Date: Thu, 23 Jan 2014 17:29:37 +0100
Message-ID: <s5h1tzyu0em.wl%tiwai@suse.de> (sfid-20140123_172943_295420_43D95950)
From: Takashi Iwai <tiwai@suse.de>
To: LANGLOIS Olivier PIS -EXT
	<olivier.pis.langlois@transport.alstom.com>
Cc: Olivier Langlois <olivier@trillion01.com>,
	"alsa-devel@alsa-project.org" <alsa-devel@alsa-project.org>,
	"linux-wireless@vger.kernel.org" <linux-wireless@vger.kernel.org>,
	"James Cameron" <quozl@laptop.org>,
	"daniel@zonque.org" <daniel@zonque.org>
Subject: Re: [alsa-devel] [ISSUE] rtl8192ce appears to interfere with	ALSA	playback
In-Reply-To: <89F8E4011FD7234ABFE48744E300CBC206BCE732@041-DB3MPN1-092.041d.mgd.msft.net>
References: <1389767820.2436.45.camel@Wailaba2>
	<52D6BEEF.5050705@lwfinger.net>
	<52D76D56.6060906@trillion01.com>
	<20140116055547.GU19393@us.netrek.org>
	<1390464800.2436.111.camel@Wailaba2>
	<s5htxcvt718.wl%tiwai@suse.de>
	<89F8E4011FD7234ABFE48744E300CBC206BCE732@041-DB3MPN1-092.041d.mgd.msft.net>
MIME-Version: 1.0 (generated by SEMI 1.14.6 - "Maruoka")
Content-Type: text/plain; charset=US-ASCII
Sender: linux-wireless-owner@vger.kernel.org

At Thu, 23 Jan 2014 16:15:19 +0000,
LANGLOIS Olivier PIS -EXT wrote:
> 
> Takashi,
> 
> > > 2. I was expecting to see XRUN entries in dmesg but I didn't. The
> > > closest that I have is [ 1700.756017] PCM: Lost interrupts? [Q]
> > > (stream=0, delta=16108, new_hw_ptr=383756, old_hw_ptr=367648) 3. I
> > > didn't realise that the default ALSA device was a plugin using dmix. I
> > > have no underruns by using directly the ALSA HW device (hw:0,0)
> >
> > The dmix uses the fixed configuration.  Did you try playback using hw but
> > with the same configuration (period size, buffer size, format)?
> >
> this is a very good observation.
> 
> With the same buffer/period sizes at a sampling rate of 44100 (this slightly modify the buffer/period times), I have 0 underruns. As soon as I resample my wav file to 48kHz, boom the underruns happens.
> 
> So if I understand correctly by reducing the buffer time correctly from 371504 usec to 341328 usec and by assuming that the buffer is full, if the snd_hda_intel irq/timer are delayed to a greater amount of time than 341328 usec, this is what is causing the underruns.

Which machine and codec?  For example, Lenovo IdeaPad with ALC269 is
known to be broken with 48kHz, so we have a fixed rate limitation in
the driver.

Other than the hardware restriction, you may play more with
position_fix or bdl_pos_adj option of snd-hda-intel driver.


Takashi


> 
> Maybe it is important to the discussion that my CONFIG_HZ=1000 and
> 
> $ zgrep PREEMPT /proc/config.gz
> CONFIG_TREE_PREEMPT_RCU=y
> CONFIG_PREEMPT_RCU=y
> # CONFIG_PREEMPT_NONE is not set
> # CONFIG_PREEMPT_VOLUNTARY is not set
> CONFIG_PREEMPT=y
> CONFIG_PREEMPT_COUNT=y
> # CONFIG_DEBUG_PREEMPT is not set
> # CONFIG_PREEMPT_TRACER is not set
> 
> $ aplay -Dhw:0,0 --period-size=1024 --buffer-size=16384 -v sine.wav
> Playing WAVE 'sine.wav' : Signed 16 bit Little Endian, Rate 44100 Hz, Stereo
> Hardware PCM card 0 'HDA Intel' device 0 subdevice 0
> Its setup is:
>   stream       : PLAYBACK
>   access       : RW_INTERLEAVED
>   format       : S16_LE
>   subformat    : STD
>   channels     : 2
>   rate         : 44100
>   exact rate   : 44100 (44100/1)
>   msbits       : 16
>   buffer_size  : 16384
>   period_size  : 1024
>   period_time  : 23219
>   tstamp_mode  : NONE
>   period_step  : 1
>   avail_min    : 1024
>   period_event : 0
>   start_threshold  : 16384
>   stop_threshold   : 16384
>   silence_threshold: 0
>   silence_size : 0
>   boundary     : 1073741824
>   appl_ptr     : 0
>   hw_ptr       : 0
> 
> $ aplay -Dhw:0,0 --period-size=1024 --buffer-size=16384 -v sine_48000.wav
> Playing WAVE 'sine_48000.wav' : Signed 16 bit Little Endian, Rate 48000 Hz, Stereo
> Hardware PCM card 0 'HDA Intel' device 0 subdevice 0
> Its setup is:
>   stream       : PLAYBACK
>   access       : RW_INTERLEAVED
>   format       : S16_LE
>   subformat    : STD
>   channels     : 2
>   rate         : 48000
>   exact rate   : 48000 (48000/1)
>   msbits       : 16
>   buffer_size  : 16384
>   period_size  : 1024
>   period_time  : 21333
>   tstamp_mode  : NONE
>   period_step  : 1
>   avail_min    : 1024
>   period_event : 0
>   start_threshold  : 16384
>   stop_threshold   : 16384
>   silence_threshold: 0
>   silence_size : 0
>   boundary     : 1073741824
>   appl_ptr     : 0
>   hw_ptr       : 0
> underrun!!! (at least 0.075 ms long)
> Status:
>   state       : XRUN
>   trigger_time: 2155.298508067
>   tstamp      : 2155.299242462
>   delay       : 0
>   avail       : 16815
>   avail_max   : 16815
> 
> 
> ________________________________
> CONFIDENTIALITY : This e-mail and any attachments are confidential and may be privileged. If you are not a named recipient, please notify the sender immediately and do not disclose the contents to another person, use it for any purpose or store or copy the information in any medium.
>