caio/clang-p2996
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
12ccdd67aaf56bd05ab3375d81c58dfd5a2db02c
clang-p2996/llvm/test/CodeGen/AMDGPU/GlobalISel/inst-select-pattern-umed3.s16.mir
Joe Nash 18ed279a3a [AMDGPU] gfx11 subtarget features & early tests 
Tablegen definitions for subtarget features and cpp predicate functions to
access the features.
New Sub-TargetProcessors and common latencies.
Simple changes to MIR codegen tests which pass on gfx11 because they have the
same output as previous subtargets or operate on pseudo instructions which
are reused from previous subtargets.

Contributors:
Jay Foad <jay.foad@amd.com>
Petar Avramovic <Petar.Avramovic@amd.com>

Patch 4/N for upstreaming of AMDGPU gfx11 architecture

Depends on D124538

Reviewed By: Petar.Avramovic, foad

Differential Revision: https://reviews.llvm.org/D125261
2022-05-11 10:31:49 -04:00

171 lines
7.7 KiB
YAML
Raw Blame History

# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
# RUN: llc -march=amdgcn -mcpu=fiji -run-pass=instruction-select -verify-machineinstrs %s -o - | FileCheck -check-prefix=GFX8 %s
# RUN: llc -march=amdgcn -mcpu=gfx900 -run-pass=instruction-select -verify-machineinstrs %s -o - | FileCheck -check-prefix=GFX9 %s
# RUN: llc -march=amdgcn -mcpu=gfx1010 -run-pass=instruction-select -verify-machineinstrs %s -o - | FileCheck -check-prefix=GFX9 %s
# RUN: llc -march=amdgcn -mcpu=gfx1100 -run-pass=instruction-select -verify-machineinstrs %s -o - | FileCheck -check-prefix=GFX9 %s
---
name: umed3_s16_vvv
legalized: true
regBankSelected: true
body: |
bb.0:
liveins: $vgpr0, $vgpr1, $vgpr2
; GFX8-LABEL: name: umed3_s16_vvv
; GFX8: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX8: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX8: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX8: [[V_MAX_U16_e64_:%[0-9]+]]:vgpr_32 = V_MAX_U16_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX8: [[V_MIN_U16_e64_:%[0-9]+]]:vgpr_32 = V_MIN_U16_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX8: [[V_MAX_U16_e64_1:%[0-9]+]]:vgpr_32 = V_MAX_U16_e64 [[V_MIN_U16_e64_]], [[COPY2]], implicit $exec
; GFX8: [[V_MIN_U16_e64_1:%[0-9]+]]:vgpr_32 = V_MIN_U16_e64 [[V_MAX_U16_e64_]], [[V_MAX_U16_e64_1]], implicit $exec
; GFX8: S_ENDPGM 0, implicit [[V_MIN_U16_e64_1]]
; GFX9-LABEL: name: umed3_s16_vvv
; GFX9: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX9: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX9: [[V_MED3_U16_e64_:%[0-9]+]]:vgpr_32 = V_MED3_U16_e64 0, [[COPY]], 0, [[COPY1]], 0, [[COPY2]], 0, 0, implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_MED3_U16_e64_]]
%0:vgpr(s32) = COPY $vgpr0
%1:vgpr(s32) = COPY $vgpr1
%2:vgpr(s32) = COPY $vgpr2
%3:vgpr(s16) = G_TRUNC %0
%4:vgpr(s16) = G_TRUNC %1
%5:vgpr(s16) = G_TRUNC %2
%6:vgpr(s16) = G_UMAX %3, %4
%7:vgpr(s16) = G_UMIN %3, %4
