On Tue, Feb 22, 2022 at 03:10:52PM +0100, Tim Wiederhake wrote:
On Thu, 2022-02-17 at 13:15 +0000, Daniel P. Berrangé wrote:
On Thu, Feb 17, 2022 at 02:01:01PM +0100, Tim Wiederhake wrote:
Really should be giving an explanation of why this change is required.
Added locally to commit message:
Making the mutex static and independent of the lifetime of the driver object allows for simplification of mutex handling during the object's initialization and cleanup using VIR_LOCK_GUARD's unlock-on-scope-exit behavior.
Note, the locking during init/cleanup is pointless because we rely on serialization and a mere mutex doesn't give us safety in cleanup, as any concurrent thread will be accessing a freed driver anyway. I guess that's pre-existing issue though. Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|