[PATCH 0/4] block: fix partition use-after-free and optimization

[PATCH 0/4] block: fix partition use-after-free and optimization

[Ming Lei]

Hi,

The 1st patch fixes one use-after-free on cached last_lookup partition.

The other 3 patches optimizes partition uses in IO path.



Ming Lei (4):
  block: fix use-after-free on cached last_lookup partition
  block: only define 'nr_sects_seq' in hd_part for 32bit SMP
  block: re-organize fields of 'struct hd_part'
  block: don't hold part0's refcount in IO path

 block/blk-core.c        | 15 ++-------------
 block/genhd.c           |  7 +++++--
 block/partitions/core.c | 14 ++++++++++++--
 include/linux/genhd.h   | 25 ++++++++++++++++++-------
 4 files changed, 37 insertions(+), 24 deletions(-)

Cc: Yufen Yu <yuyufen@huawei.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Hou Tao <houtao1@huawei.com>
-- 
2.25.2

[PATCH 1/4] block: fix use-after-free on cached last_lookup partition

[Ming Lei]

delete_partition() clears the cached last_lookup partition. However the
.last_lookup cache may be overwritten by one IO path after it is cleared
from delete_partition(). Then another IO path may use the cached deleting
partition after hd_struct_free() is called, then use-after-free is triggered
on the cached partition.

Fixes the issue by the following approach:

1) always get the partition's refcount via hd_struct_try_get() before
setting .last_lookup

2) move clearing .last_lookup from delete_partition() to hd_struct_free()
which is the release handle of the partition's percpu-refcount, so that no
IO path can overwrite .last_lookup after it is cleared in hd_struct_free().

It is one candidate approach of Yufen's patch[1] which adds overhead
in fast path by indirect lookup which may introduce one extra cacheline
in IO path. Also this patch relies on percpu-refcount's protection, and
it is easier to understand and verify.

[1] https://lore.kernel.org/linux-block/20200109013551.GB9655@ming.t460p/T/#t

Reported-by: Yufen Yu <yuyufen@huawei.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Hou Tao <houtao1@huawei.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
---
 block/blk-core.c        | 12 ------------
 block/genhd.c           |  6 +++++-
 block/partitions/core.c | 12 +++++++++++-
 include/linux/genhd.h   |  1 +
 4 files changed, 17 insertions(+), 14 deletions(-)

diff --git a/block/blk-core.c b/block/blk-core.c
index ec50d7e6be21..826a8980997d 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -1363,18 +1363,6 @@ void blk_account_io_start(struct request *rq, bool new_io)
 		part_stat_inc(part, merges[rw]);
 	} else {
 		part = disk_map_sector_rcu(rq->rq_disk, blk_rq_pos(rq));
-		if (!hd_struct_try_get(part)) {
-			/*
-			 * The partition is already being removed,
-			 * the request will be accounted on the disk only
-			 *
-			 * We take a reference on disk->part0 although that
-			 * partition will never be deleted, so we can treat
-			 * it as any other partition.
-			 */
-			part = &rq->rq_disk->part0;
-			hd_struct_get(part);
-		}
 		part_inc_in_flight(rq->q, part, rw);
 		rq->part = part;
 	}
diff --git a/block/genhd.c b/block/genhd.c
index c05d509877fa..8b33f0e54356 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -359,17 +359,21 @@ struct hd_struct *disk_map_sector_rcu(struct gendisk *disk, sector_t sector)
 	ptbl = rcu_dereference(disk->part_tbl);
 
 	part = rcu_dereference(ptbl->last_lookup);
-	if (part && sector_in_part(part, sector))
+	if (part && sector_in_part(part, sector) && hd_struct_try_get(part))
 		return part;
 
 	for (i = 1; i < ptbl->len; i++) {
 		part = rcu_dereference(ptbl->part[i]);
 
 		if (part && sector_in_part(part, sector)) {
+			if (!hd_struct_try_get(part))
+				goto exit;
 			rcu_assign_pointer(ptbl->last_lookup, part);
 			return part;
 		}
 	}
+ exit:
+	hd_struct_get(&disk->part0);
 	return &disk->part0;
 }
 
diff --git a/block/partitions/core.c b/block/partitions/core.c
index c085bf85509b..b6cbc9b98426 100644
--- a/block/partitions/core.c
+++ b/block/partitions/core.c
@@ -288,6 +288,11 @@ static void hd_struct_free_work(struct work_struct *work)
 static void hd_struct_free(struct percpu_ref *ref)
 {
 	struct hd_struct *part = container_of(ref, struct hd_struct, ref);
+	struct disk_part_tbl *ptbl =
+		rcu_dereference_protected(part->disk->part_tbl, 1);
+
+	rcu_assign_pointer(ptbl->last_lookup, NULL);
+	put_device(disk_to_dev(part->disk));
 
