]> git.ipfire.org Git - thirdparty/kernel/stable.git/commit
wifi: wlcore: Return -ENOMEM instead of -EAGAIN if there is not enough headroom
authorGuenter Roeck <linux@roeck-us.net>
Wed, 18 Mar 2026 06:46:36 +0000 (23:46 -0700)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 25 Mar 2026 10:08:54 +0000 (11:08 +0100)
commitcfa64e2b3717be1da7c4c1aff7268a009e8c1610
tree3d537e3463021fefffc32fadb21ffc202ca53c4b
parenta90279e7f7ea0b7e923a1c5ebee9a6b78b6d1004
wifi: wlcore: Return -ENOMEM instead of -EAGAIN if there is not enough headroom

[ Upstream commit deb353d9bb009638b7762cae2d0b6e8fdbb41a69 ]

Since upstream commit e75665dd0968 ("wifi: wlcore: ensure skb headroom
before skb_push"), wl1271_tx_allocate() and with it
wl1271_prepare_tx_frame() returns -EAGAIN if pskb_expand_head() fails.
However, in wlcore_tx_work_locked(), a return value of -EAGAIN from
wl1271_prepare_tx_frame() is interpreted as the aggregation buffer being
full. This causes the code to flush the buffer, put the skb back at the
head of the queue, and immediately retry the same skb in a tight while
loop.

Because wlcore_tx_work_locked() holds wl->mutex, and the retry happens
immediately with GFP_ATOMIC, this will result in an infinite loop and a
CPU soft lockup. Return -ENOMEM instead so the packet is dropped and
the loop terminates.

The problem was found by an experimental code review agent based on
gemini-3.1-pro while reviewing backports into v6.18.y.

Assisted-by: Gemini:gemini-3.1-pro
Fixes: e75665dd0968 ("wifi: wlcore: ensure skb headroom before skb_push")
Cc: Peter Astrand <astrand@lysator.liu.se>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Link: https://patch.msgid.link/20260318064636.3065925-1-linux@roeck-us.net
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
drivers/net/wireless/ti/wlcore/tx.c