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
79 lines
3.8 KiB
LLVM
79 lines
3.8 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
|
|
; RUN: llc < %s -mtriple=x86_64-- | FileCheck %s
|
|
|
|
; This testcase was written to demonstrate an instruction-selection problem,
|
|
; however it also happens to expose a limitation in the DAGCombiner's
|
|
; expression reassociation which causes it to miss opportunities for
|
|
; constant folding due to the intermediate adds having multiple uses.
|
|
; The Reassociate pass has similar limitations. If these limitations are
|
|
; fixed, the test commands above will need to be updated to expect fewer
|
|
; lea instructions.
|
|
|
|
@g0 = weak dso_local global [1000 x i32] zeroinitializer, align 32 ; <ptr> [#uses=8]
|
|
@g1 = weak dso_local global [1000 x i32] zeroinitializer, align 32 ; <ptr> [#uses=7]
|
|
|
|
define dso_local void @foo() {
|
|
; CHECK-LABEL: foo:
|
|
; CHECK: # %bb.0: # %entry
|
|
; CHECK-NEXT: movl g0(%rip), %eax
|
|
; CHECK-NEXT: movl g1(%rip), %ecx
|
|
; CHECK-NEXT: leal (%rax,%rcx), %edx
|
|
; CHECK-NEXT: leal 1(%rax,%rcx), %eax
|
|
; CHECK-NEXT: movl %eax, g0+4(%rip)
|
|
; CHECK-NEXT: movl g1+4(%rip), %eax
|
|
; CHECK-NEXT: leal 1(%rax,%rdx), %ecx
|
|
; CHECK-NEXT: leal 2(%rax,%rdx), %eax
|
|
; CHECK-NEXT: movl %eax, g0+8(%rip)
|
|
; CHECK-NEXT: movl g1+8(%rip), %eax
|
|
; CHECK-NEXT: leal 1(%rax,%rcx), %edx
|
|
; CHECK-NEXT: leal 2(%rax,%rcx), %eax
|
|
; CHECK-NEXT: movl %eax, g0+12(%rip)
|
|
; CHECK-NEXT: movl g1+12(%rip), %eax
|
|
; CHECK-NEXT: leal 1(%rax,%rdx), %ecx
|
|
; CHECK-NEXT: leal 2(%rax,%rdx), %eax
|
|
; CHECK-NEXT: movl %eax, g0+16(%rip)
|
|
; CHECK-NEXT: movl g1+16(%rip), %eax
|
|
; CHECK-NEXT: leal 1(%rax,%rcx), %edx
|
|
; CHECK-NEXT: leal 2(%rax,%rcx), %eax
|
|
; CHECK-NEXT: movl %eax, g0+20(%rip)
|
|
; CHECK-NEXT: movl g1+20(%rip), %eax
|
|
; CHECK-NEXT: leal 1(%rax,%rdx), %ecx
|
|
; CHECK-NEXT: leal 2(%rax,%rdx), %eax
|
|
; CHECK-NEXT: movl %eax, g0+24(%rip)
|
|
; CHECK-NEXT: movl g1+24(%rip), %eax
|
|
; CHECK-NEXT: leal 2(%rax,%rcx), %eax
|
|
; CHECK-NEXT: movl %eax, g0+28(%rip)
|
|
; CHECK-NEXT: retq
|
|
entry:
|
|
%tmp4 = load i32, ptr @g0 ; <i32> [#uses=1]
|
|
%tmp8 = load i32, ptr @g1 ; <i32> [#uses=1]
|
|
%tmp9 = add i32 %tmp4, 1 ; <i32> [#uses=1]
|
|
%tmp10 = add i32 %tmp9, %tmp8 ; <i32> [#uses=2]
|
|
store i32 %tmp10, ptr getelementptr ([1000 x i32], ptr @g0, i32 0, i32 1)
|
|
%tmp8.1 = load i32, ptr getelementptr ([1000 x i32], ptr @g1, i32 0, i32 1) ; <i32> [#uses=1]
|
|
%tmp9.1 = add i32 %tmp10, 1 ; <i32> [#uses=1]
|
|
%tmp10.1 = add i32 %tmp9.1, %tmp8.1 ; <i32> [#uses=2]
|
|
store i32 %tmp10.1, ptr getelementptr ([1000 x i32], ptr @g0, i32 0, i32 2)
|
|
%tmp8.2 = load i32, ptr getelementptr ([1000 x i32], ptr @g1, i32 0, i32 2) ; <i32> [#uses=1]
|
|
%tmp9.2 = add i32 %tmp10.1, 1 ; <i32> [#uses=1]
|
|
%tmp10.2 = add i32 %tmp9.2, %tmp8.2 ; <i32> [#uses=2]
|
|
store i32 %tmp10.2, ptr getelementptr ([1000 x i32], ptr @g0, i32 0, i32 3)
|
|
%tmp8.3 = load i32, ptr getelementptr ([1000 x i32], ptr @g1, i32 0, i32 3) ; <i32> [#uses=1]
|
|
%tmp9.3 = add i32 %tmp10.2, 1 ; <i32> [#uses=1]
|
|
%tmp10.3 = add i32 %tmp9.3, %tmp8.3 ; <i32> [#uses=2]
|
|
store i32 %tmp10.3, ptr getelementptr ([1000 x i32], ptr @g0, i32 0, i32 4)
|
|
%tmp8.4 = load i32, ptr getelementptr ([1000 x i32], ptr @g1, i32 0, i32 4) ; <i32> [#uses=1]
|
|
%tmp9.4 = add i32 %tmp10.3, 1 ; <i32> [#uses=1]
|
|
%tmp10.4 = add i32 %tmp9.4, %tmp8.4 ; <i32> [#uses=2]
|
|
store i32 %tmp10.4, ptr getelementptr ([1000 x i32], ptr @g0, i32 0, i32 5)
|
|
%tmp8.5 = load i32, ptr getelementptr ([1000 x i32], ptr @g1, i32 0, i32 5) ; <i32> [#uses=1]
|
|
%tmp9.5 = add i32 %tmp10.4, 1 ; <i32> [#uses=1]
|
|
%tmp10.5 = add i32 %tmp9.5, %tmp8.5 ; <i32> [#uses=2]
|
|
store i32 %tmp10.5, ptr getelementptr ([1000 x i32], ptr @g0, i32 0, i32 6)
|
|
%tmp8.6 = load i32, ptr getelementptr ([1000 x i32], ptr @g1, i32 0, i32 6) ; <i32> [#uses=1]
|
|
%tmp9.6 = add i32 %tmp10.5, 1 ; <i32> [#uses=1]
|
|
%tmp10.6 = add i32 %tmp9.6, %tmp8.6 ; <i32> [#uses=1]
|
|
store i32 %tmp10.6, ptr getelementptr ([1000 x i32], ptr @g0, i32 0, i32 7)
|
|
ret void
|
|
}
|