On Wed, Aug 08, 2012 at 05:52:58PM -0700, Leo Alterman wrote: > lockfile_lock() accepts a timeout argument but, aside from unit tests > pertaining to timeout, its value is always 0. Since this feature relies on > a periodic SIGALRM signal, which is no longer a given after refactoring > timeval.c, the cleanest solution is to just remove it. > > Signed-off-by: Leo Alterman <lalter...@nicira.com>
I think that this fixes a failure in the unit tests that the first patch causes. If so, then that means that this patch should be the first in the series, not the second, so that there is no point at which the testsuite is broken. (One way to reorder the commits is to run "git rebase -i origin/master", then reorder the two "pick" lines in the editor that comes up and save and exit the editor.) The lockfile_lock() comment should mention that it won't block waiting for the lock. I believe that lockfile_try_lock()'s 'block' parameter may now be removed because it is always passed as false. Thanks, Ben. (I'll look at patch 1/2 now.) _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev
