mirror of
https://github.com/c64scene-ar/llvm-6502.git
synced 2024-12-15 04:30:12 +00:00
7c9c6ed761
Essentially the same as the GEP change in r230786.
A similar migration script can be used to update test cases, though a few more
test case improvements/changes were required this time around: (r229269-r229278)
import fileinput
import sys
import re
pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)")
for line in sys.stdin:
sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line))
Reviewers: rafael, dexonsmith, grosser
Differential Revision: http://reviews.llvm.org/D7649
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230794 91177308-0d34-0410-b5e6-96231b3b80d8
108 lines
4.4 KiB
LLVM
108 lines
4.4 KiB
LLVM
; RUN: llc -march=amdgcn -mcpu=tahiti -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
|
|
; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
|
|
|
|
declare double @llvm.fabs.f64(double) #0
|
|
|
|
; SI-LABEL: {{^}}fsub_f64:
|
|
; SI: v_add_f64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], -v\[[0-9]+:[0-9]+\]}}
|
|
define void @fsub_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
|
|
double addrspace(1)* %in2) {
|
|
%r0 = load double, double addrspace(1)* %in1
|
|
%r1 = load double, double addrspace(1)* %in2
|
|
%r2 = fsub double %r0, %r1
|
|
store double %r2, double addrspace(1)* %out
|
|
ret void
|
|
}
|
|
|
|
; SI-LABEL: {{^}}fsub_fabs_f64:
|
|
; SI: v_add_f64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], -\|v\[[0-9]+:[0-9]+\]\|}}
|
|
define void @fsub_fabs_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
|
|
double addrspace(1)* %in2) {
|
|
%r0 = load double, double addrspace(1)* %in1
|
|
%r1 = load double, double addrspace(1)* %in2
|
|
%r1.fabs = call double @llvm.fabs.f64(double %r1) #0
|
|
%r2 = fsub double %r0, %r1.fabs
|
|
store double %r2, double addrspace(1)* %out
|
|
ret void
|
|
}
|
|
|
|
; SI-LABEL: {{^}}fsub_fabs_inv_f64:
|
|
; SI: v_add_f64 {{v\[[0-9]+:[0-9]+\], |v\[[0-9]+:[0-9]+\]|, -v\[[0-9]+:[0-9]+\]}}
|
|
define void @fsub_fabs_inv_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
|
|
double addrspace(1)* %in2) {
|
|
%r0 = load double, double addrspace(1)* %in1
|
|
%r1 = load double, double addrspace(1)* %in2
|
|
%r0.fabs = call double @llvm.fabs.f64(double %r0) #0
|
|
%r2 = fsub double %r0.fabs, %r1
|
|
store double %r2, double addrspace(1)* %out
|
|
ret void
|
|
}
|
|
|
|
; SI-LABEL: {{^}}s_fsub_f64:
|
|
; SI: v_add_f64 {{v\[[0-9]+:[0-9]+\], s\[[0-9]+:[0-9]+\], -v\[[0-9]+:[0-9]+\]}}
|
|
define void @s_fsub_f64(double addrspace(1)* %out, double %a, double %b) {
|
|
%sub = fsub double %a, %b
|
|
store double %sub, double addrspace(1)* %out
|
|
ret void
|
|
}
|
|
|
|
; SI-LABEL: {{^}}s_fsub_imm_f64:
|
|
