00:57karolherbst: anholt: fyi, I've wrote a patch to fix that crash on tegra: https://patchwork.freedesktop.org/patch/479134/ I don't think it helps witht he iommu issue though
03:55imirkin: anholt: is there a way for me to see what my original commit was on https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8544 ?
03:55imirkin: i see some dodgy stuff in there, want to check if it's you or me :)
03:56anholt: I don't know why the UI doesn't give it to you on the main MR page, but you can go to pipelines and see the first pipeline's commit https://gitlab.freedesktop.org/mesa/mesa/-/commit/1950c39004e8ccded587eb521b0bb29ad6f59c62
03:56imirkin: anholt: cool thanks
03:56imirkin: didn't think about the pipelines thing
03:56imirkin: hrmph. not surprising, but looks like "me"
03:56imirkin: (about vp->translated not getting set, or maybe used? anymore)
03:56imirkin: need to check what that's about
03:57imirkin: anholt: oh. and also... "remotes/gitlab/nv30-shader-state" on my local client. heh
03:59imirkin: i guess that only works since i never fetch from that remote
04:02imirkin: anholt: if you remind me about it on the weekend, i'll definitely look at it. tired now, and it requires a slight investigation to remember what it's all about
04:03imirkin: also need to make sure this doesn't interact poorly with draw somehow
04:04imirkin: (sometimes we'll plug "draw" instead of VS, if we fall of the "fast" path)
04:06anholt: sigh. apparently I can't run tgsi-mode shader-db any more for nv92, because src/mesa/state_tracker/st_glsl_to_tgsi.cpp:1648 triggers all over the place.