%8:vgpr(s16) = G_UMAX %7, %5
%9:vgpr(s16) = G_UMIN %6, %8
S_ENDPGM 0, implicit %9
...
---
name: umed3_s16_vvv_multiuse0
legalized: true
regBankSelected: true
body: |
bb.0:
liveins: $vgpr0, $vgpr1, $vgpr2
; GFX8-LABEL: name: umed3_s16_vvv_multiuse0
; GFX8: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX8: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX8: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX8: [[V_MAX_U16_e64_:%[0-9]+]]:vgpr_32 = V_MAX_U16_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX8: [[V_MIN_U16_e64_:%[0-9]+]]:vgpr_32 = V_MIN_U16_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX8: [[V_MAX_U16_e64_1:%[0-9]+]]:vgpr_32 = V_MAX_U16_e64 [[V_MIN_U16_e64_]], [[COPY2]], implicit $exec
; GFX8: [[V_MIN_U16_e64_1:%[0-9]+]]:vgpr_32 = V_MIN_U16_e64 [[V_MAX_U16_e64_]], [[V_MAX_U16_e64_1]], implicit $exec
; GFX8: S_ENDPGM 0, implicit [[V_MIN_U16_e64_1]], implicit [[V_MAX_U16_e64_]]
; GFX9-LABEL: name: umed3_s16_vvv_multiuse0
; GFX9: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX9: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX9: [[V_MAX_U16_e64_:%[0-9]+]]:vgpr_32 = V_MAX_U16_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX9: [[V_MIN_U16_e64_:%[0-9]+]]:vgpr_32 = V_MIN_U16_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX9: [[V_MAX_U16_e64_1:%[0-9]+]]:vgpr_32 = V_MAX_U16_e64 [[V_MIN_U16_e64_]], [[COPY2]], implicit $exec
; GFX9: [[V_MIN_U16_e64_1:%[0-9]+]]:vgpr_32 = V_MIN_U16_e64 [[V_MAX_U16_e64_]], [[V_MAX_U16_e64_1]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_MIN_U16_e64_1]], implicit [[V_MAX_U16_e64_]]
%0:vgpr(s32) = COPY $vgpr0
%1:vgpr(s32) = COPY $vgpr1
%2:vgpr(s32) = COPY $vgpr2
%3:vgpr(s16) = G_TRUNC %0
%4:vgpr(s16) = G_TRUNC %1
%5:vgpr(s16) = G_TRUNC %2
%6:vgpr(s16) = G_UMAX %3, %4
%7:vgpr(s16) = G_UMIN %3, %4
%8:vgpr(s16) = G_UMAX %7, %5
%9:vgpr(s16) = G_UMIN %6, %8
S_ENDPGM 0, implicit %9, implicit %6
...
---
name: umed3_s16_vvv_multiuse1
legalized: true
regBankSelected: true
body: |
bb.0:
liveins: $vgpr0, $vgpr1, $vgpr2
; GFX8-LABEL: name: umed3_s16_vvv_multiuse1
; GFX8: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX8: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX8: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX8: [[V_MAX_U16_e64_:%[0-9]+]]:vgpr_32 = V_MAX_U16_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX8: [[V_MIN_U16_e64_:%[0-9]+]]:vgpr_32 = V_MIN_U16_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX8: [[V_MAX_U16_e64_1:%[0-9]+]]:vgpr_32 = V_MAX_U16_e64 [[V_MIN_U16_e64_]], [[COPY2]], implicit $exec
; GFX8: [[V_MIN_U16_e64_1:%[0-9]+]]:vgpr_32 = V_MIN_U16_e64 [[V_MAX_U16_e64_]], [[V_MAX_U16_e64_1]], implicit $exec
; GFX8: S_ENDPGM 0, implicit [[V_MIN_U16_e64_1]], implicit [[V_MIN_U16_e64_]]
; GFX9-LABEL: name: umed3_s16_vvv_multiuse1
