linux/block/blk-core.c

1259 lines
34 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0
/*
* Copyright (C) 1991, 1992 Linus Torvalds
* Copyright (C) 1994, Karl Keyte: Added support for disk statistics
* Elevator latency, (C) 2000 Andrea Arcangeli <andrea@suse.de> SuSE
* Queue request tables / lock, selectable elevator, Jens Axboe <axboe@suse.de>
* kernel-doc documentation started by NeilBrown <neilb@cse.unsw.edu.au>
* - July2000
* bio rewrite, highmem i/o, etc, Jens Axboe <axboe@suse.de> - may 2001
*/
/*
* This handles all read/write requests to block devices
*/
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/bio.h>
#include <linux/blkdev.h>
scsi: block: Do not accept any requests while suspended blk_queue_enter() accepts BLK_MQ_REQ_PM requests independent of the runtime power management state. Now that SCSI domain validation no longer depends on this behavior, modify the behavior of blk_queue_enter() as follows: - Do not accept any requests while suspended. - Only process power management requests while suspending or resuming. Submitting BLK_MQ_REQ_PM requests to a device that is runtime suspended causes runtime-suspended devices not to resume as they should. The request which should cause a runtime resume instead gets issued directly, without resuming the device first. Of course the device can't handle it properly, the I/O fails, and the device remains suspended. The problem is fixed by checking that the queue's runtime-PM status isn't RPM_SUSPENDED before allowing a request to be issued, and queuing a runtime-resume request if it is. In particular, the inline blk_pm_request_resume() routine is renamed blk_pm_resume_queue() and the code is unified by merging the surrounding checks into the routine. If the queue isn't set up for runtime PM, or there currently is no restriction on allowed requests, the request is allowed. Likewise if the BLK_MQ_REQ_PM flag is set and the status isn't RPM_SUSPENDED. Otherwise a runtime resume is queued and the request is blocked until conditions are more suitable. [ bvanassche: modified commit message and removed Cc: stable because without the previous patches from this series this patch would break parallel SCSI domain validation + introduced queue_rpm_status() ] Link: https://lore.kernel.org/r/20201209052951.16136-9-bvanassche@acm.org Cc: Jens Axboe <axboe@kernel.dk> Cc: Christoph Hellwig <hch@lst.de> Cc: Hannes Reinecke <hare@suse.de> Cc: Can Guo <cang@codeaurora.org> Cc: Stanley Chu <stanley.chu@mediatek.com> Cc: Ming Lei <ming.lei@redhat.com> Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Reported-and-tested-by: Martin Kepplinger <martin.kepplinger@puri.sm> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Can Guo <cang@codeaurora.org> Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-12-09 05:29:51 +00:00
#include <linux/blk-pm.h>
#include <linux/blk-integrity.h>
#include <linux/highmem.h>
#include <linux/mm.h>
mm: move readahead prototypes from mm.h Patch series "Change readahead API", v11. This series adds a readahead address_space operation to replace the readpages operation. The key difference is that pages are added to the page cache as they are allocated (and then looked up by the filesystem) instead of passing them on a list to the readpages operation and having the filesystem add them to the page cache. It's a net reduction in code for each implementation, more efficient than walking a list, and solves the direct-write vs buffered-read problem reported by yu kuai at http://lkml.kernel.org/r/20200116063601.39201-1-yukuai3@huawei.com The only unconverted filesystems are those which use fscache. Their conversion is pending Dave Howells' rewrite which will make the conversion substantially easier. This should be completed by the end of the year. I want to thank the reviewers/testers; Dave Chinner, John Hubbard, Eric Biggers, Johannes Thumshirn, Dave Sterba, Zi Yan, Christoph Hellwig and Miklos Szeredi have done a marvellous job of providing constructive criticism. These patches pass an xfstests run on ext4, xfs & btrfs with no regressions that I can tell (some of the tests seem a little flaky before and remain flaky afterwards). This patch (of 25): The readahead code is part of the page cache so should be found in the pagemap.h file. force_page_cache_readahead is only used within mm, so move it to mm/internal.h instead. Remove the parameter names where they add no value, and rename the ones which were actively misleading. Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Reviewed-by: John Hubbard <jhubbard@nvidia.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: William Kucharski <william.kucharski@oracle.com> Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Cc: Chao Yu <yuchao0@huawei.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Darrick J. Wong <darrick.wong@oracle.com> Cc: Dave Chinner <dchinner@redhat.com> Cc: Eric Biggers <ebiggers@google.com> Cc: Gao Xiang <gaoxiang25@huawei.com> Cc: Jaegeuk Kim <jaegeuk@kernel.org> Cc: Joseph Qi <joseph.qi@linux.alibaba.com> Cc: Junxiao Bi <junxiao.bi@oracle.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Zi Yan <ziy@nvidia.com> Cc: Miklos Szeredi <mszeredi@redhat.com> Link: http://lkml.kernel.org/r/20200414150233.24495-1-willy@infradead.org Link: http://lkml.kernel.org/r/20200414150233.24495-2-willy@infradead.org Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-06-02 04:46:07 +00:00
#include <linux/pagemap.h>
#include <linux/kernel_stat.h>
#include <linux/string.h>
#include <linux/init.h>
#include <linux/completion.h>
#include <linux/slab.h>
#include <linux/swap.h>
#include <linux/writeback.h>
#include <linux/task_io_accounting_ops.h>
#include <linux/fault-inject.h>
#include <linux/list_sort.h>
#include <linux/delay.h>
#include <linux/ratelimit.h>
#include <linux/pm_runtime.h>
#include <linux/t10-pi.h>
#include <linux/debugfs.h>
#include <linux/bpf.h>
#include <linux/part_stat.h>
#include <linux/sched/sysctl.h>
block: Inline encryption support for blk-mq We must have some way of letting a storage device driver know what encryption context it should use for en/decrypting a request. However, it's the upper layers (like the filesystem/fscrypt) that know about and manages encryption contexts. As such, when the upper layer submits a bio to the block layer, and this bio eventually reaches a device driver with support for inline encryption, the device driver will need to have been told the encryption context for that bio. We want to communicate the encryption context from the upper layer to the storage device along with the bio, when the bio is submitted to the block layer. To do this, we add a struct bio_crypt_ctx to struct bio, which can represent an encryption context (note that we can't use the bi_private field in struct bio to do this because that field does not function to pass information across layers in the storage stack). We also introduce various functions to manipulate the bio_crypt_ctx and make the bio/request merging logic aware of the bio_crypt_ctx. We also make changes to blk-mq to make it handle bios with encryption contexts. blk-mq can merge many bios into the same request. These bios need to have contiguous data unit numbers (the necessary changes to blk-merge are also made to ensure this) - as such, it suffices to keep the data unit number of just the first bio, since that's all a storage driver needs to infer the data unit number to use for each data block in each bio in a request. blk-mq keeps track of the encryption context to be used for all the bios in a request with the request's rq_crypt_ctx. When the first bio is added to an empty request, blk-mq will program the encryption context of that bio into the request_queue's keyslot manager, and store the returned keyslot in the request's rq_crypt_ctx. All the functions to operate on encryption contexts are in blk-crypto.c. Upper layers only need to call bio_crypt_set_ctx with the encryption key, algorithm and data_unit_num; they don't have to worry about getting a keyslot for each encryption context, as blk-mq/blk-crypto handles that. Blk-crypto also makes it possible for request-based layered devices like dm-rq to make use of inline encryption hardware by cloning the rq_crypt_ctx and programming a keyslot in the new request_queue when necessary. Note that any user of the block layer can submit bios with an encryption context, such as filesystems, device-mapper targets, etc. Signed-off-by: Satya Tangirala <satyat@google.com> Reviewed-by: Eric Biggers <ebiggers@google.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2020-05-14 00:37:18 +00:00
#include <linux/blk-crypto.h>
tracing/events: convert block trace points to TRACE_EVENT() TRACE_EVENT is a more generic way to define tracepoints. Doing so adds these new capabilities to this tracepoint: - zero-copy and per-cpu splice() tracing - binary tracing without printf overhead - structured logging records exposed under /debug/tracing/events - trace events embedded in function tracer output and other plugins - user-defined, per tracepoint filter expressions ... Cons: - no dev_t info for the output of plug, unplug_timer and unplug_io events. no dev_t info for getrq and sleeprq events if bio == NULL. no dev_t info for rq_abort,...,rq_requeue events if rq->rq_disk == NULL. This is mainly because we can't get the deivce from a request queue. But this may change in the future. - A packet command is converted to a string in TP_assign, not TP_print. While blktrace do the convertion just before output. Since pc requests should be rather rare, this is not a big issue. - In blktrace, an event can have 2 different print formats, but a TRACE_EVENT has a unique format, which means we have some unused data in a trace entry. The overhead is minimized by using __dynamic_array() instead of __array(). I've benchmarked the ioctl blktrace vs the splice based TRACE_EVENT tracing: dd dd + ioctl blktrace dd + TRACE_EVENT (splice) 1 7.36s, 42.7 MB/s 7.50s, 42.0 MB/s 7.41s, 42.5 MB/s 2 7.43s, 42.3 MB/s 7.48s, 42.1 MB/s 7.43s, 42.4 MB/s 3 7.38s, 42.6 MB/s 7.45s, 42.2 MB/s 7.41s, 42.5 MB/s So the overhead of tracing is very small, and no regression when using those trace events vs blktrace. And the binary output of TRACE_EVENT is much smaller than blktrace: # ls -l -h -rw-r--r-- 1 root root 8.8M 06-09 13:24 sda.blktrace.0 -rw-r--r-- 1 root root 195K 06-09 13:24 sda.blktrace.1 -rw-r--r-- 1 root root 2.7M 06-09 13:25 trace_splice.out Following are some comparisons between TRACE_EVENT and blktrace: plug: kjournald-480 [000] 303.084981: block_plug: [kjournald] kjournald-480 [000] 303.084981: 8,0 P N [kjournald] unplug_io: kblockd/0-118 [000] 300.052973: block_unplug_io: [kblockd/0] 1 kblockd/0-118 [000] 300.052974: 8,0 U N [kblockd/0] 1 remap: kjournald-480 [000] 303.085042: block_remap: 8,0 W 102736992 + 8 <- (8,8) 33384 kjournald-480 [000] 303.085043: 8,0 A W 102736992 + 8 <- (8,8) 33384 bio_backmerge: kjournald-480 [000] 303.085086: block_bio_backmerge: 8,0 W 102737032 + 8 [kjournald] kjournald-480 [000] 303.085086: 8,0 M W 102737032 + 8 [kjournald] getrq: kjournald-480 [000] 303.084974: block_getrq: 8,0 W 102736984 + 8 [kjournald] kjournald-480 [000] 303.084975: 8,0 G W 102736984 + 8 [kjournald] bash-2066 [001] 1072.953770: 8,0 G N [bash] bash-2066 [001] 1072.953773: block_getrq: 0,0 N 0 + 0 [bash] rq_complete: konsole-2065 [001] 300.053184: block_rq_complete: 8,0 W () 103669040 + 16 [0] konsole-2065 [001] 300.053191: 8,0 C W 103669040 + 16 [0] ksoftirqd/1-7 [001] 1072.953811: 8,0 C N (5a 00 08 00 00 00 00 00 24 00) [0] ksoftirqd/1-7 [001] 1072.953813: block_rq_complete: 0,0 N (5a 00 08 00 00 00 00 00 24 00) 0 + 0 [0] rq_insert: kjournald-480 [000] 303.084985: block_rq_insert: 8,0 W 0 () 102736984 + 8 [kjournald] kjournald-480 [000] 303.084986: 8,0 I W 102736984 + 8 [kjournald] Changelog from v2 -> v3: - use the newly introduced __dynamic_array(). Changelog from v1 -> v2: - use __string() instead of __array() to minimize the memory required to store hex dump of rq->cmd(). - support large pc requests. - add missing blk_fill_rwbs_rq() in block_rq_requeue TRACE_EVENT. - some cleanups. Signed-off-by: Li Zefan <lizf@cn.fujitsu.com> LKML-Reference: <4A2DF669.5070905@cn.fujitsu.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2009-06-09 05:43:05 +00:00
#define CREATE_TRACE_POINTS
#include <trace/events/block.h>
#include "blk.h"
#include "blk-mq-sched.h"
#include "blk-pm.h"
#include "blk-cgroup.h"
#include "blk-throttle.h"
#include "blk-ioprio.h"
struct dentry *blk_debugfs_root;
EXPORT_TRACEPOINT_SYMBOL_GPL(block_bio_remap);
EXPORT_TRACEPOINT_SYMBOL_GPL(block_rq_remap);
EXPORT_TRACEPOINT_SYMBOL_GPL(block_bio_complete);
EXPORT_TRACEPOINT_SYMBOL_GPL(block_split);
EXPORT_TRACEPOINT_SYMBOL_GPL(block_unplug);
EXPORT_TRACEPOINT_SYMBOL_GPL(block_rq_insert);
static DEFINE_IDA(blk_queue_ida);
/*
* For queue allocation
*/
static struct kmem_cache *blk_requestq_cachep;
/*
* Controlling structure to kblockd
*/
static struct workqueue_struct *kblockd_workqueue;
/**
* blk_queue_flag_set - atomically set a queue flag
* @flag: flag to be set
* @q: request queue
*/
void blk_queue_flag_set(unsigned int flag, struct request_queue *q)
{
set_bit(flag, &q->queue_flags);
}
EXPORT_SYMBOL(blk_queue_flag_set);
/**
* blk_queue_flag_clear - atomically clear a queue flag
* @flag: flag to be cleared
* @q: request queue
*/
void blk_queue_flag_clear(unsigned int flag, struct request_queue *q)
{
clear_bit(flag, &q->queue_flags);
}
EXPORT_SYMBOL(blk_queue_flag_clear);
/**
* blk_queue_flag_test_and_set - atomically test and set a queue flag
* @flag: flag to be set
* @q: request queue
*
* Returns the previous value of @flag - 0 if the flag was not set and 1 if
* the flag was already set.
*/
bool blk_queue_flag_test_and_set(unsigned int flag, struct request_queue *q)
{
return test_and_set_bit(flag, &q->queue_flags);
}
EXPORT_SYMBOL_GPL(blk_queue_flag_test_and_set);
#define REQ_OP_NAME(name) [REQ_OP_##name] = #name
static const char *const blk_op_name[] = {
REQ_OP_NAME(READ),
REQ_OP_NAME(WRITE),
REQ_OP_NAME(FLUSH),
REQ_OP_NAME(DISCARD),
REQ_OP_NAME(SECURE_ERASE),
REQ_OP_NAME(ZONE_RESET),
REQ_OP_NAME(ZONE_RESET_ALL),
2019-10-27 14:05:45 +00:00
REQ_OP_NAME(ZONE_OPEN),
REQ_OP_NAME(ZONE_CLOSE),
REQ_OP_NAME(ZONE_FINISH),
block: Introduce REQ_OP_ZONE_APPEND Define REQ_OP_ZONE_APPEND to append-write sectors to a zone of a zoned block device. This is a no-merge write operation. A zone append write BIO must: * Target a zoned block device * Have a sector position indicating the start sector of the target zone * The target zone must be a sequential write zone * The BIO must not cross a zone boundary * The BIO size must not be split to ensure that a single range of LBAs is written with a single command. Implement these checks in generic_make_request_checks() using the helper function blk_check_zone_append(). To avoid write append BIO splitting, introduce the new max_zone_append_sectors queue limit attribute and ensure that a BIO size is always lower than this limit. Export this new limit through sysfs and check these limits in bio_full(). Also when a LLDD can't dispatch a request to a specific zone, it will return BLK_STS_ZONE_RESOURCE indicating this request needs to be delayed, e.g. because the zone it will be dispatched to is still write-locked. If this happens set the request aside in a local list to continue trying dispatching requests such as READ requests or a WRITE/ZONE_APPEND requests targetting other zones. This way we can still keep a high queue depth without starving other requests even if one request can't be served due to zone write-locking. Finally, make sure that the bio sector position indicates the actual write position as indicated by the device on completion. Signed-off-by: Keith Busch <kbusch@kernel.org> [ jth: added zone-append specific add_page and merge_page helpers ] Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2020-05-12 08:55:47 +00:00
REQ_OP_NAME(ZONE_APPEND),
REQ_OP_NAME(WRITE_ZEROES),
REQ_OP_NAME(DRV_IN),
REQ_OP_NAME(DRV_OUT),
};
#undef REQ_OP_NAME
/**
* blk_op_str - Return string XXX in the REQ_OP_XXX.
* @op: REQ_OP_XXX.
*
* Description: Centralize block layer function to convert REQ_OP_XXX into
* string format. Useful in the debugging and tracing bio or request. For
* invalid REQ_OP_XXX it returns string "UNKNOWN".
*/
inline const char *blk_op_str(enum req_op op)
{
const char *op_str = "UNKNOWN";
if (op < ARRAY_SIZE(blk_op_name) && blk_op_name[op])
op_str = blk_op_name[op];
return op_str;
}
EXPORT_SYMBOL_GPL(blk_op_str);
static const struct {
int errno;
const char *name;
} blk_errors[] = {
[BLK_STS_OK] = { 0, "" },
[BLK_STS_NOTSUPP] = { -EOPNOTSUPP, "operation not supported" },
[BLK_STS_TIMEOUT] = { -ETIMEDOUT, "timeout" },
[BLK_STS_NOSPC] = { -ENOSPC, "critical space allocation" },
[BLK_STS_TRANSPORT] = { -ENOLINK, "recoverable transport" },
[BLK_STS_TARGET] = { -EREMOTEIO, "critical target" },
[BLK_STS_RESV_CONFLICT] = { -EBADE, "reservation conflict" },
[BLK_STS_MEDIUM] = { -ENODATA, "critical medium" },
[BLK_STS_PROTECTION] = { -EILSEQ, "protection" },
[BLK_STS_RESOURCE] = { -ENOMEM, "kernel resource" },
blk-mq: introduce BLK_STS_DEV_RESOURCE This status is returned from driver to block layer if device related resource is unavailable, but driver can guarantee that IO dispatch will be triggered in future when the resource is available. Convert some drivers to return BLK_STS_DEV_RESOURCE. Also, if driver returns BLK_STS_RESOURCE and SCHED_RESTART is set, rerun queue after a delay (BLK_MQ_DELAY_QUEUE) to avoid IO stalls. BLK_MQ_DELAY_QUEUE is 3 ms because both scsi-mq and nvmefc are using that magic value. If a driver can make sure there is in-flight IO, it is safe to return BLK_STS_DEV_RESOURCE because: 1) If all in-flight IOs complete before examining SCHED_RESTART in blk_mq_dispatch_rq_list(), SCHED_RESTART must be cleared, so queue is run immediately in this case by blk_mq_dispatch_rq_list(); 2) if there is any in-flight IO after/when examining SCHED_RESTART in blk_mq_dispatch_rq_list(): - if SCHED_RESTART isn't set, queue is run immediately as handled in 1) - otherwise, this request will be dispatched after any in-flight IO is completed via blk_mq_sched_restart() 3) if SCHED_RESTART is set concurently in context because of BLK_STS_RESOURCE, blk_mq_delay_run_hw_queue() will cover the above two cases and make sure IO hang can be avoided. One invariant is that queue will be rerun if SCHED_RESTART is set. Suggested-by: Jens Axboe <axboe@kernel.dk> Tested-by: Laurence Oberman <loberman@redhat.com> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Mike Snitzer <snitzer@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2018-01-31 03:04:57 +00:00
[BLK_STS_DEV_RESOURCE] = { -EBUSY, "device resource" },
[BLK_STS_AGAIN] = { -EAGAIN, "nonblocking retry" },
[BLK_STS_OFFLINE] = { -ENODEV, "device offline" },
/* device mapper special case, should not leak out: */
[BLK_STS_DM_REQUEUE] = { -EREMCHG, "dm internal retry" },
/* zone device specific errors */
[BLK_STS_ZONE_OPEN_RESOURCE] = { -ETOOMANYREFS, "open zones exceeded" },
[BLK_STS_ZONE_ACTIVE_RESOURCE] = { -EOVERFLOW, "active zones exceeded" },
/* Command duration limit device-side timeout */
[BLK_STS_DURATION_LIMIT] = { -ETIME, "duration limit exceeded" },
/* everything else not covered above: */
[BLK_STS_IOERR] = { -EIO, "I/O" },
};
blk_status_t errno_to_blk_status(int errno)
{
int i;
for (i = 0; i < ARRAY_SIZE(blk_errors); i++) {
if (blk_errors[i].errno == errno)
return (__force blk_status_t)i;
}
return BLK_STS_IOERR;
}
EXPORT_SYMBOL_GPL(errno_to_blk_status);
int blk_status_to_errno(blk_status_t status)
{
int idx = (__force int)status;
if (WARN_ON_ONCE(idx >= ARRAY_SIZE(blk_errors)))
return -EIO;
return blk_errors[idx].errno;
}
EXPORT_SYMBOL_GPL(blk_status_to_errno);
const char *blk_status_to_str(blk_status_t status)
{
int idx = (__force int)status;
if (WARN_ON_ONCE(idx >= ARRAY_SIZE(blk_errors)))
return "<null>";
return blk_errors[idx].name;
}
EXPORT_SYMBOL_GPL(blk_status_to_str);
/**
* blk_sync_queue - cancel any pending callbacks on a queue
* @q: the queue
*
* Description:
* The block layer may perform asynchronous callback activity
* on a queue, such as calling the unplug function after a timeout.
