af_unix: Set gc_in_progress to true in unix_gc().
Subsystems (3)
PATCH 1/1 af_unix: Set gc_in_progress to true in unix_gc().
From: Kuniyuki Iwashima <[email protected]>
[ Upstream commit d82ba05263c69fa2437fe93e4e561cc40f4c03af ]
Igor Ushakov reported that unix_gc() could run with gc_in_progress being false if the work is scheduled while running:
Thread 1 Thread 2 Thread 3
-------- -------- --------
unix_schedule_gc() unix_schedule_gc()
`- if (!gc_in_progress) `- if (!gc_in_progress)
|- gc_in_progress = true |
`- queue_work() |
unix_gc() <----------------/ |
| |- gc_in_progress = true
... `- queue_work()
| |
`- gc_in_progress = false |
| unix_gc() <---------------------------------------------' |
... /* gc_in_progress == false */
|
`- gc_in_progress = false
unix_peek_fpl() relies on gc_in_progress not to confuse GC by MSG_PEEK.
Let's set gc_in_progress to true in unix_gc().
--- net/unix/garbage.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
net/unix/garbage.c +2 -1
@@ -583,6 +583,8 @@ static void __unix_gc(struct work_struct *work)struct sk_buff_head hitlist;struct sk_buff *skb; +WRITE_ONCE (gc_in_progress, true); +spin_lock (&unix_gc_lock);if (!unix_graph_maybe_cyclic) {
@@ -613,7 +615,6 @@ static DECLARE_WORK(unix_gc_work, __unix_gc);void unix_gc (void ) { -WRITE_ONCE (gc_in_progress, true);queue_work (system_unbound_wq, &unix_gc_work); }
Signature · 1 line
2.47.3
Sasha Levin
maintainerQuoted 1 line
[ move
WRITE_ONCE(gc_in_progress, true)into the__unix_gc()work function and drop it fromunix_gc(). ]
Dropping the set from unix_gc() and doing it in the work function is fine upstream, but only because later refactors made wait_for_unix_gc() gate flush_work() on unix_graph_cyclic_sccs. On these trees it still gates on gc_in_progress, so this brings back a window where the over-limit throttle stops waiting for GC.
Can you respin keeping gc_in_progress set in unix_gc() before queue_work()? The code is identical across all three, so one version covers them.
--
Thanks,
Sasha
PATCH v2 1/1 af_unix: Set gc_in_progress to true in unix_gc().
From: Kuniyuki Iwashima <[email protected]>
[ Upstream commit d82ba05263c69fa2437fe93e4e561cc40f4c03af ]
Igor Ushakov reported that unix_gc() could run with gc_in_progress being false if the work is scheduled while running:
Thread 1 Thread 2 Thread 3
-------- -------- --------
unix_schedule_gc() unix_schedule_gc()
`- if (!gc_in_progress) `- if (!gc_in_progress)
|- gc_in_progress = true |
`- queue_work() |
unix_gc() <----------------/ |
| |- gc_in_progress = true
... `- queue_work()
| |
`- gc_in_progress = false |
| unix_gc() <---------------------------------------------' |
... /* gc_in_progress == false */
|
`- gc_in_progress = false
unix_peek_fpl() relies on gc_in_progress not to confuse GC by MSG_PEEK.
Let's set gc_in_progress to true in unix_gc().
[ Add setting gc_in_progress in __unix_gc(). Keep the existing
set in unix_gc() for wait_for_unix_gc() over-limit throttling. ]
--- net/unix/garbage.c | 2 ++ 1 file changed, 2 insertions(+)
net/unix/garbage.c +2 -0
@@ -583,6 +583,8 @@ static void __unix_gc(struct work_struct *work)struct sk_buff_head hitlist;struct sk_buff *skb; +WRITE_ONCE (gc_in_progress, true); +spin_lock (&unix_gc_lock);if (!unix_graph_maybe_cyclic) {
Signature · 1 line
2.47.3