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
60 lines
1.6 KiB
LLVM
60 lines
1.6 KiB
LLVM
; RUN: llc -mtriple=hexagon < %s | FileCheck %s
|
|
|
|
; DAG combiner folds sequences of shifts, which can sometimes obscure
|
|
; optimization opportunities. For example
|
|
;
|
|
; unsigned int c(unsigned int b, unsigned int *a) {
|
|
; unsigned int bitidx = b >> 5;
|
|
; return a[bitidx];
|
|
; }
|
|
;
|
|
; produces
|
|
; (add x (shl (srl y 5) 2))
|
|
; which is then folded into
|
|
; (add x (and (srl y 3) 1FFFFFFC))
|
|
;
|
|
; That results in a constant-extended and:
|
|
; r0 = and(##536870908,lsr(r0,#3))
|
|
; r0 = memw(r1+r0<<#0)
|
|
; whereas
|
|
; r0 = lsr(r0,#5)
|
|
; r0 = memw(r1+r0<<#2)
|
|
; is more desirable.
|
|
|
|
target triple = "hexagon"
|
|
|
|
; CHECK-LABEL: load_0
|
|
; CHECK: memw(r{{[0-9]+}}+r{{[0-9]}}<<#2)
|
|
define i32 @load_0(i32 %b, ptr nocapture readonly %a) #0 {
|
|
entry:
|
|
%shr = lshr i32 %b, 5
|
|
%arrayidx = getelementptr inbounds i32, ptr %a, i32 %shr
|
|
%0 = load i32, ptr %arrayidx, align 4
|
|
ret i32 %0
|
|
}
|
|
|
|
; This would require r0<<#3, which is not legal.
|
|
; CHECK-LABEL: load_1
|
|
; CHECK: memw(r{{[0-9]+}}+r{{[0-9]}}<<#0)
|
|
define i32 @load_1(i32 %b, ptr nocapture readonly %a) #0 {
|
|
entry:
|
|
%shr = lshr i32 %b, 5
|
|
%arrayidx = getelementptr inbounds [3 x i32], ptr %a, i32 %shr, i32 0
|
|
%0 = load i32, ptr %arrayidx, align 4
|
|
ret i32 %0
|
|
}
|
|
|
|
; CHECK-LABEL: store_0
|
|
; CHECK: memw(r{{[0-9]+}}+r{{[0-9]}}<<#2)
|
|
define void @store_0(i32 %b, ptr nocapture %a, i32 %v) #1 {
|
|
entry:
|
|
%shr = lshr i32 %b, 5
|
|
%arrayidx = getelementptr inbounds i32, ptr %a, i32 %shr
|
|
store i32 %v, ptr %arrayidx, align 4
|
|
ret void
|
|
}
|
|
|
|
attributes #0 = { norecurse nounwind readonly "target-cpu"="hexagonv60" "target-features"="-hvx,-long-calls" }
|
|
attributes #1 = { norecurse nounwind "target-cpu"="hexagonv60" "target-features"="-hvx,-long-calls" }
|
|
|