RFR: 8307084: C2: Vectorized drain loop is not executed for some small trip counts
Emanuel Peter
epeter at openjdk.org
Tue Aug 26 12:50:37 UTC 2025
On Thu, 21 Aug 2025 10:30:40 GMT, Fei Gao <fgao at openjdk.org> wrote:
>> Thanks for the updates. I gave it a quick scan and proposed some changes. I can look at it again once you repond to these :)
>> (we currently have lots of reviews, so I need to do a little round-robin here 🙈 )
>
> Hi @eme64 , I’ve addressed some corner case failures and refined parts of the code in the new commit. Would you like to review it? Thanks!
@fg1417 Yes, I'd love to review! I'll try to have a look in the next days.
It would be good if you re-ran the benchmarks. It seems the last ones you did in December of 2024.
We should see that we have various benchmarks, both for array and MemorySegment.
You could look at the array benchmarks from here: https://github.com/openjdk/jdk/pull/22070
-------------
PR Comment: https://git.openjdk.org/jdk/pull/22629#issuecomment-3224041030
More information about the hotspot-compiler-dev
mailing list