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/RISCV/GlobalISel/rv32zbb.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

850 lines
23 KiB
LLVM
Raw Blame History

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=riscv32 -global-isel -verify-machineinstrs < %s \
; RUN: | FileCheck %s -check-prefixes=CHECK,RV32I
; RUN: llc -mtriple=riscv32 -global-isel -mattr=+zbb -verify-machineinstrs < %s \
; RUN: | FileCheck %s -check-prefixes=CHECK,RV32ZBB
define i32 @ctlz_i32(i32 %a) nounwind {
; RV32I-LABEL: ctlz_i32:
; RV32I: # %bb.0:
; RV32I-NEXT: beqz a0, .LBB0_2
; RV32I-NEXT: # %bb.1: # %cond.false
; RV32I-NEXT: srli a1, a0, 1
; RV32I-NEXT: lui a2, 349525
; RV32I-NEXT: or a0, a0, a1
; RV32I-NEXT: addi a1, a2, 1365
; RV32I-NEXT: srli a2, a0, 2
; RV32I-NEXT: or a0, a0, a2
; RV32I-NEXT: srli a2, a0, 4
; RV32I-NEXT: or a0, a0, a2
; RV32I-NEXT: srli a2, a0, 8
; RV32I-NEXT: or a0, a0, a2
; RV32I-NEXT: srli a2, a0, 16
; RV32I-NEXT: or a0, a0, a2
; RV32I-NEXT: srli a2, a0, 1
; RV32I-NEXT: and a1, a2, a1
; RV32I-NEXT: lui a2, 209715
; RV32I-NEXT: addi a2, a2, 819
; RV32I-NEXT: sub a0, a0, a1
; RV32I-NEXT: srli a1, a0, 2
; RV32I-NEXT: and a0, a0, a2
; RV32I-NEXT: and a1, a1, a2
; RV32I-NEXT: lui a2, 61681
; RV32I-NEXT: addi a2, a2, -241
; RV32I-NEXT: add a0, a1, a0
; RV32I-NEXT: srli a1, a0, 4
; RV32I-NEXT: add a0, a1, a0
; RV32I-NEXT: and a0, a0, a2
; RV32I-NEXT: slli a1, a0, 8
; RV32I-NEXT: add a0, a0, a1
; RV32I-NEXT: slli a1, a0, 16
; RV32I-NEXT: add a0, a0, a1
; RV32I-NEXT: srli a0, a0, 24
; RV32I-NEXT: li a1, 32
; RV32I-NEXT: sub a0, a1, a0
; RV32I-NEXT: ret
; RV32I-NEXT: .LBB0_2:
; RV32I-NEXT: li a0, 32
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: ctlz_i32:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: clz a0, a0
; RV32ZBB-NEXT: ret
%1 = call i32 @llvm.ctlz.i32(i32 %a, i1 false)
ret i32 %1
}
define i64 @ctlz_i64(i64 %a) nounwind {
; RV32I-LABEL: ctlz_i64:
; RV32I: # %bb.0:
; RV32I-NEXT: or a2, a0, a1
; RV32I-NEXT: beqz a2, .LBB1_3
; RV32I-NEXT: # %bb.1: # %cond.false
; RV32I-NEXT: lui a2, 349525
; RV32I-NEXT: lui a3, 209715
; RV32I-NEXT: lui a6, 61681
; RV32I-NEXT: addi a5, a2, 1365
; RV32I-NEXT: addi a4, a3, 819
; RV32I-NEXT: addi a3, a6, -241
; RV32I-NEXT: li a2, 32
; RV32I-NEXT: beqz a1, .LBB1_4
; RV32I-NEXT: # %bb.2: # %cond.false
; RV32I-NEXT: srli a0, a1, 1
; RV32I-NEXT: or a0, a1, a0
; RV32I-NEXT: srli a1, a0, 2
; RV32I-NEXT: or a0, a0, a1
; RV32I-NEXT: srli a1, a0, 4
; RV32I-NEXT: or a0, a0, a1
; RV32I-NEXT: srli a1, a0, 8
; RV32I-NEXT: or a0, a0, a1
; RV32I-NEXT: srli a1, a0, 16
; RV32I-NEXT: or a0, a0, a1
; RV32I-NEXT: srli a1, a0, 1
; RV32I-NEXT: and a1, a1, a5
; RV32I-NEXT: sub a0, a0, a1
; RV32I-NEXT: srli a1, a0, 2
; RV32I-NEXT: and a1, a1, a4
