On 03/19/2014 11:29 AM, Michal Privoznik wrote:
Currently, we use pthread_sigmask(SIG_BLOCK, ...) prior to calling
poll(). This is okay, as we don't want poll() to be interrupted.
However, then - immediately as we fall out from the poll() - we try to
restore the original sigmask - again using SIG_BLOCK. But as the man
page says, SIG_BLOCK adds signals to the signal mask:
SIG_BLOCK
The set of blocked signals is the union of the current set and the set argument.
Therefore, when restoring the original mask, we need to completely
overwrite the one we set earlier and hence we should be using:
SIG_SETMASK
The set of blocked signals is set to the argument set.
Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com>
---
Notes:
This is a very old bug. It's worth backporting onto all maint branches we have.
src/rpc/virnetclient.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
ACK. And yes, it needs to be backported.
Bug was introduced in virnetclient.c at its creation in commit 434de30
(v0.9.3); that code was generalized from src/remote/remote_driver.c, but
at that time, remote_driver.c was doing things correctly. Not sure how
I missed the corruption (obviously, the refactoring to add the rpc
client code wasn't quite a copy and paste, for that typo to have snuck in).
--
Eric Blake eblake redhat com +1-919-301-3266
Libvirt virtualization library
http://libvirt.org