pr_err("Can't attach %s: cache set not found", buf); return v; } diff --git a/drivers/md/bcache/util.c b/drivers/md/bcache/util.c index fc479b026d6d..b15256bcf0e7 100644 --- a/drivers/md/bcache/util.c +++ b/drivers/md/bcache/util.c @@ -200,7 +200,7 @@ uint64_t bch_next_delay(struct bch_ratelimit *d, uint64_t done) { uint64_t now = local_clock(); - d->next += div_u64(done * NSEC_PER_SEC, d->rate); + d->next += div_u64(done * NSEC_PER_SEC, atomic_long_read(&d->rate)); /* Bound the time. Don't let us fall further than 2 seconds behind * (this prevents unnecessary backlog that would make it impossible diff --git a/drivers/md/bcache/util.h b/drivers/md/bcache/util.h index cced87f8eb27..f7b0133c9d2f 100644 --- a/drivers/md/bcache/util.h +++ b/drivers/md/bcache/util.h @@ -442,7 +442,7 @@ struct bch_ratelimit { * Rate at which we want to do work, in units per second * The units here correspond to the units passed to bch_next_delay() */ - uint32_t rate; + atomic_long_t rate; }; static inline void bch_ratelimit_reset(struct bch_ratelimit *d) diff --git a/drivers/md/bcache/writeback.c b/drivers/md/bcache/writeback.c index ad45ebe1a74b..481d4cf38ac0 100644 --- a/drivers/md/bcache/writeback.c +++ b/drivers/md/bcache/writeback.c @@ -27,7 +27,7 @@ static uint64_t __calc_target_rate(struct cached_dev *dc) * flash-only devices */ uint64_t cache_sectors = c->nbuckets * c->sb.bucket_size - - bcache_flash_devs_sectors_dirty(c); + atomic_long_read(&c->flash_dev_dirty_sectors); /* * Unfortunately there is no control of global dirty data. If the @@ -104,11 +104,56 @@ static void __update_writeback_rate(struct cached_dev *dc) dc->writeback_rate_proportional = proportional_scaled; dc->writeback_rate_integral_scaled = integral_scaled; - dc->writeback_rate_change = new_rate - dc->writeback_rate.rate; - dc->writeback_rate.rate = new_rate; + dc->writeback_rate_change = new_rate - + atomic_long_read(&dc->writeback_rate.rate); + atomic_long_set(&dc->writeback_rate.rate, new_rate); dc->writeback_rate_target = target; } +static bool set_at_max_writeback_rate(struct cache_set *c, + struct cached_dev *dc) +{ + /* + * Idle_counter is increased everytime when update_writeback_rate() is + * called. If all backing devices attached to the same cache set have + * identical dc->writeback_rate_update_seconds values, it is about 6 + * rounds of update_writeback_rate() on each backing device before + * c->at_max_writeback_rate is set to 1, and then max wrteback rate set + * to each dc->writeback_rate.rate. + * In order to avoid extra locking cost for counting exact dirty cached + * devices number, c->attached_dev_nr is used to calculate the idle + * throushold. It might be bigger if not all cached device are in write- + * back mode, but it still works well with limited extra rounds of + * update_writeback_rate(). + */ + if (atomic_inc_return(&c->idle_counter) < + atomic_read(&c->attached_dev_nr) * 6) + return false; + + if (atomic_read(&c->at_max_writeback_rate) != 1) + atomic_set(&c->at_max_writeback_rate, 1); + + atomic_long_set(&dc->writeback_rate.rate, INT_MAX); + + /* keep writeback_rate_target as existing value */ + dc->writeback_rate_proportional = 0; + dc->writeback_rate_integral_scaled = 0; + dc->writeback_rate_change = 0; + + /* + * Check c->idle_counter and c->at_max_writeback_rate agagain in case + * new I/O arrives during before set_at_max_writeback_rate() returns. + * Then the writeback rate is set to 1, and its new value should be + * decided via __update_writeback_rate(). + */ + if ((atomic_read(&c->idle_counter) < + atomic_read(&c->attached_dev_nr) * 6) || + !atomic_read(&c->at_max_writeback_rate)) + return false; + + return true; +} + static void update_writeback_rate(struct work_struct *work) { struct cached_dev *dc = container_of(to_delayed_work(work), @@ -136,13 +181,20 @@ static void update_writeback_rate(struct work_struct *work) return; } - down_read(&dc->writeback_lock); - - if (atomic_read(&dc->has_dirty) && - dc->writeback_percent) - __update_writeback_rate(dc); + if (atomic_read(&dc->has_dirty) && dc->writeback_percent) { + /* + * If the whole cache set is idle, set_at_max_writeback_rate() + * will set writeback rate to a max number. Then it is + * unncessary to update writeback rate for an idle cache set + * in maximum writeback rate number(s). + */ + if (!set_at_max_writeback_rate(c, dc)) { + down_read(&dc->writeback_lock); + __update_writeback_rate(dc); + up_read(&dc->writeback_lock); + } + } - up_read(&dc->writeback_lock); /* * CACHE_SET_IO_DISABLE might be set via sysfs interface, @@ -422,27 +474,6 @@ static void read_dirty(struct cached_dev *dc) delay = writeback_delay(dc, size); - /* If the control system would wait for at least half a - * second, and there's been no reqs hitting the backing disk - * for awhile: use an alternate mode where we have at most - * one contiguous set of writebacks in flight at a time. If - * someone wants to do IO it will be quick, as it will only - * have to contend with one operation in flight, and we'll - * be round-tripping data to the backing disk as quickly as - * it can accept it. - */ - if (delay >= HZ / 2) { - /* 3 means at least 1.5 seconds, up to 7.5 if we - * have slowed way down. - */ - if (atomic_inc_return(&dc->backing_idle) >= 3) { - /* Wait for current I/Os to finish */ - closure_sync(&cl); - /* And immediately launch a new set. */ - delay = 0; - } - } - while (!kthread_should_stop() && !test_bit(CACHE_SET_IO_DISABLE, &dc->disk.c->flags) && delay) { @@ -476,6 +507,9 @@ void bcache_dev_sectors_dirty_add(struct cache_set *c, unsigned inode, if (!d) return; + if (UUID_FLASH_ONLY(&c->uuids[inode])) + atomic_long_add(nr_sectors, &c->flash_dev_dirty_sectors); + stripe = offset_to_stripe(d, offset); stripe_offset = offset & (d->stripe_size - 1); @@ -673,10 +707,14 @@ static int bch_writeback_thread(void *arg) } /* Init */ +#define INIT_KEYS_EACH_TIME 500000 +#define INIT_KEYS_SLEEP_MS 100 struct sectors_dirty_init { struct btree_op op; unsigned inode; + size_t count; + struct bkey start; }; static int sectors_dirty_init_fn(struct btree_op *_op, struct btree *b, @@ -691,18 +729,37 @@ static int sectors_dirty_init_fn(struct btree_op *_op, struct btree *b, bcache_dev_sectors_dirty_add(b->c, KEY_INODE(k), KEY_START(k), KEY_SIZE(k)); + op->count++; + if (atomic_read(&b->c->search_inflight) && + !(op->count % INIT_KEYS_EACH_TIME)) { + bkey_copy_key(&op->start, k); + return -EAGAIN; + } + return MAP_CONTINUE; } void bch_sectors_dirty_init(struct bcache_device *d) { struct sectors_dirty_init op; + int ret; bch_btree_op_init(&op.op, -1); op.inode = d->id; - - bch_btree_map_keys(&op.op, d->c, &KEY(op.inode, 0, 0), - sectors_dirty_init_fn, 0); + op.count = 0; + op.start = KEY(op.inode, 0, 0); + + do { + ret = bch_btree_map_keys(&op.op, d->c, &op.start, + sectors_dirty_init_fn, 0); + if (ret == -EAGAIN) + schedule_timeout_interruptible( + msecs_to_jiffies(INIT_KEYS_SLEEP_MS)); + else if (ret < 0) { + pr_warn("sectors dirty init failed, ret=%d!", ret); + break; + } + } while (ret == -EAGAIN); } void bch_cached_dev_writeback_init(struct cached_dev *dc) @@ -715,7 +772,7 @@ void bch_cached_dev_writeback_init(struct cached_dev *dc) dc->writeback_running = true; dc->writeback_percent = 10; dc->writeback_delay = 30; - dc->writeback_rate.rate = 1024; + atomic_long_set(&dc->writeback_rate.rate, 1024); dc->writeback_rate_minimum = 8; dc->writeback_rate_update_seconds = WRITEBACK_RATE_UPDATE_SECS_DEFAULT; diff --git a/drivers/md/bcache/writeback.h b/drivers/md/bcache/writeback.h index 610fb01de629..3745d7004c47 100644 --- a/drivers/md/bcache/writeback.h +++ b/drivers/md/bcache/writeback.h @@ -28,25 +28,6 @@ static inline uint64_t bcache_dev_sectors_dirty(struct bcache_device *d) return ret; } -static inline uint64_t bcache_flash_devs_sectors_dirty(struct cache_set *c) -{ - uint64_t i, ret = 0; - - mutex_lock(&bch_register_lock); - - for (i = 0; i < c->devices_max_used; i++) { - struct bcache_device *d = c->devices[i]; - - if (!d || !UUID_FLASH_ONLY(&c->uuids[i])) - continue; - ret += bcache_dev_sectors_dirty(d); - } - - mutex_unlock(&bch_register_lock); - - return ret; -} - static inline unsigned offset_to_stripe(struct bcache_device *d, uint64_t offset) { diff --git a/drivers/md/dm.c b/drivers/md/dm.c index b0dd7027848b..20f7e4ef5342 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -609,7 +609,8 @@ static void start_io_acct(struct dm_io *io) io->start_time = jiffies; - generic_start_io_acct(md->queue, rw, bio_sectors(bio), &dm_disk(md)->part0); + generic_start_io_acct(md->queue, bio_op(bio), bio_sectors(bio), + &dm_disk(md)->part0); atomic_set(&dm_disk(md)->part0.in_flight[rw], atomic_inc_return(&md->pending[rw])); @@ -628,7 +629,8 @@ static void end_io_acct(struct dm_io *io) int pending; int rw = bio_data_dir(bio); - generic_end_io_acct(md->queue, rw, &dm_disk(md)->part0, io->start_time); + generic_end_io_acct(md->queue, bio_op(bio), &dm_disk(md)->part0, + io->start_time); if (unlikely(dm_stats_used(&md->stats))) dm_stats_account_io(&md->stats, bio_data_dir(bio), diff --git a/drivers/md/md.c b/drivers/md/md.c index 994aed2f9dff..cb4eb5faa519 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -204,10 +204,6 @@ static int start_readonly; */ static bool create_on_open = true; -/* bio_clone_mddev - * like bio_clone_bioset, but with a local bio set - */ - struct bio *bio_alloc_mddev(gfp_t gfp_mask, int nr_iovecs, struct mddev *mddev) { @@ -335,6 +331,7 @@ EXPORT_SYMBOL(md_handle_request); static blk_qc_t md_make_request(struct request_queue *q, struct bio *bio) { const int rw = bio_data_dir(bio); + const int sgrp = op_stat_group(bio_op(bio)); struct mddev *mddev = q->queuedata; unsigned int sectors; int cpu; @@ -363,8 +360,8 @@ static blk_qc_t md_make_request(struct request_queue *q, struct bio *bio) md_handle_request(mddev, bio); cpu = part_stat_lock(); - part_stat_inc(cpu, &mddev->gendisk->part0, ios[rw]); - part_stat_add(cpu, &mddev->gendisk->part0, sectors[rw], sectors); + part_stat_inc(cpu, &mddev->gendisk->part0, ios[sgrp]); + part_stat_add(cpu, &mddev->gendisk->part0, sectors[sgrp], sectors); part_stat_unlock(); return BLK_QC_T_NONE; @@ -8046,8 +8043,7 @@ static int is_mddev_idle(struct mddev *mddev, int init) rcu_read_lock(); rdev_for_each_rcu(rdev, mddev) { struct gendisk *disk = rdev->bdev->bd_contains->bd_disk; - curr_events = (int)part_stat_read(&disk->part0, sectors[0]) + - (int)part_stat_read(&disk->part0, sectors[1]) - + curr_events = (int)part_stat_read_accum(&disk->part0, sectors) - atomic_read(&disk->sync_io); /* sync IO will cause sync_io to increase before the disk_stats * as sync_io is counted when a request starts, and diff --git a/drivers/nvdimm/btt.c b/drivers/nvdimm/btt.c index 85de8053aa34..0360c015f658 100644 --- a/drivers/nvdimm/btt.c +++ b/drivers/nvdimm/btt.c @@ -1423,11 +1423,11 @@ static int btt_write_pg(struct btt *btt, struct bio_integrity_payload *bip, static int btt_do_bvec(struct btt *btt, struct bio_integrity_payload *bip, struct page *page, unsigned int len, unsigned int off, - bool is_write, sector_t sector) + unsigned int op, sector_t sector) { int ret; - if (!is_write) { + if (!op_is_write(op)) { ret = btt_read_pg(btt, bip, page, off, sector, len); flush_dcache_page(page); } else { @@ -1464,7 +1464,7 @@ static blk_qc_t btt_make_request(struct request_queue *q, struct bio *bio) } err = btt_do_bvec(btt, bip, bvec.bv_page, len, bvec.bv_offset, - op_is_write(bio_op(bio)), iter.bi_sector); + bio_op(bio), iter.bi_sector); if (err) { dev_err(&btt->nd_btt->dev, "io error in %s sector %lld, len %d,\n", @@ -1483,16 +1483,16 @@ static blk_qc_t btt_make_request(struct request_queue *q, struct bio *bio) } static int btt_rw_page(struct block_device *bdev, sector_t sector, - struct page *page, bool is_write) + struct page *page, unsigned int op) { struct btt *btt = bdev->bd_disk->private_data; int rc; unsigned int len; len = hpage_nr_pages(page) * PAGE_SIZE; - rc = btt_do_bvec(btt, NULL, page, len, 0, is_write, sector); + rc = btt_do_bvec(btt, NULL, page, len, 0, op, sector); if (rc == 0) - page_endio(page, is_write, 0); + page_endio(page, op_is_write(op), 0); return rc; } diff --git a/drivers/nvdimm/nd.h b/drivers/nvdimm/nd.h index 32e0364b48b9..6ee7fd7e4bbd 100644 --- a/drivers/nvdimm/nd.h +++ b/drivers/nvdimm/nd.h @@ -396,16 +396,15 @@ static inline bool nd_iostat_start(struct bio *bio, unsigned long *start) return false; *start = jiffies; - generic_start_io_acct(disk->queue, bio_data_dir(bio), - bio_sectors(bio), &disk->part0); + generic_start_io_acct(disk->queue, bio_op(bio), bio_sectors(bio), + &disk->part0); return true; } static inline void nd_iostat_end(struct bio *bio, unsigned long start) { struct gendisk *disk = bio->bi_disk; - generic_end_io_acct(disk->queue, bio_data_dir(bio), &disk->part0, - start); + generic_end_io_acct(disk->queue, bio_op(bio), &disk->part0, start); } static inline bool is_bad_pmem(struct badblocks *bb, sector_t sector, unsigned int len) diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c index 8b1fd7f1a224..dd17acd8fe68 100644 --- a/drivers/nvdimm/pmem.c +++ b/drivers/nvdimm/pmem.c @@ -120,7 +120,7 @@ static blk_status_t read_pmem(struct page *page, unsigned int off, } static blk_status_t pmem_do_bvec(struct pmem_device *pmem, struct page *page, - unsigned int len, unsigned int off, bool is_write, + unsigned int len, unsigned int off, unsigned int op, sector_t sector) { blk_status_t rc = BLK_STS_OK; @@ -131,7 +131,7 @@ static blk_status_t pmem_do_bvec(struct pmem_device *pmem, struct page *page, if (unlikely(is_bad_pmem(&pmem->bb, sector, len))) bad_pmem = true; - if (!is_write) { + if (!op_is_write(op)) { if (unlikely(bad_pmem)) rc = BLK_STS_IOERR; else { @@ -180,8 +180,7 @@ static blk_qc_t pmem_make_request(struct request_queue *q, struct bio *bio) do_acct = nd_iostat_start(bio, &start); bio_for_each_segment(bvec, bio, iter) { rc = pmem_do_bvec(pmem, bvec.bv_page, bvec.bv_len, - bvec.bv_offset, op_is_write(bio_op(bio)), - iter.bi_sector); + bvec.bv_offset, bio_op(bio), iter.bi_sector); if (rc) { bio->bi_status = rc; break; @@ -198,13 +197,13 @@ static blk_qc_t pmem_make_request(struct request_queue *q, struct bio *bio) } static int pmem_rw_page(struct block_device *bdev, sector_t sector, - struct page *page, bool is_write) + struct page *page, unsigned int op) { struct pmem_device *pmem = bdev->bd_queue->queuedata; blk_status_t rc; rc = pmem_do_bvec(pmem, page, hpage_nr_pages(page) * PAGE_SIZE, - 0, is_write, sector); + 0, op, sector); /* * The ->rw_page interface is subtle and tricky. The core @@ -213,7 +212,7 @@ static int pmem_rw_page(struct block_device *bdev, sector_t sector, * caused by double completion. */ if (rc == 0) - page_endio(page, is_write, 0); + page_endio(page, op_is_write(op), 0); return blk_status_to_errno(rc); } diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index bf65501e6ed6..dd8ec1dd9219 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -252,7 +252,8 @@ void nvme_complete_rq(struct request *req) trace_nvme_complete_rq(req); if (unlikely(status != BLK_STS_OK && nvme_req_needs_retry(req))) { - if (nvme_req_needs_failover(req, status)) { + if ((req->cmd_flags & REQ_NVME_MPATH) && + blk_path_error(status)) { nvme_failover_req(req); return; } @@ -617,6 +618,8 @@ static inline blk_status_t nvme_setup_rw(struct nvme_ns *ns, if (WARN_ON_ONCE(!nvme_ns_has_pi(ns))) return BLK_STS_NOTSUPP; control |= NVME_RW_PRINFO_PRACT; + } else if (req_op(req) == REQ_OP_WRITE) { + t10_pi_prepare(req, ns->pi_type); } switch (ns->pi_type) { @@ -627,8 +630,7 @@ static inline blk_status_t nvme_setup_rw(struct nvme_ns *ns, case NVME_NS_DPS_PI_TYPE2: control |= NVME_RW_PRINFO_PRCHK_GUARD | NVME_RW_PRINFO_PRCHK_REF; - cmnd->rw.reftag = cpu_to_le32( - nvme_block_nr(ns, blk_rq_pos(req))); + cmnd->rw.reftag = cpu_to_le32(t10_pi_ref_tag(req)); break; } } @@ -638,6 +640,22 @@ static inline blk_status_t nvme_setup_rw(struct nvme_ns *ns, return 0; } +void nvme_cleanup_cmd(struct request *req) +{ + if (blk_integrity_rq(req) && req_op(req) == REQ_OP_READ && + nvme_req(req)->status == 0) { + struct nvme_ns *ns = req->rq_disk->private_data; + + t10_pi_complete(req, ns->pi_type, + blk_rq_bytes(req) >> ns->lba_shift); + } + if (req->rq_flags & RQF_SPECIAL_PAYLOAD) { + kfree(page_address(req->special_vec.bv_page) + + req->special_vec.bv_offset); + } +} +EXPORT_SYMBOL_GPL(nvme_cleanup_cmd); + blk_status_t nvme_setup_cmd(struct nvme_ns *ns, struct request *req, struct nvme_command *cmd) { @@ -668,10 +686,7 @@ blk_status_t nvme_setup_cmd(struct nvme_ns *ns, struct request *req, } cmd->common.command_id = req->tag; - if (ns) - trace_nvme_setup_nvm_cmd(req->q->id, cmd); - else - trace_nvme_setup_admin_cmd(cmd); + trace_nvme_setup_cmd(req, cmd); return ret; } EXPORT_SYMBOL_GPL(nvme_setup_cmd); @@ -864,9 +879,6 @@ static void nvme_start_keep_alive(struct nvme_ctrl *ctrl) if (unlikely(ctrl->kato == 0)) return; - INIT_DELAYED_WORK(&ctrl->ka_work, nvme_keep_alive_work); - memset(&ctrl->ka_cmd, 0, sizeof(ctrl->ka_cmd)); - ctrl->ka_cmd.common.opcode = nvme_admin_keep_alive; schedule_delayed_work(&ctrl->ka_work, ctrl->kato * HZ); } @@ -1056,7 +1068,7 @@ int nvme_set_queue_count(struct nvme_ctrl *ctrl, int *count) EXPORT_SYMBOL_GPL(nvme_set_queue_count); #define NVME_AEN_SUPPORTED \ - (NVME_AEN_CFG_NS_ATTR | NVME_AEN_CFG_FW_ACT) + (NVME_AEN_CFG_NS_ATTR | NVME_AEN_CFG_FW_ACT | NVME_AEN_CFG_ANA_CHANGE) static void nvme_enable_aen(struct nvme_ctrl *ctrl) { @@ -1472,6 +1484,12 @@ static void nvme_update_disk_info(struct gendisk *disk, set_capacity(disk, capacity); nvme_config_discard(ns); + + if (id->nsattr & (1 << 0)) + set_disk_ro(disk, true); + else + set_disk_ro(disk, false); + blk_mq_unfreeze_queue(disk->queue); } @@ -2270,21 +2288,16 @@ out_unlock: return ret; } -int nvme_get_log_ext(struct nvme_ctrl *ctrl, struct nvme_ns *ns, - u8 log_page, void *log, - size_t size, u64 offset) +int nvme_get_log(struct nvme_ctrl *ctrl, u32 nsid, u8 log_page, u8 lsp, + void *log, size_t size, u64 offset) { struct nvme_command c = { }; unsigned long dwlen = size / 4 - 1; c.get_log_page.opcode = nvme_admin_get_log_page; - - if (ns) - c.get_log_page.nsid = cpu_to_le32(ns->head->ns_id); - else - c.get_log_page.nsid = cpu_to_le32(NVME_NSID_ALL); - + c.get_log_page.nsid = cpu_to_le32(nsid); c.get_log_page.lid = log_page; + c.get_log_page.lsp = lsp; c.get_log_page.numdl = cpu_to_le16(dwlen & ((1 << 16) - 1)); c.get_log_page.numdu = cpu_to_le16(dwlen >> 16); c.get_log_page.lpol = cpu_to_le32(lower_32_bits(offset)); @@ -2293,12 +2306,6 @@ int nvme_get_log_ext(struct nvme_ctrl *ctrl, struct nvme_ns *ns, return nvme_submit_sync_cmd(ctrl->admin_q, &c, log, size); } -static int nvme_get_log(struct nvme_ctrl *ctrl, u8 log_page, void *log, - size_t size) -{ - return nvme_get_log_ext(ctrl, NULL, log_page, log, size, 0); -} - static int nvme_get_effects_log(struct nvme_ctrl *ctrl) { int ret; @@ -2309,8 +2316,8 @@ static int nvme_get_effects_log(struct nvme_ctrl *ctrl) if (!ctrl->effects) return 0; - ret = nvme_get_log(ctrl, NVME_LOG_CMD_EFFECTS, ctrl->effects, - sizeof(*ctrl->effects)); + ret = nvme_get_log(ctrl, NVME_NSID_ALL, NVME_LOG_CMD_EFFECTS, 0, + ctrl->effects, sizeof(*ctrl->effects), 0); if (ret) { kfree(ctrl->effects); ctrl->effects = NULL; @@ -2401,6 +2408,7 @@ int nvme_init_identify(struct nvme_ctrl *ctrl) nvme_set_queue_limits(ctrl, ctrl->admin_q); ctrl->sgls = le32_to_cpu(id->sgls); ctrl->kas = le16_to_cpu(id->kas); + ctrl->max_namespaces = le32_to_cpu(id->mnan); if (id->rtd3e) { /* us -> s */ @@ -2460,8 +2468,12 @@ int nvme_init_identify(struct nvme_ctrl *ctrl) ctrl->hmmaxd = le16_to_cpu(id->hmmaxd); } + ret = nvme_mpath_init(ctrl, id); kfree(id); + if (ret < 0) + return ret; + if (ctrl->apst_enabled && !prev_apst_enabled) dev_pm_qos_expose_latency_tolerance(ctrl->device); else if (!ctrl->apst_enabled && prev_apst_enabled) @@ -2680,6 +2692,10 @@ static struct attribute *nvme_ns_id_attrs[] = { &dev_attr_nguid.attr, &dev_attr_eui.attr, &dev_attr_nsid.attr, +#ifdef CONFIG_NVME_MULTIPATH + &dev_attr_ana_grpid.attr, + &dev_attr_ana_state.attr, +#endif NULL, }; @@ -2702,6 +2718,14 @@ static umode_t nvme_ns_id_attrs_are_visible(struct kobject *kobj, if (!memchr_inv(ids->eui64, 0, sizeof(ids->eui64))) return 0; } +#ifdef CONFIG_NVME_MULTIPATH + if (a == &dev_attr_ana_grpid.attr || a == &dev_attr_ana_state.attr) { + if (dev_to_disk(dev)->fops != &nvme_fops) /* per-path attr */ + return 0; + if (!nvme_ctrl_use_ana(nvme_get_ns_from_dev(dev)->ctrl)) + return 0; + } +#endif return a->mode; } @@ -3075,8 +3099,6 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid) nvme_get_ctrl(ctrl); - kfree(id); - device_add_disk(ctrl->device, ns->disk); if (sysfs_create_group(&disk_to_dev(ns->disk)->kobj, &nvme_ns_id_attr_group)) @@ -3086,8 +3108,10 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid) pr_warn("%s: failed to register lightnvm sysfs group for identification\n", ns->disk->disk_name); - nvme_mpath_add_disk(ns->head); + nvme_mpath_add_disk(ns, id); nvme_fault_inject_init(ns); + kfree(id); + return; out_unlink_ns: mutex_lock(&ctrl->subsys->lock); @@ -3229,7 +3253,8 @@ static void nvme_clear_changed_ns_log(struct nvme_ctrl *ctrl) * raced with us in reading the log page, which could cause us to miss * updates. */ - error = nvme_get_log(ctrl, NVME_LOG_CHANGED_NS, log, log_size); + error = nvme_get_log(ctrl, NVME_NSID_ALL, NVME_LOG_CHANGED_NS, 0, log, + log_size, 0); if (error) dev_warn(ctrl->device, "reading changed ns log failed: %d\n", error); @@ -3346,9 +3371,9 @@ static void nvme_get_fw_slot_info(struct nvme_ctrl *ctrl) if (!log) return; - if (nvme_get_log(ctrl, NVME_LOG_FW_SLOT, log, sizeof(*log))) - dev_warn(ctrl->device, - "Get FW SLOT INFO log error\n"); + if (nvme_get_log(ctrl, NVME_NSID_ALL, 0, NVME_LOG_FW_SLOT, log, + sizeof(*log), 0)) + dev_warn(ctrl->device, "Get FW SLOT INFO log error\n"); kfree(log); } @@ -3394,6 +3419,13 @@ static void nvme_handle_aen_notice(struct nvme_ctrl *ctrl, u32 result) case NVME_AER_NOTICE_FW_ACT_STARTING: queue_work(nvme_wq, &ctrl->fw_act_work); break; +#ifdef CONFIG_NVME_MULTIPATH + case NVME_AER_NOTICE_ANA: + if (!ctrl->ana_log_buf) + break; + queue_work(nvme_wq, &ctrl->ana_work); + break; +#endif default: dev_warn(ctrl->device, "async event result %08x\n", result); } @@ -3426,6 +3458,7 @@ EXPORT_SYMBOL_GPL(nvme_complete_async_event); void nvme_stop_ctrl(struct nvme_ctrl *ctrl) { + nvme_mpath_stop(ctrl); nvme_stop_keep_alive(ctrl); flush_work(&ctrl->async_event_work); flush_work(&ctrl->scan_work); @@ -3463,6 +3496,7 @@ static void nvme_free_ctrl(struct device *dev) ida_simple_remove(&nvme_instance_ida, ctrl->instance); kfree(ctrl->effects); + nvme_mpath_uninit(ctrl); if (subsys) { mutex_lock(&subsys->lock); @@ -3499,6 +3533,10 @@ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev, INIT_WORK(&ctrl->fw_act_work, nvme_fw_act_work); INIT_WORK(&ctrl->delete_work, nvme_delete_ctrl_work); + INIT_DELAYED_WORK(&ctrl->ka_work, nvme_keep_alive_work); + memset(&ctrl->ka_cmd, 0, sizeof(ctrl->ka_cmd)); + ctrl->ka_cmd.common.opcode = nvme_admin_keep_alive; + ret = ida_simple_get(&nvme_instance_ida, 0, 0, GFP_KERNEL); if (ret < 0) goto out; diff --git a/drivers/nvme/host/fabrics.c b/drivers/nvme/host/fabrics.c index f7efe5a58cc7..206d63cb1afc 100644 --- a/drivers/nvme/host/fabrics.c +++ b/drivers/nvme/host/fabrics.c @@ -474,7 +474,7 @@ EXPORT_SYMBOL_GPL(nvmf_connect_io_queue); bool nvmf_should_reconnect(struct nvme_ctrl *ctrl) { - if (ctrl->opts->max_reconnects != -1 && + if (ctrl->opts->max_reconnects == -1 || ctrl->nr_reconnects < ctrl->opts->max_reconnects) return true; diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c index 9bac912173ba..611e70cae754 100644 --- a/drivers/nvme/host/fc.c +++ b/drivers/nvme/host/fc.c @@ -1737,6 +1737,7 @@ nvme_fc_init_request(struct blk_mq_tag_set *set, struct request *rq, int queue_idx = (set == &ctrl->tag_set) ? hctx_idx + 1 : 0; struct nvme_fc_queue *queue = &ctrl->queues[queue_idx]; + nvme_req(rq)->ctrl = &ctrl->ctrl; return __nvme_fc_init_request(ctrl, queue, op, rq, queue->rqcnt++); } diff --git a/drivers/nvme/host/lightnvm.c b/drivers/nvme/host/lightnvm.c index 41279da799ed..6fe5923c95d4 100644 --- a/drivers/nvme/host/lightnvm.c +++ b/drivers/nvme/host/lightnvm.c @@ -414,12 +414,6 @@ static int nvme_nvm_setup_20(struct nvme_nvm_id20 *id, /* Set compacted version for upper layers */ geo->version = NVM_OCSSD_SPEC_20; - if (!(geo->major_ver_id == 2 && geo->minor_ver_id == 0)) { - pr_err("nvm: OCSSD version not supported (v%d.%d)\n", - geo->major_ver_id, geo->minor_ver_id); - return -EINVAL; - } - geo->num_ch = le16_to_cpu(id->num_grp); geo->num_lun = le16_to_cpu(id->num_pu); geo->all_luns = geo->num_ch * geo->num_lun; @@ -583,7 +577,13 @@ static int nvme_nvm_get_chk_meta(struct nvm_dev *ndev, struct ppa_addr ppa; size_t left = nchks * sizeof(struct nvme_nvm_chk_meta); size_t log_pos, offset, len; - int ret, i; + int ret, i, max_len; + + /* + * limit requests to maximum 256K to avoid issuing arbitrary large + * requests when the device does not specific a maximum transfer size. + */ + max_len = min_t(unsigned int, ctrl->max_hw_sectors << 9, 256 * 1024); /* Normalize lba address space to obtain log offset */ ppa.ppa = slba; @@ -596,10 +596,11 @@ static int nvme_nvm_get_chk_meta(struct nvm_dev *ndev, offset = log_pos * sizeof(struct nvme_nvm_chk_meta); while (left) { - len = min_t(unsigned int, left, ctrl->max_hw_sectors << 9); + len = min_t(unsigned int, left, max_len); - ret = nvme_get_log_ext(ctrl, ns, NVME_NVM_LOG_REPORT_CHUNK, - dev_meta, len, offset); + ret = nvme_get_log(ctrl, ns->head->ns_id, + NVME_NVM_LOG_REPORT_CHUNK, 0, dev_meta, len, + offset); if (ret) { dev_err(ctrl->device, "Get REPORT CHUNK log error\n"); break; @@ -662,12 +663,10 @@ static struct request *nvme_nvm_alloc_request(struct request_queue *q, rq->cmd_flags &= ~REQ_FAILFAST_DRIVER; - if (rqd->bio) { + if (rqd->bio) blk_init_request_from_bio(rq, rqd->bio); - } else { + else rq->ioprio = IOPRIO_PRIO_VALUE(IOPRIO_CLASS_BE, IOPRIO_NORM); - rq->__data_len = 0; - } return rq; } diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c index 1ffd3e8b13a1..5a9562881d4e 100644 --- a/drivers/nvme/host/multipath.c +++ b/drivers/nvme/host/multipath.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2017 Christoph Hellwig. + * Copyright (c) 2017-2018 Christoph Hellwig. * * This program is free software; you can redistribute it and/or modify it * under the terms and conditions of the GNU General Public License, @@ -20,6 +20,11 @@ module_param(multipath, bool, 0444); MODULE_PARM_DESC(multipath, "turn on native support for multiple controllers per subsystem"); +inline bool nvme_ctrl_use_ana(struct nvme_ctrl *ctrl) +{ + return multipath && ctrl->subsys && (ctrl->subsys->cmic & (1 << 3)); +} + /* * If multipathing is enabled we need to always use the subsystem instance * number for numbering our devices to avoid conflicts between subsystems that @@ -45,6 +50,7 @@ void nvme_set_disk_name(char *disk_name, struct nvme_ns *ns, void nvme_failover_req(struct request *req) { struct nvme_ns *ns = req->q->queuedata; + u16 status = nvme_req(req)->status; unsigned long flags; spin_lock_irqsave(&ns->head->requeue_lock, flags); @@ -52,15 +58,35 @@ void nvme_failover_req(struct request *req) spin_unlock_irqrestore(&ns->head->requeue_lock, flags); blk_mq_end_request(req, 0); - nvme_reset_ctrl(ns->ctrl); - kblockd_schedule_work(&ns->head->requeue_work); -} + switch (status & 0x7ff) { + case NVME_SC_ANA_TRANSITION: + case NVME_SC_ANA_INACCESSIBLE: + case NVME_SC_ANA_PERSISTENT_LOSS: + /* + * If we got back an ANA error we know the controller is alive, + * but not ready to serve this namespaces. The spec suggests + * we should update our general state here, but due to the fact + * that the admin and I/O queues are not serialized that is + * fundamentally racy. So instead just clear the current path, + * mark the the path as pending and kick of a re-read of the ANA + * log page ASAP. + */ + nvme_mpath_clear_current_path(ns); + if (ns->ctrl->ana_log_buf) { + set_bit(NVME_NS_ANA_PENDING, &ns->flags); + queue_work(nvme_wq, &ns->ctrl->ana_work); + } + break; + default: + /* + * Reset the controller for any non-ANA error as we don't know + * what caused the error. + */ + nvme_reset_ctrl(ns->ctrl); + break; + } -bool nvme_req_needs_failover(struct request *req, blk_status_t error) -{ - if (!(req->cmd_flags & REQ_NVME_MPATH)) - return false; - return blk_path_error(error); + kblockd_schedule_work(&ns->head->requeue_work); } void nvme_kick_requeue_lists(struct nvme_ctrl *ctrl) @@ -75,25 +101,51 @@ void nvme_kick_requeue_lists(struct nvme_ctrl *ctrl) up_read(&ctrl->namespaces_rwsem); } +static const char *nvme_ana_state_names[] = { + [0] = "invalid state", + [NVME_ANA_OPTIMIZED] = "optimized", + [NVME_ANA_NONOPTIMIZED] = "non-optimized", + [NVME_ANA_INACCESSIBLE] = "inaccessible", + [NVME_ANA_PERSISTENT_LOSS] = "persistent-loss", + [NVME_ANA_CHANGE] = "change", +}; + static struct nvme_ns *__nvme_find_path(struct nvme_ns_head *head) { - struct nvme_ns *ns; + struct nvme_ns *ns, *fallback = NULL; list_for_each_entry_rcu(ns, &head->list, siblings) { - if (ns->ctrl->state == NVME_CTRL_LIVE) { + if (ns->ctrl->state != NVME_CTRL_LIVE || + test_bit(NVME_NS_ANA_PENDING, &ns->flags)) + continue; + switch (ns->ana_state) { + case NVME_ANA_OPTIMIZED: rcu_assign_pointer(head->current_path, ns); return ns; + case NVME_ANA_NONOPTIMIZED: + fallback = ns; + break; + default: + break; } } - return NULL; + if (fallback) + rcu_assign_pointer(head->current_path, fallback); + return fallback; +} + +static inline bool nvme_path_is_optimized(struct nvme_ns *ns) +{ + return ns->ctrl->state == NVME_CTRL_LIVE && + ns->ana_state == NVME_ANA_OPTIMIZED; } inline struct nvme_ns *nvme_find_path(struct nvme_ns_head *head) { struct nvme_ns *ns = srcu_dereference(head->current_path, &head->srcu); - if (unlikely(!ns || ns->ctrl->state != NVME_CTRL_LIVE)) + if (unlikely(!ns || !nvme_path_is_optimized(ns))) ns = __nvme_find_path(head); return ns; } @@ -142,7 +194,7 @@ static bool nvme_ns_head_poll(struct request_queue *q, blk_qc_t qc) srcu_idx = srcu_read_lock(&head->srcu); ns = srcu_dereference(head->current_path, &head->srcu); - if (likely(ns && ns->ctrl->state == NVME_CTRL_LIVE)) + if (likely(ns && nvme_path_is_optimized(ns))) found = ns->queue->poll_fn(q, qc); srcu_read_unlock(&head->srcu, srcu_idx); return found; @@ -176,6 +228,7 @@ int nvme_mpath_alloc_disk(struct nvme_ctrl *ctrl, struct nvme_ns_head *head) struct request_queue *q; bool vwc = false; + mutex_init(&head->lock); bio_list_init(&head->requeue_list); spin_lock_init(&head->requeue_lock); INIT_WORK(&head->requeue_work, nvme_requeue_work); @@ -220,29 +273,232 @@ out: return -ENOMEM; } -void nvme_mpath_add_disk(struct nvme_ns_head *head) +static void nvme_mpath_set_live(struct nvme_ns *ns) { + struct nvme_ns_head *head = ns->head; + + lockdep_assert_held(&ns->head->lock); + if (!head->disk) return; - mutex_lock(&head->subsys->lock); if (!(head->disk->flags & GENHD_FL_UP)) { device_add_disk(&head->subsys->dev, head->disk); if (sysfs_create_group(&disk_to_dev(head->disk)->kobj, &nvme_ns_id_attr_group)) - pr_warn("%s: failed to create sysfs group for identification\n", - head->disk->disk_name); + dev_warn(&head->subsys->dev, + "failed to create id group.\n"); + } + + kblockd_schedule_work(&ns->head->requeue_work); +} + +static int nvme_parse_ana_log(struct nvme_ctrl *ctrl, void *data, + int (*cb)(struct nvme_ctrl *ctrl, struct nvme_ana_group_desc *, + void *)) +{ + void *base = ctrl->ana_log_buf; + size_t offset = sizeof(struct nvme_ana_rsp_hdr); + int error, i; + + lockdep_assert_held(&ctrl->ana_lock); + + for (i = 0; i < le16_to_cpu(ctrl->ana_log_buf->ngrps); i++) { + struct nvme_ana_group_desc *desc = base + offset; + u32 nr_nsids = le32_to_cpu(desc->nnsids); + size_t nsid_buf_size = nr_nsids * sizeof(__le32); + + if (WARN_ON_ONCE(desc->grpid == 0)) + return -EINVAL; + if (WARN_ON_ONCE(le32_to_cpu(desc->grpid) > ctrl->anagrpmax)) + return -EINVAL; + if (WARN_ON_ONCE(desc->state == 0)) + return -EINVAL; + if (WARN_ON_ONCE(desc->state > NVME_ANA_CHANGE)) + return -EINVAL; + + offset += sizeof(*desc); + if (WARN_ON_ONCE(offset > ctrl->ana_log_size - nsid_buf_size)) + return -EINVAL; + + error = cb(ctrl, desc, data); + if (error) + return error; + + offset += nsid_buf_size; + if (WARN_ON_ONCE(offset > ctrl->ana_log_size - sizeof(*desc))) + return -EINVAL; + } + + return 0; +} + +static inline bool nvme_state_is_live(enum nvme_ana_state state) +{ + return state == NVME_ANA_OPTIMIZED || state == NVME_ANA_NONOPTIMIZED; +} + +static void nvme_update_ns_ana_state(struct nvme_ana_group_desc *desc, + struct nvme_ns *ns) +{ + enum nvme_ana_state old; + + mutex_lock(&ns->head->lock); + old = ns->ana_state; + ns->ana_grpid = le32_to_cpu(desc->grpid); + ns->ana_state = desc->state; + clear_bit(NVME_NS_ANA_PENDING, &ns->flags); + + if (nvme_state_is_live(ns->ana_state) && !nvme_state_is_live(old)) + nvme_mpath_set_live(ns); + mutex_unlock(&ns->head->lock); +} + +static int nvme_update_ana_state(struct nvme_ctrl *ctrl, + struct nvme_ana_group_desc *desc, void *data) +{ + u32 nr_nsids = le32_to_cpu(desc->nnsids), n = 0; + unsigned *nr_change_groups = data; + struct nvme_ns *ns; + + dev_info(ctrl->device, "ANA group %d: %s.\n", + le32_to_cpu(desc->grpid), + nvme_ana_state_names[desc->state]); + + if (desc->state == NVME_ANA_CHANGE) + (*nr_change_groups)++; + + if (!nr_nsids) + return 0; + + down_write(&ctrl->namespaces_rwsem); + list_for_each_entry(ns, &ctrl->namespaces, list) { + if (ns->head->ns_id != le32_to_cpu(desc->nsids[n])) + continue; + nvme_update_ns_ana_state(desc, ns); + if (++n == nr_nsids) + break; + } + up_write(&ctrl->namespaces_rwsem); + WARN_ON_ONCE(n < nr_nsids); + return 0; +} + +static int nvme_read_ana_log(struct nvme_ctrl *ctrl, bool groups_only) +{ + u32 nr_change_groups = 0; + int error; + + mutex_lock(&ctrl->ana_lock); + error = nvme_get_log(ctrl, NVME_NSID_ALL, NVME_LOG_ANA, + groups_only ? NVME_ANA_LOG_RGO : 0, + ctrl->ana_log_buf, ctrl->ana_log_size, 0); + if (error) { + dev_warn(ctrl->device, "Failed to get ANA log: %d\n", error); + goto out_unlock; + } + + error = nvme_parse_ana_log(ctrl, &nr_change_groups, + nvme_update_ana_state); + if (error) + goto out_unlock; + + /* + * In theory we should have an ANATT timer per group as they might enter + * the change state at different times. But that is a lot of overhead + * just to protect against a target that keeps entering new changes + * states while never finishing previous ones. But we'll still + * eventually time out once all groups are in change state, so this + * isn't a big deal. + * + * We also double the ANATT value to provide some slack for transports + * or AEN processing overhead. + */ + if (nr_change_groups) + mod_timer(&ctrl->anatt_timer, ctrl->anatt * HZ * 2 + jiffies); + else + del_timer_sync(&ctrl->anatt_timer); +out_unlock: + mutex_unlock(&ctrl->ana_lock); + return error; +} + +static void nvme_ana_work(struct work_struct *work) +{ + struct nvme_ctrl *ctrl = container_of(work, struct nvme_ctrl, ana_work); + + nvme_read_ana_log(ctrl, false); +} + +static void nvme_anatt_timeout(struct timer_list *t) +{ + struct nvme_ctrl *ctrl = from_timer(ctrl, t, anatt_timer); + + dev_info(ctrl->device, "ANATT timeout, resetting controller.\n"); + nvme_reset_ctrl(ctrl); +} + +void nvme_mpath_stop(struct nvme_ctrl *ctrl) +{ + if (!nvme_ctrl_use_ana(ctrl)) + return; + del_timer_sync(&ctrl->anatt_timer); + cancel_work_sync(&ctrl->ana_work); +} + +static ssize_t ana_grpid_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + return sprintf(buf, "%d\n", nvme_get_ns_from_dev(dev)->ana_grpid); +} +DEVICE_ATTR_RO(ana_grpid); + +static ssize_t ana_state_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct nvme_ns *ns = nvme_get_ns_from_dev(dev); + + return sprintf(buf, "%s\n", nvme_ana_state_names[ns->ana_state]); +} +DEVICE_ATTR_RO(ana_state); + +static int nvme_set_ns_ana_state(struct nvme_ctrl *ctrl, + struct nvme_ana_group_desc *desc, void *data) +{ + struct nvme_ns *ns = data; + + if (ns->ana_grpid == le32_to_cpu(desc->grpid)) { + nvme_update_ns_ana_state(desc, ns); + return -ENXIO; /* just break out of the loop */ + } + + return 0; +} + +void nvme_mpath_add_disk(struct nvme_ns *ns, struct nvme_id_ns *id) +{ + if (nvme_ctrl_use_ana(ns->ctrl)) { + mutex_lock(&ns->ctrl->ana_lock); + ns->ana_grpid = le32_to_cpu(id->anagrpid); + nvme_parse_ana_log(ns->ctrl, ns, nvme_set_ns_ana_state); + mutex_unlock(&ns->ctrl->ana_lock); + } else { + mutex_lock(&ns->head->lock); + ns->ana_state = NVME_ANA_OPTIMIZED; + nvme_mpath_set_live(ns); + mutex_unlock(&ns->head->lock); } - mutex_unlock(&head->subsys->lock); } void nvme_mpath_remove_disk(struct nvme_ns_head *head) { if (!head->disk) return; - sysfs_remove_group(&disk_to_dev(head->disk)->kobj, - &nvme_ns_id_attr_group); - del_gendisk(head->disk); + if (head->disk->flags & GENHD_FL_UP) { + sysfs_remove_group(&disk_to_dev(head->disk)->kobj, + &nvme_ns_id_attr_group); + del_gendisk(head->disk); + } blk_set_queue_dying(head->disk->queue); /* make sure all pending bios are cleaned up */ kblockd_schedule_work(&head->requeue_work); @@ -250,3 +506,52 @@ void nvme_mpath_remove_disk(struct nvme_ns_head *head) blk_cleanup_queue(head->disk->queue); put_disk(head->disk); } + +int nvme_mpath_init(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id) +{ + int error; + + if (!nvme_ctrl_use_ana(ctrl)) + return 0; + + ctrl->anacap = id->anacap; + ctrl->anatt = id->anatt; + ctrl->nanagrpid = le32_to_cpu(id->nanagrpid); + ctrl->anagrpmax = le32_to_cpu(id->anagrpmax); + + mutex_init(&ctrl->ana_lock); + timer_setup(&ctrl->anatt_timer, nvme_anatt_timeout, 0); + ctrl->ana_log_size = sizeof(struct nvme_ana_rsp_hdr) + + ctrl->nanagrpid * sizeof(struct nvme_ana_group_desc); + if (!(ctrl->anacap & (1 << 6))) + ctrl->ana_log_size += ctrl->max_namespaces * sizeof(__le32); + + if (ctrl->ana_log_size > ctrl->max_hw_sectors << SECTOR_SHIFT) { + dev_err(ctrl->device, + "ANA log page size (%zd) larger than MDTS (%d).\n", + ctrl->ana_log_size, + ctrl->max_hw_sectors << SECTOR_SHIFT); + dev_err(ctrl->device, "disabling ANA support.\n"); + return 0; + } + + INIT_WORK(&ctrl->ana_work, nvme_ana_work); + ctrl->ana_log_buf = kmalloc(ctrl->ana_log_size, GFP_KERNEL); + if (!ctrl->ana_log_buf) + goto out; + + error = nvme_read_ana_log(ctrl, true); + if (error) + goto out_free_ana_log_buf; + return 0; +out_free_ana_log_buf: + kfree(ctrl->ana_log_buf); +out: + return -ENOMEM; +} + +void nvme_mpath_uninit(struct nvme_ctrl *ctrl) +{ + kfree(ctrl->ana_log_buf); +} + diff --git a/drivers/nvme/host/nvme.h b/drivers/nvme/host/nvme.h index 0c4a33df3b2f..bb4a2003c097 100644 --- a/drivers/nvme/host/nvme.h +++ b/drivers/nvme/host/nvme.h @@ -102,6 +102,7 @@ struct nvme_request { u8 retries; u8 flags; u16 status; + struct nvme_ctrl *ctrl; }; /* @@ -119,6 +120,13 @@ static inline struct nvme_request *nvme_req(struct request *req) return blk_mq_rq_to_pdu(req); } +static inline u16 nvme_req_qid(struct request *req) +{ + if (!req->rq_disk) + return 0; + return blk_mq_unique_tag_to_hwq(blk_mq_unique_tag(req)) + 1; +} + /* The below value is the specific amount of delay needed before checking * readiness in case of the PCI_DEVICE(0x1c58, 0x0003), which needs the * NVME_QUIRK_DELAY_BEFORE_CHK_RDY quirk enabled. The value (in ms) was @@ -175,6 +183,7 @@ struct nvme_ctrl { u16 oacs; u16 nssa; u16 nr_streams; + u32 max_namespaces; atomic_t abort_limit; u8 vwc; u32 vs; @@ -197,6 +206,19 @@ struct nvme_ctrl { struct work_struct fw_act_work; unsigned long events; +#ifdef CONFIG_NVME_MULTIPATH + /* asymmetric namespace access: */ + u8 anacap; + u8 anatt; + u32 anagrpmax; + u32 nanagrpid; + struct mutex ana_lock; + struct nvme_ana_rsp_hdr *ana_log_buf; + size_t ana_log_size; + struct timer_list anatt_timer; + struct work_struct ana_work; +#endif + /* Power saving configuration */ u64 ps_max_latency_us; bool apst_enabled; @@ -261,6 +283,7 @@ struct nvme_ns_head { struct bio_list requeue_list; spinlock_t requeue_lock; struct work_struct requeue_work; + struct mutex lock; #endif struct list_head list; struct srcu_struct srcu; @@ -287,6 +310,10 @@ struct nvme_ns { struct nvme_ctrl *ctrl; struct request_queue *queue; struct gendisk *disk; +#ifdef CONFIG_NVME_MULTIPATH + enum nvme_ana_state ana_state; + u32 ana_grpid; +#endif struct list_head siblings; struct nvm_dev *ndev; struct kref kref; @@ -299,8 +326,9 @@ struct nvme_ns { bool ext; u8 pi_type; unsigned long flags; -#define NVME_NS_REMOVING 0 -#define NVME_NS_DEAD 1 +#define NVME_NS_REMOVING 0 +#define NVME_NS_DEAD 1 +#define NVME_NS_ANA_PENDING 2 u16 noiob; #ifdef CONFIG_FAULT_INJECTION_DEBUG_FS @@ -356,14 +384,6 @@ static inline u64 nvme_block_nr(struct nvme_ns *ns, sector_t sector) return (sector >> (ns->lba_shift - 9)); } -static inline void nvme_cleanup_cmd(struct request *req) -{ - if (req->rq_flags & RQF_SPECIAL_PAYLOAD) { - kfree(page_address(req->special_vec.bv_page) + - req->special_vec.bv_offset); - } -} - static inline void nvme_end_request(struct request *req, __le16 status, union nvme_result result) { @@ -420,6 +440,7 @@ void nvme_start_freeze(struct nvme_ctrl *ctrl); #define NVME_QID_ANY -1 struct request *nvme_alloc_request(struct request_queue *q, struct nvme_command *cmd, blk_mq_req_flags_t flags, int qid); +void nvme_cleanup_cmd(struct request *req); blk_status_t nvme_setup_cmd(struct nvme_ns *ns, struct request *req, struct nvme_command *cmd); int nvme_submit_sync_cmd(struct request_queue *q, struct nvme_command *cmd, @@ -435,21 +456,24 @@ int nvme_reset_ctrl_sync(struct nvme_ctrl *ctrl); int nvme_delete_ctrl(struct nvme_ctrl *ctrl); int nvme_delete_ctrl_sync(struct nvme_ctrl *ctrl); -int nvme_get_log_ext(struct nvme_ctrl *ctrl, struct nvme_ns *ns, - u8 log_page, void *log, size_t size, u64 offset); +int nvme_get_log(struct nvme_ctrl *ctrl, u32 nsid, u8 log_page, u8 lsp, + void *log, size_t size, u64 offset); extern const struct attribute_group nvme_ns_id_attr_group; extern const struct block_device_operations nvme_ns_head_ops; #ifdef CONFIG_NVME_MULTIPATH +bool