On Thu, Oct 10, 2013 at 05:17:31PM -0300, Paulo Zanoni wrote: > 2013/10/10 Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>: > > If the hardware does not support package C8, then do not even schedule > > work to enable it. Thereby we can eliminate a bunch of dangerous work. > > As I already explained, this should not be a problem since non-Haswell > platforms don't have a way to make the refcount become zero (unless we > have a bug). I also asked people's opinions about this specific > decision in one of my cover letters, but no one said anything: > http://lists.freedesktop.org/archives/intel-gfx/2013-August/031440.html. > > Quoting the email: "Another thing worth mentioning is that all this > code doesn't have IS_HASWELL checks, and on non-Haswell platforms the > refcount will never reach 0, so we won't ever try to enable PC8. I'm > not sure if that's what we want, so please comment on that.". > > That said, I'm not against your changes. If they don't actually fix anything, they are low priority as they only remove a mutex lock at most 10Hz. Maybe a comment would be good to remind the next person that nothing gets enabled except on hsw. > > +#define HAS_PC8(dev) (IS_HASWELL(dev)) /* XXX HSW:ULX */ > > What exactly do you mean with this comment? Did you actually mean > "IS_ULT()"? Even though only ULT has PC8-10 residencies, non-ULT seems > to work fine with this code, so I thought it wouldn't be a problem. It means I didn't actually check the valid restrictions :) -Chris -- Chris Wilson, Intel Open Source Technology Centre _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/intel-gfx
