Re: [PATCH RFC 16/26] ALSA: rme32: Convert to copy_silence ops

["René Rebe" <rene@exactcode.de>]

Hi,

On 18 Jul 2018, at 12:56, Takashi Iwai <tiwai@suse.de> wrote:

> On Wed, 18 Jul 2018 12:22:11 +0200,
> René Rebe wrote:
>> 
>> Hello everyone,
>> 
>> to have another digital audio i/o card for our studio / office, I got a pair of RME32 the other week from ebay. (mostly as reference to implement ADAT for the RAD1 Sgi/Octane ALSA driver, …)
>> 
>> Unfortunately they do not work with Linux. They are recognised and all the usual devices and /proc/… entries show up, however, the hardware pointer does not move during playback or capture no matter what clock source I choose.
>> I tried attaching coax s/pdif as well as an 8-channel Behringer Ultragain ADAT source w/ clock.
> 
> Does the /proc/asound/card*/rme32 entry show the right setup?

Well, I think it never showed external clock when I set it to external ADAT, right now, freshly booted (without the ext. ADAT) just trying to play stereo on the s/pdif it shows:

RME Digi32/8 (Rev. 101) at 0xec000000, irq 5 (index #1)

General settings
  Half-duplex mode
  receiver: CS8412
  format: 16 bit, Stereo

Input settings
  input: optical
  sample rate: no valid signal

Output settings
  output signal: normal playback (muted)
  sample rate: 44100 Hz
  sample clock source: Internal
  format: IEC958 (consumer)
  emphasis: off

Which should be right for internally word clock’ed s/pdif output. However, aplay does not play anything, after some 20? seconds or so of hanging it even prints:

aplay: pcm_write:2011: write error: Input/output error

While it is “trying to playback” /proc shows:

root@hostname:/proc/asound/card0# grep . pcm*p/sub0/*
pcm0p/sub0/hw_params:access: RW_INTERLEAVED
pcm0p/sub0/hw_params:format: S16_LE
pcm0p/sub0/hw_params:subformat: STD
pcm0p/sub0/hw_params:channels: 2
pcm0p/sub0/hw_params:rate: 44100 (44100/1)
pcm0p/sub0/hw_params:period_size: 2048
pcm0p/sub0/hw_params:buffer_size: 32768
pcm0p/sub0/info:card: 0
pcm0p/sub0/info:device: 0
pcm0p/sub0/info:subdevice: 0
pcm0p/sub0/info:stream: PLAYBACK
pcm0p/sub0/info:id: Digi32 IEC958
pcm0p/sub0/info:name: Digi32 IEC958
pcm0p/sub0/info:subname: subdevice #0
pcm0p/sub0/info:class: 0
pcm0p/sub0/info:subclass: 0
pcm0p/sub0/info:subdevices_count: 1
pcm0p/sub0/info:subdevices_avail: 0
pcm0p/sub0/status:state: RUNNING
pcm0p/sub0/status:owner_pid   : 753
pcm0p/sub0/status:trigger_time: 1195.174967384
pcm0p/sub0/status:tstamp      : 0.000000000
pcm0p/sub0/status:delay       : 32768
pcm0p/sub0/status:avail       : 0
pcm0p/sub0/status:avail_max   : 30720
pcm0p/sub0/status:-----
pcm0p/sub0/status:hw_ptr      : 10243
pcm0p/sub0/status:appl_ptr    : 43011
pcm0p/sub0/sw_params:tstamp_mode: NONE
pcm0p/sub0/sw_params:period_step: 1
pcm0p/sub0/sw_params:avail_min: 2048
pcm0p/sub0/sw_params:start_threshold: 32768
pcm0p/sub0/sw_params:stop_threshold: 32768
pcm0p/sub0/sw_params:silence_threshold: 0
pcm0p/sub0/sw_params:silence_size: 0
pcm0p/sub0/sw_params:boundary: 1073741824

I tried many kernels, down to 2.6.31, which I had initially on the box.
So unless some this hwparams never worked, this driver appears to not work a looong looooong time.

I doubt they are fully defect, as both behave the same and at least output “something” on Windows XP, ...

> RME32 seems to have only few registers, and it behaves differently for
> read and write.  Maybe you should try to watch the register 0x20000.
> The hwptr is the LSB 33 bits.