nvme_ctrl_use_ana(struct nvme_ctrl *ctrl); void nvme_set_disk_name(char *disk_name, struct nvme_ns *ns, struct nvme_ctrl *ctrl, int *flags); void nvme_failover_req(struct request *req); -bool nvme_req_needs_failover(struct request *req, blk_status_t error); void nvme_kick_requeue_lists(struct nvme_ctrl *ctrl); int nvme_mpath_alloc_disk(struct nvme_ctrl *ctrl,struct nvme_ns_head *head); -void nvme_mpath_add_disk(struct nvme_ns_head *head); +void nvme_mpath_add_disk(struct nvme_ns *ns, struct nvme_id_ns *id); void nvme_mpath_remove_disk(struct nvme_ns_head *head); +int nvme_mpath_init(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id); +void nvme_mpath_uninit(struct nvme_ctrl *ctrl); +void nvme_mpath_stop(struct nvme_ctrl *ctrl); static inline void nvme_mpath_clear_current_path(struct nvme_ns *ns) { @@ -468,7 +492,14 @@ static inline void nvme_mpath_check_last_path(struct nvme_ns *ns) kblockd_schedule_work(&head->requeue_work); } +extern struct device_attribute dev_attr_ana_grpid; +extern struct device_attribute dev_attr_ana_state; + #else +static inline bool nvme_ctrl_use_ana(struct nvme_ctrl *ctrl) +{ + return false; +} /* * Without the multipath code enabled, multiple controller per subsystems are * visible as devices and thus we cannot use the subsystem instance. @@ -482,11 +513,6 @@ static inline void nvme_set_disk_name(char *disk_name, struct nvme_ns *ns, static inline void nvme_failover_req(struct request *req) { } -static inline bool nvme_req_needs_failover(struct request *req, - blk_status_t error) -{ - return false; -} static inline void nvme_kick_requeue_lists(struct nvme_ctrl *ctrl) { } @@ -495,7 +521,8 @@ static inline int nvme_mpath_alloc_disk(struct nvme_ctrl *ctrl, { return 0; } -static inline void nvme_mpath_add_disk(struct nvme_ns_head *head) +static inline void nvme_mpath_add_disk(struct nvme_ns *ns, + struct nvme_id_ns *id) { } static inline void nvme_mpath_remove_disk(struct nvme_ns_head *head) @@ -507,6 +534,17 @@ static inline void nvme_mpath_clear_current_path(struct nvme_ns *ns) static inline void nvme_mpath_check_last_path(struct nvme_ns *ns) { } +static inline int nvme_mpath_init(struct nvme_ctrl *ctrl, + struct nvme_id_ctrl *id) +{ + return 0; +} +static inline void nvme_mpath_uninit(struct nvme_ctrl *ctrl) +{ +} +static inline void nvme_mpath_stop(struct nvme_ctrl *ctrl) +{ +} #endif /* CONFIG_NVME_MULTIPATH */ #ifdef CONFIG_NVM diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index ddd441b1516a..1b9951d2067e 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -418,6 +418,8 @@ static int nvme_init_request(struct blk_mq_tag_set *set, struct request *req, BUG_ON(!nvmeq); iod->nvmeq = nvmeq; + + nvme_req(req)->ctrl = &dev->ctrl; return 0; } @@ -535,73 +537,6 @@ static void nvme_free_iod(struct nvme_dev *dev, struct request *req) mempool_free(iod->sg, dev->iod_mempool); } -#ifdef CONFIG_BLK_DEV_INTEGRITY -static void nvme_dif_prep(u32 p, u32 v, struct t10_pi_tuple *pi) -{ - if (be32_to_cpu(pi->ref_tag) == v) - pi->ref_tag = cpu_to_be32(p); -} - -static void nvme_dif_complete(u32 p, u32 v, struct t10_pi_tuple *pi) -{ - if (be32_to_cpu(pi->ref_tag) == p) - pi->ref_tag = cpu_to_be32(v); -} - -/** - * nvme_dif_remap - remaps ref tags to bip seed and physical lba - * - * The virtual start sector is the one that was originally submitted by the - * block layer. Due to partitioning, MD/DM cloning, etc. the actual physical - * start sector may be different. Remap protection information to match the - * physical LBA on writes, and back to the original seed on reads. - * - * Type 0 and 3 do not have a ref tag, so no remapping required. - */ -static void nvme_dif_remap(struct request *req, - void (*dif_swap)(u32 p, u32 v, struct t10_pi_tuple *pi)) -{ - struct nvme_ns *ns = req->rq_disk->private_data; - struct bio_integrity_payload *bip; - struct t10_pi_tuple *pi; - void *p, *pmap; - u32 i, nlb, ts, phys, virt; - - if (!ns->pi_type || ns->pi_type == NVME_NS_DPS_PI_TYPE3) - return; - - bip = bio_integrity(req->bio); - if (!bip) - return; - - pmap = kmap_atomic(bip->bip_vec->bv_page) + bip->bip_vec->bv_offset; - - p = pmap; - virt = bip_get_seed(bip); - phys = nvme_block_nr(ns, blk_rq_pos(req)); - nlb = (blk_rq_bytes(req) >> ns->lba_shift); - ts = ns->disk->queue->integrity.tuple_size; - - for (i = 0; i < nlb; i++, virt++, phys++) { - pi = (struct t10_pi_tuple *)p; - dif_swap(phys, virt, pi); - p += ts; - } - kunmap_atomic(pmap); -} -#else /* CONFIG_BLK_DEV_INTEGRITY */ -static void nvme_dif_remap(struct request *req, - void (*dif_swap)(u32 p, u32 v, struct t10_pi_tuple *pi)) -{ -} -static void nvme_dif_prep(u32 p, u32 v, struct t10_pi_tuple *pi) -{ -} -static void nvme_dif_complete(u32 p, u32 v, struct t10_pi_tuple *pi) -{ -} -#endif - static void nvme_print_sgl(struct scatterlist *sgl, int nents) { int i; @@ -827,9 +762,6 @@ static blk_status_t nvme_map_data(struct nvme_dev *dev, struct request *req, if (blk_rq_map_integrity_sg(q, req->bio, &iod->meta_sg) != 1) goto out_unmap; - if (req_op(req) == REQ_OP_WRITE) - nvme_dif_remap(req, nvme_dif_prep); - if (!dma_map_sg(dev->dev, &iod->meta_sg, 1, dma_dir)) goto out_unmap; } @@ -852,11 +784,8 @@ static void nvme_unmap_data(struct nvme_dev *dev, struct request *req) if (iod->nents) { dma_unmap_sg(dev->dev, iod->sg, iod->nents, dma_dir); - if (blk_integrity_rq(req)) { - if (req_op(req) == REQ_OP_READ) - nvme_dif_remap(req, nvme_dif_complete); + if (blk_integrity_rq(req)) dma_unmap_sg(dev->dev, &iod->meta_sg, 1, dma_dir); - } } nvme_cleanup_cmd(req); diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index 66ec5985c9f3..0805fa6215ee 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -40,13 +40,14 @@ #define NVME_RDMA_MAX_SEGMENTS 256 -#define NVME_RDMA_MAX_INLINE_SEGMENTS 1 +#define NVME_RDMA_MAX_INLINE_SEGMENTS 4 struct nvme_rdma_device { struct ib_device *dev; struct ib_pd *pd; struct kref ref; struct list_head entry; + unsigned int num_inline_segments; }; struct nvme_rdma_qe { @@ -117,6 +118,7 @@ struct nvme_rdma_ctrl { struct sockaddr_storage src_addr; struct nvme_ctrl ctrl; + bool use_inline_data; }; static inline struct nvme_rdma_ctrl *to_rdma_ctrl(struct nvme_ctrl *ctrl) @@ -249,7 +251,7 @@ static int nvme_rdma_create_qp(struct nvme_rdma_queue *queue, const int factor) /* +1 for drain */ init_attr.cap.max_recv_wr = queue->queue_size + 1; init_attr.cap.max_recv_sge = 1; - init_attr.cap.max_send_sge = 1 + NVME_RDMA_MAX_INLINE_SEGMENTS; + init_attr.cap.max_send_sge = 1 + dev->num_inline_segments; init_attr.sq_sig_type = IB_SIGNAL_REQ_WR; init_attr.qp_type = IB_QPT_RC; init_attr.send_cq = queue->ib_cq; @@ -286,6 +288,7 @@ static int nvme_rdma_init_request(struct blk_mq_tag_set *set, struct ib_device *ibdev = dev->dev; int ret; + nvme_req(rq)->ctrl = &ctrl->ctrl; ret = nvme_rdma_alloc_qe(ibdev, &req->sqe, sizeof(struct nvme_command), DMA_TO_DEVICE); if (ret) @@ -374,6 +377,8 @@ nvme_rdma_find_get_device(struct rdma_cm_id *cm_id) goto out_free_pd; } + ndev->num_inline_segments = min(NVME_RDMA_MAX_INLINE_SEGMENTS, + ndev->dev->attrs.max_sge - 1); list_add(&ndev->entry, &device_list); out_unlock: mutex_unlock(&device_list_mutex); @@ -868,6 +873,31 @@ out_free_io_queues: return ret; } +static void nvme_rdma_teardown_admin_queue(struct nvme_rdma_ctrl *ctrl, + bool remove) +{ + blk_mq_quiesce_queue(ctrl->ctrl.admin_q); + nvme_rdma_stop_queue(&ctrl->queues[0]); + blk_mq_tagset_busy_iter(&ctrl->admin_tag_set, nvme_cancel_request, + &ctrl->ctrl); + blk_mq_unquiesce_queue(ctrl->ctrl.admin_q); + nvme_rdma_destroy_admin_queue(ctrl, remove); +} + +static void nvme_rdma_teardown_io_queues(struct nvme_rdma_ctrl *ctrl, + bool remove) +{ + if (ctrl->ctrl.queue_count > 1) { + nvme_stop_queues(&ctrl->ctrl); + nvme_rdma_stop_io_queues(ctrl); + blk_mq_tagset_busy_iter(&ctrl->tag_set, nvme_cancel_request, + &ctrl->ctrl); + if (remove) + nvme_start_queues(&ctrl->ctrl); + nvme_rdma_destroy_io_queues(ctrl, remove); + } +} + static void nvme_rdma_stop_ctrl(struct nvme_ctrl *nctrl) { struct nvme_rdma_ctrl *ctrl = to_rdma_ctrl(nctrl); @@ -912,21 +942,44 @@ static void nvme_rdma_reconnect_or_remove(struct nvme_rdma_ctrl *ctrl) } } -static void nvme_rdma_reconnect_ctrl_work(struct work_struct *work) +static int nvme_rdma_setup_ctrl(struct nvme_rdma_ctrl *ctrl, bool new) { - struct nvme_rdma_ctrl *ctrl = container_of(to_delayed_work(work), - struct nvme_rdma_ctrl, reconnect_work); + int ret = -EINVAL; bool changed; - int ret; - ++ctrl->ctrl.nr_reconnects; - - ret = nvme_rdma_configure_admin_queue(ctrl, false); + ret = nvme_rdma_configure_admin_queue(ctrl, new); if (ret) - goto requeue; + return ret; + + if (ctrl->ctrl.icdoff) { + dev_err(ctrl->ctrl.device, "icdoff is not supported!\n"); + goto destroy_admin; + } + + if (!(ctrl->ctrl.sgls & (1 << 2))) { + dev_err(ctrl->ctrl.device, + "Mandatory keyed sgls are not supported!\n"); + goto destroy_admin; + } + + if (ctrl->ctrl.opts->queue_size > ctrl->ctrl.sqsize + 1) { + dev_warn(ctrl->ctrl.device, + "queue_size %zu > ctrl sqsize %u, clamping down\n", + ctrl->ctrl.opts->queue_size, ctrl->ctrl.sqsize + 1); + } + + if (ctrl->ctrl.sqsize + 1 > ctrl->ctrl.maxcmd) { + dev_warn(ctrl->ctrl.device, + "sqsize %u > ctrl maxcmd %u, clamping down\n", + ctrl->ctrl.sqsize + 1, ctrl->ctrl.maxcmd); + ctrl->ctrl.sqsize = ctrl->ctrl.maxcmd - 1; + } + + if (ctrl->ctrl.sgls & (1 << 20)) + ctrl->use_inline_data = true; if (ctrl->ctrl.queue_count > 1) { - ret = nvme_rdma_configure_io_queues(ctrl, false); + ret = nvme_rdma_configure_io_queues(ctrl, new); if (ret) goto destroy_admin; } @@ -935,10 +988,31 @@ static void nvme_rdma_reconnect_ctrl_work(struct work_struct *work) if (!changed) { /* state change failure is ok if we're in DELETING state */ WARN_ON_ONCE(ctrl->ctrl.state != NVME_CTRL_DELETING); - return; + ret = -EINVAL; + goto destroy_io; } nvme_start_ctrl(&ctrl->ctrl); + return 0; + +destroy_io: + if (ctrl->ctrl.queue_count > 1) + nvme_rdma_destroy_io_queues(ctrl, new); +destroy_admin: + nvme_rdma_stop_queue(&ctrl->queues[0]); + nvme_rdma_destroy_admin_queue(ctrl, new); + return ret; +} + +static void nvme_rdma_reconnect_ctrl_work(struct work_struct *work) +{ + struct nvme_rdma_ctrl *ctrl = container_of(to_delayed_work(work), + struct nvme_rdma_ctrl, reconnect_work); + + ++ctrl->ctrl.nr_reconnects; + + if (nvme_rdma_setup_ctrl(ctrl, false)) + goto requeue; dev_info(ctrl->ctrl.device, "Successfully reconnected (%d attempts)\n", ctrl->ctrl.nr_reconnects); @@ -947,9 +1021,6 @@ static void nvme_rdma_reconnect_ctrl_work(struct work_struct *work) return; -destroy_admin: - nvme_rdma_stop_queue(&ctrl->queues[0]); - nvme_rdma_destroy_admin_queue(ctrl, false); requeue: dev_info(ctrl->ctrl.device, "Failed reconnect attempt %d\n", ctrl->ctrl.nr_reconnects); @@ -962,27 +1033,9 @@ static void nvme_rdma_error_recovery_work(struct work_struct *work) struct nvme_rdma_ctrl, err_work); nvme_stop_keep_alive(&ctrl->ctrl); - - if (ctrl->ctrl.queue_count > 1) { - nvme_stop_queues(&ctrl->ctrl); - nvme_rdma_stop_io_queues(ctrl); - blk_mq_tagset_busy_iter(&ctrl->tag_set, - nvme_cancel_request, &ctrl->ctrl); - nvme_rdma_destroy_io_queues(ctrl, false); - } - - blk_mq_quiesce_queue(ctrl->ctrl.admin_q); - nvme_rdma_stop_queue(&ctrl->queues[0]); - blk_mq_tagset_busy_iter(&ctrl->admin_tag_set, - nvme_cancel_request, &ctrl->ctrl); - nvme_rdma_destroy_admin_queue(ctrl, false); - - /* - * queues are not a live anymore, so restart the queues to fail fast - * new IO - */ - blk_mq_unquiesce_queue(ctrl->ctrl.admin_q); + nvme_rdma_teardown_io_queues(ctrl, false); nvme_start_queues(&ctrl->ctrl); + nvme_rdma_teardown_admin_queue(ctrl, false); if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING)) { /* state change failure is ok if we're in DELETING state */ @@ -1090,19 +1143,27 @@ static int nvme_rdma_set_sg_null(struct nvme_command *c) } static int nvme_rdma_map_sg_inline(struct nvme_rdma_queue *queue, - struct nvme_rdma_request *req, struct nvme_command *c) + struct nvme_rdma_request *req, struct nvme_command *c, + int count) { struct nvme_sgl_desc *sg = &c->common.dptr.sgl; + struct scatterlist *sgl = req->sg_table.sgl; + struct ib_sge *sge = &req->sge[1]; + u32 len = 0; + int i; - req->sge[1].addr = sg_dma_address(req->sg_table.sgl); - req->sge[1].length = sg_dma_len(req->sg_table.sgl); - req->sge[1].lkey = queue->device->pd->local_dma_lkey; + for (i = 0; i < count; i++, sgl++, sge++) { + sge->addr = sg_dma_address(sgl); + sge->length = sg_dma_len(sgl); + sge->lkey = queue->device->pd->local_dma_lkey; + len += sge->length; + } sg->addr = cpu_to_le64(queue->ctrl->ctrl.icdoff); - sg->length = cpu_to_le32(sg_dma_len(req->sg_table.sgl)); + sg->length = cpu_to_le32(len); sg->type = (NVME_SGL_FMT_DATA_DESC << 4) | NVME_SGL_FMT_OFFSET; - req->num_sge++; + req->num_sge += count; return 0; } @@ -1195,15 +1256,16 @@ static int nvme_rdma_map_data(struct nvme_rdma_queue *queue, goto out_free_table; } - if (count == 1) { + if (count <= dev->num_inline_segments) { if (rq_data_dir(rq) == WRITE && nvme_rdma_queue_idx(queue) && + queue->ctrl->use_inline_data && blk_rq_payload_bytes(rq) <= nvme_rdma_inline_data_size(queue)) { - ret = nvme_rdma_map_sg_inline(queue, req, c); + ret = nvme_rdma_map_sg_inline(queue, req, c, count); goto out; } - if (dev->pd->flags & IB_PD_UNSAFE_GLOBAL_RKEY) { + if (count == 1 && dev->pd->flags & IB_PD_UNSAFE_GLOBAL_RKEY) { ret = nvme_rdma_map_sg_single(queue, req, c); goto out; } @@ -1574,6 +1636,7 @@ static int nvme_rdma_cm_handler(struct rdma_cm_id *cm_id, case RDMA_CM_EVENT_CONNECT_ERROR: case RDMA_CM_EVENT_UNREACHABLE: nvme_rdma_destroy_queue_ib(queue); + /* fall through */ case RDMA_CM_EVENT_ADDR_ERROR: dev_dbg(queue->ctrl->ctrl.device, "CM error event %d\n", ev->event); @@ -1736,25 +1799,12 @@ static const struct blk_mq_ops nvme_rdma_admin_mq_ops = { static void nvme_rdma_shutdown_ctrl(struct nvme_rdma_ctrl *ctrl, bool shutdown) { - if (ctrl->ctrl.queue_count > 1) { - nvme_stop_queues(&ctrl->ctrl); - nvme_rdma_stop_io_queues(ctrl); - blk_mq_tagset_busy_iter(&ctrl->tag_set, - nvme_cancel_request, &ctrl->ctrl); - nvme_rdma_destroy_io_queues(ctrl, shutdown); - } - + nvme_rdma_teardown_io_queues(ctrl, shutdown); if (shutdown) nvme_shutdown_ctrl(&ctrl->ctrl); else nvme_disable_ctrl(&ctrl->ctrl, ctrl->ctrl.cap); - - blk_mq_quiesce_queue(ctrl->ctrl.admin_q); - nvme_rdma_stop_queue(&ctrl->queues[0]); - blk_mq_tagset_busy_iter(&ctrl->admin_tag_set, - nvme_cancel_request, &ctrl->ctrl); - blk_mq_unquiesce_queue(ctrl->ctrl.admin_q); - nvme_rdma_destroy_admin_queue(ctrl, shutdown); + nvme_rdma_teardown_admin_queue(ctrl, shutdown); } static void nvme_rdma_delete_ctrl(struct nvme_ctrl *ctrl) @@ -1766,8 +1816,6 @@ static void nvme_rdma_reset_ctrl_work(struct work_struct *work) { struct nvme_rdma_ctrl *ctrl = container_of(work, struct nvme_rdma_ctrl, ctrl.reset_work); - int ret; - bool changed; nvme_stop_ctrl(&ctrl->ctrl); nvme_rdma_shutdown_ctrl(ctrl, false); @@ -1778,25 +1826,9 @@ static void nvme_rdma_reset_ctrl_work(struct work_struct *work) return; } - ret = nvme_rdma_configure_admin_queue(ctrl, false); - if (ret) + if (nvme_rdma_setup_ctrl(ctrl, false)) goto out_fail; - if (ctrl->ctrl.queue_count > 1) { - ret = nvme_rdma_configure_io_queues(ctrl, false); - if (ret) - goto out_fail; - } - - changed = nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_LIVE); - if (!changed) { - /* state change failure is ok if we're in DELETING state */ - WARN_ON_ONCE(ctrl->ctrl.state != NVME_CTRL_DELETING); - return; - } - - nvme_start_ctrl(&ctrl->ctrl); - return; out_fail: @@ -1959,49 +1991,10 @@ static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev, changed = nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING); WARN_ON_ONCE(!changed); - ret = nvme_rdma_configure_admin_queue(ctrl, true); + ret = nvme_rdma_setup_ctrl(ctrl, true); if (ret) goto out_uninit_ctrl; - /* sanity check icdoff */ - if (ctrl->ctrl.icdoff) { - dev_err(ctrl->ctrl.device, "icdoff is not supported!\n"); - ret = -EINVAL; - goto out_remove_admin_queue; - } - - /* sanity check keyed sgls */ - if (!(ctrl->ctrl.sgls & (1 << 2))) { - dev_err(ctrl->ctrl.device, - "Mandatory keyed sgls are not supported!\n"); - ret = -EINVAL; - goto out_remove_admin_queue; - } - - /* only warn if argument is too large here, will clamp later */ - if (opts->queue_size > ctrl->ctrl.sqsize + 1) { - dev_warn(ctrl->ctrl.device, - "queue_size %zu > ctrl sqsize %u, clamping down\n", - opts->queue_size, ctrl->ctrl.sqsize + 1); - } - - /* warn if maxcmd is lower than sqsize+1 */ - if (ctrl->ctrl.sqsize + 1 > ctrl->ctrl.maxcmd) { - dev_warn(ctrl->ctrl.device, - "sqsize %u > ctrl maxcmd %u, clamping down\n", - ctrl->ctrl.sqsize + 1, ctrl->ctrl.maxcmd); - ctrl->ctrl.sqsize = ctrl->ctrl.maxcmd - 1; - } - - if (opts->nr_io_queues) { - ret = nvme_rdma_configure_io_queues(ctrl, true); - if (ret) - goto out_remove_admin_queue; - } - - changed = nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_LIVE); - WARN_ON_ONCE(!changed); - dev_info(ctrl->ctrl.device, "new ctrl: NQN \"%s\", addr %pISpcs\n", ctrl->ctrl.opts->subsysnqn, &ctrl->addr); @@ -2011,13 +2004,8 @@ static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev, list_add_tail(&ctrl->list, &nvme_rdma_ctrl_list); mutex_unlock(&nvme_rdma_ctrl_mutex); - nvme_start_ctrl(&ctrl->ctrl); - return &ctrl->ctrl; -out_remove_admin_queue: - nvme_rdma_stop_queue(&ctrl->queues[0]); - nvme_rdma_destroy_admin_queue(ctrl, true); out_uninit_ctrl: nvme_uninit_ctrl(&ctrl->ctrl); nvme_put_ctrl(&ctrl->ctrl); diff --git a/drivers/nvme/host/trace.c b/drivers/nvme/host/trace.c index 41944bbef835..25b0e310f4a8 100644 --- a/drivers/nvme/host/trace.c +++ b/drivers/nvme/host/trace.c @@ -128,3 +128,14 @@ const char *nvme_trace_parse_nvm_cmd(struct trace_seq *p, return nvme_trace_common(p, cdw10); } } + +const char *nvme_trace_disk_name(struct trace_seq *p, char *name) +{ + const char *ret = trace_seq_buffer_ptr(p); + + if (*name) + trace_seq_printf(p, "disk=%s, ", name); + trace_seq_putc(p, 0); + + return ret; +} diff --git a/drivers/nvme/host/trace.h b/drivers/nvme/host/trace.h index 01390f0e1671..a490790d6691 100644 --- a/drivers/nvme/host/trace.h +++ b/drivers/nvme/host/trace.h @@ -50,13 +50,8 @@ nvme_admin_opcode_name(nvme_admin_security_recv), \ nvme_admin_opcode_name(nvme_admin_sanitize_nvm)) -const char *nvme_trace_parse_admin_cmd(struct trace_seq *p, u8 opcode, - u8 *cdw10); -#define __parse_nvme_admin_cmd(opcode, cdw10) \ - nvme_trace_parse_admin_cmd(p, opcode, cdw10) - #define nvme_opcode_name(opcode) { opcode, #opcode } -#define show_opcode_name(val) \ +#define show_nvm_opcode_name(val) \ __print_symbolic(val, \ nvme_opcode_name(nvme_cmd_flush), \ nvme_opcode_name(nvme_cmd_write), \ @@ -70,85 +65,92 @@ const char *nvme_trace_parse_admin_cmd(struct trace_seq *p, u8 opcode, nvme_opcode_name(nvme_cmd_resv_acquire), \ nvme_opcode_name(nvme_cmd_resv_release)) -const char *nvme_trace_parse_nvm_cmd(struct trace_seq *p, u8 opcode, - u8 *cdw10); -#define __parse_nvme_cmd(opcode, cdw10) \ - nvme_trace_parse_nvm_cmd(p, opcode, cdw10) - -TRACE_EVENT(nvme_setup_admin_cmd, - TP_PROTO(struct nvme_command *cmd), - TP_ARGS(cmd), - TP_STRUCT__entry( - __field(u8, opcode) - __field(u8, flags) - __field(u16, cid) - __field(u64, metadata) - __array(u8, cdw10, 24) - ), - TP_fast_assign( - __entry->opcode = cmd->common.opcode; - __entry->flags = cmd->common.flags; - __entry->cid = cmd->common.command_id; - __entry->metadata = le64_to_cpu(cmd->common.metadata); - memcpy(__entry->cdw10, cmd->common.cdw10, - sizeof(__entry->cdw10)); - ), - TP_printk(" cmdid=%u, flags=0x%x, meta=0x%llx, cmd=(%s %s)", - __entry->cid, __entry->flags, __entry->metadata, - show_admin_opcode_name(__entry->opcode), - __parse_nvme_admin_cmd(__entry->opcode, __entry->cdw10)) -); - +#define show_opcode_name(qid, opcode) \ + (qid ? show_nvm_opcode_name(opcode) : show_admin_opcode_name(opcode)) -TRACE_EVENT(nvme_setup_nvm_cmd, - TP_PROTO(int qid, struct nvme_command *cmd), - TP_ARGS(qid, cmd), +const char *nvme_trace_parse_admin_cmd(struct trace_seq *p, u8 opcode, + u8 *cdw10); +const char *nvme_trace_parse_nvm_cmd(struct trace_seq *p, u8 opcode, + u8 *cdw10); + +#define parse_nvme_cmd(qid, opcode, cdw10) \ + (qid ? \ + nvme_trace_parse_nvm_cmd(p, opcode, cdw10) : \ + nvme_trace_parse_admin_cmd(p, opcode, cdw10)) + +const char *nvme_trace_disk_name(struct trace_seq *p, char *name); +#define __print_disk_name(name) \ + nvme_trace_disk_name(p, name) + +#ifndef TRACE_HEADER_MULTI_READ +static inline void __assign_disk_name(char *name, struct gendisk *disk) +{ + if (disk) + memcpy(name, disk->disk_name, DISK_NAME_LEN); + else + memset(name, 0, DISK_NAME_LEN); +} +#endif + +TRACE_EVENT(nvme_setup_cmd, + TP_PROTO(struct request *req, struct nvme_command *cmd), + TP_ARGS(req, cmd), TP_STRUCT__entry( - __field(int, qid) - __field(u8, opcode) - __field(u8, flags) - __field(u16, cid) - __field(u32, nsid) - __field(u64, metadata) - __array(u8, cdw10, 24) + __array(char, disk, DISK_NAME_LEN) + __field(int, ctrl_id) + __field(int, qid) + __field(u8, opcode) + __field(u8, flags) + __field(u16, cid) + __field(u32, nsid) + __field(u64, metadata) + __array(u8, cdw10, 