Made possible adding multiple AArch64 extensions (CPU features) to GCC "-march" #4449
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
maximmasiutin
force-pushed
the
ampere-altra
branch
2 times, most recently
from
3bf2e9b
to
5934bd0
Compare
maximmasiutin
force-pushed
the
ampere-altra
branch
from
5934bd0
to
164c004
Compare
|
I know it's late, but may I ask what exactly this PR is trying to fix or improve? Only early revisions of Cortex-A53s have this specific bugs, and compilers have already managed to mitigate this. (GCC for example: https://github.com/gcc-mirror/gcc/blob/master/gcc/config/aarch64/aarch64-errata.h) Even if someone is using outdated compilers, I believe armv8 (without dotprod) is already enough. Is there any particular reason for the use of ARMv8.4? |
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Outdated suggestions cannot be applied.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.
Added support for multiple AArch64 extensions (CPU features) instead of a single one. Now the features like "+dotprod" are accumulated in the
armextvariable before finally adding it to the "-march" parameter. When specifying "ARCH=armv8" make parameter, the Makefile didn't contain the provisions to specify -march=armv8.2-a parameter to gcc. Now it does. Also added ARMv8.4-A architecture for Amazon Graviton 3 which is faster than the initial version (ARMv8-A). Tested Amazon Graviton 3 and Ampere Altra locally in 'stockfish bench' and as Fishtest workers.