From: Jun'ichi Nomura <junichi.nom...@nec.com> [ Upstream commit 7d4553b69fb335496c597c31590e982485ebe071 ]
The devmap bulk queue is allocated with GFP_ATOMIC and the allocation may fail if there is no available space in existing percpu pool. Since commit 75ccae62cb8d42 ("xdp: Move devmap bulk queue into struct net_device") moved the bulk queue allocation to NETDEV_REGISTER callback, whose context is allowed to sleep, use GFP_KERNEL instead of GFP_ATOMIC to let percpu allocator extend the pool when needed and avoid possible failure of netdev registration. As the required alignment is natural, we can simply use alloc_percpu(). Fixes: 75ccae62cb8d42 ("xdp: Move devmap bulk queue into struct net_device") Signed-off-by: Jun'ichi Nomura <junichi.nom...@nec.com> Signed-off-by: Daniel Borkmann <dan...@iogearbox.net> Cc: Toke Høiland-Jørgensen <t...@redhat.com> Link: https://lore.kernel.org/bpf/20210209082451.ga44...@jeru.linux.bs1.fc.nec.co.jp Signed-off-by: Sasha Levin <sas...@kernel.org> --- kernel/bpf/devmap.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c index 2b5ca93c17dec..b5be9659ab590 100644 --- a/kernel/bpf/devmap.c +++ b/kernel/bpf/devmap.c @@ -815,9 +815,7 @@ static int dev_map_notification(struct notifier_block *notifier, break; /* will be freed in free_netdev() */ - netdev->xdp_bulkq = - __alloc_percpu_gfp(sizeof(struct xdp_dev_bulk_queue), - sizeof(void *), GFP_ATOMIC); + netdev->xdp_bulkq = alloc_percpu(struct xdp_dev_bulk_queue); if (!netdev->xdp_bulkq) return NOTIFY_BAD; -- 2.27.0