19:00:03 #startmeeting 19:00:03 Meeting started Wed Aug 5 19:00:03 2026 UTC. The chair is waldi. Information about MeetBot at https://wiki.debian.org/MeetBot. 19:00:03 Useful Commands: #action #agreed #help #info #idea #link #topic. 19:00:10 Hi 19:00:11 welcome to todays meeting 19:00:13 Hi 19:00:15 o/ 19:00:18 #chair bwh carnil ukleinek 19:00:18 Current chairs: bwh carnil ukleinek waldi 19:00:36 anything we should discuss first? 19:00:47 hi 19:00:48 nothing I'm aware of 19:01:02 Nothing urgent from me 19:01:12 nothing 19:01:21 then we can go straight into our long list of bugs 19:01:33 #topic #1126671: (S, M) linux-image-6.17.13+deb13-amd64: Disconnect root (path=/) during heavy load on NvME, partial corruption on NTFS.crash soon but not yet 19:01:48 this waits for the submitter 19:02:00 oh cool 19:02:01 right 19:02:19 #topic #1141969: (i, M) linux-image-7.1.3+deb14-amd64: Cannot shutdown computer 19:02:42 here i got a bit lost 19:03:02 is it a regression? 19:03:11 Yes, from 7.0 to 7.1 19:03:57 not fixed in 7.2, but they consider bisection risky 19:05:02 I wouldn't fear bisecting, but I think it would be unfair to talk them into it. 19:05:17 right 19:05:35 Since this is reproducible in 7.2 can we take it upstream anyway? 19:05:48 and those are two possibly unrelated problems in one report 19:05:56 yes 19:05:57 I wonder if there are kernel messages about the issue 19:06:50 waldi: Ah, I missed that this is not the original reporter 19:08:18 Now we have 3 reporters, maybe with all different bugs 19:08:19 there are even three people contributing to that bug 19:09:24 I think maybe someone should ping the original reporter 19:09:33 and with hardware of vastly different vintage. the submitter from 2022, the others from 2010 19:09:37 and split the bug in 3? 19:10:02 is it clear that it's a regression for all from 7.0 even? 19:10:30 no. the submitter did not teil when or if it ever worked 19:11:01 But probably was for them, as this was "the latest kernel update" 19:11:02 four reporters even 19:11:11 For "simon" it's not clear 19:12:34 for Martin Steigerwald and Ole, definitely reported as a regression from 7.0 to 7.1 19:12:40 #action ukleinek responds to the original submitter of #1141969 and asks the other contributors to create a separate bug 19:12:57 should we simply try to rebootstrap the bug with the original reporter and mention to the other involved to please fill new bugreport if the experience similar issues so we can disentangle bit things? 19:13:01 ok 19:13:18 #topic #1142099: (i, M) /usr/share/bug/linux-image-6.12.95+deb13-amd64-unsigned/include-1cmdline: Kernel 6.12.95 does not start boot sequence without setting acpi=off or recovery mode (logs in this mode) 19:13:45 Still waiting for the reporter 19:13:57 #topic #1142863: (i, M) linux-image-6.12.96+deb13-amd64: S3 (deep) suspend powers the machine off instead of suspending 19:14:06 also waiting 19:14:20 #topic #1143268: (i, Mu) linux-image-6.12.100+deb13-amd64: HID++ devices broken, ghost keyboard events, systemd-modules-load failure 19:15:24 this one is very odd, the bisect result does not make sense so maybe the whole problem is a followup problem, we did not have a full kernel log right? 19:15:44 no log provided, no hardware information 19:16:01 (/me is bit fed up with reports which are lengthly and feels like ai assistent generated) 19:16:09 I also got that impression 19:16:11 nvidia hardware and the reporter talks about the problem being a followup problem to not being able to load the nvidia modules 19:16:55 ah, but nvidia was ruled out as culprit 19:17:45 ack, bisection result looks bogous 19:18:31 "The system becomes completely unusable as no keyboard input is recognized, not even at the GRUB level on subsequent boots." 19:18:32 "this might indicate a build/patch ordering issue in the 6.12.100 stable release" is non-sense 19:18:46 I can't work out what's real information and what's slop 19:18:50 I vote to close it 19:19:12 +1 19:19:13 +1 19:19:25 please do 19:19:27 #action bwh will close #1143268 as incoherent slop 19:19:35 #topic #1143269: (i, ) linux: Fatal error during GPU init 19:20:01 this is also weird and we lack a log that shows what is claimed. slop as well? 19:20:37 the xrandr reference makes no sense at all 19:20:42 The diagnostic is slop 19:21:20 The rest looks plausible but they've bypassed the reportbug script somehow :-/ 19:21:55 does the graphical installer also use the amdgpu driver? 19:22:15 I don't think so 19:23:00 Let's ask for lspci (+ whatever options we use in the bug script) and kernel log 19:23:08 let them test the backports kernel? 19:23:09 Is there a way to check whether the sender used the reportbug tool in the debian mailing list? 19:23:45 yunseong[m]: I don't understand the idea 19:23:50 Pretty sure this comes through reportbug, but they reported against "linux" and not the right binary package 19:23:52 X-Mailer: reportbug 13.2.0 19:24:09 Maybe we should ask the reportbug maintianers to map "linux" to linux-image-$(uname -r), the same as it does for "kernel" 19:24:16 so yes, they used reportbug, but neither the correct package nor version 19:24:34 can't we provide /usr/share/bugs/linux? 19:25:12 anyway, what to do? 19:25:35 I guess we should go with: 19:25:36 < bwh> Let's ask for lspci (+ whatever options we use in the bug script) and kernel log 19:25:41 ukleinek: Aha, It seems that many bug reports are being generated with AI rather than submitted via reportbug, and therefore do not include the necessary attachments.. 19:25:43 and then see what happens 19:26:02 I can do 19:26:24 sounds good, thanks. Maybe ask for a test for 7.whateverwecurrentlyhaveinbackports 19:26:32 ack 19:27:10 #action carnil responds to #1143269 asking for lspci (with options from bug scripts), kernel log, plus testing with newer versions from backports 19:27:17 thx 19:27:25 #topic #1143500: (i, u) The Lenovo Legion Y9000X 2021 (Realtek ALC287 + Cirrus CS35L41) has no sound output through external speakers, but the headphones work normally 19:27:37 #action bwh will look at making "reportbug linux" DTRT 19:28:24 "I was unable to 19:28:26 reproduce this bug using the Debian version of Plasma." 19:29:18 so the problem is in KDE newer than packaged for stable and not the kernel? 19:30:04 ukleinek: Not sure which bug you are reading that in? 19:30:30 This one seems like legit missing hardware support 19:30:33 right, not 1143500 19:31:04 I suppose we should ask to test a backports kernel 19:32:08 ok 19:33:29 no volunters 19:33:30 * ukleinek agrees, sounds like it needs a quirk 19:33:30 I will ask 19:34:04 #action carnil ask reporter of #1143500 if newer kernel from backports properly support the device 19:34:10 #topic #1139265: (n, M) linux-image-7.0.10+deb14-amd64: QXL graphics crash, processes stuck in D state, NULL-dereference errors, TTY-subsystem malfunction 19:34:35 this is the bug I looked at before 19:34:42 possible fix found. submitter will test when it landed 19:34:44 yes 19:34:45 right 19:35:04 #topic #1141869: (n, M) linux-image-6.12.95+deb13-amd64: USB-C monitor that worked on Debian 12 fails on Debian 13 with "no signal" 19:35:26 The reporter needs help with bisection 19:35:49 yes seen the reply today, I can followup 19:36:05 I'll remove the moreinfo tag for now 19:36:53 yes 19:37:30 #topic #1142419: (n, u) 7.1.3 does not resume "i915 0000:00:02.0: [drm] pipe state doesn't match!" 19:37:40 Waiting on upstream 19:37:58 #topic #1142489: (n, M) linux-image-7.1.3+deb14-amd64: nouveau 0000:01:00.0: Xorg[1491]: failed to idle channel 4 [Xorg[1491]] 19:38:14 waiting on submitter 19:38:17 right 19:38:23 #topic #1142626: (n, M) linux-image-6.12.96+deb13-amd64: Debian 13 kernel has poor disk I/O for Windows guest on QEMU 19:38:51 waiting on submitter 19:38:51 the #1142489 submitter is using xorg and should consider wayland? 19:40:56 #topic #1142881: (n, ) linux-image-armmp > 6.1 boot error UUID=... or LABEL=... does not exist but exists 19:41:35 This is either a necessary driver missing from the armmp config, or from the initramfs 19:42:33 #action ukleinek looks into #1142881 and tries to find the missing driver 19:42:43 Thank you 19:42:54 #topic #1142912: (n, +) linux-image-6.12.96+deb13-amd64: CONFIG_MISC_ALCOR_PCI not set, no driver for Alcor Micro AU6625 card reader 19:43:23 03linux 05debian/latest 06Ben Hutchings * [update] merge request !2039: Add support for Alcor Micro AU6625 PCI-E Flash card reader * 14https://salsa.debian.org/kernel-team/linux/-/merge_requests/2039 19:43:57 I just tagged that for backporting 19:44:12 I can check and merge it later this evening 19:44:17 * ukleinek wonders about (Closes: ...) twice in the MR 19:44:34 #topic #1143143: (n, M) RAID10 rebuild high CPU low throughput after upgrade from Bookworm 19:45:10 no information, submitter asked to recheck with latest version. nothing to do 19:45:21 * ukleinek nods 19:45:27 #topic #1143161: (n, +M) linux-image-6.12.96+deb13-amd64: X does not start on DisplayPort of NVIDIA Quadro 600 with nouveau 19:45:48 waldi: Raptor only made one model of computer AFAIK :-) 19:46:23 wait on submitter communicating with upstream 19:46:32 OK 19:46:38 #topic #1143255: (n, M) Bug in kernel 6.12.100 19:46:51 Great bug subject 19:47:02 asolutely :) 19:47:10 + "NVIDIA proprietary driver" 19:47:26 yes. wasn't there a problem to build the nvidia stuff with 6.12.100? 19:47:41 + repro requires proprietary game 19:48:03 waldi: yes for nvidia-open-kernel-dkms 19:48:27 "dear reporter, please provide a license for that game to the kernel team" 19:49:47 submitter was asked to provide more information, so wait 19:50:03 #topic #1143450: (n, ) linux-image-6.12.94+deb13-amd64: hibernate on toughbook CF-C2 does not power off fully 19:50:31 submitter was asked to provide more information, so wait 19:50:41 #topic #1143506: (n, ) linux-image-7.1.3+deb13-amd64: fails to complete shutdown after extended uptime, session-2.scope stop timeout 19:51:09 bug text in X-Debbugs-Cc is a new variant 19:51:52 This might be related to (one of) the bugs in the report we wanted to split 19:52:09 or to the external modules 19:52:14 yes 19:52:31 a whole lot of vendor magic 19:53:06 Is it worth asking to test without those, or should we suggest reporting to Clevo? 19:53:18 did anyone find a reference to that session-2.scope somwhere in a log? 19:54:16 +1 for testing without clevo OOT modules 19:54:45 #action waldi responds to #1143506, asking for test without external modules or report to clevo 19:54:59 #topic #1143545: (n, ) kernel: performance regression in 6.12.x series kernel 19:55:38 we had a disk speed regression already before, right? 19:56:15 Yes, #1143143 19:56:31 giben the reporter claims to have found the regression introducing commit and fixing commit (not backported to 6.12.y) ask to test and let him ask stable list to backport the commit? 19:56:51 s/giben/given/ 19:57:00 and maybe let the reporter of #1143143 test /sys/kernel/mm/lru_gen/enabled=0 19:57:16 but I really doubt these are related 19:57:44 this is specifically about VM paging 19:58:01 I wouldn't be surprised and the test is trivial 19:58:16 RAID syncing should not have anything to do with that 19:59:02 * ukleinek doesn't understand what the sysfs file does, so I believe you 19:59:50 what is surprising 20:00:05 the 14aa8b2d5c2e ("mm/mglru: don't sync disk for each aging cycle") commit is in v6.1-rc1 20:00:20 so how the claim "After upgrading from 6.1 series kernel to 6.12.x series kernel [...]"? 20:00:39 ah maybe we enabled MGLRU only later 20:01:02 14aa8b2d5c2e is the designated regression introducer 20:01:24 v6.1-rc1: 14aa8b2d5c2ebead01b542f62d68029023054774 mm/mglru: don't sync disk for each aging cycle 20:01:43 linux (6.4~rc7-1~exp1) experimental; urgency=medium 20:01:43 * mm: Enable Multi-Gen LRU implementation (by default) (Closes: #1030617) 20:04:41 so I sill think: ask the reporter to confirm he tested the commit to fix the issue, and secondly let him ask stable maintainers to pick the backport? I'm not sure we should take reponsibility for backporting mm/ changes, this should be validated by Andrew Morton and other involved in the change IMHO 20:04:52 (and keep us looped in) 20:05:00 so we can close the bug once it lands in a stable eries 20:05:07 in a new 6.12.y stable series version 20:05:18 #action bwh will ask for backport to fix #1143545 20:05:29 thx 20:05:50 #topic AoB 20:05:58 anything else we should talk about? 20:06:06 * ukleinek can chair next week 20:06:24 #action ukleinek chairs next week 20:06:47 thanks 20:06:59 thank you all for attending 20:07:02 #endmeeting