From: Yu Kuai yukuai3@huawei.com

mainline inclusion from md-next commit ddc489e066cd267b383c0eed4f576f6bdb154588 category: performance bugzilla: https://gitee.com/src-openeuler/kernel/issues/I5PRMO CVE: NA

Reference: https://git.kernel.org/pub/scm/linux/kernel/git/song/md.git/commit/?h=md-nex...

---------------------

Currently, wait_barrier() will hold 'resync_lock' to read 'conf->barrier', and io can't be dispatched until 'barrier' is dropped.

Since holding the 'barrier' is not common, convert 'resync_lock' to use seqlock so that holding lock can be avoided in fast path.

Signed-off-by: Yu Kuai yukuai3@huawei.com Reviewed-and-tested-by: Logan Gunthorpe logang@deltatee.com Signed-off-by: Song Liu song@kernel.org Reviewed-by: Jason Yan yanaijie@huawei.com Signed-off-by: Yongqiang Liu liuyongqiang13@huawei.com --- drivers/md/raid10.c | 88 ++++++++++++++++++++++++++++++--------------- drivers/md/raid10.h | 2 +- 2 files changed, 61 insertions(+), 29 deletions(-)

diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index 13e0b4a462fc..f8fea9593955 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -112,6 +112,21 @@ static void end_reshape(struct r10conf *conf);

#include "raid1-10.c"

+#define NULL_CMD +#define cmd_before(conf, cmd) \ + do { \ + write_sequnlock_irq(&(conf)->resync_lock); \ + cmd; \ + } while (0) +#define cmd_after(conf) write_seqlock_irq(&(conf)->resync_lock) + +#define wait_event_barrier_cmd(conf, cond, cmd) \ + wait_event_cmd((conf)->wait_barrier, cond, cmd_before(conf, cmd), \ + cmd_after(conf)) + +#define wait_event_barrier(conf, cond) \ + wait_event_barrier_cmd(conf, cond, NULL_CMD) + /* * for resync bio, r10bio pointer can be retrieved from the per-bio * 'struct resync_pages'. @@ -971,35 +986,54 @@ static void flush_pending_writes(struct r10conf *conf) static void raise_barrier(struct r10conf *conf, int force) { BUG_ON(force && !conf->barrier); - spin_lock_irq(&conf->resync_lock); + write_seqlock_irq(&conf->resync_lock);

/* Wait until no block IO is waiting (unless 'force') */ - wait_event_lock_irq(conf->wait_barrier, force || !conf->nr_waiting, - conf->resync_lock); + wait_event_barrier(conf, force || !conf->nr_waiting);

/* block any new IO from starting */ - conf->barrier++; + WRITE_ONCE(conf->barrier, conf->barrier + 1);

/* Now wait for all pending IO to complete */ - wait_event_lock_irq(conf->wait_barrier, - !atomic_read(&conf->nr_pending) && conf->barrier < RESYNC_DEPTH, - conf->resync_lock); + wait_event_barrier(conf, !atomic_read(&conf->nr_pending) && + conf->barrier < RESYNC_DEPTH);

- spin_unlock_irq(&conf->resync_lock); + write_sequnlock_irq(&conf->resync_lock); }

static void lower_barrier(struct r10conf *conf) { unsigned long flags; - spin_lock_irqsave(&conf->resync_lock, flags); - conf->barrier--; - spin_unlock_irqrestore(&conf->resync_lock, flags); + + write_seqlock_irqsave(&conf->resync_lock, flags); + WRITE_ONCE(conf->barrier, conf->barrier - 1); + write_sequnlock_irqrestore(&conf->resync_lock, flags); wake_up(&conf->wait_barrier); }

