On Fri, 23 Jan 2026 17:47:38 GMT, Chris Plummer <[email protected]> wrote:
>> src/jdk.hotspot.agent/linux/native/libsaproc/symtab.c line 555:
>>
>>> 553: static bool is_in(uintptr_t offset, struct elf_symbol* sym) {
>>> 554: if (offset == sym->offset) {
>>> 555: // offset points the top of the symbol
>>
>> Suggestion:
>>
>> // offset points to the top of the symbol
>
> This code is now confusing to read without knowledge that it is trying to
> solve the issue of a 0 sized symbol. I would suggest having it check
> `(sym->size == 0 && offset == sym->offset)`, and add a comment explaining
> why, and then the check below should be reverted back to `offset >= 0`
> instead of `offset > 0`. It is confusing to not have it check the full
> address range, especially since the comment indicates that it does.
I fixed the conditions, and updated comments. Could you review again?
-------------
PR Review Comment: https://git.openjdk.org/jdk/pull/29023#discussion_r2723428565
