derek/TheRock
1
0 Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity
Files
1a182e2dac58879f215c0fad0f5f78c76dcafe56
TheRock/compiler/amd-llvm/llvm/test/CodeGen/AMDGPU/GlobalISel/combine-rsq.ll
T
derek.holloway 7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
Details
CodeQL / Analyze (python) (push) Canceled after 0s
Details
pre-commit / pre-commit (push) Canceled after 0s
Details
Multi-Arch CI / setup (push) Canceled after 0s
Details
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Details
Setup (Multi-Arch) / setup (push) Canceled after 0s
Details
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Details
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
Details
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Details
Multi-Arch CI / CI Summary (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Details
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Details
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Details
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Details
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Details
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
Details
Flatten subprojects into a single base monorepo
2026-09-02 16:34:47 -07:00

302 lines
11 KiB
LLVM
Raw Blame History

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -global-isel -new-reg-bank-select -mtriple=amdgcn -mcpu=gfx1010 %s -o - | FileCheck -check-prefix=GCN %s
define amdgpu_cs float @div_sqrt(float inreg %arg1) {
; GCN-LABEL: div_sqrt:
; GCN: ; %bb.0: ; %.entry
; GCN-NEXT: v_mul_f32_e64 v0, 0x4f800000, s0
; GCN-NEXT: v_cmp_gt_f32_e64 s1, 0xf800000, s0
; GCN-NEXT: v_readfirstlane_b32 s2, v0
; GCN-NEXT: s_cmp_lg_u32 s1, 0
; GCN-NEXT: s_cselect_b32 s1, 1, 0
; GCN-NEXT: s_cselect_b32 s2, s2, s0
; GCN-NEXT: v_sqrt_f32_e32 v0, s2
; GCN-NEXT: v_readfirstlane_b32 s3, v0
; GCN-NEXT: s_add_i32 s4, s3, -1
; GCN-NEXT: s_add_i32 s5, s3, 1
; GCN-NEXT: v_fma_f32 v1, -s4, v0, s2
; GCN-NEXT: v_fma_f32 v0, -s5, v0, s2
; GCN-NEXT: v_cmp_ge_f32_e32 vcc_lo, 0, v1
; GCN-NEXT: v_cmp_lt_f32_e64 s0, 0, v0
; GCN-NEXT: s_cmp_lg_u32 vcc_lo, 0
; GCN-NEXT: s_cselect_b32 s3, s4, s3
; GCN-NEXT: s_cmp_lg_u32 s0, 0
; GCN-NEXT: v_cmp_class_f32_e64 s4, s2, 0x260
; GCN-NEXT: s_cselect_b32 s0, s5, s3
; GCN-NEXT: s_cmp_lg_u32 s1, 0
; GCN-NEXT: v_mul_f32_e64 v0, 0x37800000, s0
; GCN-NEXT: v_readfirstlane_b32 s3, v0
