On Tue, May 26, 2015 at 10:20:26AM +0200, Daniel Vetter wrote: > On Tue, May 26, 2015 at 08:57:13AM +0100, Chris Wilson wrote: > > On Tue, May 26, 2015 at 07:47:22AM +0200, Daniel Vetter wrote: > > > This reverts commit 118182e9d7d5afa0c7c10f568afb46ab78b462e9. > > > > > > It's causing too much trouble when compile-testing for non-i915 folks. > > > > What's the argument against using COMPILE_TEST as a proxy for detecting > > all(yes|mod)config as in Damien's patch? > > I also thought about random users bisecting who'll then also might trip > over this. Figure that overall the benefit isn't good enough given the > trouble this seems to have caused immediately. 0-day is pretty fast at > screaming around about new compiler warnings anyway, for the cases where I > miss them. You haven't fixed the issue of breaking a bisect using allyesconfig unless you modify history. Either way they need to do a couple of skips. -Chris -- Chris Wilson, Intel Open Source Technology Centre _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/intel-gfx
