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
199 lines
7.8 KiB
LLVM
199 lines
7.8 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
|
|
; RUN: llc -mtriple=aarch64 %s -o - | FileCheck %s --check-prefixes=CHECK,CHECK-SD
|
|
; RUN: llc -mtriple=aarch64 -global-isel=true -global-isel-abort=2 %s -o - 2>&1 | FileCheck %s --check-prefixes=CHECK,CHECK-GI
|
|
|
|
; CHECK-GI: warning: Instruction selection used fallback path for unused_div_fpexcept_strict
|
|
; CHECK-GI-NEXT: warning: Instruction selection used fallback path for unused_div_round_dynamic
|
|
; CHECK-GI-NEXT: warning: Instruction selection used fallback path for add_twice_fpexcept_strict
|
|
; CHECK-GI-NEXT: warning: Instruction selection used fallback path for add_twice_round_dynamic
|
|
; CHECK-GI-NEXT: warning: Instruction selection used fallback path for set_rounding
|
|
; CHECK-GI-NEXT: warning: Instruction selection used fallback path for set_rounding_fpexcept_strict
|
|
; CHECK-GI-NEXT: warning: Instruction selection used fallback path for set_rounding_round_dynamic
|
|
|
|
; Div whose result is unused should be removed unless we have strict exceptions
|
|
|
|
define void @unused_div(float %x, float %y) {
|
|
; CHECK-LABEL: unused_div:
|
|
; CHECK: // %bb.0: // %entry
|
|
; CHECK-NEXT: ret
|
|
entry:
|
|
%add = fdiv float %x, %y
|
|
ret void
|
|
}
|
|
|
|
define void @unused_div_fpexcept_strict(float %x, float %y) #0 {
|
|
; CHECK-LABEL: unused_div_fpexcept_strict:
|
|
; CHECK: // %bb.0: // %entry
|
|
; CHECK-NEXT: fdiv s0, s0, s1
|
|
; CHECK-NEXT: ret
|
|
entry:
|
|
%add = call float @llvm.experimental.constrained.fdiv.f32(float %x, float %y, metadata !"round.tonearest", metadata !"fpexcept.strict") #0
|
|
ret void
|
|
}
|
|
|
|
define void @unused_div_round_dynamic(float %x, float %y) #0 {
|
|
; CHECK-LABEL: unused_div_round_dynamic:
|
|
; CHECK: // %bb.0: // %entry
|
|
; CHECK-NEXT: ret
|
|
entry:
|
|
%add = call float @llvm.experimental.constrained.fdiv.f32(float %x, float %y, metadata !"round.dynamic", metadata !"fpexcept.ignore") #0
|
|
ret void
|
|
}
|
|
|
|
|
|
; Machine CSE should eliminate the second add unless we have strict exceptions
|
|
define float @add_twice(float %x, float %y, i32 %n) {
|
|
; CHECK-LABEL: add_twice:
|
|
; CHECK: // %bb.0: // %entry
|
|
; CHECK-NEXT: fadd s0, s0, s1
|
|
; CHECK-NEXT: cmp w0, #0
|
|
; CHECK-NEXT: fmul s1, s0, s0
|
|
; CHECK-NEXT: fcsel s0, s0, s1, eq
|
|
; CHECK-NEXT: ret
|
|
entry:
|
|
%add = fadd float %x, %y
|
|
%tobool.not = icmp eq i32 %n, 0
|
|
br i1 %tobool.not, label %if.end, label %if.then
