[jdk11u-dev] RFR: 8292713: Unsafe.allocateInstance should be intrinsified without UseUnalignedAccesses [v2]
Fei Yang
fyang at openjdk.org
Mon Sep 11 14:29:22 UTC 2023
> Backport to improve arm32 performance in 11u and riscv-port in: https://github.com/openjdk/riscv-port-jdk11u/pull/3.
> This is not a clean backport as the function has been moved to new file and renamed.
>
> I checked the 11u related code and I think the following analysis on the JBS description is still true for 11u:
> "The intrinsic is currently only implemented in C2. I see no reason for this intrinsic to depend on `UseUnalignedAccesses`: what `LibraryCallKit::inline_unsafe_allocate()` does seems similar to what the regular `Parse::do_new()` does, namely doing the clinit barrier and then delegating into `GraphKit::new_instance`."
>
> Tested with the initial riscv-port in: https://github.com/openjdk/riscv-port-jdk11u/pull/3
Fei Yang has updated the pull request with a new target base due to a merge or a rebase. The incremental webrev excludes the unrelated changes brought in by the merge/rebase. The pull request contains two additional commits since the last revision:
- Merge remote-tracking branch 'upstream/master' into backport-8292713
- 8292713: Unsafe.allocateInstance should be intrinsified without UseUnalignedAccesses
-------------
Changes:
- all: https://git.openjdk.org/jdk11u-dev/pull/2125/files
- new: https://git.openjdk.org/jdk11u-dev/pull/2125/files/22895ae1..719596c5
Webrevs:
- full: https://webrevs.openjdk.org/?repo=jdk11u-dev&pr=2125&range=01
- incr: https://webrevs.openjdk.org/?repo=jdk11u-dev&pr=2125&range=00-01
Stats: 110 lines in 13 files changed: 38 ins; 25 del; 47 mod
Patch: https://git.openjdk.org/jdk11u-dev/pull/2125.diff
Fetch: git fetch https://git.openjdk.org/jdk11u-dev.git pull/2125/head:pull/2125
PR: https://git.openjdk.org/jdk11u-dev/pull/2125
More information about the jdk-updates-dev
mailing list