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
91 lines
3.5 KiB
LLVM
91 lines
3.5 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
|
|
; RUN: llc < %s -mtriple=nvptx64 -mcpu=sm_20 -verify-machineinstrs | FileCheck %s
|
|
; RUN: llc < %s -mtriple=nvptx64 -mcpu=sm_30 -verify-machineinstrs | FileCheck %s
|
|
; RUN: %if ptxas %{ llc < %s -mtriple=nvptx64 -mcpu=sm_20 -verify-machineinstrs | %ptxas-verify %}
|
|
; RUN: %if ptxas %{ llc < %s -mtriple=nvptx64 -mcpu=sm_30 -verify-machineinstrs | %ptxas-verify %}
|
|
|
|
|
|
target triple = "nvptx-unknown-cuda"
|
|
|
|
declare { float, float, float, float } @llvm.nvvm.tex.unified.1d.v4f32.s32(i64, i32)
|
|
declare i64 @llvm.nvvm.texsurf.handle.internal.p1(ptr addrspace(1))
|
|
|
|
define ptx_kernel void @foo(i64 %img, ptr %red, i32 %idx) {
|
|
; CHECK-LABEL: foo(
|
|
; CHECK: {
|
|
; CHECK-NEXT: .reg .b32 %r<6>;
|
|
; CHECK-NEXT: .reg .b64 %rd<4>;
|
|
; CHECK-EMPTY:
|
|
; CHECK-NEXT: // %bb.0:
|
|
; CHECK-NEXT: ld.param.b64 %rd1, [foo_param_0];
|
|
; CHECK-NEXT: ld.param.b64 %rd2, [foo_param_1];
|
|
; CHECK-NEXT: cvta.to.global.u64 %rd3, %rd2;
|
|
; CHECK-NEXT: ld.param.b32 %r1, [foo_param_2];
|
|
; CHECK-NEXT: tex.1d.v4.f32.s32 {%r2, %r3, %r4, %r5}, [%rd1, {%r1}];
|
|
; CHECK-NEXT: st.global.b32 [%rd3], %r2;
|
|
; CHECK-NEXT: ret;
|
|
%val = tail call { float, float, float, float } @llvm.nvvm.tex.unified.1d.v4f32.s32(i64 %img, i32 %idx)
|
|
%ret = extractvalue { float, float, float, float } %val, 0
|
|
store float %ret, ptr %red
|
|
ret void
|
|
}
|
|
|
|
|
|
@tex0 = internal addrspace(1) global i64 0, align 8
|
|
|
|
define ptx_kernel void @bar(ptr %red, i32 %idx) {
|
|
; CHECK-LABEL: bar(
|
|
; CHECK: {
|
|
; CHECK-NEXT: .reg .b32 %r<6>;
|
|
; CHECK-NEXT: .reg .b64 %rd<3>;
|
|
; CHECK-EMPTY:
|
|
; CHECK-NEXT: // %bb.0:
|
|
; CHECK-NEXT: ld.param.b64 %rd1, [bar_param_0];
|
|
; CHECK-NEXT: cvta.to.global.u64 %rd2, %rd1;
|
|
; CHECK-NEXT: ld.param.b32 %r1, [bar_param_1];
|
|
; CHECK-NEXT: tex.1d.v4.f32.s32 {%r2, %r3, %r4, %r5}, [tex0, {%r1}];
|
|
; CHECK-NEXT: st.global.b32 [%rd2], %r2;
|
|
; CHECK-NEXT: ret;
|
|
%texHandle = tail call i64 @llvm.nvvm.texsurf.handle.internal.p1(ptr addrspace(1) @tex0)
|
|
%val = tail call { float, float, float, float } @llvm.nvvm.tex.unified.1d.v4f32.s32(i64 %texHandle, i32 %idx)
|
|
%ret = extractvalue { float, float, float, float } %val, 0
|
|
store float %ret, ptr %red
|
|
ret void
|
|
}
|
|
|
|
declare float @texfunc(i64)
|
|
|
|
define ptx_kernel void @baz(ptr %red, i32 %idx) {
|
|
; CHECK-LABEL: baz(
|
|
; CHECK: {
|
|
; CHECK-NEXT: .reg .b32 %r<8>;
|
|
; CHECK-NEXT: .reg .b64 %rd<4>;
|
|
; CHECK-EMPTY:
|
|
; CHECK-NEXT: // %bb.0:
|
|
; CHECK-NEXT: ld.param.b64 %rd1, [baz_param_0];
|
|
; CHECK-NEXT: cvta.to.global.u64 %rd2, %rd1;
|
|
; CHECK-NEXT: ld.param.b32 %r1, [baz_param_1];
|
|
; CHECK-NEXT: mov.u64 %rd3, tex0;
|
|
; CHECK-NEXT: tex.1d.v4.f32.s32 {%r2, %r3, %r4, %r5}, [tex0, {%r1}];
|
|
; CHECK-NEXT: { // callseq 0, 0
|
|
; CHECK-NEXT: .param .b64 param0;
|
|
; CHECK-NEXT: .param .b32 retval0;
|
|
; CHECK-NEXT: st.param.b64 [param0], %rd3;
|
|
; CHECK-NEXT: call.uni (retval0), texfunc, (param0);
|
|
; CHECK-NEXT: ld.param.b32 %r6, [retval0];
|
|
; CHECK-NEXT: } // callseq 0
|
|
; CHECK-NEXT: add.rn.f32 %r7, %r2, %r6;
|
|
; CHECK-NEXT: st.global.b32 [%rd2], %r7;
|
|
; CHECK-NEXT: ret;
|
|
%texHandle = tail call i64 @llvm.nvvm.texsurf.handle.internal.p1(ptr addrspace(1) @tex0)
|
|
%val = tail call { float, float, float, float } @llvm.nvvm.tex.unified.1d.v4f32.s32(i64 %texHandle, i32 %idx)
|
|
%ret = extractvalue { float, float, float, float } %val, 0
|
|
%texcall = tail call float @texfunc(i64 %texHandle)
|
|
%ret2 = fadd float %ret, %texcall
|
|
store float %ret2, ptr %red
|
|
ret void
|
|
}
|
|
|
|
!nvvm.annotations = !{!1}
|
|
!1 = !{ptr addrspace(1) @tex0, !"texture", i32 1}
|