+static bool wait_barrier_nolock(struct r10conf *conf) +{ + unsigned int seq = read_seqbegin(&conf->resync_lock); + + if (READ_ONCE(conf->barrier)) + return false; + + atomic_inc(&conf->nr_pending); + if (!read_seqretry(&conf->resync_lock, seq)) + return true; + + if (atomic_dec_and_test(&conf->nr_pending)) + wake_up_barrier(conf); + + return false; +} + static void wait_barrier(struct r10conf *conf) { - spin_lock_irq(&conf->resync_lock); + if (wait_barrier_nolock(conf)) + return; + + write_seqlock_irq(&conf->resync_lock); if (conf->barrier) { conf->nr_waiting++; /* Wait for the barrier to drop. @@ -1012,19 +1046,18 @@ static void wait_barrier(struct r10conf *conf) * count down. */ raid10_log(conf->mddev, "wait barrier"); - wait_event_lock_irq(conf->wait_barrier, + wait_event_barrier(conf, !conf->barrier || (atomic_read(&conf->nr_pending) && current->bio_list && (!bio_list_empty(&current->bio_list[0]) || - !bio_list_empty(&current->bio_list[1]))), - conf->resync_lock); + !bio_list_empty(&current->bio_list[1])))); conf->nr_waiting--; if (!conf->nr_waiting) wake_up(&conf->wait_barrier); } atomic_inc(&conf->nr_pending); - spin_unlock_irq(&conf->resync_lock); + write_sequnlock_irq(&conf->resync_lock); }

static void allow_barrier(struct r10conf *conf) @@ -1048,27 +1081,26 @@ static void freeze_array(struct r10conf *conf, int extra) * must match the number of pending IOs (nr_pending) before * we continue. */ - spin_lock_irq(&conf->resync_lock); + write_seqlock_irq(&conf->resync_lock); conf->array_freeze_pending++; - conf->barrier++; + WRITE_ONCE(conf->barrier, conf->barrier + 1); conf->nr_waiting++; - wait_event_lock_irq_cmd(conf->wait_barrier, - atomic_read(&conf->nr_pending) == conf->nr_queued+extra, - conf->resync_lock, - flush_pending_writes(conf)); + wait_event_barrier_cmd(conf, + atomic_read(&conf->nr_pending) == conf->nr_queued+extra, + flush_pending_writes(conf));

conf->array_freeze_pending--; - spin_unlock_irq(&conf->resync_lock); + write_sequnlock_irq(&conf->resync_lock); }

static void unfreeze_array(struct r10conf *conf) { /* reverse the effect of the freeze */ - spin_lock_irq(&conf->resync_lock); - conf->barrier--; + write_seqlock_irq(&conf->resync_lock); + WRITE_ONCE(conf->barrier, conf->barrier - 1); conf->nr_waiting--; wake_up(&conf->wait_barrier); - spin_unlock_irq(&conf->resync_lock); + write_sequnlock_irq(&conf->resync_lock); }

static sector_t choose_data_offset(struct r10bio *r10_bio, @@ -3740,7 +3772,7 @@ static struct r10conf *setup_conf(struct mddev *mddev) INIT_LIST_HEAD(&conf->retry_list); INIT_LIST_HEAD(&conf->bio_end_io_list);

- spin_lock_init(&conf->resync_lock); + seqlock_init(&conf->resync_lock); init_waitqueue_head(&conf->wait_barrier); atomic_set(&conf->nr_pending, 0);

@@ -4080,7 +4112,7 @@ static void *raid10_takeover_raid0(struct mddev *mddev, sector_t size, int devs) rdev->new_raid_disk = rdev->raid_disk * 2; rdev->sectors = size; } - conf->barrier = 1; + WRITE_ONCE(conf->barrier, 1); }

return conf; diff --git a/drivers/md/raid10.h b/drivers/md/raid10.h index d3eaaf3eb1bc..e368a92f37fd 100644 --- a/drivers/md/raid10.h +++ b/drivers/md/raid10.h @@ -77,7 +77,7 @@ struct r10conf { struct bio_list pending_bio_list; int pending_count;

- spinlock_t resync_lock; + seqlock_t resync_lock; atomic_t nr_pending; int nr_waiting; int nr_queued;

From: Yu Kuai yukuai3@huawei.com

hulk inclusion category: performance bugzilla: https://gitee.com/openeuler/kernel/issues/I5QK5M CVE: NA

--------------------------------

In some architecture memory access latency is very bad across nodes compare to local node. For consequence, io performance is rather bad while users issue io from multiple nodes if lock contention exist in the driver.

