From: Junio C Hamano Date: Mon, 14 Dec 2015 21:00:42 GMT Subject: Re: [PATCH 0/8] Rerolling sb/submodule-parallel-fetch for the time after 2.7 Message-ID: In-Reply-To: <566F2952.8000307@kdbg.org> Johannes Sixt writes: > Am 14.12.2015 um 20:37 schrieb Stefan Beller: >> I am sending out a new version for replacing sb/submodule-parallel-fetch for >> the time after the 2.7 release. >> >> The content are >> * all patches as in the branch sb/submodule-parallel-fetch >> * inlcuding the fixups as suggested by Hannes, >> * write a message to the debug log for better testing and debugging purposes >> (a patch cherry picked from the series which is supposed to build on top of this) >> >> The patches themselves were rebased such that there are no fixup commits >> any more, but we get things right the first time. >> >> The commit message of "run-command: add an asynchronous parallel child processor" >> has slightly been updated to mention the fact that we don't want to use waitpid(-1) >> but rather use the assumption of child's stderr living as long as the child itself. > > Thanks! I rebased a version of sb/submodule-parallel-fetch that > includes my suggested improvements, and the result is identical to > this series except for the trace output mentioned in the last bullet > point. > > With or without addressing my note about the commit message in 6/8: > > Acked-by: Johannes Sixt Thanks, both. This round does look very clean to build on top.