SYCLomatic
SYCLomatic copied to clipboard
oneapi-src
Could you please bring the SYCL support to the compiler people for cudaDeviceGetDefaultMemPool, cudaMallocAsync, cudaFreeAsync ? Thanks Reference https://developer.nvidia.com/blog/using-cuda-stream-ordered-memory-allocator-part-1/ https://github.com/NVIDIA/cuda-samples/tree/master/Samples/2_Concepts_and_Techniques/streamOrderedAllocation
Is there any way to print the regsPerBlock of the device on SYCLomatic? I don't want to da any operation with them but only to print this data.
PR for Store header functions for Block API (related later to https://github.com/oneapi-src/SYCLomatic/pull/1305 ) Linked with Load: #1640 cc @yihanwg @danhoeflinger @mmichel11
Signed-off-by: Tang, Jiajun [email protected]
Is "mask" missing when migrating __syncwarp(mask) ? ``` // void __syncwarp(unsigned mask=0xffffffff) void a(unsigned mask, const sycl::nd_item &item_ct1) { sycl::group_barrier(item_ct1.get_sub_group()); } ```
warning: DPCT1023:7: The SYCL sub-group does not support mask options for sycl::reduce_over_group. temp = __reduce_max_sync(same_threads, diag_maxHH[(m
Please see the migration result. dpct version 17.0.0. Codebase:(426a0560b12914f001e8980152d9447a28ed7a4b) Thanks. ``` input file: test.cuh #include template inline __device__ T shfl(const T val, int srcLane, int width = 32) { #if...
…mory can be cached.
Add Migration of CUB Block Radix Sort API. cc @yihanwg