[PATCH] D55570: [AMDGPU] Improve SDWA generation for V_OR_B32_E32.

[Stanislav Mekhanoshin via Phabricator via llvm-commits]

rampitec added inline comments.


================
Comment at: test/CodeGen/AMDGPU/add.v2i16.ll:130
 ; VI-NOT: v_add_u16
-; VI: v_or_b32_e32
+; VI: v_or_b32_sdwa
 define amdgpu_kernel void @v_test_add_v2i16_inline_lo_zero_hi(<2 x i16> addrspace(1)* %out, <2 x i16> addrspace(1)* %in0) #1 {
----------------
Check for specific sdwa operand.


================
Comment at: test/CodeGen/AMDGPU/insert_vector_elt.v2i16.ll:2
+; RUN: llc -verify-machineinstrs -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -enable-amdgpu-aa=0 -mattr=+flat-for-global,-fp64-fp16-denormals < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GFX9,GFX89,GFX9NOTFIJI %s
+; RUN: llc -verify-machineinstrs -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -enable-amdgpu-aa=0 -mattr=+flat-for-global < %s | FileCheck -enable-var-scope -check-prefixes=GCN,CIVI,VI,GFX89,CIVIFIJI %s
+; RUN: llc -verify-machineinstrs -mtriple=amdgcn-amd-amdhsa -mcpu=hawaii -enable-amdgpu-aa=0 -mattr=+flat-for-global < %s | FileCheck -enable-var-scope -check-prefixes=GCN,CIVI,CIVINOTFIJI,CI %s
----------------
Fiji is VI, CIVIFIJI makes no sense.
Also please use VI or GFX8, do not add FIJI checks.


================
Comment at: test/CodeGen/AMDGPU/insert_vector_elt.v2i16.ll:219
+
+; CIVIFIJI: v_or_b32_sdwa [[RES:v[0-9]+]], v{{[0-9]}}, v{{[0-9]}}
 
----------------
Check for sdwa operands.


Repository:
  rL LLVM

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D55570/new/

https://reviews.llvm.org/D55570