; GFX9: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX9: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX9: [[V_MAX_U16_e64_:%[0-9]+]]:vgpr_32 = V_MAX_U16_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX9: [[V_MIN_U16_e64_:%[0-9]+]]:vgpr_32 = V_MIN_U16_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX9: [[V_MAX_U16_e64_1:%[0-9]+]]:vgpr_32 = V_MAX_U16_e64 [[V_MIN_U16_e64_]], [[COPY2]], implicit $exec
; GFX9: [[V_MIN_U16_e64_1:%[0-9]+]]:vgpr_32 = V_MIN_U16_e64 [[V_MAX_U16_e64_]], [[V_MAX_U16_e64_1]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_MIN_U16_e64_1]], implicit [[V_MIN_U16_e64_]]
%0:vgpr(s32) = COPY $vgpr0
%1:vgpr(s32) = COPY $vgpr1
%2:vgpr(s32) = COPY $vgpr2
%3:vgpr(s16) = G_TRUNC %0
%4:vgpr(s16) = G_TRUNC %1
%5:vgpr(s16) = G_TRUNC %2
%6:vgpr(s16) = G_UMAX %3, %4
%7:vgpr(s16) = G_UMIN %3, %4
%8:vgpr(s16) = G_UMAX %7, %5
%9:vgpr(s16) = G_UMIN %6, %8
S_ENDPGM 0, implicit %9, implicit %7
...
---
name: umed3_s16_vvv_multiuse2
legalized: true
regBankSelected: true
body: |
bb.0:
liveins: $vgpr0, $vgpr1, $vgpr2
; GFX8-LABEL: name: umed3_s16_vvv_multiuse2
; GFX8: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX8: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX8: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX8: [[V_MAX_U16_e64_:%[0-9]+]]:vgpr_32 = V_MAX_U16_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX8: [[V_MIN_U16_e64_:%[0-9]+]]:vgpr_32 = V_MIN_U16_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX8: [[V_MAX_U16_e64_1:%[0-9]+]]:vgpr_32 = V_MAX_U16_e64 [[V_MIN_U16_e64_]], [[COPY2]], implicit $exec
; GFX8: [[V_MIN_U16_e64_1:%[0-9]+]]:vgpr_32 = V_MIN_U16_e64 [[V_MAX_U16_e64_]], [[V_MAX_U16_e64_1]], implicit $exec
; GFX8: S_ENDPGM 0, implicit [[V_MIN_U16_e64_1]], implicit [[V_MAX_U16_e64_1]]
; GFX9-LABEL: name: umed3_s16_vvv_multiuse2
; GFX9: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
; GFX9: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
; GFX9: [[COPY2:%[0-9]+]]:vgpr_32 = COPY $vgpr2
; GFX9: [[V_MAX_U16_e64_:%[0-9]+]]:vgpr_32 = V_MAX_U16_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX9: [[V_MIN_U16_e64_:%[0-9]+]]:vgpr_32 = V_MIN_U16_e64 [[COPY]], [[COPY1]], implicit $exec
; GFX9: [[V_MAX_U16_e64_1:%[0-9]+]]:vgpr_32 = V_MAX_U16_e64 [[V_MIN_U16_e64_]], [[COPY2]], implicit $exec
; GFX9: [[V_MIN_U16_e64_1:%[0-9]+]]:vgpr_32 = V_MIN_U16_e64 [[V_MAX_U16_e64_]], [[V_MAX_U16_e64_1]], implicit $exec
; GFX9: S_ENDPGM 0, implicit [[V_MIN_U16_e64_1]], implicit [[V_MAX_U16_e64_1]]
%0:vgpr(s32) = COPY $vgpr0
%1:vgpr(s32) = COPY $vgpr1
%2:vgpr(s32) = COPY $vgpr2
%3:vgpr(s16) = G_TRUNC %0
%4:vgpr(s16) = G_TRUNC %1
%5:vgpr(s16) = G_TRUNC %2
%6:vgpr(s16) = G_UMAX %3, %4
%7:vgpr(s16) = G_UMIN %3, %4
%8:vgpr(s16) = G_UMAX %7, %5
%9:vgpr(s16) = G_UMIN %6, %8
S_ENDPGM 0, implicit %9, implicit %8
...
Reference in New Issue View Git Blame Copy Permalink