; GCN-NEXT: s_cselect_b32 s0, s3, s0
; GCN-NEXT: s_cmp_lg_u32 s4, 0
; GCN-NEXT: s_cselect_b32 s0, s2, s0
; GCN-NEXT: v_rcp_f32_e32 v0, s0
; GCN-NEXT: ; return to shader part epilog
.entry:
%a = call float @llvm.sqrt.f32(float %arg1)
%b = fdiv afn float 1.000000e+00, %a
ret float %b
}
define amdgpu_cs float @sqrt_div(float inreg %arg1) {
; GCN-LABEL: sqrt_div:
; GCN: ; %bb.0: ; %.entry
; GCN-NEXT: v_rcp_f32_e32 v0, s0
; GCN-NEXT: v_mul_f32_e32 v1, 0x4f800000, v0
; GCN-NEXT: v_cmp_gt_f32_e32 vcc_lo, 0xf800000, v0
; GCN-NEXT: v_readfirstlane_b32 s0, v0
; GCN-NEXT: v_readfirstlane_b32 s1, v1
; GCN-NEXT: s_cmp_lg_u32 vcc_lo, 0
; GCN-NEXT: s_cselect_b32 s2, 1, 0
; GCN-NEXT: s_cselect_b32 s1, s1, s0
; GCN-NEXT: v_sqrt_f32_e32 v0, s1
; GCN-NEXT: v_readfirstlane_b32 s3, v0
; GCN-NEXT: s_add_i32 s4, s3, -1
; GCN-NEXT: s_add_i32 s5, s3, 1
; GCN-NEXT: v_fma_f32 v1, -s4, v0, s1
; GCN-NEXT: v_fma_f32 v0, -s5, v0, s1
; GCN-NEXT: v_cmp_ge_f32_e32 vcc_lo, 0, v1
; GCN-NEXT: v_cmp_lt_f32_e64 s0, 0, v0
; GCN-NEXT: s_cmp_lg_u32 vcc_lo, 0
; GCN-NEXT: s_cselect_b32 s3, s4, s3
; GCN-NEXT: s_cmp_lg_u32 s0, 0
; GCN-NEXT: v_cmp_class_f32_e64 s4, s1, 0x260
; GCN-NEXT: s_cselect_b32 s0, s5, s3
; GCN-NEXT: s_cmp_lg_u32 s2, 0
; GCN-NEXT: v_mul_f32_e64 v0, 0x37800000, s0
; GCN-NEXT: v_readfirstlane_b32 s3, v0
; GCN-NEXT: s_cselect_b32 s0, s3, s0
; GCN-NEXT: s_cmp_lg_u32 s4, 0
; GCN-NEXT: s_cselect_b32 s0, s1, s0
; GCN-NEXT: v_mov_b32_e32 v0, s0
; GCN-NEXT: ; return to shader part epilog
.entry:
%a = fdiv afn float 1.000000e+00, %arg1
%b = call float @llvm.sqrt.f32(float %a)
ret float %b
}
define amdgpu_cs float @rcp_sqrt(float inreg %arg1) {
; GCN-LABEL: rcp_sqrt:
; GCN: ; %bb.0: ; %.entry
; GCN-NEXT: v_mul_f32_e64 v0, 0x4f800000, s0
; GCN-NEXT: v_cmp_gt_f32_e64 s1, 0xf800000, s0
; GCN-NEXT: v_readfirstlane_b32 s2, v0
; GCN-NEXT: s_cmp_lg_u32 s1, 0
; GCN-NEXT: s_cselect_b32 s1, 1, 0
; GCN-NEXT: s_cselect_b32 s2, s2, s0
; GCN-NEXT: v_sqrt_f32_e32 v0, s2
; GCN-NEXT: v_readfirstlane_b32 s3, v0
; GCN-NEXT: s_add_i32 s4, s3, -1
; GCN-NEXT: s_add_i32 s5, s3, 1
; GCN-NEXT: v_fma_f32 v1, -s4, v0, s2
; GCN-NEXT: v_fma_f32 v0, -s5, v0, s2
; GCN-NEXT: v_cmp_ge_f32_e32 vcc_lo, 0, v1
; GCN-NEXT: v_cmp_lt_f32_e64 s0, 0, v0
; GCN-NEXT: s_cmp_lg_u32 vcc_lo, 0
; GCN-NEXT: s_cselect_b32 s3, s4, s3
; GCN-NEXT: s_cmp_lg_u32 s0, 0
; GCN-NEXT: v_cmp_class_f32_e64 s4, s2, 0x260
