[alsa-devel] Aplay Fails When Reading From stdin

[Jay Foster]

On 5/15/2018 1:21 PM, Takashi Iwai wrote:
> On Mon, 07 May 2018 17:37:44 +0200,
> Jay Foster wrote:
>>
>>
>> On 5/4/2018 11:51 PM, Takashi Iwai wrote:
>>> On Fri, 04 May 2018 21:57:35 +0200,
>>> Jay Foster wrote:
>>>>
>>>> On 5/4/2018 12:47 PM, Takashi Iwai wrote:
>>>>> On Fri, 04 May 2018 19:51:05 +0200,
>>>>> Jay Foster wrote:
>>>>>> I recently updated may alsa from 1.1.4.1 to 1.1.6.  I now noticed that
>>>>>> aplay does not work properly (the same as 1.1.4.1) when reading the
>>>>>> sound file data from stdin.  This might have something to do with the
>>>>>> recent change with reading the sound file header.
>>>>>>
>>>>>> With the previous version of aplay, aplay would report:
>>>>>>
>>>>>> Playing WAVE 'stdin' : Signed 16 bit Little Endian, Rate 22050 Hz, Mono
>>>>>>
>>>>>> With 1.1.6, aplay reports:
>>>>>>
>>>>>> Playing raw data 'stdin' : Unsigned 8 bit, Rate 8000 Hz, Mono
>>>>>>
>>>>>> This results in static/noise output.  If I explicitly add the '-f
>>>>>> S16_LE -r 22050' options to aplay, then it does play correctly.
>>>>>>
>>>>>> Is this a known bug and is there a fix?
>>>>> It works fine on my system.
>>>>>
>>>>> % aplay test.wav
>>>>> Playing WAVE 'test.wav' : Signed 16 bit Little Endian, Rate 44100 Hz, Stereo
>>>>>
>>>>> % aplay < ~/test/test.wav
>>>>> Playing WAVE 'stdin' : Signed 16 bit Little Endian, Rate 44100 Hz, Stereo
>>>>>
>>>>> % aplay --version
>>>>> aplay: version 1.1.6 by Jaroslav Kysela <perex at perex.cz>
>>>>>
>>>>>
>>>>> Takashi
>>>>>
>>>>>
>>>> Well, it does not work for me:
>>>>
>>>> cat bird-calls.wav | /usr/bin/aplay -Dplughw:0 --
>>>> Playing raw data 'stdin' : Unsigned 8 bit, Rate 8000 Hz, Mono
>>>> ^CAborted by signal Interrupt...
>>>> aplay: pcm_write:2051: write error: Interrupted system call
>>>>
>>>> aplay -Dplughw:0 bird-calls.wav
>>>> Playing WAVE 'bird-calls.wav' : Signed 16 bit Little Endian, Rate
>>>> 44100 Hz, Stereo
>>>>
>>>> Looks like aplay may now have problems reading from stdin if stdin is
>>>> a pipe.
>>> OK, now I see it.  It's in the safety check of file stat.
>>> The fix patch is below.
>>>
>> Thanks.  That works.
> Actually the fix below would be much simpler and better.
> I'm going to apply it.
>
>
> Takashi
>
> -- 8< --
> From: Takashi Iwai <tiwai at suse.de>
> Subject: [PATCH] aplay: Fix invalid file size check for non-regular files
>
> aplay tries to check the file size via fstat() at parsing the format
> headers and avoids parsing when the size is shorter than the given
> size.  This works fine for regular files, but when a special file like
> pipe is passed, it fails, eventually leading to the fallback mode
> wrongly.
>
> A proper fix is to do this sanity check only for a regular file.
>
> Reported-by: Jay Foster <jay at systech.com>
> Signed-off-by: Takashi Iwai <tiwai at suse.de>
> ---
>   aplay/aplay.c | 3 ++-
>   1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/aplay/aplay.c b/aplay/aplay.c
> index bbd7fffa04fc..63ec9efbebc1 100644
> --- a/aplay/aplay.c
> +++ b/aplay/aplay.c
> @@ -2821,7 +2821,8 @@ static int read_header(int *loaded, int header_size)
>   
>   	/* don't be adventurous, get out if file size is smaller than
>   	 * requested header size */
> -	if (buf.st_size < header_size)
> +	if ((buf.st_mode & S_IFMT) == S_IFREG &&
> +	    buf.st_size < header_size)
>   		return -1;
>   
>   	if (*loaded < header_size) {
Thanks.  Your new patch also works.
Jay