RFR: 8365163: [ubsan] left-shift issue in globalDefinitions.hpp [v3]
Afshin Zafari
azafari at openjdk.org
Tue Aug 26 12:34:38 UTC 2025
On Tue, 26 Aug 2025 11:38:03 GMT, Stefan Karlsson <stefank at openjdk.org> wrote:
> > Do we need to consider endianess here?
>
> I guess the answer is no because the changed code and tests only operators on the full sized variables and doesn't try to extract and operate on parts of the full sized variables.
>
> Did you see that the new test failed on Windows?
That is right, no need to consider endianness.
Windows failure is due to 'long' casting instead of 'long long'. Fixed it and is under tests.
-------------
PR Comment: https://git.openjdk.org/jdk/pull/26809#issuecomment-3223985684
More information about the hotspot-dev
mailing list