; GCN-NEXT: s_cselect_b32 s0, s5, s3
; GCN-NEXT: s_cmp_lg_u32 s1, 0
; GCN-NEXT: v_mul_f32_e64 v0, 0x37800000, s0
; GCN-NEXT: v_readfirstlane_b32 s3, v0
; GCN-NEXT: s_cselect_b32 s0, s3, s0
; GCN-NEXT: s_cmp_lg_u32 s4, 0
; GCN-NEXT: s_cselect_b32 s0, s2, s0
; GCN-NEXT: v_rcp_f32_e32 v0, s0
; GCN-NEXT: ; return to shader part epilog
.entry:
%a = call float @llvm.sqrt.f32(float %arg1)
%b = call float @llvm.amdgcn.rcp.f32(float %a)
ret float %b
}
define amdgpu_cs float @sqrt_rcp(float inreg %arg1) {
; GCN-LABEL: sqrt_rcp:
; GCN: ; %bb.0: ; %.entry
; GCN-NEXT: v_rcp_f32_e32 v0, s0
; GCN-NEXT: v_mul_f32_e32 v1, 0x4f800000, v0
; GCN-NEXT: v_cmp_gt_f32_e32 vcc_lo, 0xf800000, v0
; GCN-NEXT: v_readfirstlane_b32 s0, v0
; GCN-NEXT: v_readfirstlane_b32 s1, v1
; GCN-NEXT: s_cmp_lg_u32 vcc_lo, 0
; GCN-NEXT: s_cselect_b32 s2, 1, 0
; GCN-NEXT: s_cselect_b32 s1, s1, s0
; GCN-NEXT: v_sqrt_f32_e32 v0, s1
; GCN-NEXT: v_readfirstlane_b32 s3, v0
; GCN-NEXT: s_add_i32 s4, s3, -1
; GCN-NEXT: s_add_i32 s5, s3, 1
; GCN-NEXT: v_fma_f32 v1, -s4, v0, s1
; GCN-NEXT: v_fma_f32 v0, -s5, v0, s1
; GCN-NEXT: v_cmp_ge_f32_e32 vcc_lo, 0, v1
; GCN-NEXT: v_cmp_lt_f32_e64 s0, 0, v0
; GCN-NEXT: s_cmp_lg_u32 vcc_lo, 0
; GCN-NEXT: s_cselect_b32 s3, s4, s3
; GCN-NEXT: s_cmp_lg_u32 s0, 0
; GCN-NEXT: v_cmp_class_f32_e64 s4, s1, 0x260
; GCN-NEXT: s_cselect_b32 s0, s5, s3
; GCN-NEXT: s_cmp_lg_u32 s2, 0
; GCN-NEXT: v_mul_f32_e64 v0, 0x37800000, s0
; GCN-NEXT: v_readfirstlane_b32 s3, v0
; GCN-NEXT: s_cselect_b32 s0, s3, s0
; GCN-NEXT: s_cmp_lg_u32 s4, 0
; GCN-NEXT: s_cselect_b32 s0, s1, s0
; GCN-NEXT: v_mov_b32_e32 v0, s0
; GCN-NEXT: ; return to shader part epilog
.entry:
%a = call float @llvm.amdgcn.rcp.f32(float %arg1)
%b = call float @llvm.sqrt.f32(float %a)
ret float %b
}
define amdgpu_cs float @div_sqrt_contract(float inreg %arg1) {
; GCN-LABEL: div_sqrt_contract:
; GCN: ; %bb.0: ; %.entry
; GCN-NEXT: v_mul_f32_e64 v0, 0x4f800000, s0
; GCN-NEXT: v_cmp_gt_f32_e64 s1, 0xf800000, s0
; GCN-NEXT: v_readfirstlane_b32 s2, v0
; GCN-NEXT: s_cmp_lg_u32 s1, 0
; GCN-NEXT: s_cselect_b32 s1, 1, 0
; GCN-NEXT: s_cselect_b32 s2, s2, s0
; GCN-NEXT: v_sqrt_f32_e32 v0, s2
; GCN-NEXT: v_readfirstlane_b32 s3, v0
; GCN-NEXT: s_add_i32 s4, s3, -1
; GCN-NEXT: s_add_i32 s5, s3, 1
; GCN-NEXT: v_fma_f32 v1, -s4, v0, s2
; GCN-NEXT: v_fma_f32 v0, -s5, v0, s2
; GCN-NEXT: v_cmp_ge_f32_e32 vcc_lo, 0, v1