* A block device may call blk_sync_queue to ensure that any
* such activity is cancelled, thus allowing it to release resources
* that the callbacks might use. The caller must already have made sure
* that its ->submit_bio will not re-add plugging prior to calling
* this function.
*
block: Move blk_throtl_exit() call to blk_cleanup_queue() Move blk_throtl_exit() in blk_cleanup_queue() as blk_throtl_exit() is written in such a way that it needs queue lock. In blk_release_queue() there is no gurantee that ->queue_lock is still around. Initially blk_throtl_exit() was in blk_cleanup_queue() but Ingo reported one problem. https://lkml.org/lkml/2010/10/23/86 And a quick fix moved blk_throtl_exit() to blk_release_queue(). commit 7ad58c028652753814054f4e3ac58f925e7343f4 Author: Jens Axboe <jaxboe@fusionio.com> Date: Sat Oct 23 20:40:26 2010 +0200 block: fix use-after-free bug in blk throttle code This patch reverts above change and does not try to shutdown the throtl work in blk_sync_queue(). By avoiding call to throtl_shutdown_timer_wq() from blk_sync_queue(), we should also avoid the problem reported by Ingo. blk_sync_queue() seems to be used only by md driver and it seems to be using it to make sure q->unplug_fn is not called as md registers its own unplug functions and it is about to free up the data structures used by unplug_fn(). Block throttle does not call back into unplug_fn() or into md. So there is no need to cancel blk throttle work. In fact I think cancelling block throttle work is bad because it might happen that some bios are throttled and scheduled to be dispatched later with the help of pending work and if work is cancelled, these bios might never be dispatched. Block layer also uses blk_sync_queue() during blk_cleanup_queue() and blk_release_queue() time. That should be safe as we are also calling blk_throtl_exit() which should make sure all the throttling related data structures are cleaned up. Signed-off-by: Vivek Goyal <vgoyal@redhat.com> Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
2011-03-03 00:05:33 +00:00
* This function does not cancel any asynchronous activity arising
* out of elevator or throttling code. That would require elevator_exit()
* and blkcg_exit_queue() to be called with queue lock initialized.
block: Move blk_throtl_exit() call to blk_cleanup_queue() Move blk_throtl_exit() in blk_cleanup_queue() as blk_throtl_exit() is written in such a way that it needs queue lock. In blk_release_queue() there is no gurantee that ->queue_lock is still around. Initially blk_throtl_exit() was in blk_cleanup_queue() but Ingo reported one problem. https://lkml.org/lkml/2010/10/23/86 And a quick fix moved blk_throtl_exit() to blk_release_queue(). commit 7ad58c028652753814054f4e3ac58f925e7343f4 Author: Jens Axboe <jaxboe@fusionio.com> Date: Sat Oct 23 20:40:26 2010 +0200 block: fix use-after-free bug in blk throttle code This patch reverts above change and does not try to shutdown the throtl work in blk_sync_queue(). By avoiding call to throtl_shutdown_timer_wq() from blk_sync_queue(), we should also avoid the problem reported by Ingo. blk_sync_queue() seems to be used only by md driver and it seems to be using it to make sure q->unplug_fn is not called as md registers its own unplug functions and it is about to free up the data structures used by unplug_fn(). Block throttle does not call back into unplug_fn() or into md. So there is no need to cancel blk throttle work. In fact I think cancelling block throttle work is bad because it might happen that some bios are throttled and scheduled to be dispatched later with the help of pending work and if work is cancelled, these bios might never be dispatched. Block layer also uses blk_sync_queue() during blk_cleanup_queue() and blk_release_queue() time. That should be safe as we are also calling blk_throtl_exit() which should make sure all the throttling related data structures are cleaned up. Signed-off-by: Vivek Goyal <vgoyal@redhat.com> Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
2011-03-03 00:05:33 +00:00
*
*/
void blk_sync_queue(struct request_queue *q)
{
del_timer_sync(&q->timeout);
cancel_work_sync(&q->timeout_work);
}
EXPORT_SYMBOL(blk_sync_queue);
/**
* blk_set_pm_only - increment pm_only counter
* @q: request queue pointer
*/
void blk_set_pm_only(struct request_queue *q)
{
atomic_inc(&q->pm_only);
}
EXPORT_SYMBOL_GPL(blk_set_pm_only);
void blk_clear_pm_only(struct request_queue *q)
{
int pm_only;
pm_only = atomic_dec_return(&q->pm_only);
WARN_ON_ONCE(pm_only < 0);
if (pm_only == 0)
wake_up_all(&q->mq_freeze_wq);
}
EXPORT_SYMBOL_GPL(blk_clear_pm_only);
static void blk_free_queue_rcu(struct rcu_head *rcu_head)
{
struct request_queue *q = container_of(rcu_head,
struct request_queue, rcu_head);
percpu_ref_exit(&q->q_usage_counter);
kmem_cache_free(blk_requestq_cachep, q);
}
static void blk_free_queue(struct request_queue *q)
{
blk_free_queue_stats(q->stats);
if (queue_is_mq(q))
blk_mq_release(q);
ida_free(&blk_queue_ida, q->id);
call_rcu(&q->rcu_head, blk_free_queue_rcu);
}
/**
* blk_put_queue - decrement the request_queue refcount
* @q: the request_queue structure to decrement the refcount for
*
* Decrements the refcount of the request_queue and free it when the refcount
* reaches 0.
*/
void blk_put_queue(struct request_queue *q)
{
if (refcount_dec_and_test(&q->refs))
blk_free_queue(q);
}
EXPORT_SYMBOL(blk_put_queue);
void blk_queue_start_drain(struct request_queue *q)
{
/*
* When queue DYING flag is set, we need to block new req
* entering queue, so we call blk_freeze_queue_start() to
* prevent I/O from crossing blk_queue_enter().
*/
blk_freeze_queue_start(q);
if (queue_is_mq(q))
blk_mq_wake_waiters(q);
/* Make blk_queue_enter() reexamine the DYING flag. */
wake_up_all(&q->mq_freeze_wq);
}
block, scsi: Make SCSI quiesce and resume work reliably The contexts from which a SCSI device can be quiesced or resumed are: * Writing into /sys/class/scsi_device/*/device/state. * SCSI parallel (SPI) domain validation. * The SCSI device power management methods. See also scsi_bus_pm_ops. It is essential during suspend and resume that neither the filesystem state nor the filesystem metadata in RAM changes. This is why while the hibernation image is being written or restored that SCSI devices are quiesced. The SCSI core quiesces devices through scsi_device_quiesce() and scsi_device_resume(). In the SDEV_QUIESCE state execution of non-preempt requests is deferred. This is realized by returning BLKPREP_DEFER from inside scsi_prep_state_check() for quiesced SCSI devices. Avoid that a full queue prevents power management requests to be submitted by deferring allocation of non-preempt requests for devices in the quiesced state. This patch has been tested by running the following commands and by verifying that after each resume the fio job was still running: for ((i=0; i<10; i++)); do ( cd /sys/block/md0/md && while true; do [ "$(<sync_action)" = "idle" ] && echo check > sync_action sleep 1 done ) & pids=($!) for d in /sys/class/block/sd*[a-z]; do bdev=${d#/sys/class/block/} hcil=$(readlink "$d/device") hcil=${hcil#../../../} echo 4 > "$d/queue/nr_requests" echo 1 > "/sys/class/scsi_device/$hcil/device/queue_depth" fio --name="$bdev" --filename="/dev/$bdev" --buffered=0 --bs=512 \ --rw=randread --ioengine=libaio --numjobs=4 --iodepth=16 \ --iodepth_batch=1 --thread --loops=$((2**31)) & pids+=($!) done sleep 1 echo "$(date) Hibernating ..." >>hibernate-test-log.txt systemctl hibernate sleep 10 kill "${pids[@]}" echo idle > /sys/block/md0/md/sync_action wait echo "$(date) Done." >>hibernate-test-log.txt done Reported-by: Oleksandr Natalenko <oleksandr@natalenko.name> References: "I/O hangs after resuming from suspend-to-ram" (https://marc.info/?l=linux-block&m=150340235201348). Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com> Reviewed-by: Hannes Reinecke <hare@suse.com> Tested-by: Martin Steigerwald <martin@lichtvoll.de> Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name> Cc: Martin K. Petersen <martin.petersen@oracle.com> Cc: Ming Lei <ming.lei@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Cc: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-09 18:49:58 +00:00
/**
* blk_queue_enter() - try to increase q->q_usage_counter
* @q: request queue pointer
* @flags: BLK_MQ_REQ_NOWAIT and/or BLK_MQ_REQ_PM
block, scsi: Make SCSI quiesce and resume work reliably The contexts from which a SCSI device can be quiesced or resumed are: * Writing into /sys/class/scsi_device/*/device/state. * SCSI parallel (SPI) domain validation. * The SCSI device power management methods. See also scsi_bus_pm_ops. It is essential during suspend and resume that neither the filesystem state nor the filesystem metadata in RAM changes. This is why while the hibernation image is being written or restored that SCSI devices are quiesced. The SCSI core quiesces devices through scsi_device_quiesce() and scsi_device_resume(). In the SDEV_QUIESCE state execution of non-preempt requests is deferred. This is realized by returning BLKPREP_DEFER from inside scsi_prep_state_check() for quiesced SCSI devices. Avoid that a full queue prevents power management requests to be submitted by deferring allocation of non-preempt requests for devices in the quiesced state. This patch has been tested by running the following commands and by verifying that after each resume the fio job was still running: for ((i=0; i<10; i++)); do ( cd /sys/block/md0/md && while true; do [ "$(<sync_action)" = "idle" ] && echo check > sync_action sleep 1 done ) & pids=($!) for d in /sys/class/block/sd*[a-z]; do bdev=${d#/sys/class/block/} hcil=$(readlink "$d/device") hcil=${hcil#../../../} echo 4 > "$d/queue/nr_requests" echo 1 > "/sys/class/scsi_device/$hcil/device/queue_depth" fio --name="$bdev" --filename="/dev/$bdev" --buffered=0 --bs=512 \ --rw=randread --ioengine=libaio --numjobs=4 --iodepth=16 \ --iodepth_batch=1 --thread --loops=$((2**31)) & pids+=($!) done sleep 1 echo "$(date) Hibernating ..." >>hibernate-test-log.txt systemctl hibernate sleep 10 kill "${pids[@]}" echo idle > /sys/block/md0/md/sync_action wait echo "$(date) Done." >>hibernate-test-log.txt done Reported-by: Oleksandr Natalenko <oleksandr@natalenko.name> References: "I/O hangs after resuming from suspend-to-ram" (https://marc.info/?l=linux-block&m=150340235201348). Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com> Reviewed-by: Hannes Reinecke <hare@suse.com> Tested-by: Martin Steigerwald <martin@lichtvoll.de> Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name> Cc: Martin K. Petersen <martin.petersen@oracle.com> Cc: Ming Lei <ming.lei@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Cc: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-09 18:49:58 +00:00
*/
int blk_queue_enter(struct request_queue *q, blk_mq_req_flags_t flags)
block: generic request_queue reference counting Allow pmem, and other synchronous/bio-based block drivers, to fallback on a per-cpu reference count managed by the core for tracking queue live/dead state. The existing per-cpu reference count for the blk_mq case is promoted to be used in all block i/o scenarios. This involves initializing it by default, waiting for it to drop to zero at exit, and holding a live reference over the invocation of q->make_request_fn() in generic_make_request(). The blk_mq code continues to take its own reference per blk_mq request and retains the ability to freeze the queue, but the check that the queue is frozen is moved to generic_make_request(). This fixes crash signatures like the following: BUG: unable to handle kernel paging request at ffff880140000000 [..] Call Trace: [<ffffffff8145e8bf>] ? copy_user_handle_tail+0x5f/0x70 [<ffffffffa004e1e0>] pmem_do_bvec.isra.11+0x70/0xf0 [nd_pmem] [<ffffffffa004e331>] pmem_make_request+0xd1/0x200 [nd_pmem] [<ffffffff811c3162>] ? mempool_alloc+0x72/0x1a0 [<ffffffff8141f8b6>] generic_make_request+0xd6/0x110 [<ffffffff8141f966>] submit_bio+0x76/0x170 [<ffffffff81286dff>] submit_bh_wbc+0x12f/0x160 [<ffffffff81286e62>] submit_bh+0x12/0x20 [<ffffffff813395bd>] jbd2_write_superblock+0x8d/0x170 [<ffffffff8133974d>] jbd2_mark_journal_empty+0x5d/0x90 [<ffffffff813399cb>] jbd2_journal_destroy+0x24b/0x270 [<ffffffff810bc4ca>] ? put_pwq_unlocked+0x2a/0x30 [<ffffffff810bc6f5>] ? destroy_workqueue+0x225/0x250 [<ffffffff81303494>] ext4_put_super+0x64/0x360 [<ffffffff8124ab1a>] generic_shutdown_super+0x6a/0xf0 Cc: Jens Axboe <axboe@kernel.dk> Cc: Keith Busch <keith.busch@intel.com> Cc: Ross Zwisler <ross.zwisler@linux.intel.com> Suggested-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Christoph Hellwig <hch@lst.de> Tested-by: Ross Zwisler <ross.zwisler@linux.intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-10-21 17:20:12 +00:00
{
const bool pm = flags & BLK_MQ_REQ_PM;
block, scsi: Make SCSI quiesce and resume work reliably The contexts from which a SCSI device can be quiesced or resumed are: * Writing into /sys/class/scsi_device/*/device/state. * SCSI parallel (SPI) domain validation. * The SCSI device power management methods. See also scsi_bus_pm_ops. It is essential during suspend and resume that neither the filesystem state nor the filesystem metadata in RAM changes. This is why while the hibernation image is being written or restored that SCSI devices are quiesced. The SCSI core quiesces devices through scsi_device_quiesce() and scsi_device_resume(). In the SDEV_QUIESCE state execution of non-preempt requests is deferred. This is realized by returning BLKPREP_DEFER from inside scsi_prep_state_check() for quiesced SCSI devices. Avoid that a full queue prevents power management requests to be submitted by deferring allocation of non-preempt requests for devices in the quiesced state. This patch has been tested by running the following commands and by verifying that after each resume the fio job was still running: for ((i=0; i<10; i++)); do ( cd /sys/block/md0/md && while true; do [ "$(<sync_action)" = "idle" ] && echo check > sync_action sleep 1 done ) & pids=($!) for d in /sys/class/block/sd*[a-z]; do bdev=${d#/sys/class/block/} hcil=$(readlink "$d/device") hcil=${hcil#../../../} echo 4 > "$d/queue/nr_requests" echo 1 > "/sys/class/scsi_device/$hcil/device/queue_depth" fio --name="$bdev" --filename="/dev/$bdev" --buffered=0 --bs=512 \ --rw=randread --ioengine=libaio --numjobs=4 --iodepth=16 \ --iodepth_batch=1 --thread --loops=$((2**31)) & pids+=($!) done sleep 1 echo "$(date) Hibernating ..." >>hibernate-test-log.txt systemctl hibernate sleep 10 kill "${pids[@]}" echo idle > /sys/block/md0/md/sync_action wait echo "$(date) Done." >>hibernate-test-log.txt done Reported-by: Oleksandr Natalenko <oleksandr@natalenko.name> References: "I/O hangs after resuming from suspend-to-ram" (https://marc.info/?l=linux-block&m=150340235201348). Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com> Reviewed-by: Hannes Reinecke <hare@suse.com> Tested-by: Martin Steigerwald <martin@lichtvoll.de> Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name> Cc: Martin K. Petersen <martin.petersen@oracle.com> Cc: Ming Lei <ming.lei@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Cc: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-09 18:49:58 +00:00
while (!blk_try_enter_queue(q, pm)) {
block, scsi: Make SCSI quiesce and resume work reliably The contexts from which a SCSI device can be quiesced or resumed are: * Writing into /sys/class/scsi_device/*/device/state. * SCSI parallel (SPI) domain validation. * The SCSI device power management methods. See also scsi_bus_pm_ops. It is essential during suspend and resume that neither the filesystem state nor the filesystem metadata in RAM changes. This is why while the hibernation image is being written or restored that SCSI devices are quiesced. The SCSI core quiesces devices through scsi_device_quiesce() and scsi_device_resume(). In the SDEV_QUIESCE state execution of non-preempt requests is deferred. This is realized by returning BLKPREP_DEFER from inside scsi_prep_state_check() for quiesced SCSI devices. Avoid that a full queue prevents power management requests to be submitted by deferring allocation of non-preempt requests for devices in the quiesced state. This patch has been tested by running the following commands and by verifying that after each resume the fio job was still running: for ((i=0; i<10; i++)); do ( cd /sys/block/md0/md && while true; do [ "$(<sync_action)" = "idle" ] && echo check > sync_action sleep 1 done ) & pids=($!) for d in /sys/class/block/sd*[a-z]; do bdev=${d#/sys/class/block/} hcil=$(readlink "$d/device") hcil=${hcil#../../../} echo 4 > "$d/queue/nr_requests" echo 1 > "/sys/class/scsi_device/$hcil/device/queue_depth" fio --name="$bdev" --filename="/dev/$bdev" --buffered=0 --bs=512 \ --rw=randread --ioengine=libaio --numjobs=4 --iodepth=16 \ --iodepth_batch=1 --thread --loops=$((2**31)) & pids+=($!) done sleep 1 echo "$(date) Hibernating ..." >>hibernate-test-log.txt systemctl hibernate sleep 10 kill "${pids[@]}" echo idle > /sys/block/md0/md/sync_action wait echo "$(date) Done." >>hibernate-test-log.txt done Reported-by: Oleksandr Natalenko <oleksandr@natalenko.name> References: "I/O hangs after resuming from suspend-to-ram" (https://marc.info/?l=linux-block&m=150340235201348). Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com> Reviewed-by: Hannes Reinecke <hare@suse.com> Tested-by: Martin Steigerwald <martin@lichtvoll.de> Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name> Cc: Martin K. Petersen <martin.petersen@oracle.com> Cc: Ming Lei <ming.lei@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Cc: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-09 18:49:58 +00:00
if (flags & BLK_MQ_REQ_NOWAIT)
return -EAGAIN;
block: generic request_queue reference counting Allow pmem, and other synchronous/bio-based block drivers, to fallback on a per-cpu reference count managed by the core for tracking queue live/dead state. The existing per-cpu reference count for the blk_mq case is promoted to be used in all block i/o scenarios. This involves initializing it by default, waiting for it to drop to zero at exit, and holding a live reference over the invocation of q->make_request_fn() in generic_make_request(). The blk_mq code continues to take its own reference per blk_mq request and retains the ability to freeze the queue, but the check that the queue is frozen is moved to generic_make_request(). This fixes crash signatures like the following: BUG: unable to handle kernel paging request at ffff880140000000 [..] Call Trace: [<ffffffff8145e8bf>] ? copy_user_handle_tail+0x5f/0x70 [<ffffffffa004e1e0>] pmem_do_bvec.isra.11+0x70/0xf0 [nd_pmem] [<ffffffffa004e331>] pmem_make_request+0xd1/0x200 [nd_pmem] [<ffffffff811c3162>] ? mempool_alloc+0x72/0x1a0 [<ffffffff8141f8b6>] generic_make_request+0xd6/0x110 [<ffffffff8141f966>] submit_bio+0x76/0x170 [<ffffffff81286dff>] submit_bh_wbc+0x12f/0x160 [<ffffffff81286e62>] submit_bh+0x12/0x20 [<ffffffff813395bd>] jbd2_write_superblock+0x8d/0x170 [<ffffffff8133974d>] jbd2_mark_journal_empty+0x5d/0x90 [<ffffffff813399cb>] jbd2_journal_destroy+0x24b/0x270 [<ffffffff810bc4ca>] ? put_pwq_unlocked+0x2a/0x30 [<ffffffff810bc6f5>] ? destroy_workqueue+0x225/0x250 [<ffffffff81303494>] ext4_put_super+0x64/0x360 [<ffffffff8124ab1a>] generic_shutdown_super+0x6a/0xf0 Cc: Jens Axboe <axboe@kernel.dk> Cc: Keith Busch <keith.busch@intel.com> Cc: Ross Zwisler <ross.zwisler@linux.intel.com> Suggested-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Christoph Hellwig <hch@lst.de> Tested-by: Ross Zwisler <ross.zwisler@linux.intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-10-21 17:20:12 +00:00
/*
* read pair of barrier in blk_freeze_queue_start(), we need to
* order reading __PERCPU_REF_DEAD flag of .q_usage_counter and
* reading .mq_freeze_depth or queue dying flag, otherwise the
* following wait may never return if the two reads are
* reordered.