; RV32I-NEXT: and a0, a0, a4
; RV32I-NEXT: add a0, a1, a0
; RV32I-NEXT: srli a1, a0, 4
; RV32I-NEXT: add a0, a1, a0
; RV32I-NEXT: and a0, a0, a3
; RV32I-NEXT: slli a1, a0, 8
; RV32I-NEXT: add a0, a0, a1
; RV32I-NEXT: slli a1, a0, 16
; RV32I-NEXT: add a0, a0, a1
; RV32I-NEXT: srli a0, a0, 24
; RV32I-NEXT: sub a0, a2, a0
; RV32I-NEXT: li a1, 0
; RV32I-NEXT: ret
; RV32I-NEXT: .LBB1_3:
; RV32I-NEXT: li a1, 0
; RV32I-NEXT: li a0, 64
; RV32I-NEXT: ret
; RV32I-NEXT: .LBB1_4:
; RV32I-NEXT: srli a1, a0, 1
; RV32I-NEXT: or a0, a0, a1
; RV32I-NEXT: srli a1, a0, 2
; RV32I-NEXT: or a0, a0, a1
; RV32I-NEXT: srli a1, a0, 4
; RV32I-NEXT: or a0, a0, a1
; RV32I-NEXT: srli a1, a0, 8
; RV32I-NEXT: or a0, a0, a1
; RV32I-NEXT: srli a1, a0, 16
; RV32I-NEXT: or a0, a0, a1
; RV32I-NEXT: srli a1, a0, 1
; RV32I-NEXT: and a1, a1, a5
; RV32I-NEXT: sub a0, a0, a1
; RV32I-NEXT: srli a1, a0, 2
; RV32I-NEXT: and a1, a1, a4
; RV32I-NEXT: and a0, a0, a4
; RV32I-NEXT: add a0, a1, a0
; RV32I-NEXT: srli a1, a0, 4
; RV32I-NEXT: add a0, a1, a0
; RV32I-NEXT: and a0, a0, a3
; RV32I-NEXT: slli a1, a0, 8
; RV32I-NEXT: add a0, a0, a1
; RV32I-NEXT: slli a1, a0, 16
; RV32I-NEXT: add a0, a0, a1
; RV32I-NEXT: srli a0, a0, 24
; RV32I-NEXT: sub a2, a2, a0
; RV32I-NEXT: addi a0, a2, 32
; RV32I-NEXT: li a1, 0
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: ctlz_i64:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: beqz a1, .LBB1_2
; RV32ZBB-NEXT: # %bb.1:
; RV32ZBB-NEXT: clz a0, a1
; RV32ZBB-NEXT: li a1, 0
; RV32ZBB-NEXT: ret
; RV32ZBB-NEXT: .LBB1_2:
; RV32ZBB-NEXT: clz a0, a0
; RV32ZBB-NEXT: addi a0, a0, 32
; RV32ZBB-NEXT: li a1, 0
; RV32ZBB-NEXT: ret
%1 = call i64 @llvm.ctlz.i64(i64 %a, i1 false)
ret i64 %1
}
define i32 @cttz_i32(i32 %a) nounwind {
; RV32I-LABEL: cttz_i32:
; RV32I: # %bb.0:
; RV32I-NEXT: beqz a0, .LBB2_2
; RV32I-NEXT: # %bb.1: # %cond.false
; RV32I-NEXT: not a1, a0
; RV32I-NEXT: addi a0, a0, -1
; RV32I-NEXT: lui a2, 349525
; RV32I-NEXT: and a0, a1, a0
; RV32I-NEXT: addi a1, a2, 1365
; RV32I-NEXT: srli a2, a0, 1
; RV32I-NEXT: and a1, a2, a1
; RV32I-NEXT: lui a2, 209715
; RV32I-NEXT: addi a2, a2, 819
; RV32I-NEXT: sub a0, a0, a1
; RV32I-NEXT: srli a1, a0, 2
; RV32I-NEXT: and a0, a0, a2
; RV32I-NEXT: and a1, a1, a2
; RV32I-NEXT: lui a2, 61681
; RV32I-NEXT: add a0, a1, a0
; RV32I-NEXT: srli a1, a0, 4
; RV32I-NEXT: add a0, a1, a0
; RV32I-NEXT: addi a1, a2, -241
; RV32I-NEXT: and a0, a0, a1
; RV32I-NEXT: slli a1, a0, 8
; RV32I-NEXT: add a0, a0, a1
; RV32I-NEXT: slli a1, a0, 16
; RV32I-NEXT: add a0, a0, a1
; RV32I-NEXT: srli a0, a0, 24
; RV32I-NEXT: ret
; RV32I-NEXT: .LBB2_2:
; RV32I-NEXT: li a0, 32
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: cttz_i32:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: ctz a0, a0
