Files
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
Flatten subprojects into a single base monorepo
2026-09-02 16:34:47 -07:00

158 lines
4.1 KiB
LLVM

; Test load-and-trap instructions (LAT/LGAT)
;
; RUN: llc < %s -mtriple=s390x-linux-gnu -mcpu=zEC12 | FileCheck %s
declare void @llvm.trap()
; Check LAT with no displacement.
define i32 @f1(ptr %ptr) {
; CHECK-LABEL: f1:
; CHECK: lat %r2, 0(%r2)
; CHECK: br %r14
entry:
%val = load i32, ptr %ptr
%cmp = icmp eq i32 %val, 0
br i1 %cmp, label %if.then, label %if.end
if.then: ; preds = %entry
tail call void @llvm.trap()
unreachable
if.end: ; preds = %entry
ret i32 %val
}
; Check the high end of the LAT range.
define i32 @f2(ptr %src) {
; CHECK-LABEL: f2:
; CHECK: lat %r2, 524284(%r2)
; CHECK: br %r14
%ptr = getelementptr i32, ptr %src, i64 131071
%val = load i32, ptr %ptr
%cmp = icmp eq i32 %val, 0
br i1 %cmp, label %if.then, label %if.end
if.then: ; preds = %entry
tail call void @llvm.trap()
unreachable
if.end: ; preds = %entry
ret i32 %val
}
; Check the next word up, which needs separate address logic.
; Other sequences besides this one would be OK.
define i32 @f3(ptr %src) {
; CHECK-LABEL: f3:
; CHECK: agfi %r2, 524288
; CHECK: lat %r2, 0(%r2)
; CHECK: br %r14
%ptr = getelementptr i32, ptr %src, i64 131072
%val = load i32, ptr %ptr
%cmp = icmp eq i32 %val, 0
br i1 %cmp, label %if.then, label %if.end
if.then: ; preds = %entry
tail call void @llvm.trap()
unreachable
if.end: ; preds = %entry
ret i32 %val
}
; Check that LAT allows an index.
define i32 @f4(i64 %src, i64 %index) {
; CHECK-LABEL: f4:
; CHECK: lat %r2, 524287(%r3,%r2)
; CHECK: br %r14
%add1 = add i64 %src, %index
%add2 = add i64 %add1, 524287
%ptr = inttoptr i64 %add2 to ptr
%val = load i32, ptr %ptr
%cmp = icmp eq i32 %val, 0
br i1 %cmp, label %if.then, label %if.end
if.then: ; preds = %entry
tail call void @llvm.trap()
unreachable
if.end: ; preds = %entry
ret i32 %val
}
; Check LGAT with no displacement.
define i64 @f5(ptr %ptr) {
; CHECK-LABEL: f5:
; CHECK: lgat %r2, 0(%r2)
; CHECK: br %r14
entry:
%val = load i64, ptr %ptr
%cmp = icmp eq i64 %val, 0
br i1 %cmp, label %if.then, label %if.end
if.then: ; preds = %entry
tail call void @llvm.trap()
unreachable
if.end: ; preds = %entry
ret i64 %val
}
; Check the high end of the LGAT range.
define i64 @f6(ptr %src) {
; CHECK-LABEL: f6:
; CHECK: lgat %r2, 524280(%r2)
; CHECK: br %r14
%ptr = getelementptr i64, ptr %src, i64 65535
%val = load i64, ptr %ptr
%cmp = icmp eq i64 %val, 0
br i1 %cmp, label %if.then, label %if.end
if.then: ; preds = %entry
tail call void @llvm.trap()
unreachable
if.end: ; preds = %entry
ret i64 %val
}
; Check the next word up, which needs separate address logic.
; Other sequences besides this one would be OK.
define i64 @f7(ptr %src) {
; CHECK-LABEL: f7:
; CHECK: agfi %r2, 524288
; CHECK: lgat %r2, 0(%r2)
; CHECK: br %r14
%ptr = getelementptr i64, ptr %src, i64 65536
%val = load i64, ptr %ptr
%cmp = icmp eq i64 %val, 0
br i1 %cmp, label %if.then, label %if.end
if.then: ; preds = %entry
tail call void @llvm.trap()
unreachable
if.end: ; preds = %entry
ret i64 %val
}
; Check that LGAT allows an index.
define i64 @f8(i64 %src, i64 %index) {
; CHECK-LABEL: f8:
; CHECK: lgat %r2, 524287(%r3,%r2)
; CHECK: br %r14
%add1 = add i64 %src, %index
%add2 = add i64 %add1, 524287
%ptr = inttoptr i64 %add2 to ptr
%val = load i64, ptr %ptr
%cmp = icmp eq i64 %val, 0
br i1 %cmp, label %if.then, label %if.end
if.then: ; preds = %entry
tail call void @llvm.trap()
unreachable
if.end: ; preds = %entry
ret i64 %val
}