This is an archive of the discontinued LLVM Phabricator instance.

Differential D63392

AMDGPU: Explicitly define a triple for some tests
ClosedPublic

Authored by nhaehnle on Jun 16 2019, 4:37 PM.

Details

Reviewers
arsenm
Commits
rG8af7198c6caa: AMDGPU: Explicitly define a triple for some tests
rL363600: AMDGPU: Explicitly define a triple for some tests
Summary

This is related to the changes to the groupstaticsize intrinsic in
D61494 which would otherwise make the related tests in these files
fail or much less useful.

Note that for some reason, SOPK generation is less effective in the
amdhsa OS, which is why I chose PAL. I haven't investigated this
deeper.

Change-Id: I6bb99569338f7a433c28b4c9eb1e3e036b00d166

Diff Detail

Repository
rL LLVM

Event Timeline

nhaehnle created this revision.Jun 16 2019, 4:37 PM
Herald added a project: Restricted Project. · View Herald TranscriptJun 16 2019, 4:37 PM
Herald added subscribers: t-tye, tpr, dstuttard and 4 others. · View Herald Transcript
Harbormaster completed remote builds in B33456: Diff 204975.Jun 16 2019, 4:38 PM
arsenm accepted this revision.Jun 16 2019, 4:51 PM
This revision is now accepted and ready to land.Jun 16 2019, 4:51 PM
nhaehnle added a child revision: D61494: AMDGPU: Write LDS objects out as global symbols in code generation.Jun 16 2019, 6:03 PM
Closed by commit rL363600: AMDGPU: Explicitly define a triple for some tests (authored by nha). · Explain WhyJun 17 2019, 12:24 PM
This revision was automatically updated to reflect the committed changes.

Revision Contents

PathSize
llvm/
trunk/
test/
CodeGen/
AMDGPU/
4 lines
7 lines
4 lines
4 lines
4 lines

Diff 205148

llvm/trunk/test/CodeGen/AMDGPU/constant-fold-mi-operands.ll

; RUN: llc -march=amdgcn -mcpu=verde -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s ; RUN: llc -mtriple=amdgcn--amdpal -mcpu=verde -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
; RUN: llc -march=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s ; RUN: llc -mtriple=amdgcn--amdpal -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
; GCN-LABEL: {{^}}fold_mi_v_and_0: ; GCN-LABEL: {{^}}fold_mi_v_and_0:
; GCN: v_mov_b32_e32 [[RESULT:v[0-9]+]], 0{{$}} ; GCN: v_mov_b32_e32 [[RESULT:v[0-9]+]], 0{{$}}
; GCN-NOT: [[RESULT]] ; GCN-NOT: [[RESULT]]
; GCN: buffer_store_dword [[RESULT]] ; GCN: buffer_store_dword [[RESULT]]
define amdgpu_kernel void @fold_mi_v_and_0(i32 addrspace(1)* %out) { define amdgpu_kernel void @fold_mi_v_and_0(i32 addrspace(1)* %out) {
%x = call i32 @llvm.amdgcn.mbcnt.lo(i32 -1, i32 0) #0 %x = call i32 @llvm.amdgcn.mbcnt.lo(i32 -1, i32 0) #0
%size = call i32 @llvm.amdgcn.groupstaticsize() %size = call i32 @llvm.amdgcn.groupstaticsize()
▲ Show 20 Lines • Show All 134 Lines • Show Last 20 Lines

llvm/trunk/test/CodeGen/AMDGPU/s_addk_i32.ll

; RUN: llc -march=amdgcn -mcpu=tahiti -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s ; RUN: llc -mtriple=amdgcn--amdpal -mcpu=tahiti -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
; RUN: llc -march=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s ; RUN: llc -mtriple=amdgcn--amdpal -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
; TODO: Some of those tests fail with OS == amdhsa due to unreasonable register
; allocation differences.
; SI-LABEL: {{^}}s_addk_i32_k0: ; SI-LABEL: {{^}}s_addk_i32_k0:
; SI: s_load_dword [[VAL:s[0-9]+]] ; SI: s_load_dword [[VAL:s[0-9]+]]
; SI: s_addk_i32 [[VAL]], 0x41 ; SI: s_addk_i32 [[VAL]], 0x41
; SI: v_mov_b32_e32 [[VRESULT:v[0-9]+]], [[VAL]] ; SI: v_mov_b32_e32 [[VRESULT:v[0-9]+]], [[VAL]]
; SI: buffer_store_dword [[VRESULT]] ; SI: buffer_store_dword [[VRESULT]]
; SI: s_endpgm ; SI: s_endpgm
define amdgpu_kernel void @s_addk_i32_k0(i32 addrspace(1)* %out, i32 %b) { define amdgpu_kernel void @s_addk_i32_k0(i32 addrspace(1)* %out, i32 %b) {
▲ Show 20 Lines • Show All 108 Lines • Show Last 20 Lines