; SI: v_add_f64 {{v\[[0-9]+:[0-9]+\], 4.0, -s\[[0-9]+:[0-9]+\]}}
|
|
define void @s_fsub_imm_f64(double addrspace(1)* %out, double %a, double %b) {
|
|
%sub = fsub double 4.0, %a
|
|
store double %sub, double addrspace(1)* %out
|
|
ret void
|
|
}
|
|
|
|
; SI-LABEL: {{^}}s_fsub_imm_inv_f64:
|
|
; SI: v_add_f64 {{v\[[0-9]+:[0-9]+\], -4.0, s\[[0-9]+:[0-9]+\]}}
|
|
define void @s_fsub_imm_inv_f64(double addrspace(1)* %out, double %a, double %b) {
|
|
%sub = fsub double %a, 4.0
|
|
store double %sub, double addrspace(1)* %out
|
|
ret void
|
|
}
|
|
|
|
; SI-LABEL: {{^}}s_fsub_self_f64:
|
|
; SI: v_add_f64 {{v\[[0-9]+:[0-9]+\], s\[[0-9]+:[0-9]+\], -s\[[0-9]+:[0-9]+\]}}
|
|
define void @s_fsub_self_f64(double addrspace(1)* %out, double %a) {
|
|
%sub = fsub double %a, %a
|
|
store double %sub, double addrspace(1)* %out
|
|
ret void
|
|
}
|
|
|
|
; SI-LABEL: {{^}}fsub_v2f64:
|
|
; SI: v_add_f64 {{v\[[0-9]+:[0-9]+\], s\[[0-9]+:[0-9]+\], -v\[[0-9]+:[0-9]+\]}}
|
|
; SI: v_add_f64 {{v\[[0-9]+:[0-9]+\], s\[[0-9]+:[0-9]+\], -v\[[0-9]+:[0-9]+\]}}
|
|
define void @fsub_v2f64(<2 x double> addrspace(1)* %out, <2 x double> %a, <2 x double> %b) {
|
|
%sub = fsub <2 x double> %a, %b
|
|
store <2 x double> %sub, <2 x double> addrspace(1)* %out
|
|
ret void
|
|
}
|
|
|
|
; SI-LABEL: {{^}}fsub_v4f64:
|
|
; SI: v_add_f64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], -v\[[0-9]+:[0-9]+\]}}
|
|
; SI: v_add_f64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], -v\[[0-9]+:[0-9]+\]}}
|
|
; SI: v_add_f64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], -v\[[0-9]+:[0-9]+\]}}
|
|
; SI: v_add_f64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], -v\[[0-9]+:[0-9]+\]}}
|
|
define void @fsub_v4f64(<4 x double> addrspace(1)* %out, <4 x double> addrspace(1)* %in) {
|
|
%b_ptr = getelementptr <4 x double>, <4 x double> addrspace(1)* %in, i32 1
|
|
%a = load <4 x double>, <4 x double> addrspace(1)* %in
|
|
%b = load <4 x double>, <4 x double> addrspace(1)* %b_ptr
|
|
%result = fsub <4 x double> %a, %b
|
|
store <4 x double> %result, <4 x double> addrspace(1)* %out
|
|
ret void
|
|
}
|
|
|
|
; SI-LABEL: {{^}}s_fsub_v4f64:
|
|
; SI: v_add_f64 {{v\[[0-9]+:[0-9]+\], s\[[0-9]+:[0-9]+\], -v\[[0-9]+:[0-9]+\]}}
|
|
; SI: v_add_f64 {{v\[[0-9]+:[0-9]+\], s\[[0-9]+:[0-9]+\], -v\[[0-9]+:[0-9]+\]}}
|
|
; SI: v_add_f64 {{v\[[0-9]+:[0-9]+\], s\[[0-9]+:[0-9]+\], -v\[[0-9]+:[0-9]+\]}}
|
|
; SI: v_add_f64 {{v\[[0-9]+:[0-9]+\], s\[[0-9]+:[0-9]+\], -v\[[0-9]+:[0-9]+\]}}
|
|
define void @s_fsub_v4f64(<4 x double> addrspace(1)* %out, <4 x double> %a, <4 x double> %b) {
|
|
%result = fsub <4 x double> %a, %b
|
|
store <4 x double> %result, <4 x double> addrspace(1)* %out, align 16
|
|
ret void
|
|
}
|
|
|
|
attributes #0 = { nounwind readnone }
|