*/
smp_rmb();
wait_event(q->mq_freeze_wq,
blk-mq: fix hang caused by freeze/unfreeze sequence The following is a description of a hang in blk_mq_freeze_queue_wait(). The hang happens on attempt to freeze a queue while another task does queue unfreeze. The root cause is an incorrect sequence of percpu_ref_resurrect() and percpu_ref_kill() and as a result those two can be swapped: CPU#0 CPU#1 ---------------- ----------------- q1 = blk_mq_init_queue(shared_tags) q2 = blk_mq_init_queue(shared_tags): blk_mq_add_queue_tag_set(shared_tags): blk_mq_update_tag_set_depth(shared_tags): list_for_each_entry() blk_mq_freeze_queue(q1) > percpu_ref_kill() > blk_mq_freeze_queue_wait() blk_cleanup_queue(q1) blk_mq_freeze_queue(q1) > percpu_ref_kill() ^^^^^^ freeze_depth can't guarantee the order blk_mq_unfreeze_queue() > percpu_ref_resurrect() > blk_mq_freeze_queue_wait() ^^^^^^ Hang here!!!! This wrong sequence raises kernel warning: percpu_ref_kill_and_confirm called more than once on blk_queue_usage_counter_release! WARNING: CPU: 0 PID: 11854 at lib/percpu-refcount.c:336 percpu_ref_kill_and_confirm+0x99/0xb0 But the most unpleasant effect is a hang of a blk_mq_freeze_queue_wait(), which waits for a zero of a q_usage_counter, which never happens because percpu-ref was reinited (instead of being killed) and stays in PERCPU state forever. How to reproduce: - "insmod null_blk.ko shared_tags=1 nr_devices=0 queue_mode=2" - cpu0: python Script.py 0; taskset the corresponding process running on cpu0 - cpu1: python Script.py 1; taskset the corresponding process running on cpu1 Script.py: ------ #!/usr/bin/python3 import os import sys while True: on = "echo 1 > /sys/kernel/config/nullb/%s/power" % sys.argv[1] off = "echo 0 > /sys/kernel/config/nullb/%s/power" % sys.argv[1] os.system(on) os.system(off) ------ This bug was first reported and fixed by Roman, previous discussion: [1] Message id: 1443287365-4244-7-git-send-email-akinobu.mita@gmail.com [2] Message id: 1443563240-29306-6-git-send-email-tj@kernel.org [3] https://patchwork.kernel.org/patch/9268199/ Reviewed-by: Hannes Reinecke <hare@suse.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Roman Pen <roman.penyaev@profitbricks.com> Signed-off-by: Bob Liu <bob.liu@oracle.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2019-05-21 03:25:55 +00:00
(!q->mq_freeze_depth &&
scsi: block: Do not accept any requests while suspended blk_queue_enter() accepts BLK_MQ_REQ_PM requests independent of the runtime power management state. Now that SCSI domain validation no longer depends on this behavior, modify the behavior of blk_queue_enter() as follows: - Do not accept any requests while suspended. - Only process power management requests while suspending or resuming. Submitting BLK_MQ_REQ_PM requests to a device that is runtime suspended causes runtime-suspended devices not to resume as they should. The request which should cause a runtime resume instead gets issued directly, without resuming the device first. Of course the device can't handle it properly, the I/O fails, and the device remains suspended. The problem is fixed by checking that the queue's runtime-PM status isn't RPM_SUSPENDED before allowing a request to be issued, and queuing a runtime-resume request if it is. In particular, the inline blk_pm_request_resume() routine is renamed blk_pm_resume_queue() and the code is unified by merging the surrounding checks into the routine. If the queue isn't set up for runtime PM, or there currently is no restriction on allowed requests, the request is allowed. Likewise if the BLK_MQ_REQ_PM flag is set and the status isn't RPM_SUSPENDED. Otherwise a runtime resume is queued and the request is blocked until conditions are more suitable. [ bvanassche: modified commit message and removed Cc: stable because without the previous patches from this series this patch would break parallel SCSI domain validation + introduced queue_rpm_status() ] Link: https://lore.kernel.org/r/20201209052951.16136-9-bvanassche@acm.org Cc: Jens Axboe <axboe@kernel.dk> Cc: Christoph Hellwig <hch@lst.de> Cc: Hannes Reinecke <hare@suse.de> Cc: Can Guo <cang@codeaurora.org> Cc: Stanley Chu <stanley.chu@mediatek.com> Cc: Ming Lei <ming.lei@redhat.com> Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Reported-and-tested-by: Martin Kepplinger <martin.kepplinger@puri.sm> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Can Guo <cang@codeaurora.org> Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-12-09 05:29:51 +00:00
blk_pm_resume_queue(pm, q)) ||
blk_queue_dying(q));
block: generic request_queue reference counting Allow pmem, and other synchronous/bio-based block drivers, to fallback on a per-cpu reference count managed by the core for tracking queue live/dead state. The existing per-cpu reference count for the blk_mq case is promoted to be used in all block i/o scenarios. This involves initializing it by default, waiting for it to drop to zero at exit, and holding a live reference over the invocation of q->make_request_fn() in generic_make_request(). The blk_mq code continues to take its own reference per blk_mq request and retains the ability to freeze the queue, but the check that the queue is frozen is moved to generic_make_request(). This fixes crash signatures like the following: BUG: unable to handle kernel paging request at ffff880140000000 [..] Call Trace: [<ffffffff8145e8bf>] ? copy_user_handle_tail+0x5f/0x70 [<ffffffffa004e1e0>] pmem_do_bvec.isra.11+0x70/0xf0 [nd_pmem] [<ffffffffa004e331>] pmem_make_request+0xd1/0x200 [nd_pmem] [<ffffffff811c3162>] ? mempool_alloc+0x72/0x1a0 [<ffffffff8141f8b6>] generic_make_request+0xd6/0x110 [<ffffffff8141f966>] submit_bio+0x76/0x170 [<ffffffff81286dff>] submit_bh_wbc+0x12f/0x160 [<ffffffff81286e62>] submit_bh+0x12/0x20 [<ffffffff813395bd>] jbd2_write_superblock+0x8d/0x170 [<ffffffff8133974d>] jbd2_mark_journal_empty+0x5d/0x90 [<ffffffff813399cb>] jbd2_journal_destroy+0x24b/0x270 [<ffffffff810bc4ca>] ? put_pwq_unlocked+0x2a/0x30 [<ffffffff810bc6f5>] ? destroy_workqueue+0x225/0x250 [<ffffffff81303494>] ext4_put_super+0x64/0x360 [<ffffffff8124ab1a>] generic_shutdown_super+0x6a/0xf0 Cc: Jens Axboe <axboe@kernel.dk> Cc: Keith Busch <keith.busch@intel.com> Cc: Ross Zwisler <ross.zwisler@linux.intel.com> Suggested-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Christoph Hellwig <hch@lst.de> Tested-by: Ross Zwisler <ross.zwisler@linux.intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-10-21 17:20:12 +00:00
if (blk_queue_dying(q))
return -ENODEV;
}
return 0;
block: generic request_queue reference counting Allow pmem, and other synchronous/bio-based block drivers, to fallback on a per-cpu reference count managed by the core for tracking queue live/dead state. The existing per-cpu reference count for the blk_mq case is promoted to be used in all block i/o scenarios. This involves initializing it by default, waiting for it to drop to zero at exit, and holding a live reference over the invocation of q->make_request_fn() in generic_make_request(). The blk_mq code continues to take its own reference per blk_mq request and retains the ability to freeze the queue, but the check that the queue is frozen is moved to generic_make_request(). This fixes crash signatures like the following: BUG: unable to handle kernel paging request at ffff880140000000 [..] Call Trace: [<ffffffff8145e8bf>] ? copy_user_handle_tail+0x5f/0x70 [<ffffffffa004e1e0>] pmem_do_bvec.isra.11+0x70/0xf0 [nd_pmem] [<ffffffffa004e331>] pmem_make_request+0xd1/0x200 [nd_pmem] [<ffffffff811c3162>] ? mempool_alloc+0x72/0x1a0 [<ffffffff8141f8b6>] generic_make_request+0xd6/0x110 [<ffffffff8141f966>] submit_bio+0x76/0x170 [<ffffffff81286dff>] submit_bh_wbc+0x12f/0x160 [<ffffffff81286e62>] submit_bh+0x12/0x20 [<ffffffff813395bd>] jbd2_write_superblock+0x8d/0x170 [<ffffffff8133974d>] jbd2_mark_journal_empty+0x5d/0x90 [<ffffffff813399cb>] jbd2_journal_destroy+0x24b/0x270 [<ffffffff810bc4ca>] ? put_pwq_unlocked+0x2a/0x30 [<ffffffff810bc6f5>] ? destroy_workqueue+0x225/0x250 [<ffffffff81303494>] ext4_put_super+0x64/0x360 [<ffffffff8124ab1a>] generic_shutdown_super+0x6a/0xf0 Cc: Jens Axboe <axboe@kernel.dk> Cc: Keith Busch <keith.busch@intel.com> Cc: Ross Zwisler <ross.zwisler@linux.intel.com> Suggested-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Christoph Hellwig <hch@lst.de> Tested-by: Ross Zwisler <ross.zwisler@linux.intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-10-21 17:20:12 +00:00
}
int __bio_queue_enter(struct request_queue *q, struct bio *bio)
{
while (!blk_try_enter_queue(q, false)) {
struct gendisk *disk = bio->bi_bdev->bd_disk;
if (bio->bi_opf & REQ_NOWAIT) {
if (test_bit(GD_DEAD, &disk->state))
goto dead;
bio_wouldblock_error(bio);
return -EAGAIN;
}
/*
* read pair of barrier in blk_freeze_queue_start(), we need to
* order reading __PERCPU_REF_DEAD flag of .q_usage_counter and
* reading .mq_freeze_depth or queue dying flag, otherwise the
* following wait may never return if the two reads are
* reordered.
*/
smp_rmb();
wait_event(q->mq_freeze_wq,
(!q->mq_freeze_depth &&
blk_pm_resume_queue(false, q)) ||
test_bit(GD_DEAD, &disk->state));
if (test_bit(GD_DEAD, &disk->state))
goto dead;
}
return 0;
dead:
bio_io_error(bio);
return -ENODEV;
}
block: generic request_queue reference counting Allow pmem, and other synchronous/bio-based block drivers, to fallback on a per-cpu reference count managed by the core for tracking queue live/dead state. The existing per-cpu reference count for the blk_mq case is promoted to be used in all block i/o scenarios. This involves initializing it by default, waiting for it to drop to zero at exit, and holding a live reference over the invocation of q->make_request_fn() in generic_make_request(). The blk_mq code continues to take its own reference per blk_mq request and retains the ability to freeze the queue, but the check that the queue is frozen is moved to generic_make_request(). This fixes crash signatures like the following: BUG: unable to handle kernel paging request at ffff880140000000 [..] Call Trace: [<ffffffff8145e8bf>] ? copy_user_handle_tail+0x5f/0x70 [<ffffffffa004e1e0>] pmem_do_bvec.isra.11+0x70/0xf0 [nd_pmem] [<ffffffffa004e331>] pmem_make_request+0xd1/0x200 [nd_pmem] [<ffffffff811c3162>] ? mempool_alloc+0x72/0x1a0 [<ffffffff8141f8b6>] generic_make_request+0xd6/0x110 [<ffffffff8141f966>] submit_bio+0x76/0x170 [<ffffffff81286dff>] submit_bh_wbc+0x12f/0x160 [<ffffffff81286e62>] submit_bh+0x12/0x20 [<ffffffff813395bd>] jbd2_write_superblock+0x8d/0x170 [<ffffffff8133974d>] jbd2_mark_journal_empty+0x5d/0x90 [<ffffffff813399cb>] jbd2_journal_destroy+0x24b/0x270 [<ffffffff810bc4ca>] ? put_pwq_unlocked+0x2a/0x30 [<ffffffff810bc6f5>] ? destroy_workqueue+0x225/0x250 [<ffffffff81303494>] ext4_put_super+0x64/0x360 [<ffffffff8124ab1a>] generic_shutdown_super+0x6a/0xf0 Cc: Jens Axboe <axboe@kernel.dk> Cc: Keith Busch <keith.busch@intel.com> Cc: Ross Zwisler <ross.zwisler@linux.intel.com> Suggested-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Christoph Hellwig <hch@lst.de> Tested-by: Ross Zwisler <ross.zwisler@linux.intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-10-21 17:20:12 +00:00
void blk_queue_exit(struct request_queue *q)
{
percpu_ref_put(&q->q_usage_counter);
}
static void blk_queue_usage_counter_release(struct percpu_ref *ref)
{
struct request_queue *q =
container_of(ref, struct request_queue, q_usage_counter);
wake_up_all(&q->mq_freeze_wq);
}
static void blk_rq_timed_out_timer(struct timer_list *t)
{
struct request_queue *q = from_timer(q, t, timeout);
kblockd_schedule_work(&q->timeout_work);
}
static void blk_timeout_work(struct work_struct *work)
{
}
struct request_queue *blk_alloc_queue(struct queue_limits *lim, int node_id)
{
struct request_queue *q;
int error;
q = kmem_cache_alloc_node(blk_requestq_cachep, GFP_KERNEL | __GFP_ZERO,
node_id);
if (!q)
return ERR_PTR(-ENOMEM);
q->last_merge = NULL;
q->id = ida_alloc(&blk_queue_ida, GFP_KERNEL);
if (q->id < 0) {
error = q->id;
goto fail_q;
}
q->stats = blk_alloc_queue_stats();
if (!q->stats) {
error = -ENOMEM;
goto fail_id;
}
error = blk_set_default_limits(lim);
if (error)
goto fail_stats;
q->limits = *lim;
q->node = node_id;
atomic_set(&q->nr_active_requests_shared_tags, 0);
timer_setup(&q->timeout, blk_rq_timed_out_timer, 0);
INIT_WORK(&q->timeout_work, blk_timeout_work);
INIT_LIST_HEAD(&q->icq_list);
refcount_set(&q->refs, 1);
mutex_init(&q->debugfs_mutex);
mutex_init(&q->sysfs_lock);
block: split .sysfs_lock into two locks The kernfs built-in lock of 'kn->count' is held in sysfs .show/.store path. Meantime, inside block's .show/.store callback, q->sysfs_lock is required. However, when mq & iosched kobjects are removed via blk_mq_unregister_dev() & elv_unregister_queue(), q->sysfs_lock is held too. This way causes AB-BA lock because the kernfs built-in lock of 'kn-count' is required inside kobject_del() too, see the lockdep warning[1]. On the other hand, it isn't necessary to acquire q->sysfs_lock for both blk_mq_unregister_dev() & elv_unregister_queue() because clearing REGISTERED flag prevents storing to 'queue/scheduler' from being happened. Also sysfs write(store) is exclusive, so no necessary to hold the lock for elv_unregister_queue() when it is called in switching elevator path. So split .sysfs_lock into two: one is still named as .sysfs_lock for covering sync .store, the other one is named as .sysfs_dir_lock for covering kobjects and related status change. sysfs itself can handle the race between add/remove kobjects and showing/storing attributes under kobjects. For switching scheduler via storing to 'queue/scheduler', we use the queue flag of QUEUE_FLAG_REGISTERED with .sysfs_lock for avoiding the race, then we can avoid to hold .sysfs_lock during removing/adding kobjects. [1] lockdep warning ====================================================== WARNING: possible circular locking dependency detected 5.3.0-rc3-00044-g73277fc75ea0 #1380 Not tainted ------------------------------------------------------ rmmod/777 is trying to acquire lock: 00000000ac50e981 (kn->count#202){++++}, at: kernfs_remove_by_name_ns+0x59/0x72 but task is already holding lock: 00000000fb16ae21 (&q->sysfs_lock){+.+.}, at: blk_unregister_queue+0x78/0x10b which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (&q->sysfs_lock){+.+.}: __lock_acquire+0x95f/0xa2f lock_acquire+0x1b4/0x1e8 __mutex_lock+0x14a/0xa9b blk_mq_hw_sysfs_show+0x63/0xb6 sysfs_kf_seq_show+0x11f/0x196 seq_read+0x2cd/0x5f2 vfs_read+0xc7/0x18c ksys_read+0xc4/0x13e do_syscall_64+0xa7/0x295 entry_SYSCALL_64_after_hwframe+0x49/0xbe -> #0 (kn->count#202){++++}: check_prev_add+0x5d2/0xc45 validate_chain+0xed3/0xf94 __lock_acquire+0x95f/0xa2f lock_acquire+0x1b4/0x1e8 __kernfs_remove+0x237/0x40b kernfs_remove_by_name_ns+0x59/0x72 remove_files+0x61/0x96 sysfs_remove_group+0x81/0xa4 sysfs_remove_groups+0x3b/0x44 kobject_del+0x44/0x94 blk_mq_unregister_dev+0x83/0xdd blk_unregister_queue+0xa0/0x10b del_gendisk+0x259/0x3fa null_del_dev+0x8b/0x1c3 [null_blk] null_exit+0x5c/0x95 [null_blk] __se_sys_delete_module+0x204/0x337 do_syscall_64+0xa7/0x295 entry_SYSCALL_64_after_hwframe+0x49/0xbe other info that might help us debug this: Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(&q->sysfs_lock); lock(kn->count#202); lock(&q->sysfs_lock); lock(kn->count#202); *** DEADLOCK *** 2 locks held by rmmod/777: #0: 00000000e69bd9de (&lock){+.+.}, at: null_exit+0x2e/0x95 [null_blk] #1: 00000000fb16ae21 (&q->sysfs_lock){+.+.}, at: blk_unregister_queue+0x78/0x10b stack backtrace: CPU: 0 PID: 777 Comm: rmmod Not tainted 5.3.0-rc3-00044-g73277fc75ea0 #1380 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS ?-20180724_192412-buildhw-07.phx4 Call Trace: dump_stack+0x9a/0xe6 check_noncircular+0x207/0x251 ? print_circular_bug+0x32a/0x32a ? find_usage_backwards+0x84/0xb0 check_prev_add+0x5d2/0xc45 validate_chain+0xed3/0xf94 ? check_prev_add+0xc45/0xc45 ? mark_lock+0x11b/0x804 ? check_usage_forwards+0x1ca/0x1ca __lock_acquire+0x95f/0xa2f lock_acquire+0x1b4/0x1e8 ? kernfs_remove_by_name_ns+0x59/0x72 __kernfs_remove+0x237/0x40b ? kernfs_remove_by_name_ns+0x59/0x72 ? kernfs_next_descendant_post+0x7d/0x7d ? strlen+0x10/0x23 ? strcmp+0x22/0x44 kernfs_remove_by_name_ns+0x59/0x72 remove_files+0x61/0x96 sysfs_remove_group+0x81/0xa4 sysfs_remove_groups+0x3b/0x44 kobject_del+0x44/0x94 blk_mq_unregister_dev+0x83/0xdd blk_unregister_queue+0xa0/0x10b del_gendisk+0x259/0x3fa ? disk_events_poll_msecs_store+0x12b/0x12b ? check_flags+0x1ea/0x204 ? mark_held_locks+0x1f/0x7a null_del_dev+0x8b/0x1c3 [null_blk] null_exit+0x5c/0x95 [null_blk] __se_sys_delete_module+0x204/0x337 ? free_module+0x39f/0x39f ? blkcg_maybe_throttle_current+0x8a/0x718 ? rwlock_bug+0x62/0x62 ? __blkcg_punt_bio_submit+0xd0/0xd0 ? trace_hardirqs_on_thunk+0x1a/0x20 ? mark_held_locks+0x1f/0x7a ? do_syscall_64+0x4c/0x295 do_syscall_64+0xa7/0x295 entry_SYSCALL_64_after_hwframe+0x49/0xbe RIP: 0033:0x7fb696cdbe6b Code: 73 01 c3 48 8b 0d 1d 20 0c 00 f7 d8 64 89 01 48 83 c8 ff c3 66 2e 0f 1f 84 00 00 008 RSP: 002b:00007ffec9588788 EFLAGS: 00000206 ORIG_RAX: 00000000000000b0 RAX: ffffffffffffffda RBX: 0000559e589137c0 RCX: 00007fb696cdbe6b RDX: 000000000000000a RSI: 0000000000000800 RDI: 0000559e58913828 RBP: 0000000000000000 R08: 00007ffec9587701 R09: 0000000000000000 R10: 00007fb696d4eae0 R11: 0000000000000206 R12: 00007ffec95889b0 R13: 00007ffec95896b3 R14: 0000559e58913260 R15: 0000559e589137c0 Cc: Christoph Hellwig <hch@infradead.org> Cc: Hannes Reinecke <hare@suse.com> Cc: Greg KH <gregkh@linuxfoundation.org> Cc: Mike Snitzer <snitzer@redhat.com> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2019-08-27 11:01:48 +00:00
