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
8253015: Aarch64: Move linux code out from generic CPU feature detection #154
Conversation
|
@AntonKozlov The following label will be automatically applied to this pull request: When this pull request is ready to be reviewed, an RFR email will be sent to the corresponding mailing list. If you would like to change these labels, use the |
Webrevs
|
I've verified that this code works for the Windows-AArch64 port with minimal changes to the new |
A quick follow-up on getting a review for this patch. We're starting to take dependency on this patch for the Windows-AArch64 and macOS-AArch64 (7bf0aab). /cc @theRealAph |
Mailing list message from Andrew Haley on hotspot-dev: On 24/09/2020 18:10, Ludovic Henry wrote:
It's a large and complex patch. It looks OK, so I have read through and reviewed it, but I haven't tested -- |
Mailing list message from Anton Kozlov on hotspot-dev: Hello Andrew, On 25.09.2020 15:45, Andrew Haley wrote:
Sorry, do you refer to JDK-8248238 (Implementation of JEP: Windows AArch64) or The intention of this CPU feature detection patch is to do a preparation and Could you and community review this (cpu feature detection) patch as well? Thanks, |
@AntonKozlov This change now passes all automated pre-integration checks. After integration, the commit message for the final commit will be:
You can use pull request commands such as /summary, /contributor and /issue to adjust it as needed. At the time when this comment was updated there had been 191 new commits pushed to the
As there are no conflicts, your changes will automatically be rebased on top of these commits when integrating. If you prefer to avoid this automatic rebasing, please check the documentation for the /integrate command for further details. As you do not have Committer status in this project an existing Committer must agree to sponsor your change. Possible candidates are the reviewers of this PR (@theRealAph) but any other Committer may sponsor as well.
|
/integrate |
@AntonKozlov |
Mailing list message from Andrew Haley on hotspot-dev: On 28/09/2020 12:31, Anton Kozlov wrote:
That's very weird. One of the nice things about the Github web
Sure, done. I hope it sticks this time. -- |
/sponsor |
@theRealAph @AntonKozlov Since your change was applied there have been 191 commits pushed to the
Your commit was automatically rebased without conflicts. Pushed as commit ec9bee6. |
Please review a change to remove Linux-specific code from hotspot/cpu/aarch64.
The change is made to prepare for non-linux aarch64 platforms.
vm_version no longer contains any "raw" values that were obtained directly from the platform registers. Some of them may be unavailable on certain architectures, like
ctr_el0
is not available on windows [1].Few opportunities to improve linux code were taken:
HWCAP_
flags now explicitly mapped toCPU_
_dcpop
boolean was replaced withCPU_DCPOP
Graal-related changes: oracle/graal#2861
Testing:
isDcZvaProhibited
andzvaLength
values in Graal (not committed) -- have not changed./graal-compiler-micro-benchmarks.jar -p size=128,256,512,1024,2048,4096,8192,16384,32768,65536,131072 ArrayAllocationBenchmark.arrayAllocate
(oracle/graal@dcb4506) -- the benchmark is noisy in my environment, probably too many threads are used. Manually compared iterations, there is no regression spotted.[1] https://mail.openjdk.java.net/pipermail/aarch64-port-dev/2020-September/009690.html
Progress
Issue
Reviewers
Download
$ git fetch https://git.openjdk.java.net/jdk pull/154/head:pull/154
$ git checkout pull/154