; GCN-NEXT: v_cmp_lt_f32_e64 s0, 0, v0
; GCN-NEXT: s_cmp_lg_u32 vcc_lo, 0
; GCN-NEXT: s_cselect_b32 s3, s4, s3
; GCN-NEXT: s_cmp_lg_u32 s0, 0
; GCN-NEXT: v_cmp_class_f32_e64 s4, s2, 0x260
; GCN-NEXT: s_cselect_b32 s0, s5, s3
; GCN-NEXT: s_cmp_lg_u32 s1, 0
; GCN-NEXT: v_mul_f32_e64 v0, 0x37800000, s0
; GCN-NEXT: v_readfirstlane_b32 s3, v0
; GCN-NEXT: s_cselect_b32 s0, s3, s0
; GCN-NEXT: s_cmp_lg_u32 s4, 0
; GCN-NEXT: s_cselect_b32 s0, s2, s0
; GCN-NEXT: v_rcp_f32_e32 v0, s0
; GCN-NEXT: ; return to shader part epilog
.entry:
%a = call contract float @llvm.sqrt.f32(float %arg1)
%b = fdiv afn contract float 1.000000e+00, %a
ret float %b
}
define amdgpu_cs float @sqrt_div_contract(float inreg %arg1) {
; GCN-LABEL: sqrt_div_contract:
; GCN: ; %bb.0: ; %.entry
; GCN-NEXT: v_rcp_f32_e32 v0, s0
; GCN-NEXT: v_mul_f32_e32 v1, 0x4f800000, v0
; GCN-NEXT: v_cmp_gt_f32_e32 vcc_lo, 0xf800000, v0
; GCN-NEXT: v_readfirstlane_b32 s0, v0
; GCN-NEXT: v_readfirstlane_b32 s1, v1
; GCN-NEXT: s_cmp_lg_u32 vcc_lo, 0
; GCN-NEXT: s_cselect_b32 s2, 1, 0
; GCN-NEXT: s_cselect_b32 s1, s1, s0
; GCN-NEXT: v_sqrt_f32_e32 v0, s1
; GCN-NEXT: v_readfirstlane_b32 s3, v0
; GCN-NEXT: s_add_i32 s4, s3, -1
; GCN-NEXT: s_add_i32 s5, s3, 1
; GCN-NEXT: v_fma_f32 v1, -s4, v0, s1
; GCN-NEXT: v_fma_f32 v0, -s5, v0, s1
; GCN-NEXT: v_cmp_ge_f32_e32 vcc_lo, 0, v1
; GCN-NEXT: v_cmp_lt_f32_e64 s0, 0, v0
; GCN-NEXT: s_cmp_lg_u32 vcc_lo, 0
; GCN-NEXT: s_cselect_b32 s3, s4, s3
; GCN-NEXT: s_cmp_lg_u32 s0, 0
; GCN-NEXT: v_cmp_class_f32_e64 s4, s1, 0x260
; GCN-NEXT: s_cselect_b32 s0, s5, s3
; GCN-NEXT: s_cmp_lg_u32 s2, 0
; GCN-NEXT: v_mul_f32_e64 v0, 0x37800000, s0
; GCN-NEXT: v_readfirstlane_b32 s3, v0
; GCN-NEXT: s_cselect_b32 s0, s3, s0
; GCN-NEXT: s_cmp_lg_u32 s4, 0
; GCN-NEXT: s_cselect_b32 s0, s1, s0
; GCN-NEXT: v_mov_b32_e32 v0, s0
; GCN-NEXT: ; return to shader part epilog
.entry:
%a = fdiv afn contract float 1.000000e+00, %arg1
%b = call contract float @llvm.sqrt.f32(float %a)
ret float %b
}
define amdgpu_cs float @rcp_sqrt_contract(float inreg %arg1) {
; GCN-LABEL: rcp_sqrt_contract:
; GCN: ; %bb.0: ; %.entry
; GCN-NEXT: v_mul_f32_e64 v0, 0x4f800000, s0
; GCN-NEXT: v_cmp_gt_f32_e64 s1, 0xf800000, s0
; GCN-NEXT: v_readfirstlane_b32 s2, v0
; GCN-NEXT: s_cmp_lg_u32 s1, 0
; GCN-NEXT: s_cselect_b32 s1, 1, 0
; GCN-NEXT: s_cselect_b32 s2, s2, s0
; GCN-NEXT: v_sqrt_f32_e32 v0, s2