mutex_init(&q->sysfs_dir_lock);
mutex_init(&q->limits_lock);
block/rq_qos: protect rq_qos apis with a new lock commit 50e34d78815e ("block: disable the elevator int del_gendisk") move rq_qos_exit() from disk_release() to del_gendisk(), this will introduce some problems: 1) If rq_qos_add() is triggered by enabling iocost/iolatency through cgroupfs, then it can concurrent with del_gendisk(), it's not safe to write 'q->rq_qos' concurrently. 2) Activate cgroup policy that is relied on rq_qos will call rq_qos_add() and blkcg_activate_policy(), and if rq_qos_exit() is called in the middle, null-ptr-dereference will be triggered in blkcg_activate_policy(). 3) blkg_conf_open_bdev() can call blkdev_get_no_open() first to find the disk, then if rq_qos_exit() from del_gendisk() is done before rq_qos_add(), then memory will be leaked. This patch add a new disk level mutex 'rq_qos_mutex': 1) The lock will protect rq_qos_exit() directly. 2) For wbt that doesn't relied on blk-cgroup, rq_qos_add() can only be called from disk initialization for now because wbt can't be destructed until rq_qos_exit(), so it's safe not to protect wbt for now. Hoever, in case that rq_qos dynamically destruction is supported in the furture, this patch also protect rq_qos_add() from wbt_init() directly, this is enough because blk-sysfs already synchronize writers with disk removal. 3) For iocost and iolatency, in order to synchronize disk removal and cgroup configuration, the lock is held after blkdev_get_no_open() from blkg_conf_open_bdev(), and is released in blkg_conf_exit(). In order to fix the above memory leak, disk_live() is checked after holding the new lock. Fixes: 50e34d78815e ("block: disable the elevator int del_gendisk") Signed-off-by: Yu Kuai <yukuai3@huawei.com> Acked-by: Tejun Heo <tj@kernel.org> Link: https://lore.kernel.org/r/20230414084008.2085155-1-yukuai1@huaweicloud.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2023-04-14 08:40:08 +00:00
mutex_init(&q->rq_qos_mutex);
spin_lock_init(&q->queue_lock);
block: Initialize ->queue_lock to internal lock at queue allocation time There does not seem to be a clear convention whether q->queue_lock is initialized or not when blk_cleanup_queue() is called. In the past it was not necessary but now blk_throtl_exit() takes up queue lock by default and needs queue lock to be available. In fact elevator_exit() code also has similar requirement just that it is less stringent in the sense that elevator_exit() is called only if elevator is initialized. Two problems have been noticed because of ambiguity about spin lock status. - If a driver calls blk_alloc_queue() and then soon calls blk_cleanup_queue() almost immediately, (because some other driver structure allocation failed or some other error happened) then blk_throtl_exit() will run into issues as queue lock is not initialized. Loop driver ran into this issue recently and I noticed error paths in md driver too. Similar error paths should exist in other drivers too. - If some driver provided external spin lock and zapped the lock before blk_cleanup_queue(), then it can lead to issues. So this patch initializes the default queue lock at queue allocation time. block throttling code is one of the users of queue lock and it is initialized at the queue allocation time, so it makes sense to initialize ->queue_lock also to internal lock. A driver can overide that lock later. This will take care of the issue where a driver does not have to worry about initializing the queue lock to default before calling blk_cleanup_queue() Signed-off-by: Vivek Goyal <vgoyal@redhat.com> Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
2011-03-03 00:04:42 +00:00
blk-mq: new multi-queue block IO queueing mechanism Linux currently has two models for block devices: - The classic request_fn based approach, where drivers use struct request units for IO. The block layer provides various helper functionalities to let drivers share code, things like tag management, timeout handling, queueing, etc. - The "stacked" approach, where a driver squeezes in between the block layer and IO submitter. Since this bypasses the IO stack, driver generally have to manage everything themselves. With drivers being written for new high IOPS devices, the classic request_fn based driver doesn't work well enough. The design dates back to when both SMP and high IOPS was rare. It has problems with scaling to bigger machines, and runs into scaling issues even on smaller machines when you have IOPS in the hundreds of thousands per device. The stacked approach is then most often selected as the model for the driver. But this means that everybody has to re-invent everything, and along with that we get all the problems again that the shared approach solved. This commit introduces blk-mq, block multi queue support. The design is centered around per-cpu queues for queueing IO, which then funnel down into x number of hardware submission queues. We might have a 1:1 mapping between the two, or it might be an N:M mapping. That all depends on what the hardware supports. blk-mq provides various helper functions, which include: - Scalable support for request tagging. Most devices need to be able to uniquely identify a request both in the driver and to the hardware. The tagging uses per-cpu caches for freed tags, to enable cache hot reuse. - Timeout handling without tracking request on a per-device basis. Basically the driver should be able to get a notification, if a request happens to fail. - Optional support for non 1:1 mappings between issue and submission queues. blk-mq can redirect IO completions to the desired location. - Support for per-request payloads. Drivers almost always need to associate a request structure with some driver private command structure. Drivers can tell blk-mq this at init time, and then any request handed to the driver will have the required size of memory associated with it. - Support for merging of IO, and plugging. The stacked model gets neither of these. Even for high IOPS devices, merging sequential IO reduces per-command overhead and thus increases bandwidth. For now, this is provided as a potential 3rd queueing model, with the hope being that, as it matures, it can replace both the classic and stacked model. That would get us back to having just 1 real model for block devices, leaving the stacked approach to dm/md devices (as it was originally intended). Contributions in this patch from the following people: Shaohua Li <shli@fusionio.com> Alexander Gordeev <agordeev@redhat.com> Christoph Hellwig <hch@infradead.org> Mike Christie <michaelc@cs.wisc.edu> Matias Bjorling <m@bjorling.me> Jeff Moyer <jmoyer@redhat.com> Acked-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2013-10-24 08:20:05 +00:00
init_waitqueue_head(&q->mq_freeze_wq);
blk-mq: fix hang caused by freeze/unfreeze sequence The following is a description of a hang in blk_mq_freeze_queue_wait(). The hang happens on attempt to freeze a queue while another task does queue unfreeze. The root cause is an incorrect sequence of percpu_ref_resurrect() and percpu_ref_kill() and as a result those two can be swapped: CPU#0 CPU#1 ---------------- ----------------- q1 = blk_mq_init_queue(shared_tags) q2 = blk_mq_init_queue(shared_tags): blk_mq_add_queue_tag_set(shared_tags): blk_mq_update_tag_set_depth(shared_tags): list_for_each_entry() blk_mq_freeze_queue(q1) > percpu_ref_kill() > blk_mq_freeze_queue_wait() blk_cleanup_queue(q1) blk_mq_freeze_queue(q1) > percpu_ref_kill() ^^^^^^ freeze_depth can't guarantee the order blk_mq_unfreeze_queue() > percpu_ref_resurrect() > blk_mq_freeze_queue_wait() ^^^^^^ Hang here!!!! This wrong sequence raises kernel warning: percpu_ref_kill_and_confirm called more than once on blk_queue_usage_counter_release! WARNING: CPU: 0 PID: 11854 at lib/percpu-refcount.c:336 percpu_ref_kill_and_confirm+0x99/0xb0 But the most unpleasant effect is a hang of a blk_mq_freeze_queue_wait(), which waits for a zero of a q_usage_counter, which never happens because percpu-ref was reinited (instead of being killed) and stays in PERCPU state forever. How to reproduce: - "insmod null_blk.ko shared_tags=1 nr_devices=0 queue_mode=2" - cpu0: python Script.py 0; taskset the corresponding process running on cpu0 - cpu1: python Script.py 1; taskset the corresponding process running on cpu1 Script.py: ------ #!/usr/bin/python3 import os import sys while True: on = "echo 1 > /sys/kernel/config/nullb/%s/power" % sys.argv[1] off = "echo 0 > /sys/kernel/config/nullb/%s/power" % sys.argv[1] os.system(on) os.system(off) ------ This bug was first reported and fixed by Roman, previous discussion: [1] Message id: 1443287365-4244-7-git-send-email-akinobu.mita@gmail.com [2] Message id: 1443563240-29306-6-git-send-email-tj@kernel.org [3] https://patchwork.kernel.org/patch/9268199/ Reviewed-by: Hannes Reinecke <hare@suse.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Roman Pen <roman.penyaev@profitbricks.com> Signed-off-by: Bob Liu <bob.liu@oracle.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2019-05-21 03:25:55 +00:00
mutex_init(&q->mq_freeze_lock);
blk-mq: new multi-queue block IO queueing mechanism Linux currently has two models for block devices: - The classic request_fn based approach, where drivers use struct request units for IO. The block layer provides various helper functionalities to let drivers share code, things like tag management, timeout handling, queueing, etc. - The "stacked" approach, where a driver squeezes in between the block layer and IO submitter. Since this bypasses the IO stack, driver generally have to manage everything themselves. With drivers being written for new high IOPS devices, the classic request_fn based driver doesn't work well enough. The design dates back to when both SMP and high IOPS was rare. It has problems with scaling to bigger machines, and runs into scaling issues even on smaller machines when you have IOPS in the hundreds of thousands per device. The stacked approach is then most often selected as the model for the driver. But this means that everybody has to re-invent everything, and along with that we get all the problems again that the shared approach solved. This commit introduces blk-mq, block multi queue support. The design is centered around per-cpu queues for queueing IO, which then funnel down into x number of hardware submission queues. We might have a 1:1 mapping between the two, or it might be an N:M mapping. That all depends on what the hardware supports. blk-mq provides various helper functions, which include: - Scalable support for request tagging. Most devices need to be able to uniquely identify a request both in the driver and to the hardware. The tagging uses per-cpu caches for freed tags, to enable cache hot reuse. - Timeout handling without tracking request on a per-device basis. Basically the driver should be able to get a notification, if a request happens to fail. - Optional support for non 1:1 mappings between issue and submission queues. blk-mq can redirect IO completions to the desired location. - Support for per-request payloads. Drivers almost always need to associate a request structure with some driver private command structure. Drivers can tell blk-mq this at init time, and then any request handed to the driver will have the required size of memory associated with it. - Support for merging of IO, and plugging. The stacked model gets neither of these. Even for high IOPS devices, merging sequential IO reduces per-command overhead and thus increases bandwidth. For now, this is provided as a potential 3rd queueing model, with the hope being that, as it matures, it can replace both the classic and stacked model. That would get us back to having just 1 real model for block devices, leaving the stacked approach to dm/md devices (as it was originally intended). Contributions in this patch from the following people: Shaohua Li <shli@fusionio.com> Alexander Gordeev <agordeev@redhat.com> Christoph Hellwig <hch@infradead.org> Mike Christie <michaelc@cs.wisc.edu> Matias Bjorling <m@bjorling.me> Jeff Moyer <jmoyer@redhat.com> Acked-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2013-10-24 08:20:05 +00:00
blkg_init_queue(q);
block: generic request_queue reference counting Allow pmem, and other synchronous/bio-based block drivers, to fallback on a per-cpu reference count managed by the core for tracking queue live/dead state. The existing per-cpu reference count for the blk_mq case is promoted to be used in all block i/o scenarios. This involves initializing it by default, waiting for it to drop to zero at exit, and holding a live reference over the invocation of q->make_request_fn() in generic_make_request(). The blk_mq code continues to take its own reference per blk_mq request and retains the ability to freeze the queue, but the check that the queue is frozen is moved to generic_make_request(). This fixes crash signatures like the following: BUG: unable to handle kernel paging request at ffff880140000000 [..] Call Trace: [<ffffffff8145e8bf>] ? copy_user_handle_tail+0x5f/0x70 [<ffffffffa004e1e0>] pmem_do_bvec.isra.11+0x70/0xf0 [nd_pmem] [<ffffffffa004e331>] pmem_make_request+0xd1/0x200 [nd_pmem] [<ffffffff811c3162>] ? mempool_alloc+0x72/0x1a0 [<ffffffff8141f8b6>] generic_make_request+0xd6/0x110 [<ffffffff8141f966>] submit_bio+0x76/0x170 [<ffffffff81286dff>] submit_bh_wbc+0x12f/0x160 [<ffffffff81286e62>] submit_bh+0x12/0x20 [<ffffffff813395bd>] jbd2_write_superblock+0x8d/0x170 [<ffffffff8133974d>] jbd2_mark_journal_empty+0x5d/0x90 [<ffffffff813399cb>] jbd2_journal_destroy+0x24b/0x270 [<ffffffff810bc4ca>] ? put_pwq_unlocked+0x2a/0x30 [<ffffffff810bc6f5>] ? destroy_workqueue+0x225/0x250 [<ffffffff81303494>] ext4_put_super+0x64/0x360 [<ffffffff8124ab1a>] generic_shutdown_super+0x6a/0xf0 Cc: Jens Axboe <axboe@kernel.dk> Cc: Keith Busch <keith.busch@intel.com> Cc: Ross Zwisler <ross.zwisler@linux.intel.com> Suggested-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Christoph Hellwig <hch@lst.de> Tested-by: Ross Zwisler <ross.zwisler@linux.intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-10-21 17:20:12 +00:00
/*
* Init percpu_ref in atomic mode so that it's faster to shutdown.
* See blk_register_queue() for details.
*/
error = percpu_ref_init(&q->q_usage_counter,
block: generic request_queue reference counting Allow pmem, and other synchronous/bio-based block drivers, to fallback on a per-cpu reference count managed by the core for tracking queue live/dead state. The existing per-cpu reference count for the blk_mq case is promoted to be used in all block i/o scenarios. This involves initializing it by default, waiting for it to drop to zero at exit, and holding a live reference over the invocation of q->make_request_fn() in generic_make_request(). The blk_mq code continues to take its own reference per blk_mq request and retains the ability to freeze the queue, but the check that the queue is frozen is moved to generic_make_request(). This fixes crash signatures like the following: BUG: unable to handle kernel paging request at ffff880140000000 [..] Call Trace: [<ffffffff8145e8bf>] ? copy_user_handle_tail+0x5f/0x70 [<ffffffffa004e1e0>] pmem_do_bvec.isra.11+0x70/0xf0 [nd_pmem] [<ffffffffa004e331>] pmem_make_request+0xd1/0x200 [nd_pmem] [<ffffffff811c3162>] ? mempool_alloc+0x72/0x1a0 [<ffffffff8141f8b6>] generic_make_request+0xd6/0x110 [<ffffffff8141f966>] submit_bio+0x76/0x170 [<ffffffff81286dff>] submit_bh_wbc+0x12f/0x160 [<ffffffff81286e62>] submit_bh+0x12/0x20 [<ffffffff813395bd>] jbd2_write_superblock+0x8d/0x170 [<ffffffff8133974d>] jbd2_mark_journal_empty+0x5d/0x90 [<ffffffff813399cb>] jbd2_journal_destroy+0x24b/0x270 [<ffffffff810bc4ca>] ? put_pwq_unlocked+0x2a/0x30 [<ffffffff810bc6f5>] ? destroy_workqueue+0x225/0x250 [<ffffffff81303494>] ext4_put_super+0x64/0x360 [<ffffffff8124ab1a>] generic_shutdown_super+0x6a/0xf0 Cc: Jens Axboe <axboe@kernel.dk> Cc: Keith Busch <keith.busch@intel.com> Cc: Ross Zwisler <ross.zwisler@linux.intel.com> Suggested-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Christoph Hellwig <hch@lst.de> Tested-by: Ross Zwisler <ross.zwisler@linux.intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-10-21 17:20:12 +00:00
blk_queue_usage_counter_release,
PERCPU_REF_INIT_ATOMIC, GFP_KERNEL);
if (error)
goto fail_stats;
q->nr_requests = BLKDEV_DEFAULT_RQ;
return q;
fail_stats:
blk_free_queue_stats(q->stats);
fail_id:
ida_free(&blk_queue_ida, q->id);
fail_q:
kmem_cache_free(blk_requestq_cachep, q);
return ERR_PTR(error);
}
/**
* blk_get_queue - increment the request_queue refcount
* @q: the request_queue structure to increment the refcount for
*
* Increment the refcount of the request_queue kobject.
