On Thu, Mar 07, 2013 at 05:10:36PM -0700, Eric Blake wrote:
On 03/06/2013 05:49 AM, Daniel P. Berrange wrote:
From: "Daniel P. Berrange" <berrange@redhat.com>
Add a virThreadCancel function. This functional is inherantly
s/inherantly/inherently/
dangerous and not something we want to use in general, but integration with SELinux requires that we provide this stub. We leave out any Win32 impl to discourage further use and because obviously SELinux isn't enabled on Win32
Signed-off-by: Daniel P. Berrange <berrange@redhat.com> --- src/libvirt_private.syms | 1 + src/util/virthread.h | 1 + src/util/virthreadpthread.c | 5 +++++ 3 files changed, 7 insertions(+)
+++ b/src/util/virthreadpthread.c @@ -236,6 +236,11 @@ void virThreadJoin(virThreadPtr thread) pthread_join(thread->thread, NULL); }
+void virThreadCancel(virThreadPtr thread)
It would help to add documentation right before this implementation that mentions that the function exists SOLELY for use by threads created by third-party libraries which are prepared for cancellation, and that most libvirt threads do not qualify. Having the discouraging comment tucked away solely in the commit message won't help a future coder who doesn't realize the dangers.
ACK with the comment added.
I'll add it in the header file since that's where people will probably find the function first. Daniel -- |: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|