|
|
|
|
if.then:
|
|
%add1 = fadd float %x, %y
|
|
%mul = fmul float %add, %add1
|
|
br label %if.end
|
|
|
|
if.end:
|
|
%a.0 = phi float [ %mul, %if.then ], [ %add, %entry ]
|
|
ret float %a.0
|
|
}
|
|
|
|
define float @add_twice_fpexcept_strict(float %x, float %y, i32 %n) #0 {
|
|
; CHECK-LABEL: add_twice_fpexcept_strict:
|
|
; CHECK: // %bb.0: // %entry
|
|
; CHECK-NEXT: fmov s2, s0
|
|
; CHECK-NEXT: fadd s0, s0, s1
|
|
; CHECK-NEXT: cbz w0, .LBB4_2
|
|
; CHECK-NEXT: // %bb.1: // %if.then
|
|
; CHECK-NEXT: fadd s1, s2, s1
|
|
; CHECK-NEXT: fmul s0, s0, s1
|
|
; CHECK-NEXT: .LBB4_2: // %if.end
|
|
; CHECK-NEXT: ret
|
|
entry:
|
|
%add = call float @llvm.experimental.constrained.fadd.f32(float %x, float %y, metadata !"round.tonearest", metadata !"fpexcept.strict") #0
|
|
%tobool.not = icmp eq i32 %n, 0
|
|
br i1 %tobool.not, label %if.end, label %if.then
|
|
|
|
if.then:
|
|
%add1 = call float @llvm.experimental.constrained.fadd.f32(float %x, float %y, metadata !"round.tonearest", metadata !"fpexcept.strict") #0
|
|
%mul = call float @llvm.experimental.constrained.fmul.f32(float %add, float %add1, metadata !"round.tonearest", metadata !"fpexcept.strict") #0
|
|
br label %if.end
|
|
|
|
if.end:
|
|
%a.0 = phi float [ %mul, %if.then ], [ %add, %entry ]
|
|
ret float %a.0
|
|
}
|
|
|
|
define float @add_twice_round_dynamic(float %x, float %y, i32 %n) #0 {
|
|
; CHECK-LABEL: add_twice_round_dynamic:
|
|
; CHECK: // %bb.0: // %entry
|
|
; CHECK-NEXT: fadd s0, s0, s1
|
|
; CHECK-NEXT: cbz w0, .LBB5_2
|
|
; CHECK-NEXT: // %bb.1: // %if.then
|
|
; CHECK-NEXT: fmul s0, s0, s0
|
|
; CHECK-NEXT: .LBB5_2: // %if.end
|
|
; CHECK-NEXT: ret
|
|
entry:
|
|
%add = call float @llvm.experimental.constrained.fadd.f32(float %x, float %y, metadata !"round.dynamic", metadata !"fpexcept.ignore") #0
|
|
%tobool.not = icmp eq i32 %n, 0
|
|
br i1 %tobool.not, label %if.end, label %if.then
|
|
|
|
if.then:
|
|
%add1 = call float @llvm.experimental.constrained.fadd.f32(float %x, float %y, metadata !"round.dynamic", metadata !"fpexcept.ignore") #0
|
|
%mul = call float @llvm.experimental.constrained.fmul.f32(float %add, float %add1, metadata !"round.dynamic", metadata !"fpexcept.ignore") #0
|
|
br label %if.end
|
|
|
|
if.end:
|
|
%a.0 = phi float [ %mul, %if.then ], [ %add, %entry ]
|
|
ret float %a.0
|
|
}
|
|
|
|
|
|
; Two adds separated by llvm.set.rounding should be preserved when rounding is
|
|
; dynamic (as they may give different results) or when we have strict exceptions
|
|
; (the llvm.set.rounding is irrelevant, but both could trap).