*
* Context: Any context.
*/
bool blk_get_queue(struct request_queue *q)
{
if (unlikely(blk_queue_dying(q)))
return false;
refcount_inc(&q->refs);
return true;
}
EXPORT_SYMBOL(blk_get_queue);
#ifdef CONFIG_FAIL_MAKE_REQUEST
static DECLARE_FAULT_ATTR(fail_make_request);
static int __init setup_fail_make_request(char *str)
{
return setup_fault_attr(&fail_make_request, str);
}
__setup("fail_make_request=", setup_fail_make_request);
bool should_fail_request(struct block_device *part, unsigned int bytes)
{
return bdev_test_flag(part, BD_MAKE_IT_FAIL) &&
should_fail(&fail_make_request, bytes);
}
static int __init fail_make_request_debugfs(void)
{
struct dentry *dir = fault_create_debugfs_attr("fail_make_request",
NULL, &fail_make_request);
return PTR_ERR_OR_ZERO(dir);
}
late_initcall(fail_make_request_debugfs);
#endif /* CONFIG_FAIL_MAKE_REQUEST */
static inline void bio_check_ro(struct bio *bio)
{
if (op_is_write(bio_op(bio)) && bdev_read_only(bio->bi_bdev)) {
if (op_is_flush(bio->bi_opf) && !bio_sectors(bio))
return;
if (bdev_test_flag(bio->bi_bdev, BD_RO_WARNED))
return;
bdev_set_flag(bio->bi_bdev, BD_RO_WARNED);
/*
* Use ioctl to set underlying disk of raid/dm to read-only
* will trigger this.
*/
pr_warn("Trying to write to read-only block-device %pg\n",
bio->bi_bdev);
}
}
static noinline int should_fail_bio(struct bio *bio)
{
if (should_fail_request(bdev_whole(bio->bi_bdev), bio->bi_iter.bi_size))
return -EIO;
return 0;
}
ALLOW_ERROR_INJECTION(should_fail_bio, ERRNO);
/*
* Check whether this bio extends beyond the end of the device or partition.
* This may well happen - the kernel calls bread() without checking the size of
* the device, e.g., when mounting a file system.
*/
static inline int bio_check_eod(struct bio *bio)
{
sector_t maxsector = bdev_nr_sectors(bio->bi_bdev);
unsigned int nr_sectors = bio_sectors(bio);
if (nr_sectors &&
(nr_sectors > maxsector ||
bio->bi_iter.bi_sector > maxsector - nr_sectors)) {
pr_info_ratelimited("%s: attempt to access beyond end of device\n"
"%pg: rw=%d, sector=%llu, nr_sectors = %u limit=%llu\n",
current->comm, bio->bi_bdev, bio->bi_opf,
bio->bi_iter.bi_sector, nr_sectors, maxsector);
return -EIO;
}
return 0;
}
/*
* Remap block n of partition p to block n+start(p) of the disk.
*/
static int blk_partition_remap(struct bio *bio)
{
struct block_device *p = bio->bi_bdev;
if (unlikely(should_fail_request(p, bio->bi_iter.bi_size)))
return -EIO;
if (bio_sectors(bio)) {
bio->bi_iter.bi_sector += p->bd_start_sect;
trace_block_bio_remap(bio, p->bd_dev,
bio->bi_iter.bi_sector -
p->bd_start_sect);
}
bio_set_flag(bio, BIO_REMAPPED);
return 0;
}
block: Introduce REQ_OP_ZONE_APPEND Define REQ_OP_ZONE_APPEND to append-write sectors to a zone of a zoned block device. This is a no-merge write operation. A zone append write BIO must: * Target a zoned block device * Have a sector position indicating the start sector of the target zone * The target zone must be a sequential write zone * The BIO must not cross a zone boundary * The BIO size must not be split to ensure that a single range of LBAs is written with a single command. Implement these checks in generic_make_request_checks() using the helper function blk_check_zone_append(). To avoid write append BIO splitting, introduce the new max_zone_append_sectors queue limit attribute and ensure that a BIO size is always lower than this limit. Export this new limit through sysfs and check these limits in bio_full(). Also when a LLDD can't dispatch a request to a specific zone, it will return BLK_STS_ZONE_RESOURCE indicating this request needs to be delayed, e.g. because the zone it will be dispatched to is still write-locked. If this happens set the request aside in a local list to continue trying dispatching requests such as READ requests or a WRITE/ZONE_APPEND requests targetting other zones. This way we can still keep a high queue depth without starving other requests even if one request can't be served due to zone write-locking. Finally, make sure that the bio sector position indicates the actual write position as indicated by the device on completion. Signed-off-by: Keith Busch <kbusch@kernel.org> [ jth: added zone-append specific add_page and merge_page helpers ] Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2020-05-12 08:55:47 +00:00
/*
* Check write append to a zoned block device.
*/
static inline blk_status_t blk_check_zone_append(struct request_queue *q,
struct bio *bio)
{
int nr_sectors = bio_sectors(bio);
/* Only applicable to zoned block devices */
if (!bdev_is_zoned(bio->bi_bdev))
block: Introduce REQ_OP_ZONE_APPEND Define REQ_OP_ZONE_APPEND to append-write sectors to a zone of a zoned block device. This is a no-merge write operation. A zone append write BIO must: * Target a zoned block device * Have a sector position indicating the start sector of the target zone * The target zone must be a sequential write zone * The BIO must not cross a zone boundary * The BIO size must not be split to ensure that a single range of LBAs is written with a single command. Implement these checks in generic_make_request_checks() using the helper function blk_check_zone_append(). To avoid write append BIO splitting, introduce the new max_zone_append_sectors queue limit attribute and ensure that a BIO size is always lower than this limit. Export this new limit through sysfs and check these limits in bio_full(). Also when a LLDD can't dispatch a request to a specific zone, it will return BLK_STS_ZONE_RESOURCE indicating this request needs to be delayed, e.g. because the zone it will be dispatched to is still write-locked. If this happens set the request aside in a local list to continue trying dispatching requests such as READ requests or a WRITE/ZONE_APPEND requests targetting other zones. This way we can still keep a high queue depth without starving other requests even if one request can't be served due to zone write-locking. Finally, make sure that the bio sector position indicates the actual write position as indicated by the device on completion. Signed-off-by: Keith Busch <kbusch@kernel.org> [ jth: added zone-append specific add_page and merge_page helpers ] Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2020-05-12 08:55:47 +00:00
return BLK_STS_NOTSUPP;
/* The bio sector must point to the start of a sequential zone */
if (!bdev_is_zone_start(bio->bi_bdev, bio->bi_iter.bi_sector))
block: Introduce REQ_OP_ZONE_APPEND Define REQ_OP_ZONE_APPEND to append-write sectors to a zone of a zoned block device. This is a no-merge write operation. A zone append write BIO must: * Target a zoned block device * Have a sector position indicating the start sector of the target zone * The target zone must be a sequential write zone * The BIO must not cross a zone boundary * The BIO size must not be split to ensure that a single range of LBAs is written with a single command. Implement these checks in generic_make_request_checks() using the helper function blk_check_zone_append(). To avoid write append BIO splitting, introduce the new max_zone_append_sectors queue limit attribute and ensure that a BIO size is always lower than this limit. Export this new limit through sysfs and check these limits in bio_full(). Also when a LLDD can't dispatch a request to a specific zone, it will return BLK_STS_ZONE_RESOURCE indicating this request needs to be delayed, e.g. because the zone it will be dispatched to is still write-locked. If this happens set the request aside in a local list to continue trying dispatching requests such as READ requests or a WRITE/ZONE_APPEND requests targetting other zones. This way we can still keep a high queue depth without starving other requests even if one request can't be served due to zone write-locking. Finally, make sure that the bio sector position indicates the actual write position as indicated by the device on completion. Signed-off-by: Keith Busch <kbusch@kernel.org> [ jth: added zone-append specific add_page and merge_page helpers ] Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2020-05-12 08:55:47 +00:00
return BLK_STS_IOERR;
/*
* Not allowed to cross zone boundaries. Otherwise, the BIO will be
* split and could result in non-contiguous sectors being written in
* different zones.
*/
if (nr_sectors > q->limits.chunk_sectors)
return BLK_STS_IOERR;
/* Make sure the BIO is small enough and will not get split */
block: Allow zero value of max_zone_append_sectors queue limit In preparation for adding a generic zone append emulation using zone write plugging, allow device drivers supporting zoned block device to set a the max_zone_append_sectors queue limit of a device to 0 to indicate the lack of native support for zone append operations and that the block layer should emulate these operations using regular write operations. blk_queue_max_zone_append_sectors() is modified to allow passing 0 as the max_zone_append_sectors argument. The function queue_max_zone_append_sectors() is also modified to ensure that the minimum of the max_hw_sectors and chunk_sectors limit is used whenever the max_zone_append_sectors limit is 0. This minimum is consistent with the value set for the max_zone_append_sectors limit by the function blk_validate_zoned_limits() when limits for a queue are validated. The helper functions queue_emulates_zone_append() and bdev_emulates_zone_append() are added to test if a queue (or block device) emulates zone append operations. In order for blk_revalidate_disk_zones() to accept zoned block devices relying on zone append emulation, the direct check to the max_zone_append_sectors queue limit of the disk is replaced by a check using the value returned by queue_max_zone_append_sectors(). Similarly, queue_zone_append_max_show() is modified to use the same accessor so that the sysfs attribute advertizes the non-zero limit that will be used, regardless if it is for native or emulated commands. For stacking drivers, a top device should not need to care if the underlying devices have native or emulated zone append operations. blk_stack_limits() is thus modified to set the top device max_zone_append_sectors limit using the new accessor queue_limits_max_zone_append_sectors(). queue_max_zone_append_sectors() is modified to use this function as well. Stacking drivers that require zone append emulation, e.g. dm-crypt, can still request this feature by calling blk_queue_max_zone_append_sectors() with a 0 limit. Signed-off-by: Damien Le Moal <dlemoal@kernel.org> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Tested-by: Hans Holmberg <hans.holmberg@wdc.com> Tested-by: Dennis Maisenbacher <dennis.maisenbacher@wdc.com> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com> Link: https://lore.kernel.org/r/20240408014128.205141-10-dlemoal@kernel.org Signed-off-by: Jens Axboe <axboe@kernel.dk>
2024-04-08 01:41:09 +00:00
if (nr_sectors > queue_max_zone_append_sectors(q))
block: Introduce REQ_OP_ZONE_APPEND Define REQ_OP_ZONE_APPEND to append-write sectors to a zone of a zoned block device. This is a no-merge write operation. A zone append write BIO must: * Target a zoned block device * Have a sector position indicating the start sector of the target zone * The target zone must be a sequential write zone * The BIO must not cross a zone boundary * The BIO size must not be split to ensure that a single range of LBAs is written with a single command. Implement these checks in generic_make_request_checks() using the helper function blk_check_zone_append(). To avoid write append BIO splitting, introduce the new max_zone_append_sectors queue limit attribute and ensure that a BIO size is always lower than this limit. Export this new limit through sysfs and check these limits in bio_full(). Also when a LLDD can't dispatch a request to a specific zone, it will return BLK_STS_ZONE_RESOURCE indicating this request needs to be delayed, e.g. because the zone it will be dispatched to is still write-locked. If this happens set the request aside in a local list to continue trying dispatching requests such as READ requests or a WRITE/ZONE_APPEND requests targetting other zones. This way we can still keep a high queue depth without starving other requests even if one request can't be served due to zone write-locking. Finally, make sure that the bio sector position indicates the actual write position as indicated by the device on completion. Signed-off-by: Keith Busch <kbusch@kernel.org> [ jth: added zone-append specific add_page and merge_page helpers ] Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2020-05-12 08:55:47 +00:00
return BLK_STS_IOERR;
bio->bi_opf |= REQ_NOMERGE;
return BLK_STS_OK;
}
static void __submit_bio(struct bio *bio)
{
block: fix lost bio for plug enabled bio based device With the following two conditions, bio will be lost: 1) blk plug is not enabled, for example, __blkdev_direct_IO_simple() and __blkdev_direct_IO_async(); 2) bio plug is enabled, for example write IO for raid1/raid10 while bitmap is enabled; Root cause is that blk_finish_plug() will add the bio to curent->bio_list, while such bio will not be handled: __submit_bio_noacct current->bio_list = bio_list_on_stack; blk_start_plug do { dm_submit_bio md_handle_request raid10_write_request -> generate new bio for underlying disks raid1_add_bio_to_plug -> bio is added to plug } while ((bio = bio_list_pop(&bio_list_on_stack[0]))) -> previous bio are all handled blk_finish_plug raid10_unplug raid1_submit_write submit_bio_noacct if (current->bio_list) bio_list_add(&current->bio_list[0], bio) -> add new bio current->bio_list = NULL -> new bio is lost Fix the problem by moving the plug into the while loop, so that current->bio_list will still be handled after blk_finish_plug(). By the way, enable plug for raid1/raid10 in this case will also prevent delay IO handling into daemon thread, which should also improve IO performance. Fixes: 060406c61c7c ("block: add plug while submitting IO") Reported-by: Changhui Zhong <czhong@redhat.com> Closes: https://lore.kernel.org/all/CAGVVp+Xsmzy2G9YuEatfMT6qv1M--YdOCQ0g7z7OVmcTbBxQAg@mail.gmail.com/ Signed-off-by: Yu Kuai <yukuai3@huawei.com> Tested-by: Changhui Zhong <czhong@redhat.com> Link: https://lore.kernel.org/r/20240521200308.983986-1-yukuai1@huaweicloud.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2024-05-21 20:03:08 +00:00
/* If plug is not used, add new plug here to cache nsecs time. */
struct blk_plug plug;
if (unlikely(!blk_crypto_bio_prep(&bio)))
return;
block: fix lost bio for plug enabled bio based device With the following two conditions, bio will be lost: 1) blk plug is not enabled, for example, __blkdev_direct_IO_simple() and __blkdev_direct_IO_async(); 2) bio plug is enabled, for example write IO for raid1/raid10 while bitmap is enabled; Root cause is that blk_finish_plug() will add the bio to curent->bio_list, while such bio will not be handled: __submit_bio_noacct current->bio_list = bio_list_on_stack; blk_start_plug do { dm_submit_bio md_handle_request raid10_write_request -> generate new bio for underlying disks raid1_add_bio_to_plug -> bio is added to plug } while ((bio = bio_list_pop(&bio_list_on_stack[0]))) -> previous bio are all handled blk_finish_plug raid10_unplug raid1_submit_write submit_bio_noacct if (current->bio_list) bio_list_add(&current->bio_list[0], bio) -> add new bio current->bio_list = NULL -> new bio is lost Fix the problem by moving the plug into the while loop, so that current->bio_list will still be handled after blk_finish_plug(). By the way, enable plug for raid1/raid10 in this case will also prevent delay IO handling into daemon thread, which should also improve IO performance. Fixes: 060406c61c7c ("block: add plug while submitting IO") Reported-by: Changhui Zhong <czhong@redhat.com> Closes: https://lore.kernel.org/all/CAGVVp+Xsmzy2G9YuEatfMT6qv1M--YdOCQ0g7z7OVmcTbBxQAg@mail.gmail.com/ Signed-off-by: Yu Kuai <yukuai3@huawei.com> Tested-by: Changhui Zhong <czhong@redhat.com> Link: https://lore.kernel.org/r/20240521200308.983986-1-yukuai1@huaweicloud.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2024-05-21 20:03:08 +00:00
blk_start_plug(&plug);
if (!bdev_test_flag(bio->bi_bdev, BD_HAS_SUBMIT_BIO)) {
blk_mq_submit_bio(bio);
} else if (likely(bio_queue_enter(bio) == 0)) {
struct gendisk *disk = bio->bi_bdev->bd_disk;
disk->fops->submit_bio(bio);
blk_queue_exit(disk->queue);
}
block: fix lost bio for plug enabled bio based device With the following two conditions, bio will be lost: 1) blk plug is not enabled, for example, __blkdev_direct_IO_simple() and __blkdev_direct_IO_async(); 2) bio plug is enabled, for example write IO for raid1/raid10 while bitmap is enabled; Root cause is that blk_finish_plug() will add the bio to curent->bio_list, while such bio will not be handled: __submit_bio_noacct current->bio_list = bio_list_on_stack; blk_start_plug do { dm_submit_bio md_handle_request raid10_write_request -> generate new bio for underlying disks raid1_add_bio_to_plug -> bio is added to plug } while ((bio = bio_list_pop(&bio_list_on_stack[0]))) -> previous bio are all handled blk_finish_plug raid10_unplug raid1_submit_write submit_bio_noacct if (current->bio_list) bio_list_add(&current->bio_list[0], bio) -> add new bio current->bio_list = NULL -> new bio is lost Fix the problem by moving the plug into the while loop, so that current->bio_list will still be handled after blk_finish_plug(). By the way, enable plug for raid1/raid10 in this case will also prevent delay IO handling into daemon thread, which should also improve IO performance. Fixes: 060406c61c7c ("block: add plug while submitting IO") Reported-by: Changhui Zhong <czhong@redhat.com> Closes: https://lore.kernel.org/all/CAGVVp+Xsmzy2G9YuEatfMT6qv1M--YdOCQ0g7z7OVmcTbBxQAg@mail.gmail.com/ Signed-off-by: Yu Kuai <yukuai3@huawei.com> Tested-by: Changhui Zhong <czhong@redhat.com> Link: https://lore.kernel.org/r/20240521200308.983986-1-yukuai1@huaweicloud.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2024-05-21 20:03:08 +00:00
blk_finish_plug(&plug);
}
/*
* The loop in this function may be a bit non-obvious, and so deserves some
* explanation:
*
* - Before entering the loop, bio->bi_next is NULL (as all callers ensure
* that), so we have a list with a single bio.
* - We pretend that we have just taken it off a longer list, so we assign
* bio_list to a pointer to the bio_list_on_stack, thus initialising the
* bio_list of new bios to be added. ->submit_bio() may indeed add some more
* bios through a recursive call to submit_bio_noacct. If it did, we find a
* non-NULL value in bio_list and re-enter the loop from the top.
