00:00 <clover[m]> It's güd

04:52 <steev> i think, at this point, most stuff is in? venus is still not quite in (maybe 6.7?) and webcam is in progress, but also requires userland reworking

04:55 <steev> robclark: btw, something i notice, and i'm not sure where the issue is exactly... on X13s, in gnome with *xorg* backend, using a terminal (doesn't matter which, i see it on both gnome-terminal and alacritty with the x11 winit backend), occasionally, running a command will seem to not refresh for a little bit

04:55 <steev> e.g. git pull won't actually show output until you hit enter a second time

04:56 <steev> it's only in the xorg backend though

05:00 <steev> it makes me doubt i've hit enter sometimes, except that when i hit enter the second time, i get a blank line, so i know i did

05:20 <steev> hm, and we seem to lose touchpad for some reason in 6.6.0-rc2 (was definitely working in rc1)

05:20 <steev> haven't looked into and work meeting in the morning, so need to head to bed

05:22 <steev> i'll push out what i have, but i haven't kept up on patchsets so some may be outdated - clover[m] if you want to play with it (but keep in mind the touchpad might go poof, it's lenovo-x13s-v6.6.0-rc2

07:25 <ema> steev: indeed it was :)

13:10 <jhovold> Here's an updated wip branch for the X13s based on 6.6-rc2:

13:10 <jhovold> - https://github.com/jhovold/linux/tree/wip/sc8280xp-v6.6-rc2

13:10 <jhovold> Changes since the previous 6.5 branch include:

13:10 <jhovold> - fix touchpad regression in 6.6-rc1

13:10 <jhovold> - fix frambuffer console regression in 6.6-rc1

13:10 <jhovold> - drop audio fw path series

13:10 <jhovold> - NOTE: depends on fw link just merged to linux-firmware

13:10 <jhovold> - ln -s /lib/firmware/qcom/sc8280xp/LENOVO/21BX/audioreach-tplg.bin /lib/firmware/qcom/sc8280xp/SC8280XP-LENOVO-X13S-tplg.bin

13:20 <steev> oh nice

14:08 <steev> jhovold: confirming i have touchpad with 6.6.0-rc2, thank you for tracking that down! i'd seen the message in rc1, but my touchpad still worked so i didn't think it was related

15:01 <steev> clover[m]: updated rc2 with the fixes jhovold mentioned above. so it's that + lower cluster idle states, aspm, and acv enable mask basically

15:14 <jhovold> steev: yeah, random changes in timing could determine whether probe succeeds for fails

17:11 <clover[m]> 6.6.0-rc2 boots, but dummy output on audio, maybe i need to do that fw link thing

17:12 <steev> you do

17:12 <steev> because it's not trying to load the firmware without that link since we aren't saying where to find it in the dtb anymore

17:14 <clover[m]> Yep that worked

17:25 <clover[m]> steev: now im getting this for bt

17:25 <clover[m]> sudo btmgmt public-addr 00:25:81:17:62:36

17:25 <clover[m]> Set Public Address for hci0 failed with status 0x11 (Invalid Index)

17:25 <clover[m]> any ideas?

17:26 <clover[m]> hm reboot fixed it

17:26 <clover[m]> will let you know if i hit it more often

17:29 <clover[m]> okay, i updated x13s-firmware so arch people can use that along with linux-x13s-rc to get the latest and greatest features

17:32 <steev> ah right, 6.6 drops my nvm patch, i need to look at tim's newest patch to see what it's doing and modify again

17:33 <clover[m]> getting some crackles coming out of the speakers that i hadn't noticed before while paying music. fyi srinik

17:36 <clover[m]> i wonder if i should drop that set-digital-gain-at-0dB.patch in the alsa-ucm-conf

17:44 <clover[m]> @O_o: just hit the reboot after gdm login.

18:13 <jhovold> clover[m]: srinik has updated his volume-fixes PR, i only use that, I suggest you do the same

18:14 <clover[m]> yes i dropped the patch im carrying and updated the git source to his latest commit

18:14 <clover[m]> thanks

18:14 <jhovold> haven't heard any crackles, except for the occasional click when stopping puleaudio, which is a known issue

22:00 <robclark> HdkR: CS:GO menu is still a bit of a slide-show.. but that's still without thunking

22:00 <HdkR> robclark: With the driver bug fixed?

22:01 <robclark> driver bug?

22:01 <HdkR> https://gist.github.com/Sonicadvance1/713a136c80b0016552ece40c34ff2d36 This convert_ubyte driver overhead

22:02 <HdkR> That's the CS:GO menu completely consumed by the driver and then kernel time

22:02 <robclark> ahh, right.. no changes for that.. but not sure there is any bug there.. or at least I don't think we have any missing format support

22:04 <robclark> hmm, how do I even update mesa without thunking.. I guess my cross-compile setup isn't the greatest

22:04 <HdkR> Would be updating the mesa install inside the rootfs without thunks

22:05 <robclark> right, which means cross-compile for x86

22:05 <robclark> hmm, can you get an apitrace of that?

22:06 <HdkR> Potentially, let's see if it works

22:12 <HdkR> robclark: https://drive.google.com/file/d/10QBSjuPJb9m-nuz6ly1-fysL9gxiRZcm/view?usp=sharing Looks like it captured and replayed correctly

22:12 <robclark> not a small trace

22:12 <robclark> lets see what I can figure out from that

22:13 <HdkR> Killed it as soon as it reached the menu