On Thu, Sep 15, 2011 at 07:08:56PM -0700, Ben Widawsky wrote: > FWIW, I thought my original patch was the easiest to read, but this certainly > removes the most magic. These had a ton of input from Daniel Vetter, and Chris > Wilson. The goal is only to make the code more readable, and easier to > veryify/debug.
With the bloat gone: Reviewed-by: Daniel Vetter <[email protected]> for patches 2-4 But in then end I think it boils down to having a bunch of magic values in a table or having a bunch or magic calculations in a few tiny functions. So after the ring->id cleanup (which imo is independant cleanup, albeit just a tiny improvement) I don't care with what approach we'll be going. And imo engineering for future products isn't a good idea either - we just don't know what they'll look like exactly. In the end it's likely just yet another function pointer with completely different code, and having less chipset specific magic constant in struct intel_ring_buffer might even help. Now for the trace patch, I definitely want that. But that has a tiny review comment ;-) -Daniel -- Daniel Vetter Mail: [email protected] Mobile: +41 (0)79 365 57 48 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