* - In this case we really did just take the bio of the top of the list (no
* pretending) and so remove it from bio_list, and call into ->submit_bio()
* again.
*
* bio_list_on_stack[0] contains bios submitted by the current ->submit_bio.
* bio_list_on_stack[1] contains bios that were submitted before the current
* ->submit_bio, but that haven't been processed yet.
*/
static void __submit_bio_noacct(struct bio *bio)
{
struct bio_list bio_list_on_stack[2];
BUG_ON(bio->bi_next);
bio_list_init(&bio_list_on_stack[0]);
current->bio_list = bio_list_on_stack;
do {
struct request_queue *q = bdev_get_queue(bio->bi_bdev);
struct bio_list lower, same;
/*
* Create a fresh bio_list for all subordinate requests.
*/
bio_list_on_stack[1] = bio_list_on_stack[0];
bio_list_init(&bio_list_on_stack[0]);
__submit_bio(bio);
/*
* Sort new bios into those for a lower level and those for the
* same level.
*/
bio_list_init(&lower);
bio_list_init(&same);
while ((bio = bio_list_pop(&bio_list_on_stack[0])) != NULL)
if (q == bdev_get_queue(bio->bi_bdev))
bio_list_add(&same, bio);
else
bio_list_add(&lower, bio);
/*
* Now assemble so we handle the lowest level first.
*/
bio_list_merge(&bio_list_on_stack[0], &lower);
bio_list_merge(&bio_list_on_stack[0], &same);
bio_list_merge(&bio_list_on_stack[0], &bio_list_on_stack[1]);
} while ((bio = bio_list_pop(&bio_list_on_stack[0])));
current->bio_list = NULL;
}
static void __submit_bio_noacct_mq(struct bio *bio)
{
struct bio_list bio_list[2] = { };
current->bio_list = bio_list;
do {
__submit_bio(bio);
} while ((bio = bio_list_pop(&bio_list[0])));
current->bio_list = NULL;
}
void submit_bio_noacct_nocheck(struct bio *bio)
When stacked block devices are in-use (e.g. md or dm), the recursive calls to generic_make_request can use up a lot of space, and we would rather they didn't. As generic_make_request is a void function, and as it is generally not expected that it will have any effect immediately, it is safe to delay any call to generic_make_request until there is sufficient stack space available. As ->bi_next is reserved for the driver to use, it can have no valid value when generic_make_request is called, and as __make_request implicitly assumes it will be NULL (ELEVATOR_BACK_MERGE fork of switch) we can be certain that all callers set it to NULL. We can therefore safely use bi_next to link pending requests together, providing we clear it before making the real call. So, we choose to allow each thread to only be active in one generic_make_request at a time. If a subsequent (recursive) call is made, the bio is linked into a per-thread list, and is handled when the active call completes. As the list of pending bios is per-thread, there are no locking issues to worry about. I say above that it is "safe to delay any call...". There are, however, some behaviours of a make_request_fn which would make it unsafe. These include any behaviour that assumes anything will have changed after a recursive call to generic_make_request. These could include: - waiting for that call to finish and call it's bi_end_io function. md use to sometimes do this (marking the superblock dirty before completing a write) but doesn't any more - inspecting the bio for fields that generic_make_request might change, such as bi_sector or bi_bdev. It is hard to see a good reason for this, and I don't think anyone actually does it. - inspecing the queue to see if, e.g. it is 'full' yet. Again, I think this is very unlikely to be useful, or to be done. Signed-off-by: Neil Brown <neilb@suse.de> Cc: Jens Axboe <axboe@kernel.dk> Cc: <dm-devel@redhat.com> Alasdair G Kergon <agk@redhat.com> said: I can see nothing wrong with this in principle. For device-mapper at the moment though it's essential that, while the bio mappings may now get delayed, they still get processed in exactly the same order as they were passed to generic_make_request(). My main concern is whether the timing changes implicit in this patch will make the rare data-corrupting races in the existing snapshot code more likely. (I'm working on a fix for these races, but the unfinished patch is already several hundred lines long.) It would be helpful if some people on this mailing list would test this patch in various scenarios and report back. Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2007-05-01 07:53:42 +00:00
{
blk_cgroup_bio_start(bio);
blkcg_bio_issue_init(bio);
if (!bio_flagged(bio, BIO_TRACE_COMPLETION)) {
trace_block_bio_queue(bio);
/*
* Now that enqueuing has been traced, we need to trace
* completion as well.
*/
bio_set_flag(bio, BIO_TRACE_COMPLETION);
}
/*
* We only want one ->submit_bio to be active at a time, else stack
* usage with stacked devices could be a problem. Use current->bio_list
* to collect a list of requests submited by a ->submit_bio method while
* it is active, and then process them after it returned.
*/
if (current->bio_list)
bio_list_add(&current->bio_list[0], bio);
else if (!bdev_test_flag(bio->bi_bdev, BD_HAS_SUBMIT_BIO))
__submit_bio_noacct_mq(bio);
else
__submit_bio_noacct(bio);
When stacked block devices are in-use (e.g. md or dm), the recursive calls to generic_make_request can use up a lot of space, and we would rather they didn't. As generic_make_request is a void function, and as it is generally not expected that it will have any effect immediately, it is safe to delay any call to generic_make_request until there is sufficient stack space available. As ->bi_next is reserved for the driver to use, it can have no valid value when generic_make_request is called, and as __make_request implicitly assumes it will be NULL (ELEVATOR_BACK_MERGE fork of switch) we can be certain that all callers set it to NULL. We can therefore safely use bi_next to link pending requests together, providing we clear it before making the real call. So, we choose to allow each thread to only be active in one generic_make_request at a time. If a subsequent (recursive) call is made, the bio is linked into a per-thread list, and is handled when the active call completes. As the list of pending bios is per-thread, there are no locking issues to worry about. I say above that it is "safe to delay any call...". There are, however, some behaviours of a make_request_fn which would make it unsafe. These include any behaviour that assumes anything will have changed after a recursive call to generic_make_request. These could include: - waiting for that call to finish and call it's bi_end_io function. md use to sometimes do this (marking the superblock dirty before completing a write) but doesn't any more - inspecting the bio for fields that generic_make_request might change, such as bi_sector or bi_bdev. It is hard to see a good reason for this, and I don't think anyone actually does it. - inspecing the queue to see if, e.g. it is 'full' yet. Again, I think this is very unlikely to be useful, or to be done. Signed-off-by: Neil Brown <neilb@suse.de> Cc: Jens Axboe <axboe@kernel.dk> Cc: <dm-devel@redhat.com> Alasdair G Kergon <agk@redhat.com> said: I can see nothing wrong with this in principle. For device-mapper at the moment though it's essential that, while the bio mappings may now get delayed, they still get processed in exactly the same order as they were passed to generic_make_request(). My main concern is whether the timing changes implicit in this patch will make the rare data-corrupting races in the existing snapshot code more likely. (I'm working on a fix for these races, but the unfinished patch is already several hundred lines long.) It would be helpful if some people on this mailing list would test this patch in various scenarios and report back. Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2007-05-01 07:53:42 +00:00
}
/**
* submit_bio_noacct - re-submit a bio to the block device layer for I/O
* @bio: The bio describing the location in memory and on the device.
*
* This is a version of submit_bio() that shall only be used for I/O that is
* resubmitted to lower level drivers by stacking block drivers. All file
* systems and other upper level users of the block layer should use
* submit_bio() instead.
*/
void submit_bio_noacct(struct bio *bio)
{
struct block_device *bdev = bio->bi_bdev;
struct request_queue *q = bdev_get_queue(bdev);
blk_status_t status = BLK_STS_IOERR;
might_sleep();
/*
* For a REQ_NOWAIT based request, return -EOPNOTSUPP
* if queue does not support NOWAIT.
*/
if ((bio->bi_opf & REQ_NOWAIT) && !bdev_nowait(bdev))
goto not_supported;
if (should_fail_bio(bio))
goto end_io;
bio_check_ro(bio);
if (!bio_flagged(bio, BIO_REMAPPED)) {
if (unlikely(bio_check_eod(bio)))
goto end_io;
if (bdev_is_partition(bdev) &&
unlikely(blk_partition_remap(bio)))
goto end_io;
}
/*
* Filter flush bio's early so that bio based drivers without flush
* support don't have to worry about them.
*/
if (op_is_flush(bio->bi_opf)) {
if (WARN_ON_ONCE(bio_op(bio) != REQ_OP_WRITE &&
bio_op(bio) != REQ_OP_ZONE_APPEND))
goto end_io;
if (!test_bit(QUEUE_FLAG_WC, &q->queue_flags)) {
bio->bi_opf &= ~(REQ_PREFLUSH | REQ_FUA);
if (!bio_sectors(bio)) {
status = BLK_STS_OK;
goto end_io;
}
}
}
if (!test_bit(QUEUE_FLAG_POLL, &q->queue_flags))
bio_clear_polled(bio);
switch (bio_op(bio)) {
case REQ_OP_READ:
case REQ_OP_WRITE:
break;
case REQ_OP_FLUSH:
/*
* REQ_OP_FLUSH can't be submitted through bios, it is only
* synthetized in struct request by the flush state machine.
*/
goto not_supported;
case REQ_OP_DISCARD:
if (!bdev_max_discard_sectors(bdev))
goto not_supported;
break;
case REQ_OP_SECURE_ERASE:
if (!bdev_max_secure_erase_sectors(bdev))
goto not_supported;
break;
block: Introduce REQ_OP_ZONE_APPEND Define REQ_OP_ZONE_APPEND to append-write sectors to a zone of a zoned block device. This is a no-merge write operation. A zone append write BIO must: * Target a zoned block device * Have a sector position indicating the start sector of the target zone * The target zone must be a sequential write zone * The BIO must not cross a zone boundary * The BIO size must not be split to ensure that a single range of LBAs is written with a single command. Implement these checks in generic_make_request_checks() using the helper function blk_check_zone_append(). To avoid write append BIO splitting, introduce the new max_zone_append_sectors queue limit attribute and ensure that a BIO size is always lower than this limit. Export this new limit through sysfs and check these limits in bio_full(). Also when a LLDD can't dispatch a request to a specific zone, it will return BLK_STS_ZONE_RESOURCE indicating this request needs to be delayed, e.g. because the zone it will be dispatched to is still write-locked. If this happens set the request aside in a local list to continue trying dispatching requests such as READ requests or a WRITE/ZONE_APPEND requests targetting other zones. This way we can still keep a high queue depth without starving other requests even if one request can't be served due to zone write-locking. Finally, make sure that the bio sector position indicates the actual write position as indicated by the device on completion. Signed-off-by: Keith Busch <kbusch@kernel.org> [ jth: added zone-append specific add_page and merge_page helpers ] Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2020-05-12 08:55:47 +00:00
case REQ_OP_ZONE_APPEND:
status = blk_check_zone_append(q, bio);
if (status != BLK_STS_OK)
goto end_io;
break;
case REQ_OP_WRITE_ZEROES:
if (!q->limits.max_write_zeroes_sectors)
goto not_supported;
break;
case REQ_OP_ZONE_RESET:
2019-10-27 14:05:45 +00:00
case REQ_OP_ZONE_OPEN:
case REQ_OP_ZONE_CLOSE:
case REQ_OP_ZONE_FINISH:
if (!bdev_is_zoned(bio->bi_bdev))
goto not_supported;
break;
case REQ_OP_ZONE_RESET_ALL:
if (!bdev_is_zoned(bio->bi_bdev) || !blk_queue_zone_resetall(q))
goto not_supported;
break;
case REQ_OP_DRV_IN:
case REQ_OP_DRV_OUT:
/*
* Driver private operations are only used with passthrough
* requests.
*/
fallthrough;
default:
goto not_supported;
}
blkcg: Remove extra blkcg_bio_issue_init KASAN reports a use-after-free report when doing block test: ================================================================== [10050.967049] BUG: KASAN: use-after-free in submit_bio_checks+0x1539/0x1550 [10050.977638] Call Trace: [10050.978190] dump_stack+0x9b/0xce [10050.979674] print_address_description.constprop.6+0x3e/0x60 [10050.983510] kasan_report.cold.9+0x22/0x3a [10050.986089] submit_bio_checks+0x1539/0x1550 [10050.989576] submit_bio_noacct+0x83/0xc80 [10050.993714] submit_bio+0xa7/0x330 [10050.994435] mpage_readahead+0x380/0x500 [10050.998009] read_pages+0x1c1/0xbf0 [10051.002057] page_cache_ra_unbounded+0x4c2/0x6f0 [10051.007413] do_page_cache_ra+0xda/0x110 [10051.008207] force_page_cache_ra+0x23d/0x3d0 [10051.009087] page_cache_sync_ra+0xca/0x300 [10051.009970] generic_file_buffered_read+0xbea/0x2130 [10051.012685] generic_file_read_iter+0x315/0x490 [10051.014472] blkdev_read_iter+0x113/0x1b0 [10051.015300] aio_read+0x2ad/0x450 [10051.023786] io_submit_one+0xc8e/0x1d60 [10051.029855] __se_sys_io_submit+0x125/0x350 [10051.033442] do_syscall_64+0x2d/0x40 [10051.034156] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [10051.048733] Allocated by task 18598: [10051.049482] kasan_save_stack+0x19/0x40 [10051.050263] __kasan_kmalloc.constprop.1+0xc1/0xd0 [10051.051230] kmem_cache_alloc+0x146/0x440 [10051.052060] mempool_alloc+0x125/0x2f0 [10051.052818] bio_alloc_bioset+0x353/0x590 [10051.053658] mpage_alloc+0x3b/0x240 [10051.054382] do_mpage_readpage+0xddf/0x1ef0 [10051.055250] mpage_readahead+0x264/0x500 [10051.056060] read_pages+0x1c1/0xbf0 [10051.056758] page_cache_ra_unbounded+0x4c2/0x6f0 [10051.057702] do_page_cache_ra+0xda/0x110 [10051.058511] force_page_cache_ra+0x23d/0x3d0 [10051.059373] page_cache_sync_ra+0xca/0x300 [10051.060198] generic_file_buffered_read+0xbea/0x2130 [10051.061195] generic_file_read_iter+0x315/0x490 [10051.062189] blkdev_read_iter+0x113/0x1b0 [10051.063015] aio_read+0x2ad/0x450 [10051.063686] io_submit_one+0xc8e/0x1d60 [10051.064467] __se_sys_io_submit+0x125/0x350 [10051.065318] do_syscall_64+0x2d/0x40 [10051.066082] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [10051.067455] Freed by task 13307: [10051.068136] kasan_save_stack+0x19/0x40 [10051.068931] kasan_set_track+0x1c/0x30 [10051.069726] kasan_set_free_info+0x1b/0x30 [10051.070621] __kasan_slab_free+0x111/0x160 [10051.071480] kmem_cache_free+0x94/0x460 [10051.072256] mempool_free+0xd6/0x320 [10051.072985] bio_free+0xe0/0x130 [10051.073630] bio_put+0xab/0xe0 [10051.074252] bio_endio+0x3a6/0x5d0 [10051.074984] blk_update_request+0x590/0x1370 [10051.075870] scsi_end_request+0x7d/0x400 [10051.076667] scsi_io_completion+0x1aa/0xe50 [10051.077503] scsi_softirq_done+0x11b/0x240 [10051.078344] blk_mq_complete_request+0xd4/0x120 [10051.079275] scsi_mq_done+0xf0/0x200 [10051.080036] virtscsi_vq_done+0xbc/0x150 [10051.080850] vring_interrupt+0x179/0x390 [10051.081650] __handle_irq_event_percpu+0xf7/0x490 [10051.082626] handle_irq_event_percpu+0x7b/0x160 [10051.083527] handle_irq_event+0xcc/0x170 [10051.084297] handle_edge_irq+0x215/0xb20 [10051.085122] asm_call_irq_on_stack+0xf/0x20 [10051.085986] common_interrupt+0xae/0x120 [10051.086830] asm_common_interrupt+0x1e/0x40 ================================================================== Bio will be checked at beginning of submit_bio_noacct(). If bio needs to be throttled, it will start the timer and stop submit bio directly. Bio will submit in blk_throtl_dispatch_work_fn() when the timer expires. But in the current process, if bio is throttled, it will still set bio issue->value by blkcg_bio_issue_init(). This is redundant and may cause the above use-after-free. CPU0 CPU1 submit_bio submit_bio_noacct submit_bio_checks blk_throtl_bio() <=mod_timer(&sq->pending_timer blk_throtl_dispatch_work_fn submit_bio_noacct() <= bio have throttle tag, will throw directly and bio issue->value will be set here bio_endio() bio_put() bio_free() <= free this bio blkcg_bio_issue_init(bio) <= bio has been freed and will lead to UAF return BLK_QC_T_NONE Fix this by remove extra blkcg_bio_issue_init. Fixes: e439bedf6b24 (blkcg: consolidate bio_issue_init() to be a part of core) Signed-off-by: Laibin Qiu <qiulaibin@huawei.com> Link: https://lore.kernel.org/r/20211112093354.3581504-1-qiulaibin@huawei.com Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2021-11-12 09:33:54 +00:00
if (blk_throtl_bio(bio))
return;
submit_bio_noacct_nocheck(bio);
return;
not_supported:
status = BLK_STS_NOTSUPP;
end_io:
bio->bi_status = status;
bio_endio(bio);
When stacked block devices are in-use (e.g. md or dm), the recursive calls to generic_make_request can use up a lot of space, and we would rather they didn't. As generic_make_request is a void function, and as it is generally not expected that it will have any effect immediately, it is safe to delay any call to generic_make_request until there is sufficient stack space available. As ->bi_next is reserved for the driver to use, it can have no valid value when generic_make_request is called, and as __make_request implicitly assumes it will be NULL (ELEVATOR_BACK_MERGE fork of switch) we can be certain that all callers set it to NULL. We can therefore safely use bi_next to link pending requests together, providing we clear it before making the real call. So, we choose to allow each thread to only be active in one generic_make_request at a time. If a subsequent (recursive) call is made, the bio is linked into a per-thread list, and is handled when the active call completes. As the list of pending bios is per-thread, there are no locking issues to worry about. I say above that it is "safe to delay any call...". There are, however, some behaviours of a make_request_fn which would make it unsafe. These include any behaviour that assumes anything will have changed after a recursive call to generic_make_request. These could include: - waiting for that call to finish and call it's bi_end_io function. md use to sometimes do this (marking the superblock dirty before completing a write) but doesn't any more - inspecting the bio for fields that generic_make_request might change, such as bi_sector or bi_bdev. It is hard to see a good reason for this, and I don't think anyone actually does it. - inspecing the queue to see if, e.g. it is 'full' yet. Again, I think this is very unlikely to be useful, or to be done. Signed-off-by: Neil Brown <neilb@suse.de> Cc: Jens Axboe <axboe@kernel.dk> Cc: <dm-devel@redhat.com> Alasdair G Kergon <agk@redhat.com> said: I can see nothing wrong with this in principle. For device-mapper at the moment though it's essential that, while the bio mappings may now get delayed, they still get processed in exactly the same order as they were passed to generic_make_request(). My main concern is whether the timing changes implicit in this patch will make the rare data-corrupting races in the existing snapshot code more likely. (I'm working on a fix for these races, but the unfinished patch is already several hundred lines long.) It would be helpful if some people on this mailing list would test this patch in various scenarios and report back. Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2007-05-01 07:53:42 +00:00
}
EXPORT_SYMBOL(submit_bio_noacct);
static void bio_set_ioprio(struct bio *bio)
{
/* Nobody set ioprio so far? Initialize it based on task's nice value */
if (IOPRIO_PRIO_CLASS(bio->bi_ioprio) == IOPRIO_CLASS_NONE)
bio->bi_ioprio = get_current_ioprio();
blkcg_set_ioprio(bio);
}
/**
* submit_bio - submit a bio to the block device layer for I/O
* @bio: The &struct bio which describes the I/O
*
* submit_bio() is used to submit I/O requests to block devices. It is passed a
* fully set up &struct bio that describes the I/O that needs to be done. The
* bio will be send to the device described by the bi_bdev field.