24) ), TP_fast_assign( - __entry->qid = qid; - __entry->opcode = cmd->common.opcode; - __entry->flags = cmd->common.flags; - __entry->cid = cmd->common.command_id; - __entry->nsid = le32_to_cpu(cmd->common.nsid); - __entry->metadata = le64_to_cpu(cmd->common.metadata); - memcpy(__entry->cdw10, cmd->common.cdw10, - sizeof(__entry->cdw10)); + __entry->ctrl_id = nvme_req(req)->ctrl->instance; + __entry->qid = nvme_req_qid(req); + __entry->opcode = cmd->common.opcode; + __entry->flags = cmd->common.flags; + __entry->cid = cmd->common.command_id; + __entry->nsid = le32_to_cpu(cmd->common.nsid); + __entry->metadata = le64_to_cpu(cmd->common.metadata); + __assign_disk_name(__entry->disk, req->rq_disk); + memcpy(__entry->cdw10, cmd->common.cdw10, + sizeof(__entry->cdw10)); ), - TP_printk("qid=%d, nsid=%u, cmdid=%u, flags=0x%x, meta=0x%llx, cmd=(%s %s)", - __entry->qid, __entry->nsid, __entry->cid, + TP_printk("nvme%d: %sqid=%d, cmdid=%u, nsid=%u, flags=0x%x, meta=0x%llx, cmd=(%s %s)", + __entry->ctrl_id, __print_disk_name(__entry->disk), + __entry->qid, __entry->cid, __entry->nsid, __entry->flags, __entry->metadata, - show_opcode_name(__entry->opcode), - __parse_nvme_cmd(__entry->opcode, __entry->cdw10)) + show_opcode_name(__entry->qid, __entry->opcode), + parse_nvme_cmd(__entry->qid, __entry->opcode, __entry->cdw10)) ); TRACE_EVENT(nvme_complete_rq, TP_PROTO(struct request *req), TP_ARGS(req), TP_STRUCT__entry( - __field(int, qid) - __field(int, cid) - __field(u64, result) - __field(u8, retries) - __field(u8, flags) - __field(u16, status) + __array(char, disk, DISK_NAME_LEN) + __field(int, ctrl_id) + __field(int, qid) + __field(int, cid) + __field(u64, result) + __field(u8, retries) + __field(u8, flags) + __field(u16, status) ), TP_fast_assign( - __entry->qid = req->q->id; - __entry->cid = req->tag; - __entry->result = le64_to_cpu(nvme_req(req)->result.u64); - __entry->retries = nvme_req(req)->retries; - __entry->flags = nvme_req(req)->flags; - __entry->status = nvme_req(req)->status; + __entry->ctrl_id = nvme_req(req)->ctrl->instance; + __entry->qid = nvme_req_qid(req); + __entry->cid = req->tag; + __entry->result = le64_to_cpu(nvme_req(req)->result.u64); + __entry->retries = nvme_req(req)->retries; + __entry->flags = nvme_req(req)->flags; + __entry->status = nvme_req(req)->status; + __assign_disk_name(__entry->disk, req->rq_disk); ), - TP_printk("qid=%d, cmdid=%u, res=%llu, retries=%u, flags=0x%x, status=%u", + TP_printk("nvme%d: %sqid=%d, cmdid=%u, res=%llu, retries=%u, flags=0x%x, status=%u", + __entry->ctrl_id, __print_disk_name(__entry->disk), __entry->qid, __entry->cid, __entry->result, __entry->retries, __entry->flags, __entry->status) diff --git a/drivers/nvme/target/admin-cmd.c b/drivers/nvme/target/admin-cmd.c index 38803576d5e1..a21caea1e080 100644 --- a/drivers/nvme/target/admin-cmd.c +++ b/drivers/nvme/target/admin-cmd.c @@ -19,6 +19,19 @@ #include <asm/unaligned.h> #include "nvmet.h" +/* + * This helper allows us to clear the AEN based on the RAE bit, + * Please use this helper when processing the log pages which are + * associated with the AEN. + */ +static inline void nvmet_clear_aen(struct nvmet_req *req, u32 aen_bit) +{ + int rae = le32_to_cpu(req->cmd->common.cdw10[0]) & 1 << 15; + + if (!rae) + clear_bit(aen_bit, &req->sq->ctrl->aen_masked); +} + u32 nvmet_get_log_page_len(struct nvme_command *cmd) { u32 len = le16_to_cpu(cmd->get_log_page.numdu); @@ -128,6 +141,36 @@ out: nvmet_req_complete(req, status); } +static void nvmet_execute_get_log_cmd_effects_ns(struct nvmet_req *req) +{ + u16 status = NVME_SC_INTERNAL; + struct nvme_effects_log *log; + + log = kzalloc(sizeof(*log), GFP_KERNEL); + if (!log) + goto out; + + log->acs[nvme_admin_get_log_page] = cpu_to_le32(1 << 0); + log->acs[nvme_admin_identify] = cpu_to_le32(1 << 0); + log->acs[nvme_admin_abort_cmd] = cpu_to_le32(1 << 0); + log->acs[nvme_admin_set_features] = cpu_to_le32(1 << 0); + log->acs[nvme_admin_get_features] = cpu_to_le32(1 << 0); + log->acs[nvme_admin_async_event] = cpu_to_le32(1 << 0); + log->acs[nvme_admin_keep_alive] = cpu_to_le32(1 << 0); + + log->iocs[nvme_cmd_read] = cpu_to_le32(1 << 0); + log->iocs[nvme_cmd_write] = cpu_to_le32(1 << 0); + log->iocs[nvme_cmd_flush] = cpu_to_le32(1 << 0); + log->iocs[nvme_cmd_dsm] = cpu_to_le32(1 << 0); + log->iocs[nvme_cmd_write_zeroes] = cpu_to_le32(1 << 0); + + status = nvmet_copy_to_sgl(req, 0, log, sizeof(*log)); + + kfree(log); +out: + nvmet_req_complete(req, status); +} + static void nvmet_execute_get_log_changed_ns(struct nvmet_req *req) { struct nvmet_ctrl *ctrl = req->sq->ctrl; @@ -146,12 +189,76 @@ static void nvmet_execute_get_log_changed_ns(struct nvmet_req *req) if (!status) status = nvmet_zero_sgl(req, len, req->data_len - len); ctrl->nr_changed_ns = 0; - clear_bit(NVME_AEN_CFG_NS_ATTR, &ctrl->aen_masked); + nvmet_clear_aen(req, NVME_AEN_CFG_NS_ATTR); mutex_unlock(&ctrl->lock); out: nvmet_req_complete(req, status); } +static u32 nvmet_format_ana_group(struct nvmet_req *req, u32 grpid, + struct nvme_ana_group_desc *desc) +{ + struct nvmet_ctrl *ctrl = req->sq->ctrl; + struct nvmet_ns *ns; + u32 count = 0; + + if (!(req->cmd->get_log_page.lsp & NVME_ANA_LOG_RGO)) { + rcu_read_lock(); + list_for_each_entry_rcu(ns, &ctrl->subsys->namespaces, dev_link) + if (ns->anagrpid == grpid) + desc->nsids[count++] = cpu_to_le32(ns->nsid); + rcu_read_unlock(); + } + + desc->grpid = cpu_to_le32(grpid); + desc->nnsids = cpu_to_le32(count); + desc->chgcnt = cpu_to_le64(nvmet_ana_chgcnt); + desc->state = req->port->ana_state[grpid]; + memset(desc->rsvd17, 0, sizeof(desc->rsvd17)); + return sizeof(struct nvme_ana_group_desc) + count * sizeof(__le32); +} + +static void nvmet_execute_get_log_page_ana(struct nvmet_req *req) +{ + struct nvme_ana_rsp_hdr hdr = { 0, }; + struct nvme_ana_group_desc *desc; + size_t offset = sizeof(struct nvme_ana_rsp_hdr); /* start beyond hdr */ + size_t len; + u32 grpid; + u16 ngrps = 0; + u16 status; + + status = NVME_SC_INTERNAL; + desc = kmalloc(sizeof(struct nvme_ana_group_desc) + + NVMET_MAX_NAMESPACES * sizeof(__le32), GFP_KERNEL); + if (!desc) + goto out; + + down_read(&nvmet_ana_sem); + for (grpid = 1; grpid <= NVMET_MAX_ANAGRPS; grpid++) { + if (!nvmet_ana_group_enabled[grpid]) + continue; + len = nvmet_format_ana_group(req, grpid, desc); + status = nvmet_copy_to_sgl(req, offset, desc, len); + if (status) + break; + offset += len; + ngrps++; + } + + hdr.chgcnt = cpu_to_le64(nvmet_ana_chgcnt); + hdr.ngrps = cpu_to_le16(ngrps); + nvmet_clear_aen(req, NVME_AEN_CFG_ANA_CHANGE); + up_read(&nvmet_ana_sem); + + kfree(desc); + + /* copy the header last once we know the number of groups */ + status = nvmet_copy_to_sgl(req, 0, &hdr, sizeof(hdr)); +out: + nvmet_req_complete(req, status); +} + static void nvmet_execute_identify_ctrl(struct nvmet_req *req) { struct nvmet_ctrl *ctrl = req->sq->ctrl; @@ -183,8 +290,8 @@ static void nvmet_execute_identify_ctrl(struct nvmet_req *req) * the safest is to leave it as zeroes. */ - /* we support multiple ports and multiples hosts: */ - id->cmic = (1 << 0) | (1 << 1); + /* we support multiple ports, multiples hosts and ANA: */ + id->cmic = (1 << 0) | (1 << 1) | (1 << 3); /* no limit on data transfer sizes for now */ id->mdts = 0; @@ -208,7 +315,7 @@ static void nvmet_execute_identify_ctrl(struct nvmet_req *req) /* first slot is read-only, only one slot supported */ id->frmw = (1 << 0) | (1 << 1); - id->lpa = (1 << 0) | (1 << 2); + id->lpa = (1 << 0) | (1 << 1) | (1 << 2); id->elpe = NVMET_ERROR_LOG_SLOTS - 1; id->npss = 0; @@ -222,6 +329,7 @@ static void nvmet_execute_identify_ctrl(struct nvmet_req *req) id->maxcmd = cpu_to_le16(NVMET_MAX_CMD); id->nn = cpu_to_le32(ctrl->subsys->max_nsid); + id->mnan = cpu_to_le32(NVMET_MAX_NAMESPACES); id->oncs = cpu_to_le16(NVME_CTRL_ONCS_DSM | NVME_CTRL_ONCS_WRITE_ZEROES); @@ -238,19 +346,24 @@ static void nvmet_execute_identify_ctrl(struct nvmet_req *req) id->sgls = cpu_to_le32(1 << 0); /* we always support SGLs */ if (ctrl->ops->has_keyed_sgls) id->sgls |= cpu_to_le32(1 << 2); - if (ctrl->ops->sqe_inline_size) + if (req->port->inline_data_size) id->sgls |= cpu_to_le32(1 << 20); strcpy(id->subnqn, ctrl->subsys->subsysnqn); /* Max command capsule size is sqe + single page of in-capsule data */ id->ioccsz = cpu_to_le32((sizeof(struct nvme_command) + - ctrl->ops->sqe_inline_size) / 16); + req->port->inline_data_size) / 16); /* Max response capsule size is cqe */ id->iorcsz = cpu_to_le32(sizeof(struct nvme_completion) / 16); id->msdbd = ctrl->ops->msdbd; + id->anacap = (1 << 0) | (1 << 1) | (1 << 2) | (1 << 3) | (1 << 4); + id->anatt = 10; /* random value */ + id->anagrpmax = cpu_to_le32(NVMET_MAX_ANAGRPS); + id->nanagrpid = cpu_to_le32(NVMET_MAX_ANAGRPS); + /* * Meh, we don't really support any power state. Fake up the same * values that qemu does. @@ -259,6 +372,8 @@ static void nvmet_execute_identify_ctrl(struct nvmet_req *req) id->psd[0].entry_lat = cpu_to_le32(0x10); id->psd[0].exit_lat = cpu_to_le32(0x4); + id->nwpc = 1 << 0; /* write protect and no write protect */ + status = nvmet_copy_to_sgl(req, 0, id, sizeof(*id)); kfree(id); @@ -292,8 +407,15 @@ static void nvmet_execute_identify_ns(struct nvmet_req *req) * nuse = ncap = nsze isn't always true, but we have no way to find * that out from the underlying device. */ - id->ncap = id->nuse = id->nsze = - cpu_to_le64(ns->size >> ns->blksize_shift); + id->ncap = id->nsze = cpu_to_le64(ns->size >> ns->blksize_shift); + switch (req->port->ana_state[ns->anagrpid]) { + case NVME_ANA_INACCESSIBLE: + case NVME_ANA_PERSISTENT_LOSS: + break; + default: + id->nuse = id->nsze; + break; + } /* * We just provide a single LBA format that matches what the @@ -307,11 +429,14 @@ static void nvmet_execute_identify_ns(struct nvmet_req *req) * controllers, but also with any other user of the block device. */ id->nmic = (1 << 0); + id->anagrpid = cpu_to_le32(ns->anagrpid); - memcpy(&id->nguid, &ns->nguid, sizeof(uuid_le)); + memcpy(&id->nguid, &ns->nguid, sizeof(id->nguid)); id->lbaf[0].ds = ns->blksize_shift; + if (ns->readonly) + id->nsattr |= (1 << 0); nvmet_put_namespace(ns); done: status = nvmet_copy_to_sgl(req, 0, id, sizeof(*id)); @@ -424,6 +549,52 @@ static void nvmet_execute_abort(struct nvmet_req *req) nvmet_req_complete(req, 0); } +static u16 nvmet_write_protect_flush_sync(struct nvmet_req *req) +{ + u16 status; + + if (req->ns->file) + status = nvmet_file_flush(req); + else + status = nvmet_bdev_flush(req); + + if (status) + pr_err("write protect flush failed nsid: %u\n", req->ns->nsid); + return status; +} + +static u16 nvmet_set_feat_write_protect(struct nvmet_req *req) +{ + u32 write_protect = le32_to_cpu(req->cmd->common.cdw10[1]); + struct nvmet_subsys *subsys = req->sq->ctrl->subsys; + u16 status = NVME_SC_FEATURE_NOT_CHANGEABLE; + + req->ns = nvmet_find_namespace(req->sq->ctrl, req->cmd->rw.nsid); + if (unlikely(!req->ns)) + return status; + + mutex_lock(&subsys->lock); + switch (write_protect) { + case NVME_NS_WRITE_PROTECT: + req->ns->readonly = true; + status = nvmet_write_protect_flush_sync(req); + if (status) + req->ns->readonly = false; + break; + case NVME_NS_NO_WRITE_PROTECT: + req->ns->readonly = false; + status = 0; + break; + default: + break; + } + + if (!status) + nvmet_ns_changed(subsys, req->ns->nsid); + mutex_unlock(&subsys->lock); + return status; +} + static void nvmet_execute_set_features(struct nvmet_req *req) { struct nvmet_subsys *subsys = req->sq->ctrl->subsys; @@ -454,6 +625,9 @@ static void nvmet_execute_set_features(struct nvmet_req *req) case NVME_FEAT_HOST_ID: status = NVME_SC_CMD_SEQ_ERROR | NVME_SC_DNR; break; + case NVME_FEAT_WRITE_PROTECT: + status = nvmet_set_feat_write_protect(req); + break; default: status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; break; @@ -462,6 +636,26 @@ static void nvmet_execute_set_features(struct nvmet_req *req) nvmet_req_complete(req, status); } +static u16 nvmet_get_feat_write_protect(struct nvmet_req *req) +{ + struct nvmet_subsys *subsys = req->sq->ctrl->subsys; + u32 result; + + req->ns = nvmet_find_namespace(req->sq->ctrl, req->cmd->common.nsid); + if (!req->ns) + return NVME_SC_INVALID_NS | NVME_SC_DNR; + + mutex_lock(&subsys->lock); + if (req->ns->readonly == true) + result = NVME_NS_WRITE_PROTECT; + else + result = NVME_NS_NO_WRITE_PROTECT; + nvmet_set_result(req, result); + mutex_unlock(&subsys->lock); + + return 0; +} + static void nvmet_execute_get_features(struct nvmet_req *req) { struct nvmet_subsys *subsys = req->sq->ctrl->subsys; @@ -513,6 +707,9 @@ static void nvmet_execute_get_features(struct nvmet_req *req) status = nvmet_copy_to_sgl(req, 0, &req->sq->ctrl->hostid, sizeof(req->sq->ctrl->hostid)); break; + case NVME_FEAT_WRITE_PROTECT: + status = nvmet_get_feat_write_protect(req); + break; default: status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; break; @@ -586,6 +783,12 @@ u16 nvmet_parse_admin_cmd(struct nvmet_req *req) case NVME_LOG_CHANGED_NS: req->execute = nvmet_execute_get_log_changed_ns; return 0; + case NVME_LOG_CMD_EFFECTS: + req->execute = nvmet_execute_get_log_cmd_effects_ns; + return 0; + case NVME_LOG_ANA: + req->execute = nvmet_execute_get_log_page_ana; + return 0; } break; case nvme_admin_identify: diff --git a/drivers/nvme/target/configfs.c b/drivers/nvme/target/configfs.c index ebea1373d1b7..b37a8e3e3f80 100644 --- a/drivers/nvme/target/configfs.c +++ b/drivers/nvme/target/configfs.c @@ -218,6 +218,35 @@ static ssize_t nvmet_addr_trsvcid_store(struct config_item *item, CONFIGFS_ATTR(nvmet_, addr_trsvcid); +static ssize_t nvmet_param_inline_data_size_show(struct config_item *item, + char *page) +{ + struct nvmet_port *port = to_nvmet_port(item); + + return snprintf(page, PAGE_SIZE, "%d\n", port->inline_data_size); +} + +static ssize_t nvmet_param_inline_data_size_store(struct config_item *item, + const char *page, size_t count) +{ + struct nvmet_port *port = to_nvmet_port(item); + int ret; + + if (port->enabled) { + pr_err("Cannot modify inline_data_size while port enabled\n"); + pr_err("Disable the port before modifying\n"); + return -EACCES; + } + ret = kstrtoint(page, 0, &port->inline_data_size); + if (ret) { + pr_err("Invalid value '%s' for inline_data_size\n", page); + return -EINVAL; + } + return count; +} + +CONFIGFS_ATTR(nvmet_, param_inline_data_size); + static ssize_t nvmet_addr_trtype_show(struct config_item *item, char *page) { @@ -387,6 +416,39 @@ out_unlock: CONFIGFS_ATTR(nvmet_ns_, device_nguid); +static ssize_t nvmet_ns_ana_grpid_show(struct config_item *item, char *page) +{ + return sprintf(page, "%u\n", to_nvmet_ns(item)->anagrpid); +} + +static ssize_t nvmet_ns_ana_grpid_store(struct config_item *item, + const char *page, size_t count) +{ + struct nvmet_ns *ns = to_nvmet_ns(item); + u32 oldgrpid, newgrpid; + int ret; + + ret = kstrtou32(page, 0, &newgrpid); + if (ret) + return ret; + + if (newgrpid < 1 || newgrpid > NVMET_MAX_ANAGRPS) + return -EINVAL; + + down_write(&nvmet_ana_sem); + oldgrpid = ns->anagrpid; + nvmet_ana_group_enabled[newgrpid]++; + ns->anagrpid = newgrpid; + nvmet_ana_group_enabled[oldgrpid]--; + nvmet_ana_chgcnt++; + up_write(&nvmet_ana_sem); + + nvmet_send_ana_event(ns->subsys, NULL); + return count; +} + +CONFIGFS_ATTR(nvmet_ns_, ana_grpid); + static ssize_t nvmet_ns_enable_show(struct config_item *item, char *page) { return sprintf(page, "%d\n", to_nvmet_ns(item)->enabled); @@ -412,11 +474,41 @@ static ssize_t nvmet_ns_enable_store(struct config_item *item, CONFIGFS_ATTR(nvmet_ns_, enable); +static ssize_t nvmet_ns_buffered_io_show(struct config_item *item, char *page) +{ + return sprintf(page, "%d\n", to_nvmet_ns(item)->buffered_io); +} + +static ssize_t nvmet_ns_buffered_io_store(struct config_item *item, + const char *page, size_t count) +{ + struct nvmet_ns *ns = to_nvmet_ns(item); + bool val; + + if (strtobool(page, &val)) + return -EINVAL; + + mutex_lock(&ns->subsys->lock); + if (ns->enabled) { + pr_err("disable ns before setting buffered_io value.\n"); + mutex_unlock(&ns->subsys->lock); + return -EINVAL; + } + + ns->buffered_io = val; + mutex_unlock(&ns->subsys->lock); + return count; +} + +CONFIGFS_ATTR(nvmet_ns_, buffered_io); + static struct configfs_attribute *nvmet_ns_attrs[] = { &nvmet_ns_attr_device_path, &nvmet_ns_attr_device_nguid, &nvmet_ns_attr_device_uuid, + &nvmet_ns_attr_ana_grpid, &nvmet_ns_attr_enable, + &nvmet_ns_attr_buffered_io, NULL, }; @@ -863,6 +955,134 @@ static const struct config_item_type nvmet_referrals_type = { .ct_group_ops = &nvmet_referral_group_ops, }; +static struct { + enum nvme_ana_state state; + const char *name; +} nvmet_ana_state_names[] = { + { NVME_ANA_OPTIMIZED, "optimized" }, + { NVME_ANA_NONOPTIMIZED, "non-optimized" }, + { NVME_ANA_INACCESSIBLE, "inaccessible" }, + { NVME_ANA_PERSISTENT_LOSS, "persistent-loss" }, + { NVME_ANA_CHANGE, "change" }, +}; + +static ssize_t nvmet_ana_group_ana_state_show(struct config_item *item, + char *page) +{ + struct nvmet_ana_group *grp = to_ana_group(item); + enum nvme_ana_state state = grp->port->ana_state[grp->grpid]; + int i; + + for (i = 0; i < ARRAY_SIZE(nvmet_ana_state_names); i++) { + if (state != nvmet_ana_state_names[i].state) + continue; + return sprintf(page, "%s\n", nvmet_ana_state_names[i].name); + } + + return sprintf(page, "\n"); +} + +static ssize_t nvmet_ana_group_ana_state_store(struct config_item *item, + const char *page, size_t count) +{ + struct nvmet_ana_group *grp = to_ana_group(item); + int i; + + for (i = 0; i < ARRAY_SIZE(nvmet_ana_state_names); i++) { + if (sysfs_streq(page, nvmet_ana_state_names[i].name)) + goto found; + } + + pr_err("Invalid value '%s' for ana_state\n", page); + return -EINVAL; + +found: + down_write(&nvmet_ana_sem); + grp->port->ana_state[grp->grpid] = nvmet_ana_state_names[i].state; + nvmet_ana_chgcnt++; + up_write(&nvmet_ana_sem); + + nvmet_port_send_ana_event(grp->port); + return count; +} + +CONFIGFS_ATTR(nvmet_ana_group_, ana_state); + +static struct configfs_attribute *nvmet_ana_group_attrs[] = { + &nvmet_ana_group_attr_ana_state, + NULL, +}; + +static void nvmet_ana_group_release(struct config_item *item) +{ + struct nvmet_ana_group *grp = to_ana_group(item); + + if (grp == &grp->port->ana_default_group) + return; + + down_write(&nvmet_ana_sem); + grp->port->ana_state[grp->grpid] = NVME_ANA_INACCESSIBLE; + nvmet_ana_group_enabled[grp->grpid]--; + up_write(&nvmet_ana_sem); + + nvmet_port_send_ana_event(grp->port); + kfree(grp); +} + +static struct configfs_item_operations nvmet_ana_group_item_ops = { + .release = nvmet_ana_group_release, +}; + +static const struct config_item_type nvmet_ana_group_type = { + .ct_item_ops = &nvmet_ana_group_item_ops, + .ct_attrs = nvmet_ana_group_attrs, + .ct_owner = THIS_MODULE, +}; + +static struct config_group *nvmet_ana_groups_make_group( + struct config_group *group, const char *name) +{ + struct nvmet_port *port = ana_groups_to_port(&group->cg_item); + struct nvmet_ana_group *grp; + u32 grpid; + int ret; + + ret = kstrtou32(name, 0, &grpid); + if (ret) + goto out; + + ret = -EINVAL; + if (grpid <= 1 || grpid > NVMET_MAX_ANAGRPS) + goto out; + + ret = -ENOMEM; + grp = kzalloc(sizeof(*grp), GFP_KERNEL); + if (!grp) + goto out; + grp->port = port; + grp->grpid = grpid; + + down_write(&nvmet_ana_sem); + nvmet_ana_group_enabled[grpid]++; + up_write(&nvmet_ana_sem); + + nvmet_port_send_ana_event(grp->port); + + config_group_init_type_name(&grp->group, name, &nvmet_ana_group_type); + return &grp->group; +out: + return ERR_PTR(ret); +} + +static struct configfs_group_operations nvmet_ana_groups_group_ops = { + .make_group = nvmet_ana_groups_make_group, +}; + +static const struct config_item_type nvmet_ana_groups_type = { + .ct_group_ops = &nvmet_ana_groups_group_ops, + .ct_owner = THIS_MODULE, +}; + /* * Ports definitions. */ @@ -870,6 +1090,7 @@ static void nvmet_port_release(struct config_item *item) { struct nvmet_port *port = to_nvmet_port(item); + kfree(port->ana_state); kfree(port); } @@ -879,6 +1100,7 @@ static struct configfs_attribute *nvmet_port_attrs[] = { &nvmet_attr_addr_traddr, &nvmet_attr_addr_trsvcid, &nvmet_attr_addr_trtype, + &nvmet_attr_param_inline_data_size, NULL, }; @@ -897,6 +1119,7 @@ static struct config_group *nvmet_ports_make(struct config_group *group, { struct nvmet_port *port; u16 portid; + u32 i; if (kstrtou16(name, 0, &portid)) return ERR_PTR(-EINVAL); @@ -905,9 +1128,24 @@ static struct config_group *nvmet_ports_make(struct config_group *group, if (!port) return ERR_PTR(-ENOMEM); + port->ana_state = kcalloc(NVMET_MAX_ANAGRPS + 1, + sizeof(*port->ana_state), GFP_KERNEL); + if (!port->ana_state) { + kfree(port); + return ERR_PTR(-ENOMEM); + } + + for (i = 1; i <= NVMET_MAX_ANAGRPS; i++) { + if (i == NVMET_DEFAULT_ANA_GRPID) + port->ana_state[1] = NVME_ANA_OPTIMIZED; + else + port->ana_state[i] = NVME_ANA_INACCESSIBLE; + } + INIT_LIST_HEAD(&port->entry); INIT_LIST_HEAD(&port->subsystems); INIT_LIST_HEAD(&port->referrals); + port->inline_data_size = -1; /* < 0 == let the transport choose */ port->disc_addr.portid = cpu_to_le16(portid); config_group_init_type_name(&port->group, name, &nvmet_port_type); @@ -920,6 +1158,18 @@ static struct config_group *nvmet_ports_make(struct config_group *group, "referrals", &nvmet_referrals_type); configfs_add_default_group(&port->referrals_group, &port->group); + config_group_init_type_name(&port->ana_groups_group, + "ana_groups", &nvmet_ana_groups_type); + configfs_add_default_group(&port->ana_groups_group, &port->group); + + port->ana_default_group.port = port; + port->ana_default_group.grpid = NVMET_DEFAULT_ANA_GRPID; + config_group_init_type_name(&port->ana_default_group.group, + __stringify(NVMET_DEFAULT_ANA_GRPID), + &nvmet_ana_group_type); + configfs_add_default_group(&port->ana_default_group.group, + &port->ana_groups_group); + return &port->group; } diff --git a/drivers/nvme/target/core.c b/drivers/nvme/target/core.c index 9838103f2d62..ebf3e7a6c49e 100644 --- a/drivers/nvme/target/core.c +++ b/drivers/nvme/target/core.c @@ -18,6 +18,7 @@ #include "nvmet.h" +struct workqueue_struct *buffered_io_wq; static const struct nvmet_fabrics_ops *nvmet_transports[NVMF_TRTYPE_MAX]; static DEFINE_IDA(cntlid_ida); @@ -39,6 +40,10 @@ static DEFINE_IDA(cntlid_ida); */ DECLARE_RWSEM(nvmet_config_sem); +u32 nvmet_ana_group_enabled[NVMET_MAX_ANAGRPS + 1]; +u64 nvmet_ana_chgcnt; +DECLARE_RWSEM(nvmet_ana_sem); + static struct nvmet_subsys *nvmet_find_get_subsys(struct nvmet_port *port, const char *subsysnqn); @@ -175,7 +180,7 @@ out_unlock: mutex_unlock(&ctrl->lock); } -static void nvmet_ns_changed(struct nvmet_subsys *subsys, u32 nsid) +void nvmet_ns_changed(struct nvmet_subsys *subsys, u32 nsid) { struct nvmet_ctrl *ctrl; @@ -189,6 +194,33 @@ static void nvmet_ns_changed(struct nvmet_subsys *subsys, u32 nsid) } } +void nvmet_send_ana_event(struct nvmet_subsys *subsys, + struct nvmet_port *port) +{ + struct nvmet_ctrl *ctrl; + + mutex_lock(&subsys->lock); + list_for_each_entry(ctrl, &subsys->ctrls, subsys_entry) { + if (port && ctrl->port != port) + continue; + if (nvmet_aen_disabled(ctrl, NVME_AEN_CFG_ANA_CHANGE)) + continue; + nvmet_add_async_event(ctrl, NVME_AER_TYPE_NOTICE, + NVME_AER_NOTICE_ANA, NVME_LOG_ANA); + } + mutex_unlock(&subsys->lock); +} + +void nvmet_port_send_ana_event(struct nvmet_port *port) +{ + struct nvmet_subsys_link *p; + + down_read(&nvmet_config_sem); + list_for_each_entry(p, &port->subsystems, entry) + nvmet_send_ana_event(p->subsys, port); + up_read(&nvmet_config_sem); +} + int nvmet_register_transport(const struct nvmet_fabrics_ops *ops) { int ret = 0; @@ -241,6 +273,10 @@ int nvmet_enable_port(struct nvmet_port *port) return ret; } + /* If the transport didn't set inline_data_size, then disable it. */ + if (port->inline_data_size < 0) + port->inline_data_size = 0; + port->enabled = true; return 0; } @@ -332,9 +368,13 @@ static void nvmet_ns_dev_disable(struct nvmet_ns *ns) int nvmet_ns_enable(struct nvmet_ns *ns) { struct nvmet_subsys *subsys = ns->subsys; - int ret = 0; + int ret; mutex_lock(&subsys->lock); + ret = -EMFILE; + if (subsys->nr_namespaces == NVMET_MAX_NAMESPACES) + goto out_unlock; + ret = 0; if (ns->enabled) goto out_unlock; @@ -369,6 +409,7 @@ int nvmet_ns_enable(struct nvmet_ns *ns) list_add_tail_rcu(&ns->dev_link, &old->dev_link); } + subsys->nr_namespaces++; nvmet_ns_changed(subsys, ns->nsid); ns->enabled = true; @@ -409,6 +450,7 @@ void nvmet_ns_disable(struct nvmet_ns *ns) percpu_ref_exit(&ns->ref); mutex_lock(&subsys->lock); + subsys->nr_namespaces--; nvmet_ns_changed(subsys, ns->nsid); nvmet_ns_dev_disable(ns); out_unlock: @@ -419,6 +461,10 @@ void nvmet_ns_free(struct nvmet_ns *ns) { nvmet_ns_disable(ns); + down_write(&nvmet_ana_sem); + nvmet_ana_group_enabled[ns->anagrpid]--; + up_write(&nvmet_ana_sem); + kfree(ns->device_path); kfree(ns); } @@ -436,7 +482,14 @@ struct nvmet_ns *nvmet_ns_alloc(struct nvmet_subsys *subsys, u32 nsid) ns->nsid = nsid; ns->subsys = subsys; + + down_write(&nvmet_ana_sem); + ns->anagrpid = NVMET_DEFAULT_ANA_GRPID; + nvmet_ana_group_enabled[ns->anagrpid]++; + up_write(&nvmet_ana_sem); + uuid_gen(&ns->uuid); + ns->buffered_io = false; return ns; } @@ -542,6 +595,35 @@ int nvmet_sq_init(struct nvmet_sq *sq) } EXPORT_SYMBOL_GPL(nvmet_sq_init); +static inline u16 nvmet_check_ana_state(struct nvmet_port *port, + struct nvmet_ns *ns) +{ + enum nvme_ana_state state = port->ana_state[ns->anagrpid]; + + if (unlikely(state == NVME_ANA_INACCESSIBLE)) + return NVME_SC_ANA_INACCESSIBLE; + if (unlikely(state == NVME_ANA_PERSISTENT_LOSS)) + return NVME_SC_ANA_PERSISTENT_LOSS; + if (unlikely(state == NVME_ANA_CHANGE)) + return NVME_SC_ANA_TRANSITION; + return 0; +} + +static inline u16 nvmet_io_cmd_check_access(struct nvmet_req *req) +{ + if (unlikely(req->ns->readonly)) { + switch (req->cmd->common.opcode) { + case nvme_cmd_read: + case nvme_cmd_flush: + break; + default: + return NVME_SC_NS_WRITE_PROTECTED; + } + } + + return 0; +} + static u16 nvmet_parse_io_cmd(struct nvmet_req *req) { struct nvme_command *cmd = req->cmd; @@ -554,6 +636,12 @@ static u16 nvmet_parse_io_cmd(struct nvmet_req *req) req->ns = nvmet_find_namespace(req->sq->ctrl, cmd->rw.nsid); if (unlikely(!req->ns)) return NVME_SC_INVALID_NS | NVME_SC_DNR; + ret = nvmet_check_ana_state(req->port, req->ns); + if (unlikely(ret)) + return ret; + ret = nvmet_io_cmd_check_access(req); + if (unlikely(ret)) + return ret; if (req->ns->file) return nvmet_file_parse_io_cmd(req); @@ -870,6 +958,8 @@ u16 nvmet_alloc_ctrl(const char *subsysnqn, const char *hostnqn, nvmet_init_cap(ctrl); + ctrl->port = req->port; + INIT_WORK(&ctrl->async_event_work, nvmet_async_event_work); INIT_LIST_HEAD(&ctrl->async_events); @@ -1109,6 +1199,15 @@ static int __init nvmet_init(void) { int error; + nvmet_ana_group_enabled[NVMET_DEFAULT_ANA_GRPID] = 1; + + buffered_io_wq = alloc_workqueue("nvmet-buffered-io-wq", + WQ_MEM_RECLAIM, 0); + if (!buffered_io_wq) { + error = -ENOMEM; + goto out; + } + error = nvmet_init_discovery(); if (error) goto out; @@ -1129,6 +1228,7 @@ static void __exit nvmet_exit(void) nvmet_exit_configfs(); nvmet_exit_discovery(); ida_destroy(&cntlid_ida); + destroy_workqueue(buffered_io_wq); BUILD_BUG_ON(sizeof(struct nvmf_disc_rsp_page_entry) != 1024); BUILD_BUG_ON(sizeof(struct nvmf_disc_rsp_page_hdr) != 1024); diff --git a/drivers/nvme/target/discovery.c b/drivers/nvme/target/discovery.c index 08656b849bd6..eae29f493a07 100644 --- a/drivers/nvme/target/discovery.c +++ b/drivers/nvme/target/discovery.c @@ -171,7 +171,7 @@ static void nvmet_execute_identify_disc_ctrl(struct nvmet_req *req) id->sgls = cpu_to_le32(1 << 0); /* we always support SGLs */ if (ctrl->ops->has_keyed_sgls) id->sgls |= cpu_to_le32(1 << 2); - if (ctrl->ops->sqe_inline_size) + if (req->port->inline_data_size) id->sgls |= cpu_to_le32(1 << 20); strcpy(id->subnqn, ctrl->subsys->subsysnqn); diff --git a/drivers/nvme/target/io-cmd-bdev.c b/drivers/nvme/target/io-cmd-bdev.c index e0b0f7df70c2..7bc9f6240432 100644 --- a/drivers/nvme/target/io-cmd-bdev.c +++ b/drivers/nvme/target/io-cmd-bdev.c @@ -124,6 +124,13 @@ static void nvmet_bdev_execute_flush(struct nvmet_req *req) submit_bio(bio); } +u16 nvmet_bdev_flush(struct nvmet_req *req) +{ + if (blkdev_issue_flush(req->ns->bdev, GFP_KERNEL, NULL)) + return NVME_SC_INTERNAL | NVME_SC_DNR; + return 0; +} + static u16 nvmet_bdev_discard_range(struct nvmet_ns *ns, struct nvme_dsm_range *range, struct bio **bio) { diff --git a/drivers/nvme/target/io-cmd-file.c b/drivers/nvme/target/io-cmd-file.c index 8c42b3a8c420..81a9dc5290a8 100644 --- a/drivers/nvme/target/io-cmd-file.c +++ b/drivers/nvme/target/io-cmd-file.c @@ -16,6 +16,8 @@ void nvmet_file_ns_disable(struct nvmet_ns *ns) { if (ns->file) { + if (ns->buffered_io) + flush_workqueue(buffered_io_wq); mempool_destroy(ns->bvec_pool); ns->bvec_pool = NULL; kmem_cache_destroy(ns->bvec_cache); @@ -27,11 +29,14 @@ void nvmet_file_ns_disable(struct nvmet_ns *ns) int nvmet_file_ns_enable(struct nvmet_ns *ns) { - int ret; + int flags = O_RDWR | O_LARGEFILE; struct kstat stat; + int ret; + + if (!ns->buffered_io) + flags |= O_DIRECT; - ns->file = filp_open(ns->device_path, - O_RDWR | O_LARGEFILE | O_DIRECT, 0); + ns->file = filp_open(ns->device_path, flags, 0); if (IS_ERR(ns->file)) { pr_err("failed to open file %s: (%ld)\n", ns->device_path, PTR_ERR(ns->file)); @@ -100,7 +105,7 @@ static ssize_t nvmet_file_submit_bvec(struct nvmet_req *req, loff_t pos, iocb->ki_pos = pos; iocb->ki_filp = req->ns->file; - iocb->ki_flags = IOCB_DIRECT | ki_flags; + iocb->ki_flags = ki_flags | iocb_flags(req->ns->file); ret = call_iter(iocb, &iter); @@ -140,6 +145,12 @@ static void nvmet_file_execute_rw(struct nvmet_req *req) return; } + pos = le64_to_cpu(req->cmd->rw.slba) << req->ns->blksize_shift; + if (unlikely(pos + req->data_len > req->ns->size)) { + nvmet_req_complete(req, NVME_SC_LBA_RANGE | NVME_SC_DNR); + return; + } + if (nr_bvec > NVMET_MAX_INLINE_BIOVEC) req->f.bvec = kmalloc_array(nr_bvec, sizeof(struct bio_vec), GFP_KERNEL); @@ -155,8 +166,6 @@ static void nvmet_file_execute_rw(struct nvmet_req *req) is_sync = true; } - pos = le64_to_cpu(req->cmd->rw.slba) << req->ns->blksize_shift; - memset(&req->f.iocb, 0, sizeof(struct kiocb)); for_each_sg_page(req->sg, &sg_pg_iter, req->sg_cnt, 0) { nvmet_file_init_bvec(&req->f.bvec[bv_cnt], &sg_pg_iter); @@ -189,14 +198,31 @@ out: nvmet_file_submit_bvec(req, pos, bv_cnt, total_len); } -static void nvmet_file_flush_work(struct work_struct *w) +static void nvmet_file_buffered_io_work(struct work_struct *w) { struct nvmet_req *req = container_of(w, struct nvmet_req, f.work); - int ret; - ret = vfs_fsync(req->ns->file, 1); + nvmet_file_execute_rw(req); +} - nvmet_req_complete(req, ret < 0 ? NVME_SC_INTERNAL | NVME_SC_DNR : 0); +static void nvmet_file_execute_rw_buffered_io(struct nvmet_req *req) +{ + INIT_WORK(&req->f.work, nvmet_file_buffered_io_work); + queue_work(buffered_io_wq, &req->f.work); +} + +u16 nvmet_file_flush(struct nvmet_req *req) +{ + if (vfs_fsync(req->ns->file, 1) < 0) + return NVME_SC_INTERNAL | NVME_SC_DNR; + return 0; +} + +static void nvmet_file_flush_work(struct work_struct *w) +{ + struct nvmet_req *req = container_of(w, struct nvmet_req, f.work); + + nvmet_req_complete(req, nvmet_file_flush(req)); } static void nvmet_file_execute_flush(struct nvmet_req *req) @@ -209,22 +235,30 @@ static void nvmet_file_execute_discard(struct nvmet_req *req) { int mode = FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE; struct nvme_dsm_range range; - loff_t offset; - loff_t len; - int i, ret; + loff_t offset, len; + u16 ret; + int i; for (i = 0; i <= le32_to_cpu(req->cmd->dsm.nr); i++) { - if (nvmet_copy_from_sgl(req, i * sizeof(range), &range, - sizeof(range))) + ret = nvmet_copy_from_sgl(req, i * sizeof(range), &range, + sizeof(range)); + if (ret) break; + offset = le64_to_cpu(range.slba) << req->ns->blksize_shift; len = le32_to_cpu(range.nlb) << req->ns->blksize_shift; - ret = vfs_fallocate(req->ns->file, mode, offset, len); - if (ret) + if (offset + len > req->ns->size) { + ret = NVME_SC_LBA_RANGE | NVME_SC_DNR; break; + } + + if (vfs_fallocate(req->ns->file, mode, offset, len)) { + ret = NVME_SC_INTERNAL | NVME_SC_DNR; + break; + } } - nvmet_req_complete(req, ret < 0 ? NVME_SC_INTERNAL | NVME_SC_DNR : 0); + nvmet_req_complete(req, ret); } static void nvmet_file_dsm_work(struct work_struct *w) @@ -263,6 +297,11 @@ static void nvmet_file_write_zeroes_work(struct work_struct *w) len = (((sector_t)le16_to_cpu(write_zeroes->length) + 1) << req->ns->blksize_shift); + if (unlikely(offset + len > req->ns->size)) { + nvmet_req_complete(req, NVME_SC_LBA_RANGE | NVME_SC_DNR); + return; + } + ret = vfs_fallocate(req->ns->file, mode, offset, len); nvmet_req_complete(req, ret < 0 ? NVME_SC_INTERNAL | NVME_SC_DNR : 0); } @@ -280,7 +319,10 @@ u16 nvmet_file_parse_io_cmd(struct nvmet_req *req) switch (cmd->common.opcode) { case nvme_cmd_read: case nvme_cmd_write: - req->execute = nvmet_file_execute_rw; + if (req->ns->buffered_io) + req->execute = nvmet_file_execute_rw_buffered_io; + else + req->execute = nvmet_file_execute_rw; req->data_len = nvmet_rw_len(req); return 0; case nvme_cmd_flush: diff --git a/drivers/nvme/target/loop.c b/drivers/nvme/target/loop.c index ae7586b8be07..9908082b32c4 100644 --- a/drivers/nvme/target/loop.c +++ b/drivers/nvme/target/loop.c @@ -227,6 +227,7 @@ static int nvme_loop_init_request(struct blk_mq_tag_set *set, { struct nvme_loop_ctrl *ctrl = set->driver_data; + nvme_req(req)->ctrl = &ctrl->ctrl; return nvme_loop_init_iod(ctrl, blk_mq_rq_to_pdu(req), (set == &ctrl->tag_set) ? hctx_idx + 1 : 0); } diff --git a/drivers/nvme/target/nvmet.h b/drivers/nvme/target/nvmet.h index 480dfe10fad9..ec9af4ee03b6 100644 --- a/drivers/nvme/target/nvmet.h +++ b/drivers/nvme/target/nvmet.h @@ -30,12 +30,11 @@ #define NVMET_ASYNC_EVENTS 4 #define NVMET_ERROR_LOG_SLOTS 128 - /* * Supported optional AENs: */ #define NVMET_AEN_CFG_OPTIONAL \ - NVME_AEN_CFG_NS_ATTR + (NVME_AEN_CFG_NS_ATTR | NVME_AEN_CFG_ANA_CHANGE) /* * Plus mandatory SMART AENs (we'll never send them, but allow enabling them): @@ -59,12 +58,15 @@ struct nvmet_ns { struct percpu_ref ref; struct block_device *bdev; struct file *file; + bool readonly; u32 nsid; u32 blksize_shift; loff_t size; u8 nguid[16]; uuid_t uuid; + u32 anagrpid; + bool buffered_io; bool enabled; struct nvmet_subsys *subsys; const char *device_path; @@ -97,6 +99,18 @@ struct nvmet_sq { struct completion confirm_done; }; +struct nvmet_ana_group { + struct config_group group; + struct nvmet_port *port; + u32 grpid; +}; + +static inline struct nvmet_ana_group *to_ana_group(struct config_item *item) +{ + return container_of(to_config_group(item), struct nvmet_ana_group, + group); +} + /** * struct nvmet_port - Common structure to keep port * information for the target. @@ -114,8 +128,12 @@ struct nvmet_port { struct list_head subsystems; struct config_group referrals_group; struct list_head referrals; + struct config_group ana_groups_group; + struct nvmet_ana_group ana_default_group; + enum nvme_ana_state *ana_state; void *priv; bool enabled; + int inline_data_size; }; static inline struct nvmet_port *to_nvmet_port(struct config_item *item) @@ -124,6 +142,13 @@ static inline struct nvmet_port *to_nvmet_port(struct config_item *item) group); } +static inline struct nvmet_port *ana_groups_to_port( + struct config_item *item) +{ + return container_of(to_config_group(item), struct nvmet_port, + ana_groups_group); +} + struct nvmet_ctrl { struct nvmet_subsys *subsys; struct nvmet_cq **cqs; @@ -138,6 +163,8 @@ struct nvmet_ctrl { u16 cntlid; u32 kato; + struct nvmet_port *port; + u32 aen_enabled; unsigned long aen_masked; struct nvmet_req *async_event_cmds[NVMET_ASYNC_EVENTS]; @@ -166,6 +193,7 @@ struct nvmet_subsys { struct kref ref; struct list_head namespaces; + unsigned int nr_namespaces; unsigned int max_nsid; struct list_head ctrls; @@ -225,7 +253,6 @@ struct nvmet_req; struct nvmet_fabrics_ops { struct module *owner; unsigned int type; - unsigned int sqe_inline_size; unsigned int msdbd; bool has_keyed_sgls : 1; void (*queue_response)(struct nvmet_req *req); @@ -269,6 +296,8 @@ struct nvmet_req { const struct nvmet_fabrics_ops *ops; }; +extern struct workqueue_struct *buffered_io_wq; + static inline void nvmet_set_status(struct nvmet_req *req, u16 status) { req->rsp->status = cpu_to_le16(status << 1); @@ -337,6 +366,10 @@ void nvmet_ns_disable(struct nvmet_ns *ns); struct nvmet_ns *nvmet_ns_alloc(struct nvmet_subsys *subsys, u32 nsid); void nvmet_ns_free(struct nvmet_ns *ns); +void nvmet_send_ana_event(struct nvmet_subsys *subsys, + struct nvmet_port *port); +void nvmet_port_send_ana_event(struct nvmet_port *port); + int nvmet_register_transport(const struct nvmet_fabrics_ops *ops); void nvmet_unregister_transport(const struct nvmet_fabrics_ops *ops); @@ -357,6 +390,22 @@ u32 nvmet_get_log_page_len(struct nvme_command *cmd); #define NVMET_QUEUE_SIZE 1024 #define NVMET_NR_QUEUES 128 #define NVMET_MAX_CMD NVMET_QUEUE_SIZE + +/* + * Nice round number that makes a list of nsids fit into a page. + * Should become tunable at some point in the future. + */ +#define NVMET_MAX_NAMESPACES 1024 + +/* + * 0 is not a valid ANA group ID, so we start numbering at 1. + * + * ANA Group 1 exists without manual intervention, has namespaces assigned to it + * by default, and is available in an optimized state through all ports. + */ +#define NVMET_MAX_ANAGRPS 128 +#define NVMET_DEFAULT_ANA_GRPID 1 + #define NVMET_KAS 10 #define NVMET_DISC_KATO 120 @@ -370,6 +419,10 @@ extern struct nvmet_subsys *nvmet_disc_subsys; extern u64 nvmet_genctr; extern struct rw_semaphore nvmet_config_sem; +extern u32 nvmet_ana_group_enabled[NVMET_MAX_ANAGRPS + 1]; +extern u64 nvmet_ana_chgcnt; +extern struct rw_semaphore nvmet_ana_sem; + bool nvmet_host_allowed(struct nvmet_req *req, struct nvmet_subsys *subsys, const char *hostnqn); @@ -377,6 +430,9 @@ int nvmet_bdev_ns_enable(struct nvmet_ns *ns); int nvmet_file_ns_enable(struct nvmet_ns *ns); void nvmet_bdev_ns_disable(struct nvmet_ns *ns); void nvmet_file_ns_disable(struct nvmet_ns *ns); +u16 nvmet_bdev_flush(struct nvmet_req *req); +u16 nvmet_file_flush(struct nvmet_req *req); +void nvmet_ns_changed(struct nvmet_subsys *subsys, u32 nsid); static inline u32 nvmet_rw_len(struct nvmet_req *req) { diff --git a/drivers/nvme/target/rdma.c b/drivers/nvme/target/rdma.c index 52e0c5d579a7..e7f43d1e1779 100644 --- a/drivers/nvme/target/rdma.c +++ b/drivers/nvme/target/rdma.c @@ -33,16 +33,17 @@ #include "nvmet.h" /* - * We allow up to a page of inline data to go with the SQE + * We allow at least 1 page, up to 4 SGEs, and up to 16KB of inline data */ -#define NVMET_RDMA_INLINE_DATA_SIZE PAGE_SIZE +#define NVMET_RDMA_DEFAULT_INLINE_DATA_SIZE PAGE_SIZE +#define NVMET_RDMA_MAX_INLINE_SGE 4 +#define NVMET_RDMA_MAX_INLINE_DATA_SIZE max_t(int, SZ_16K, PAGE_SIZE) struct nvmet_rdma_cmd { - struct ib_sge sge[2]; + struct ib_sge sge[NVMET_RDMA_MAX_INLINE_SGE + 1]; struct ib_cqe cqe; struct ib_recv_wr wr; - struct scatterlist inline_sg; - struct page *inline_page; + struct scatterlist inline_sg[NVMET_RDMA_MAX_INLINE_SGE]; struct nvme_command *nvme_cmd; struct nvmet_rdma_queue *queue; }; @@ -116,6 +117,8 @@ struct nvmet_rdma_device { size_t srq_size; struct kref ref; struct list_head entry; + int inline_data_size; + int inline_page_count; }; static bool nvmet_rdma_use_srq; @@ -138,6 +141,11 @@ static void nvmet_rdma_queue_disconnect(struct nvmet_rdma_queue *queue); static const struct nvmet_fabrics_ops nvmet_rdma_ops; +static int num_pages(int len) +{ + return 1 + (((len - 1) & PAGE_MASK) >> PAGE_SHIFT); +} + /* XXX: really should move to a generic header sooner or later.. */ static inline u32 get_unaligned_le24(const u8 *p) { @@ -184,6 +192,71 @@ nvmet_rdma_put_rsp(struct nvmet_rdma_rsp *rsp) spin_unlock_irqrestore(&rsp->queue->rsps_lock, flags); } +static void nvmet_rdma_free_inline_pages(struct nvmet_rdma_device *ndev, + struct nvmet_rdma_cmd *c) +{ + struct scatterlist *sg; + struct ib_sge *sge; + int i; + + if (!ndev->inline_data_size) + return; + + sg = c->inline_sg; + sge = &c->sge[1]; + + for (i = 0; i < ndev->inline_page_count; i++, sg++, sge++) { + if (sge->length) + ib_dma_unmap_page(ndev->device, sge->addr, + sge->length, DMA_FROM_DEVICE); + if (sg_page(sg)) + __free_page(sg_page(sg)); + } +} + +static int nvmet_rdma_alloc_inline_pages(struct nvmet_rdma_device *ndev, + struct nvmet_rdma_cmd *c) +{ + struct scatterlist *sg; + struct ib_sge *sge; + struct page *pg; + int len; + int i; + + if (!ndev->inline_data_size) + return 0; + + sg = c->inline_sg; + sg_init_table(sg, ndev->inline_page_count); + sge = &c->sge[1]; + len = ndev->inline_data_size; + + for (i = 0; i < ndev->inline_page_count; i++, sg++, sge++) { + pg = alloc_page(GFP_KERNEL); + if (!pg) + goto out_err; + sg_assign_page(sg, pg); + sge->addr = ib_dma_map_page(ndev->device, + pg, 0, PAGE_SIZE, DMA_FROM_DEVICE); + if (ib_dma_mapping_error(ndev->device, sge->addr)) + goto out_err; + sge->length = min_t(int, len, PAGE_SIZE); + sge->lkey = ndev->pd->local_dma_lkey; + len -= sge->length; + } + + return 0; +out_err: + for (; i >= 0; i--, sg--, sge--) { + if (sge->length) + ib_dma_unmap_page(ndev->device, sge->addr, + sge->length, DMA_FROM_DEVICE); + if (sg_page(sg)) + __free_page(sg_page(sg)); + } + return -ENOMEM; +} + static int nvmet_rdma_alloc_cmd(struct nvmet_rdma_device *ndev, struct nvmet_rdma_cmd *c, bool admin) { @@ -200,33 +273,17 @@ static int nvmet_rdma_alloc_cmd(struct nvmet_rdma_device *ndev, c->sge[0].length = sizeof(*c->nvme_cmd); c->sge[0].lkey = ndev->pd->local_dma_lkey; - if (!admin) { - c->inline_page = alloc_pages(GFP_KERNEL, - get_order(NVMET_RDMA_INLINE_DATA_SIZE)); - if (!c->inline_page) - goto out_unmap_cmd; - c->sge[1].addr = ib_dma_map_page(ndev->device, - c->inline_page, 0, NVMET_RDMA_INLINE_DATA_SIZE, - DMA_FROM_DEVICE); - if (ib_dma_mapping_error(ndev->device, c->sge[1].addr)) - goto out_free_inline_page; - c->sge[1].length = NVMET_RDMA_INLINE_DATA_SIZE; - c->sge[1].lkey = ndev->pd->local_dma_lkey; - } + if (!admin && nvmet_rdma_alloc_inline_pages(ndev, c)) + goto out_unmap_cmd; c->cqe.done = nvmet_rdma_recv_done; c->wr.wr_cqe = &c->cqe; c->wr.sg_list = c->sge; - c->wr.num_sge = admin ? 