00:15 <alyssa> urjaman: Please get a backtrace from Xorg, if you can

00:31 <alyssa> gdb -batch -ex "run" -ex "bt" /usr/lib/xorg/Xorg.wrap -ex "q" 2>&1 >err & sleep 3 && DISPLAY=:0;0 /usr/bin/i3

00:31 <alyssa> :0.0 i mean

00:31 <alyssa> something like that?

09:47 <urjaman> alyssa: https://urja.dev/Xorg_mousepad_bt.txt

09:48 <urjaman> i decided to instead ssh into the laptop and just attach gdb to the running Xorg

10:00 <urjaman> hmm interesting, i decided (just because curious) to have gdb store a core file of this (so i dont need to keep a frozen Xorg on the laptop while i investigate) and when i opened the core file it told me that 'list_addtail (list=0x0, item=0x1aa0240)'

10:00 <urjaman> and yeah this is kinda fun, i have a core file so I can poke around it if you like

10:16 <warpme_> guys: what is progress with mali-t720? Non-supported GPU makes my H6 boxes useless despite generally is seems to be quite nice and well linux supported SoC....

12:34 <alyssa> urjaman: oh, yeah, that's the crash I was looking into the other night

13:07 <robmur01> FWIW that kernel oops looks like the timeout-reset thing that Steve's patch is meant to fix

13:12 <robmur01> warpme_: kernel fixes for 720/820 are on their way upstream for 5.4 (and should hopefully get backported to stable)

13:13 <robmur01> userspace sort-of-works if you hack out the whitelist check - I think it's mostly incomplete SFBD support that's going to bite

13:17 <warpme_> oh that's perfect. i'm already on 5.4-rc2 and have all working H6 (except black screen). Are these fixes already in mainline 5.4rc2?

13:20 <robmur01> not yet; the IOMMU maintainer has picked them up, so hopefully will send a fixes pull to Linus this week

13:26 <warpme_> ah ok - so rc3 might be nice bunch?

13:31 <warpme_> may You pls give me link to this IOMMU pull request?

15:00 <urjaman> robmur01: link to that patch? (sorry i havent read mailing lists in a while right now...)

15:18 <robmur01> urjaman: https://patchwork.freedesktop.org/patch/335022/

15:18 <robmur01> warpme_: No IOMMU PR yet, but the submaintainer PR is here - https://lore.kernel.org/linux-iommu/20191008085153.GA2514@8bytes.org/T/#t

15:26 <urjaman> robmur01: thanks, i'll apply it to my kernel and see if it helps

15:30 <robmur01> ideally xorg/mesa shouldn't get into the timeout condition in the first place, but at least we should handle it without oopsing ;)

15:32 <urjaman> yup

17:33 <warpme_> robmur01: isn't it this link about iommu for panfrost: https://patchwork.kernel.org/patch/11166777/ ?

17:35 <robmur01> yeah, those are the patches. When you asked about the pull request I took it literally, sorry :)

17:41 <warpme_> ok - so my 5.4-rc2 has already applied them from some time. still black screen.....