This patch make io dispatch asynchronously to specific kthread that is bind to cpus that are belong to the same node, so that memory access across nodes in driver can be avoided.

Signed-off-by: Yu Kuai yukuai3@huawei.com Reviewed-by: Jason Yan yanaijie@huawei.com Signed-off-by: Yongqiang Liu liuyongqiang13@huawei.com --- block/blk-core.c | 188 ++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 187 insertions(+), 1 deletion(-)

diff --git a/block/blk-core.c b/block/blk-core.c index bffecc437fbc..bed4fbfbd337 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -85,6 +85,31 @@ struct kmem_cache *blk_requestq_cachep; */ static struct workqueue_struct *kblockd_workqueue;

+#define BIO_DISPATCH_MAX_LOOP 16 + +/* prevent false sharing */ +#define BIO_ASYNC_LIST_SHIFT 2 +#define BIO_ASYNC_LOCK_SHIFT 4 +#define bio_async_list(ctl, i) (&(ctl)->list[(i) << BIO_ASYNC_LIST_SHIFT]) +#define bio_async_lock(ctl, i) (&(ctl)->lock[(i) << BIO_ASYNC_LOCK_SHIFT]) + +struct bio_dispatch_async_ctl { + /* + * Vector size is nr_cpu_ids, list stores bio dispatched from other cpu, + * such bio will be dispatched asynchronously to the cpu this structure + * is serviced. + */ + struct bio_list *list; + /* list is protected by lock */ + spinlock_t *lock; + /* kthread to dispatch bio asynchronously */ + struct task_struct *thread; + /* thread will wait here if there are no bios in list */ + wait_queue_head_t wait; +}; + +static struct bio_dispatch_async_ctl __percpu **bio_dispatch_async_ctl; + static int blk_alloc_queue_dispatch_async(struct request_queue *q) { int cpu; @@ -106,6 +131,165 @@ void blk_free_queue_dispatch_async(struct request_queue *q) free_percpu(q->last_dispatch_cpu); }

