Vegard Nossum found a unix socket OOM was possible, posting an exploit
My analysis is we can eat all LOWMEM memory before unix_gc() being
called from unix_release_sock(). Moreover, the thread blocked in
unix_gc() can consume huge amount of time to perform cleanup because of
huge working set.
One way to handle this is to have a sensible limit on unix_tot_inflight,
tested from wait_for_unix_gc() and to force a call to unix_gc() if this
limit is hit.
This solves the OOM and also reduce overall latencies, and should not
slowdown normal workloads.
Reported-by: Vegard Nossum <firstname.lastname@example.org>
Signed-off-by: Eric Dumazet <email@example.com>
Signed-off-by: David S. Miller <firstname.lastname@example.org>
Signed-off-by: Brad Figg <email@example.com>
net/unix/garbage.c | 7 +++++++
1 files changed, 7 insertions(+), 0 deletions(-)
static int gc_in_progress = 0;
+#define UNIX_INFLIGHT_TRIGGER_GC 16000
+ * If number of inflight sockets is insane,
+ * force a garbage collect right now.
+ if (atomic_read(&unix_tot_inflight) > UNIX_INFLIGHT_TRIGGER_GC && !gc_in_progress)
wait_event(unix_gc_wait, gc_in_progress == 0);
kernel-team mailing list