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
1.2 KiB
LLVM
34 lines
1.2 KiB
LLVM
; RUN: llc < %s -march=nvptx64 -mcpu=sm_30 -mattr=+ptx60 | FileCheck %s
|
|
; RUN: %if ptxas %{ llc < %s -march=nvptx64 -mcpu=sm_30 -mattr=+ptx60 | %ptxas-verify %if !ptxas-11.0 %{-arch=sm_30%} %}
|
|
|
|
declare void @llvm.nvvm.bar.warp.sync(i32)
|
|
declare void @llvm.nvvm.barrier.sync(i32)
|
|
declare void @llvm.nvvm.barrier.sync.cnt(i32, i32)
|
|
|
|
; CHECK-LABEL: .func{{.*}}barrier_sync
|
|
define void @barrier_sync(i32 %id, i32 %cnt) {
|
|
; CHECK: ld.param.u32 [[ID:%r[0-9]+]], [barrier_sync_param_0];
|
|
; CHECK: ld.param.u32 [[CNT:%r[0-9]+]], [barrier_sync_param_1];
|
|
|
|
; CHECK: barrier.sync [[ID]], [[CNT]];
|
|
call void @llvm.nvvm.barrier.sync.cnt(i32 %id, i32 %cnt)
|
|
; CHECK: barrier.sync [[ID]], 32;
|
|
call void @llvm.nvvm.barrier.sync.cnt(i32 %id, i32 32)
|
|
; CHECK: barrier.sync 3, [[CNT]];
|
|
call void @llvm.nvvm.barrier.sync.cnt(i32 3, i32 %cnt)
|
|
; CHECK: barrier.sync 4, 64;
|
|
call void @llvm.nvvm.barrier.sync.cnt(i32 4, i32 64)
|
|
|
|
; CHECK: barrier.sync [[ID]];
|
|
call void @llvm.nvvm.barrier.sync(i32 %id)
|
|
; CHECK: barrier.sync 1;
|
|
call void @llvm.nvvm.barrier.sync(i32 1)
|
|
|
|
; CHECK: bar.warp.sync [[ID]];
|
|
call void @llvm.nvvm.bar.warp.sync(i32 %id)
|
|
; CHECK: bar.warp.sync 6;
|
|
call void @llvm.nvvm.bar.warp.sync(i32 6)
|
|
ret void;
|
|
}
|
|
|