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
34 lines
719 B
LLVM
34 lines
719 B
LLVM
; RUN: llc < %s -march=nvptx -mcpu=sm_20 | FileCheck %s
|
|
; RUN: %if ptxas %{ llc < %s -march=nvptx -mcpu=sm_20 | %ptxas-verify %}
|
|
|
|
|
|
; CHECK: bfe0
|
|
define i32 @bfe0(i32 %a) {
|
|
; CHECK: bfe.u32 %r{{[0-9]+}}, %r{{[0-9]+}}, 4, 4
|
|
; CHECK-NOT: shr
|
|
; CHECK-NOT: and
|
|
%val0 = ashr i32 %a, 4
|
|
%val1 = and i32 %val0, 15
|
|
ret i32 %val1
|
|
}
|
|
|
|
; CHECK: bfe1
|
|
define i32 @bfe1(i32 %a) {
|
|
; CHECK: bfe.u32 %r{{[0-9]+}}, %r{{[0-9]+}}, 3, 3
|
|
; CHECK-NOT: shr
|
|
; CHECK-NOT: and
|
|
%val0 = ashr i32 %a, 3
|
|
%val1 = and i32 %val0, 7
|
|
ret i32 %val1
|
|
}
|
|
|
|
; CHECK: bfe2
|
|
define i32 @bfe2(i32 %a) {
|
|
; CHECK: bfe.u32 %r{{[0-9]+}}, %r{{[0-9]+}}, 5, 3
|
|
; CHECK-NOT: shr
|
|
; CHECK-NOT: and
|
|
%val0 = ashr i32 %a, 5
|
|
%val1 = and i32 %val0, 7
|
|
ret i32 %val1
|
|
}
|