derek.holloway
7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
104 lines
5.3 KiB
LLVM
104 lines
5.3 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
|
|
; RUN: llc -mtriple=amdgcn -mcpu=gfx1100 -show-mc-encoding < %s | FileCheck %s --check-prefixes=GFX11
|
|
|
|
; RUN: not llc -global-isel=0 -mtriple=amdgcn -mcpu=gfx900 -filetype=null < %s 2>&1 | FileCheck -check-prefix=ERR %s
|
|
; RUN: not llc -global-isel=1 -global-isel-abort=0 -mtriple=amdgcn -mcpu=gfx900 -filetype=null < %s 2>&1 | FileCheck -check-prefix=ERR %s
|
|
; RUN: not llc -global-isel=0 -mtriple=amdgcn -mcpu=gfx1010 -filetype=null < %s 2>&1 | FileCheck -check-prefix=ERR %s
|
|
; RUN: not llc -global-isel=1 -global-isel-abort=0 -mtriple=amdgcn -mcpu=gfx1010 -filetype=null < %s 2>&1 | FileCheck -check-prefix=ERR %s
|
|
; RUN: not llc -global-isel=0 -mtriple=amdgcn -mcpu=gfx950 -filetype=null < %s 2>&1 | FileCheck -check-prefix=ERR %s
|
|
; RUN: not llc -global-isel=1 -global-isel-abort=0 -mtriple=amdgcn -mcpu=gfx950 -filetype=null < %s 2>&1 | FileCheck -check-prefix=ERR %s
|
|
|
|
; ERR: error: <unknown>:0:0: in function test_llvm_amdgcn_sudot4_uu i32 (i32, i32, i32): intrinsic not supported on subtarget
|
|
|
|
declare i32 @llvm.amdgcn.sudot4(i1 %asign, i32 %a, i1 %bsign, i32 %b, i32 %c, i1 %clamp)
|
|
|
|
define i32 @test_llvm_amdgcn_sudot4_uu(i32 %a, i32 %b, i32 %c) {
|
|
; GFX11-LABEL: test_llvm_amdgcn_sudot4_uu:
|
|
; GFX11: ; %bb.0: ; %entry
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) ; encoding: [0x00,0x00,0x89,0xbf]
|
|
; GFX11-NEXT: v_dot4_i32_iu8 v0, v0, v1, v2 ; encoding: [0x00,0x40,0x16,0xcc,0x00,0x03,0x0a,0x1c]
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31] ; encoding: [0x1e,0x48,0x80,0xbe]
|
|
entry:
|
|
%ret = call i32 @llvm.amdgcn.sudot4(i1 0, i32 %a, i1 0, i32 %b, i32 %c, i1 0)
|
|
ret i32 %ret
|
|
}
|
|
|
|
define i32 @test_llvm_amdgcn_sudot4_us(i32 %a, i32 %b, i32 %c) {
|
|
; GFX11-LABEL: test_llvm_amdgcn_sudot4_us:
|
|
; GFX11: ; %bb.0: ; %entry
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) ; encoding: [0x00,0x00,0x89,0xbf]
|
|
; GFX11-NEXT: v_dot4_i32_iu8 v0, v0, v1, v2 neg_lo:[0,1,0] ; encoding: [0x00,0x40,0x16,0xcc,0x00,0x03,0x0a,0x5c]
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31] ; encoding: [0x1e,0x48,0x80,0xbe]
|
|
entry:
|
|
%ret = call i32 @llvm.amdgcn.sudot4(i1 0, i32 %a, i1 1, i32 %b, i32 %c, i1 0)
|
|
ret i32 %ret
|
|
}
|
|
|
|
define i32 @test_llvm_amdgcn_sudot4_su(i32 %a, i32 %b, i32 %c) {
|
|
; GFX11-LABEL: test_llvm_amdgcn_sudot4_su:
|
|
; GFX11: ; %bb.0: ; %entry
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) ; encoding: [0x00,0x00,0x89,0xbf]
|
|
; GFX11-NEXT: v_dot4_i32_iu8 v0, v0, v1, v2 neg_lo:[1,0,0] ; encoding: [0x00,0x40,0x16,0xcc,0x00,0x03,0x0a,0x3c]
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31] ; encoding: [0x1e,0x48,0x80,0xbe]
|
|
entry:
|
|
%ret = call i32 @llvm.amdgcn.sudot4(i1 1, i32 %a, i1 0, i32 %b, i32 %c, i1 0)
