]> git.ipfire.org Git - thirdparty/kernel/stable.git/commit
net: usb: aqc111: Do not perform PM inside suspend callback
authorNikola Z. Ivanov <zlatistiv@gmail.com>
Fri, 13 Mar 2026 14:16:43 +0000 (16:16 +0200)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 25 Mar 2026 10:08:54 +0000 (11:08 +0100)
commit3267bcb744ee8a2feabaa7ab69473f086f67fd71
tree46f5acfa10733be20da5a17f8545b3e08f802af3
parent37bef86e5428d59f70a4da82b80f9a8f252fecbe
net: usb: aqc111: Do not perform PM inside suspend callback

[ Upstream commit 069c8f5aebe4d5224cf62acc7d4b3486091c658a ]

syzbot reports "task hung in rpm_resume"

This is caused by aqc111_suspend calling
the PM variant of its write_cmd routine.

The simplified call trace looks like this:

rpm_suspend()
  usb_suspend_both() - here udev->dev.power.runtime_status == RPM_SUSPENDING
    aqc111_suspend() - called for the usb device interface
      aqc111_write32_cmd()
        usb_autopm_get_interface()
          pm_runtime_resume_and_get()
            rpm_resume() - here we call rpm_resume() on our parent
              rpm_resume() - Here we wait for a status change that will never happen.

At this point we block another task which holds
rtnl_lock and locks up the whole networking stack.

Fix this by replacing the write_cmd calls with their _nopm variants

Reported-by: syzbot+48dc1e8dfc92faf1124c@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=48dc1e8dfc92faf1124c
Fixes: e58ba4544c77 ("net: usb: aqc111: Add support for wake on LAN by MAGIC packet")
Signed-off-by: Nikola Z. Ivanov <zlatistiv@gmail.com>
Link: https://patch.msgid.link/20260313141643.1181386-1-zlatistiv@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
drivers/net/usb/aqc111.c