+static int collect_bio(struct bio_dispatch_async_ctl *ctl, + struct bio_list *list) +{ + int count = 0; + int cpu; + struct bio *bio; + + for_each_possible_cpu(cpu) { + spin_lock_irq(bio_async_lock(ctl, cpu)); + while ((bio = bio_list_pop(bio_async_list(ctl, cpu)))) { + bio_list_add(list, bio); + count++; + } + spin_unlock_irq(bio_async_lock(ctl, cpu)); + } + + return count; +} + +static int bio_dispatch_work(void *data) +{ + int loop_count = 0; + int cpu = smp_processor_id(); + struct bio_dispatch_async_ctl *ctl = + *per_cpu_ptr(bio_dispatch_async_ctl, cpu); + + for (;; loop_count++) { + struct bio_list bio_list_on_stack; + struct blk_plug plug; + struct bio *bio; + int count; + + bio_list_init(&bio_list_on_stack); + count = collect_bio(ctl, &bio_list_on_stack); + + if (!count) { + DEFINE_WAIT(wait); + + for (;;) { + prepare_to_wait(&ctl->wait, &wait, + TASK_INTERRUPTIBLE); + count = collect_bio(ctl, &bio_list_on_stack); + if (count) + break; + schedule(); + loop_count = 0; + } + finish_wait(&ctl->wait, &wait); + + } + + blk_start_plug(&plug); + while ((bio = bio_list_pop(&bio_list_on_stack))) { + struct request_queue *q = bio->bi_disk->queue; + + q->make_request_fn(q, bio); + } + blk_finish_plug(&plug); + + /* prevent soft lockup */ + if (loop_count >= BIO_DISPATCH_MAX_LOOP) { + loop_count = 0; + cond_resched(); + } + } + + return 0; +} + +static int get_dispatch_cpu(struct request_queue *q, int cpu) +{ + int *last_dispatch_cpu = per_cpu_ptr(q->last_dispatch_cpu, cpu); + + cpu = cpumask_next(*last_dispatch_cpu, &q->dispatch_async_cpus); + if (cpu >= nr_cpu_ids) + cpu = cpumask_first(&q->dispatch_async_cpus); + + *last_dispatch_cpu = cpu; + + return cpu; +} + +static void blk_queue_make_request_async(struct bio *bio) +{ + struct request_queue *q = bio->bi_disk->queue; + int cpu = smp_processor_id(); + int dispatch_cpu = get_dispatch_cpu(q, cpu); + struct bio_dispatch_async_ctl *ctl = + *per_cpu_ptr(bio_dispatch_async_ctl, dispatch_cpu); + + spin_lock_irq(bio_async_lock(ctl, cpu)); + bio_list_add(bio_async_list(ctl, cpu), bio); + spin_unlock_irq(bio_async_lock(ctl, cpu)); + + if (wq_has_sleeper(&ctl->wait)) + wake_up(&ctl->wait); +} + +static blk_qc_t blk_queue_do_make_request(struct bio *bio) +{ + struct request_queue *q = bio->bi_disk->queue; + int cpu = smp_processor_id(); + + /* + * Don't dispatch bio asynchronously in following cases: + * + * 1) QUEUE_FLAG_DISPATCH_ASYNC is not set; + * 2) current cpu is the target cpu; + * 3) bio is flagged no wait; + * 4) TODO: return value of submit_bio() will be used in io polling. + */ + if (!test_bit(QUEUE_FLAG_DISPATCH_ASYNC, &q->queue_flags) || + cpumask_test_cpu(cpu, &q->dispatch_async_cpus) || + bio->bi_opf & REQ_NOWAIT) + return q->make_request_fn(q, bio); + + /* return value is not concerned */ + blk_queue_make_request_async(bio); + return BLK_QC_T_NONE; +} + +static void init_blk_queue_async_dispatch(void) +{ + int cpu; + + bio_dispatch_async_ctl = alloc_percpu(struct bio_dispatch_async_ctl *); + if (!bio_dispatch_async_ctl) + panic("Failed to alloc bio_dispatch_async_ctl\n"); + + for_each_possible_cpu(cpu) { + int i; + struct bio_dispatch_async_ctl *ctl = + kmalloc(sizeof(struct bio_dispatch_async_ctl), + GFP_KERNEL | __GFP_NOFAIL); + + *per_cpu_ptr(bio_dispatch_async_ctl, cpu) = ctl; + + ctl->thread = + kthread_create_on_cpu(bio_dispatch_work, NULL, cpu, + "bio_dispatch_work_%u"); + if (IS_ERR_OR_NULL(ctl->thread)) + panic("Failed to create bio dispatch thread\n"); + + ctl->list = kmalloc_array(nr_cpu_ids, + sizeof(struct bio_list) << BIO_ASYNC_LIST_SHIFT, + GFP_KERNEL | __GFP_NOFAIL); + ctl->lock = kmalloc_array(nr_cpu_ids, + sizeof(spinlock_t) << BIO_ASYNC_LOCK_SHIFT, + GFP_KERNEL | __GFP_NOFAIL); + for (i = 0; i < nr_cpu_ids; ++i) { + bio_list_init(bio_async_list(ctl, i)); + spin_lock_init(bio_async_lock(ctl, i)); + } + + wake_up_process(ctl->thread); + init_waitqueue_head(&ctl->wait); + } +} + /** * blk_queue_flag_set - atomically set a queue flag * @flag: flag to be set @@ -2514,7 +2698,7 @@ blk_qc_t generic_make_request(struct bio *bio) /* 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]); - ret = q->make_request_fn(q, bio); + ret = blk_queue_do_make_request(bio);

/* sort new bios into those for a lower level * and those for the same level @@ -4054,6 +4238,8 @@ int __init blk_dev_init(void) sizeof(struct request_queue_wrapper), 0, SLAB_PANIC, NULL);

+ init_blk_queue_async_dispatch(); + #ifdef CONFIG_DEBUG_FS blk_debugfs_root = debugfs_create_dir("block", NULL); #endif

From: Yu Kuai yukuai3@huawei.com

hulk inclusion category: performance bugzilla: 187597, https://gitee.com/openeuler/kernel/issues/I5QK5M CVE: NA

--------------------------------

Try to improve performance for raid when user issues io concurrently from multiple nodes.

