On Fri, 13 Jul 2012 06:56:44 -0700, Ben Widawsky <ben at bwidawsk.net> wrote: > On Fri, 13 Jul 2012 08:47:15 +0100 > Chris Wilson <chris at chris-wilson.co.uk> wrote: > > > On Thu, 12 Jul 2012 23:16:13 -0700, Ben Widawsky <ben at bwidawsk.net> wrote: > > > mostly for convenience, this will help us clear up a bit of the code in > > > intel_ringbuffer.c > > > > I don't think your couple of use-cases is a strong enough argument to > > justify an extra pointer on thousands of objects. > > > > If you wanted, you could make the ilk pc w/a use the status page > > instead... > > -Chris > > > > Actually, my original code went a step further than this. It combined > all the ring data into 1 object, and then used a mini allocator for the > pipe control, ring status, and the few dwords I need. I was _sure_ you > would hate that, so I went to this. > > Can you swallow the one object for everything + allocator idea? Yes. I'll happily pay a little one-off extra complexity to reduce the number of pages we have allocated for a smattering of dwords. -Chris -- Chris Wilson, Intel Open Source Technology Centre
