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
76 lines
2.7 KiB
LLVM
76 lines
2.7 KiB
LLVM
; RUN: llc < %s -mtriple=nvptx64 -mcpu=sm_53 -mattr=+ptx42 | FileCheck %s
|
|
; RUN: %if ptxas-sm_53 && ptxas-isa-4.2 %{ llc < %s -mtriple=nvptx64 -mcpu=sm_53 -mattr=+ptx42 | %ptxas-verify -arch=sm_53 %}
|
|
|
|
declare half @llvm.nvvm.fma.rn.f16(half, half, half)
|
|
declare half @llvm.nvvm.fma.rn.ftz.f16(half, half, half)
|
|
declare half @llvm.nvvm.fma.rn.sat.f16(half, half, half)
|
|
declare half @llvm.nvvm.fma.rn.ftz.sat.f16(half, half, half)
|
|
declare <2 x half> @llvm.nvvm.fma.rn.f16x2(<2 x half>, <2 x half>, <2 x half>)
|
|
declare <2 x half> @llvm.nvvm.fma.rn.ftz.f16x2(<2 x half>, <2 x half>, <2 x half>)
|
|
declare <2 x half> @llvm.nvvm.fma.rn.sat.f16x2(<2 x half>, <2 x half>, <2 x half>)
|
|
declare <2 x half> @llvm.nvvm.fma.rn.ftz.sat.f16x2(<2 x half>, <2 x half>, <2 x half>)
|
|
|
|
; CHECK-LABEL: fma_rn_f16
|
|
define half @fma_rn_f16(half %0, half %1, half %2) {
|
|
; CHECK-NOT: call
|
|
; CHECK: fma.rn.f16
|
|
%res = call half @llvm.nvvm.fma.rn.f16(half %0, half %1, half %2)
|
|
ret half %res
|
|
}
|
|
|
|
; CHECK-LABEL: fma_rn_ftz_f16
|
|
define half @fma_rn_ftz_f16(half %0, half %1, half %2) {
|
|
; CHECK-NOT: call
|
|
; CHECK: fma.rn.ftz.f16
|
|
%res = call half @llvm.nvvm.fma.rn.ftz.f16(half %0, half %1, half %2)
|
|
ret half %res
|
|
}
|
|
|
|
; CHECK-LABEL: fma_rn_sat_f16
|
|
define half @fma_rn_sat_f16(half %0, half %1, half %2) {
|
|
; CHECK-NOT: call
|
|
; CHECK: fma.rn.sat.f16
|
|
%res = call half @llvm.nvvm.fma.rn.sat.f16(half %0, half %1, half %2)
|
|
ret half %res
|
|
}
|
|
|
|
; CHECK-LABEL: fma_rn_ftz_sat_f16
|
|
define half @fma_rn_ftz_sat_f16(half %0, half %1, half %2) {
|
|
; CHECK-NOT: call
|
|
; CHECK: fma.rn.ftz.sat.f16
|
|
%res = call half @llvm.nvvm.fma.rn.ftz.sat.f16(half %0, half %1, half %2)
|
|
ret half %res
|
|
}
|
|
|
|
; CHECK-LABEL: fma_rn_f16x2
|
|
define <2 x half> @fma_rn_f16x2(<2 x half> %0, <2 x half> %1, <2 x half> %2) {
|
|
; CHECK-NOT: call
|
|
; CHECK: fma.rn.f16x2
|
|
%res = call <2 x half> @llvm.nvvm.fma.rn.f16x2(<2 x half> %0, <2 x half> %1, <2 x half> %2)
|
|
ret <2 x half> %res
|
|
}
|
|
|
|
; CHECK-LABEL: fma_rn_ftz_f16x2
|
|
define <2 x half> @fma_rn_ftz_f16x2(<2 x half> %0, <2 x half> %1, <2 x half> %2) {
|
|
; CHECK-NOT: call
|
|
; CHECK: fma.rn.ftz.f16x2
|
|
%res = call <2 x half> @llvm.nvvm.fma.rn.ftz.f16x2(<2 x half> %0, <2 x half> %1, <2 x half> %2)
|
|
ret <2 x half> %res
|
|
}
|
|
|
|
; CHECK-LABEL: fma_rn_sat_f16x2
|
|
define <2 x half> @fma_rn_sat_f16x2(<2 x half> %0, <2 x half> %1, <2 x half> %2) {
|
|
; CHECK-NOT: call
|
|
; CHECK: fma.rn.sat.f16x2
|
|
%res = call <2 x half> @llvm.nvvm.fma.rn.sat.f16x2(<2 x half> %0, <2 x half> %1, <2 x half> %2)
|
|
ret <2 x half> %res
|
|
}
|
|
|
|
; CHECK-LABEL: fma_rn_ftz_sat_f16x2
|
|
define <2 x half> @fma_rn_ftz_sat_f16x2(<2 x half> %0, <2 x half> %1, <2 x half> %2) {
|
|
; CHECK-NOT: call
|
|
; CHECK: fma.rn.ftz.sat.f16x2
|
|
%res = call <2 x half> @llvm.nvvm.fma.rn.ftz.sat.f16x2(<2 x half> %0, <2 x half> %1, <2 x half> %2)
|
|
ret <2 x half> %res
|
|
}
|