; RV32ZBB-NEXT: ret
%1 = call i32 @llvm.cttz.i32(i32 %a, i1 false)
ret i32 %1
}
define i64 @cttz_i64(i64 %a) nounwind {
; RV32I-LABEL: cttz_i64:
; RV32I: # %bb.0:
; RV32I-NEXT: or a2, a0, a1
; RV32I-NEXT: beqz a2, .LBB3_3
; RV32I-NEXT: # %bb.1: # %cond.false
; RV32I-NEXT: lui a2, 349525
; RV32I-NEXT: lui a3, 209715
; RV32I-NEXT: lui a5, 61681
; RV32I-NEXT: addi a4, a2, 1365
; RV32I-NEXT: addi a3, a3, 819
; RV32I-NEXT: addi a2, a5, -241
; RV32I-NEXT: beqz a0, .LBB3_4
; RV32I-NEXT: # %bb.2: # %cond.false
; RV32I-NEXT: not a1, a0
; RV32I-NEXT: addi a0, a0, -1
; RV32I-NEXT: and a0, a1, a0
; RV32I-NEXT: srli a1, a0, 1
; RV32I-NEXT: and a1, a1, a4
; RV32I-NEXT: sub a0, a0, a1
; RV32I-NEXT: srli a1, a0, 2
; RV32I-NEXT: and a1, a1, a3
; RV32I-NEXT: and a0, a0, a3
; RV32I-NEXT: add a0, a1, a0
; RV32I-NEXT: srli a1, a0, 4
; RV32I-NEXT: add a0, a1, a0
; RV32I-NEXT: and a0, a0, a2
; RV32I-NEXT: slli a1, a0, 8
; RV32I-NEXT: add a0, a0, a1
; RV32I-NEXT: slli a1, a0, 16
; RV32I-NEXT: add a0, a0, a1
; RV32I-NEXT: srli a0, a0, 24
; RV32I-NEXT: li a1, 0
; RV32I-NEXT: ret
; RV32I-NEXT: .LBB3_3:
; RV32I-NEXT: li a1, 0
; RV32I-NEXT: li a0, 64
; RV32I-NEXT: ret
; RV32I-NEXT: .LBB3_4:
; RV32I-NEXT: not a0, a1
; RV32I-NEXT: addi a1, a1, -1
; RV32I-NEXT: and a0, a0, a1
; RV32I-NEXT: srli a1, a0, 1
; RV32I-NEXT: and a1, a1, a4
; RV32I-NEXT: sub a0, a0, a1
; RV32I-NEXT: srli a1, a0, 2
; RV32I-NEXT: and a1, a1, a3
; RV32I-NEXT: and a0, a0, a3
; RV32I-NEXT: add a0, a1, a0
; RV32I-NEXT: srli a1, a0, 4
; RV32I-NEXT: add a0, a1, a0
; RV32I-NEXT: and a0, a0, a2
; RV32I-NEXT: slli a1, a0, 8
; RV32I-NEXT: add a0, a0, a1
; RV32I-NEXT: slli a1, a0, 16
; RV32I-NEXT: add a0, a0, a1
; RV32I-NEXT: srli a0, a0, 24
; RV32I-NEXT: addi a0, a0, 32
; RV32I-NEXT: li a1, 0
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: cttz_i64:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: beqz a0, .LBB3_2
; RV32ZBB-NEXT: # %bb.1:
; RV32ZBB-NEXT: ctz a0, a0
; RV32ZBB-NEXT: li a1, 0
; RV32ZBB-NEXT: ret
; RV32ZBB-NEXT: .LBB3_2:
; RV32ZBB-NEXT: ctz a0, a1
; RV32ZBB-NEXT: addi a0, a0, 32
; RV32ZBB-NEXT: li a1, 0
; RV32ZBB-NEXT: ret
%1 = call i64 @llvm.cttz.i64(i64 %a, i1 false)
ret i64 %1
}
define i32 @ctpop_i32(i32 %a) nounwind {
; RV32I-LABEL: ctpop_i32:
; RV32I: # %bb.0:
; RV32I-NEXT: srli a1, a0, 1
; RV32I-NEXT: lui a2, 349525
; RV32I-NEXT: addi a2, a2, 1365
; RV32I-NEXT: and a1, a1, a2
; RV32I-NEXT: lui a2, 209715
; RV32I-NEXT: addi a2, a2, 819
; RV32I-NEXT: sub a0, a0, a1
; RV32I-NEXT: srli a1, a0, 2
; RV32I-NEXT: and a0, a0, a2
; RV32I-NEXT: and a1, a1, a2
; RV32I-NEXT: lui a2, 61681
; RV32I-NEXT: add a0, a1, a0
; RV32I-NEXT: srli a1, a0, 4
; RV32I-NEXT: add a0, a1, a0
; RV32I-NEXT: addi a1, a2, -241