1 : 2; + c->wr.num_sge = admin ? 1 : ndev->inline_page_count + 1; return 0; -out_free_inline_page: - if (!admin) { - __free_pages(c->inline_page, - get_order(NVMET_RDMA_INLINE_DATA_SIZE)); - } out_unmap_cmd: ib_dma_unmap_single(ndev->device, c->sge[0].addr, sizeof(*c->nvme_cmd), DMA_FROM_DEVICE); @@ -240,12 +297,8 @@ out: static void nvmet_rdma_free_cmd(struct nvmet_rdma_device *ndev, struct nvmet_rdma_cmd *c, bool admin) { - if (!admin) { - ib_dma_unmap_page(ndev->device, c->sge[1].addr, - NVMET_RDMA_INLINE_DATA_SIZE, DMA_FROM_DEVICE); - __free_pages(c->inline_page, - get_order(NVMET_RDMA_INLINE_DATA_SIZE)); - } + if (!admin) + nvmet_rdma_free_inline_pages(ndev, c); ib_dma_unmap_single(ndev->device, c->sge[0].addr, sizeof(*c->nvme_cmd), DMA_FROM_DEVICE); kfree(c->nvme_cmd); @@ -383,14 +436,21 @@ static int nvmet_rdma_post_recv(struct nvmet_rdma_device *ndev, struct nvmet_rdma_cmd *cmd) { struct ib_recv_wr *bad_wr; + int ret; ib_dma_sync_single_for_device(ndev->device, cmd->sge[0].addr, cmd->sge[0].length, DMA_FROM_DEVICE); if (ndev->srq) - return ib_post_srq_recv(ndev->srq, &cmd->wr, &bad_wr); - return ib_post_recv(cmd->queue->cm_id->qp, &cmd->wr, &bad_wr); + ret = ib_post_srq_recv(ndev->srq, &cmd->wr, &bad_wr); + else + ret = ib_post_recv(cmd->queue->cm_id->qp, &cmd->wr, &bad_wr); + + if (unlikely(ret)) + pr_err("post_recv cmd failed\n"); + + return ret; } static void nvmet_rdma_process_wr_wait_list(struct nvmet_rdma_queue *queue) @@ -429,7 +489,7 @@ static void nvmet_rdma_release_rsp(struct nvmet_rdma_rsp *rsp) rsp->req.sg_cnt, nvmet_data_dir(&rsp->req)); } - if (rsp->req.sg != &rsp->cmd->inline_sg) + if (rsp->req.sg != rsp->cmd->inline_sg) sgl_free(rsp->req.sg); if (unlikely(!list_empty_careful(&queue->rsp_wr_wait_list))) @@ -493,7 +553,7 @@ static void nvmet_rdma_queue_response(struct nvmet_req *req) rsp->send_sge.addr, rsp->send_sge.length, DMA_TO_DEVICE); - if (ib_post_send(cm_id->qp, first_wr, &bad_wr)) { + if (unlikely(ib_post_send(cm_id->qp, first_wr, &bad_wr))) { pr_err("sending cmd response failed\n"); nvmet_rdma_release_rsp(rsp); } @@ -529,10 +589,25 @@ static void nvmet_rdma_read_data_done(struct ib_cq *cq, struct ib_wc *wc) static void nvmet_rdma_use_inline_sg(struct nvmet_rdma_rsp *rsp, u32 len, u64 off) { - sg_init_table(&rsp->cmd->inline_sg, 1); - sg_set_page(&rsp->cmd->inline_sg, rsp->cmd->inline_page, len, off); - rsp->req.sg = &rsp->cmd->inline_sg; - rsp->req.sg_cnt = 1; + int sg_count = num_pages(len); + struct scatterlist *sg; + int i; + + sg = rsp->cmd->inline_sg; + for (i = 0; i < sg_count; i++, sg++) { + if (i < sg_count - 1) + sg_unmark_end(sg); + else + sg_mark_end(sg); + sg->offset = off; + sg->length = min_t(int, len, PAGE_SIZE - off); + len -= sg->length; + if (!i) + off = 0; + } + + rsp->req.sg = rsp->cmd->inline_sg; + rsp->req.sg_cnt = sg_count; } static u16 nvmet_rdma_map_sgl_inline(struct nvmet_rdma_rsp *rsp) @@ -544,7 +619,7 @@ static u16 nvmet_rdma_map_sgl_inline(struct nvmet_rdma_rsp *rsp) if (!nvme_is_write(rsp->req.cmd)) return NVME_SC_INVALID_FIELD | NVME_SC_DNR; - if (off + len > NVMET_RDMA_INLINE_DATA_SIZE) { + if (off + len > rsp->queue->dev->inline_data_size) { pr_err("invalid inline data offset!\n"); return NVME_SC_SGL_INVALID_OFFSET | NVME_SC_DNR; } @@ -743,7 +818,7 @@ static int nvmet_rdma_init_srq(struct nvmet_rdma_device *ndev) srq_size = 4095; /* XXX: tune */ srq_attr.attr.max_wr = srq_size; - srq_attr.attr.max_sge = 2; + srq_attr.attr.max_sge = 1 + ndev->inline_page_count; srq_attr.attr.srq_limit = 0; srq_attr.srq_type = IB_SRQT_BASIC; srq = ib_create_srq(ndev->pd, &srq_attr); @@ -765,11 +840,16 @@ static int nvmet_rdma_init_srq(struct nvmet_rdma_device *ndev) ndev->srq = srq; ndev->srq_size = srq_size; - for (i = 0; i < srq_size; i++) - nvmet_rdma_post_recv(ndev, &ndev->srq_cmds[i]); + for (i = 0; i < srq_size; i++) { + ret = nvmet_rdma_post_recv(ndev, &ndev->srq_cmds[i]); + if (ret) + goto out_free_cmds; + } return 0; +out_free_cmds: + nvmet_rdma_free_cmds(ndev, ndev->srq_cmds, ndev->srq_size, false); out_destroy_srq: ib_destroy_srq(srq); return ret; @@ -793,7 +873,10 @@ static void nvmet_rdma_free_dev(struct kref *ref) static struct nvmet_rdma_device * nvmet_rdma_find_get_device(struct rdma_cm_id *cm_id) { + struct nvmet_port *port = cm_id->context; struct nvmet_rdma_device *ndev; + int inline_page_count; + int inline_sge_count; int ret; mutex_lock(&device_list_mutex); @@ -807,6 +890,18 @@ nvmet_rdma_find_get_device(struct rdma_cm_id *cm_id) if (!ndev) goto out_err; + inline_page_count = num_pages(port->inline_data_size); + inline_sge_count = max(cm_id->device->attrs.max_sge_rd, + cm_id->device->attrs.max_sge) - 1; + if (inline_page_count > inline_sge_count) { + pr_warn("inline_data_size %d cannot be supported by device %s. Reducing to %lu.\n", + port->inline_data_size, cm_id->device->name, + inline_sge_count * PAGE_SIZE); + port->inline_data_size = inline_sge_count * PAGE_SIZE; + inline_page_count = inline_sge_count; + } + ndev->inline_data_size = port->inline_data_size; + ndev->inline_page_count = inline_page_count; ndev->device = cm_id->device; kref_init(&ndev->ref); @@ -881,7 +976,7 @@ static int nvmet_rdma_create_queue_ib(struct nvmet_rdma_queue *queue) } else { /* +1 for drain */ qp_attr.cap.max_recv_wr = 1 + queue->recv_queue_size; - qp_attr.cap.max_recv_sge = 2; + qp_attr.cap.max_recv_sge = 1 + ndev->inline_page_count; } ret = rdma_create_qp(queue->cm_id, ndev->pd, &qp_attr); @@ -899,13 +994,17 @@ static int nvmet_rdma_create_queue_ib(struct nvmet_rdma_queue *queue) if (!ndev->srq) { for (i = 0; i < queue->recv_queue_size; i++) { queue->cmds[i].queue = queue; - nvmet_rdma_post_recv(ndev, &queue->cmds[i]); + ret = nvmet_rdma_post_recv(ndev, &queue->cmds[i]); + if (ret) + goto err_destroy_qp; } } out: return ret; +err_destroy_qp: + rdma_destroy_qp(queue->cm_id); err_destroy_cq: ib_free_cq(queue->cq); goto out; @@ -1379,6 +1478,15 @@ static int nvmet_rdma_add_port(struct nvmet_port *port) return -EINVAL; } + if (port->inline_data_size < 0) { + port->inline_data_size = NVMET_RDMA_DEFAULT_INLINE_DATA_SIZE; + } else if (port->inline_data_size > NVMET_RDMA_MAX_INLINE_DATA_SIZE) { + pr_warn("inline_data_size %u is too large, reducing to %u\n", + port->inline_data_size, + NVMET_RDMA_MAX_INLINE_DATA_SIZE); + port->inline_data_size = NVMET_RDMA_MAX_INLINE_DATA_SIZE; + } + ret = inet_pton_with_scope(&init_net, af, port->disc_addr.traddr, port->disc_addr.trsvcid, &addr); if (ret) { @@ -1456,7 +1564,6 @@ static void nvmet_rdma_disc_port_addr(struct nvmet_req *req, static const struct nvmet_fabrics_ops nvmet_rdma_ops = { .owner = THIS_MODULE, .type = NVMF_TRTYPE_RDMA, - .sqe_inline_size = NVMET_RDMA_INLINE_DATA_SIZE, .msdbd = 1, .has_keyed_sgls = 1, .add_port = nvmet_rdma_add_port, diff --git a/drivers/scsi/Makefile b/drivers/scsi/Makefile index 80aca2456353..768953881c9e 100644 --- a/drivers/scsi/Makefile +++ b/drivers/scsi/Makefile @@ -21,6 +21,7 @@ CFLAGS_gdth.o = # -DDEBUG_GDTH=2 -D__SERIAL__ -D__COM2__ -DGDTH_STATISTICS obj-$(CONFIG_PCMCIA) += pcmcia/ obj-$(CONFIG_SCSI) += scsi_mod.o +obj-$(CONFIG_BLK_SCSI_REQUEST) += scsi_common.o obj-$(CONFIG_RAID_ATTRS) += raid_class.o @@ -156,7 +157,6 @@ obj-$(CONFIG_SCSI_HISI_SAS) += hisi_sas/ obj-$(CONFIG_SCSI_DEBUG) += scsi_debug.o scsi_mod-y += scsi.o hosts.o scsi_ioctl.o \ scsicam.o scsi_error.o scsi_lib.o -scsi_mod-y += scsi_common.o scsi_mod-$(CONFIG_SCSI_CONSTANTS) += constants.o scsi_mod-$(CONFIG_SCSI_DMA) += scsi_lib_dma.o scsi_mod-y += scsi_scan.o scsi_sysfs.o scsi_devinfo.o diff --git a/drivers/scsi/cxlflash/superpipe.c b/drivers/scsi/cxlflash/superpipe.c index e489d89cbb45..379890c4500b 100644 --- a/drivers/scsi/cxlflash/superpipe.c +++ b/drivers/scsi/cxlflash/superpipe.c @@ -339,7 +339,6 @@ static int read_cap16(struct scsi_device *sdev, struct llun_info *lli) struct scsi_sense_hdr sshdr; u8 *cmd_buf = NULL; u8 *scsi_cmd = NULL; - u8 *sense_buf = NULL; int rc = 0; int result = 0; int retry_cnt = 0; @@ -348,8 +347,7 @@ static int read_cap16(struct scsi_device *sdev, struct llun_info *lli) retry: cmd_buf = kzalloc(CMD_BUFSIZE, GFP_KERNEL); scsi_cmd = kzalloc(MAX_COMMAND_SIZE, GFP_KERNEL); - sense_buf = kzalloc(SCSI_SENSE_BUFFERSIZE, GFP_KERNEL); - if (unlikely(!cmd_buf || !scsi_cmd || !sense_buf)) { + if (unlikely(!cmd_buf || !scsi_cmd)) { rc = -ENOMEM; goto out; } @@ -364,7 +362,7 @@ retry: /* Drop the ioctl read semahpore across lengthy call */ up_read(&cfg->ioctl_rwsem); result = scsi_execute(sdev, scsi_cmd, DMA_FROM_DEVICE, cmd_buf, - CMD_BUFSIZE, sense_buf, &sshdr, to, CMD_RETRIES, + CMD_BUFSIZE, NULL, &sshdr, to, CMD_RETRIES, 0, 0, NULL); down_read(&cfg->ioctl_rwsem); rc = check_state(cfg); @@ -395,7 +393,6 @@ retry: if (retry_cnt++ < 1) { kfree(cmd_buf); kfree(scsi_cmd); - kfree(sense_buf); goto retry; } } @@ -426,7 +423,6 @@ retry: out: kfree(cmd_buf); kfree(scsi_cmd); - kfree(sense_buf); dev_dbg(dev, "%s: maxlba=%lld blklen=%d rc=%d\n", __func__, gli->max_lba, gli->blk_len, rc); diff --git a/drivers/scsi/cxlflash/vlun.c b/drivers/scsi/cxlflash/vlun.c index 66e445a17d6c..2c904bf16b65 100644 --- a/drivers/scsi/cxlflash/vlun.c +++ b/drivers/scsi/cxlflash/vlun.c @@ -426,7 +426,6 @@ static int write_same16(struct scsi_device *sdev, { u8 *cmd_buf = NULL; u8 *scsi_cmd = NULL; - u8 *sense_buf = NULL; int rc = 0; int result = 0; u64 offset = lba; @@ -440,8 +439,7 @@ static int write_same16(struct scsi_device *sdev, cmd_buf = kzalloc(CMD_BUFSIZE, GFP_KERNEL); scsi_cmd = kzalloc(MAX_COMMAND_SIZE, GFP_KERNEL); - sense_buf = kzalloc(SCSI_SENSE_BUFFERSIZE, GFP_KERNEL); - if (unlikely(!cmd_buf || !scsi_cmd || !sense_buf)) { + if (unlikely(!cmd_buf || !scsi_cmd)) { rc = -ENOMEM; goto out; } @@ -457,7 +455,7 @@ static int write_same16(struct scsi_device *sdev, /* Drop the ioctl read semahpore across lengthy call */ up_read(&cfg->ioctl_rwsem); result = scsi_execute(sdev, scsi_cmd, DMA_TO_DEVICE, cmd_buf, - CMD_BUFSIZE, sense_buf, NULL, to, + CMD_BUFSIZE, NULL, NULL, to, CMD_RETRIES, 0, 0, NULL); down_read(&cfg->ioctl_rwsem); rc = check_state(cfg); @@ -482,7 +480,6 @@ static int write_same16(struct scsi_device *sdev, out: kfree(cmd_buf); kfree(scsi_cmd); - kfree(sense_buf); dev_dbg(dev, "%s: returning rc=%d\n", __func__, rc); return rc; } diff --git a/drivers/scsi/mpt3sas/mpt3sas_scsih.c b/drivers/scsi/mpt3sas/mpt3sas_scsih.c index b8d131a455d0..dd738ae5c75b 100644 --- a/drivers/scsi/mpt3sas/mpt3sas_scsih.c +++ b/drivers/scsi/mpt3sas/mpt3sas_scsih.c @@ -4568,7 +4568,7 @@ _scsih_setup_eedp(struct MPT3SAS_ADAPTER *ioc, struct scsi_cmnd *scmd, MPI2_SCSIIO_EEDPFLAGS_CHECK_REFTAG | MPI2_SCSIIO_EEDPFLAGS_CHECK_GUARD; mpi_request->CDB.EEDP32.PrimaryReferenceTag = - cpu_to_be32(scsi_prot_ref_tag(scmd)); + cpu_to_be32(t10_pi_ref_tag(scmd->request)); break; case SCSI_PROT_DIF_TYPE3: diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index 41e9ac9fc138..9cb9a166fa0c 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -238,7 +238,7 @@ void scsi_queue_insert(struct scsi_cmnd *cmd, int reason) /** - * scsi_execute - insert request and wait for the result + * __scsi_execute - insert request and wait for the result * @sdev: scsi device * @cmd: scsi command * @data_direction: data direction @@ -255,7 +255,7 @@ void scsi_queue_insert(struct scsi_cmnd *cmd, int reason) * Returns the scsi_cmnd result field if a command was executed, or a negative * Linux error code if we didn't get that far. */ -int scsi_execute(struct scsi_device *sdev, const unsigned char *cmd, +int __scsi_execute(struct scsi_device *sdev, const unsigned char *cmd, int data_direction, void *buffer, unsigned bufflen, unsigned char *sense, struct scsi_sense_hdr *sshdr, int timeout, int retries, u64 flags, req_flags_t rq_flags, @@ -309,7 +309,7 @@ int scsi_execute(struct scsi_device *sdev, const unsigned char *cmd, return ret; } -EXPORT_SYMBOL(scsi_execute); +EXPORT_SYMBOL(__scsi_execute); /* * Function: scsi_init_cmd_errh() diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c index 9421d9877730..bbebdc3769b0 100644 --- a/drivers/scsi/sd.c +++ b/drivers/scsi/sd.c @@ -1119,7 +1119,7 @@ static int sd_setup_read_write_cmnd(struct scsi_cmnd *SCpnt) SCpnt->cmnd[0] = WRITE_6; if (blk_integrity_rq(rq)) - sd_dif_prepare(SCpnt); + t10_pi_prepare(SCpnt->request, sdkp->protection_type); } else if (rq_data_dir(rq) == READ) { SCpnt->cmnd[0] = READ_6; @@ -2047,8 +2047,10 @@ static int sd_done(struct scsi_cmnd *SCpnt) "sd_done: completed %d of %d bytes\n", good_bytes, scsi_bufflen(SCpnt))); - if (rq_data_dir(SCpnt->request) == READ && scsi_prot_sg_count(SCpnt)) - sd_dif_complete(SCpnt, good_bytes); + if (rq_data_dir(SCpnt->request) == READ && scsi_prot_sg_count(SCpnt) && + good_bytes) + t10_pi_complete(SCpnt->request, sdkp->protection_type, + good_bytes / scsi_prot_interval(SCpnt)); return good_bytes; } diff --git a/drivers/scsi/sd.h b/drivers/scsi/sd.h index 392c7d078ae3..a7d4f50b67d4 100644 --- a/drivers/scsi/sd.h +++ b/drivers/scsi/sd.h @@ -254,21 +254,12 @@ static inline unsigned int sd_prot_flag_mask(unsigned int prot_op) #ifdef CONFIG_BLK_DEV_INTEGRITY extern void sd_dif_config_host(struct scsi_disk *); -extern void sd_dif_prepare(struct scsi_cmnd *scmd); -extern void sd_dif_complete(struct scsi_cmnd *, unsigned int); #else /* CONFIG_BLK_DEV_INTEGRITY */ static inline void sd_dif_config_host(struct scsi_disk *disk) { } -static inline int sd_dif_prepare(struct scsi_cmnd *scmd) -{ - return 0; -} -static inline void sd_dif_complete(struct scsi_cmnd *cmd, unsigned int a) -{ -} #endif /* CONFIG_BLK_DEV_INTEGRITY */ diff --git a/drivers/scsi/sd_dif.c b/drivers/scsi/sd_dif.c index 9035380c0dda..db72c82486e3 100644 --- a/drivers/scsi/sd_dif.c +++ b/drivers/scsi/sd_dif.c @@ -95,116 +95,3 @@ out: blk_integrity_register(disk, &bi); } -/* - * The virtual start sector is the one that was originally submitted - * by the block layer. Due to partitioning, MD/DM cloning, etc. the - * actual physical start sector is likely to be different. Remap - * protection information to match the physical LBA. - * - * From a protocol perspective there's a slight difference between - * Type 1 and 2. The latter uses 32-byte CDBs exclusively, and the - * reference tag is seeded in the CDB. This gives us the potential to - * avoid virt->phys remapping during write. However, at read time we - * don't know whether the virt sector is the same as when we wrote it - * (we could be reading from real disk as opposed to MD/DM device. So - * we always remap Type 2 making it identical to Type 1. - * - * Type 3 does not have a reference tag so no remapping is required. - */ -void sd_dif_prepare(struct scsi_cmnd *scmd) -{ - const int tuple_sz = sizeof(struct t10_pi_tuple); - struct bio *bio; - struct scsi_disk *sdkp; - struct t10_pi_tuple *pi; - u32 phys, virt; - - sdkp = scsi_disk(scmd->request->rq_disk); - - if (sdkp->protection_type == T10_PI_TYPE3_PROTECTION) - return; - - phys = scsi_prot_ref_tag(scmd); - - __rq_for_each_bio(bio, scmd->request) { - struct bio_integrity_payload *bip = bio_integrity(bio); - struct bio_vec iv; - struct bvec_iter iter; - unsigned int j; - - /* Already remapped? */ - if (bip->bip_flags & BIP_MAPPED_INTEGRITY) - break; - - virt = bip_get_seed(bip) & 0xffffffff; - - bip_for_each_vec(iv, bip, iter) { - pi = kmap_atomic(iv.bv_page) + iv.bv_offset; - - for (j = 0; j < iv.bv_len; j += tuple_sz, pi++) { - - if (be32_to_cpu(pi->ref_tag) == virt) - pi->ref_tag = cpu_to_be32(phys); - - virt++; - phys++; - } - - kunmap_atomic(pi); - } - - bip->bip_flags |= BIP_MAPPED_INTEGRITY; - } -} - -/* - * Remap physical sector values in the reference tag to the virtual - * values expected by the block layer. - */ -void sd_dif_complete(struct scsi_cmnd *scmd, unsigned int good_bytes) -{ - const int tuple_sz = sizeof(struct t10_pi_tuple); - struct scsi_disk *sdkp; - struct bio *bio; - struct t10_pi_tuple *pi; - unsigned int j, intervals; - u32 phys, virt; - - sdkp = scsi_disk(scmd->request->rq_disk); - - if (sdkp->protection_type == T10_PI_TYPE3_PROTECTION || good_bytes == 0) - return; - - intervals = good_bytes / scsi_prot_interval(scmd); - phys = scsi_prot_ref_tag(scmd); - - __rq_for_each_bio(bio, scmd->request) { - struct bio_integrity_payload *bip = bio_integrity(bio); - struct bio_vec iv; - struct bvec_iter iter; - - virt = bip_get_seed(bip) & 0xffffffff; - - bip_for_each_vec(iv, bip, iter) { - pi = kmap_atomic(iv.bv_page) + iv.bv_offset; - - for (j = 0; j < iv.bv_len; j += tuple_sz, pi++) { - - if (intervals == 0) { - kunmap_atomic(pi); - return; - } - - if (be32_to_cpu(pi->ref_tag) == phys) - pi->ref_tag = cpu_to_be32(virt); - - virt++; - phys++; - intervals--; - } - - kunmap_atomic(pi); - } - } -} - diff --git a/drivers/scsi/sr_ioctl.c b/drivers/scsi/sr_ioctl.c index 35fab1e18adc..ffcf902da390 100644 --- a/drivers/scsi/sr_ioctl.c +++ b/drivers/scsi/sr_ioctl.c @@ -186,14 +186,13 @@ static int sr_play_trkind(struct cdrom_device_info *cdi, int sr_do_ioctl(Scsi_CD *cd, struct packet_command *cgc) { struct scsi_device *SDev; - struct scsi_sense_hdr sshdr; + struct scsi_sense_hdr local_sshdr, *sshdr = &local_sshdr; int result, err = 0, retries = 0; - unsigned char sense_buffer[SCSI_SENSE_BUFFERSIZE], *senseptr = NULL; SDev = cd->device; - if (cgc->sense) - senseptr = sense_buffer; + if (cgc->sshdr) + sshdr = cgc->sshdr; retry: if (!scsi_block_when_processing_errors(SDev)) { @@ -202,15 +201,12 @@ int sr_do_ioctl(Scsi_CD *cd, struct packet_command *cgc) } result = scsi_execute(SDev, cgc->cmd, cgc->data_direction, - cgc->buffer, cgc->buflen, senseptr, &sshdr, + cgc->buffer, cgc->buflen, NULL, sshdr, cgc->timeout, IOCTL_RETRIES, 0, 0, NULL); - if (cgc->sense) - memcpy(cgc->sense, sense_buffer, sizeof(*cgc->sense)); - /* Minimal error checking. Ignore cases we know about, and report the rest. */ if (driver_byte(result) != 0) { - switch (sshdr.sense_key) { + switch (sshdr->sense_key) { case UNIT_ATTENTION: SDev->changed = 1; if (!cgc->quiet) @@ -221,8 +217,8 @@ int sr_do_ioctl(Scsi_CD *cd, struct packet_command *cgc) err = -ENOMEDIUM; break; case NOT_READY: /* This happens if there is no disc in drive */ - if (sshdr.asc == 0x04 && - sshdr.ascq == 0x01) { + if (sshdr->asc == 0x04 && + sshdr->ascq == 0x01) { /* sense: Logical unit is in process of becoming ready */ if (!cgc->quiet) sr_printk(KERN_INFO, cd, @@ -245,8 +241,8 @@ int sr_do_ioctl(Scsi_CD *cd, struct packet_command *cgc) break; case ILLEGAL_REQUEST: err = -EIO; - if (sshdr.asc == 0x20 && - sshdr.ascq == 0x00) + if (sshdr->asc == 0x20 && + sshdr->ascq == 0x00) /* sense: Invalid command operation code */ err = -EDRIVE_CANT_DO_THIS; break; diff --git a/drivers/scsi/virtio_scsi.c b/drivers/scsi/virtio_scsi.c index 6dc8891ccb74..1c72db94270e 100644 --- a/drivers/scsi/virtio_scsi.c +++ b/drivers/scsi/virtio_scsi.c @@ -513,12 +513,12 @@ static void virtio_scsi_init_hdr_pi(struct virtio_device *vdev, if (sc->sc_data_direction == DMA_TO_DEVICE) cmd_pi->pi_bytesout = cpu_to_virtio32(vdev, - blk_rq_sectors(rq) * - bi->tuple_size); + bio_integrity_bytes(bi, + blk_rq_sectors(rq))); else if (sc->sc_data_direction == DMA_FROM_DEVICE) cmd_pi->pi_bytesin = cpu_to_virtio32(vdev, - blk_rq_sectors(rq) * - bi->tuple_size); + bio_integrity_bytes(bi, + blk_rq_sectors(rq))); } #endif diff --git a/drivers/target/Kconfig b/drivers/target/Kconfig index 4c44d7bed01a..cb6f32ce7de8 100644 --- a/drivers/target/Kconfig +++ b/drivers/target/Kconfig @@ -1,10 +1,10 @@ menuconfig TARGET_CORE tristate "Generic Target Core Mod (TCM) and ConfigFS Infrastructure" - depends on SCSI && BLOCK + depends on BLOCK select CONFIGFS_FS select CRC_T10DIF - select BLK_SCSI_REQUEST # only for scsi_command_size_tbl.. + select BLK_SCSI_REQUEST select SGL_ALLOC default n help @@ -29,6 +29,7 @@ config TCM_FILEIO config TCM_PSCSI tristate "TCM/pSCSI Subsystem Plugin for Linux/SCSI" + depends on SCSI help Say Y here to enable the TCM/pSCSI subsystem plugin for non-buffered passthrough access to Linux/SCSI device diff --git a/drivers/target/loopback/Kconfig b/drivers/target/loopback/Kconfig index abe8ecbcdf06..158ee9d522f7 100644 --- a/drivers/target/loopback/Kconfig +++ b/drivers/target/loopback/Kconfig @@ -1,5 +1,6 @@ config LOOPBACK_TARGET tristate "TCM Virtual SAS target and Linux/SCSI LDD fabric loopback module" + depends on SCSI help Say Y here to enable the TCM Virtual SAS target and Linux/SCSI LLD fabric loopback module. diff --git a/fs/block_dev.c b/fs/block_dev.c index aba25414231a..38b8ce05cbc7 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -666,7 +666,8 @@ int bdev_read_page(struct block_device *bdev, sector_t sector, result = blk_queue_enter(bdev->bd_queue, 0); if (result) return result; - result = ops->rw_page(bdev, sector + get_start_sect(bdev), page, false); + result = ops->rw_page(bdev, sector + get_start_sect(bdev), page, + REQ_OP_READ); blk_queue_exit(bdev->bd_queue); return result; } @@ -704,7 +705,8 @@ int bdev_write_page(struct block_device *bdev, sector_t sector, return result; set_page_writeback(page); - result = ops->rw_page(bdev, sector + get_start_sect(bdev), page, true); + result = ops->rw_page(bdev, sector + get_start_sect(bdev), page, + REQ_OP_WRITE); if (result) { end_page_writeback(page); } else { diff --git a/fs/exofs/ore.c b/fs/exofs/ore.c index 1b8b44637e70..5331a15a61f1 100644 --- a/fs/exofs/ore.c +++ b/fs/exofs/ore.c @@ -873,8 +873,8 @@ static int _write_mirror(struct ore_io_state *ios, int cur_comp) struct bio *bio; if (per_dev != master_dev) { - bio = bio_clone_kmalloc(master_dev->bio, - GFP_KERNEL); + bio = bio_clone_fast(master_dev->bio, + GFP_KERNEL, NULL); if (unlikely(!bio)) { ORE_DBGMSG( "Failed to allocate BIO size=%u\n", diff --git a/fs/ext4/super.c b/fs/ext4/super.c index f7750bc5b85a..5863fd22e90b 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -3529,7 +3529,7 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) sbi->s_sb_block = sb_block; if (sb->s_bdev->bd_part) sbi->s_sectors_written_start = - part_stat_read(sb->s_bdev->bd_part, sectors[1]); + part_stat_read(sb->s_bdev->bd_part, sectors[STAT_WRITE]); /* Cleanup superblock name */ strreplace(sb->s_id, '/', '!'); @@ -4838,7 +4838,8 @@ static int ext4_commit_super(struct super_block *sb, int sync) if (sb->s_bdev->bd_part) es->s_kbytes_written = cpu_to_le64(EXT4_SB(sb)->s_kbytes_written + - ((part_stat_read(sb->s_bdev->bd_part, sectors[1]) - + ((part_stat_read(sb->s_bdev->bd_part, + sectors[STAT_WRITE]) - EXT4_SB(sb)->s_sectors_written_start) >> 1)); else es->s_kbytes_written = diff --git a/fs/ext4/sysfs.c b/fs/ext4/sysfs.c index e60cc5e89023..9212a026a1f1 100644 --- a/fs/ext4/sysfs.c +++ b/fs/ext4/sysfs.c @@ -58,7 +58,8 @@ static ssize_t session_write_kbytes_show(struct ext4_sb_info *sbi, char *buf) if (!sb->s_bdev->bd_part) return snprintf(buf, PAGE_SIZE, "0\n"); return snprintf(buf, PAGE_SIZE, "%lu\n", - (part_stat_read(sb->s_bdev->bd_part, sectors[1]) - + (part_stat_read(sb->s_bdev->bd_part, + sectors[STAT_WRITE]) - sbi->s_sectors_written_start) >> 1); } @@ -70,7 +71,8 @@ static ssize_t lifetime_write_kbytes_show(struct ext4_sb_info *sbi, char *buf) return snprintf(buf, PAGE_SIZE, "0\n"); return snprintf(buf, PAGE_SIZE, "%llu\n", (unsigned long long)(sbi->s_kbytes_written + - ((part_stat_read(sb->s_bdev->bd_part, sectors[1]) - + ((part_stat_read(sb->s_bdev->bd_part, + sectors[STAT_WRITE]) - EXT4_SB(sb)->s_sectors_written_start) >> 1))); } diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 4d8b1de83143..6799c3fc44e3 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -1304,7 +1304,7 @@ static inline bool time_to_inject(struct f2fs_sb_info *sbi, int type) * and the return value is in kbytes. s is of struct f2fs_sb_info. */ #define BD_PART_WRITTEN(s) \ -(((u64)part_stat_read((s)->sb->s_bdev->bd_part, sectors[1]) - \ +(((u64)part_stat_read((s)->sb->s_bdev->bd_part, sectors[STAT_WRITE]) - \ (s)->sectors_written_start) >> 1) static inline void f2fs_update_time(struct f2fs_sb_info *sbi, int type) diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index 3995e926ba3a..17bcff789c08 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -2882,7 +2882,8 @@ try_onemore: /* For write statistics */ if (sb->s_bdev->bd_part) sbi->sectors_written_start = - (u64)part_stat_read(sb->s_bdev->bd_part, sectors[1]); + (u64)part_stat_read(sb->s_bdev->bd_part, + sectors[STAT_WRITE]); /* Read accumulated write IO statistics if exists */ seg_i = CURSEG_I(sbi, CURSEG_HOT_NODE); diff --git a/fs/mpage.c b/fs/mpage.c index b7e7f570733a..b73638db9866 100644 --- a/fs/mpage.c +++ b/fs/mpage.c @@ -51,8 +51,8 @@ static void mpage_end_io(struct bio *bio) bio_for_each_segment_all(bv, bio, i) { struct page *page = bv->bv_page; - page_endio(page, op_is_write(bio_op(bio)), - blk_status_to_errno(bio->bi_status)); + page_endio(page, bio_op(bio), + blk_status_to_errno(bio->bi_status)); } bio_put(bio); diff --git a/include/linux/bio.h b/include/linux/bio.h index f08f5fe7bd08..51371740d2a8 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -429,7 +429,6 @@ extern void bio_put(struct bio *); extern void __bio_clone_fast(struct bio *, struct bio *); extern struct bio *bio_clone_fast(struct bio *, gfp_t, struct bio_set *); -extern struct bio *bio_clone_bioset(struct bio *, gfp_t, struct bio_set *bs); extern struct bio_set fs_bio_set; @@ -443,12 +442,6 @@ static inline struct bio *bio_kmalloc(gfp_t gfp_mask, unsigned int nr_iovecs) return bio_alloc_bioset(gfp_mask, nr_iovecs, NULL); } -static inline struct bio *bio_clone_kmalloc(struct bio *bio, gfp_t gfp_mask) -{ - return bio_clone_bioset(bio, gfp_mask, NULL); - -} - extern blk_qc_t submit_bio(struct bio *); extern void bio_endio(struct bio *); @@ -496,9 +489,9 @@ extern struct bio *bio_copy_kern(struct request_queue *, void *, unsigned int, extern void bio_set_pages_dirty(struct bio *bio); extern void bio_check_pages_dirty(struct bio *bio); -void generic_start_io_acct(struct request_queue *q, int rw, +void generic_start_io_acct(struct request_queue *q, int op, unsigned long sectors, struct hd_struct *part); -void generic_end_io_acct(struct request_queue *q, int rw, +void generic_end_io_acct(struct request_queue *q, int op, struct hd_struct *part, unsigned long start_time); @@ -553,8 +546,16 @@ do { \ #define bio_dev(bio) \ disk_devt((bio)->bi_disk) +#if defined(CONFIG_MEMCG) && defined(CONFIG_BLK_CGROUP) +int bio_associate_blkcg_from_page(struct bio *bio, struct page *page); +#else +static inline int bio_associate_blkcg_from_page(struct bio *bio, + struct page *page) { return 0; } +#endif + #ifdef CONFIG_BLK_CGROUP int bio_associate_blkcg(struct bio *bio, struct cgroup_subsys_state *blkcg_css); +int bio_associate_blkg(struct bio *bio, struct blkcg_gq *blkg); void bio_disassociate_task(struct bio *bio); void bio_clone_blkcg_association(struct bio *dst, struct bio *src); #else /* CONFIG_BLK_CGROUP */ diff --git a/include/linux/blk-cgroup.h b/include/linux/blk-cgroup.h index 6c666fd7de3c..34aec30e06c7 100644 --- a/include/linux/blk-cgroup.h +++ b/include/linux/blk-cgroup.h @@ -35,6 +35,7 @@ enum blkg_rwstat_type { BLKG_RWSTAT_WRITE, BLKG_RWSTAT_SYNC, BLKG_RWSTAT_ASYNC, + BLKG_RWSTAT_DISCARD, BLKG_RWSTAT_NR, BLKG_RWSTAT_TOTAL = BLKG_RWSTAT_NR, @@ -136,6 +137,12 @@ struct blkcg_gq { struct blkg_policy_data *pd[BLKCG_MAX_POLS]; struct rcu_head rcu_head; + + atomic_t use_delay; + atomic64_t delay_nsec; + atomic64_t delay_start; + u64 last_delay; + int last_use; }; typedef struct blkcg_policy_data *(blkcg_pol_alloc_cpd_fn)(gfp_t gfp); @@ -148,6 +155,8 @@ typedef void (blkcg_pol_online_pd_fn)(struct blkg_policy_data *pd); typedef void (blkcg_pol_offline_pd_fn)(struct blkg_policy_data *pd); typedef void (blkcg_pol_free_pd_fn)(struct blkg_policy_data *pd); typedef void (blkcg_pol_reset_pd_stats_fn)(struct blkg_policy_data *pd); +typedef size_t (blkcg_pol_stat_pd_fn)(struct blkg_policy_data *pd, char *buf, + size_t size); struct blkcg_policy { int plid; @@ -167,6 +176,7 @@ struct blkcg_policy { blkcg_pol_offline_pd_fn *pd_offline_fn; blkcg_pol_free_pd_fn *pd_free_fn; blkcg_pol_reset_pd_stats_fn *pd_reset_stats_fn; + blkcg_pol_stat_pd_fn *pd_stat_fn; }; extern struct blkcg blkcg_root; @@ -238,6 +248,42 @@ static inline struct blkcg *bio_blkcg(struct bio *bio) return css_to_blkcg(task_css(current, io_cgrp_id)); } +static inline bool blk_cgroup_congested(void) +{ + struct cgroup_subsys_state *css; + bool ret = false; + + rcu_read_lock(); + css = kthread_blkcg(); + if (!css) + css = task_css(current, io_cgrp_id); + while (css) { + if (atomic_read(&css->cgroup->congestion_count)) { + ret = true; + break; + } + css = css->parent; + } + rcu_read_unlock(); + return ret; +} + +/** + * bio_issue_as_root_blkg - see if this bio needs to be issued as root blkg + * @return: true if this bio needs to be submitted with the root blkg context. + * + * In order to avoid priority inversions we sometimes need to issue a bio as if + * it were attached to the root blkg, and then backcharge to the actual owning + * blkg. The idea is we do bio_blkcg() to look up the actual context for the + * bio and attach the appropriate blkg to the bio. Then we call this helper and + * if it is true run with the root blkg for that queue and then do any + * backcharging to the originating cgroup once the io is complete. + */ +static inline bool bio_issue_as_root_blkg(struct bio *bio) +{ + return (bio->bi_opf & (REQ_META | REQ_SWAP)) != 0; +} + /** * blkcg_parent - get the parent of a blkcg * @blkcg: blkcg of interest @@ -296,6 +342,17 @@ static inline struct blkcg_gq *blkg_lookup(struct blkcg *blkcg, } /** + * blk_queue_root_blkg - return blkg for the (blkcg_root, @q) pair + * @q: request_queue of interest + * + * Lookup blkg for @q at the root level. See also blkg_lookup(). + */ +static inline struct blkcg_gq *blk_queue_root_blkg(struct request_queue *q) +{ + return q->root_blkg; +} + +/** * blkg_to_pdata - get policy private data * @blkg: blkg of interest * @pol: policy of interest @@ -355,6 +412,21 @@ static inline void blkg_get(struct blkcg_gq *blkg) atomic_inc(&blkg->refcnt); } +/** + * blkg_try_get - try and get a blkg reference + * @blkg: blkg to get + * + * This is for use when doing an RCU lookup of the blkg. We may be in the midst + * of freeing this blkg, so we can only use it if the refcnt is not zero. + */ +static inline struct blkcg_gq *blkg_try_get(struct blkcg_gq *blkg) +{ + if (atomic_inc_not_zero(&blkg->refcnt)) + return blkg; + return NULL; +} + + void __blkg_release_rcu(struct rcu_head *rcu); /** @@ -589,7 +661,9 @@ static inline void blkg_rwstat_add(struct blkg_rwstat *rwstat, { struct percpu_counter *cnt; - if (op_is_write(op)) + if (op_is_discard(op)) + cnt = &rwstat->cpu_cnt[BLKG_RWSTAT_DISCARD]; + else if (op_is_write(op)) cnt = &rwstat->cpu_cnt[BLKG_RWSTAT_WRITE]; else cnt = &rwstat->cpu_cnt[BLKG_RWSTAT_READ]; @@ -706,8 +780,14 @@ static inline bool blkcg_bio_issue_check(struct request_queue *q, if (!throtl) { blkg = blkg ?: q->root_blkg; - blkg_rwstat_add(&blkg->stat_bytes, bio->bi_opf, - bio->bi_iter.bi_size); + /* + * If the bio is flagged with BIO_QUEUE_ENTERED it means this + * is a split bio and we would have already accounted for the + * size of the bio. + */ + if (!bio_flagged(bio, BIO_QUEUE_ENTERED)) + blkg_rwstat_add(&blkg->stat_bytes, bio->bi_opf, + bio->bi_iter.bi_size); blkg_rwstat_add(&blkg->stat_ios, bio->bi_opf, 1); } @@ -715,6 +795,59 @@ static inline bool blkcg_bio_issue_check(struct request_queue *q, return !throtl; } +static inline void blkcg_use_delay(struct blkcg_gq *blkg) +{ + if (atomic_add_return(1, &blkg->use_delay) == 1) + atomic_inc(&blkg->blkcg->css.cgroup->congestion_count); +} + +static inline int blkcg_unuse_delay(struct blkcg_gq *blkg) +{ + int old = atomic_read(&blkg->use_delay); + + if (old == 0) + return 0; + + /* + * We do this song and dance because we can race with somebody else + * adding or removing delay. If we just did an atomic_dec we'd end up + * negative and we'd already be in trouble. We need to subtract 1 and + * then check to see if we were the last delay so we can drop the + * congestion count on the cgroup. + */ + while (old) { + int cur = atomic_cmpxchg(&blkg->use_delay, old, old - 1); + if (cur == old) + break; + old = cur; + } + + if (old == 0) + return 0; + if (old == 1) + atomic_dec(&blkg->blkcg->css.cgroup->congestion_count); + return 1; +} + +static inline void blkcg_clear_delay(struct blkcg_gq *blkg) +{ + int old = atomic_read(&blkg->use_delay); + if (!old) + return; + /* We only want 1 person clearing the congestion count for this blkg. */ + while (old) { + int cur = atomic_cmpxchg(&blkg->use_delay, old, 0); + if (cur == old) { + atomic_dec(&blkg->blkcg->css.cgroup->congestion_count); + break; + } + old = cur; + } +} + +void blkcg_add_delay(struct blkcg_gq *blkg, u64 now, u64 delta); +void blkcg_schedule_throttle(struct request_queue *q, bool use_memdelay); +void blkcg_maybe_throttle_current(void); #else /* CONFIG_BLK_CGROUP */ struct blkcg { @@ -734,9 +867,16 @@ struct blkcg_policy { #define blkcg_root_css ((struct cgroup_subsys_state *)ERR_PTR(-EINVAL)) +static inline void blkcg_maybe_throttle_current(void) { } +static inline bool blk_cgroup_congested(void) { return false; } + #ifdef CONFIG_BLOCK +static inline void blkcg_schedule_throttle(struct request_queue *q, bool use_memdelay) { } + static inline struct blkcg_gq *blkg_lookup(struct blkcg *blkcg, void *key) { return NULL; } +static inline struct blkcg_gq *blk_queue_root_blkg(struct request_queue *q) +{ return NULL; } static inline int blkcg_init_queue(struct request_queue *q) { return 0; } static inline void blkcg_drain_queue(struct request_queue *q) { } static inline void blkcg_exit_queue(struct request_queue *q) { } diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index ca3f2c2edd85..1da59c16f637 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -35,10 +35,12 @@ struct blk_mq_hw_ctx { struct sbitmap ctx_map; struct blk_mq_ctx *dispatch_from; + unsigned int dispatch_busy; - struct blk_mq_ctx **ctxs; unsigned int nr_ctx; + struct blk_mq_ctx **ctxs; + spinlock_t dispatch_wait_lock; wait_queue_entry_t dispatch_wait; atomic_t wait_index; diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h index 3c4f390aea4b..f6dfb30737d8 100644 --- a/include/linux/blk_types.h +++ b/include/linux/blk_types.h @@ -179,11 +179,9 @@ struct bio { */ struct io_context *bi_ioc; struct cgroup_subsys_state *bi_css; -#ifdef CONFIG_BLK_DEV_THROTTLING_LOW - void *bi_cg_private; + struct blkcg_gq *bi_blkg; struct bio_issue bi_issue; #endif -#endif union { #if defined(CONFIG_BLK_DEV_INTEGRITY) struct bio_integrity_payload *bi_integrity; /* data integrity */ @@ -329,7 +327,7 @@ enum req_flag_bits { /* for driver use */ __REQ_DRV, - + __REQ_SWAP, /* swapping request. */ __REQ_NR_BITS, /* stops here */ }; @@ -351,6 +349,7 @@ enum req_flag_bits { #define REQ_NOUNMAP (1ULL << __REQ_NOUNMAP) #define REQ_DRV (1ULL << __REQ_DRV) +#define REQ_SWAP (1ULL << __REQ_SWAP) #define REQ_FAILFAST_MASK \ (REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT | REQ_FAILFAST_DRIVER) @@ -358,6 +357,14 @@ enum req_flag_bits { #define REQ_NOMERGE_FLAGS \ (REQ_NOMERGE | REQ_PREFLUSH | REQ_FUA) +enum stat_group { + STAT_READ, + STAT_WRITE, + STAT_DISCARD, + + NR_STAT_GROUPS +}; + #define bio_op(bio) \ ((bio)->bi_opf & REQ_OP_MASK) #define req_op(req) \ @@ -395,6 +402,18 @@ static inline bool op_is_sync(unsigned int op) (op & (REQ_SYNC | REQ_FUA | REQ_PREFLUSH)); } +static inline bool op_is_discard(unsigned int op) +{ + return (op & REQ_OP_MASK) == REQ_OP_DISCARD; +} + +static inline int op_stat_group(unsigned int op) +{ + if (op_is_discard(op)) + return STAT_DISCARD; + return op_is_write(op); +} + typedef unsigned int blk_qc_t; #define BLK_QC_T_NONE -1U #define BLK_QC_T_SHIFT 16 diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 79226ca8f80f..d6869e0e2b64 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -27,8 +27,6 @@ #include <linux/percpu-refcount.h> #include <linux/scatterlist.h> #include <linux/blkzoned.h> -#include <linux/seqlock.h> -#include <linux/u64_stats_sync.h> struct module; struct scsi_ioctl_command; @@ -42,7 +40,7 @@ struct bsg_job; struct blkcg_gq; struct blk_flush_queue; struct pr_ops; -struct rq_wb; +struct rq_qos; struct blk_queue_stats; struct blk_stat_callback; @@ -442,10 +440,8 @@ struct request_queue { int nr_rqs[2]; /* # allocated [a]sync rqs */ int nr_rqs_elvpriv; /* # allocated rqs w/ elvpriv */ - atomic_t shared_hctx_restart; - struct blk_queue_stats *stats; - struct rq_wb *rq_wb; + struct rq_qos *rq_qos; /* * If blkcg is not used, @q->root_rl serves all requests. If blkcg @@ -592,6 +588,7 @@ struct request_queue { struct queue_limits limits; +#ifdef CONFIG_BLK_DEV_ZONED /* * Zoned block device information for request dispatch control. * nr_zones is the total number of zones of the device. This is always @@ -612,6 +609,7 @@ struct request_queue { unsigned int nr_zones; unsigned long *seq_zones_bitmap; unsigned long *seq_zones_wlock; +#endif /* CONFIG_BLK_DEV_ZONED */ /* * sg stuff @@ -800,11 +798,7 @@ static inline unsigned int blk_queue_zone_sectors(struct request_queue *q) return blk_queue_is_zoned(q) ? q->limits.chunk_sectors : 0; } -static inline unsigned int blk_queue_nr_zones(struct request_queue *q) -{ - return q->nr_zones; -} - +#ifdef CONFIG_BLK_DEV_ZONED static inline unsigned int blk_queue_zone_no(struct request_queue *q, sector_t sector) { @@ -820,6 +814,7 @@ static inline bool blk_queue_zone_is_seq(struct request_queue *q, return false; return test_bit(blk_queue_zone_no(q, sector), q->seq_zones_bitmap); } +#endif /* CONFIG_BLK_DEV_ZONED */ static inline bool rq_is_sync(struct request *rq) { @@ -1070,6 +1065,7 @@ static inline unsigned int blk_rq_cur_sectors(const struct request *rq) return blk_rq_cur_bytes(rq) >> SECTOR_SHIFT; } +#ifdef CONFIG_BLK_DEV_ZONED static inline unsigned int blk_rq_zone_no(struct request *rq) { return blk_queue_zone_no(rq->q, blk_rq_pos(rq)); @@ -1079,6 +1075,7 @@ static inline unsigned int blk_rq_zone_is_seq(struct request *rq) { return blk_queue_zone_is_seq(rq->q, blk_rq_pos(rq)); } +#endif /* CONFIG_BLK_DEV_ZONED */ /* * Some commands like WRITE SAME have a payload or data transfer size which @@ -1437,8 +1434,6 @@ enum blk_default_limits { BLK_SEG_BOUNDARY_MASK = 0xFFFFFFFFUL, }; -#define blkdev_entry_to_request(entry) list_entry((entry), struct request, queuelist) - static inline unsigned long queue_segment_boundary(struct request_queue *q) { return q->limits.seg_boundary_mask; @@ -1639,15 +1634,6 @@ static inline unsigned int bdev_zone_sectors(struct block_device *bdev) return 0; } -static inline unsigned int bdev_nr_zones(struct block_device *bdev) -{ - struct request_queue *q = bdev_get_queue(bdev); - - if (q) - return blk_queue_nr_zones(q); - return 0; -} - static inline int queue_dma_alignment(struct request_queue *q) { return q ? q->dma_alignment : 511; @@ -1877,6 +1863,28 @@ static inline bool integrity_req_gap_front_merge(struct request *req, bip_next->bip_vec[0].bv_offset); } +/** + * bio_integrity_intervals - Return number of integrity intervals for a bio + * @bi: blk_integrity profile for device + * @sectors: Size of the bio in 512-byte sectors + * + * Description: The block layer calculates everything in 512 byte + * sectors but integrity metadata is done in terms of the data integrity + * interval size of the storage device. Convert the block layer sectors + * to the appropriate number of integrity intervals. + */ +static inline unsigned int bio_integrity_intervals(struct blk_integrity *bi, + unsigned int sectors) +{ + return sectors >> (bi->interval_exp - 9); +} + +static inline unsigned int bio_integrity_bytes(struct blk_integrity *bi, + unsigned int sectors) +{ + return bio_integrity_intervals(bi, sectors) * bi->tuple_size; +} + #else /* CONFIG_BLK_DEV_INTEGRITY */ struct bio; @@ -1950,12 +1958,24 @@ static inline bool integrity_req_gap_front_merge(struct request *req, return false; } +static inline unsigned int bio_integrity_intervals(struct blk_integrity *bi, + unsigned int sectors) +{ + return 0; +} + +static inline unsigned int bio_integrity_bytes(struct blk_integrity *bi, + unsigned int sectors) +{ + return 0; +} + #endif /* CONFIG_BLK_DEV_INTEGRITY */ struct block_device_operations { int (*open) (struct block_device *, fmode_t); void (*release) (struct gendisk *, fmode_t); - int (*rw_page)(struct block_device *, sector_t, struct page *, bool); + int (*rw_page)(struct block_device *, sector_t, struct page *, unsigned int); int (*ioctl) (struct block_device *, fmode_t, unsigned, unsigned long); int (*compat_ioctl) (struct block_device *, fmode_t, unsigned, unsigned long); unsigned int (*check_events) (struct gendisk *disk, diff --git a/include/linux/cdrom.h b/include/linux/cdrom.h index e75dfd1f1dec..528271c60018 100644 --- a/include/linux/cdrom.h +++ b/include/linux/cdrom.h @@ -13,6 +13,7 @@ #include <linux/fs.h> /* not really needed, later.. */ #include <linux/list.h> +#include <scsi/scsi_common.h> #include <uapi/linux/cdrom.h> struct packet_command @@ -21,7 +22,7 @@ struct packet_command unsigned char *buffer; unsigned int buflen; int stat; - struct request_sense *sense; + struct scsi_sense_hdr *sshdr; unsigned char data_direction; int quiet; int timeout; diff --git a/include/linux/cgroup-defs.h b/include/linux/cgroup-defs.h index c0e68f903011..ff20b677fb9f 100644 --- a/include/linux/cgroup-defs.h +++ b/include/linux/cgroup-defs.h @@ -438,6 +438,9 @@ struct cgroup { /* used to store eBPF programs */ struct cgroup_bpf bpf; + /* If there is block congestion on this cgroup. */ + atomic_t congestion_count; + /* ids of the ancestors at each level including self */ int ancestor_ids[]; }; diff --git a/include/linux/genhd.h b/include/linux/genhd.h index 6cb8a5789668..57864422a2c8 100644 --- a/include/linux/genhd.h +++ b/include/linux/genhd.h @@ -16,6 +16,7 @@ #include <linux/slab.h> #include <linux/percpu-refcount.h> #include <linux/uuid.h> +#include <linux/blk_types.h> #ifdef CONFIG_BLOCK @@ -82,10 +83,10 @@ struct partition { } __attribute__((packed)); struct