Re: emu10k1 bug and patch (0.9.0rc7)

[Takashi Iwai <tiwai@suse.de>]

At Thu, 13 Feb 2003 21:30:17 +0100 (CET),
Jaroslav wrote:
> 
> On Mon, 10 Feb 2003, Arnaud de Bossoreille de Ribou wrote:
> 
> > So the bug looks like a signedness problem since sw_ready is unsigned
> > and there is a while(sw_ready > 0), which explain the constant delay,
> > next in the "snd_emu10k1_fx8010_playback_transfer" function.
> > 
> > So the emu10k1.patch file attached fixes the problem and seems not to
> > introduce new ones.
> 
> Please, could you try this patch, if it also fixes your problem? Thanks.

i don't think the patch fixes the original problem.
if there are only two periods, the diff will be either period_size or
-period_size, i.e. (buffer_size/2) or -(buffer_size/2).

assume the buffer size = 200
	appl_ptr

	0

	100	diff = 100 - 0 = 100

	0	diff = 0 - 100 = -100

	100


>  	if (diff) {
>  		if (diff < -(snd_pcm_sframes_t) (runtime->boundary / 2))
>  			diff += runtime->boundary;
> -		pcm->sw_ready += diff;
> +		frames += diff;
>  	}

and since boundary = buffer_size as default, the above condition will
be not satisfied (i.e. diff is equal with -(runtime->boundary/2) and
it is never smaller).

as i wrote, the condition should be <= .


btw, can the rewound length be over -(boundary/2) ?


Takashi


-------------------------------------------------------
This SF.NET email is sponsored by: FREE  SSL Guide from Thawte
are you planning your Web Server Security? Click here to get a FREE
Thawte SSL guide and find the answers to all your  SSL security issues.
http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0026en