From: Wang Liang Date: Thu, 27 Feb 2025 08:10:52 +0000 (+0800) Subject: xsk: Fix __xsk_generic_xmit() error code when cq is full X-Git-Tag: nvme-6.15-2025-04-10~23^2~3 X-Git-Url: https://www.infradead.org/git/?a=commitdiff_plain;h=5d0b204654de25615cf712be86c3192eca68ed80;p=nvme.git xsk: Fix __xsk_generic_xmit() error code when cq is full When the cq reservation is failed, the error code is not set which is initialized to zero in __xsk_generic_xmit(). That means the packet is not send successfully but sendto() return ok. Considering the impact on uapi, return -EAGAIN is a good idea. The cq is full usually because it is not released in time, try to send msg again is appropriate. The bug was at the very early implementation of xsk, so the Fixes tag targets the commit that introduced the changes in xsk_cq_reserve_addr_locked where this fix depends on. Fixes: e6c4047f5122 ("xsk: Use xsk_buff_pool directly for cq functions") Suggested-by: Magnus Karlsson Signed-off-by: Wang Liang Signed-off-by: Martin KaFai Lau Acked-by: Stanislav Fomichev Link: https://patch.msgid.link/20250227081052.4096337-1-wangliang74@huawei.com Signed-off-by: Alexei Starovoitov --- diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c index e5d104ce7b82..5696af45bcf7 100644 --- a/net/xdp/xsk.c +++ b/net/xdp/xsk.c @@ -806,8 +806,11 @@ static int __xsk_generic_xmit(struct sock *sk) * if there is space in it. This avoids having to implement * any buffering in the Tx path. */ - if (xsk_cq_reserve_addr_locked(xs->pool, desc.addr)) + err = xsk_cq_reserve_addr_locked(xs->pool, desc.addr); + if (err) { + err = -EAGAIN; goto out; + } skb = xsk_build_skb(xs, &desc); if (IS_ERR(skb)) {