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
63 lines
2.5 KiB
LLVM
63 lines
2.5 KiB
LLVM
; RUN: llc < %s -mtriple=nvptx64-nvidia-cuda -mcpu=sm_20 -O3 | FileCheck %s
|
|
; RUN: %if ptxas %{ llc < %s -mtriple=nvptx64 -mcpu=sm_20 | %ptxas-verify %}
|
|
|
|
target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v16:16:16-v32:32:32-v64:64:64-v128:128:128-n16:32:64"
|
|
target triple = "nvptx64-nvidia-cuda"
|
|
|
|
; Verify that both %input and %output are converted to global pointers and then
|
|
; addrspacecast'ed back to the original type.
|
|
define ptx_kernel void @kernel(ptr %input, ptr %output) {
|
|
; CHECK-LABEL: .visible .entry kernel(
|
|
; CHECK: cvta.to.global.u64
|
|
; CHECK: cvta.to.global.u64
|
|
%1 = load float, ptr %input, align 4
|
|
; CHECK: ld.global.b32
|
|
store float %1, ptr %output, align 4
|
|
; CHECK: st.global.b32
|
|
ret void
|
|
}
|
|
|
|
define ptx_kernel void @kernel2(ptr addrspace(1) %input, ptr addrspace(1) %output) {
|
|
; CHECK-LABEL: .visible .entry kernel2(
|
|
; CHECK-NOT: cvta.to.global.u64
|
|
%1 = load float, ptr addrspace(1) %input, align 4
|
|
; CHECK: ld.global.b32
|
|
store float %1, ptr addrspace(1) %output, align 4
|
|
; CHECK: st.global.b32
|
|
ret void
|
|
}
|
|
|
|
%struct.S = type { ptr, ptr }
|
|
|
|
define ptx_kernel void @ptr_in_byval_kernel(ptr byval(%struct.S) %input, ptr %output) {
|
|
; CHECK-LABEL: .visible .entry ptr_in_byval_kernel(
|
|
; CHECK: ld.param.b64 %[[optr:rd.*]], [ptr_in_byval_kernel_param_1]
|
|
; CHECK: cvta.to.global.u64 %[[optr_g:.*]], %[[optr]];
|
|
; CHECK: ld.param.b64 %[[iptr:rd.*]], [ptr_in_byval_kernel_param_0+8]
|
|
; CHECK: cvta.to.global.u64 %[[iptr_g:.*]], %[[iptr]];
|
|
%b_ptr = getelementptr inbounds %struct.S, ptr %input, i64 0, i32 1
|
|
%b = load ptr, ptr %b_ptr, align 8
|
|
%v = load i32, ptr %b, align 4
|
|
; CHECK: ld.global.b32 %[[val:.*]], [%[[iptr_g]]]
|
|
store i32 %v, ptr %output, align 4
|
|
; CHECK: st.global.b32 [%[[optr_g]]], %[[val]]
|
|
ret void
|
|
}
|
|
|
|
; Regular functions lower byval arguments differently. We need to make
|
|
; sure that we're loading byval argument data using [symbol+offset].
|
|
; There's also no assumption that all pointers within are in global space.
|
|
define void @ptr_in_byval_func(ptr byval(%struct.S) %input, ptr %output) {
|
|
; CHECK-LABEL: .visible .func ptr_in_byval_func(
|
|
; CHECK: ld.param.b64 %[[optr:rd.*]], [ptr_in_byval_func_param_1]
|
|
; CHECK: ld.param.b64 %[[iptr:rd.*]], [ptr_in_byval_func_param_0+8]
|
|
%b_ptr = getelementptr inbounds %struct.S, ptr %input, i64 0, i32 1
|
|
%b = load ptr, ptr %b_ptr, align 8
|
|
%v = load i32, ptr %b, align 4
|
|
; CHECK: ld.b32 %[[val:.*]], [%[[iptr]]]
|
|
store i32 %v, ptr %output, align 4
|
|
; CHECK: st.b32 [%[[optr]]], %[[val]]
|
|
ret void
|
|
}
|
|
|