Hello,
The initial goal was to verify that a KSYM_NAME_LEN of 512 was working as expected. It also fixed an issue in /x86/tools/insn_decoder_test.c.
This fix is indeed quite trivial, I just implemented the suggested ideas. Amend the test using KSYM_NAME_LEN-8 doesn't really seem to me to fix the underlying problem (I also don't know if there is any magic value).
Take into account that he actual warnings that have been reported in the CI ("Please increase KSYM_NAME_LEN both in kernel and kallsyms.c") are not from this test, but from https://elixir.bootlin.com/linux/v6.15/source/scripts/kallsyms.c#L146, the test has just make that warning evident. Would removing that comment be a better solution?
Nevertheless, acknowledging that we don't have full control over the symbols' names, is also an interesting insight.
I will be happy to address the issue in one way or another, no problem there.
Regards, Sergio