0f1b5f115a
ptxas is a proprietary compiler from Nvidia that can compile PTX to machine code (SASS). It has a lot of diagnostics to catch errors in PTX, which can be used to verify PTX output from llc. Set -DPXTAS_EXECUTABLE=/path/to/ptxas CMake option to enable it. If this option is not set, then ptxas is substituted to true which effectively disables all ptxas RUN lines. LLVM_PTXAS_EXECUTABLE environment variable takes precedence over the CMake option, and allows to override ptxas executable that is used for LIT without complete re-configuration. Differential Revision: https://reviews.llvm.org/D121727
46 lines
1.2 KiB
LLVM
46 lines
1.2 KiB
LLVM
; RUN: llc < %s -march=nvptx -mcpu=sm_20 -verify-machineinstrs | FileCheck %s
|
|
; RUN: %if ptxas %{ llc < %s -march=nvptx -mcpu=sm_20 -verify-machineinstrs | %ptxas-verify %}
|
|
|
|
target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v16:16:16-v32:32:32-v64:64:64-v128:128:128-n16:32:64"
|
|
|
|
declare i16 @llvm.cttz.i16(i16, i1) readnone
|
|
declare i32 @llvm.cttz.i32(i32, i1) readnone
|
|
declare i64 @llvm.cttz.i64(i64, i1) readnone
|
|
|
|
define i32 @myctpop(i32 %a) {
|
|
; CHECK: popc.b32
|
|
%val = call i32 @llvm.cttz.i32(i32 %a, i1 false) readnone
|
|
ret i32 %val
|
|
}
|
|
|
|
define i16 @myctpop16(i16 %a) {
|
|
; CHECK: popc.b32
|
|
%val = call i16 @llvm.cttz.i16(i16 %a, i1 false) readnone
|
|
ret i16 %val
|
|
}
|
|
|
|
define i64 @myctpop64(i64 %a) {
|
|
; CHECK: popc.b64
|
|
%val = call i64 @llvm.cttz.i64(i64 %a, i1 false) readnone
|
|
ret i64 %val
|
|
}
|
|
|
|
|
|
define i32 @myctpop_2(i32 %a) {
|
|
; CHECK: popc.b32
|
|
%val = call i32 @llvm.cttz.i32(i32 %a, i1 true) readnone
|
|
ret i32 %val
|
|
}
|
|
|
|
define i16 @myctpop16_2(i16 %a) {
|
|
; CHECK: popc.b32
|
|
%val = call i16 @llvm.cttz.i16(i16 %a, i1 true) readnone
|
|
ret i16 %val
|
|
}
|
|
|
|
define i64 @myctpop64_2(i64 %a) {
|
|
; CHECK: popc.b64
|
|
%val = call i64 @llvm.cttz.i64(i64 %a, i1 true) readnone
|
|
ret i64 %val
|
|
}
|