Yeah, guess I have to instrument this driver to track what is happening, maybe register not flushed to the hw?

> Takashi
> 
>> The two cards came from the same seller, look ok and both behave the same. 
>> I went so far to install a Windows XP test install where both cards work “more”.
>> (They are not perfect in windows, however, at least s/pdif can come out with or with-out external ADAT clock source. However, the digital signal strangely unclean, but that may be a bug in the window system sound device emulation, I only tested with foobar2k and not some Pro audio app which I do not really have).
>> 
>> Long story short, does someone still have such a card, or not moving hardware ptr (or missing interrupts?) does ring a bell regarding this RME generation? Instead of a working ADAT reference card, I apparently have two ALSA driver to hack on, … ;-)
>> 
>> I tested two different “PC” boards and the results were the same, too.
>> 
>> Thanks,
>> 	René
>> 
>> On 11 May 2017, at 23:09, Takashi Iwai <tiwai@suse.de> wrote:
>> 
>>> Replace the copy and the silence ops with the new merged ops.
>>> The conversion is straightforward with standard helper functions.
>>> 
>>> Signed-off-by: Takashi Iwai <tiwai@suse.de>
>>> ---
>>> sound/pci/rme32.c | 49 ++++++++++++++++++++++++-------------------------
>>> 1 file changed, 24 insertions(+), 25 deletions(-)
>>> 
>>> diff --git a/sound/pci/rme32.c b/sound/pci/rme32.c
>>> index 96d15db65dfd..d2b4a3ef0bd3 100644
>>> --- a/sound/pci/rme32.c
>>> +++ b/sound/pci/rme32.c
>>> @@ -253,41 +253,42 @@ static inline unsigned int snd_rme32_pcm_byteptr(struct rme32 * rme32)
>>> 		& RME32_RCR_AUDIO_ADDR_MASK);
>>> }
>>> 
>>> -/* silence callback for halfduplex mode */
>>> -static int snd_rme32_playback_silence(struct snd_pcm_substream *substream, int channel,	/* not used (interleaved data) */
>>> -				      snd_pcm_uframes_t pos,
>>> -				      snd_pcm_uframes_t count)
>>> -{
>>> -	struct rme32 *rme32 = snd_pcm_substream_chip(substream);
>>> -	count <<= rme32->playback_frlog;
>>> -	pos <<= rme32->playback_frlog;
>>> -	memset_io(rme32->iobase + RME32_IO_DATA_BUFFER + pos, 0, count);
>>> -	return 0;
>>> -}
>>> -
>>> /* copy callback for halfduplex mode */
>>> -static int snd_rme32_playback_copy(struct snd_pcm_substream *substream, int channel,	/* not used (interleaved data) */
>>> +static int snd_rme32_playback_copy(struct snd_pcm_substream *substream,
>>> +				   int channel,	/* not used (interleaved data) */
>>> 				   snd_pcm_uframes_t pos,
>>> -				   void __user *src, snd_pcm_uframes_t count)
>>> +				   void __user *src, snd_pcm_uframes_t count,
>>> +				   bool in_kernel)
>>> {
>>> 	struct rme32 *rme32 = snd_pcm_substream_chip(substream);
>>> 	count <<= rme32->playback_frlog;
>>> 	pos <<= rme32->playback_frlog;
>>> -	if (copy_from_user_toio(rme32->iobase + RME32_IO_DATA_BUFFER + pos,
>>> -			    src, count))
>>> +	if (!src)
>>> +		memset_io(rme32->iobase + RME32_IO_DATA_BUFFER + pos, 0, count);
>>> +	else if (in_kernel)
>>> +		memcpy_toio(rme32->iobase + RME32_IO_DATA_BUFFER + pos,
>>> +			    (void *)src, count);
>>> +	else if (copy_from_user_toio(rme32->iobase + RME32_IO_DATA_BUFFER + pos,
>>> +				     src, count))
>>> 		return -EFAULT;
