13:30 <zmike> Soroush: I had meant to open a todo ticket for cleaning up emit_tex when I wrote match_tex_dests, but I think I didn't get around to it

15:58 <zmike> anholt: if you check out zmike/test3 now I've implemented ZINK_DEBUG=shaderdb on top of my precompile series

15:59 <anholt> \o/

15:59 <zmike> unfortunately this won't work out of the box due to https://gitlab.freedesktop.org/mesa/mesa/-/issues/5404

15:59 <zmike> you can work around this locally by uncommenting the vk_logd in anv's compiler_debug_log()

15:59 <anholt> oh, you're using debug output not the pipeline stats?

16:00 <zmike> uhhhh

16:00 <zmike> for now at least

16:00 <zmike> time during conference is limited

16:02 <zmike> I figured it's all the same prints that iris does so it should be suitable for initial analysis

16:08 <anholt> high recommend for using the pipeline stats, it's what the fossilize tools use.

16:11 <zmike> yep

16:12 <zmike> but I figured I could throw together something you could maybe use today instead of waiting until next week

20:45 <nebadon> Is there a place to file bug reports specifically about Zink?

20:49 <Sachiel> https://gitlab.freedesktop.org/mesa/mesa/-/issues

20:50 <nebadon> ok thanks! wasn't sure if there was a separate support for Zink

20:51 <nebadon> Tomb Raider no longer works for me with Zink its just artifacts galore, Bioshock Infinite seems ok though, tomb raider was working fine not long ago

20:53 <zmike> you could try bisecting to find when it broke

21:06 <nebadon> I have been running oibaf ppa for ubuntu so far, have not compiled it myself

21:08 <nebadon> if I can get some time I might try that, not sure what it entails with compiling mesa or if I even have the environment setup for that at the moment