19:00:47 #startmeeting 19:00:47 Meeting started Wed Apr 30 19:00:47 2025 UTC. The chair is bwh. Information about MeetBot at http://wiki.debian.org/MeetBot. 19:00:47 Useful Commands: #action #agreed #help #info #idea #link #topic. 19:01:16 If anyone else besides ukleinek is attending, please speak up 19:03:09 OK, let's go to the first topic 19:03:16 * ukleinek nods 19:03:19 #topic Bug #1086638: (i, ) linux-image-6.11.5: usbguard-daemon invalid opcode: 0000, usb's not usable 19:04:04 I think we need to ask to re-test on the latest kernel in unstable, and then (if not fixed) forward upstream 19:04:06 When reading the report I wondered how relevant usbguard is. Is it worth to ask for a test to reproduce without usbguard 19:04:25 I think there was a test without usbguard 19:05:06 ok, then ack, we ask for a test with a recent kernel and depending on the outcome close or forward upstream. 19:05:21 In message #34 I think tests #1-6 were done without the usbguard rules 19:05:53 at least all the rules were removed 19:05:53 #action bwh will request re-testing #1086638 with a newer kernel version, then forward upstream 19:06:02 👍 19:06:06 #topic Bug #1092103: (i, ) Linux 6.12 freeze on small machine 19:06:08 hi 19:06:15 Hi carnil! 19:06:23 sorry for beeing late 19:06:37 No problem, you did say not to expect you 19:07:30 that bug should indeed be easy to reproduce 19:07:54 It sounds like it. Are you willing to try that? 19:08:02 I could try to do that tomorrow morning 19:08:08 okay ukleinek is fine too 19:08:15 * ukleinek could look into it though I don't know much about x86, so I wouldn't mind if someone else looks 19:08:55 #action ukleinek tries to reproduce #1092103 19:09:08 carnil: Please do not work on May 1st 19:09:34 #topic Bug #1100153: (i, M) linux: hangs on resume from suspend or hibernation with 6.12-6.13 on ThinkPad X1 Carbon Gen 10 19:09:38 ok, that is my chance to be quicker, will do it tomorrow! :-) 19:09:41 I'm in hamburg :) I'm expected to do some work ;-) 19:10:02 Nothing new on this one since yesterday 19:10:15 yes, tagged moreinfo, next! 19:10:28 #topic Bug #1101944: (i, ) linux-image-6.1.0-32-amd64: list_del corruption ... kernel BUG at lib/list_debug.c:62! 19:10:41 upstream's response wasn't very informative 19:11:16 For the first point, the "[cut here]" is misplaced and we actually also want the line before it 19:11:29 * ukleinek also wondered why we didn't have the error message in the log. But apart from that I'm as clueless as before upstream's response 19:12:32 We could ask the reporter if they can still find the message in their journal 19:12:45 I think we can use decode_stacktrace.sh with vmlinux from the debuginfo package, to get line numbers 19:13:02 ukleinek: yes 19:13:24 Matthew Wilcox already decoded the line number?! 19:13:45 Not for the backtrace 19:14:24 OK, who will take this one? 19:14:26 #action ukleinek to check backtrace for details (decode-stacktrace) and ask reporter for more kernel log from before the cut line 19:14:31 thanks 19:14:47 #topic Bug #1102175: (i, ) Repeated lockups in AMDGPU DRM driver 19:15:47 It seems like there is a cascade of failures and it's not that clear what is the trigger 19:16:51 The PCI device that gets stuck in D3 cold might be part of a USB 4 / Thunderbolt bridge, which would also interact with the amdgpu driver 19:17:15 or rather, it would interact with the GPU, and indirectly with the amdgpu driver 19:17:58 I'm not sure what we can ask next, or where to forward it 19:18:29 there are some drm debug messages that can be enabled, but no clue if that would help 19:20:03 Forward to the dri-devel@l.f.o or amd-gfx@l.f.o? 19:20:56 If there is a setting that can prevent the card going into D3cold, that could be suggested as a workaround to the reporter 19:21:33 The thing is, D3cold means the card has been fully powered off 19:22:14 which I would only expect to happen in S3 or deeper system power state 19:22:57 A regular driver can put it in D3hot but not D3cold 19:23:22 could be that the machine goes into S3 when idle. Without knowing about what S3 is, at least gnome puts the machine to sleep after a while. 