From: Guennadi Liakhovetski guennadi.liakhovetski@linux.intel.com
If HDA probing on Intel platforms fails, all the devices, that have previously been powered up, have to be suspended again.
Signed-off-by: Guennadi Liakhovetski guennadi.liakhovetski@linux.intel.com ---
I came across HDA not suspending when no codec driver has successfully probed while working on SOF. Then I decided to look how this is handled in the mainline Intel HDA code and I think something like this patch is needed, however, I don't have a set up to test it. I haven't looked at the tegra HDA implementation, possibly they need something similar. Please, comment.
sound/pci/hda/hda_intel.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index e784130..d3caa37 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -2229,14 +2229,14 @@ static int azx_probe_continue(struct azx *chip) /* create codec instances */ err = azx_probe_codecs(chip, azx_max_codecs[chip->driver_type]); if (err < 0) - goto out_free; + goto out_power_down;
#ifdef CONFIG_SND_HDA_PATCH_LOADER if (chip->fw) { err = snd_hda_load_patch(&chip->bus, chip->fw->size, chip->fw->data); if (err < 0) - goto out_free; + goto out_power_down; #ifndef CONFIG_PM release_firmware(chip->fw); /* no longer needed */ chip->fw = NULL; @@ -2246,12 +2246,12 @@ static int azx_probe_continue(struct azx *chip) if ((probe_only[dev] & 1) == 0) { err = azx_codec_configure(chip); if (err < 0) - goto out_free; + goto out_power_down; }
err = snd_card_register(chip->card); if (err < 0) - goto out_free; + goto out_power_down;
setup_vga_switcheroo_runtime_pm(chip);
@@ -2260,6 +2260,7 @@ static int azx_probe_continue(struct azx *chip)
set_default_power_save(chip);
+out_power_down: if (azx_has_pm_runtime(chip)) pm_runtime_put_autosuspend(&pci->dev);