Created attachment 300588 [details] SND_HDA_* excrept from kernel config On PCI device discovery the Intel HDA codec doesn't bind to appropriate codec (if that is the correct term). The device codec (Conexant) is loaded, but sound doesn't work. Reloading the main module (snd_hda_intel) fixes the issue: $ echo 1 > /sys/bus/pci/devices/0000:00:1b.0/remove $ lsmod | sort Module Size Used by configs 45056 0 e1000e 225280 0 grace 16384 1 lockd irqbypass 16384 1 kvm iwlwifi 274432 0 kvm 569344 1 kvm_intel kvm_intel 229376 0 lockd 98304 1 nfs mei 90112 1 mei_me mei_me 28672 0 nfs 184320 3 nfsv4 nfsv4 380928 2 sunrpc 274432 9 nfsv4,lockd,nfs uvcvideo 110592 0 videobuf2_common 45056 4 videobuf2_vmalloc,videobuf2_v4l2,uvcvideo,videobuf2_memops videobuf2_memops 16384 1 videobuf2_vmalloc videobuf2_v4l2 24576 1 uvcvideo videobuf2_vmalloc 16384 1 uvcvideo $ echo 1 > /sys/bus/pci/rescan # dmesg output [ 9218.563134] pci 0000:00:1b.0: [8086:3b56] type 00 class 0x040300 [ 9218.563178] pci 0000:00:1b.0: reg 0x10: [mem 0xc0004000-0xc0007fff 64bit] [ 9218.563344] pci 0000:00:1b.0: PME# supported from D0 D3hot D3cold [ 9218.563818] pci 0000:00:1b.0: BAR 0: assigned [mem 0xc0004000-0xc0007fff 64bit] [ 9218.563846] pci 0000:00:1e.0: PCI bridge to [bus 0e] [ 9218.598927] snd_hda_intel 0000:00:1b.0: bound 0000:00:02.0 (ops 0xffffffff824a44e0) [ 9218.609798] snd_hda_intel 0000:00:1b.0: Cannot probe codecs, giving up $ lsmod | sort Module Size Used by configs 45056 0 e1000e 225280 0 grace 16384 1 lockd irqbypass 16384 1 kvm iwlwifi 274432 0 kvm 569344 1 kvm_intel kvm_intel 229376 0 lockd 98304 1 nfs mei 90112 1 mei_me mei_me 28672 0 nfs 184320 3 nfsv4 nfsv4 380928 2 snd 77824 7 snd_hda_codec_generic,snd_hda_codec_conexant,snd_hwdep,snd_hda_intel,snd_hda_codec,snd_timer,snd_pcm snd_hda_codec 118784 3 snd_hda_codec_generic,snd_hda_codec_conexant,snd_hda_intel snd_hda_codec_conexant 20480 0 snd_hda_codec_generic 77824 1 snd_hda_codec_conexant snd_hda_core 65536 4 snd_hda_codec_generic,snd_hda_codec_conexant,snd_hda_intel,snd_hda_codec snd_hda_intel 36864 0 snd_hwdep 16384 1 snd_hda_codec snd_intel_dspcfg 16384 1 snd_hda_intel snd_pcm 110592 3 snd_hda_intel,snd_hda_codec,snd_hda_core snd_timer 32768 1 snd_pcm soundcore 16384 1 snd sunrpc 274432 9 nfsv4,lockd,nfs uvcvideo 110592 0 videobuf2_common 45056 4 videobuf2_vmalloc,videobuf2_v4l2,uvcvideo,videobuf2_memops videobuf2_memops 16384 1 videobuf2_vmalloc videobuf2_v4l2 24576 1 uvcvideo videobuf2_vmalloc 16384 1 uvcvideo $ modprobe -rv snd_hda_intel; modprobe -v snd_hda_intel # dmesg output [ 9230.886362] snd_hda_intel 0000:00:1b.0: bound 0000:00:02.0 (ops 0xffffffff824a44e0) [ 9230.895721] snd_hda_codec_conexant hdaudioC0D0: CX20585: BIOS auto-probing. [ 9230.896677] snd_hda_codec_conexant hdaudioC0D0: autoconfig for CX20585: line_outs=1 (0x1f/0x0/0x0/0x0/0x0) type:speaker [ 9230.896691] snd_hda_codec_conexant hdaudioC0D0: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0) [ 9230.896700] snd_hda_codec_conexant hdaudioC0D0: hp_outs=2 (0x1c/0x19/0x0/0x0/0x0) [ 9230.896707] snd_hda_codec_conexant hdaudioC0D0: mono: mono_out=0x0 [ 9230.896713] snd_hda_codec_conexant hdaudioC0D0: inputs: [ 9230.896718] snd_hda_codec_conexant hdaudioC0D0: Internal Mic=0x23 [ 9230.896724] snd_hda_codec_conexant hdaudioC0D0: Mic=0x1b [ 9230.896729] snd_hda_codec_conexant hdaudioC0D0: Dock Mic=0x1a [ 