Comment 327 for bug 1801540

FYI, the 'solution' for this issue that is now in the kernel breaks capturing screen+currrently playing audio with ffmpeg, for me at least, on AMD Raven hardware.

I tried many things but could not achieve good audio capture that is synchronised, not when capturing from Pulse anyway.

I'm guessing it's being in BATCH mode, and Pulse not operating with tsched, that is the problem, not the rest of the patchwork.

Nonetheless I just wanted things to work as before without any doubt, so I now have my kernel patched with the following (restoring 1022:15e3 to use the parameter sets it used to before all this) and everything is working fine now:

diff -Naur a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c
--- a/sound/pci/hda/hda_intel.c 2021-02-10 19:29:23.000000000 +1100
+++ b/sound/pci/hda/hda_intel.c 2021-02-14 02:42:38.714796278 +1100
@@ -2601,7 +2601,8 @@
     AZX_DCAPS_PM_RUNTIME },
  /* AMD Raven */
  { PCI_DEVICE(0x1022, 0x15e3),
- .driver_data = AZX_DRIVER_GENERIC | AZX_DCAPS_PRESET_AMD_SB },
+ .driver_data = AZX_DRIVER_GENERIC | AZX_DCAPS_PRESET_ATI_SB |
+ AZX_DCAPS_PM_RUNTIME },
  /* ATI HDMI */
  { PCI_DEVICE(0x1002, 0x0002),
    .driver_data = AZX_DRIVER_ATIHDMI_NS | AZX_DCAPS_PRESET_ATI_HDMI_NS },

--
While I greatly appreciate the work on the microphone input crackling issue (which I have experienced and been annoyed by myself over the years), i'm concerned that the current solution should've been further tested and developed before being mainlined.