|
|
ret i32 %ret
|
|
}
|
|
|
|
define i32 @test_llvm_amdgcn_sudot4_ss(i32 %a, i32 %b, i32 %c) {
|
|
; GFX11-LABEL: test_llvm_amdgcn_sudot4_ss:
|
|
; GFX11: ; %bb.0: ; %entry
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) ; encoding: [0x00,0x00,0x89,0xbf]
|
|
; GFX11-NEXT: v_dot4_i32_iu8 v0, v0, v1, v2 neg_lo:[1,1,0] ; encoding: [0x00,0x40,0x16,0xcc,0x00,0x03,0x0a,0x7c]
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31] ; encoding: [0x1e,0x48,0x80,0xbe]
|
|
entry:
|
|
%ret = call i32 @llvm.amdgcn.sudot4(i1 1, i32 %a, i1 1, i32 %b, i32 %c, i1 0)
|
|
ret i32 %ret
|
|
}
|
|
|
|
|
|
|
|
define i32 @test_llvm_amdgcn_sudot4_uu_clamp(i32 %a, i32 %b, i32 %c) {
|
|
; GFX11-LABEL: test_llvm_amdgcn_sudot4_uu_clamp:
|
|
; GFX11: ; %bb.0: ; %entry
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) ; encoding: [0x00,0x00,0x89,0xbf]
|
|
; GFX11-NEXT: v_dot4_i32_iu8 v0, v0, v1, v2 clamp ; encoding: [0x00,0xc0,0x16,0xcc,0x00,0x03,0x0a,0x1c]
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31] ; encoding: [0x1e,0x48,0x80,0xbe]
|
|
entry:
|
|
%ret = call i32 @llvm.amdgcn.sudot4(i1 0, i32 %a, i1 0, i32 %b, i32 %c, i1 1)
|
|
ret i32 %ret
|
|
}
|
|
|
|
define i32 @test_llvm_amdgcn_sudot4_us_clamp(i32 %a, i32 %b, i32 %c) {
|
|
; GFX11-LABEL: test_llvm_amdgcn_sudot4_us_clamp:
|
|
; GFX11: ; %bb.0: ; %entry
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) ; encoding: [0x00,0x00,0x89,0xbf]
|
|
; GFX11-NEXT: v_dot4_i32_iu8 v0, v0, v1, v2 neg_lo:[0,1,0] clamp ; encoding: [0x00,0xc0,0x16,0xcc,0x00,0x03,0x0a,0x5c]
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31] ; encoding: [0x1e,0x48,0x80,0xbe]
|
|
entry:
|
|
%ret = call i32 @llvm.amdgcn.sudot4(i1 0, i32 %a, i1 1, i32 %b, i32 %c, i1 1)
|
|
ret i32 %ret
|
|
}
|
|
|
|
define i32 @test_llvm_amdgcn_sudot4_su_clamp(i32 %a, i32 %b, i32 %c) {
|
|
; GFX11-LABEL: test_llvm_amdgcn_sudot4_su_clamp:
|
|
; GFX11: ; %bb.0: ; %entry
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) ; encoding: [0x00,0x00,0x89,0xbf]
|
|
; GFX11-NEXT: v_dot4_i32_iu8 v0, v0, v1, v2 neg_lo:[1,0,0] clamp ; encoding: [0x00,0xc0,0x16,0xcc,0x00,0x03,0x0a,0x3c]
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31] ; encoding: [0x1e,0x48,0x80,0xbe]
|
|
entry:
|
|
%ret = call i32 @llvm.amdgcn.sudot4(i1 1, i32 %a, i1 0, i32 %b, i32 %c, i1 1)
|
|
ret i32 %ret
|
|
}
|
|
|
|
define i32 @test_llvm_amdgcn_sudot4_ss_clamp(i32 %a, i32 %b, i32 %c) {
|
|
; GFX11-LABEL: test_llvm_amdgcn_sudot4_ss_clamp:
|
|
; GFX11: ; %bb.0: ; %entry
|
|
; GFX11-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0) ; encoding: [0x00,0x00,0x89,0xbf]
|
|
; GFX11-NEXT: v_dot4_i32_iu8 v0, v0, v1, v2 neg_lo:[1,1,0] clamp ; encoding: [0x00,0xc0,0x16,0xcc,0x00,0x03,0x0a,0x7c]
|
|
; GFX11-NEXT: s_setpc_b64 s[30:31] ; encoding: [0x1e,0x48,0x80,0xbe]
|
|
entry:
|
|
%ret = call i32 @llvm.amdgcn.sudot4(i1 1, i32 %a, i1 1, i32 %b, i32 %c, i1 1)
|
|
ret i32 %ret
|
|
}
|