xsk: Do not sleep in poll() when need_wakeup set
authorMagnus Karlsson <magnus.karlsson@intel.com>
Tue, 14 Dec 2021 10:26:07 +0000 (11:26 +0100)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 22 Dec 2021 08:32:50 +0000 (09:32 +0100)
commite8d38fcd0a1ca4abd4b132d098c32bc13d6f54f0
tree6bc4aa4c14e54cbc1ad398c65d4871e24e8d1cce
parent6b2ee1002c6b96ce06a6f6bf9c23d08aa47190f8
xsk: Do not sleep in poll() when need_wakeup set

commit bd0687c18e635b63233dc87f38058cd728802ab4 upstream.

Do not sleep in poll() when the need_wakeup flag is set. When this
flag is set, the application needs to explicitly wake up the driver
with a syscall (poll, recvmsg, sendmsg, etc.) to guarantee that Rx
and/or Tx processing will be processed promptly. But the current code
in poll(), sleeps first then wakes up the driver. This means that no
driver processing will occur (baring any interrupts) until the timeout
has expired.

Fix this by checking the need_wakeup flag first and if set, wake the
driver and return to the application. Only if need_wakeup is not set
should the process sleep if there is a timeout set in the poll() call.

Fixes: 77cd0d7b3f25 ("xsk: add support for need_wakeup flag in AF_XDP rings")
Reported-by: Keith Wiles <keith.wiles@intel.com>
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://lore.kernel.org/bpf/20211214102607.7677-1-magnus.karlsson@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
net/xdp/xsk.c