|
|
|
|
define float @set_rounding(float %x, float %y) {
|
|
; CHECK-LABEL: set_rounding:
|
|
; CHECK: // %bb.0: // %entry
|
|
; CHECK-NEXT: mrs x8, FPCR
|
|
; CHECK-NEXT: fadd s0, s0, s1
|
|
; CHECK-NEXT: orr x8, x8, #0xc00000
|
|
; CHECK-NEXT: msr FPCR, x8
|
|
; CHECK-NEXT: mrs x8, FPCR
|
|
; CHECK-NEXT: and x8, x8, #0xffffffffff3fffff
|
|
; CHECK-NEXT: msr FPCR, x8
|
|
; CHECK-NEXT: fsub s0, s0, s0
|
|
; CHECK-NEXT: ret
|
|
entry:
|
|
%add1 = fadd float %x, %y
|
|
call void @llvm.set.rounding(i32 0)
|
|
%add2 = fadd float %x, %y
|
|
call void @llvm.set.rounding(i32 1)
|
|
%sub = fsub float %add1, %add2
|
|
ret float %sub
|
|
}
|
|
|
|
define float @set_rounding_fpexcept_strict(float %x, float %y) #0 {
|
|
; CHECK-LABEL: set_rounding_fpexcept_strict:
|
|
; CHECK: // %bb.0: // %entry
|
|
; CHECK-NEXT: fadd s2, s0, s1
|
|
; CHECK-NEXT: mrs x8, FPCR
|
|
; CHECK-NEXT: orr x8, x8, #0xc00000
|
|
; CHECK-NEXT: msr FPCR, x8
|
|
; CHECK-NEXT: fadd s0, s0, s1
|
|
; CHECK-NEXT: mrs x8, FPCR
|
|
; CHECK-NEXT: and x8, x8, #0xffffffffff3fffff
|
|
; CHECK-NEXT: msr FPCR, x8
|
|
; CHECK-NEXT: fsub s0, s2, s0
|
|
; CHECK-NEXT: ret
|
|
entry:
|
|
%add1 = call float @llvm.experimental.constrained.fadd.f32(float %x, float %y, metadata !"round.tonearest", metadata !"fpexcept.strict") #0
|
|
call void @llvm.set.rounding(i32 0) #0
|
|
%add2 = call float @llvm.experimental.constrained.fadd.f32(float %x, float %y, metadata !"round.tonearest", metadata !"fpexcept.strict") #0
|
|
call void @llvm.set.rounding(i32 1) #0
|
|
%sub = call float @llvm.experimental.constrained.fsub.f32(float %add1, float %add2, metadata !"round.tonearest", metadata !"fpexcept.strict") #0
|
|
ret float %sub
|
|
}
|
|
|
|
define float @set_rounding_round_dynamic(float %x, float %y) #0 {
|
|
; CHECK-LABEL: set_rounding_round_dynamic:
|
|
; CHECK: // %bb.0: // %entry
|
|
; CHECK-NEXT: mrs x8, FPCR
|
|
; CHECK-NEXT: fadd s2, s0, s1
|
|
; CHECK-NEXT: orr x8, x8, #0xc00000
|
|
; CHECK-NEXT: msr FPCR, x8
|
|
; CHECK-NEXT: fadd s0, s0, s1
|
|
; CHECK-NEXT: mrs x8, FPCR
|
|
; CHECK-NEXT: and x8, x8, #0xffffffffff3fffff
|
|
; CHECK-NEXT: msr FPCR, x8
|
|
; CHECK-NEXT: fsub s0, s2, s0
|
|
; CHECK-NEXT: ret
|
|
entry:
|
|
%add1 = call float @llvm.experimental.constrained.fadd.f32(float %x, float %y, metadata !"round.dynamic", metadata !"fpexcept.ignore") #0
|
|
call void @llvm.set.rounding(i32 0) #0
|
|
%add2 = call float @llvm.experimental.constrained.fadd.f32(float %x, float %y, metadata !"round.dynamic", metadata !"fpexcept.ignore") #0
|
|
call void @llvm.set.rounding(i32 1) #0
|
|
%sub = call float @llvm.experimental.constrained.fsub.f32(float %add1, float %add2, metadata !"round.dynamic", metadata !"fpexcept.ignore") #0
|
|
ret float %sub
|
|
}
|
|
|
|
declare float @llvm.experimental.constrained.fadd.f32(float, float, metadata, metadata)
|
|
declare float @llvm.experimental.constrained.fsub.f32(float, float, metadata, metadata)
|
|
declare float @llvm.experimental.constrained.fmul.f32(float, float, metadata, metadata)
|
|
declare float @llvm.experimental.constrained.fdiv.f32(float, float, metadata, metadata)
|
|
declare i32 @llvm.get.rounding()
|
|
declare void @llvm.set.rounding(i32)
|
|
|
|
attributes #0 = { strictfp }
|
|
;; NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line:
|
|
; CHECK-GI: {{.*}}
|
|
; CHECK-SD: {{.*}}
|