mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-24 03:33:20 +01:00
ab043ff680
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
llvm-svn: 230794
64 lines
1.5 KiB
LLVM
64 lines
1.5 KiB
LLVM
; RUN: llc < %s -march=nvptx -mcpu=sm_20 | FileCheck %s --check-prefix=PTX32
|
|
; RUN: llc < %s -march=nvptx64 -mcpu=sm_20 | FileCheck %s --check-prefix=PTX64
|
|
|
|
|
|
;; i8
|
|
define i8 @ld_global_i8(i8 addrspace(0)* %ptr) {
|
|
; PTX32: ld.u8 %r{{[0-9]+}}, [%r{{[0-9]+}}]
|
|
; PTX32: ret
|
|
; PTX64: ld.u8 %r{{[0-9]+}}, [%rd{{[0-9]+}}]
|
|
; PTX64: ret
|
|
%a = load i8, i8 addrspace(0)* %ptr
|
|
ret i8 %a
|
|
}
|
|
|
|
;; i16
|
|
define i16 @ld_global_i16(i16 addrspace(0)* %ptr) {
|
|
; PTX32: ld.u16 %r{{[0-9]+}}, [%r{{[0-9]+}}]
|
|
; PTX32: ret
|
|
; PTX64: ld.u16 %r{{[0-9]+}}, [%rd{{[0-9]+}}]
|
|
; PTX64: ret
|
|
%a = load i16, i16 addrspace(0)* %ptr
|
|
ret i16 %a
|
|
}
|
|
|
|
;; i32
|
|
define i32 @ld_global_i32(i32 addrspace(0)* %ptr) {
|
|
; PTX32: ld.u32 %r{{[0-9]+}}, [%r{{[0-9]+}}]
|
|
; PTX32: ret
|
|
; PTX64: ld.u32 %r{{[0-9]+}}, [%rd{{[0-9]+}}]
|
|
; PTX64: ret
|
|
%a = load i32, i32 addrspace(0)* %ptr
|
|
ret i32 %a
|
|
}
|
|
|
|
;; i64
|
|
define i64 @ld_global_i64(i64 addrspace(0)* %ptr) {
|
|
; PTX32: ld.u64 %rd{{[0-9]+}}, [%r{{[0-9]+}}]
|
|
; PTX32: ret
|
|
; PTX64: ld.u64 %rd{{[0-9]+}}, [%rd{{[0-9]+}}]
|
|
; PTX64: ret
|
|
%a = load i64, i64 addrspace(0)* %ptr
|
|
ret i64 %a
|
|
}
|
|
|
|
;; f32
|
|
define float @ld_global_f32(float addrspace(0)* %ptr) {
|
|
; PTX32: ld.f32 %f{{[0-9]+}}, [%r{{[0-9]+}}]
|
|
; PTX32: ret
|
|
; PTX64: ld.f32 %f{{[0-9]+}}, [%rd{{[0-9]+}}]
|
|
; PTX64: ret
|
|
%a = load float, float addrspace(0)* %ptr
|
|
ret float %a
|
|
}
|
|
|
|
;; f64
|
|
define double @ld_global_f64(double addrspace(0)* %ptr) {
|
|
; PTX32: ld.f64 %fd{{[0-9]+}}, [%r{{[0-9]+}}]
|
|
; PTX32: ret
|
|
; PTX64: ld.f64 %fd{{[0-9]+}}, [%rd{{[0-9]+}}]
|
|
; PTX64: ret
|
|
%a = load double, double addrspace(0)* %ptr
|
|
ret double %a
|
|
}
|