Re: [V4L] Re: v4l2 + select() + read()

[Ben Bridgwater]

Alan Cox wrote:

> > If/when Linux gets async IO that will be true, but until then there's no
> > way to queue multiple reads, so mmap() is needed for that reason, as
> > well as for backwards compatability (although I guess we *could* break
> > compatability in 2.5).
>
> Why do you need to queue multiple reads

If mmap is the performance interfaces then you don't, but you need *some*
mechanism for queued "reads", and the deep reason is because Linux is not hard
real-time. Of course you also need at least one "queued" read to take advantage
of acquisition time.

> (and a threaded app can do it btw in
> fact glibc 2.2 has full aio using threads) - you have a whole frame to get
> around to the next one plus any internal buffering

Sure, and you can get around lack of select() support with threads too. But
it's not very convenient or efficient.

Ben




_______________________________________________
Video4linux-list mailing list
[EMAIL PROTECTED]
https://listman.redhat.com/mailman/listinfo/video4linux-list