Fix: urcu-qsbr: futex wait: handle spurious futex wakeups
authorMathieu Desnoyers <mathieu.desnoyers@efficios.com>
Wed, 22 Jun 2022 20:49:11 +0000 (16:49 -0400)
committerMathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mon, 27 Jun 2022 14:30:13 +0000 (10:30 -0400)
commit4974ad5f21d2cb17d313504f8d1252738526d588
treec157fafc996bb382b2100de87604bb8ee6934503
parenta18d0428065b301086d4e8a85bc057dc39946891
Fix: urcu-qsbr: futex wait: handle spurious futex wakeups

Observed issue
==============

The urcu-qsbr wait_gp() implements a futex wait/wakeup scheme identical to
the workqueue code, which has an issue with spurious wakeups.

A spurious wakeup on wait_gp can cause wait_gp to return with a
urcu_qsbr_gp.futex state of -1, which is unexpected. It would cause the
following loops in wait_for_readers() to decrement the
urcu_qsbr_gp.futex to values below -1, thus actively using CPU as values
will be decremented to very low negative values until it reaches 0
through underflow, or until the input_readers list is found to be empty.
The state is restored to 0 when the input_readers list is found to be
empty, which restores the futex state to a correct state for the
following calls to wait_for_readers().

This issue will cause spurious unexpected high CPU use, but will not
lead to data corruption.

Cause
=====

From futex(5):

       FUTEX_WAIT
              Returns 0 if the caller was woken up.  Note that a  wake-up  can
              also  be caused by common futex usage patterns in unrelated code
              that happened to have previously used the  futex  word's  memory
              location  (e.g., typical futex-based implementations of Pthreads
              mutexes can cause this under some conditions).  Therefore, call‐
              ers should always conservatively assume that a return value of 0
              can mean a spurious wake-up, and  use  the  futex  word's  value
              (i.e.,  the user-space synchronization scheme) to decide whether
              to continue to block or not.

Solution
========

We therefore need to validate whether the value differs from -1 in
user-space after the call to FUTEX_WAIT returns 0.

Known drawbacks
===============

None.

Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Change-Id: I87f7cd3b02820cefe850c3bdb8da27fb2f9be9b2
src/urcu-qsbr.c
This page took 0.025259 seconds and 4 git commands to generate.