-
-
Notifications
You must be signed in to change notification settings - Fork 11.3k
ENH, SIMD: Extend universal intrinsics to support IBMZ #20913
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Merged
Merged
Conversation
This file contains hidden or 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
b66ba82
to
75d77af
Compare
6224f8c
to
50407a6
Compare
873ea68
to
982ba4b
Compare
It covers SIMD operations for all datatypes starting from z/Arch11 a.k.a IBM Z13, except for single-precision which requires minimum z/Arch12 a.k.a IBMZ 14 to be dispatched. This patch rename the branch /simd/vsx to /simd/vec, the new the path is hold the definitions of universal intrinsics for both Power and Z architectures. This patch also adds new preprocessor identifiers: * NPY_SIMD_BIGENDIAN: 1 if the enabled SIMD extension is running on big-endian mode otherwise 0. * NPY_SIMD_F32: 1 if the enabled SIMD extension supports single-precision otherwise 0.
Github doesn't allow big body of the pr desc, the rest as follows: VX
|
cc @mattip |
Let's put this in now, at the beginning of the 1.24 release cycle, in hopes that some IBMZ users can chime in. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Labels
01 - Enhancement
56 - Needs Release Note.
Needs an entry in doc/release/upcoming_changes
component: SIMD
Issues in SIMD (fast instruction sets) code or machinery
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Outdated suggestions cannot be applied.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.
Extend universal intrinsics to support IBMZ
It covers SIMD operations for all datatypes starting
from z/Arch11 a.k.a IBM Z13, except for single-precision
which requires minimum z/Arch12 a.k.a IBMZ 14 to be dispatched.
This patch rename the branch /simd/vsx to /simd/vec, the new
the path holds the definitions of universal intrinsics for
both Power and Z architectures.
This patch also adds new preprocessor identifiers:
NPY_SIMD_BIGENDIAN: 1 if the enabled SIMD extension
is running on big-endian mode otherwise 0.
NPY_SIMD_F32: 1 if the enabled SIMD extension
supports single-precision otherwise 0.
TODO:
Benchmark
The following benchmark is inferential and does not accurately reflect the true change, since it was accomplished using an unstable VM.
CPU
OS
Linux numpy 5.4.0-104-generic #118-Ubuntu SMP Wed Mar 2 19:02:13 UTC 2022 s390x s390x s390x GNU/Linux Python 3.8.10 gcc (Ubuntu 11.1.0-1ubuntu1~20.04) 11.1.0
Benchmark
VXE2
unset NPY_DISABLE_CPU_FEATURES python runtests.py --bench-compare parent/main
VXE