; RV32I-NEXT: and a0, a0, a1
; RV32I-NEXT: slli a1, a0, 8
; RV32I-NEXT: add a0, a0, a1
; RV32I-NEXT: slli a1, a0, 16
; RV32I-NEXT: add a0, a0, a1
; RV32I-NEXT: srli a0, a0, 24
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: ctpop_i32:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: cpop a0, a0
; RV32ZBB-NEXT: ret
%1 = call i32 @llvm.ctpop.i32(i32 %a)
ret i32 %1
}
define i64 @ctpop_i64(i64 %a) nounwind {
; RV32I-LABEL: ctpop_i64:
; RV32I: # %bb.0:
; RV32I-NEXT: srli a2, a0, 1
; RV32I-NEXT: lui a3, 349525
; RV32I-NEXT: lui a4, 209715
; RV32I-NEXT: srli a5, a1, 1
; RV32I-NEXT: addi a3, a3, 1365
; RV32I-NEXT: and a2, a2, a3
; RV32I-NEXT: and a3, a5, a3
; RV32I-NEXT: lui a5, 61681
; RV32I-NEXT: addi a4, a4, 819
; RV32I-NEXT: addi a5, a5, -241
; RV32I-NEXT: sub a0, a0, a2
; RV32I-NEXT: sub a1, a1, a3
; RV32I-NEXT: srli a2, a0, 2
; RV32I-NEXT: and a0, a0, a4
; RV32I-NEXT: srli a3, a1, 2
; RV32I-NEXT: and a1, a1, a4
; RV32I-NEXT: and a2, a2, a4
; RV32I-NEXT: and a3, a3, a4
; RV32I-NEXT: add a0, a2, a0
; RV32I-NEXT: add a1, a3, a1
; RV32I-NEXT: srli a2, a0, 4
; RV32I-NEXT: srli a3, a1, 4
; RV32I-NEXT: add a0, a2, a0
; RV32I-NEXT: add a1, a3, a1
; RV32I-NEXT: and a0, a0, a5
; RV32I-NEXT: and a1, a1, a5
; RV32I-NEXT: slli a2, a0, 8
; RV32I-NEXT: slli a3, a1, 8
; RV32I-NEXT: add a0, a0, a2
; RV32I-NEXT: add a1, a1, a3
; RV32I-NEXT: slli a2, a0, 16
; RV32I-NEXT: slli a3, a1, 16
; RV32I-NEXT: add a0, a0, a2
; RV32I-NEXT: add a1, a1, a3
; RV32I-NEXT: srli a0, a0, 24
; RV32I-NEXT: srli a1, a1, 24
; RV32I-NEXT: add a0, a1, a0
; RV32I-NEXT: li a1, 0
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: ctpop_i64:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: cpop a0, a0
; RV32ZBB-NEXT: cpop a1, a1
; RV32ZBB-NEXT: add a0, a1, a0
; RV32ZBB-NEXT: li a1, 0
; RV32ZBB-NEXT: ret
%1 = call i64 @llvm.ctpop.i64(i64 %a)
ret i64 %1
}
define i1 @ctpop_i64_ugt_two(i64 %a) nounwind {
; RV32I-LABEL: ctpop_i64_ugt_two:
; RV32I: # %bb.0:
; RV32I-NEXT: addi a2, a0, -1
; RV32I-NEXT: addi a3, a1, -1
; RV32I-NEXT: sltiu a4, a2, -1
; RV32I-NEXT: add a3, a3, a4
; RV32I-NEXT: and a0, a0, a2
; RV32I-NEXT: and a1, a1, a3
; RV32I-NEXT: or a0, a0, a1
; RV32I-NEXT: seqz a0, a0
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: ctpop_i64_ugt_two:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: j .LBB6_2
; RV32ZBB-NEXT: # %bb.1:
; RV32ZBB-NEXT: li a0, 0
; RV32ZBB-NEXT: ret
; RV32ZBB-NEXT: .LBB6_2:
; RV32ZBB-NEXT: cpop a0, a0
; RV32ZBB-NEXT: cpop a1, a1
; RV32ZBB-NEXT: add a0, a1, a0
; RV32ZBB-NEXT: sltiu a0, a0, 2
; RV32ZBB-NEXT: ret
%1 = call i64 @llvm.ctpop.i64(i64 %a)
%2 = icmp ult i64 %1, 2
ret i1 %2
}
define i1 @ctpop_i64_ugt_one(i64 %a) nounwind {
; RV32I-LABEL: ctpop_i64_ugt_one:
; RV32I: # %bb.0:
