RFR: 8343789: Move mutable nmethod data out of CodeCache [v14]
Vladimir Kozlov
kvn at openjdk.org
Tue Apr 29 21:20:03 UTC 2025
On Tue, 29 Apr 2025 20:55:40 GMT, Dean Long <dlong at openjdk.org> wrote:
>> I think in 8331087, I think only `_jvmci_data_offset` was subject to the [narrowing cast](https://github.com/openjdk/jdk/pull/18984/files#diff-c345a29edc19eb49f833e007143e00f897bb762b0f451a202e1d2f5304ed2308R1496).
>> I've opened https://bugs.openjdk.org/browse/JDK-8355896.
>
> Yes, my mistake. I was thinking `_jvmci_data_offset` was used to compute `jvmci_data_end()`, not `jvmci_data_begin()`.
Yes we should use 32 bits. Even if we revert back to using _jvmci_data_offset we can **NOT** use uint16_t because size of relocation (after which JVMCI data is placed) data is bigger.
-------------
PR Review Comment: https://git.openjdk.org/jdk/pull/21276#discussion_r2067442878
More information about the hotspot-compiler-dev
mailing list