On Tue, 24 Mar 2015, Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxxxxxxxx> wrote: > On 03/24/2015 01:16 PM, Jani Nikula wrote: >> On Tue, 24 Mar 2015, Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxxxxxxxx> wrote: >>> From: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx> >>> >>> It should have been negative since it is returned with ERR_PTR(). >> >> Please always reference the commit that introduced the issue. > > Is there some more precisely defined criteria for "always"? Always when you fix a bug that was introduced by another commit? I need the commit reference to check if I need to queue the fix to current development kernel (i.e. v4.0-rcN). If someone backports the commit, it's easier to check if the commit is referenced by a later commit potentially fixing issues in it. > In this case this is at the moment dead code freshly merged to -nightly > only so ideally it would even be better to fix it up in the original > patch if possible? I typically wouldn't know this by looking at a patch. The commit reference helps. And to squash it into another commit, the reference helps. Reducing the burden from the maintainers helps you too in the end! BR, Jani. > > Regards, > > Tvrtko -- Jani Nikula, Intel Open Source Technology Center _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/intel-gfx