19:23:30 yes 19:24:10 * ukleinek idles in #dri-devel, I can ask there if someone has an idea 19:24:27 Are you OK to take this bug then? 19:24:40 * ukleinek looks at his list and nods 19:25:37 #action ukleinek will try to make progress on #1102175 19:25:48 #topic Bug #1103397: (i, u) linux-image-amd64: Kernel Panic: copy_fpstate_to_sigfrane crashes. 19:26:26 Let's wait a bit more on upstream 19:26:33 OK 19:26:51 #topic Bug #1104269: (i, uM) linux-image-6.12.22-amd64: amdgpu flip_done and commit wait timeouts during normal use - causes session lockup 19:27:11 No new information since yesterday 19:27:22 #topic Bug #991712: (i, U) Intel wifi card "Killer Wi-Fi 6 AX1650 (2x2)" in new Dell XPS 15 doesn't work on bullseye's instaleld kernel 19:27:38 Ah, carnil closed it 19:27:47 * ukleinek was about to write that 19:27:58 #topic Bug #1104165: (i, u) linux-image-6.12.22-amd64: Freezing with 'amdgpu_dm_plane_helper_prepare_fb bind failed' some time after returning from suspend 19:29:09 what is guake? Something challenging for the hardware? 19:29:22 It looks there's GPU memory leak triggered by some change after resume 19:29:41 Quake-style console 19:29:44 Maybe in the error path of the failed bind 19:30:18 right 19:30:30 I think this one can be forwarded upstream 19:30:48 #action bwh will forward #1104165 upstream 19:31:01 #topic Bug #1104327: (i, u) firmware-amd-graphics: after laptop resume, AMD graphics jsut show large patterns on all terminals 19:31:16 (It's a driver bug, not a firmware bug) 19:31:41 * ukleinek sometimes wonders if that are just different symptoms of similar problems. amdgpu is quite present in your bug list. 19:32:02 It's possible 19:32:33 s/your/our/ 19:33:00 I think for this one we need to ask for a kernel log and device info (lspci) 19:33:12 and reassign to linux 19:33:18 I already did that 19:33:24 ah indeed 19:33:46 So who will ask? 19:34:04 I can add that to my question in #dri-devel 19:34:34 Well it probably needs more info from the reporter first, doesn't it? 19:35:45 not necessarily 19:35:51 OK 19:36:24 * ukleinek already asked in #dri-devel 19:36:49 #action ukleinek will try to progress #1104327 19:37:01 #topic Bug #1035878: (n, +u) rpi400: visual speckling on 'faulty' HDMI port during mouse movement 19:37:41 This has a patch from the RPi kernel but it's not upstream 19:37:55 So someone will need to help get it applied upstream 19:38:28 #action bwh will forward #1035878 and its patch upstream 19:38:36 bwh: thanks 19:38:50 #topic Bug #1100105: (n, M) linux-image-amd64: Issues with Framework audio module 19:39:17 We are waiting on the reporter to confirm whether this was faulty hardwae or still reporducible with a replacement 19:39:26 * ukleinek nods 19:39:35 #topic Bug #1103277: (n, Mu) linux: CVE-2024-38541 for 6.1 branch 19:40:02 that patch was taken by the stable team, included in the review round currently. 19:40:05 OK 19:40:06 the patch was queued and now in the 6.1.136 review 19:40:26 which means we have it in the next update 19:40:27 Then I guess the moreinfo and wontfix tags are no longer correct? 19:40:29 * ukleinek expects this to resolve itself and carnil already said he will care for the bug colser 19:40:54 yes the moreinfo and wontfix can be removed again I think 19:41:11 * ukleinek runs bts tag 1103277 - moreinfo wontfix 19:41:16 me too 19:41:18 (as side note, I still do think the security scanners are wrong in assessing it as critical issue ;-)) 19:41:36 #topic Bug #1103830: (n, ) linux-image-6.12.22-amd64: No keyboard or touchpad functionality on a Macbook Pro 14,1 19:41:59 carnil: blindly believing automatic security scanners is always wrong 19:42:44 > [ 538.721827] dmar_fault: 73361 callbacks suppressed 19:42:48 eek 19:43:25 Anyway, this is confirmed to be a regression since bookworm 19:44:05 so we ask for a bisection, to at least identify the first broken debian package? 