 	INIT_RCU_WORK(&part->rcu_work, hd_struct_free_work);
 	queue_rcu_work(system_wq, &part->rcu_work);
@@ -309,8 +314,12 @@ void delete_partition(struct gendisk *disk, struct hd_struct *part)
 	struct disk_part_tbl *ptbl =
 		rcu_dereference_protected(disk->part_tbl, 1);
 
+	/*
+	 * make sure disk won't be gone because ->part_tbl is referenced in
+	 * this part's release handler
+	 */
+	get_device(disk_to_dev(disk));
 	rcu_assign_pointer(ptbl->part[part->partno], NULL);
-	rcu_assign_pointer(ptbl->last_lookup, NULL);
 	kobject_put(part->holder_dir);
 	device_del(part_to_dev(part));
 
@@ -393,6 +402,7 @@ static struct hd_struct *add_partition(struct gendisk *disk, int partno,
 	p->nr_sects = len;
 	p->partno = partno;
 	p->policy = get_disk_ro(disk);
+	p->disk = disk;
 
 	if (info) {
 		struct partition_meta_info *pinfo;
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index f9c226f9546a..c28d1d9bfa72 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -85,6 +85,7 @@ struct hd_struct {
 	struct disk_stats dkstats;
 #endif
 	struct percpu_ref ref;
+	struct gendisk *disk;
 	struct rcu_work rcu_work;
 };
 
-- 
2.25.2

[PATCH 2/4] block: only define 'nr_sects_seq' in hd_part for 32bit SMP

[Ming Lei]

The seqcount of 'nr_sects_seq' is only needed in case of 32bit SMP,
so define it just for 32bit SMP.

Cc: Yufen Yu <yuyufen@huawei.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Hou Tao <houtao1@huawei.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
---
 block/genhd.c           | 2 +-
 block/partitions/core.c | 2 +-
 include/linux/genhd.h   | 9 +++++++++
 3 files changed, 11 insertions(+), 2 deletions(-)

diff --git a/block/genhd.c b/block/genhd.c
index 8b33f0e54356..c0288b89a7ad 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -1724,7 +1724,7 @@ struct gendisk *__alloc_disk_node(int minors, int node_id)
 		 * TODO: Ideally set_capacity() and get_capacity() should be
 		 * converted to make use of bd_mutex and sequence counters.
 		 */
-		seqcount_init(&disk->part0.nr_sects_seq);
+		hd_sects_seq_init(&disk->part0);
 		if (hd_ref_init(&disk->part0)) {
 			hd_free_part(&disk->part0);
 			kfree(disk);
diff --git a/block/partitions/core.c b/block/partitions/core.c
index b6cbc9b98426..efccb1be4c5c 100644
--- a/block/partitions/core.c
+++ b/block/partitions/core.c
@@ -391,7 +391,7 @@ static struct hd_struct *add_partition(struct gendisk *disk, int partno,
 		goto out_free;
 	}
 
-	seqcount_init(&p->nr_sects_seq);
+	hd_sects_seq_init(p);
 	pdev = part_to_dev(p);
 
 	p->start_sect = start;
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index c28d1d9bfa72..2732120751e8 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -68,7 +68,9 @@ struct hd_struct {
 	 * can be non-atomic on 32bit machines with 64bit sector_t.
 	 */
 	sector_t nr_sects;
+#if BITS_PER_LONG==32 && defined(CONFIG_SMP)
 	seqcount_t nr_sects_seq;
+#endif
 	sector_t alignment_offset;
 	unsigned int discard_alignment;
 	struct device __dev;
@@ -275,6 +277,13 @@ static inline void disk_put_part(struct hd_struct *part)
 		put_device(part_to_dev(part));
 }
 
+static inline void hd_sects_seq_init(struct hd_struct *p)
+{
+#if BITS_PER_LONG==32 && defined(CONFIG_SMP)
+	seqcount_init(&p->nr_sects_seq);
+#endif
+}
+
 /*
  * Smarter partition iterator without context limits.
  */
-- 
2.25.2

[PATCH 3/4] block: re-organize fields of 'struct hd_part'

[Ming Lei]

Put all fields accessed in IO path together at the beginning
of the struct, so that all can be fetched in single cacheline.

Cc: Yufen Yu <yuyufen@huawei.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Hou Tao <houtao1@huawei.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
---
 include/linux/genhd.h | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index 2732120751e8..e0fab519ce37 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -71,6 +71,14 @@ struct hd_struct {
 #if BITS_PER_LONG==32 && defined(CONFIG_SMP)
 	seqcount_t nr_sects_seq;
 #endif
+	unsigned long stamp;
+#ifdef	CONFIG_SMP
+	struct disk_stats __percpu *dkstats;
+#else
+	struct disk_stats dkstats;
+#endif
+	struct percpu_ref ref;
+
 	sector_t alignment_offset;
 	unsigned int discard_alignment;
 	struct device __dev;
@@ -80,13 +88,6 @@ struct hd_struct {
 #ifdef CONFIG_FAIL_MAKE_REQUEST
 	int make_it_fail;
 #endif
-	unsigned long stamp;
-#ifdef	CONFIG_SMP
-	struct disk_stats __percpu *dkstats;
-#else
-	struct disk_stats dkstats;
-#endif
-	struct percpu_ref ref;
 	struct gendisk *disk;
 	struct rcu_work rcu_work;
 };
-- 
2.25.2

[PATCH 4/4] block: don't hold part0's refcount in IO path

[Ming Lei]

gendisk can't be gone when there is IO activity, so not hold
part0's refcount in IO path.

Cc: Yufen Yu <yuyufen@huawei.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Hou Tao <houtao1@huawei.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
---
 block/blk-core.c | 3 ++-
 block/genhd.c    | 1 -
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/block/blk-core.c b/block/blk-core.c
index 826a8980997d..56cc61354671 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -1343,7 +1343,8 @@ void blk_account_io_done(struct request *req, u64 now)
 		part_stat_add(part, nsecs[sgrp], now - req->start_time_ns);
 		part_dec_in_flight(req->q, part, rq_data_dir(req));
 
-		hd_struct_put(part);
+		if (part->partno)
+			hd_struct_put(part);
 		part_stat_unlock();
 	}
 }
diff --git a/block/genhd.c b/block/genhd.c
index c0288b89a7ad..af8641351510 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -373,7 +373,6 @@ struct hd_struct *disk_map_sector_rcu(struct gendisk *disk, sector_t sector)
 		}
 	}
  exit:
-	hd_struct_get(&disk->part0);
 	return &disk->part0;
 }
 
-- 
2.25.2

Re: [PATCH 1/4] block: fix use-after-free on cached last_lookup partition

[Christoph Hellwig]

On Thu, May 07, 2020 at 04:52:36PM +0800, Ming Lei wrote:
>  	for (i = 1; i < ptbl->len; i++) {
>  		part = rcu_dereference(ptbl->part[i]);
>  
>  		if (part && sector_in_part(part, sector)) {
> +			if (!hd_struct_try_get(part))
> +				goto exit;

I think this needs the comment that was in blk_account_io_start to
explain the logic.  Also no need for the goto, this can just be a break.

> -	rcu_assign_pointer(ptbl->last_lookup, NULL);
>  	kobject_put(part->holder_dir);
>  	device_del(part_to_dev(part));
>  
> @@ -393,6 +402,7 @@ static struct hd_struct *add_partition(struct gendisk *disk, int partno,
>  	p->nr_sects = len;
>  	p->partno = partno;
>  	p->policy = get_disk_ro(disk);
> +	p->disk = disk;

Do we really need this pointer?  Can't we use part_to_disk in the
free path for some reason?

Re: [PATCH 2/4] block: only define 'nr_sects_seq' in hd_part for 32bit SMP

[Christoph Hellwig]

On Thu, May 07, 2020 at 04:52:37PM +0800, Ming Lei wrote:
> The seqcount of 'nr_sects_seq' is only needed in case of 32bit SMP,
> so define it just for 32bit SMP.

Looks good,

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH 3/4] block: re-organize fields of 'struct hd_part'

[Christoph Hellwig]

On Thu, May 07, 2020 at 04:52:38PM +0800, Ming Lei wrote:
> Put all fields accessed in IO path together at the beginning
> of the struct, so that all can be fetched in single cacheline.

Looks good,

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH 4/4] block: don't hold part0's refcount in IO path

[Christoph Hellwig]

On Thu, May 07, 2020 at 04:52:39PM +0800, Ming Lei wrote:
> gendisk can't be gone when there is IO activity, so not hold
> part0's refcount in IO path.
> 
> Cc: Yufen Yu <yuyufen@huawei.com>
> Cc: Christoph Hellwig <hch@infradead.org>
> Cc: Hou Tao <houtao1@huawei.com>
> Signed-off-by: Ming Lei <ming.lei@redhat.com>
> ---
>  block/blk-core.c | 3 ++-
>  block/genhd.c    | 1 -
>  2 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/block/blk-core.c b/block/blk-core.c
> index 826a8980997d..56cc61354671 100644
> --- a/block/blk-core.c
> +++ b/block/blk-core.c
> @@ -1343,7 +1343,8 @@ void blk_account_io_done(struct request *req, u64 now)
>  		part_stat_add(part, nsecs[sgrp], now - req->start_time_ns);
>  		part_dec_in_flight(req->q, part, rq_data_dir(req));
>  
> -		hd_struct_put(part);
> +		if (part->partno)
> +			hd_struct_put(part);
>  		part_stat_unlock();

Doesn't blk_account_io_merge needs the check as well?