Signed-off-by: Yu Kuai yukuai3@huawei.com Reviewed-by: Jason Yan yanaijie@huawei.com Signed-off-by: Yongqiang Liu liuyongqiang13@huawei.com --- drivers/md/md.c | 13 +++++++++++++ 1 file changed, 13 insertions(+)

diff --git a/drivers/md/md.c b/drivers/md/md.c index 9c94190769b8..ac2bdb4665b8 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -66,6 +66,7 @@ #include <linux/raid/md_u.h> #include <linux/slab.h> #include <linux/percpu-refcount.h> +#include <linux/arch_topology.h>

#include <trace/events/block.h> #include "md.h" @@ -5543,6 +5544,16 @@ static void md_safemode_timeout(struct timer_list *t)

static int start_dirty_degraded;

+#define MIN_DISPATCH_ASYNC_CPUS 16 +static void queue_init_dispatch_async_cpus(struct request_queue *q, int node) +{ + arch_get_preferred_sibling_cpumask(node, &q->dispatch_async_cpus); + if (cpumask_weight(&q->dispatch_async_cpus) >= MIN_DISPATCH_ASYNC_CPUS) + blk_queue_flag_set(QUEUE_FLAG_DISPATCH_ASYNC, q); + else + cpumask_setall(&q->dispatch_async_cpus); +} + int md_run(struct mddev *mddev) { int err; @@ -5786,6 +5797,8 @@ int md_run(struct mddev *mddev) if (mddev->sb_flags) md_update_sb(mddev, 0);

+ if (mddev->queue && pers->level == 10) + queue_init_dispatch_async_cpus(mddev->queue, 1); md_new_event(mddev); return 0;

From: Yu Kuai yukuai3@huawei.com

hulk inclusion category: performance bugzilla: 187597, https://gitee.com/openeuler/kernel/issues/I5QK5M CVE: NA

--------------------------------

If CONFIG_BLK_BIO_DISPATCH_ASYNC is enabled, and driver support QUEUE_FLAG_DISPATCH_ASYNC, bios will be dispatched asynchronously to specific CPUs to avoid across nodes memory access in driver.

Signed-off-by: Yu Kuai yukuai3@huawei.com Reviewed-by: Jason Yan yanaijie@huawei.com Signed-off-by: Yongqiang Liu liuyongqiang13@huawei.com --- block/Kconfig | 13 +++++++++++++ block/blk-core.c | 21 ++++++++++++++++++++- block/blk-sysfs.c | 4 ++++ block/blk.h | 8 ++++++++ include/linux/blkdev.h | 7 +++++++ 5 files changed, 52 insertions(+), 1 deletion(-)

diff --git a/block/Kconfig b/block/Kconfig index 1f2469a0123c..da71e56f8682 100644 --- a/block/Kconfig +++ b/block/Kconfig @@ -200,6 +200,19 @@ config BLK_SED_OPAL Enabling this option enables users to setup/unlock/lock Locking ranges for SED devices using the Opal protocol.

+config BLK_BIO_DISPATCH_ASYNC + bool "Dispatch bios asynchronously on specific cpus" + default n + depends on BLOCK=y + help + If there are multiple nodes, memory access across nodes is rather bad + compare to local node. And if some drivers are using internal spin + locks, io performance will be bad if bios are issued concurrently from + different nodes. This feature will dispatch bio asynchronously to the + specific CPUs to avoid across nodes memory access in driver, noted this + feature will require special care in the driver to work. If unsure, + say N here. + menu "Partition Types"

source "block/partitions/Kconfig" diff --git a/block/blk-core.c b/block/blk-core.c index 08e7a875b229..2ce4ad4619c1 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -35,7 +35,6 @@ #include <linux/blk-cgroup.h> #include <linux/debugfs.h> #include <linux/bpf.h> -#include <linux/arch_topology.h>