*
* The success/failure status of the request, along with notification of
* completion, is delivered asynchronously through the ->bi_end_io() callback
* in @bio. The bio must NOT be touched by the caller until ->bi_end_io() has
* been called.
*/
void submit_bio(struct bio *bio)
{
if (bio_op(bio) == REQ_OP_READ) {
task_io_account_read(bio->bi_iter.bi_size);
count_vm_events(PGPGIN, bio_sectors(bio));
} else if (bio_op(bio) == REQ_OP_WRITE) {
count_vm_events(PGPGOUT, bio_sectors(bio));
}
bio_set_ioprio(bio);
submit_bio_noacct(bio);
}
EXPORT_SYMBOL(submit_bio);
/**
* bio_poll - poll for BIO completions
* @bio: bio to poll for
* @iob: batches of IO
* @flags: BLK_POLL_* flags that control the behavior
*
* Poll for completions on queue associated with the bio. Returns number of
* completed entries found.
*
* Note: the caller must either be the context that submitted @bio, or
* be in a RCU critical section to prevent freeing of @bio.
*/
int bio_poll(struct bio *bio, struct io_comp_batch *iob, unsigned int flags)
{
blk_qc_t cookie = READ_ONCE(bio->bi_cookie);
block: be a bit more careful in checking for NULL bdev while polling Wei reports a crash with an application using polled IO: PGD 14265e067 P4D 14265e067 PUD 47ec50067 PMD 0 Oops: 0000 [#1] SMP CPU: 0 PID: 21915 Comm: iocore_0 Kdump: loaded Tainted: G S 5.12.0-0_fbk12_clang_7346_g1bb6f2e7058f #1 Hardware name: Wiwynn Delta Lake MP T8/Delta Lake-Class2, BIOS Y3DLM08 04/10/2022 RIP: 0010:bio_poll+0x25/0x200 Code: 0f 1f 44 00 00 0f 1f 44 00 00 55 41 57 41 56 41 55 41 54 53 48 83 ec 28 65 48 8b 04 25 28 00 00 00 48 89 44 24 20 48 8b 47 08 <48> 8b 80 70 02 00 00 4c 8b 70 50 8b 6f 34 31 db 83 fd ff 75 25 65 RSP: 0018:ffffc90005fafdf8 EFLAGS: 00010292 RAX: 0000000000000000 RBX: 0000000000000000 RCX: 74b43cd65dd66600 RDX: 0000000000000003 RSI: ffffc90005fafe78 RDI: ffff8884b614e140 RBP: ffff88849964df78 R08: 0000000000000000 R09: 0000000000000008 R10: 0000000000000000 R11: 0000000000000000 R12: ffff88849964df00 R13: ffffc90005fafe78 R14: ffff888137d3c378 R15: 0000000000000001 FS: 00007fd195000640(0000) GS:ffff88903f400000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000270 CR3: 0000000466121001 CR4: 00000000007706f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: iocb_bio_iopoll+0x1d/0x30 io_do_iopoll+0xac/0x250 __se_sys_io_uring_enter+0x3c5/0x5a0 ? __x64_sys_write+0x89/0xd0 do_syscall_64+0x2d/0x40 entry_SYSCALL_64_after_hwframe+0x44/0xae RIP: 0033:0x94f225d Code: 24 cc 00 00 00 41 8b 84 24 d0 00 00 00 c1 e0 04 83 e0 10 41 09 c2 8b 33 8b 53 04 4c 8b 43 18 4c 63 4b 0c b8 aa 01 00 00 0f 05 <85> c0 0f 88 85 00 00 00 29 03 45 84 f6 0f 84 88 00 00 00 41 f6 c7 RSP: 002b:00007fd194ffcd88 EFLAGS: 00000202 ORIG_RAX: 00000000000001aa RAX: ffffffffffffffda RBX: 00007fd194ffcdc0 RCX: 00000000094f225d RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000007 RBP: 00007fd194ffcdb0 R08: 0000000000000000 R09: 0000000000000008 R10: 0000000000000001 R11: 0000000000000202 R12: 00007fd269d68030 R13: 0000000000000000 R14: 0000000000000001 R15: 0000000000000000 which is due to bio->bi_bdev being NULL. This can happen if we have two tasks doing polled IO, and task B ends up completing IO from task A if they are sharing a poll queue. If task B completes the IO and puts the bio into our cache, then it can allocate that bio again before task A is done polling for it. As that would necessitate a preempt between the two tasks, it's enough to just be a bit more careful in checking for whether or not bio->bi_bdev is NULL. Reported-and-tested-by: Wei Zhang <wzhang@meta.com> Cc: stable@vger.kernel.org Fixes: be4d234d7aeb ("bio: add allocation cache abstraction") Reviewed-by: Keith Busch <kbusch@kernel.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2023-02-24 17:01:19 +00:00
struct block_device *bdev;
struct request_queue *q;
int ret = 0;
block: be a bit more careful in checking for NULL bdev while polling Wei reports a crash with an application using polled IO: PGD 14265e067 P4D 14265e067 PUD 47ec50067 PMD 0 Oops: 0000 [#1] SMP CPU: 0 PID: 21915 Comm: iocore_0 Kdump: loaded Tainted: G S 5.12.0-0_fbk12_clang_7346_g1bb6f2e7058f #1 Hardware name: Wiwynn Delta Lake MP T8/Delta Lake-Class2, BIOS Y3DLM08 04/10/2022 RIP: 0010:bio_poll+0x25/0x200 Code: 0f 1f 44 00 00 0f 1f 44 00 00 55 41 57 41 56 41 55 41 54 53 48 83 ec 28 65 48 8b 04 25 28 00 00 00 48 89 44 24 20 48 8b 47 08 <48> 8b 80 70 02 00 00 4c 8b 70 50 8b 6f 34 31 db 83 fd ff 75 25 65 RSP: 0018:ffffc90005fafdf8 EFLAGS: 00010292 RAX: 0000000000000000 RBX: 0000000000000000 RCX: 74b43cd65dd66600 RDX: 0000000000000003 RSI: ffffc90005fafe78 RDI: ffff8884b614e140 RBP: ffff88849964df78 R08: 0000000000000000 R09: 0000000000000008 R10: 0000000000000000 R11: 0000000000000000 R12: ffff88849964df00 R13: ffffc90005fafe78 R14: ffff888137d3c378 R15: 0000000000000001 FS: 00007fd195000640(0000) GS:ffff88903f400000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000270 CR3: 0000000466121001 CR4: 00000000007706f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: iocb_bio_iopoll+0x1d/0x30 io_do_iopoll+0xac/0x250 __se_sys_io_uring_enter+0x3c5/0x5a0 ? __x64_sys_write+0x89/0xd0 do_syscall_64+0x2d/0x40 entry_SYSCALL_64_after_hwframe+0x44/0xae RIP: 0033:0x94f225d Code: 24 cc 00 00 00 41 8b 84 24 d0 00 00 00 c1 e0 04 83 e0 10 41 09 c2 8b 33 8b 53 04 4c 8b 43 18 4c 63 4b 0c b8 aa 01 00 00 0f 05 <85> c0 0f 88 85 00 00 00 29 03 45 84 f6 0f 84 88 00 00 00 41 f6 c7 RSP: 002b:00007fd194ffcd88 EFLAGS: 00000202 ORIG_RAX: 00000000000001aa RAX: ffffffffffffffda RBX: 00007fd194ffcdc0 RCX: 00000000094f225d RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000007 RBP: 00007fd194ffcdb0 R08: 0000000000000000 R09: 0000000000000008 R10: 0000000000000001 R11: 0000000000000202 R12: 00007fd269d68030 R13: 0000000000000000 R14: 0000000000000001 R15: 0000000000000000 which is due to bio->bi_bdev being NULL. This can happen if we have two tasks doing polled IO, and task B ends up completing IO from task A if they are sharing a poll queue. If task B completes the IO and puts the bio into our cache, then it can allocate that bio again before task A is done polling for it. As that would necessitate a preempt between the two tasks, it's enough to just be a bit more careful in checking for whether or not bio->bi_bdev is NULL. Reported-and-tested-by: Wei Zhang <wzhang@meta.com> Cc: stable@vger.kernel.org Fixes: be4d234d7aeb ("bio: add allocation cache abstraction") Reviewed-by: Keith Busch <kbusch@kernel.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2023-02-24 17:01:19 +00:00
bdev = READ_ONCE(bio->bi_bdev);
if (!bdev)
return 0;
q = bdev_get_queue(bdev);
if (cookie == BLK_QC_T_NONE ||
!test_bit(QUEUE_FLAG_POLL, &q->queue_flags))
return 0;
blk_flush_plug(current->plug, false);
block: treat poll queue enter similarly to timeouts We ran into an issue where a production workload would randomly grind to a halt and not continue until the pending IO had timed out. This turned out to be a complicated interaction between queue freezing and polled IO: 1) You have an application that does polled IO. At any point in time, there may be polled IO pending. 2) You have a monitoring application that issues a passthrough command, which is marked with side effects such that it needs to freeze the queue. 3) Passthrough command is started, which calls blk_freeze_queue_start() on the device. At this point the queue is marked frozen, and any attempt to enter the queue will fail (for non-blocking) or block. 4) Now the driver calls blk_mq_freeze_queue_wait(), which will return when the queue is quiesced and pending IO has completed. 5) The pending IO is polled IO, but any attempt to poll IO through the normal iocb_bio_iopoll() -> bio_poll() will fail when it gets to bio_queue_enter() as the queue is frozen. Rather than poll and complete IO, the polling threads will sit in a tight loop attempting to poll, but failing to enter the queue to do so. The end result is that progress for either application will be stalled until all pending polled IO has timed out. This causes obvious huge latency issues for the application doing polled IO, but also long delays for passthrough command. Fix this by treating queue enter for polled IO just like we do for timeouts. This allows quick quiesce of the queue as we still poll and complete this IO, while still disallowing queueing up new IO. Reviewed-by: Keith Busch <kbusch@kernel.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2023-01-20 14:51:07 +00:00
/*
* We need to be able to enter a frozen queue, similar to how
* timeouts also need to do that. If that is blocked, then we can
* have pending IO when a queue freeze is started, and then the
* wait for the freeze to finish will wait for polled requests to
* timeout as the poller is preventer from entering the queue and
* completing them. As long as we prevent new IO from being queued,
* that should be all that matters.
*/
if (!percpu_ref_tryget(&q->q_usage_counter))
return 0;
if (queue_is_mq(q)) {
ret = blk_mq_poll(q, cookie, iob, flags);
} else {
struct gendisk *disk = q->disk;
if (disk && disk->fops->poll_bio)
ret = disk->fops->poll_bio(bio, iob, flags);
}
blk_queue_exit(q);
return ret;
}
EXPORT_SYMBOL_GPL(bio_poll);
/*
* Helper to implement file_operations.iopoll. Requires the bio to be stored
* in iocb->private, and cleared before freeing the bio.
*/
int iocb_bio_iopoll(struct kiocb *kiocb, struct io_comp_batch *iob,
unsigned int flags)
{
struct bio *bio;
int ret = 0;
/*
* Note: the bio cache only uses SLAB_TYPESAFE_BY_RCU, so bio can
* point to a freshly allocated bio at this point. If that happens
* we have a few cases to consider:
*
* 1) the bio is beeing initialized and bi_bdev is NULL. We can just
* simply nothing in this case
* 2) the bio points to a not poll enabled device. bio_poll will catch
* this and return 0
* 3) the bio points to a poll capable device, including but not
* limited to the one that the original bio pointed to. In this
* case we will call into the actual poll method and poll for I/O,
* even if we don't need to, but it won't cause harm either.
*
* For cases 2) and 3) above the RCU grace period ensures that bi_bdev
* is still allocated. Because partitions hold a reference to the whole
* device bdev and thus disk, the disk is also still valid. Grabbing
* a reference to the queue in bio_poll() ensures the hctxs and requests
* are still valid as well.
*/
rcu_read_lock();
bio = READ_ONCE(kiocb->private);
block: be a bit more careful in checking for NULL bdev while polling Wei reports a crash with an application using polled IO: PGD 14265e067 P4D 14265e067 PUD 47ec50067 PMD 0 Oops: 0000 [#1] SMP CPU: 0 PID: 21915 Comm: iocore_0 Kdump: loaded Tainted: G S 5.12.0-0_fbk12_clang_7346_g1bb6f2e7058f #1 Hardware name: Wiwynn Delta Lake MP T8/Delta Lake-Class2, BIOS Y3DLM08 04/10/2022 RIP: 0010:bio_poll+0x25/0x200 Code: 0f 1f 44 00 00 0f 1f 44 00 00 55 41 57 41 56 41 55 41 54 53 48 83 ec 28 65 48 8b 04 25 28 00 00 00 48 89 44 24 20 48 8b 47 08 <48> 8b 80 70 02 00 00 4c 8b 70 50 8b 6f 34 31 db 83 fd ff 75 25 65 RSP: 0018:ffffc90005fafdf8 EFLAGS: 00010292 RAX: 0000000000000000 RBX: 0000000000000000 RCX: 74b43cd65dd66600 RDX: 0000000000000003 RSI: ffffc90005fafe78 RDI: ffff8884b614e140 RBP: ffff88849964df78 R08: 0000000000000000 R09: 0000000000000008 R10: 0000000000000000 R11: 0000000000000000 R12: ffff88849964df00 R13: ffffc90005fafe78 R14: ffff888137d3c378 R15: 0000000000000001 FS: 00007fd195000640(0000) GS:ffff88903f400000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000270 CR3: 0000000466121001 CR4: 00000000007706f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: iocb_bio_iopoll+0x1d/0x30 io_do_iopoll+0xac/0x250 __se_sys_io_uring_enter+0x3c5/0x5a0 ? __x64_sys_write+0x89/0xd0 do_syscall_64+0x2d/0x40 entry_SYSCALL_64_after_hwframe+0x44/0xae RIP: 0033:0x94f225d Code: 24 cc 00 00 00 41 8b 84 24 d0 00 00 00 c1 e0 04 83 e0 10 41 09 c2 8b 33 8b 53 04 4c 8b 43 18 4c 63 4b 0c b8 aa 01 00 00 0f 05 <85> c0 0f 88 85 00 00 00 29 03 45 84 f6 0f 84 88 00 00 00 41 f6 c7 RSP: 002b:00007fd194ffcd88 EFLAGS: 00000202 ORIG_RAX: 00000000000001aa RAX: ffffffffffffffda RBX: 00007fd194ffcdc0 RCX: 00000000094f225d RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000007 RBP: 00007fd194ffcdb0 R08: 0000000000000000 R09: 0000000000000008 R10: 0000000000000001 R11: 0000000000000202 R12: 00007fd269d68030 R13: 0000000000000000 R14: 0000000000000001 R15: 0000000000000000 which is due to bio->bi_bdev being NULL. This can happen if we have two tasks doing polled IO, and task B ends up completing IO from task A if they are sharing a poll queue. If task B completes the IO and puts the bio into our cache, then it can allocate that bio again before task A is done polling for it. As that would necessitate a preempt between the two tasks, it's enough to just be a bit more careful in checking for whether or not bio->bi_bdev is NULL. Reported-and-tested-by: Wei Zhang <wzhang@meta.com> Cc: stable@vger.kernel.org Fixes: be4d234d7aeb ("bio: add allocation cache abstraction") Reviewed-by: Keith Busch <kbusch@kernel.org> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2023-02-24 17:01:19 +00:00
if (bio)
ret = bio_poll(bio, iob, flags);
rcu_read_unlock();
return ret;
}
EXPORT_SYMBOL_GPL(iocb_bio_iopoll);
void update_io_ticks(struct block_device *part, unsigned long now, bool end)
{
unsigned long stamp;
again:
stamp = READ_ONCE(part->bd_stamp);
block: support to account io_ticks precisely Currently, io_ticks is accounted based on sampling, specifically update_io_ticks() will always account io_ticks by 1 jiffies from bdev_start_io_acct()/blk_account_io_start(), and the result can be inaccurate, for example(HZ is 250): Test script: fio -filename=/dev/sda -bs=4k -rw=write -direct=1 -name=test -thinktime=4ms Test result: util is about 90%, while the disk is really idle. This behaviour is introduced by commit 5b18b5a73760 ("block: delete part_round_stats and switch to less precise counting"), however, there was a key point that is missed that this patch also improve performance a lot: Before the commit: part_round_stats: if (part->stamp != now) stats |= 1; part_in_flight() -> there can be lots of task here in 1 jiffies. part_round_stats_single() __part_stat_add() part->stamp = now; After the commit: update_io_ticks: stamp = part->bd_stamp; if (time_after(now, stamp)) if (try_cmpxchg()) __part_stat_add() -> only one task can reach here in 1 jiffies. Hence in order to account io_ticks precisely, we only need to know if there are IO inflight at most once in one jiffies. Noted that for rq-based device, iterating tags should not be used here because 'tags->lock' is grabbed in blk_mq_find_and_get_req(), hence part_stat_lock_inc/dec() and part_in_flight() is used to trace inflight. The additional overhead is quite little: - per cpu add/dec for each IO for rq-based device; - per cpu sum for each jiffies; And it's verified by null-blk that there are no performance degration under heavy IO pressure. Fixes: 5b18b5a73760 ("block: delete part_round_stats and switch to less precise counting") Signed-off-by: Yu Kuai <yukuai3@huawei.com> Link: https://lore.kernel.org/r/20240509123717.3223892-2-yukuai1@huaweicloud.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2024-05-09 12:37:16 +00:00
if (unlikely(time_after(now, stamp)) &&
likely(try_cmpxchg(&part->bd_stamp, &stamp, now)) &&
(end || part_in_flight(part)))
__part_stat_add(part, io_ticks, now - stamp);
if (bdev_is_partition(part)) {
part = bdev_whole(part);
goto again;
}
}
unsigned long bdev_start_io_acct(struct block_device *bdev, enum req_op op,
unsigned long start_time)
{
part_stat_lock();
update_io_ticks(bdev, start_time, false);
part_stat_local_inc(bdev, in_flight[op_is_write(op)]);
part_stat_unlock();
blk-mq: new multi-queue block IO queueing mechanism Linux currently has two models for block devices: - The classic request_fn based approach, where drivers use struct request units for IO. The block layer provides various helper functionalities to let drivers share code, things like tag management, timeout handling, queueing, etc. - The "stacked" approach, where a driver squeezes in between the block layer and IO submitter. Since this bypasses the IO stack, driver generally have to manage everything themselves. With drivers being written for new high IOPS devices, the classic request_fn based driver doesn't work well enough. The design dates back to when both SMP and high IOPS was rare. It has problems with scaling to bigger machines, and runs into scaling issues even on smaller machines when you have IOPS in the hundreds of thousands per device. The stacked approach is then most often selected as the model for the driver. But this means that everybody has to re-invent everything, and along with that we get all the problems again that the shared approach solved. This commit introduces blk-mq, block multi queue support. The design is centered around per-cpu queues for queueing IO, which then funnel down into x number of hardware submission queues. We might have a 1:1 mapping between the two, or it might be an N:M mapping. That all depends on what the hardware supports. blk-mq provides various helper functions, which include: - Scalable support for request tagging. Most devices need to be able to uniquely identify a request both in the driver and to the hardware. The tagging uses per-cpu caches for freed tags, to enable cache hot reuse. - Timeout handling without tracking request on a per-device basis. Basically the driver should be able to get a notification, if a request happens to fail. - Optional support for non 1:1 mappings between issue and submission queues. blk-mq can redirect IO completions to the desired location. - Support for per-request payloads. Drivers almost always need to associate a request structure with some driver private command structure. Drivers can tell blk-mq this at init time, and then any request handed to the driver will have the required size of memory associated with it. - Support for merging of IO, and plugging. The stacked model gets neither of these. Even for high IOPS devices, merging sequential IO reduces per-command overhead and thus increases bandwidth. For now, this is provided as a potential 3rd queueing model, with the hope being that, as it matures, it can replace both the classic and stacked model. That would get us back to having just 1 real model for block devices, leaving the stacked approach to dm/md devices (as it was originally intended). Contributions in this patch from the following people: Shaohua Li <shli@fusionio.com> Alexander Gordeev <agordeev@redhat.com> Christoph Hellwig <hch@infradead.org> Mike Christie <michaelc@cs.wisc.edu> Matias Bjorling <m@bjorling.me> Jeff Moyer <jmoyer@redhat.com> Acked-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2013-10-24 08:20:05 +00:00
return start_time;
}
EXPORT_SYMBOL(bdev_start_io_acct);
/**
* bio_start_io_acct - start I/O accounting for bio based drivers
* @bio: bio to start account for
*
* Returns the start time that should be passed back to bio_end_io_acct().
