Re: v4l2 + kernel

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]



On Tue, Apr 24, 2001 at 07:45:58AM -0300, Gleicon S. Moraes wrote:
> Gerd Knorr wrote:
> > 
> > For userland buffers the application has specify that tuple:  either
> > register a fixed set of buffers first (QUERYBUF) or fill them for every
> > QBUF.  Not sure yet which way is better.  
> 
> Does the buffer numbers depend on the hardware ?

Depends.  The v4l(1) API allows 32 buffers max.  It might be limited
due to hardware capabilies too.  For bttv it is a insmod config option.
As the bt848 writes directly to the host's memory there is no hard
limit other than the available RAM.

> The fixed set of
> buffers seems ok for me if the hardware has this kind of "parameter".
> The later one has a high cost.

It is costy for bttv too.  bttv has to lock down the memory to prevent
the VM from swapping out these pages, has to build risc code for the
buffer, ...

> Could we set up both modes,

I don't want to do that.  I want to keep the API simple.


> > for the v4l driver nor for the application which might want to put stuff
> > directly into X11 MIT SHM blocks for example).
> > 
> 
> Did I understand it right ? Would be possible to assign a previously
> malloc'ed or reserved buffer to v4l2 driver capture overthere ? Like X11
> MIT SHM you said, giver the proper format ( that you could set up
> previously ) and dump it to a given address ?

That is the plan.  Of course it will be optional for a driver to support
that because it requires hardware which can do scather-gather.

  Gerd





[Index of Archives]     [Linux DVB]     [Video Disk Recorder]     [Asterisk]     [Photo]     [DCCP]     [Netdev]     [Xorg]     [Util Linux NG]     [Xfree86]     [Free Photo Albums]     [Fedora Users]     [Fedora Women]     [ALSA Users]     [ALSA Devel]     [Linux USB]

Powered by Linux