#define CREATE_TRACE_POINTS #include <trace/events/block.h> @@ -86,6 +85,9 @@ struct kmem_cache *blk_requestq_cachep; */ static struct workqueue_struct *kblockd_workqueue;

+#ifdef CONFIG_BLK_BIO_DISPATCH_ASYNC +#include <linux/arch_topology.h> + #define BIO_DISPATCH_MAX_LOOP 16 /* the minimum of cpus that dispatch async can be enabled */ #define MIN_DISPATCH_ASYNC_CPUS 16 @@ -309,6 +311,23 @@ void queue_init_dispatch_async_cpus(struct request_queue *q, int node) cpumask_setall(dispatch_async_cpus); } EXPORT_SYMBOL_GPL(queue_init_dispatch_async_cpus); +#else +static int blk_alloc_queue_dispatch_async(struct request_queue *q) +{ + return 0; +} + +static blk_qc_t blk_queue_do_make_request(struct bio *bio) +{ + struct request_queue *q = bio->bi_disk->queue; + + return q->make_request_fn(q, bio); +} + +static void init_blk_queue_async_dispatch(void) +{ +} +#endif

/** * blk_queue_flag_set - atomically set a queue flag diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index cbd7de22c463..5bdae7c8fa27 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -696,6 +696,7 @@ static struct queue_sysfs_entry queue_wb_lat_entry = { .store = queue_wb_lat_store, };

+#ifdef CONFIG_BLK_BIO_DISPATCH_ASYNC static ssize_t queue_dispatch_async_cpus_show(struct request_queue *q, char *page) { @@ -731,6 +732,7 @@ static struct queue_sysfs_entry queue_dispatch_async_entry = { .attr = {.name = "dispatch_async", .mode = 0444 }, .show = queue_show_dispatch_async, }; +#endif

#ifdef CONFIG_BLK_DEV_THROTTLING_LOW static struct queue_sysfs_entry throtl_sample_time_entry = { @@ -774,8 +776,10 @@ static struct attribute *default_attrs[] = { &queue_dax_entry.attr, &queue_wb_lat_entry.attr, &queue_poll_delay_entry.attr, +#ifdef CONFIG_BLK_BIO_DISPATCH_ASYNC &queue_dispatch_async_cpus_entry.attr, &queue_dispatch_async_entry.attr, +#endif #ifdef CONFIG_BLK_DEV_THROTTLING_LOW &throtl_sample_time_entry.attr, #endif diff --git a/block/blk.h b/block/blk.h index cb1c9057d788..985e20980aff 100644 --- a/block/blk.h +++ b/block/blk.h @@ -49,9 +49,11 @@ struct request_queue_wrapper { struct mutex mq_freeze_lock; int mq_freeze_depth;

+#ifdef CONFIG_BLK_BIO_DISPATCH_ASYNC /* used when QUEUE_FLAG_DISPATCH_ASYNC is set */ struct cpumask dispatch_async_cpus; int __percpu *last_dispatch_cpu; +#endif };

#define queue_to_wrapper(q) \ @@ -464,6 +466,12 @@ extern int blk_iolatency_init(struct request_queue *q); static inline int blk_iolatency_init(struct request_queue *q) { return 0; } #endif

+#ifdef CONFIG_BLK_BIO_DISPATCH_ASYNC extern void blk_free_queue_dispatch_async(struct request_queue *q); +#else +static inline void blk_free_queue_dispatch_async(struct request_queue *q) +{ +} +#endif

#endif /* BLK_INTERNAL_H */ diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 996e00575042..41d235ee579a 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -785,7 +785,14 @@ bool blk_queue_flag_test_and_clear(unsigned int flag, struct request_queue *q);

extern void blk_set_pm_only(struct request_queue *q); extern void blk_clear_pm_only(struct request_queue *q); +#ifdef CONFIG_BLK_BIO_DISPATCH_ASYNC extern void queue_init_dispatch_async_cpus(struct request_queue *q, int node); +#else +static inline void queue_init_dispatch_async_cpus(struct request_queue *q, + int node) +{ +} +#endif

static inline int queue_in_flight(struct request_queue *q) {

From: Luo Meng luomeng12@huawei.com

hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I5TY3L CVE: NA

--------------------------------

A crash as follows:

