On Mon, Oct 02, 2023 at 03:52:44PM +0200, Alexander Lobakin wrote: > From: Andrew Kanner <andrew.kanner@xxxxxxxxx> > Date: Thu, 28 Sep 2023 23:44:40 +0300 > > > Syzkaller reported the following issue: > > [...] > > > PS: the initial number of entries is 0x20000000 in syzkaller repro: > > syscall(__NR_setsockopt, (intptr_t)r[0], 0x11b, 3, 0x20000040, 0x20); > > > > Link: https://syzkaller.appspot.com/text?tag=ReproC&x=10910f18280000 > > > > net/xdp/xsk_queue.c | 3 +++ > > 1 file changed, 3 insertions(+) > > > > diff --git a/net/xdp/xsk_queue.c b/net/xdp/xsk_queue.c > > index f8905400ee07..1bc7fb1f14ae 100644 > > --- a/net/xdp/xsk_queue.c > > +++ b/net/xdp/xsk_queue.c > > @@ -34,6 +34,9 @@ struct xsk_queue *xskq_create(u32 nentries, bool umem_queue) > > q->ring_mask = nentries - 1; > > > > size = xskq_get_ring_size(q, umem_queue); > > + if (size == SIZE_MAX) > > unlikely(). > > > + return NULL; > > + > > size = PAGE_ALIGN(size); > > > > q->ring = vmalloc_user(size); > > Thanks, > Olek Thanks, Olek. That is a reasonable optimization, I'll add it in v2. -- pw-bot: cr Andrew Kanner
