// // Copyright (c) 2017 The Khronos Group Inc. // // Licensed under the Apache License, Version 2.0 (the "License"); // you may not use this file except in compliance with the License. // You may obtain a copy of the License at // // http://www.apache.org/licenses/LICENSE-2.0 // // Unless required by applicable law or agreed to in writing, software // distributed under the License is distributed on an "AS IS" BASIS, // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. // See the License for the specific language governing permissions and // limitations under the License. // #include "../testBase.h" #include #if defined( __APPLE__ ) #include #include #include #endif const char *read1DBufferKernelSourcePattern = "__kernel void sample_kernel( read_only image1d_buffer_t inputA, read_only image1d_t inputB, sampler_t sampler, __global int *results )\n" "{\n" " int tidX = get_global_id(0);\n" " int offset = tidX;\n" " %s clr = read_image%s( inputA, tidX );\n" " int4 test = (clr != read_image%s( inputB, sampler, tidX ));\n" " if ( test.x || test.y || test.z || test.w )\n" " results[offset] = -1;\n" " else\n" " results[offset] = 0;\n" "}"; int test_read_image_1D_buffer( cl_context context, cl_command_queue queue, cl_kernel kernel, image_descriptor *imageInfo, image_sampler_data *imageSampler, ExplicitType outputType, MTdata d ) { int error; size_t threads[2]; cl_sampler actualSampler; BufferOwningPtr imageValues; generate_random_image_data( imageInfo, imageValues, d ); if ( gDebugTrace ) log_info( " - Creating 1D image from buffer %d ...\n", (int)imageInfo->width ); // Construct testing sources cl_mem image[2]; cl_image_desc image_desc; cl_mem imageBuffer = clCreateBuffer( context, CL_MEM_READ_ONLY | CL_MEM_COPY_HOST_PTR, imageInfo->rowPitch, imageValues, &error); if ( error != CL_SUCCESS ) { log_error( "ERROR: Unable to create buffer of size %d bytes (%s)\n", (int)imageInfo->rowPitch, IGetErrorString( error ) ); return error; } memset(&image_desc, 0x0, sizeof(cl_image_desc)); image_desc.image_type = CL_MEM_OBJECT_IMAGE1D_BUFFER; image_desc.image_width = imageInfo->width; image_desc.mem_object = imageBuffer; image[0] = clCreateImage( context, CL_MEM_READ_ONLY, imageInfo->format, &image_desc, NULL, &error ); if ( error != CL_SUCCESS ) { log_error( "ERROR: Unable to create IMAGE1D_BUFFER of size %d pitch %d (%s)\n", (int)imageInfo->width, (int)imageInfo->rowPitch, IGetErrorString( error ) ); return error; } cl_mem ret = NULL; error = clGetMemObjectInfo(image[0], CL_MEM_ASSOCIATED_MEMOBJECT, sizeof(ret), &ret, NULL); if ( error != CL_SUCCESS ) { log_error( "ERROR: Unable to query CL_MEM_ASSOCIATED_MEMOBJECT\n", IGetErrorString( error ) ); return error; } if (ret != imageBuffer) { log_error("ERROR: clGetImageInfo for CL_IMAGE_BUFFER returned wrong value\n"); return -1; } memset(&image_desc, 0x0, sizeof(cl_image_desc)); image_desc.image_type = CL_MEM_OBJECT_IMAGE1D; image_desc.image_width = imageInfo->width; image[1] = clCreateImage( context, CL_MEM_READ_ONLY|CL_MEM_COPY_HOST_PTR, imageInfo->format, &image_desc, imageValues, &error ); if ( error != CL_SUCCESS ) { log_error( "ERROR: Unable to create IMAGE1D of size %d pitch %d (%s)\n", (int)imageInfo->width, (int)imageInfo->rowPitch, IGetErrorString( error ) ); return error; } if ( gDebugTrace ) log_info( " - Creating kernel arguments...