BUG: unable to handle page fault for address: 000000011241cec7 sd 5:0:0:1: [sdl] Synchronizing SCSI cache #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 0 P4D 0 Oops: 0000 [#1] SMP PTI CPU: 3 PID: 2465367 Comm: multipath Kdump: loaded Tainted: G W O 5.10.0-60.18.0.50.h478.eulerosv2r11.x86_64 #1 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.1-0-ga5cab58-20220525_182517-szxrtosci10000 04/01/2014 RIP: 0010:kernfs_new_node+0x22/0x60 Code: cc cc 66 0f 1f 44 00 00 0f 1f 44 00 00 41 54 41 89 cb 0f b7 ca 48 89 f2 53 48 8b 47 08 48 89 fb 48 89 de 48 85 c0 48 0f 44 c7 <48> 8b 78 50 41 51 45 89 c1 45 89 d8 e8 4d ee ff ff 5a 49 89 c4 48 RSP: 0018:ffffa178419539e8 EFLAGS: 00010206 RAX: 000000011241ce77 RBX: ffff9596828395a0 RCX: 000000000000a1ff RDX: ffff9595ada828b0 RSI: ffff9596828395a0 RDI: ffff9596828395a0 RBP: ffff95959a9a2a80 R08: 0000000000000000 R09: 0000000000000004 R10: ffff9595ca0bf930 R11: 0000000000000000 R12: ffff9595ada828b0 R13: ffff9596828395a0 R14: 0000000000000001 R15: ffff9595948c5c80 FS: 00007f64baa10200(0000) GS:ffff9596bad80000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 000000011241cec7 CR3: 000000011923e003 CR4: 0000000000170ee0 Call Trace: kernfs_create_link+0x31/0xa0 sysfs_do_create_link_sd+0x61/0xc0 bd_link_disk_holder+0x10a/0x180 dm_get_table_device+0x10b/0x1f0 [dm_mod] __dm_get_device+0x1e2/0x280 [dm_mod] ? kmem_cache_alloc_trace+0x2fb/0x410 parse_path+0xca/0x200 [dm_multipath] parse_priority_group+0x19d/0x1f0 [dm_multipath] multipath_ctr+0x27a/0x491 [dm_multipath] dm_table_add_target+0x177/0x360 [dm_mod] table_load+0x12b/0x380 [dm_mod] ctl_ioctl+0x199/0x290 [dm_mod] ? dev_suspend+0xd0/0xd0 [dm_mod] dm_ctl_ioctl+0xa/0x20 [dm_mod] __se_sys_ioctl+0x85/0xc0 do_syscall_64+0x33/0x40 entry_SYSCALL_64_after_hwframe+0x61/0xc6

This can be easy reproduce: Add delay before ret = add_symlink(bdev->bd_part->holder_dir...) in bd_link_disk_holder() dmsetup create xxx --tabel "0 1000 linear /dev/sda 0" echo 1 > /sys/block/sda/device/delete

Delete /dev/sda will release holder_dir, but add_symlink() will use holder_dir. Therefore UAF will occur in this case.

Fix this problem by adding reference count to holder_dir.

Signed-off-by: Luo Meng luomeng12@huawei.com Reviewed-by: Jason Yan yanaijie@huawei.com Signed-off-by: Yongqiang Liu liuyongqiang13@huawei.com --- fs/block_dev.c | 2 ++ 1 file changed, 2 insertions(+)

diff --git a/fs/block_dev.c b/fs/block_dev.c index 9868b21b8ef9..7e891e08d0ce 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -1650,6 +1650,7 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) } } bdev->bd_openers++; + kobject_get(bdev->bd_part->holder_dir); if (for_part) bdev->bd_part_count++; if (mode & FMODE_WRITE) @@ -1925,6 +1926,7 @@ static void __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part) if (for_part) bdev->bd_part_count--;

+ kobject_put(bdev->bd_part->holder_dir); if (!--bdev->bd_openers) { WARN_ON_ONCE(bdev->bd_holders); sync_blockdev(bdev);