; RV32I-NEXT: addi a2, a0, -1
; RV32I-NEXT: addi a3, a1, -1
; RV32I-NEXT: sltiu a4, a2, -1
; RV32I-NEXT: add a3, a3, a4
; RV32I-NEXT: and a0, a0, a2
; RV32I-NEXT: and a1, a1, a3
; RV32I-NEXT: or a0, a0, a1
; RV32I-NEXT: snez a0, a0
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: ctpop_i64_ugt_one:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: j .LBB7_2
; RV32ZBB-NEXT: # %bb.1:
; RV32ZBB-NEXT: li a0, 0
; RV32ZBB-NEXT: ret
; RV32ZBB-NEXT: .LBB7_2:
; RV32ZBB-NEXT: cpop a0, a0
; RV32ZBB-NEXT: cpop a1, a1
; RV32ZBB-NEXT: add a0, a1, a0
; RV32ZBB-NEXT: sltiu a0, a0, 2
; RV32ZBB-NEXT: xori a0, a0, 1
; RV32ZBB-NEXT: ret
%1 = call i64 @llvm.ctpop.i64(i64 %a)
%2 = icmp ugt i64 %1, 1
ret i1 %2
}
define i1 @ctpop_i64_eq_one(i64 %a) nounwind {
; RV32I-LABEL: ctpop_i64_eq_one:
; RV32I: # %bb.0:
; RV32I-NEXT: addi a2, a0, -1
; RV32I-NEXT: sltiu a3, a2, -1
; RV32I-NEXT: addi a4, a1, -1
; RV32I-NEXT: add a3, a4, a3
; RV32I-NEXT: xor a1, a1, a3
; RV32I-NEXT: beq a1, a3, .LBB8_2
; RV32I-NEXT: # %bb.1:
; RV32I-NEXT: sltu a0, a3, a1
; RV32I-NEXT: ret
; RV32I-NEXT: .LBB8_2:
; RV32I-NEXT: xor a0, a0, a2
; RV32I-NEXT: sltu a0, a2, a0
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: ctpop_i64_eq_one:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: cpop a0, a0
; RV32ZBB-NEXT: cpop a1, a1
; RV32ZBB-NEXT: add a0, a1, a0
; RV32ZBB-NEXT: xori a0, a0, 1
; RV32ZBB-NEXT: seqz a0, a0
; RV32ZBB-NEXT: ret
%1 = call i64 @llvm.ctpop.i64(i64 %a)
%2 = icmp eq i64 %1, 1
ret i1 %2
}
define i1 @ctpop_i64_ne_one(i64 %a) nounwind {
; RV32I-LABEL: ctpop_i64_ne_one:
; RV32I: # %bb.0:
; RV32I-NEXT: addi a2, a0, -1
; RV32I-NEXT: sltiu a3, a2, -1
; RV32I-NEXT: addi a4, a1, -1
; RV32I-NEXT: add a3, a4, a3
; RV32I-NEXT: xor a1, a1, a3
; RV32I-NEXT: beq a1, a3, .LBB9_2
; RV32I-NEXT: # %bb.1:
; RV32I-NEXT: sltu a0, a3, a1
; RV32I-NEXT: xori a0, a0, 1
; RV32I-NEXT: ret
; RV32I-NEXT: .LBB9_2:
; RV32I-NEXT: xor a0, a0, a2
; RV32I-NEXT: sltu a0, a2, a0
; RV32I-NEXT: xori a0, a0, 1
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: ctpop_i64_ne_one:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: cpop a0, a0
; RV32ZBB-NEXT: cpop a1, a1
; RV32ZBB-NEXT: add a0, a1, a0
; RV32ZBB-NEXT: xori a0, a0, 1
; RV32ZBB-NEXT: snez a0, a0
; RV32ZBB-NEXT: ret
%1 = call i64 @llvm.ctpop.i64(i64 %a)
%2 = icmp ne i64 %1, 1
ret i1 %2
}
define i32 @sextb_i32(i32 %a) nounwind {
; RV32I-LABEL: sextb_i32:
; RV32I: # %bb.0:
; RV32I-NEXT: slli a0, a0, 24
; RV32I-NEXT: srai a0, a0, 24
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: sextb_i32:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: sext.b a0, a0
; RV32ZBB-NEXT: ret
%shl = shl i32 %a, 24
%shr = ashr exact i32 %shl, 24
ret i32 %shr
}
; FIXME: Combine back to back srai.