\n" ); // Create sampler to use actualSampler = clCreateSampler( context, CL_FALSE, CL_ADDRESS_NONE, CL_FILTER_NEAREST, &error ); test_error( error, "Unable to create image sampler" ); // Create results buffer cl_mem results = clCreateBuffer( context, 0, imageInfo->width * sizeof(cl_int), NULL, &error); test_error( error, "Unable to create results buffer" ); size_t resultValuesSize = imageInfo->width * sizeof(cl_int); BufferOwningPtr resultValues(malloc( resultValuesSize )); memset( resultValues, 0xff, resultValuesSize ); clEnqueueWriteBuffer( queue, results, CL_TRUE, 0, resultValuesSize, resultValues, 0, NULL, NULL ); // Set arguments int idx = 0; error = clSetKernelArg( kernel, idx++, sizeof( cl_mem ), &image[0] ); test_error( error, "Unable to set kernel arguments" ); error = clSetKernelArg( kernel, idx++, sizeof( cl_mem ), &image[1] ); test_error( error, "Unable to set kernel arguments" ); error = clSetKernelArg( kernel, idx++, sizeof( cl_sampler ), &actualSampler ); test_error( error, "Unable to set kernel arguments" ); error = clSetKernelArg( kernel, idx++, sizeof( cl_mem ), &results ); test_error( error, "Unable to set kernel arguments" ); // Run the kernel threads[0] = (size_t)imageInfo->width; error = clEnqueueNDRangeKernel( queue, kernel, 1, NULL, threads, NULL, 0, NULL, NULL ); test_error( error, "Unable to run kernel" ); if ( gDebugTrace ) log_info( " reading results, %ld kbytes\n", (unsigned long)( imageInfo->width * sizeof(cl_int) / 1024 ) ); error = clEnqueueReadBuffer( queue, results, CL_TRUE, 0, resultValuesSize, resultValues, 0, NULL, NULL ); test_error( error, "Unable to read results from kernel" ); if ( gDebugTrace ) log_info( " results read\n" ); // Check for non-zero comps bool allZeroes = true; for ( size_t ic = 0; ic < imageInfo->width; ++ic ) { if ( resultValues[ic] ) { allZeroes = false; break; } } if ( !allZeroes ) { log_error( " Sampler-less reads differ from reads with sampler.\n" ); return -1; } clReleaseSampler(actualSampler); clReleaseMemObject(results); clReleaseMemObject(image[0]); clReleaseMemObject(image[1]); clReleaseMemObject(imageBuffer); return 0; } int test_read_image_set_1D_buffer(cl_device_id device, cl_context context, cl_command_queue queue, const cl_image_format *format, image_sampler_data *imageSampler, ExplicitType outputType) { char programSrc[10240]; const char *ptr; const char *readFormat; const char *dataType; clProgramWrapper program; clKernelWrapper kernel; RandomSeed seed( gRandomSeed ); int error; // Get our operating params size_t maxWidth, maxWidth1D; cl_ulong maxAllocSize, memSize; image_descriptor imageInfo = { 0 }; size_t pixelSize; if (format->image_channel_order == CL_RGB || format->image_channel_order == CL_RGBx) { switch (format->image_channel_data_type) { case CL_UNORM_INT8: case CL_UNORM_INT16: case CL_SNORM_INT8: case CL_SNORM_INT16: case CL_HALF_FLOAT: case CL_FLOAT: case CL_SIGNED_INT8: case CL_SIGNED_INT16: case CL_SIGNED_INT32: case CL_UNSIGNED_INT8: case CL_UNSIGNED_INT16: case CL_UNSIGNED_INT32: case CL_UNORM_INT_101010: log_info( "Skipping image format: %s %s\n", GetChannelOrderName( format->image_channel_order ), GetChannelTypeName( format->image_channel_data_type )); return 0; default: break; } } imageInfo.format = format; imageInfo.height = imageInfo.depth = imageInfo.arraySize = imageInfo.slicePitch = 0; imageInfo.type = CL_MEM_OBJECT_IMAGE1D; pixelSize = get_pixel_size( imageInfo.format ); error = clGetDeviceInfo( device, CL_DEVICE_IMAGE_MAX_BUFFER_SIZE, sizeof( maxWidth ), &maxWidth, NULL ); error |= clGetDeviceInfo( device, CL_DEVICE_MAX_MEM_ALLOC_SIZE, sizeof( maxAllocSize ), &maxAllocSize, NULL ); error |= clGetDeviceInfo( device, CL_DEVICE_GLOBAL_MEM_SIZE, sizeof( memSize ), &memSize, NULL ); error |= clGetDeviceInfo( device, CL_DEVICE_IMAGE2D_MAX_WIDTH, sizeof( maxWidth ), &maxWidth1D, NULL ); test_error( error, "Unable to get max image 1D buffer size from device" ); if (memSize > (cl_ulong)SIZE_MAX) { memSize = (cl_ulong)SIZE_MAX; } // note: image_buffer test uses image1D for results validation. // So the test can't use the biggest possible size for image_buffer if it's bigger than the max image1D size maxWidth = (maxWidth > maxWidth1D) ? maxWidth1D : maxWidth; // Determine types if ( outputType == kInt ) { readFormat = "i"; dataType = "int4"; } else if ( outputType == kUInt ) { readFormat = "ui"; dataType = "uint4"; } else // kFloat { readFormat = "f"; dataType = "float4"; } sprintf( programSrc, read1DBufferKernelSourcePattern, dataType, readFormat, readFormat ); ptr = programSrc; error = create_single_kernel_helper(context, &program, &kernel, 1, &ptr, "sample_kernel"); test_error( error, "Unable to create testing kernel" ); if ( gTestSmallImages ) { for ( imageInfo.width = 1; imageInfo.width < 13; imageInfo.width++ ) { imageInfo.rowPitch = imageInfo.width * pixelSize; { if ( gDebugTrace ) log_info( " at size %d\n", (int)imageInfo.width ); int retCode = test_read_image_1D_buffer( context, queue, kernel, &imageInfo, imageSampler, outputType, seed ); if ( retCode ) return retCode; } } } else if ( gTestMaxImages ) { // Try a specific set of maximum sizes size_t numbeOfSizes; size_t sizes[100][3]; get_max_sizes(&numbeOfSizes, 100, sizes, maxWidth, 1, 1, 1, maxAllocSize, memSize, CL_MEM_OBJECT_IMAGE1D, imageInfo.format); for ( size_t idx = 0; idx < numbeOfSizes; idx++ ) { imageInfo.width = sizes[ idx ][ 0 ]; imageInfo.rowPitch = imageInfo.width * pixelSize; log_info("Testing %d\n", (int)sizes[ idx ][ 0 ]); if ( gDebugTrace ) log_info( " at max size %d\n", (int)sizes[ idx ][ 0 ] ); int retCode = test_read_image_1D_buffer( context, queue, kernel, &imageInfo, imageSampler, outputType, seed ); if ( retCode ) return retCode; } } else { for ( int i = 0; i < NUM_IMAGE_ITERATIONS; i++ ) { cl_ulong size; // Loop until we get a size that a) will fit in the max alloc size and b) that an allocation of that // image, the result array, plus offset arrays, will fit in the global ram space do { imageInfo.width = (size_t)random_log_in_range( 16, (int)maxWidth / 32, seed ); imageInfo.rowPitch = imageInfo.width * pixelSize; size = (size_t)imageInfo.rowPitch * 4; } while ( size > maxAllocSize || ( size * 3 ) > memSize ); if ( gDebugTrace ) log_info( " at size %d (row pitch %d) out of %d\n", (int)imageInfo.width, (int)imageInfo.rowPitch, (int)maxWidth ); int retCode = test_read_image_1D_buffer( context, queue, kernel, &imageInfo, imageSampler, outputType, seed ); if ( retCode ) return retCode; } } return 0; }