9230.908437] snd_hda_codec_generic hdaudioC0D3: autoconfig for Generic: line_outs=0 (0x0/0x0/0x0/0x0/0x0) type:line [ 9230.908454] snd_hda_codec_generic hdaudioC0D3: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0) [ 9230.908461] snd_hda_codec_generic hdaudioC0D3: hp_outs=0 (0x0/0x0/0x0/0x0/0x0) [ 9230.908468] snd_hda_codec_generic hdaudioC0D3: mono: mono_out=0x0 [ 9230.908473] snd_hda_codec_generic hdaudioC0D3: dig-out=0x5/0x0 [ 9230.908477] snd_hda_codec_generic hdaudioC0D3: inputs: [ 9230.916063] input: HDA Intel MID Mic as /devices/pci0000:00/0000:00:1b.0/sound/card0/input52 [ 9230.916216] input: HDA Intel MID Dock Mic as /devices/pci0000:00/0000:00:1b.0/sound/card0/input53 [ 9230.916353] input: HDA Intel MID Dock Headphone as /devices/pci0000:00/0000:00:1b.0/sound/card0/input54 [ 9230.919563] input: HDA Intel MID Headphone as /devices/pci0000:00/0000:00:1b.0/sound/card0/input55 [ 9230.919727] input: HDA Intel MID HDMI as /devices/pci0000:00/0000:00:1b.0/sound/card0/input56 $ lsmod | sort Module Size Used by configs 45056 0 e1000e 225280 0 grace 16384 1 lockd irqbypass 16384 1 kvm iwlwifi 274432 0 kvm 569344 1 kvm_intel kvm_intel 229376 0 lockd 98304 1 nfs mei 90112 1 mei_me mei_me 28672 0 nfs 184320 3 nfsv4 nfsv4 380928 2 snd 77824 7 snd_hda_codec_generic,snd_hda_codec_conexant,snd_hwdep,snd_hda_intel,snd_hda_codec,snd_timer,snd_pcm snd_hda_codec 118784 3 snd_hda_codec_generic,snd_hda_codec_conexant,snd_hda_intel snd_hda_codec_conexant 20480 1 snd_hda_codec_generic 77824 2 snd_hda_codec_conexant snd_hda_core 65536 4 snd_hda_codec_generic,snd_hda_codec_conexant,snd_hda_intel,snd_hda_codec snd_hda_intel 36864 0 snd_hwdep 16384 1 snd_hda_codec snd_intel_dspcfg 16384 1 snd_hda_intel snd_pcm 110592 3 snd_hda_intel,snd_hda_codec,snd_hda_core snd_timer 32768 1 snd_pcm soundcore 16384 1 snd sunrpc 274432 9 nfsv4,lockd,nfs uvcvideo 110592 0 videobuf2_common 45056 4 videobuf2_vmalloc,videobuf2_v4l2,uvcvideo,videobuf2_memops videobuf2_memops 16384 1 videobuf2_vmalloc videobuf2_v4l2 24576 1 uvcvideo videobuf2_vmalloc 16384 1 uvcvideo 5.15.29, have observed it already at least on 5.10.27, HW is ThinkPad x201 (x220 as well). I'm marking it as regression, since it used to work in the distant past (5.4-ish at least iirc)
Note that the used by counter for both snd_hda_codec_conexant and snd_hda_codec_generic got increased after reloading snd_hda_intel.
Could you try to bisect? At least it'd be helpful to know which kernel release starts showing that problem.
(In reply to Takashi Iwai from comment #2) > Could you try to bisect? At least it'd be helpful to know which kernel > release starts showing that problem. Will try, but it's going to take a while. In the meantime, any suggestions where this could go wrong on the system configuration side (i.e. PEBKAC)?
(In reply to Petr Cerny from comment #3) > In the meantime, any suggestions where this could go wrong on the system > configuration side (i.e. PEBKAC)? Honestly speaking, I have no idea. As the module probe succeeds, it might be outside the sound driver itself but in a deeper level.
(In reply to Petr Cerny from comment #3) > (In reply to Takashi Iwai from comment #2) > > Could you try to bisect? At least it'd be helpful to know which kernel > > release starts showing that problem. > > Will try, but it's going to take a while. Were you able to get to it, and pinpoint the problem?