You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Building 4.4.1 on aarch64-linux with gcc 13 results in:
/build/source/plugins/hawkeye/src/sbox_database.cpp: In function 'smallset_t hal::hawkeye::{anonymous}::smallset_init_full(u32)':
/build/source/plugins/hawkeye/src/sbox_database.cpp:868:56: error: cannot convert '__Uint64x2_t' to 'uint32x4_t'
868 | return {vsetq_lane_u32(0xFFFFFFFF, tmp, 0), vdupq_n_u64(0)};
| ^~~
| |
| __Uint64x2_t
etc. ad nauseum.
I'm guessing this is because you've generally been building using LLVM on aarch64, which appears to have -flax-vector-conversions enabled by default. But gcc doesn't. Builds fine with that flag added.
The text was updated successfully, but these errors were encountered:
Building 4.4.1 on aarch64-linux with gcc 13 results in:
etc. ad nauseum.
I'm guessing this is because you've generally been building using LLVM on aarch64, which appears to have
-flax-vector-conversions
enabled by default. But gcc doesn't. Builds fine with that flag added.The text was updated successfully, but these errors were encountered: