Bugzilla – Bug 1213745
cirrus logic speaker (cs35l41-hda, sof-firmware) not working after resuming from suspend
Last modified: 2024-06-25 17:52:14 UTC
Created attachment 868475 [details] dmesg output after a "fresh boot -> suspend-resume -> suspend-resume" sequence (i clicked the sound check in each time in gnome control center) Dear OpenSUSE development team, The cirrus logic speaker on my "hp zbook fury g9" laptop (a relatively new model) will stop working after the first suspend after system boot, but will work again after second/third/... suspend, though the sound volume will get significantly lower. By the way, when doing research on the internet I found Mr. Takashi Iwai's communication with a cirrus developer on LKML: https://lkml.org/lkml/2023/7/21/901 It seems that the problem will be fixed with the next kernel update, but I'm not really sure about that due to my non-technical background. And if that is so, could I ask for an rough estimation of the date for this relevant kernel update to arrive in tumbleweed? Thanks for your time and the great work.
I'm building a test kernel with the backport of those patches on top of the current TW stable git branch. It's being built in OBS home:tiwai:bsc1213745. Once after the build finishes (takes an hour or so), it'll appear at http://download.opensuse.org/repositories/home:/tiwai:/bsc1213745/standard/ Please give it a try later. Note that it's an unofficial build, hence you'd need to disable Secure Boot.
(In reply to Takashi Iwai from comment #1) > I'm building a test kernel with the backport of those patches on top of the > current TW stable git branch. It's being built in OBS home:tiwai:bsc1213745. > Once after the build finishes (takes an hour or so), it'll appear at > http://download.opensuse.org/repositories/home:/tiwai:/bsc1213745/standard/ > Please give it a try later. > > Note that it's an unofficial build, hence you'd need to disable Secure Boot. I've tried the new kernel, and can confirm the problem has been fixed. Thanks! By the way, I can see even with the new kernel, cs35l41-hda is still causing trouble for the system to successfully enter hibernation (which is another mild bug I hadn't reported). Do you by any chance know something about this related issue?
It turns out that the new test kernel is not working perfectly. When I do the first suspend and resume, everything works fine, and no errors in the dmesg output. But if I suspend the system again, it would freeze, and overheat. The same pattern will repeat after forced reboot.
Is it related with this patch set? You can check the latest kernel in OBS Kernel:stable as comparison. If it comes from the cs35l41 patches, we should ask Cirrus people to take a look.
(In reply to Takashi Iwai from comment #4) > Is it related with this patch set? You can check the latest kernel in OBS > Kernel:stable as comparison. > > If it comes from the cs35l41 patches, we should ask Cirrus people to take a > look. I believe the freeze is related to this patch. I tried the stable kernel again, the freeze during the second suspend attempt after system boot didn't happen (but of course the speaker bug came back). I'll add the error messages shown during the freeze in the attachment.
Created attachment 868481 [details] freeze during second suspend
Thanks, it looks suspicious, yes. I'll try to ping Cirrus people.
(In reply to Takashi Iwai from comment #7) > Thanks, it looks suspicious, yes. > > I'll try to ping Cirrus people. Hi Takashi. Is the above mentioned patch ( https://lkml.org/lkml/2023/7/21/901 ) actually applied to the kernels released afterwards? Thanks.
Yes, it's in Linus tree for 6.6-rc1. We'll have to fix this before 6.6 final release.
Care to check whether you still see the problem with the latest kernel in OBS Kernel:HEAD repo?
(In reply to Takashi Iwai from comment #10) > Care to check whether you still see the problem with the latest kernel in > OBS Kernel:HEAD repo? Thanks! I'll test it today.
Can you retest 6.7-rc1 kernel in OBS Kernel:HEAD repo? Now all relevant patches have been merged there.
(In reply to Takashi Iwai from comment #12) > Can you retest 6.7-rc1 kernel in OBS Kernel:HEAD repo? > Now all relevant patches have been merged there. Wow, it works perfectly with this kernel!!!
Good to hear! Then I'm going to backport the patches to 6.6.x kernel for TW. (Already backported to SLE15-SP6 branch :)
(In reply to Takashi Iwai from comment #14) > Good to hear! > Then I'm going to backport the patches to 6.6.x kernel for TW. > (Already backported to SLE15-SP6 branch :) Arigatou, Takashi san~~
I'm building a test kernel with backports. It's being built (again) in OBS home:tiwai:bsc1213745 repo. Once after the build finishes, please test it. Once after it's verified to work, I'll merge those to stable branch.
(In reply to Takashi Iwai from comment #16) > I'm building a test kernel with backports. It's being built (again) in OBS > home:tiwai:bsc1213745 repo. Once after the build finishes, please test it. > Once after it's verified to work, I'll merge those to stable branch. I can confirm that it works!
Now all pieces should have been merged. Feel free to reopen if you still encounter the problem. Thanks.