Maybe it should go into hd_struct_put and the other helpers to be
centralized?

Otherwise this looks fine to me.

Re: [PATCH 1/4] block: fix use-after-free on cached last_lookup partition

[Ming Lei]

On Thu, May 07, 2020 at 07:16:26AM -0700, Christoph Hellwig wrote:
> On Thu, May 07, 2020 at 04:52:36PM +0800, Ming Lei wrote:
> >  	for (i = 1; i < ptbl->len; i++) {
> >  		part = rcu_dereference(ptbl->part[i]);
> >  
> >  		if (part && sector_in_part(part, sector)) {
> > +			if (!hd_struct_try_get(part))
> > +				goto exit;
> 
> I think this needs the comment that was in blk_account_io_start to
> explain the logic.  Also no need for the goto, this can just be a break.

OK, not did it because I thought the logic is straightforward.

> 
> > -	rcu_assign_pointer(ptbl->last_lookup, NULL);
> >  	kobject_put(part->holder_dir);
> >  	device_del(part_to_dev(part));
> >  
> > @@ -393,6 +402,7 @@ static struct hd_struct *add_partition(struct gendisk *disk, int partno,
> >  	p->nr_sects = len;
> >  	p->partno = partno;
> >  	p->policy = get_disk_ro(disk);
> > +	p->disk = disk;
> 
> Do we really need this pointer?  Can't we use part_to_disk in the
> free path for some reason?

Looks I did miss this function, :-)

Thanks, 
Ming

Re: [PATCH 4/4] block: don't hold part0's refcount in IO path

[Ming Lei]

On Thu, May 07, 2020 at 07:20:50AM -0700, Christoph Hellwig wrote:
> On Thu, May 07, 2020 at 04:52:39PM +0800, Ming Lei wrote:
> > gendisk can't be gone when there is IO activity, so not hold
> > part0's refcount in IO path.
> > 
> > Cc: Yufen Yu <yuyufen@huawei.com>
> > Cc: Christoph Hellwig <hch@infradead.org>
> > Cc: Hou Tao <houtao1@huawei.com>
> > Signed-off-by: Ming Lei <ming.lei@redhat.com>
> > ---
> >  block/blk-core.c | 3 ++-
> >  block/genhd.c    | 1 -
> >  2 files changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/block/blk-core.c b/block/blk-core.c
> > index 826a8980997d..56cc61354671 100644
> > --- a/block/blk-core.c
> > +++ b/block/blk-core.c
> > @@ -1343,7 +1343,8 @@ void blk_account_io_done(struct request *req, u64 now)
> >  		part_stat_add(part, nsecs[sgrp], now - req->start_time_ns);
> >  		part_dec_in_flight(req->q, part, rq_data_dir(req));
> >  
> > -		hd_struct_put(part);
> > +		if (part->partno)
> > +			hd_struct_put(part);
> >  		part_stat_unlock();
> 
> Doesn't blk_account_io_merge needs the check as well?

You are right, blk_account_io_merge needs it too.

> 
> Maybe it should go into hd_struct_put and the other helpers to be
> centralized?

I think that is doable.

Thanks,
Ming

[PATCH 3/4] block: re-organize fields of 'struct hd_part'

[Ming Lei]

Put all fields accessed in IO path together at the beginning
of the struct, so that all can be fetched in single cacheline.

Cc: Yufen Yu <yuyufen@huawei.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Hou Tao <houtao1@huawei.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
---
 include/linux/genhd.h | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index 5f5718ce5e86..4c6a998a0c56 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -116,6 +116,14 @@ struct hd_struct {
 #if BITS_PER_LONG==32 && defined(CONFIG_SMP)
 	seqcount_t nr_sects_seq;
 #endif
+	unsigned long stamp;
+#ifdef	CONFIG_SMP
+	struct disk_stats __percpu *dkstats;
+#else
+	struct disk_stats dkstats;
+#endif
+	struct percpu_ref ref;
+
 	sector_t alignment_offset;
 	unsigned int discard_alignment;
 	struct device __dev;
@@ -125,13 +133,6 @@ struct hd_struct {
 #ifdef CONFIG_FAIL_MAKE_REQUEST
 	int make_it_fail;
 #endif
-	unsigned long stamp;
-#ifdef	CONFIG_SMP
-	struct disk_stats __percpu *dkstats;
-#else
-	struct disk_stats dkstats;
-#endif
-	struct percpu_ref ref;
 	struct gendisk *disk;
 	struct rcu_work rcu_work;
 };
-- 
2.20.1