How can I deal with this?

How can I deal with this?

[Adrian McMenamin]

when I play mono my device has one hardware buffer of size 0x8000, when
I play stereo it has two of size 0x8000

I have .buffer_bytes_max  and _min set to 0x8000

But for 16 bit stereo, ie 4 bytes per frame, the alsa middle layer only
expects the buffer to play 0x2000 frames when, of course, it can play
0x4000 - hence I am being deluged with bug reports and my driver
performs very poorly (this is what is really going on, not what i
suggested in a posting of yesterday about the buffer size being reset)



-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642

Re: How can I deal with this?

[Lee Revell]

On Sun, 2006-03-19 at 19:01 +0000, Adrian McMenamin wrote:
> when I play mono my device has one hardware buffer of size 0x8000, when
> I play stereo it has two of size 0x8000
> 
> I have .buffer_bytes_max  and _min set to 0x8000

Then .buffer_bytes_max should be 0x10000.  The buffer and period sizes
in your hw_info structure are for the entire buffer, not per channel.



-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642

Re: Re: How can I deal with this?

[Adrian McMenamin]

On Sun, 2006-03-19 at 14:11 -0500, Lee Revell wrote:
> On Sun, 2006-03-19 at 19:01 +0000, Adrian McMenamin wrote:
> > when I play mono my device has one hardware buffer of size 0x8000, when
> > I play stereo it has two of size 0x8000
> > 
> > I have .buffer_bytes_max  and _min set to 0x8000
> 
> Then .buffer_bytes_max should be 0x10000.  The buffer and period sizes
> in your hw_info structure are for the entire buffer, not per channel.
> 
> 
And double period max too I assume?



-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642

Re: Re: How can I deal with this?

[Lee Revell]

On Sun, 2006-03-19 at 19:16 +0000, Adrian McMenamin wrote:
> On Sun, 2006-03-19 at 14:11 -0500, Lee Revell wrote:
> > On Sun, 2006-03-19 at 19:01 +0000, Adrian McMenamin wrote:
> > > when I play mono my device has one hardware buffer of size 0x8000, when
> > > I play stereo it has two of size 0x8000
> > > 
> > > I have .buffer_bytes_max  and _min set to 0x8000
> > 
> > Then .buffer_bytes_max should be 0x10000.  The buffer and period sizes
> > in your hw_info structure are for the entire buffer, not per channel.
> > 
> > 
> And double period max too I assume?
> 
> 

If the hardware can do that, yes.

The hw_info structure should merely specify the hardware characteristics
- you are not trying to change the behavior if the middle layer.

Lee



-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642

Re: Re: How can I deal with this?

[Adrian McMenamin]

On Sun, 2006-03-19 at 14:31 -0500, Lee Revell wrote:
> On Sun, 2006-03-19 at 19:16 +0000, Adrian McMenamin wrote:
> > On Sun, 2006-03-19 at 14:11 -0500, Lee Revell wrote:
> > > On Sun, 2006-03-19 at 19:01 +0000, Adrian McMenamin wrote:
> > > > when I play mono my device has one hardware buffer of size 0x8000, when
> > > > I play stereo it has two of size 0x8000
> > > > 
> > > > I have .buffer_bytes_max  and _min set to 0x8000
> > > 
> > > Then .buffer_bytes_max should be 0x10000.  The buffer and period sizes
> > > in your hw_info structure are for the entire buffer, not per channel.
> > > 
> > > 
> > And double period max too I assume?
> > 
> > 
> 
> If the hardware can do that, yes.
> 
> The hw_info structure should merely specify the hardware characteristics
> - you are not trying to change the behavior if the middle layer.
> 

Doing any of this doesn't fix the problem in any way :(


I still get the bug reports when the play position extends beyond the
0x2000 frames point. I think I am going to have to mess about with the
firmware




-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642

Re: Re: How can I deal with this?

[Adrian McMenamin]

On Sun, 2006-03-19 at 19:49 +0000, Adrian McMenamin wrote:

> 
> 
> I still get the bug reports when the play position extends beyond the
> 0x2000 frames point. I think I am going to have to mess about with the
> firmware
> 
That seems to have fixed the problem :)

Though it poses an issue if I try to use any more of the 64 channels
this thing has - I'll have to keep shrinking the play back buffers per
channel!



-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642

Re: Re: How can I deal with this?

[Lee Revell]

On Sun, 2006-03-19 at 20:33 +0000, Adrian McMenamin wrote:
> On Sun, 2006-03-19 at 19:49 +0000, Adrian McMenamin wrote:
> 
> > 
> > 
> > I still get the bug reports when the play position extends beyond the
> > 0x2000 frames point. I think I am going to have to mess about with the
> > firmware
> > 
> That seems to have fixed the problem :)
> 
> Though it poses an issue if I try to use any more of the 64 channels
> this thing has - I'll have to keep shrinking the play back buffers per
> channel!
> 

Yep, exactly - most hardware has fixed buffer size and variable number
of channels so more channels requires smaller period sizes.

Lee



-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642