define i64 @sextb_i64(i64 %a) nounwind {
; RV32I-LABEL: sextb_i64:
; RV32I: # %bb.0:
; RV32I-NEXT: slli a1, a0, 24
; RV32I-NEXT: srai a0, a1, 24
; RV32I-NEXT: srai a1, a1, 31
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: sextb_i64:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: sext.b a0, a0
; RV32ZBB-NEXT: srai a1, a0, 31
; RV32ZBB-NEXT: ret
%shl = shl i64 %a, 56
%shr = ashr exact i64 %shl, 56
ret i64 %shr
}
define i32 @sexth_i32(i32 %a) nounwind {
; RV32I-LABEL: sexth_i32:
; RV32I: # %bb.0:
; RV32I-NEXT: slli a0, a0, 16
; RV32I-NEXT: srai a0, a0, 16
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: sexth_i32:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: sext.h a0, a0
; RV32ZBB-NEXT: ret
%shl = shl i32 %a, 16
%shr = ashr exact i32 %shl, 16
ret i32 %shr
}
; FIXME: Combine back to back srai.
define i64 @sexth_i64(i64 %a) nounwind {
; RV32I-LABEL: sexth_i64:
; RV32I: # %bb.0:
; RV32I-NEXT: slli a1, a0, 16
; RV32I-NEXT: srai a0, a1, 16
; RV32I-NEXT: srai a1, a1, 31
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: sexth_i64:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: sext.h a0, a0
; RV32ZBB-NEXT: srai a1, a0, 31
; RV32ZBB-NEXT: ret
%shl = shl i64 %a, 48
%shr = ashr exact i64 %shl, 48
ret i64 %shr
}
define i32 @min_i32(i32 %a, i32 %b) nounwind {
; RV32I-LABEL: min_i32:
; RV32I: # %bb.0:
; RV32I-NEXT: blt a0, a1, .LBB14_2
; RV32I-NEXT: # %bb.1:
; RV32I-NEXT: mv a0, a1
; RV32I-NEXT: .LBB14_2:
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: min_i32:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: min a0, a0, a1
; RV32ZBB-NEXT: ret
%cmp = icmp slt i32 %a, %b
%cond = select i1 %cmp, i32 %a, i32 %b
ret i32 %cond
}
; As we are not matching directly i64 code patterns on RV32 some i64 patterns
; don't have yet any matching bit manipulation instructions on RV32.
; This test is presented here in case future expansions of the Bitmanip
; extensions introduce instructions suitable for this pattern.
define i64 @min_i64(i64 %a, i64 %b) nounwind {
; CHECK-LABEL: min_i64:
; CHECK: # %bb.0:
; CHECK-NEXT: beq a1, a3, .LBB15_2
; CHECK-NEXT: # %bb.1:
; CHECK-NEXT: slt a4, a1, a3
; CHECK-NEXT: beqz a4, .LBB15_3
; CHECK-NEXT: j .LBB15_4
; CHECK-NEXT: .LBB15_2:
; CHECK-NEXT: sltu a4, a0, a2
; CHECK-NEXT: bnez a4, .LBB15_4
; CHECK-NEXT: .LBB15_3:
; CHECK-NEXT: mv a0, a2
; CHECK-NEXT: mv a1, a3
; CHECK-NEXT: .LBB15_4:
; CHECK-NEXT: ret
%cmp = icmp slt i64 %a, %b
%cond = select i1 %cmp, i64 %a, i64 %b
ret i64 %cond
}
define i32 @max_i32(i32 %a, i32 %b) nounwind {
; RV32I-LABEL: max_i32:
; RV32I: # %bb.0:
; RV32I-NEXT: blt a1, a0, .LBB16_2
; RV32I-NEXT: # %bb.1:
; RV32I-NEXT: mv a0, a1
; RV32I-NEXT: .LBB16_2:
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: max_i32:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: max a0, a0, a1
; RV32ZBB-NEXT: ret
%cmp = icmp sgt i32 %a, %b
%cond = select i1 %cmp, i32 %a, i32 %b
ret i32 %cond
}
; As we are not matching directly i64 code patterns on RV32 some i64 patterns
; don't have yet any matching bit manipulation instructions on RV32.
