Skip to content

Commit 5b8e541

Browse files
[ORO-0] Update. (#68)
1 parent bc3644c commit 5b8e541

48 files changed

Lines changed: 37 additions & 79 deletions

File tree

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

‎Orochi/Orochi.cpp‎

Lines changed: 4 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -338,6 +338,9 @@ oroError OROAPI oroGetDeviceProperties(oroDeviceProp* props, oroDevice dev)
338338
e = cuDeviceGetAttribute( &props->computeMode, CU_DEVICE_ATTRIBUTE_COMPUTE_MODE, deviceId );
339339
e = cuDeviceGetAttribute( &props->concurrentKernels, CU_DEVICE_ATTRIBUTE_CONCURRENT_KERNELS, deviceId );
340340
e = cuDeviceGetAttribute( &props->ECCEnabled, CU_DEVICE_ATTRIBUTE_ECC_ENABLED, deviceId );
341+
e = cuDeviceGetAttribute( &props->major, CU_DEVICE_ATTRIBUTE_COMPUTE_CAPABILITY_MAJOR, deviceId );
342+
e = cuDeviceGetAttribute( &props->minor, CU_DEVICE_ATTRIBUTE_COMPUTE_CAPABILITY_MINOR, deviceId );
343+
341344
return oroSuccess;
342345
}
343346
return oroErrorUnknown;
@@ -637,7 +640,7 @@ oroError OROAPI oroMemcpyDtoDAsync( oroDeviceptr dstDevice, oroDeviceptr srcDevi
637640
}
638641
oroError OROAPI oroMemset(oroDeviceptr dstDevice, unsigned int ui, size_t N)
639642
{
640-
__ORO_FUNC1( MemsetD8( (CUdeviceptr)dstDevice, ui, N ), Memset((void*)dstDevice, ui, N));
643+
__ORO_FUNC1( MemsetD32( (CUdeviceptr)dstDevice, ui, N/sizeof(int) ), MemsetD32( dstDevice, ui, N / sizeof(int) ) );
641644
return oroErrorUnknown;
642645
}
643646

‎Test/WMMA/main.cpp‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -74,6 +74,7 @@ int main( int argc, char** argv )
7474
oroCtx ctx;
7575
e = oroCtxCreate( &ctx, 0, device );
7676
ERROR_CHECK( e );
77+
oroCtxSetCurrent( ctx );
7778

7879
//try kernel execution
7980
oroFunction function;
Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,2 @@
1+
call hipcc --cuda-device-only --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1032 --offload-arch=gfx1033 --offload-arch=gfx1034 --offload-arch=gfx1035 --offload-arch=gfx1036 --offload-arch=gfx1010 --offload-arch=gfx1011 --offload-arch=gfx1012 --offload-arch=gfx1013 --offload-arch=gfx900 --offload-arch=gfx906 -fgpu-rdc -c --gpu-bundle-output -c -emit-llvm ../moduleTestKernel.cpp
2+
call hipcc --cuda-device-only --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1032 --offload-arch=gfx1033 --offload-arch=gfx1034 --offload-arch=gfx1035 --offload-arch=gfx1036 --offload-arch=gfx1010 --offload-arch=gfx1011 --offload-arch=gfx1012 --offload-arch=gfx1013 --offload-arch=gfx900 --offload-arch=gfx906 -fgpu-rdc -c --gpu-bundle-output -c -emit-llvm ../moduleTestFunc.cpp
Lines changed: 0 additions & 28 deletions
Original file line numberDiff line numberDiff line change
@@ -1,30 +1,2 @@
11
hipcc --cuda-device-only --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1032 --offload-arch=gfx1033 --offload-arch=gfx1034 --offload-arch=gfx1035 --offload-arch=gfx1036 --offload-arch=gfx1010 --offload-arch=gfx1011 --offload-arch=gfx1012 --offload-arch=gfx1013 --offload-arch=gfx900 --offload-arch=gfx906 -fgpu-rdc -c --gpu-bundle-output -c -emit-llvm ../moduleTestKernel.cpp
2-
32
hipcc --cuda-device-only --offload-arch=gfx1030 --offload-arch=gfx1031 --offload-arch=gfx1032 --offload-arch=gfx1033 --offload-arch=gfx1034 --offload-arch=gfx1035 --offload-arch=gfx1036 --offload-arch=gfx1010 --offload-arch=gfx1011 --offload-arch=gfx1012 --offload-arch=gfx1013 --offload-arch=gfx900 --offload-arch=gfx906 -fgpu-rdc -c --gpu-bundle-output -c -emit-llvm ../moduleTestFunc.cpp
4-
hipcc --offload-arch=gfx1030 ../moduleTestFunc.cpp -c -fgpu-rdc --cuda-device-only
5-
hipcc --offload-arch=gfx1031 ../moduleTestFunc.cpp -c -fgpu-rdc --cuda-device-only
6-
hipcc --offload-arch=gfx1032 ../moduleTestFunc.cpp -c -fgpu-rdc --cuda-device-only
7-
hipcc --offload-arch=gfx1033 ../moduleTestFunc.cpp -c -fgpu-rdc --cuda-device-only
8-
hipcc --offload-arch=gfx1034 ../moduleTestFunc.cpp -c -fgpu-rdc --cuda-device-only
9-
hipcc --offload-arch=gfx1035 ../moduleTestFunc.cpp -c -fgpu-rdc --cuda-device-only
10-
hipcc --offload-arch=gfx1036 ../moduleTestFunc.cpp -c -fgpu-rdc --cuda-device-only
11-
hipcc --offload-arch=gfx1010 ../moduleTestFunc.cpp -c -fgpu-rdc --cuda-device-only
12-
hipcc --offload-arch=gfx1011 ../moduleTestFunc.cpp -c -fgpu-rdc --cuda-device-only
13-
hipcc --offload-arch=gfx1012 ../moduleTestFunc.cpp -c -fgpu-rdc --cuda-device-only
14-
hipcc --offload-arch=gfx1013 ../moduleTestFunc.cpp -c -fgpu-rdc --cuda-device-only
15-
hipcc --offload-arch=gfx900 ../moduleTestFunc.cpp -c -fgpu-rdc --cuda-device-only
16-
hipcc --offload-arch=gfx906 ../moduleTestFunc.cpp -c -fgpu-rdc --cuda-device-only
17-
18-
hipcc --offload-arch=gfx1030 ../moduleTestKernel.cpp -c -fgpu-rdc --cuda-device-only
19-
hipcc --offload-arch=gfx1031 ../moduleTestKernel.cpp -c -fgpu-rdc --cuda-device-only
20-
hipcc --offload-arch=gfx1032 ../moduleTestKernel.cpp -c -fgpu-rdc --cuda-device-only
21-
hipcc --offload-arch=gfx1033 ../moduleTestKernel.cpp -c -fgpu-rdc --cuda-device-only
22-
hipcc --offload-arch=gfx1034 ../moduleTestKernel.cpp -c -fgpu-rdc --cuda-device-only
23-
hipcc --offload-arch=gfx1035 ../moduleTestKernel.cpp -c -fgpu-rdc --cuda-device-only
24-
hipcc --offload-arch=gfx1036 ../moduleTestKernel.cpp -c -fgpu-rdc --cuda-device-only
25-
hipcc --offload-arch=gfx1010 ../moduleTestKernel.cpp -c -fgpu-rdc --cuda-device-only
26-
hipcc --offload-arch=gfx1011 ../moduleTestKernel.cpp -c -fgpu-rdc --cuda-device-only
27-
hipcc --offload-arch=gfx1012 ../moduleTestKernel.cpp -c -fgpu-rdc --cuda-device-only
28-
hipcc --offload-arch=gfx1013 ../moduleTestKernel.cpp -c -fgpu-rdc --cuda-device-only
29-
hipcc --offload-arch=gfx900 ../moduleTestKernel.cpp -c -fgpu-rdc --cuda-device-only
30-
hipcc --offload-arch=gfx906 ../moduleTestKernel.cpp -c -fgpu-rdc --cuda-device-only

‎UnitTest/bitcodes/generate_bitcodes_gfx1100.bat‎

Lines changed: 0 additions & 7 deletions
This file was deleted.

‎UnitTest/bitcodes/generate_bitcodes_gfx1100.sh‎

Lines changed: 0 additions & 6 deletions
This file was deleted.

‎UnitTest/bitcodes/generate_bitcodes_gfx1102.bat‎

Lines changed: 0 additions & 7 deletions
This file was deleted.

‎UnitTest/bitcodes/generate_bitcodes_gfx1102.sh‎

Lines changed: 0 additions & 6 deletions
This file was deleted.
Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,2 @@
1+
nvcc -x cu -fatbin --device-c -arch=all ../moduleTestFunc.cpp
2+
nvcc -x cu -fatbin --device-c -arch=all ../moduleTestKernel.cpp
Lines changed: 2 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,2 @@
1-
nvcc -arch=compute_80 -code="sm_80,sm_86,sm_87" -fatbin --device-c ../moduleTestFunc.cu
2-
nvcc -arch=compute_80 -code="sm_80,sm_86,sm_87" -fatbin --device-c ../moduleTestKernel.cu
3-
nvcc -cubin --device-c -arch=sm_80 ../moduleTestFunc.cu
4-
nvcc -cubin --device-c -arch=sm_80 ../moduleTestKernel.cu
1+
nvcc -fatbin --device-c -arch=all ../moduleTestFunc.cu
2+
nvcc -fatbin --device-c -arch=all ../moduleTestKernel.cu

0 commit comments

Comments
 (0)