Re: [PATCH 5.4 000/168] 5.4.25-stable review

From: Holger HoffstÃtte
Date: Tue Mar 10 2020 - 10:57:22 EST


On 3/10/20 3:51 PM, Holger HoffstÃtte wrote:
On 3/10/20 3:35 PM, Greg Kroah-Hartman wrote:
On Tue, Mar 10, 2020 at 03:02:37PM +0100, Holger HoffstÃtte wrote:
On 3/10/20 1:37 PM, Greg Kroah-Hartman wrote:
This is the start of the stable review cycle for the 5.4.25 release.

This fails to compile due to broken patch 001/168:
"block, bfq: get a ref to a group when adding it to a service tree":

..
block/bfq-wf2q.c: In function 'bfq_get_entity':
./include/linux/kernel.h:994:51: error: 'struct bfq_group' has no member named 'entity'
..

The calls to bfq_get_entity::bfqg_and_blkg_get and bfq_forget_entity::bfqg_and_blkg_put
in bfq-wf2q.c need to be wrapped in #ifdef CONFIG_BFQ_GROUP_IOSCHED, otherwise
the build will fail when CONFIG_BFQ_GROUP_IOSCHED is not enabled.
This horribly error-prone #ifdef mess was finally removed in upstream commit
4d8340d0d4d9. For 5.4 we'll either need that as well or add them back.

Ick, that's a mess.

I'll go drop that patch now, odd that it passed my build tests...

Uh, please no? It fixes a rather nasty UAF when cgroups are in use.
Please just add the other upstream commit as well, I confirmed it applies
cleanly and fixes the problem.


Alternatively I've appended the version originally sent to the mailing list,
with those #ifdefs intact. That's what I had in my tree so far, you could
consider it a 5.4 backport. Other than that there's no functional difference
to the upstream version.

hth,
Holger
From: Paolo Valente <paolo.valente@xxxxxxxxxx>
To: Jens Axboe <axboe@xxxxxxxxx>
Cc: linux-block@xxxxxxxxxxxxxxx, linux-kernel@xxxxxxxxxxxxxxx, bfq-iosched@xxxxxxxxxxxxxxxx,
oleksandr@xxxxxxxxxxxxxx, patdung100@xxxxxxxxx, cevich@xxxxxxxxxx, Paolo Valente <paolo.valente@xxxxxxxxxx>
Subject: [PATCH BUGFIX 5/6] block, bfq: get a ref to a group when adding it to a service tree
Date: Fri, 31 Jan 2020 10:24:08 +0100

BFQ schedules generic entities, which may represent either bfq_queues
or groups of bfq_queues. When an entity is inserted into a service
tree, a reference must be taken, to make sure that the entity does not
disappear while still referred in the tree. Unfortunately, such a
reference is mistakenly taken only if the entity represents a
bfq_queue. This commit takes a reference also in case the entity
represents a group.

Tested-by: Chris Evich <cevich@xxxxxxxxxx>
Signed-off-by: Paolo Valente <paolo.valente@xxxxxxxxxx>
---
block/bfq-cgroup.c | 2 +-
block/bfq-iosched.h | 1 +
block/bfq-wf2q.c | 16 +++++++++++++++-
3 files changed, 17 insertions(+), 2 deletions(-)

diff --git a/block/bfq-cgroup.c b/block/bfq-cgroup.c
index c818c64766e5..f85b25fd06f2 100644
--- a/block/bfq-cgroup.c
+++ b/block/bfq-cgroup.c
@@ -332,7 +332,7 @@ static void bfqg_put(struct bfq_group *bfqg)
kfree(bfqg);
}

-static void bfqg_and_blkg_get(struct bfq_group *bfqg)
+void bfqg_and_blkg_get(struct bfq_group *bfqg)
{
/* see comments in bfq_bic_update_cgroup for why refcounting bfqg */
bfqg_get(bfqg);
diff --git a/block/bfq-iosched.h b/block/bfq-iosched.h
index f1cb89def7f8..b9627ec7007b 100644
--- a/block/bfq-iosched.h
+++ b/block/bfq-iosched.h
@@ -984,6 +984,7 @@ struct bfq_group *bfq_find_set_group(struct bfq_data *bfqd,
struct blkcg_gq *bfqg_to_blkg(struct bfq_group *bfqg);
struct bfq_group *bfqq_group(struct bfq_queue *bfqq);
struct bfq_group *bfq_create_group_hierarchy(struct bfq_data *bfqd, int node);
+void bfqg_and_blkg_get(struct bfq_group *bfqg);
void bfqg_and_blkg_put(struct bfq_group *bfqg);

#ifdef CONFIG_BFQ_GROUP_IOSCHED
diff --git a/block/bfq-wf2q.c b/block/bfq-wf2q.c
index 26776bdbdf36..ef06e0d34b5b 100644
--- a/block/bfq-wf2q.c
+++ b/block/bfq-wf2q.c
@@ -533,7 +533,13 @@ static void bfq_get_entity(struct bfq_entity *entity)
bfqq->ref++;
bfq_log_bfqq(bfqq->bfqd, bfqq, "get_entity: %p %d",
bfqq, bfqq->ref);
+#ifdef CONFIG_BFQ_GROUP_IOSCHED
+ } else
+ bfqg_and_blkg_get(container_of(entity, struct bfq_group,
+ entity));
+#else
}
+#endif
}

/**
@@ -647,8 +653,16 @@ static void bfq_forget_entity(struct bfq_service_tree *st,

entity->on_st_or_in_serv = false;
st->wsum -= entity->weight;
- if (bfqq && !is_in_service)
+ if (is_in_service)
+ return;
+
+ if (bfqq)
bfq_put_queue(bfqq);
+#ifdef CONFIG_BFQ_GROUP_IOSCHED
+ else
+ bfqg_and_blkg_put(container_of(entity, struct bfq_group,
+ entity));
+#endif
}

/**
--
2.20.1