; This test is presented here in case future expansions of the Bitmanip
; extensions introduce instructions suitable for this pattern.
define i64 @max_i64(i64 %a, i64 %b) nounwind {
; CHECK-LABEL: max_i64:
; CHECK: # %bb.0:
; CHECK-NEXT: beq a1, a3, .LBB17_2
; CHECK-NEXT: # %bb.1:
; CHECK-NEXT: slt a4, a3, a1
; CHECK-NEXT: beqz a4, .LBB17_3
; CHECK-NEXT: j .LBB17_4
; CHECK-NEXT: .LBB17_2:
; CHECK-NEXT: sltu a4, a2, a0
; CHECK-NEXT: bnez a4, .LBB17_4
; CHECK-NEXT: .LBB17_3:
; CHECK-NEXT: mv a0, a2
; CHECK-NEXT: mv a1, a3
; CHECK-NEXT: .LBB17_4:
; CHECK-NEXT: ret
%cmp = icmp sgt i64 %a, %b
%cond = select i1 %cmp, i64 %a, i64 %b
ret i64 %cond
}
define i32 @minu_i32(i32 %a, i32 %b) nounwind {
; RV32I-LABEL: minu_i32:
; RV32I: # %bb.0:
; RV32I-NEXT: bltu a0, a1, .LBB18_2
; RV32I-NEXT: # %bb.1:
; RV32I-NEXT: mv a0, a1
; RV32I-NEXT: .LBB18_2:
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: minu_i32:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: minu a0, a0, a1
; RV32ZBB-NEXT: ret
%cmp = icmp ult i32 %a, %b
%cond = select i1 %cmp, i32 %a, i32 %b
ret i32 %cond
}
; As we are not matching directly i64 code patterns on RV32 some i64 patterns
; don't have yet any matching bit manipulation instructions on RV32.
; This test is presented here in case future expansions of the Bitmanip
; extensions introduce instructions suitable for this pattern.
define i64 @minu_i64(i64 %a, i64 %b) nounwind {
; CHECK-LABEL: minu_i64:
; CHECK: # %bb.0:
; CHECK-NEXT: beq a1, a3, .LBB19_2
; CHECK-NEXT: # %bb.1:
; CHECK-NEXT: sltu a4, a1, a3
; CHECK-NEXT: beqz a4, .LBB19_3
; CHECK-NEXT: j .LBB19_4
; CHECK-NEXT: .LBB19_2:
; CHECK-NEXT: sltu a4, a0, a2
; CHECK-NEXT: bnez a4, .LBB19_4
; CHECK-NEXT: .LBB19_3:
; CHECK-NEXT: mv a0, a2
; CHECK-NEXT: mv a1, a3
; CHECK-NEXT: .LBB19_4:
; CHECK-NEXT: ret
%cmp = icmp ult i64 %a, %b
%cond = select i1 %cmp, i64 %a, i64 %b
ret i64 %cond
}
define i32 @maxu_i32(i32 %a, i32 %b) nounwind {
; RV32I-LABEL: maxu_i32:
; RV32I: # %bb.0:
; RV32I-NEXT: bltu a1, a0, .LBB20_2
; RV32I-NEXT: # %bb.1:
; RV32I-NEXT: mv a0, a1
; RV32I-NEXT: .LBB20_2:
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: maxu_i32:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: maxu a0, a0, a1
; RV32ZBB-NEXT: ret
%cmp = icmp ugt i32 %a, %b
%cond = select i1 %cmp, i32 %a, i32 %b
ret i32 %cond
}
; As we are not matching directly i64 code patterns on RV32 some i64 patterns
; don't have yet any matching bit manipulation instructions on RV32.
; This test is presented here in case future expansions of the Bitmanip
; extensions introduce instructions suitable for this pattern.
