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
84 lines
2.8 KiB
LLVM
84 lines
2.8 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
|
|
; RUN: llc < %s -mtriple=nvptx -mcpu=sm_60 -mattr=+ptx73 | FileCheck %s --check-prefix=CHECK-32
|
|
; RUN: llc < %s -mtriple=nvptx64 -mcpu=sm_60 -mattr=+ptx73 | FileCheck %s --check-prefix=CHECK-64
|
|
; RUN: llc < %s -mtriple=nvptx64 -nvptx-short-ptr -mcpu=sm_60 -mattr=+ptx73 | FileCheck %s --check-prefix=CHECK-MIXED
|
|
; RUN: %if ptxas-sm_60 && ptxas-isa-7.3 %{ llc < %s -mtriple=nvptx64 -mcpu=sm_60 -mattr=+ptx73 | %ptxas-verify -arch=sm_60 %}
|
|
|
|
target triple = "nvptx64-nvidia-cuda"
|
|
|
|
define ptr @test_save() {
|
|
; CHECK-32-LABEL: test_save(
|
|
; CHECK-32: {
|
|
; CHECK-32-NEXT: .reg .b32 %r<3>;
|
|
; CHECK-32-EMPTY:
|
|
; CHECK-32-NEXT: // %bb.0:
|
|
; CHECK-32-NEXT: stacksave.u32 %r1;
|
|
; CHECK-32-NEXT: cvta.local.u32 %r2, %r1;
|
|
; CHECK-32-NEXT: st.param.b32 [func_retval0], %r2;
|
|
; CHECK-32-NEXT: ret;
|
|
;
|
|
; CHECK-64-LABEL: test_save(
|
|
; CHECK-64: {
|
|
; CHECK-64-NEXT: .reg .b64 %rd<3>;
|
|
; CHECK-64-EMPTY:
|
|
; CHECK-64-NEXT: // %bb.0:
|
|
; CHECK-64-NEXT: stacksave.u64 %rd1;
|
|
; CHECK-64-NEXT: cvta.local.u64 %rd2, %rd1;
|
|
; CHECK-64-NEXT: st.param.b64 [func_retval0], %rd2;
|
|
; CHECK-64-NEXT: ret;
|
|
;
|
|
; CHECK-MIXED-LABEL: test_save(
|
|
; CHECK-MIXED: {
|
|
; CHECK-MIXED-NEXT: .reg .b32 %r<2>;
|
|
; CHECK-MIXED-NEXT: .reg .b64 %rd<3>;
|
|
; CHECK-MIXED-EMPTY:
|
|
; CHECK-MIXED-NEXT: // %bb.0:
|
|
; CHECK-MIXED-NEXT: stacksave.u32 %r1;
|
|
; CHECK-MIXED-NEXT: cvt.u64.u32 %rd1, %r1;
|
|
; CHECK-MIXED-NEXT: cvta.local.u64 %rd2, %rd1;
|
|
; CHECK-MIXED-NEXT: st.param.b64 [func_retval0], %rd2;
|
|
; CHECK-MIXED-NEXT: ret;
|
|
%1 = call ptr @llvm.stacksave()
|
|
ret ptr %1
|
|
}
|
|
|
|
|
|
define void @test_restore(ptr %p) {
|
|
; CHECK-32-LABEL: test_restore(
|
|
; CHECK-32: {
|
|
; CHECK-32-NEXT: .reg .b32 %r<3>;
|
|
; CHECK-32-EMPTY:
|
|
; CHECK-32-NEXT: // %bb.0:
|
|
; CHECK-32-NEXT: ld.param.b32 %r1, [test_restore_param_0];
|
|
; CHECK-32-NEXT: cvta.to.local.u32 %r2, %r1;
|
|
; CHECK-32-NEXT: stackrestore.u32 %r2;
|
|
; CHECK-32-NEXT: ret;
|
|
;
|
|
; CHECK-64-LABEL: test_restore(
|
|
; CHECK-64: {
|
|
; CHECK-64-NEXT: .reg .b64 %rd<3>;
|
|
; CHECK-64-EMPTY:
|
|
; CHECK-64-NEXT: // %bb.0:
|
|
; CHECK-64-NEXT: ld.param.b64 %rd1, [test_restore_param_0];
|
|
; CHECK-64-NEXT: cvta.to.local.u64 %rd2, %rd1;
|
|
; CHECK-64-NEXT: stackrestore.u64 %rd2;
|
|
; CHECK-64-NEXT: ret;
|
|
;
|
|
; CHECK-MIXED-LABEL: test_restore(
|
|
; CHECK-MIXED: {
|
|
; CHECK-MIXED-NEXT: .reg .b32 %r<2>;
|
|
; CHECK-MIXED-NEXT: .reg .b64 %rd<3>;
|
|
; CHECK-MIXED-EMPTY:
|
|
; CHECK-MIXED-NEXT: // %bb.0:
|
|
; CHECK-MIXED-NEXT: ld.param.b64 %rd1, [test_restore_param_0];
|
|
; CHECK-MIXED-NEXT: cvta.to.local.u64 %rd2, %rd1;
|
|
; CHECK-MIXED-NEXT: cvt.u32.u64 %r1, %rd2;
|
|
; CHECK-MIXED-NEXT: stackrestore.u32 %r1;
|
|
; CHECK-MIXED-NEXT: ret;
|
|
call void @llvm.stackrestore(ptr %p)
|
|
ret void
|
|
}
|
|
|
|
declare ptr @llvm.stacksave()
|
|
declare void @llvm.stackrestore(ptr)
|