disk_stats { - unsigned long sectors[2]; /* READs and WRITEs */ - unsigned long ios[2]; - unsigned long merges[2]; - unsigned long ticks[2]; + unsigned long sectors[NR_STAT_GROUPS]; + unsigned long ios[NR_STAT_GROUPS]; + unsigned long merges[NR_STAT_GROUPS]; + unsigned long ticks[NR_STAT_GROUPS]; unsigned long io_ticks; unsigned long time_in_queue; }; @@ -353,6 +354,11 @@ static inline void free_part_stats(struct hd_struct *part) #endif /* CONFIG_SMP */ +#define part_stat_read_accum(part, field) \ + (part_stat_read(part, field[STAT_READ]) + \ + part_stat_read(part, field[STAT_WRITE]) + \ + part_stat_read(part, field[STAT_DISCARD])) + #define part_stat_add(cpu, part, field, addnd) do { \ __part_stat_add((cpu), (part), field, addnd); \ if ((part)->partno) \ diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 6c6fb116e925..680d3395fc83 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -317,6 +317,9 @@ enum mem_cgroup_protection mem_cgroup_protected(struct mem_cgroup *root, int mem_cgroup_try_charge(struct page *page, struct mm_struct *mm, gfp_t gfp_mask, struct mem_cgroup **memcgp, bool compound); +int mem_cgroup_try_charge_delay(struct page *page, struct mm_struct *mm, + gfp_t gfp_mask, struct mem_cgroup **memcgp, + bool compound); void mem_cgroup_commit_charge(struct page *page, struct mem_cgroup *memcg, bool lrucare, bool compound); void mem_cgroup_cancel_charge(struct page *page, struct mem_cgroup *memcg, @@ -789,6 +792,16 @@ static inline int mem_cgroup_try_charge(struct page *page, struct mm_struct *mm, return 0; } +static inline int mem_cgroup_try_charge_delay(struct page *page, + struct mm_struct *mm, + gfp_t gfp_mask, + struct mem_cgroup **memcgp, + bool compound) +{ + *memcgp = NULL; + return 0; +} + static inline void mem_cgroup_commit_charge(struct page *page, struct mem_cgroup *memcg, bool lrucare, bool compound) diff --git a/include/linux/nvme.h b/include/linux/nvme.h index 2950ce957656..68e91ef5494c 100644 --- a/include/linux/nvme.h +++ b/include/linux/nvme.h @@ -242,7 +242,12 @@ struct nvme_id_ctrl { __le32 sanicap; __le32 hmminds; __le16 hmmaxd; - __u8 rsvd338[174]; + __u8 rsvd338[4]; + __u8 anatt; + __u8 anacap; + __le32 anagrpmax; + __le32 nanagrpid; + __u8 rsvd352[160]; __u8 sqes; __u8 cqes; __le16 maxcmd; @@ -254,11 +259,12 @@ struct nvme_id_ctrl { __le16 awun; __le16 awupf; __u8 nvscc; - __u8 rsvd531; + __u8 nwpc; __le16 acwu; __u8 rsvd534[2]; __le32 sgls; - __u8 rsvd540[228]; + __le32 mnan; + __u8 rsvd544[224]; char subnqn[256]; __u8 rsvd1024[768]; __le32 ioccsz; @@ -312,7 +318,11 @@ struct nvme_id_ns { __le16 nabspf; __le16 noiob; __u8 nvmcap[16]; - __u8 rsvd64[40]; + __u8 rsvd64[28]; + __le32 anagrpid; + __u8 rsvd96[3]; + __u8 nsattr; + __u8 rsvd100[4]; __u8 nguid[16]; __u8 eui64[8]; struct nvme_lbaf lbaf[16]; @@ -425,6 +435,32 @@ struct nvme_effects_log { __u8 resv[2048]; }; +enum nvme_ana_state { + NVME_ANA_OPTIMIZED = 0x01, + NVME_ANA_NONOPTIMIZED = 0x02, + NVME_ANA_INACCESSIBLE = 0x03, + NVME_ANA_PERSISTENT_LOSS = 0x04, + NVME_ANA_CHANGE = 0x0f, +}; + +struct nvme_ana_group_desc { + __le32 grpid; + __le32 nnsids; + __le64 chgcnt; + __u8 state; + __u8 rsvd17[15]; + __le32 nsids[]; +}; + +/* flag for the log specific field of the ANA log */ +#define NVME_ANA_LOG_RGO (1 << 0) + +struct nvme_ana_rsp_hdr { + __le64 chgcnt; + __le16 ngrps; + __le16 rsvd10[3]; +}; + enum { NVME_SMART_CRIT_SPARE = 1 << 0, NVME_SMART_CRIT_TEMPERATURE = 1 << 1, @@ -444,11 +480,13 @@ enum { enum { NVME_AER_NOTICE_NS_CHANGED = 0x00, NVME_AER_NOTICE_FW_ACT_STARTING = 0x01, + NVME_AER_NOTICE_ANA = 0x03, }; enum { NVME_AEN_CFG_NS_ATTR = 1 << 8, NVME_AEN_CFG_FW_ACT = 1 << 9, + NVME_AEN_CFG_ANA_CHANGE = 1 << 11, }; struct nvme_lba_range_type { @@ -749,15 +787,22 @@ enum { NVME_FEAT_HOST_MEM_BUF = 0x0d, NVME_FEAT_TIMESTAMP = 0x0e, NVME_FEAT_KATO = 0x0f, + NVME_FEAT_HCTM = 0x10, + NVME_FEAT_NOPSC = 0x11, + NVME_FEAT_RRL = 0x12, + NVME_FEAT_PLM_CONFIG = 0x13, + NVME_FEAT_PLM_WINDOW = 0x14, NVME_FEAT_SW_PROGRESS = 0x80, NVME_FEAT_HOST_ID = 0x81, NVME_FEAT_RESV_MASK = 0x82, NVME_FEAT_RESV_PERSIST = 0x83, + NVME_FEAT_WRITE_PROTECT = 0x84, NVME_LOG_ERROR = 0x01, NVME_LOG_SMART = 0x02, NVME_LOG_FW_SLOT = 0x03, NVME_LOG_CHANGED_NS = 0x04, NVME_LOG_CMD_EFFECTS = 0x05, + NVME_LOG_ANA = 0x0c, NVME_LOG_DISC = 0x70, NVME_LOG_RESERVATION = 0x80, NVME_FWACT_REPL = (0 << 3), @@ -765,6 +810,14 @@ enum { NVME_FWACT_ACTV = (2 << 3), }; +/* NVMe Namespace Write Protect State */ +enum { + NVME_NS_NO_WRITE_PROTECT = 0, + NVME_NS_WRITE_PROTECT, + NVME_NS_WRITE_PROTECT_POWER_CYCLE, + NVME_NS_WRITE_PROTECT_PERMANENT, +}; + #define NVME_MAX_CHANGED_NAMESPACES 1024 struct nvme_identify { @@ -880,7 +933,7 @@ struct nvme_get_log_page_command { __u64 rsvd2[2]; union nvme_data_ptr dptr; __u8 lid; - __u8 rsvd10; + __u8 lsp; /* upper 4 bits reserved */ __le16 numdl; __le16 numdu; __u16 rsvd11; @@ -1111,6 +1164,8 @@ enum { NVME_SC_SGL_INVALID_OFFSET = 0x16, NVME_SC_SGL_INVALID_SUBTYPE = 0x17, + NVME_SC_NS_WRITE_PROTECTED = 0x20, + NVME_SC_LBA_RANGE = 0x80, NVME_SC_CAP_EXCEEDED = 0x81, NVME_SC_NS_NOT_READY = 0x82, @@ -1180,6 +1235,13 @@ enum { NVME_SC_ACCESS_DENIED = 0x286, NVME_SC_UNWRITTEN_BLOCK = 0x287, + /* + * Path-related Errors: + */ + NVME_SC_ANA_PERSISTENT_LOSS = 0x301, + NVME_SC_ANA_INACCESSIBLE = 0x302, + NVME_SC_ANA_TRANSITION = 0x303, + NVME_SC_DNR = 0x4000, }; diff --git a/include/linux/sched.h b/include/linux/sched.h index dac5086e3815..95a5018c338e 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -734,6 +734,10 @@ struct task_struct { /* disallow userland-initiated cgroup migration */ unsigned no_cgroup_migration:1; #endif +#ifdef CONFIG_BLK_CGROUP + /* to be used once the psi infrastructure lands upstream. */ + unsigned use_memdelay:1; +#endif unsigned long atomic_flags; /* Flags requiring atomic access. */ @@ -1150,6 +1154,10 @@ struct task_struct { unsigned int memcg_nr_pages_over_high; #endif +#ifdef CONFIG_BLK_CGROUP + struct request_queue *throttle_queue; +#endif + #ifdef CONFIG_UPROBES struct uprobe_task *utask; #endif diff --git a/include/linux/swap.h b/include/linux/swap.h index c063443d8638..1a8bd05a335e 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -629,7 +629,6 @@ static inline int mem_cgroup_swappiness(struct mem_cgroup *memcg) return memcg->swappiness; } - #else static inline int mem_cgroup_swappiness(struct mem_cgroup *mem) { @@ -637,6 +636,16 @@ static inline int mem_cgroup_swappiness(struct mem_cgroup *mem) } #endif +#if defined(CONFIG_SWAP) && defined(CONFIG_MEMCG) && defined(CONFIG_BLK_CGROUP) +extern void mem_cgroup_throttle_swaprate(struct mem_cgroup *memcg, int node, + gfp_t gfp_mask); +#else +static inline void mem_cgroup_throttle_swaprate(struct mem_cgroup *memcg, + int node, gfp_t gfp_mask) +{ +} +#endif + #ifdef CONFIG_MEMCG_SWAP extern void mem_cgroup_swapout(struct page *page, swp_entry_t entry); extern int mem_cgroup_try_charge_swap(struct page *page, swp_entry_t entry); diff --git a/include/linux/t10-pi.h b/include/linux/t10-pi.h index c6aa8a3c42ed..b9626aa7e90c 100644 --- a/include/linux/t10-pi.h +++ b/include/linux/t10-pi.h @@ -37,9 +37,33 @@ struct t10_pi_tuple { #define T10_PI_APP_ESCAPE cpu_to_be16(0xffff) #define T10_PI_REF_ESCAPE cpu_to_be32(0xffffffff) +static inline u32 t10_pi_ref_tag(struct request *rq) +{ +#ifdef CONFIG_BLK_DEV_INTEGRITY + return blk_rq_pos(rq) >> + (rq->q->integrity.interval_exp - 9) & 0xffffffff; +#else + return -1U; +#endif +} + extern const struct blk_integrity_profile t10_pi_type1_crc; extern const struct blk_integrity_profile t10_pi_type1_ip; extern const struct blk_integrity_profile t10_pi_type3_crc; extern const struct blk_integrity_profile t10_pi_type3_ip; +#ifdef CONFIG_BLK_DEV_INTEGRITY +extern void t10_pi_prepare(struct request *rq, u8 protection_type); +extern void t10_pi_complete(struct request *rq, u8 protection_type, + unsigned int intervals); +#else +static inline void t10_pi_complete(struct request *rq, u8 protection_type, + unsigned int intervals) +{ +} +static inline void t10_pi_prepare(struct request *rq, u8 protection_type) +{ +} +#endif + #endif diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h index 4a8841963c2e..05589a3e37f4 100644 --- a/include/linux/tracehook.h +++ b/include/linux/tracehook.h @@ -51,6 +51,7 @@ #include <linux/security.h> #include <linux/task_work.h> #include <linux/memcontrol.h> +#include <linux/blk-cgroup.h> struct linux_binprm; /* @@ -192,6 +193,7 @@ static inline void tracehook_notify_resume(struct pt_regs *regs) task_work_run(); mem_cgroup_handle_over_high(); + blkcg_maybe_throttle_current(); } #endif /* <linux/tracehook.h> */ diff --git a/include/scsi/scsi_cmnd.h b/include/scsi/scsi_cmnd.h index aaf1e971c6a3..c891ada3c5c2 100644 --- a/include/scsi/scsi_cmnd.h +++ b/include/scsi/scsi_cmnd.h @@ -4,6 +4,7 @@ #include <linux/dma-mapping.h> #include <linux/blkdev.h> +#include <linux/t10-pi.h> #include <linux/list.h> #include <linux/types.h> #include <linux/timer.h> @@ -14,8 +15,6 @@ struct Scsi_Host; struct scsi_driver; -#include <scsi/scsi_device.h> - /* * MAX_COMMAND_SIZE is: * The longest fixed-length SCSI CDB as per the SCSI standard. @@ -120,11 +119,11 @@ struct scsi_cmnd { struct request *request; /* The command we are working on */ -#define SCSI_SENSE_BUFFERSIZE 96 unsigned char *sense_buffer; /* obtained by REQUEST SENSE when * CHECK CONDITION is received on original - * command (auto-sense) */ + * command (auto-sense). Length must be + * SCSI_SENSE_BUFFERSIZE bytes. */ /* Low-level done function - can be used by low-level driver to point * to completion function. Not used by mid/upper level code. */ @@ -313,12 +312,6 @@ static inline unsigned int scsi_prot_interval(struct scsi_cmnd *scmd) return scmd->device->sector_size; } -static inline u32 scsi_prot_ref_tag(struct scsi_cmnd *scmd) -{ - return blk_rq_pos(scmd->request) >> - (ilog2(scsi_prot_interval(scmd)) - 9) & 0xffffffff; -} - static inline unsigned scsi_prot_sg_count(struct scsi_cmnd *cmd) { return cmd->prot_sdb ? cmd->prot_sdb->table.nents : 0; diff --git a/include/scsi/scsi_device.h b/include/scsi/scsi_device.h index 4c36af6edd79..202f4d6a4342 100644 --- a/include/scsi/scsi_device.h +++ b/include/scsi/scsi_device.h @@ -17,6 +17,8 @@ struct scsi_sense_hdr; typedef __u64 __bitwise blist_flags_t; +#define SCSI_SENSE_BUFFERSIZE 96 + struct scsi_mode_data { __u32 length; __u16 block_descriptor_length; @@ -426,11 +428,21 @@ extern const char *scsi_device_state_name(enum scsi_device_state); extern int scsi_is_sdev_device(const struct device *); extern int scsi_is_target_device(const struct device *); extern void scsi_sanitize_inquiry_string(unsigned char *s, int len); -extern int scsi_execute(struct scsi_device *sdev, const unsigned char *cmd, +extern int __scsi_execute(struct scsi_device *sdev, const unsigned char *cmd, int data_direction, void *buffer, unsigned bufflen, unsigned char *sense, struct scsi_sense_hdr *sshdr, int timeout, int retries, u64 flags, req_flags_t rq_flags, int *resid); +/* Make sure any sense buffer is the correct size. */ +#define scsi_execute(sdev, cmd, data_direction, buffer, bufflen, sense, \ + sshdr, timeout, retries, flags, rq_flags, resid) \ +({ \ + BUILD_BUG_ON((sense) != NULL && \ + sizeof(sense) != SCSI_SENSE_BUFFERSIZE); \ + __scsi_execute(sdev, cmd, data_direction, buffer, bufflen, \ + sense, sshdr, timeout, retries, flags, rq_flags, \ + resid); \ +}) static inline int scsi_execute_req(struct scsi_device *sdev, const unsigned char *cmd, int data_direction, void *buffer, unsigned bufflen, struct scsi_sense_hdr *sshdr, int timeout, diff --git a/include/uapi/linux/bcache.h b/include/uapi/linux/bcache.h index 821f71a2e48f..8d19e02d752a 100644 --- a/include/uapi/linux/bcache.h +++ b/include/uapi/linux/bcache.h @@ -195,7 +195,7 @@ struct cache_sb { }; }; - __u32 last_mount; /* time_t */ + __u32 last_mount; /* time overflow in y2106 */ __u16 first_bucket; union { @@ -318,7 +318,7 @@ struct uuid_entry { struct { __u8 uuid[16]; __u8 label[32]; - __u32 first_reg; + __u32 first_reg; /* time overflow in y2106 */ __u32 last_reg; __u32 invalidated; diff --git a/include/uapi/linux/blkzoned.h b/include/uapi/linux/blkzoned.h index e3c70fe6bf0f..ff5a5db8906a 100644 --- a/include/uapi/linux/blkzoned.h +++ b/include/uapi/linux/blkzoned.h @@ -117,7 +117,7 @@ struct blk_zone_report { __u32 nr_zones; __u8 reserved[4]; struct blk_zone zones[0]; -} __packed; +}; /** * struct blk_zone_range - BLKRESETZONE ioctl request diff --git a/kernel/fork.c b/kernel/fork.c index 9d8d0e016fc6..33112315b5c0 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -866,6 +866,11 @@ static struct task_struct *dup_task_struct(struct task_struct *orig, int node) tsk->fail_nth = 0; #endif +#ifdef CONFIG_BLK_CGROUP + tsk->throttle_queue = NULL; + tsk->use_memdelay = 0; +#endif + return tsk; free_stack: diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index 987d9a9ae283..b951aa1fac61 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -494,6 +494,9 @@ static int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev, if (!buts->buf_size || !buts->buf_nr) return -EINVAL; + if (!blk_debugfs_root) + return -ENOENT; + strncpy(buts->name, name, BLKTRACE_BDEV_SIZE); buts->name[BLKTRACE_BDEV_SIZE - 1] = '\0'; @@ -518,9 +521,6 @@ static int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev, ret = -ENOENT; - if (!blk_debugfs_root) - goto err; - dir = debugfs_lookup(buts->name, blk_debugfs_root); if (!dir) bt->dir = dir = debugfs_create_dir(buts->name, blk_debugfs_root); diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 25346bd99364..a9e1e093df51 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -552,7 +552,7 @@ static int __do_huge_pmd_anonymous_page(struct vm_fault *vmf, struct page *page, VM_BUG_ON_PAGE(!PageCompound(page), page); - if (mem_cgroup_try_charge(page, vma->vm_mm, gfp, &memcg, true)) { + if (mem_cgroup_try_charge_delay(page, vma->vm_mm, gfp, &memcg, true)) { put_page(page); count_vm_event(THP_FAULT_FALLBACK); return VM_FAULT_FALLBACK; @@ -1142,7 +1142,7 @@ static int do_huge_pmd_wp_page_fallback(struct vm_fault *vmf, pmd_t orig_pmd, pages[i] = alloc_page_vma_node(GFP_HIGHUSER_MOVABLE, vma, vmf->address, page_to_nid(page)); if (unlikely(!pages[i] || - mem_cgroup_try_charge(pages[i], vma->vm_mm, + mem_cgroup_try_charge_delay(pages[i], vma->vm_mm, GFP_KERNEL, &memcg, false))) { if (pages[i]) put_page(pages[i]); @@ -1312,7 +1312,7 @@ alloc: goto out; } - if (unlikely(mem_cgroup_try_charge(new_page, vma->vm_mm, + if (unlikely(mem_cgroup_try_charge_delay(new_page, vma->vm_mm, huge_gfp, &memcg, true))) { put_page(new_page); split_huge_pmd(vma, vmf->pmd, vmf->address); diff --git a/mm/memcontrol.c b/mm/memcontrol.c index b2173f7e5164..b836e7f00309 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -5600,6 +5600,19 @@ out: return ret; } +int mem_cgroup_try_charge_delay(struct page *page, struct mm_struct *mm, + gfp_t gfp_mask, struct mem_cgroup **memcgp, + bool compound) +{ + struct mem_cgroup *memcg; + int ret; + + ret = mem_cgroup_try_charge(page, mm, gfp_mask, memcgp, compound); + memcg = *memcgp; + mem_cgroup_throttle_swaprate(memcg, page_to_nid(page), gfp_mask); + return ret; +} + /** * mem_cgroup_commit_charge - commit a page charge * @page: page to charge diff --git a/mm/memory.c b/mm/memory.c index 6d175057cfd0..348279ff6e51 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -2524,7 +2524,7 @@ static int wp_page_copy(struct vm_fault *vmf) cow_user_page(new_page, old_page, vmf->address, vma); } - if (mem_cgroup_try_charge(new_page, mm, GFP_KERNEL, &memcg, false)) + if (mem_cgroup_try_charge_delay(new_page, mm, GFP_KERNEL, &memcg, false)) goto oom_free_new; __SetPageUptodate(new_page); @@ -3024,8 +3024,8 @@ int do_swap_page(struct vm_fault *vmf) goto out_page; } - if (mem_cgroup_try_charge(page, vma->vm_mm, GFP_KERNEL, - &memcg, false)) { + if (mem_cgroup_try_charge_delay(page, vma->vm_mm, GFP_KERNEL, + &memcg, false)) { ret = VM_FAULT_OOM; goto out_page; } @@ -3186,7 +3186,8 @@ static int do_anonymous_page(struct vm_fault *vmf) if (!page) goto oom; - if (mem_cgroup_try_charge(page, vma->vm_mm, GFP_KERNEL, &memcg, false)) + if (mem_cgroup_try_charge_delay(page, vma->vm_mm, GFP_KERNEL, &memcg, + false)) goto oom_free_page; /* @@ -3682,7 +3683,7 @@ static int do_cow_fault(struct vm_fault *vmf) if (!vmf->cow_page) return VM_FAULT_OOM; - if (mem_cgroup_try_charge(vmf->cow_page, vma->vm_mm, GFP_KERNEL, + if (mem_cgroup_try_charge_delay(vmf->cow_page, vma->vm_mm, GFP_KERNEL, &vmf->memcg, false)) { put_page(vmf->cow_page); return VM_FAULT_OOM; diff --git a/mm/page_io.c b/mm/page_io.c index b41cf9644585..aafd19ec1db4 100644 --- a/mm/page_io.c +++ b/mm/page_io.c @@ -338,7 +338,8 @@ int __swap_writepage(struct page *page, struct writeback_control *wbc, ret = -ENOMEM; goto out; } - bio->bi_opf = REQ_OP_WRITE | wbc_to_write_flags(wbc); + bio->bi_opf = REQ_OP_WRITE | REQ_SWAP | wbc_to_write_flags(wbc); + bio_associate_blkcg_from_page(bio, page); count_swpout_vm_event(page); set_page_writeback(page); unlock_page(page); diff --git a/mm/readahead.c b/mm/readahead.c index e273f0de3376..a59ea70527b9 100644 --- a/mm/readahead.c +++ b/mm/readahead.c @@ -19,6 +19,7 @@ #include <linux/syscalls.h> #include <linux/file.h> #include <linux/mm_inline.h> +#include <linux/blk-cgroup.h> #include "internal.h" @@ -385,6 +386,7 @@ ondemand_readahead(struct address_space *mapping, { struct backing_dev_info *bdi = inode_to_bdi(mapping->host); unsigned long max_pages = ra->ra_pages; + unsigned long add_pages; pgoff_t prev_offset; /* @@ -474,10 +476,17 @@ readit: * Will this read hit the readahead marker made by itself? * If so, trigger the readahead marker hit now, and merge * the resulted next readahead window into the current one. + * Take care of maximum IO pages as above. */ if (offset == ra->start && ra->size == ra->async_size) { - ra->async_size = get_next_ra_size(ra, max_pages); - ra->size += ra->async_size; + add_pages = get_next_ra_size(ra, max_pages); + if (ra->size + add_pages <= max_pages) { + ra->async_size = add_pages; + ra->size += add_pages; + } else { + ra->size = max_pages; + ra->async_size = max_pages >> 1; + } } return ra_submit(ra, mapping, filp); @@ -505,6 +514,9 @@ void page_cache_sync_readahead(struct address_space *mapping, if (!ra->ra_pages) return; + if (blk_cgroup_congested()) + return; + /* be dumb */ if (filp && (filp->f_mode & FMODE_RANDOM)) { force_page_cache_readahead(mapping, filp, offset, req_size); @@ -555,6 +567,9 @@ page_cache_async_readahead(struct address_space *mapping, if (inode_read_congested(mapping->host)) return; + if (blk_cgroup_congested()) + return; + /* do read-ahead */ ondemand_readahead(mapping, ra, filp, true, offset, req_size); } diff --git a/mm/shmem.c b/mm/shmem.c index 96bcc51fb9ec..06ebe17bb924 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -1239,8 +1239,8 @@ int shmem_unuse(swp_entry_t swap, struct page *page) * the shmem_swaplist_mutex which might hold up shmem_writepage(). * Charged back to the user (not to caller) when swap account is used. */ - error = mem_cgroup_try_charge(page, current->mm, GFP_KERNEL, &memcg, - false); + error = mem_cgroup_try_charge_delay(page, current->mm, GFP_KERNEL, + &memcg, false); if (error) goto out; /* No radix_tree_preload: swap entry keeps a place for page in tree */ @@ -1713,7 +1713,7 @@ repeat: goto failed; } - error = mem_cgroup_try_charge(page, charge_mm, gfp, &memcg, + error = mem_cgroup_try_charge_delay(page, charge_mm, gfp, &memcg, false); if (!error) { error = shmem_add_to_page_cache(page, mapping, index, @@ -1819,7 +1819,7 @@ alloc_nohuge: page = shmem_alloc_and_acct_page(gfp, inode, if (sgp == SGP_WRITE) __SetPageReferenced(page); - error = mem_cgroup_try_charge(page, charge_mm, gfp, &memcg, + error = mem_cgroup_try_charge_delay(page, charge_mm, gfp, &memcg, PageTransHuge(page)); if (error) goto unacct; @@ -2292,7 +2292,7 @@ static int shmem_mfill_atomic_pte(struct mm_struct *dst_mm, __SetPageSwapBacked(page); __SetPageUptodate(page); - ret = mem_cgroup_try_charge(page, dst_mm, gfp, &memcg, false); + ret = mem_cgroup_try_charge_delay(page, dst_mm, gfp, &memcg, false); if (ret) goto out_release; diff --git a/mm/swapfile.c b/mm/swapfile.c index 18185ae4f223..8837b22c848d 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -3745,6 +3745,37 @@ static void free_swap_count_continuations(struct swap_info_struct *si) } } +#if defined(CONFIG_MEMCG) && defined(CONFIG_BLK_CGROUP) +void mem_cgroup_throttle_swaprate(struct mem_cgroup *memcg, int node, + gfp_t gfp_mask) +{ + struct swap_info_struct *si, *next; + if (!(gfp_mask & __GFP_IO) || !memcg) + return; + + if (!blk_cgroup_congested()) + return; + + /* + * We've already scheduled a throttle, avoid taking the global swap + * lock. + */ + if (current->throttle_queue) + return; + + spin_lock(&swap_avail_lock); + plist_for_each_entry_safe(si, next, &swap_avail_heads[node], + avail_lists[node]) { + if (si->bdev) { + blkcg_schedule_throttle(bdev_get_queue(si->bdev), + true); + break; + } + } + spin_unlock(&swap_avail_lock); +} +#endif + static int __init swapfile_init(void) { int nid; |