define i64 @maxu_i64(i64 %a, i64 %b) nounwind {
; CHECK-LABEL: maxu_i64:
; CHECK: # %bb.0:
; CHECK-NEXT: beq a1, a3, .LBB21_2
; CHECK-NEXT: # %bb.1:
; CHECK-NEXT: sltu a4, a3, a1
; CHECK-NEXT: beqz a4, .LBB21_3
; CHECK-NEXT: j .LBB21_4
; CHECK-NEXT: .LBB21_2:
; CHECK-NEXT: sltu a4, a2, a0
; CHECK-NEXT: bnez a4, .LBB21_4
; CHECK-NEXT: .LBB21_3:
; CHECK-NEXT: mv a0, a2
; CHECK-NEXT: mv a1, a3
; CHECK-NEXT: .LBB21_4:
; CHECK-NEXT: ret
%cmp = icmp ugt i64 %a, %b
%cond = select i1 %cmp, i64 %a, i64 %b
ret i64 %cond
}
define i32 @abs_i32(i32 %x) {
; RV32I-LABEL: abs_i32:
; RV32I: # %bb.0:
; RV32I-NEXT: srai a1, a0, 31
; RV32I-NEXT: add a0, a0, a1
; RV32I-NEXT: xor a0, a0, a1
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: abs_i32:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: neg a1, a0
; RV32ZBB-NEXT: max a0, a0, a1
; RV32ZBB-NEXT: ret
%abs = tail call i32 @llvm.abs.i32(i32 %x, i1 true)
ret i32 %abs
}
define i64 @abs_i64(i64 %x) {
; CHECK-LABEL: abs_i64:
; CHECK: # %bb.0:
; CHECK-NEXT: srai a2, a1, 31
; CHECK-NEXT: add a0, a0, a2
; CHECK-NEXT: add a1, a1, a2
; CHECK-NEXT: sltu a3, a0, a2
; CHECK-NEXT: add a1, a1, a3
; CHECK-NEXT: xor a0, a0, a2
; CHECK-NEXT: xor a1, a1, a2
; CHECK-NEXT: ret
%abs = tail call i64 @llvm.abs.i64(i64 %x, i1 true)
ret i64 %abs
}
define i32 @zexth_i32(i32 %a) nounwind {
; RV32I-LABEL: zexth_i32:
; RV32I: # %bb.0:
; RV32I-NEXT: slli a0, a0, 16
; RV32I-NEXT: srli a0, a0, 16
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: zexth_i32:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: zext.h a0, a0
; RV32ZBB-NEXT: ret
%and = and i32 %a, 65535
ret i32 %and
}
define i64 @zexth_i64(i64 %a) nounwind {
; RV32I-LABEL: zexth_i64:
; RV32I: # %bb.0:
; RV32I-NEXT: slli a0, a0, 16
; RV32I-NEXT: srli a0, a0, 16
; RV32I-NEXT: li a1, 0
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: zexth_i64:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: zext.h a0, a0
; RV32ZBB-NEXT: li a1, 0
; RV32ZBB-NEXT: ret
%and = and i64 %a, 65535
ret i64 %and
}
define i32 @bswap_i32(i32 %a) nounwind {
; RV32I-LABEL: bswap_i32:
; RV32I: # %bb.0:
; RV32I-NEXT: slli a1, a0, 24
; RV32I-NEXT: srli a2, a0, 24
; RV32I-NEXT: lui a3, 16
; RV32I-NEXT: or a1, a2, a1
; RV32I-NEXT: srli a2, a0, 8
; RV32I-NEXT: addi a3, a3, -256
; RV32I-NEXT: and a0, a0, a3
; RV32I-NEXT: and a2, a2, a3
; RV32I-NEXT: slli a0, a0, 8
; RV32I-NEXT: or a1, a1, a2
; RV32I-NEXT: or a0, a1, a0
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: bswap_i32:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: rev8 a0, a0
; RV32ZBB-NEXT: ret
%1 = tail call i32 @llvm.bswap.i32(i32 %a)
ret i32 %1
}
define i64 @bswap_i64(i64 %a) {
; RV32I-LABEL: bswap_i64:
; RV32I: # %bb.0:
; RV32I-NEXT: slli a2, a1, 24
; RV32I-NEXT: srli a3, a1, 24
; RV32I-NEXT: lui a4, 16
; RV32I-NEXT: srli a5, a1, 8
; RV32I-NEXT: slli a6, a0, 24
; RV32I-NEXT: or a2, a3, a2
; RV32I-NEXT: srli a3, a0, 24
; RV32I-NEXT: or a3, a3, a6
; RV32I-NEXT: srli a6, a0, 8
; RV32I-NEXT: addi a4, a4, -256
; RV32I-NEXT: and a1, a1, a4
; RV32I-NEXT: and a5, a5, a4
; RV32I-NEXT: and a0, a0, a4
; RV32I-NEXT: and a4, a6, a4
; RV32I-NEXT: or a2, a2, a5
; RV32I-NEXT: slli a1, a1, 8
; RV32I-NEXT: slli a5, a0, 8
; RV32I-NEXT: or a3, a3, a4
; RV32I-NEXT: or a0, a2, a1
; RV32I-NEXT: or a1, a3, a5
; RV32I-NEXT: ret
;
; RV32ZBB-LABEL: bswap_i64:
; RV32ZBB: # %bb.0:
; RV32ZBB-NEXT: rev8 a2, a1
; RV32ZBB-NEXT: rev8 a1, a0
; RV32ZBB-NEXT: mv a0, a2
; RV32ZBB-NEXT: ret
%1 = call i64 @llvm.bswap.i64(i64 %a)
ret i64 %1
}
Reference in new issue View Git Blame Copy Permalink