[PATCH] D15736: AMDGPU/SI: Select non-uniform constant addrspace loads to flat instructions for HSA

[Matt Arsenault via llvm-commits]

arsenm added inline comments.

================
Comment at: lib/Target/AMDGPU/CIInstructions.td:259-262
@@ -258,5 +258,6 @@
 
 def : FlatLoadPat <FLAT_LOAD_UBYTE, az_extloadi8_global, i32>;
 def : FlatLoadPat <FLAT_LOAD_SBYTE, sextloadi8_global, i32>;
 def : FlatLoadPat <FLAT_LOAD_USHORT, az_extloadi16_global, i32>;
 def : FlatLoadPat <FLAT_LOAD_SSHORT, sextloadi16_global, i32>;
+def : FlatLoadPat <FLAT_LOAD_DWORD, flat_load, i32>;
----------------
Do constant extloads need to be handled also?

================
Comment at: test/CodeGen/AMDGPU/salu-to-valu.ll:24
@@ -23,1 +23,3 @@
+; GCN-NOHSA: buffer_load_ubyte v{{[0-9]+}}, v[{{[0-9]+:[0-9]+}}], s[{{[0-9]+:[0-9]+}}], 0 addr64
+; GCN-HSA: flat_load_ubyte v{{[0-9]+}}, v[{{[0-9]+:[0-9]+}}
 
----------------
HSA missing 2nd load


http://reviews.llvm.org/D15736