On Wed, Mar 26, 2014 at 05:09:53PM -0700, Ben Widawsky wrote:
> On Fri, Mar 21, 2014 at 12:41:53PM +0000, Chris Wilson wrote:
> > +#define I915_READ64_2x32(lower_reg, upper_reg) ({ \
> > + u32 upper = I915_READ(upper_reg); \
> > + u32 lower = I915_READ(lower_reg); \
> > + u32 tmp = I915_READ(upper_reg); \
> > + if (upper != tmp) { \
> > + upper = tmp; \
> > + lower = I915_READ(lower_reg); \
> > + WARN_ON(I915_READ(upper_reg) != upper); \
> > + } \
> > + (u64)upper << 32 | lower; })
> > +
>
>
> May as well get the most recent value of upper:
> WARN_ON((tmp = I915_READ(upper_reg)) != upper);
> }
> return (u64)tmp << 32 | lower;
Bleh, I thought if the WARN ever fires, the result is so unstable that
it really doesn't matter what we return, or we go full-loop.
-Chris
--
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
