-
Notifications
You must be signed in to change notification settings - Fork 170
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
8271567: AArch64: AES Galois CounterMode (GCM) interleaved implementation using vector instructions #87
Conversation
👋 Welcome back aph! A progress list of the required criteria for merging this PR into |
This backport pull request has now been updated with issue from the original commit. |
Or perhaps include the patch but disable the intrinsic by default. |
@theRealAph This pull request has been inactive for more than 4 weeks and will be automatically closed if another 4 weeks passes without any activity. To avoid this, simply add a new comment to the pull request. Feel free to ask for assistance if you need help with progressing this pull request towards integration! |
Hi! Can someone look at this, please? |
I'd be happy with "include the patch but disable the intrinsic by default". That poses a pretty low risk for users who don't need this but, as you say, should a big difference for apps that need AES GCM -- enough for them to want it despite the limited upstream baking time. |
There's no point simply turning off acceleration for safety because this patch includes a partial rewrite of the code that gets executed even when the new accelerator is disabled, so it'd add risk without improving anything: a very bad bargain. This is a bit of a "Physician, heal thyself!" moment -- it's what I'm forever asking back-porters to do for JDK 11u. Oops, my bad. |
I'm closing this PR. It's replaced by https://git.openjdk.java.net/jdk17u/pull/216 |
This is a backport of the big AES/GCM patch from JDK head. It's a
major change and it's had very little time (almost a day) to mature in
head, so perhaps it shouldn't be backported for some time, However, there is a good reason for a
backport: OpenJDK on x86 has a major advantage. AES/GCM is an
important cipher, the current AArch64 implementation is much slower
than x86, and some workloads are severely impacted.
I'm open to all arguments about why this should or shouldn't be pushed,
and I'm quite happy to wait for another release cycle or two if people
think that's the best course of action.
Progress
Issue
Reviewing
Using
git
Checkout this PR locally:
$ git fetch https://git.openjdk.java.net/jdk17u pull/87/head:pull/87
$ git checkout pull/87
Update a local copy of the PR:
$ git checkout pull/87
$ git pull https://git.openjdk.java.net/jdk17u pull/87/head
Using Skara CLI tools
Checkout this PR locally:
$ git pr checkout 87
View PR using the GUI difftool:
$ git pr show -t 87
Using diff file
Download this PR as a diff file:
https://git.openjdk.java.net/jdk17u/pull/87.diff