Alderlake expansion #10
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
To allow more generic user-driven configuration of CPU patterns the new `raw_pattern()` accepts a vector of bool which CPUs should be enabled or disabled. This can be considered a wrapper to transform a list of bools to the set-based internal representation to be more in line with the existing pattern generators.
The configuration for the test system "kallisto" uses a alderlake processor, i.e. there are different types of cores. This newly added test aims to explore the behavior of these cores. It is not finished and in its current state mainly handles the pattern creation, i.e. the mapping of function to CPU number.
Add documentation and a run script for kallisto configuration. Additionally, remove useless comments and switch to production cpu patterns/frequencies.
The following *only* applies to OpenBLAS. The BLAS invocation during the matmul kernel uses BLAS-owned threads. These are created separately, and thus are *not pinned* correctly. In our kallisto (alder lake) test system this led to *100%* CPU usage, while only 1/3rd of threads where active (E-cores only). The fix is documented in the OpenBLAS FAQ under "How can I use OpenBLAS in multi-threaded applications?", recommending setting an environment variable to disable this separate thread spawning.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This is some code to support the heterogeneous intel architecture
alderlake.It has been used in production, but not upstreamed yet.
All commits are labeled, please refer to their descriptions.