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
54 lines
2.4 KiB
LLVM
54 lines
2.4 KiB
LLVM
; RUN: llc -march=hexagon -mhvx < %s | FileCheck %s
|
|
|
|
; Check that Hexagon Vector Combine do not create vmem for adjacent
|
|
; loads less than 4.
|
|
|
|
; Test 1: We have three adjacent loads. By default, we won't produce vmem
|
|
; for loads less than 4.
|
|
; CHECK-LABEL: @foo1
|
|
; CHECK: vmemu(
|
|
; CHECK-NOT: vmem(
|
|
|
|
define <32 x i32> @foo1(ptr %iptr) #0 align 32 {
|
|
entry:
|
|
%0 = load <32 x i32>, ptr %iptr, align 1
|
|
%add.ptr = getelementptr inbounds i8, ptr %iptr, i32 128
|
|
%1 = load <32 x i32>, ptr %add.ptr, align 1
|
|
%add.ptr2 = getelementptr inbounds i8, ptr %iptr, i32 256
|
|
%2 = load <32 x i32>, ptr %add.ptr2, align 1
|
|
%3 = tail call <64 x i32> @llvm.hexagon.V6.vcombine.128B(<32 x i32> %1, <32 x i32> %0)
|
|
%4 = tail call <32 x i32> @llvm.hexagon.V6.vconv.hf.qf32.128B(<64 x i32> %3)
|
|
%5 = tail call <64 x i32> @llvm.hexagon.V6.vcombine.128B(<32 x i32> %2, <32 x i32> %1)
|
|
%6 = tail call <32 x i32> @llvm.hexagon.V6.vconv.hf.qf32.128B(<64 x i32> %5)
|
|
%7 = tail call <64 x i32> @llvm.hexagon.V6.vcombine.128B(<32 x i32> %4, <32 x i32> %6)
|
|
%8 = tail call <32 x i32> @llvm.hexagon.V6.vconv.hf.qf32.128B(<64 x i32> %7)
|
|
ret <32 x i32> %8
|
|
}
|
|
|
|
; Test 2: We have four adjacent loads. Thus, we will produce vmem.
|
|
; CHECK-LABEL: @foo2
|
|
; CHECK: vmem(
|
|
; CHECK-NOT: vmemu
|
|
|
|
define <32 x i32> @foo2(ptr %iptr) #0 align 32 {
|
|
entry:
|
|
%0 = load <32 x i32>, ptr %iptr, align 1
|
|
%add.ptr = getelementptr inbounds i8, ptr %iptr, i32 128
|
|
%1 = load <32 x i32>, ptr %add.ptr, align 1
|
|
%add.ptr2 = getelementptr inbounds i8, ptr %iptr, i32 256
|
|
%2 = load <32 x i32>, ptr %add.ptr2, align 1
|
|
%add.ptr4 = getelementptr inbounds i8, ptr %iptr, i32 512
|
|
%3 = load <32 x i32>, ptr %add.ptr4, align 1
|
|
%4 = tail call <64 x i32> @llvm.hexagon.V6.vcombine.128B(<32 x i32> %1, <32 x i32> %0)
|
|
%5 = tail call <32 x i32> @llvm.hexagon.V6.vconv.hf.qf32.128B(<64 x i32> %4)
|
|
%6 = tail call <64 x i32> @llvm.hexagon.V6.vcombine.128B(<32 x i32> %3, <32 x i32> %2)
|
|
%7 = tail call <32 x i32> @llvm.hexagon.V6.vconv.hf.qf32.128B(<64 x i32> %6)
|
|
%8 = tail call <64 x i32> @llvm.hexagon.V6.vcombine.128B(<32 x i32> %5, <32 x i32> %7)
|
|
%9 = tail call <32 x i32> @llvm.hexagon.V6.vconv.hf.qf32.128B(<64 x i32> %8)
|
|
ret <32 x i32> %9
|
|
}
|
|
|
|
declare <32 x i32> @llvm.hexagon.V6.vconv.hf.qf32.128B(<64 x i32>) #1
|
|
declare <64 x i32> @llvm.hexagon.V6.vcombine.128B(<32 x i32>, <32 x i32>) #1
|
|
attributes #0 = {"target-features"="+hvx-length128b,+hvxv75" }
|