Does subgroup extension API "intel_sub_group_block_readN/writeN" have better performance than "vloadN/vstoreN"? I did some testing, but don't see much difference between them. Can you elaborate the read/write performance expectation between them?
↧
Channel:
Intel® Software - OpenCL*
X
Are you the publisher?
Claim or
contact us
about this channel.
X
0
Channel Details:
- Title: Intel® Software - OpenCL*
- Channel Number: 13188063
- Language: English
- Registered On: May 14, 2013, 2:17 pm
- Number of Articles: 1182
- Latest Snapshot: June 16, 2020, 9:12 pm
- RSS URL: http://software.intel.com/en-us/taxonomy/term/36974/feed
- Publisher: https://software.intel.com/en-us/taxonomy/term/36974
- Description: This is a place to get support, ask questions and share information with other customers on...
- Catalog: //opencl38.rssing.com/catalog.php?indx=13188063
© 2025 //www.rssing.com