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
108 lines
4.6 KiB
LLVM
108 lines
4.6 KiB
LLVM
; RUN: opt < %s -S -passes=infer-address-spaces | FileCheck %s --check-prefix=INFER
|
|
; RUN: llc < %s -mtriple=nvptx64 -mcpu=sm_90 -mattr=+ptx80 | FileCheck %s --check-prefix=PTX
|
|
; RUN: %if ptxas-sm_90 && ptxas-isa-8.0 %{ llc < %s -mtriple=nvptx64 -mcpu=sm_90 -mattr=+ptx80 | %ptxas-verify -arch=sm_90 %}
|
|
|
|
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-unknown-unknown"
|
|
|
|
@constant_tensormap = addrspace(4) global [64 x i8] zeroinitializer, align 64
|
|
|
|
; Inference from const address space
|
|
define void @test_infer_const_from_cast() {
|
|
; INFER-LABEL: @test_infer_const_from_cast
|
|
; INFER: call void @llvm.nvvm.prefetch.tensormap.p4(ptr addrspace(4) @constant_tensormap)
|
|
; PTX-LABEL: .visible .func test_infer_const_from_cast(
|
|
; PTX: mov.b64 %rd{{[0-9]+}}, constant_tensormap;
|
|
; PTX: cvta.const.u64 %rd{{[0-9]+}}, %rd{{[0-9]+}};
|
|
; PTX: prefetch.tensormap [%rd{{[0-9]+}}];
|
|
entry:
|
|
%casted = addrspacecast ptr addrspace(4) @constant_tensormap to ptr
|
|
call void @llvm.nvvm.prefetch.tensormap.p0(ptr %casted)
|
|
ret void
|
|
}
|
|
|
|
; Cast from Const space to Generic
|
|
define void @test_const_to_generic_cast(ptr addrspace(4) %const_ptr) {
|
|
; INFER-LABEL: @test_const_to_generic_cast
|
|
; INFER: call void @llvm.nvvm.prefetch.tensormap.p4(ptr addrspace(4) %const_ptr)
|
|
; PTX-LABEL: .visible .func test_const_to_generic_cast(
|
|
; PTX: prefetch.const.tensormap [%rd{{[0-9]+}}];
|
|
entry:
|
|
%cast = addrspacecast ptr addrspace(4) %const_ptr to ptr
|
|
call void @llvm.nvvm.prefetch.tensormap.p0(ptr %cast)
|
|
ret void
|
|
}
|
|
|
|
; No inference possible
|
|
define void @test_no_inference_possible(ptr %generic_ptr) {
|
|
; INFER-LABEL: @test_no_inference_possible
|
|
; INFER: call void @llvm.nvvm.prefetch.tensormap.p0(ptr %generic_ptr)
|
|
; PTX-LABEL: .visible .func test_no_inference_possible(
|
|
; PTX: prefetch.tensormap [%rd{{[0-9]+}}];
|
|
entry:
|
|
call void @llvm.nvvm.prefetch.tensormap.p0(ptr %generic_ptr)
|
|
ret void
|
|
}
|
|
|
|
; Cast from Parameter space to Generic
|
|
define void @test_param_to_generic_cast(ptr addrspace(101) %param_ptr) {
|
|
; INFER-LABEL: @test_param_to_generic_cast
|
|
; INFER: call void @llvm.nvvm.prefetch.tensormap.p101(ptr addrspace(101) %param_ptr)
|
|
; PTX-LABEL: .visible .func test_param_to_generic_cast(
|
|
; PTX: prefetch.param.tensormap [%rd{{[0-9]+}}];
|
|
entry:
|
|
%cast = addrspacecast ptr addrspace(101) %param_ptr to ptr
|
|
call void @llvm.nvvm.prefetch.tensormap.p0(ptr %cast)
|
|
ret void
|
|
}
|
|
|
|
; Multiple casts in sequence
|
|
define void @test_infer_through_multiple_casts() {
|
|
; INFER-LABEL: @test_infer_through_multiple_casts
|
|
; INFER: call void @llvm.nvvm.prefetch.tensormap.p4(ptr addrspace(4) @constant_tensormap)
|
|
; PTX-LABEL: .visible .func test_infer_through_multiple_casts(
|
|
; PTX: mov.b64 %rd{{[0-9]+}}, constant_tensormap;
|
|
; PTX: cvta.const.u64 %rd{{[0-9]+}}, %rd{{[0-9]+}};
|
|
; PTX: prefetch.tensormap [%rd{{[0-9]+}}];
|
|
entry:
|
|
%cast1 = addrspacecast ptr addrspace(4) @constant_tensormap to ptr
|
|
%cast2 = addrspacecast ptr %cast1 to ptr addrspace(4)
|
|
%cast3 = addrspacecast ptr addrspace(4) %cast2 to ptr
|
|
call void @llvm.nvvm.prefetch.tensormap.p0(ptr %cast3)
|
|
ret void
|
|
}
|
|
|
|
; Kernel Function Test
|
|
; Cast from Param space to Generic
|
|
define ptx_kernel void @test_param_to_generic_cast_kernel(ptr addrspace(101) %param_ptr) {
|
|
; INFER-LABEL: @test_param_to_generic_cast_kernel
|
|
; INFER: call void @llvm.nvvm.prefetch.tensormap.p101(ptr addrspace(101) %param_ptr)
|
|
; PTX-LABEL: .visible .entry test_param_to_generic_cast_kernel(
|
|
; PTX: prefetch.param.tensormap [%rd{{[0-9]+}}];
|
|
entry:
|
|
%cast = addrspacecast ptr addrspace(101) %param_ptr to ptr
|
|
call void @llvm.nvvm.prefetch.tensormap.p0(ptr %cast)
|
|
ret void
|
|
}
|
|
|
|
; Kernel Function Test
|
|
; Multiple casts in sequence
|
|
define ptx_kernel void @test_infer_through_multiple_casts_kernel() {
|
|
; INFER-LABEL: @test_infer_through_multiple_casts_kernel
|
|
; INFER: call void @llvm.nvvm.prefetch.tensormap.p4(ptr addrspace(4) @constant_tensormap)
|
|
; PTX-LABEL: .visible .entry test_infer_through_multiple_casts_kernel(
|
|
; PTX: mov.b64 %rd{{[0-9]+}}, constant_tensormap;
|
|
; PTX: cvta.const.u64 %rd{{[0-9]+}}, %rd{{[0-9]+}};
|
|
; PTX: prefetch.tensormap [%rd{{[0-9]+}}];
|
|
entry:
|
|
%cast1 = addrspacecast ptr addrspace(4) @constant_tensormap to ptr
|
|
%cast2 = addrspacecast ptr %cast1 to ptr addrspace(4)
|
|
%cast3 = addrspacecast ptr addrspace(4) %cast2 to ptr
|
|
call void @llvm.nvvm.prefetch.tensormap.p0(ptr %cast3)
|
|
ret void
|
|
}
|
|
|
|
declare void @llvm.nvvm.prefetch.tensormap.p0(ptr)
|
|
declare void @llvm.nvvm.prefetch.tensormap.p4(ptr addrspace(4))
|
|
declare void @llvm.nvvm.prefetch.tensormap.p101(ptr addrspace(101))
|