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
47 lines
2.3 KiB
LLVM
47 lines
2.3 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 6
|
|
|
|
; RUN: opt -passes=instcombine < %s | llc -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx1200 | FileCheck -check-prefixes=GFX12,GFX12-SDAG %s
|
|
; RUN: opt -passes=instcombine < %s | llc -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx1250 | FileCheck -check-prefixes=GFX1250,GFX1250-SDAG %s
|
|
; RUN: opt -passes=instcombine < %s | llc -global-isel -new-reg-bank-select -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx1200 | FileCheck -check-prefixes=GFX12,GFX12-GISEL %s
|
|
; RUN: opt -passes=instcombine < %s | llc -global-isel -new-reg-bank-select -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx1250 | FileCheck -check-prefixes=GFX1250,GFX1250-GISEL %s
|
|
|
|
; The address calculation can be simplified and folded because of known bits of mbcnt.
|
|
define amdgpu_ps <2 x float> @global_load_scale_add_foldable_knownbits(ptr addrspace(1) inreg %sbase) {
|
|
%v = tail call i32 @llvm.amdgcn.mbcnt.lo(i32 -1, i32 0)
|
|
%mul = shl i32 %v, 1
|
|
%add = add i32 %mul, 32
|
|
%zext.offset = zext i32 %add to i64
|
|
%gep = getelementptr inbounds float, ptr addrspace(1) %sbase, i64 %zext.offset
|
|
%load = load <2 x float>, ptr addrspace(1) %gep
|
|
ret <2 x float> %load
|
|
}
|
|
|
|
; The nsw/nuw helps the folding of address addition.
|
|
define amdgpu_ps <2 x float> @global_load_scale_add_foldable_nowrap(ptr addrspace(1) inreg %sbase, i32 %v) {
|
|
%mul = shl nsw nuw i32 %v, 3
|
|
%add = add nsw nuw i32 %mul, 128
|
|
%zext.offset = zext i32 %add to i64
|
|
%gep = getelementptr inbounds i8, ptr addrspace(1) %sbase, i64 %zext.offset
|
|
%load = load <2 x float>, ptr addrspace(1) %gep
|
|
ret <2 x float> %load
|
|
}
|
|
|
|
; Address calculation cannot be folded because possible overflow during addition.
|
|
define amdgpu_ps <2 x float> @global_load_scale_add_unfoldable(ptr addrspace(1) inreg %sbase, i32 %v) {
|
|
%mul = shl i32 %v, 3
|
|
%add = add i32 %mul, 128
|
|
%zext.offset = zext i32 %add to i64
|
|
%gep = getelementptr inbounds i8, ptr addrspace(1) %sbase, i64 %zext.offset
|
|
%load = load <2 x float>, ptr addrspace(1) %gep
|
|
ret <2 x float> %load
|
|
}
|
|
|
|
declare i32 @llvm.amdgcn.mbcnt.lo(i32, i32)
|
|
;; NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line:
|
|
; GFX12: {{.*}}
|
|
; GFX12-GISEL: {{.*}}
|
|
; GFX12-SDAG: {{.*}}
|
|
; GFX1250: {{.*}}
|
|
; GFX1250-GISEL: {{.*}}
|
|
; GFX1250-SDAG: {{.*}}
|