replica
/
hanchenye-llvm-project
2
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

r600: Cleanup barrier implementation. 

We don't have memory fences for r600 so just call group barrier directly
Make sure that barrier is called even with 0 flags

Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
Reviewed-by: Aaron Watry <awatry@gmail.com>
llvm-svn: 312492
This commit is contained in:
Jan Vesely 2017-09-04 15:52:05 +00:00
parent 1796d590c1
commit e337b30c7d
3 changed files with 5 additions and 37 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
libclc/amdgpu/lib/SOURCES
View File

@ -1,7 +1,6 @@
atomic/atomic.cl
math/nextafter.cl
math/sqrt.cl
synchronization/barrier.cl
image/get_image_width.cl
image/get_image_height.cl
image/get_image_depth.cl

10
libclc/amdgpu/lib/synchronization/barrier.cl
View File

@ -1,10 +0,0 @@
#include <clc/clc.h>
_CLC_DEF int __clc_clk_local_mem_fence() {
return CLK_LOCAL_MEM_FENCE;
}
_CLC_DEF int __clc_clk_global_mem_fence() {
return CLK_GLOBAL_MEM_FENCE;
}

31
libclc/r600/lib/synchronization/barrier_impl.ll
View File

@ -1,32 +1,11 @@
declare i32 @__clc_clk_local_mem_fence() #1
declare i32 @__clc_clk_global_mem_fence() #1
declare void @llvm.r600.group.barrier() #0
define void @barrier(i32 %flags) #2 {
barrier_local_test:
%CLK_LOCAL_MEM_FENCE = call i32 @__clc_clk_local_mem_fence()
%0 = and i32 %flags, %CLK_LOCAL_MEM_FENCE
%1 = icmp ne i32 %0, 0
br i1 %1, label %barrier_local, label %barrier_global_test
barrier_local:
call void @llvm.r600.group.barrier()
br label %barrier_global_test
barrier_global_test:
%CLK_GLOBAL_MEM_FENCE = call i32 @__clc_clk_global_mem_fence()
%2 = and i32 %flags, %CLK_GLOBAL_MEM_FENCE
%3 = icmp ne i32 %2, 0
br i1 %3, label %barrier_global, label %done
barrier_global:
call void @llvm.r600.group.barrier()
br label %done
done:
define void @barrier(i32 %flags) #1 {
entry:
; We should call mem_fence here, but that is not implemented for r600 yet
tail call void @llvm.r600.group.barrier()
ret void
}
attributes #0 = { nounwind convergent }
attributes #1 = { nounwind alwaysinline }
attributes #2 = { nounwind convergent alwaysinline }
attributes #1 = { nounwind convergent alwaysinline }