griefith/test
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

Cycles: oneAPI: Use max_compute_units in get_num_multiprocessors

Browse Source 
Instead of returning 0 in case the Intel extension for getting the count
of Execution Units isn't available, we now use
sycl::info::device::max_compute_units.

We keep using the Intel extension in priority since it logically goes
with sycl::ext::intel::info::device::gpu_hw_threads_per_eu used in
get_max_num_threads_per_multiprocessor(), for which there is no
sycl::info::device::max_threads_per_compute_unit replacement yet.
This commit is contained in:
Xavier Hallade
2025-03-26 23:15:49 +01:00
parent 06b03b0fa6
commit 7a257359f8
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
intern/cycles/device/oneapi/device_impl.cpp
View File

@@ -1591,7 +1591,7 @@ int OneapiDevice::get_num_multiprocessors()
if (device.has(sycl::aspect::ext_intel_gpu_eu_count)) {
return device.get_info<sycl::ext::intel::info::device::gpu_eu_count>();
}
return 0;
return device.get_info<sycl::info::device::max_compute_units>();
}
int OneapiDevice::get_max_num_threads_per_multiprocessor()