On Wed, May 27, 2015 at 10:26 AM, Takashi Iwai tiwai@suse.de wrote:

At Wed, 27 May 2015 15:31:02 +0200, Takashi Iwai wrote:

...

At Wed, 27 May 2015 09:23:17 -0400, Adam Goode wrote:

...

Hi,

I have this card:

05:00.0 Multimedia audio controller: C-Media Electronics Inc CMI8738/CMI8768 PCI Audio (rev 10) Subsystem: C-Media Electronics Inc CMI8738/C3DX PCI Audio Device Flags: bus master, stepping, medium devsel, latency 32, IRQ 19 I/O ports at d000 [size=256] Capabilities: [c0] Power Management version 2 Kernel driver in use: snd_cmipci Kernel modules: snd_cmipci

When I suspend/resume, the sound from the OPL3 is permanently distorted. It sounds like not all of the operators are active. I can fix this by doing a full PCI remove/rescan cycle: the driver correctly initializes the card at this time.

Any ideas before I peek into the code?

OPL3 stuff lacks of PM code completely. Maybe calling snd_opl3_init() again should enable the minimal stuff.

That said, a patch like below. The whole synth setup still needs to be reset, I suppose.

Takashi

---

diff --git a/sound/pci/cmipci.c b/sound/pci/cmipci.c index 6cf464d9043d..f03fd5f2e043 100644 --- a/sound/pci/cmipci.c +++ b/sound/pci/cmipci.c @@ -498,6 +498,8 @@ struct cmipci { /* external MIDI */ struct snd_rawmidi *rmidi;

•   struct snd_opl3 *opl3;
  

#ifdef SUPPORT_JOYSTICK struct gameport *gameport; #endif @@ -3005,6 +3007,7 @@ static int snd_cmipci_create_fm(struct cmipci *cm, long fm_port) dev_err(cm->card->dev, "cannot create OPL3 hwdep\n"); return err; }

•   cm->opl3 = opl3;
    return 0;
  
  

  disable_fm:

@@ -3386,6 +3389,8 @@ static int snd_cmipci_resume(struct device *dev) for (i = 0; i < ARRAY_SIZE(saved_mixers); i++) snd_cmipci_mixer_write(cm, saved_mixers[i], cm->saved_mixers[i]);

•   if (cm->opl3)
  

•           snd_opl3_init(cm->opl3);
    snd_power_change_state(card, SNDRV_CTL_POWER_D0);
    return 0;
  

}

Yes, it is better with this patch. I still get the same distorted sounds after resume, but a subsequent sbiload will restore the instruments correctly. Before the patch, sbiload had no effect and I had to remove/rescan the card..

Ideally, suspend/resume would restore the patches so that sbiload would not be necessary. But this is exactly as urgent as it seems. :)

Thanks,

Adam