; GCN-NEXT: v_readfirstlane_b32 s3, v0
; GCN-NEXT: s_add_i32 s4, s3, -1
; GCN-NEXT: s_add_i32 s5, s3, 1
; GCN-NEXT: v_fma_f32 v1, -s4, v0, s2
; GCN-NEXT: v_fma_f32 v0, -s5, v0, s2
; GCN-NEXT: v_cmp_ge_f32_e32 vcc_lo, 0, v1
; GCN-NEXT: v_cmp_lt_f32_e64 s0, 0, v0
; GCN-NEXT: s_cmp_lg_u32 vcc_lo, 0
; GCN-NEXT: s_cselect_b32 s3, s4, s3
; GCN-NEXT: s_cmp_lg_u32 s0, 0
; GCN-NEXT: v_cmp_class_f32_e64 s4, s2, 0x260
; GCN-NEXT: s_cselect_b32 s0, s5, s3
; GCN-NEXT: s_cmp_lg_u32 s1, 0
; GCN-NEXT: v_mul_f32_e64 v0, 0x37800000, s0
; GCN-NEXT: v_readfirstlane_b32 s3, v0
; GCN-NEXT: s_cselect_b32 s0, s3, s0
; GCN-NEXT: s_cmp_lg_u32 s4, 0
; GCN-NEXT: s_cselect_b32 s0, s2, s0
; GCN-NEXT: v_rcp_f32_e32 v0, s0
; GCN-NEXT: ; return to shader part epilog
.entry:
%a = call contract float @llvm.sqrt.f32(float %arg1)
%b = call contract float @llvm.amdgcn.rcp.f32(float %a)
ret float %b
}
define amdgpu_cs float @sqrt_rcp_contract(float inreg %arg1) {
; GCN-LABEL: sqrt_rcp_contract:
; GCN: ; %bb.0: ; %.entry
; GCN-NEXT: v_rcp_f32_e32 v0, s0
; GCN-NEXT: v_mul_f32_e32 v1, 0x4f800000, v0
; GCN-NEXT: v_cmp_gt_f32_e32 vcc_lo, 0xf800000, v0
; GCN-NEXT: v_readfirstlane_b32 s0, v0
; GCN-NEXT: v_readfirstlane_b32 s1, v1
; GCN-NEXT: s_cmp_lg_u32 vcc_lo, 0
; GCN-NEXT: s_cselect_b32 s2, 1, 0
; GCN-NEXT: s_cselect_b32 s1, s1, s0
; GCN-NEXT: v_sqrt_f32_e32 v0, s1
; GCN-NEXT: v_readfirstlane_b32 s3, v0
; GCN-NEXT: s_add_i32 s4, s3, -1
; GCN-NEXT: s_add_i32 s5, s3, 1
; GCN-NEXT: v_fma_f32 v1, -s4, v0, s1
; GCN-NEXT: v_fma_f32 v0, -s5, v0, s1
; GCN-NEXT: v_cmp_ge_f32_e32 vcc_lo, 0, v1
; GCN-NEXT: v_cmp_lt_f32_e64 s0, 0, v0
; GCN-NEXT: s_cmp_lg_u32 vcc_lo, 0
; GCN-NEXT: s_cselect_b32 s3, s4, s3
; GCN-NEXT: s_cmp_lg_u32 s0, 0
; GCN-NEXT: v_cmp_class_f32_e64 s4, s1, 0x260
; GCN-NEXT: s_cselect_b32 s0, s5, s3
; GCN-NEXT: s_cmp_lg_u32 s2, 0
; GCN-NEXT: v_mul_f32_e64 v0, 0x37800000, s0
; GCN-NEXT: v_readfirstlane_b32 s3, v0
; GCN-NEXT: s_cselect_b32 s0, s3, s0
; GCN-NEXT: s_cmp_lg_u32 s4, 0
; GCN-NEXT: s_cselect_b32 s0, s1, s0
; GCN-NEXT: v_mov_b32_e32 v0, s0
; GCN-NEXT: ; return to shader part epilog
.entry:
%a = call contract float @llvm.amdgcn.rcp.f32(float %arg1)
%b = call contract float @llvm.sqrt.f32(float %a)
ret float %b
}
declare float @llvm.sqrt.f32(float)
declare float @llvm.amdgcn.rcp.f32(float)
Reference in new issue View Git Blame Copy Permalink