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
48 lines
2.2 KiB
MLIR
48 lines
2.2 KiB
MLIR
// RUN: mlir-translate -mlir-to-llvmir %s | FileCheck %s
|
||
|
||
// f32 sqrt — all 8 forms (4 rounding modes × 2 ftz states).
|
||
llvm.func @sqrt_f32(%a : f32) -> f32 {
|
||
// CHECK-LABEL: define float @sqrt_f32(float %0) {
|
||
// CHECK: call float @llvm.nvvm.sqrt.rn.f(float %{{.*}})
|
||
// CHECK: call float @llvm.nvvm.sqrt.rz.f(float %{{.*}})
|
||
// CHECK: call float @llvm.nvvm.sqrt.rm.f(float %{{.*}})
|
||
// CHECK: call float @llvm.nvvm.sqrt.rp.f(float %{{.*}})
|
||
// CHECK: call float @llvm.nvvm.sqrt.rn.ftz.f(float %{{.*}})
|
||
// CHECK: call float @llvm.nvvm.sqrt.rz.ftz.f(float %{{.*}})
|
||
// CHECK: call float @llvm.nvvm.sqrt.rm.ftz.f(float %{{.*}})
|
||
// CHECK: call float @llvm.nvvm.sqrt.rp.ftz.f(float %{{.*}})
|
||
%r1 = nvvm.sqrt %a {rnd = #nvvm.fp_rnd_mode<rn>} : f32
|
||
%r2 = nvvm.sqrt %r1 {rnd = #nvvm.fp_rnd_mode<rz>} : f32
|
||
%r3 = nvvm.sqrt %r2 {rnd = #nvvm.fp_rnd_mode<rm>} : f32
|
||
%r4 = nvvm.sqrt %r3 {rnd = #nvvm.fp_rnd_mode<rp>} : f32
|
||
%r5 = nvvm.sqrt %r4 {rnd = #nvvm.fp_rnd_mode<rn>, ftz = true} : f32
|
||
%r6 = nvvm.sqrt %r5 {rnd = #nvvm.fp_rnd_mode<rz>, ftz = true} : f32
|
||
%r7 = nvvm.sqrt %r6 {rnd = #nvvm.fp_rnd_mode<rm>, ftz = true} : f32
|
||
%r8 = nvvm.sqrt %r7 {rnd = #nvvm.fp_rnd_mode<rp>, ftz = true} : f32
|
||
llvm.return %r8 : f32
|
||
}
|
||
|
||
// f64 sqrt — all 4 forms (4 rounding modes, no ftz).
|
||
llvm.func @sqrt_f64(%a : f64) -> f64 {
|
||
// CHECK-LABEL: define double @sqrt_f64(double %0) {
|
||
// CHECK: call double @llvm.nvvm.sqrt.rn.d(double %{{.*}})
|
||
// CHECK: call double @llvm.nvvm.sqrt.rz.d(double %{{.*}})
|
||
// CHECK: call double @llvm.nvvm.sqrt.rm.d(double %{{.*}})
|
||
// CHECK: call double @llvm.nvvm.sqrt.rp.d(double %{{.*}})
|
||
%r1 = nvvm.sqrt %a {rnd = #nvvm.fp_rnd_mode<rn>} : f64
|
||
%r2 = nvvm.sqrt %r1 {rnd = #nvvm.fp_rnd_mode<rz>} : f64
|
||
%r3 = nvvm.sqrt %r2 {rnd = #nvvm.fp_rnd_mode<rm>} : f64
|
||
%r4 = nvvm.sqrt %r3 {rnd = #nvvm.fp_rnd_mode<rp>} : f64
|
||
llvm.return %r4 : f64
|
||
}
|
||
|
||
// sqrt.approx — 2 forms.
|
||
llvm.func @sqrt_approx(%a : f32) -> f32 {
|
||
// CHECK-LABEL: define float @sqrt_approx(float %0) {
|
||
// CHECK: call float @llvm.nvvm.sqrt.approx.f(float %{{.*}})
|
||
// CHECK: call float @llvm.nvvm.sqrt.approx.ftz.f(float %{{.*}})
|
||
%r1 = nvvm.sqrt.approx %a : f32
|
||
%r2 = nvvm.sqrt.approx %r1 {ftz = true} : f32
|
||
llvm.return %r2 : f32
|
||
}
|