diff options
Diffstat (limited to 'llvm/test/CodeGen/AMDGPU/llvm.amdgcn.sched.group.barrier.iterative.ll')
| -rw-r--r-- | llvm/test/CodeGen/AMDGPU/llvm.amdgcn.sched.group.barrier.iterative.ll | 26 |
1 files changed, 13 insertions, 13 deletions
diff --git a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.sched.group.barrier.iterative.ll b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.sched.group.barrier.iterative.ll index 371b4f0..af26e7a 100644 --- a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.sched.group.barrier.iterative.ll +++ b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.sched.group.barrier.iterative.ll @@ -1,14 +1,14 @@ ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5 -; RUN: llc -mtriple=amdgcn -mcpu=gfx90a -verify-machineinstrs -misched-cluster=0 -misched=gcn-iterative-minreg < %s | FileCheck -check-prefix=GCN-MINREG %s -; RUN: llc -mtriple=amdgcn -mcpu=gfx90a -verify-machineinstrs -misched-cluster=0 -misched=gcn-iterative-max-occupancy-experimental < %s | FileCheck -check-prefix=GCN-MAXOCC %s -; RUN: llc -mtriple=amdgcn -mcpu=gfx90a -verify-machineinstrs -misched-cluster=0 -misched=gcn-iterative-ilp < %s | FileCheck -check-prefix=GCN-ILP %s +; RUN: llc -mtriple=amdgcn -mcpu=gfx90a -misched-cluster=0 -misched=gcn-iterative-minreg < %s | FileCheck -check-prefix=GCN-MINREG %s +; RUN: llc -mtriple=amdgcn -mcpu=gfx90a -misched-cluster=0 -misched=gcn-iterative-max-occupancy-experimental < %s | FileCheck -check-prefix=GCN-MAXOCC %s +; RUN: llc -mtriple=amdgcn -mcpu=gfx90a -misched-cluster=0 -misched=gcn-iterative-ilp < %s | FileCheck -check-prefix=GCN-ILP %s define amdgpu_kernel void @test_sched_group_barrier_pipeline_MFMA_interleave(ptr addrspace(3) noalias %in, ptr addrspace(3) noalias %out) #0 { ; GCN-MINREG-LABEL: test_sched_group_barrier_pipeline_MFMA_interleave: ; GCN-MINREG: ; %bb.0: ; %entry ; GCN-MINREG-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0x24 -; GCN-MINREG-NEXT: v_and_b32_e32 v0, 0x3ff, v0 ; GCN-MINREG-NEXT: v_lshlrev_b32_e32 v0, 7, v0 +; GCN-MINREG-NEXT: v_and_b32_e32 v0, 0x1ff80, v0 ; GCN-MINREG-NEXT: v_mov_b32_e32 v2, 1.0 ; GCN-MINREG-NEXT: v_mov_b32_e32 v1, 2.0 ; GCN-MINREG-NEXT: s_waitcnt lgkmcnt(0) @@ -140,8 +140,8 @@ define amdgpu_kernel void @test_sched_group_barrier_pipeline_MFMA_interleave(ptr ; GCN-MAXOCC-LABEL: test_sched_group_barrier_pipeline_MFMA_interleave: ; GCN-MAXOCC: ; %bb.0: ; %entry ; GCN-MAXOCC-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0x24 -; GCN-MAXOCC-NEXT: v_and_b32_e32 v0, 0x3ff, v0 -; GCN-MAXOCC-NEXT: v_lshlrev_b32_e32 v1, 7, v0 +; GCN-MAXOCC-NEXT: v_lshlrev_b32_e32 v0, 7, v0 +; GCN-MAXOCC-NEXT: v_and_b32_e32 v1, 0x1ff80, v0 ; GCN-MAXOCC-NEXT: v_mov_b32_e32 v2, 1.0 ; GCN-MAXOCC-NEXT: v_mov_b32_e32 v3, 2.0 ; GCN-MAXOCC-NEXT: s_waitcnt lgkmcnt(0) @@ -274,8 +274,8 @@ define amdgpu_kernel void @test_sched_group_barrier_pipeline_MFMA_interleave(ptr ; GCN-ILP-LABEL: test_sched_group_barrier_pipeline_MFMA_interleave: ; GCN-ILP: ; %bb.0: ; %entry ; GCN-ILP-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0x24 -; GCN-ILP-NEXT: v_and_b32_e32 v0, 0x3ff, v0 ; GCN-ILP-NEXT: v_lshlrev_b32_e32 v0, 7, v0 +; GCN-ILP-NEXT: v_and_b32_e32 v0, 0x1ff80, v0 ; GCN-ILP-NEXT: v_mov_b32_e32 v1, 1.0 ; GCN-ILP-NEXT: v_mov_b32_e32 v2, 2.0 ; GCN-ILP-NEXT: s_waitcnt lgkmcnt(0) @@ -469,8 +469,8 @@ define amdgpu_kernel void @test_sched_group_barrier_pipeline_MFMA_interleave_spl ; GCN-MINREG-LABEL: test_sched_group_barrier_pipeline_MFMA_interleave_split_region: ; GCN-MINREG: ; %bb.0: ; %entry ; GCN-MINREG-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0x24 -; GCN-MINREG-NEXT: v_and_b32_e32 v0, 0x3ff, v0 -; GCN-MINREG-NEXT: v_lshlrev_b32_e32 v2, 7, v0 +; GCN-MINREG-NEXT: v_lshlrev_b32_e32 v0, 7, v0 +; GCN-MINREG-NEXT: v_and_b32_e32 v2, 0x1ff80, v0 ; GCN-MINREG-NEXT: v_mov_b32_e32 v1, 1.0 ; GCN-MINREG-NEXT: v_mov_b32_e32 v0, 2.0 ; GCN-MINREG-NEXT: s_waitcnt lgkmcnt(0) @@ -604,8 +604,8 @@ define amdgpu_kernel void @test_sched_group_barrier_pipeline_MFMA_interleave_spl ; GCN-MAXOCC-LABEL: test_sched_group_barrier_pipeline_MFMA_interleave_split_region: ; GCN-MAXOCC: ; %bb.0: ; %entry ; GCN-MAXOCC-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0x24 -; GCN-MAXOCC-NEXT: v_and_b32_e32 v0, 0x3ff, v0 -; GCN-MAXOCC-NEXT: v_lshlrev_b32_e32 v3, 7, v0 +; GCN-MAXOCC-NEXT: v_lshlrev_b32_e32 v0, 7, v0 +; GCN-MAXOCC-NEXT: v_and_b32_e32 v3, 0x1ff80, v0 ; GCN-MAXOCC-NEXT: v_mov_b32_e32 v1, 1.0 ; GCN-MAXOCC-NEXT: v_mov_b32_e32 v2, 2.0 ; GCN-MAXOCC-NEXT: s_waitcnt lgkmcnt(0) @@ -739,8 +739,8 @@ define amdgpu_kernel void @test_sched_group_barrier_pipeline_MFMA_interleave_spl ; GCN-ILP-LABEL: test_sched_group_barrier_pipeline_MFMA_interleave_split_region: ; GCN-ILP: ; %bb.0: ; %entry ; GCN-ILP-NEXT: s_load_dwordx2 s[0:1], s[4:5], 0x24 -; GCN-ILP-NEXT: v_and_b32_e32 v0, 0x3ff, v0 -; GCN-ILP-NEXT: v_lshlrev_b32_e32 v2, 7, v0 +; GCN-ILP-NEXT: v_lshlrev_b32_e32 v0, 7, v0 +; GCN-ILP-NEXT: v_and_b32_e32 v2, 0x1ff80, v0 ; GCN-ILP-NEXT: v_mov_b32_e32 v0, 1.0 ; GCN-ILP-NEXT: v_mov_b32_e32 v1, 2.0 ; GCN-ILP-NEXT: s_waitcnt lgkmcnt(0) |