LLVM OpenMP Runtime Lands Improvements For Intel Meteor Lake
([LLVM] 120 Minutes Ago
Meteor Lake OpenMP Runtime)
- Reference: 0001483530
- News link: https://www.phoronix.com/news/LLVM-OpenMP-Better-Intel-MTL
- Source link:
A three month old merge request finally landed in mainline LLVM Git this past week to deliver improvements initially for Intel Meteor Lake processors.
Merged for LLVM 20 are topology and affinity changes to LLVM's OpenMP run-time that were optimized for Intel Core Ultra "Meteor Lake" processors but should benefit the upcoming Lunar Lake processors too.
This merge has various clean-ups, better handling for unknown hardware IDs, caches are now added to the topology code, different processors can now have different number of topology levels, and other alterations to improve OpenMP multi-threaded handling for Meteor Lake.
See [1]this merge request for those interested in the details around this improved OpenMP handling on Intel Core Ultra processors.
[1] https://github.com/llvm/llvm-project/pull/91012
Merged for LLVM 20 are topology and affinity changes to LLVM's OpenMP run-time that were optimized for Intel Core Ultra "Meteor Lake" processors but should benefit the upcoming Lunar Lake processors too.
This merge has various clean-ups, better handling for unknown hardware IDs, caches are now added to the topology code, different processors can now have different number of topology levels, and other alterations to improve OpenMP multi-threaded handling for Meteor Lake.
See [1]this merge request for those interested in the details around this improved OpenMP handling on Intel Core Ultra processors.
[1] https://github.com/llvm/llvm-project/pull/91012
phoronix