block: free sched's request pool in blk_cleanup_queue
In theory, IO scheduler belongs to request queue, and the request pool
of sched tags belongs to the request queue too.
However, the current tags allocation interfaces are re-used for both
driver tags and sched tags, and driver tags is definitely host wide,
and doesn't belong to any request queue, same with its request pool.
So we need tagset instance for freeing request of sched tags.
Meantime, blk_mq_free_tag_set() often follows blk_cleanup_queue() in case
of non-BLK_MQ_F_TAG_SHARED, this way requires that request pool of sched
tags to be freed before calling blk_mq_free_tag_set().
Commit 47cdee29ef9d94e ("block: move blk_exit_queue into __blk_release_queue")
moves blk_exit_queue into __blk_release_queue for simplying the fast
path in generic_make_request(), then causes oops during freeing requests
of sched tags in __blk_release_queue().
Fix the above issue by move freeing request pool of sched tags into
blk_cleanup_queue(), this way is safe becasue queue has been frozen and no any
in-queue requests at that time. Freeing sched tags has to be kept in queue's
release handler becasue there might be un-completed dispatch activity
which might refer to sched tags.
BUG=b/189007274
TEST=presubmit
SOURCE=UPSTREAM(c3e2219216c92919a6bd1711f340f5faa98695e6)
RELEASE_NOTE=Fix CVE-2019-25044 in kernel
cos-patch: security-high
Cc: Bart Van Assche <bvanassche@acm.org>
Cc: Christoph Hellwig <hch@lst.de>
Fixes: 47cdee29ef9d94e485eb08f962c74943023a5271 ("block: move blk_exit_queue into __blk_release_queue")
Tested-by: Yi Zhang <yi.zhang@redhat.com>
Reported-by: kernel test robot <rong.a.chen@intel.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Change-Id: Ibfc68beff83631126dcddf872f5fd0a44f621958
Reviewed-on: https://cos-review.googlesource.com/c/third_party/kernel/+/17173
Main-Branch-Verified: Cusky Presubmit Bot <presubmit@cos-infra-prod.iam.gserviceaccount.com>
Tested-by: Cusky Presubmit Bot <presubmit@cos-infra-prod.iam.gserviceaccount.com>
Reviewed-by: Oleksandr Tymoshenko <ovt@google.com>
Reviewed-by: Vaibhav Rustagi <vaibhavrustagi@google.com>
diff --git a/block/blk-core.c b/block/blk-core.c
index 80f3e72..0c086fa 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -726,7 +726,7 @@
*/
if (q->elevator) {
ioc_clear_queue(q);
- elevator_exit(q, q->elevator);
+ __elevator_exit(q, q->elevator);
q->elevator = NULL;
}
@@ -821,6 +821,19 @@
if (q->mq_ops)
blk_mq_exit_queue(q);
+ /*
+ * In theory, request pool of sched_tags belongs to request queue.
+ * However, the current implementation requires tag_set for freeing
+ * requests, so free the pool now.
+ *
+ * Queue has become frozen, there can't be any in-queue requests, so
+ * it is safe to free requests now.
+ */
+ mutex_lock(&q->sysfs_lock);
+ if (q->elevator)
+ blk_mq_sched_free_requests(q);
+ mutex_unlock(&q->sysfs_lock);
+
percpu_ref_exit(&q->q_usage_counter);
spin_lock_irq(lock);
diff --git a/block/blk-mq-sched.c b/block/blk-mq-sched.c
index d89a757..eca0389 100644
--- a/block/blk-mq-sched.c
+++ b/block/blk-mq-sched.c
@@ -462,14 +462,18 @@
return ret;
}
+/* called in queue's release handler, tagset has gone away */
static void blk_mq_sched_tags_teardown(struct request_queue *q)
{
- struct blk_mq_tag_set *set = q->tag_set;
struct blk_mq_hw_ctx *hctx;
int i;
- queue_for_each_hw_ctx(q, hctx, i)
- blk_mq_sched_free_tags(set, hctx, i);
+ queue_for_each_hw_ctx(q, hctx, i) {
+ if (hctx->sched_tags) {
+ blk_mq_free_rq_map(hctx->sched_tags);
+ hctx->sched_tags = NULL;
+ }
+ }
}
int blk_mq_init_sched(struct request_queue *q, struct elevator_type *e)
@@ -510,6 +514,7 @@
ret = e->ops.mq.init_hctx(hctx, i);
if (ret) {
eq = q->elevator;
+ blk_mq_sched_free_requests(q);
blk_mq_exit_sched(q, eq);
kobject_put(&eq->kobj);
return ret;
@@ -521,11 +526,30 @@
return 0;
err:
+ blk_mq_sched_free_requests(q);
blk_mq_sched_tags_teardown(q);
q->elevator = NULL;
return ret;
}
+/*
+ * called in either blk_queue_cleanup or elevator_switch, tagset
+ * is required for freeing requests
+ */
+void blk_mq_sched_free_requests(struct request_queue *q)
+{
+ struct blk_mq_hw_ctx *hctx;
+ int i;
+
+ lockdep_assert_held(&q->sysfs_lock);
+ WARN_ON(!q->elevator);
+
+ queue_for_each_hw_ctx(q, hctx, i) {
+ if (hctx->sched_tags)
+ blk_mq_free_rqs(q->tag_set, hctx->sched_tags, i);
+ }
+}
+
void blk_mq_exit_sched(struct request_queue *q, struct elevator_queue *e)
{
struct blk_mq_hw_ctx *hctx;
diff --git a/block/blk-mq-sched.h b/block/blk-mq-sched.h
index fe66076..d97e13f 100644
--- a/block/blk-mq-sched.h
+++ b/block/blk-mq-sched.h
@@ -28,6 +28,7 @@
int blk_mq_init_sched(struct request_queue *q, struct elevator_type *e);
void blk_mq_exit_sched(struct request_queue *q, struct elevator_queue *e);
+void blk_mq_sched_free_requests(struct request_queue *q);
static inline bool
blk_mq_sched_bio_merge(struct request_queue *q, struct bio *bio)
diff --git a/block/blk.h b/block/blk.h
index ae87e2a..6203fbe 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -5,6 +5,7 @@
#include <linux/idr.h>
#include <linux/blk-mq.h>
#include "blk-mq.h"
+#include "blk-mq-sched.h"
/* Amount of time in which a process may batch requests */
#define BLK_BATCH_TIME (HZ/50UL)
@@ -243,10 +244,18 @@
int elevator_init_mq(struct request_queue *q);
int elevator_switch_mq(struct request_queue *q,
struct elevator_type *new_e);
-void elevator_exit(struct request_queue *, struct elevator_queue *);
+
+void __elevator_exit(struct request_queue *, struct elevator_queue *);
int elv_register_queue(struct request_queue *q, bool uevent);
void elv_unregister_queue(struct request_queue *q);
+static inline void elevator_exit(struct request_queue *q,
+ struct elevator_queue *e)
+{
+ blk_mq_sched_free_requests(q);
+ __elevator_exit(q, e);
+}
+
struct hd_struct *__disk_get_part(struct gendisk *disk, int partno);
#ifdef CONFIG_FAIL_IO_TIMEOUT
diff --git a/block/elevator.c b/block/elevator.c
index 5b51bc5..e7f595d 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -240,7 +240,7 @@
return err;
}
-void elevator_exit(struct request_queue *q, struct elevator_queue *e)
+void __elevator_exit(struct request_queue *q, struct elevator_queue *e)
{
mutex_lock(&e->sysfs_lock);
if (e->uses_mq && e->type->ops.mq.exit_sched)
