On Wed, Sep 27, 2017 at 3:47 PM, Harry Wentland harry.wentland@amd.com wrote:
On 2017-09-27 06:15 AM, Daniel Vetter wrote:
Ok, here's one more attempt at scrolling through 130k diff. +19. The DC logger is still a rather sore thing, but I know that the DRM_DEBUG +stuff just isn't up to the challenges either. We need to figure out something +that integrates better with DRM and linux debug printing, while not being +useless with filtering output. dynamic debug printing might be an option.
Have there been any thoughts on improving log filtering in DRM? I think our KFD guys (which hopefully go upstream soon as well) are using dynamic debug prints for it but not sure if that's entirely the right approach.
Plenty of complaints for sure, but I don't think anyone has looked into it seriously. The big trouble is backwards compat with existing drm.debug (too many people know about that one), and with dynamic printk specifically, that non-debug builds don't build it (so useless as-is for debugging issues users have). That was at least the conclusion from the last few discussions I remember. -Daniel