[PATCH] D53930: [AMDGPU] Fix the new atomic optimizer in pixel shaders.

Neil Henning via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Mon Nov 5 04:05:52 PST 2018


sheredom marked 2 inline comments as done.
sheredom added inline comments.


================
Comment at: test/CodeGen/AMDGPU/atomic_optimizations_pixelshader.ll:41
+; GCN-LABEL: add_i32_varying:
+; GFX8MORE: v_mbcnt_lo_u32_b32{{(_e[0-9]+)?}} v[[mbcnt_lo:[0-9]+]], exec_lo, 0
+; GFX8MORE: v_mbcnt_hi_u32_b32{{(_e[0-9]+)?}} v[[mbcnt_hi:[0-9]+]], exec_hi, v[[mbcnt_lo]]
----------------
nhaehnle wrote:
> Why are there no check lines for GFX7LESS? What's the difference?
The opt doesn't apply < gfx8 because of lack of cross-wave opts, but I should have checked that it definitely didn't apply (so I added the GFX7LESS-NOT lines).


https://reviews.llvm.org/D53930





More information about the llvm-commits mailing list