19:44:18 Yes that might be worth doing 19:44:32 * ukleinek already handled the bug before, will ask that. 19:44:49 The changes in applespi since 6.1 are very small... 19:45:35 right, pretty sure the problem is somewhere else, maybe to do with the IOMMU 19:45:36 #action ukleinek to follow up on #1103830 19:46:04 It might be worth asking to test with the IOMMU disabled (something like intel_iommu=off) 19:46:28 #topic Bug #1104080: (n, ) linux: Please build with CONFIG_KERNEL_GZIP=y instead of CONFIG_KERNEL_XZ=y on sh4 19:47:36 cbmuser: Is xz causing out-of-memory, or is there some other problem with it? 19:48:29 tag + moreinfo? 19:48:46 #action bwh will ask for clarfication on #1104080 19:48:58 * ukleinek nods 19:49:00 #topic Bug #946791: (n, M) Please enable CONFIG_IOSCHED_BFQ=y 19:50:10 continue waiting 19:50:14 right 19:50:24 #topic Bug #1100895: (n, ) nouveau probe of GeForce 840M crashes since 6.1.119-1 19:50:55 There's more information here, and I updated the title to summarise that 19:51:32 So I think this can be forwarded upstream? 19:52:09 right 19:52:45 #action bwh will forward #1100895 upstream 19:52:57 #topic Bug #1102563: (n, M) usbhid error -71 19:53:12 Still waiting for the full kernel log 19:53:16 * ukleinek nods 19:53:34 #topic Bug #1067006: (n, uR) rpc-statd.service: State 'stop-sigterm' timed out. Killing. 19:53:43 (skipping the < normal severity bugs in linux) 19:54:20 This still seems to be waiting for a response from upstream 19:55:09 I think that's all the bugs we need to discuss 19:55:33 The forwarding to upstream is quite fresh, so let's wait to hear an answer from them, we have not modified the systemd units IIRC so it's upstream dependencies so would be good to hear from linux-nfs upstream on the issue 19:55:43 right 19:56:08 #topic Migration excuses 19:56:46 I think we're just waiting for package builds and the 10 day timer 19:57:02 looks like it, yes 19:57:24 #topic New upstream versions 19:57:57 It would be nice if someone could work on linux 6.15, but I don't think it's that high a priority 19:58:36 * ukleinek wants to have 6.15 on his NAS, but given my long list I won't promise anything 19:58:42 OK 19:58:53 #topic Merge requests 19:59:04 Did anyone want to discuss any of these? 19:59:49 I need ot re-review MRs for initramfs-tools and kernel-handbook 19:59:56 I will look through the arm specific ones when I'm done with my buglist 20:00:30 OK 20:00:37 #topic Any other business 20:00:38 ah, and I intended to look into the CONFIG_HZ_1000 one, but don't remember the details and didn't come around to do that 20:01:42 * ukleinek can chair next time again, knowing that carnil doesn't like chairing the jitsi meetings. 20:02:06 as mentioned above I have a bit of free time in the next 3 days so I will likely look at some other bugs to take action on and still aim to look at the important and normal severity bugs to see what can be closed 20:02:22 carnil: sounds good 20:02:23 I can take the chair next week as it would be my turn 20:02:32 I just mentioned it would be nice to not have to do always the jitsi one :) 20:03:00 we just need a 5th guy in the team, then ... 20:03:25 Doesn't have to be a guy 20:03:39 though, looking at the current team... :-/ 20:03:47 * ukleinek didn't want to imply a gender with "guy" 20:04:26 So, who will it be next week? 20:05:04 carnil: Schere! 20:05:11 Stein 20:05:15 I will do it ;-) 20:05:29 #agreed carnil will chair next week 20:05:31 #endmeeting