*/
unsigned long bio_start_io_acct(struct bio *bio)
{
return bdev_start_io_acct(bio->bi_bdev, bio_op(bio), jiffies);
}
EXPORT_SYMBOL_GPL(bio_start_io_acct);
void bdev_end_io_acct(struct block_device *bdev, enum req_op op,
unsigned int sectors, unsigned long start_time)
{
const int sgrp = op_stat_group(op);
unsigned long now = READ_ONCE(jiffies);
unsigned long duration = now - start_time;
part_stat_lock();
update_io_ticks(bdev, now, true);
part_stat_inc(bdev, ios[sgrp]);
part_stat_add(bdev, sectors[sgrp], sectors);
part_stat_add(bdev, nsecs[sgrp], jiffies_to_nsecs(duration));
part_stat_local_dec(bdev, in_flight[op_is_write(op)]);
blk-mq: new multi-queue block IO queueing mechanism Linux currently has two models for block devices: - The classic request_fn based approach, where drivers use struct request units for IO. The block layer provides various helper functionalities to let drivers share code, things like tag management, timeout handling, queueing, etc. - The "stacked" approach, where a driver squeezes in between the block layer and IO submitter. Since this bypasses the IO stack, driver generally have to manage everything themselves. With drivers being written for new high IOPS devices, the classic request_fn based driver doesn't work well enough. The design dates back to when both SMP and high IOPS was rare. It has problems with scaling to bigger machines, and runs into scaling issues even on smaller machines when you have IOPS in the hundreds of thousands per device. The stacked approach is then most often selected as the model for the driver. But this means that everybody has to re-invent everything, and along with that we get all the problems again that the shared approach solved. This commit introduces blk-mq, block multi queue support. The design is centered around per-cpu queues for queueing IO, which then funnel down into x number of hardware submission queues. We might have a 1:1 mapping between the two, or it might be an N:M mapping. That all depends on what the hardware supports. blk-mq provides various helper functions, which include: - Scalable support for request tagging. Most devices need to be able to uniquely identify a request both in the driver and to the hardware. The tagging uses per-cpu caches for freed tags, to enable cache hot reuse. - Timeout handling without tracking request on a per-device basis. Basically the driver should be able to get a notification, if a request happens to fail. - Optional support for non 1:1 mappings between issue and submission queues. blk-mq can redirect IO completions to the desired location. - Support for per-request payloads. Drivers almost always need to associate a request structure with some driver private command structure. Drivers can tell blk-mq this at init time, and then any request handed to the driver will have the required size of memory associated with it. - Support for merging of IO, and plugging. The stacked model gets neither of these. Even for high IOPS devices, merging sequential IO reduces per-command overhead and thus increases bandwidth. For now, this is provided as a potential 3rd queueing model, with the hope being that, as it matures, it can replace both the classic and stacked model. That would get us back to having just 1 real model for block devices, leaving the stacked approach to dm/md devices (as it was originally intended). Contributions in this patch from the following people: Shaohua Li <shli@fusionio.com> Alexander Gordeev <agordeev@redhat.com> Christoph Hellwig <hch@infradead.org> Mike Christie <michaelc@cs.wisc.edu> Matias Bjorling <m@bjorling.me> Jeff Moyer <jmoyer@redhat.com> Acked-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2013-10-24 08:20:05 +00:00
part_stat_unlock();
}
EXPORT_SYMBOL(bdev_end_io_acct);
void bio_end_io_acct_remapped(struct bio *bio, unsigned long start_time,
struct block_device *orig_bdev)
{
bdev_end_io_acct(orig_bdev, bio_op(bio), bio_sectors(bio), start_time);
}
EXPORT_SYMBOL_GPL(bio_end_io_acct_remapped);
block: add lld busy state exporting interface This patch adds an new interface, blk_lld_busy(), to check lld's busy state from the block layer. blk_lld_busy() calls down into low-level drivers for the checking if the drivers set q->lld_busy_fn() using blk_queue_lld_busy(). This resolves a performance problem on request stacking devices below. Some drivers like scsi mid layer stop dispatching request when they detect busy state on its low-level device like host/target/device. It allows other requests to stay in the I/O scheduler's queue for a chance of merging. Request stacking drivers like request-based dm should follow the same logic. However, there is no generic interface for the stacked device to check if the underlying device(s) are busy. If the request stacking driver dispatches and submits requests to the busy underlying device, the requests will stay in the underlying device's queue without a chance of merging. This causes performance problem on burst I/O load. With this patch, busy state of the underlying device is exported via q->lld_busy_fn(). So the request stacking driver can check it and stop dispatching requests if busy. The underlying device driver must return the busy state appropriately: 1: when the device driver can't process requests immediately. 0: when the device driver can process requests immediately, including abnormal situations where the device driver needs to kill all requests. Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com> Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2008-10-01 14:12:15 +00:00
/**
* blk_lld_busy - Check if underlying low-level drivers of a device are busy
* @q : the queue of the device being checked
*
* Description:
* Check if underlying low-level drivers of a device are busy.
* If the drivers want to export their busy state, they must set own
* exporting function using blk_queue_lld_busy() first.
*
* Basically, this function is used only by request stacking drivers
* to stop dispatching requests to underlying devices when underlying
* devices are busy. This behavior helps more I/O merging on the queue
* of the request stacking driver and prevents I/O throughput regression
* on burst I/O load.
*
* Return:
* 0 - Not busy (The request stacking driver should dispatch request)
* 1 - Busy (The request stacking driver should stop dispatching request)
*/
int blk_lld_busy(struct request_queue *q)
{
if (queue_is_mq(q) && q->mq_ops->busy)
return q->mq_ops->busy(q);
block: add lld busy state exporting interface This patch adds an new interface, blk_lld_busy(), to check lld's busy state from the block layer. blk_lld_busy() calls down into low-level drivers for the checking if the drivers set q->lld_busy_fn() using blk_queue_lld_busy(). This resolves a performance problem on request stacking devices below. Some drivers like scsi mid layer stop dispatching request when they detect busy state on its low-level device like host/target/device. It allows other requests to stay in the I/O scheduler's queue for a chance of merging. Request stacking drivers like request-based dm should follow the same logic. However, there is no generic interface for the stacked device to check if the underlying device(s) are busy. If the request stacking driver dispatches and submits requests to the busy underlying device, the requests will stay in the underlying device's queue without a chance of merging. This causes performance problem on burst I/O load. With this patch, busy state of the underlying device is exported via q->lld_busy_fn(). So the request stacking driver can check it and stop dispatching requests if busy. The underlying device driver must return the busy state appropriately: 1: when the device driver can't process requests immediately. 0: when the device driver can process requests immediately, including abnormal situations where the device driver needs to kill all requests. Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com> Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2008-10-01 14:12:15 +00:00
return 0;
}
EXPORT_SYMBOL_GPL(blk_lld_busy);
int kblockd_schedule_work(struct work_struct *work)
{
return queue_work(kblockd_workqueue, work);
}
EXPORT_SYMBOL(kblockd_schedule_work);
int kblockd_mod_delayed_work_on(int cpu, struct delayed_work *dwork,
unsigned long delay)
{
return mod_delayed_work_on(cpu, kblockd_workqueue, dwork, delay);
}
EXPORT_SYMBOL(kblockd_mod_delayed_work_on);
void blk_start_plug_nr_ios(struct blk_plug *plug, unsigned short nr_ios)
{
struct task_struct *tsk = current;
/*
* If this is a nested plug, don't actually assign it.
*/
if (tsk->plug)
return;
block: cache current nsec time in struct blk_plug Querying the current time is the most costly thing we do in the block layer per IO, and depending on kernel config settings, we may do it many times per IO. None of the callers actually need nsec granularity. Take advantage of that by caching the current time in the plug, with the assumption here being that any time checking will be temporally close enough that the slight loss of precision doesn't matter. If the block plug gets flushed, eg on preempt or schedule out, then we invalidate the cached clock. On a basic peak IOPS test case with iostats enabled, this changes the performance from: IOPS=108.41M, BW=52.93GiB/s, IOS/call=31/31 IOPS=108.43M, BW=52.94GiB/s, IOS/call=32/32 IOPS=108.29M, BW=52.88GiB/s, IOS/call=31/32 IOPS=108.35M, BW=52.91GiB/s, IOS/call=32/32 IOPS=108.42M, BW=52.94GiB/s, IOS/call=31/31 IOPS=108.40M, BW=52.93GiB/s, IOS/call=32/32 IOPS=108.31M, BW=52.89GiB/s, IOS/call=32/31 to IOPS=118.79M, BW=58.00GiB/s, IOS/call=31/32 IOPS=118.62M, BW=57.92GiB/s, IOS/call=31/31 IOPS=118.80M, BW=58.01GiB/s, IOS/call=32/31 IOPS=118.78M, BW=58.00GiB/s, IOS/call=32/32 IOPS=118.69M, BW=57.95GiB/s, IOS/call=32/31 IOPS=118.62M, BW=57.92GiB/s, IOS/call=32/31 IOPS=118.63M, BW=57.92GiB/s, IOS/call=31/32 which is more than a 9% improvement in performance. Looking at perf diff, we can see a huge reduction in time overhead: 10.55% -9.88% [kernel.vmlinux] [k] read_tsc 1.31% -1.22% [kernel.vmlinux] [k] ktime_get Note that since this relies on blk_plug for the caching, it's only applicable to the issue side. But this is where most of the time calls happen anyway. On the completion side, cached time stamping is done with struct io_comp patch, as long as the driver supports it. It's also worth noting that the above testing doesn't enable any of the higher cost CPU items on the block layer side, like wbt, cgroups, iocost, etc, which all would add additional time querying and hence overhead. IOW, results would likely look even better in comparison with those enabled, as distros would do. Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2024-01-15 21:46:03 +00:00
plug->cur_ktime = 0;
plug->mq_list = NULL;
plug->cached_rq = NULL;
plug->nr_ios = min_t(unsigned short, nr_ios, BLK_MAX_REQUEST_COUNT);
plug->rq_count = 0;
plug->multiple_queues = false;
plug->has_elevator = false;
INIT_LIST_HEAD(&plug->cb_list);
/*
* Store ordering should not be needed here, since a potential
* preempt will imply a full memory barrier
*/
tsk->plug = plug;
}
/**
* blk_start_plug - initialize blk_plug and track it inside the task_struct
* @plug: The &struct blk_plug that needs to be initialized
*
* Description:
* blk_start_plug() indicates to the block layer an intent by the caller
* to submit multiple I/O requests in a batch. The block layer may use
* this hint to defer submitting I/Os from the caller until blk_finish_plug()
* is called. However, the block layer may choose to submit requests
* before a call to blk_finish_plug() if the number of queued I/Os
* exceeds %BLK_MAX_REQUEST_COUNT, or if the size of the I/O is larger than
* %BLK_PLUG_FLUSH_SIZE. The queued I/Os may also be submitted early if
* the task schedules (see below).
*
* Tracking blk_plug inside the task_struct will help with auto-flushing the
* pending I/O should the task end up blocking between blk_start_plug() and
* blk_finish_plug(). This is important from a performance perspective, but
* also ensures that we don't deadlock. For instance, if the task is blocking
* for a memory allocation, memory reclaim could end up wanting to free a
* page belonging to that request that is currently residing in our private
* plug. By flushing the pending I/O when the process goes to sleep, we avoid
* this kind of deadlock.
*/
void blk_start_plug(struct blk_plug *plug)
{
blk_start_plug_nr_ios(plug, 1);
}
EXPORT_SYMBOL(blk_start_plug);
static void flush_plug_callbacks(struct blk_plug *plug, bool from_schedule)
{
LIST_HEAD(callbacks);
while (!list_empty(&plug->cb_list)) {
list_splice_init(&plug->cb_list, &callbacks);
while (!list_empty(&callbacks)) {
struct blk_plug_cb *cb = list_first_entry(&callbacks,
struct blk_plug_cb,
list);
list_del(&cb->list);
cb->callback(cb, from_schedule);
}
}
}
struct blk_plug_cb *blk_check_plugged(blk_plug_cb_fn unplug, void *data,
int size)
{
struct blk_plug *plug = current->plug;
struct blk_plug_cb *cb;
if (!plug)
return NULL;
list_for_each_entry(cb, &plug->cb_list, list)
if (cb->callback == unplug && cb->data == data)
return cb;
/* Not currently on the callback list */
BUG_ON(size < sizeof(*cb));
cb = kzalloc(size, GFP_ATOMIC);
if (cb) {
cb->data = data;
cb->callback = unplug;
list_add(&cb->list, &plug->cb_list);
}
return cb;
}
EXPORT_SYMBOL(blk_check_plugged);
void __blk_flush_plug(struct blk_plug *plug, bool from_schedule)
{
if (!list_empty(&plug->cb_list))
flush_plug_callbacks(plug, from_schedule);
blk_mq_flush_plug_list(plug, from_schedule);
/*
* Unconditionally flush out cached requests, even if the unplug
* event came from schedule. Since we know hold references to the
* queue for cached requests, we don't want a blocked task holding
* up a queue freeze/quiesce event.
*/
if (unlikely(!rq_list_empty(plug->cached_rq)))
blk_mq_free_plug_rqs(plug);
block: fix that blk_time_get_ns() doesn't update time after schedule While monitoring the throttle time of IO from iocost, it's found that such time is always zero after the io_schedule() from ioc_rqos_throttle, for example, with the following debug patch: + printk("%s-%d: %s enter %llu\n", current->comm, current->pid, __func__, blk_time_get_ns()); while (true) { set_current_state(TASK_UNINTERRUPTIBLE); if (wait.committed) break; io_schedule(); } + printk("%s-%d: %s exit %llu\n", current->comm, current->pid, __func__, blk_time_get_ns()); It can be observerd that blk_time_get_ns() always return the same time: [ 1068.096579] fio-1268: ioc_rqos_throttle enter 1067901962288 [ 1068.272587] fio-1268: ioc_rqos_throttle exit 1067901962288 [ 1068.274389] fio-1268: ioc_rqos_throttle enter 1067901962288 [ 1068.472690] fio-1268: ioc_rqos_throttle exit 1067901962288 [ 1068.474485] fio-1268: ioc_rqos_throttle enter 1067901962288 [ 1068.672656] fio-1268: ioc_rqos_throttle exit 1067901962288 [ 1068.674451] fio-1268: ioc_rqos_throttle enter 1067901962288 [ 1068.872655] fio-1268: ioc_rqos_throttle exit 1067901962288 And I think the root cause is that 'PF_BLOCK_TS' is always cleared by blk_flush_plug() before scheduel(), hence blk_plug_invalidate_ts() will never be called: blk_time_get_ns plug->cur_ktime = ktime_get_ns(); current->flags |= PF_BLOCK_TS; io_schedule: io_schedule_prepare blk_flush_plug __blk_flush_plug /* the flag is cleared, while time is not */ current->flags &= ~PF_BLOCK_TS; schedule sched_update_worker /* the flag is not set, hence plug->cur_ktime is not cleared */ if (tsk->flags & PF_BLOCK_TS) blk_plug_invalidate_ts() blk_time_get_ns /* got the time stashed before schedule */ return plug->cur_ktime; Fix the problem by clearing cached time in __blk_flush_plug(). Fixes: 06b23f92af87 ("block: update cached timestamp post schedule/preemption") Signed-off-by: Yu Kuai <yukuai3@huawei.com> Link: https://lore.kernel.org/r/20240411032349.3051233-2-yukuai1@huaweicloud.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
2024-04-11 03:23:48 +00:00
plug->cur_ktime = 0;
current->flags &= ~PF_BLOCK_TS;
}
/**
* blk_finish_plug - mark the end of a batch of submitted I/O
* @plug: The &struct blk_plug passed to blk_start_plug()
*
* Description:
* Indicate that a batch of I/O submissions is complete. This function
* must be paired with an initial call to blk_start_plug(). The intent
* is to allow the block layer to optimize I/O submission. See the
* documentation for blk_start_plug() for more information.
*/
void blk_finish_plug(struct blk_plug *plug)
{
if (plug == current->plug) {
__blk_flush_plug(plug, false);
current->plug = NULL;
}
}
EXPORT_SYMBOL(blk_finish_plug);
void blk_io_schedule(void)
{
/* Prevent hang_check timer from firing at us during very long I/O */
unsigned long timeout = sysctl_hung_task_timeout_secs * HZ / 2;
if (timeout)
io_schedule_timeout(timeout);
else
io_schedule();
}
EXPORT_SYMBOL_GPL(blk_io_schedule);
int __init blk_dev_init(void)
{
BUILD_BUG_ON((__force u32)REQ_OP_LAST >= (1 << REQ_OP_BITS));
BUILD_BUG_ON(REQ_OP_BITS + REQ_FLAG_BITS > 8 *
sizeof_field(struct request, cmd_flags));
BUILD_BUG_ON(REQ_OP_BITS + REQ_FLAG_BITS > 8 *
sizeof_field(struct bio, bi_opf));
/* used for unplugging and affects IO latency/throughput - HIGHPRI */
kblockd_workqueue = alloc_workqueue("kblockd",
WQ_MEM_RECLAIM | WQ_HIGHPRI, 0);
if (!kblockd_workqueue)
panic("Failed to create kblockd\n");
blk_requestq_cachep = KMEM_CACHE(request_queue, SLAB_PANIC);
blk_debugfs_root = debugfs_create_dir("block", NULL);
return 0;
}