llvm/trunk/test/CodeGen/AMDGPU/s_mulk_i32.ll

; RUN: llc -march=amdgcn -mcpu=tahiti -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s ; RUN: llc -mtriple=amdgcn--amdpal -mcpu=tahiti -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
; RUN: llc -march=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s ; RUN: llc -mtriple=amdgcn--amdpal -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
; SI-LABEL: {{^}}s_mulk_i32_k0: ; SI-LABEL: {{^}}s_mulk_i32_k0:
; SI: s_load_dword [[VAL:s[0-9]+]] ; SI: s_load_dword [[VAL:s[0-9]+]]
; SI: s_mulk_i32 [[VAL]], 0x41 ; SI: s_mulk_i32 [[VAL]], 0x41
; SI: v_mov_b32_e32 [[VRESULT:v[0-9]+]], [[VAL]] ; SI: v_mov_b32_e32 [[VRESULT:v[0-9]+]], [[VAL]]
; SI: buffer_store_dword [[VRESULT]] ; SI: buffer_store_dword [[VRESULT]]
; SI: s_endpgm ; SI: s_endpgm
define amdgpu_kernel void @s_mulk_i32_k0(i32 addrspace(1)* %out, i32 %b) { define amdgpu_kernel void @s_mulk_i32_k0(i32 addrspace(1)* %out, i32 %b) {
▲ Show 20 Lines • Show All 47 Lines • Show Last 20 Lines

llvm/trunk/test/CodeGen/AMDGPU/sopk-compares.ll

; RUN: llc -march=amdgcn -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=SI %s ; RUN: llc -mtriple=amdgcn--amdpal -mcpu=tahiti -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=SI %s
; RUN: llc -march=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=VI %s ; RUN: llc -mtriple=amdgcn--amdpal -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=VI %s
; Since this intrinsic is exposed as a constant after isel, use it to ; Since this intrinsic is exposed as a constant after isel, use it to
; defeat the DAG's compare with constant canonicalizations. ; defeat the DAG's compare with constant canonicalizations.
declare i32 @llvm.amdgcn.groupstaticsize() #1 declare i32 @llvm.amdgcn.groupstaticsize() #1
@lds = addrspace(3) global [512 x i32] undef, align 4 @lds = addrspace(3) global [512 x i32] undef, align 4
; GCN-LABEL: {{^}}br_scc_eq_i32_inline_imm: ; GCN-LABEL: {{^}}br_scc_eq_i32_inline_imm:
▲ Show 20 Lines • Show All 639 Lines • Show Last 20 Lines

llvm/trunk/test/CodeGen/AMDGPU/sub.i16.ll

; RUN: llc -march=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=VI -check-prefix=GCN %s ; RUN: llc -mtriple=amdgcn--amdhsa -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=VI -check-prefix=GCN %s
; RUN: llc -march=amdgcn -mcpu=bonaire -verify-machineinstrs < %s | FileCheck -check-prefix=CI -check-prefix=GCN %s ; RUN: llc -mtriple=amdgcn--amdhsa -mcpu=bonaire -verify-machineinstrs < %s | FileCheck -check-prefix=CI -check-prefix=GCN %s
; FIXME: Need to handle non-uniform case for function below (load without gep). ; FIXME: Need to handle non-uniform case for function below (load without gep).
; GCN-LABEL: {{^}}v_test_sub_i16: ; GCN-LABEL: {{^}}v_test_sub_i16:
; VI: flat_load_ushort [[A:v[0-9]+]] ; VI: flat_load_ushort [[A:v[0-9]+]]
; VI: flat_load_ushort [[B:v[0-9]+]] ; VI: flat_load_ushort [[B:v[0-9]+]]
; VI: v_sub_u16_e32 [[ADD:v[0-9]+]], [[A]], [[B]] ; VI: v_sub_u16_e32 [[ADD:v[0-9]+]], [[A]], [[B]]
; VI-NEXT: buffer_store_short [[ADD]] ; VI-NEXT: buffer_store_short [[ADD]]
define amdgpu_kernel void @v_test_sub_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %in0, i16 addrspace(1)* %in1) #1 { define amdgpu_kernel void @v_test_sub_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %in0, i16 addrspace(1)* %in1) #1 {
▲ Show 20 Lines • Show All 159 Lines • Show Last 20 Lines