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
145 lines
3.0 KiB
LLVM
145 lines
3.0 KiB
LLVM
; RUN: opt < %s -passes=sink -S | FileCheck %s
|
|
; RUN: opt < %s -aa-pipeline='basic-aa' -passes='sink' -S | FileCheck %s
|
|
|
|
@A = external global i32
|
|
@B = external global i32
|
|
|
|
; Sink should sink the load past the store (which doesn't overlap) into
|
|
; the block that uses it.
|
|
|
|
; CHECK-LABEL: @foo(
|
|
; CHECK: true:
|
|
; CHECK-NEXT: %l = load i32, ptr @A
|
|
; CHECK-NEXT: ret i32 %l
|
|
|
|
define i32 @foo(i1 %z) {
|
|
%l = load i32, ptr @A
|
|
store i32 0, ptr @B
|
|
br i1 %z, label %true, label %false
|
|
true:
|
|
ret i32 %l
|
|
false:
|
|
ret i32 0
|
|
}
|
|
|
|
; But don't sink load volatiles...
|
|
|
|
; CHECK-LABEL: @foo2(
|
|
; CHECK: load volatile
|
|
; CHECK-NEXT: store i32
|
|
|
|
define i32 @foo2(i1 %z) {
|
|
%l = load volatile i32, ptr @A
|
|
store i32 0, ptr @B
|
|
br i1 %z, label %true, label %false
|
|
true:
|
|
ret i32 %l
|
|
false:
|
|
ret i32 0
|
|
}
|
|
|
|
; Sink to the nearest post-dominator
|
|
|
|
; CHECK-LABEL: @diamond(
|
|
; CHECK: X:
|
|
; CHECK-NEXT: phi
|
|
; CHECK-NEXT: mul nsw
|
|
; CHECK-NEXT: sub
|
|
|
|
define i32 @diamond(i32 %a, i32 %b, i32 %c) {
|
|
%1 = mul nsw i32 %c, %b
|
|
%2 = icmp sgt i32 %a, 0
|
|
br i1 %2, label %B0, label %B1
|
|
|
|
B0: ; preds = %0
|
|
br label %X
|
|
|
|
B1: ; preds = %0
|
|
br label %X
|
|
|
|
X: ; preds = %5, %3
|
|
%.01 = phi i32 [ %c, %B0 ], [ %a, %B1 ]
|
|
%R = sub i32 %1, %.01
|
|
ret i32 %R
|
|
}
|
|
|
|
; We shouldn't sink constant sized allocas from the entry block, since CodeGen
|
|
; interprets allocas outside the entry block as dynamically sized stack objects.
|
|
|
|
; CHECK-LABEL: @alloca_nosink
|
|
; CHECK: entry:
|
|
; CHECK-NEXT: alloca
|
|
define i32 @alloca_nosink(i32 %a, i32 %b) {
|
|
entry:
|
|
%0 = alloca i32
|
|
%1 = icmp ne i32 %a, 0
|
|
br i1 %1, label %if, label %endif
|
|
|
|
if:
|
|
%2 = getelementptr i32, ptr %0, i32 1
|
|
store i32 0, ptr %0
|
|
store i32 1, ptr %2
|
|
%3 = getelementptr i32, ptr %0, i32 %b
|
|
%4 = load i32, ptr %3
|
|
ret i32 %4
|
|
|
|
endif:
|
|
ret i32 0
|
|
}
|
|
|
|
; Make sure we sink dynamic sized allocas
|
|
|
|
; CHECK-LABEL: @alloca_sink_dynamic
|
|
; CHECK: entry:
|
|
; CHECK-NOT: alloca
|
|
; CHECK: if:
|
|
; CHECK-NEXT: alloca
|
|
define i32 @alloca_sink_dynamic(i32 %a, i32 %b, i32 %size) {
|
|
entry:
|
|
%0 = alloca i32, i32 %size
|
|
%1 = icmp ne i32 %a, 0
|
|
br i1 %1, label %if, label %endif
|
|
|
|
if:
|
|
%2 = getelementptr i32, ptr %0, i32 1
|
|
store i32 0, ptr %0
|
|
store i32 1, ptr %2
|
|
%3 = getelementptr i32, ptr %0, i32 %b
|
|
%4 = load i32, ptr %3
|
|
ret i32 %4
|
|
|
|
endif:
|
|
ret i32 0
|
|
}
|
|
|
|
; We also want to sink allocas that are not in the entry block. These
|
|
; will already be considered as dynamically sized stack objects, so sinking
|
|
; them does no further damage.
|
|
|
|
; CHECK-LABEL: @alloca_sink_nonentry
|
|
; CHECK: if0:
|
|
; CHECK-NOT: alloca
|
|
; CHECK: if:
|
|
; CHECK-NEXT: alloca
|
|
define i32 @alloca_sink_nonentry(i32 %a, i32 %b, i32 %c) {
|
|
entry:
|
|
%cmp = icmp ne i32 %c, 0
|
|
br i1 %cmp, label %endif, label %if0
|
|
|
|
if0:
|
|
%0 = alloca i32
|
|
%1 = icmp ne i32 %a, 0
|
|
br i1 %1, label %if, label %endif
|
|
|
|
if:
|
|
%2 = getelementptr i32, ptr %0, i32 1
|
|
store i32 0, ptr %0
|
|
store i32 1, ptr %2
|
|
%3 = getelementptr i32, ptr %0, i32 %b
|
|
%4 = load i32, ptr %3
|
|
ret i32 %4
|
|
|
|
endif:
|
|
ret i32 0
|
|
}
|