>>> 	return 0;
>>> }
>>> 
>>> /* copy callback for halfduplex mode */
>>> -static int snd_rme32_capture_copy(struct snd_pcm_substream *substream, int channel,	/* not used (interleaved data) */
>>> +static int snd_rme32_capture_copy(struct snd_pcm_substream *substream,
>>> +				  int channel,	/* not used (interleaved data) */
>>> 				  snd_pcm_uframes_t pos,
>>> -				  void __user *dst, snd_pcm_uframes_t count)
>>> +				  void __user *dst, snd_pcm_uframes_t count,
>>> +				  bool in_kernel)
>>> {
>>> 	struct rme32 *rme32 = snd_pcm_substream_chip(substream);
>>> 	count <<= rme32->capture_frlog;
>>> 	pos <<= rme32->capture_frlog;
>>> -	if (copy_to_user_fromio(dst,
>>> +	if (in_kernel)
>>> +		memcpy_fromio((void *)dst,
>>> +			      rme32->iobase + RME32_IO_DATA_BUFFER + pos,
>>> +			      count);
>>> +	else if (copy_to_user_fromio(dst,
>>> 			    rme32->iobase + RME32_IO_DATA_BUFFER + pos,
>>> 			    count))
>>> 		return -EFAULT;
>>> @@ -1205,8 +1206,7 @@ static const struct snd_pcm_ops snd_rme32_playback_spdif_ops = {
>>> 	.prepare =	snd_rme32_playback_prepare,
>>> 	.trigger =	snd_rme32_pcm_trigger,
>>> 	.pointer =	snd_rme32_playback_pointer,
>>> -	.copy =		snd_rme32_playback_copy,
>>> -	.silence =	snd_rme32_playback_silence,
>>> +	.copy_silence =	snd_rme32_playback_copy,
>>> 	.mmap =		snd_pcm_lib_mmap_iomem,
>>> };
>>> 
>>> @@ -1219,7 +1219,7 @@ static const struct snd_pcm_ops snd_rme32_capture_spdif_ops = {
>>> 	.prepare =	snd_rme32_capture_prepare,
>>> 	.trigger =	snd_rme32_pcm_trigger,
>>> 	.pointer =	snd_rme32_capture_pointer,
>>> -	.copy =		snd_rme32_capture_copy,
>>> +	.copy_silence =	snd_rme32_capture_copy,
>>> 	.mmap =		snd_pcm_lib_mmap_iomem,
>>> };
>>> 
>>> @@ -1231,8 +1231,7 @@ static const struct snd_pcm_ops snd_rme32_playback_adat_ops = {
>>> 	.prepare =	snd_rme32_playback_prepare,
>>> 	.trigger =	snd_rme32_pcm_trigger,
>>> 	.pointer =	snd_rme32_playback_pointer,
>>> -	.copy =		snd_rme32_playback_copy,
>>> -	.silence =	snd_rme32_playback_silence,
>>> +	.copy_silence =	snd_rme32_playback_copy,
>>> 	.mmap =		snd_pcm_lib_mmap_iomem,
>>> };
>>> 
>>> @@ -1244,7 +1243,7 @@ static const struct snd_pcm_ops snd_rme32_capture_adat_ops = {
>>> 	.prepare =	snd_rme32_capture_prepare,
>>> 	.trigger =	snd_rme32_pcm_trigger,
>>> 	.pointer =	snd_rme32_capture_pointer,
>>> -	.copy =		snd_rme32_capture_copy,
>>> +	.copy_silence =	snd_rme32_capture_copy,
>>> 	.mmap =		snd_pcm_lib_mmap_iomem,
>>> };
>>> 
>>> -- 
>>> 2.12.2
>>> 
>>> _______________________________________________
>>> Alsa-devel mailing list
>>> Alsa-devel@alsa-project.org
>>> http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
>> 
>> -- 
>> ExactCODE GmbH, Lietzenburger Str. 42, DE-10789 Berlin
>> DE Legal: Amtsgericht Berlin (Charlottenburg) HRB 105123B, Tax-ID#: DE251602478
>> Managing Director: René Rebe
>> http://exactcode.com | http://exactscan.com | http://ocrkit.com | http://t2-project.org | http://rene.rebe.de
>> 

-- 
 ExactCODE GmbH, Lietzenburger Str. 42, DE-10789 Berlin
 DE Legal: Amtsgericht Berlin (Charlottenburg) HRB 105123B, Tax-ID#: DE251602478
 Managing Director: René Rebe
 http://exactcode.com | http://exactscan.com | http://ocrkit.com | http://t2-project.org | http://rene.rebe.de