[marcin-serwin]

Building with gcc on aarch64-linux fails otherwise.

[kd-11]

QQ - Then why was linux-arm64 CI not failing if this is a problem?
e.g latest run -> https://github.com/RPCS3/rpcs3/actions/runs/15674041842/job/44150272330

[marcin-serwin]

QQ - Then why was linux-arm64 CI not failing if this is a problem?

The linux-arm64 CI is using clang, not gcc. Perhaps a gcc workflow should also be added to match the x86 ones which test both compilers?

[Megamouse]

QQ - Then why was linux-arm64 CI not failing if this is a problem? e.g latest run ->
the sse2neon header uses this:

#if defined(__clang__) && defined(ARCH_ARM64)
#pragma clang diagnostic push
#pragma clang diagnostic ignored "-Wold-style-cast"
#endif

[kd-11]

QQ - Then why was linux-arm64 CI not failing if this is a problem?
The linux-arm64 CI is using clang, not gcc. Perhaps a gcc workflow should also be added to match the x86 ones which test both compilers?

Yes, this is a good point. We usually have both gcc and clang.