On Wed, May 27, 2015 at 10:16 PM, Ben Widawsky <benjamin.widaw...@intel.com> wrote: > AFAICT, there is no real way to make sure a send message with EOT is properly > ignored from compact, nor can I see a way to actually encode EOT while > compacting. Before the single send optimization we'd always bail because we > hit > the is_immediate && !is_compactable_immediate case. However, with single send, > is_immediate is not true, and so we end up trying to compact the > un-compactible.
I investigated, and yeah, your analysis is exactly right. The EOT bit (bit 127) is unused when an instruction doesn't have an immediate, so the compaction code doesn't read it in that case. So I think this patch (with the checking for SEND/SENDC opcodes, since they're the only instructions that can have EOT) is exactly what we want. Reviewed-by: Matt Turner <matts...@gmail.com> Thanks! _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
