On Tue, 27 Oct 2015, Daniel Vetter <[email protected]> wrote: > On Fri, Oct 23, 2015 at 8:22 AM, Ville Syrjälä > <[email protected]> wrote: >> The regressing patch change didn't add the message, so there was a clear >> change in behaviour, and now it's papered over. > > It did move around the DRM_ERROR for all the others and also added the > SDE one for consistency. At least that's how I read that patch - I > could't find the SDE DRM_ERROR in the old code. Did I miss something?
Yes. We tried and failed to point out that this is a bisected regression with a bug report. The bad commit is *NOT* when the error message was added or moved. The first bad commit is commit aaf5ec2e51ab1d9c5e962b4728a1107ed3ff7a3e Author: Sonika Jindal <[email protected]> Date: Wed Jul 8 17:07:47 2015 +0530 drm/i915: Handle HPD when it has actually occurred which triggers printing of the error message. This is all mentioned in the bug, along with a few attempts at remedying the situation. BR, Jani. -- Jani Nikula, Intel Open Source Technology Center _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
