This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug target/81647] inconsistent LTGT behavior at different optimization levels on AArch64.


https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81647

--- Comment #5 from amker at gcc dot gnu.org ---
Looks like both x86_64(with -mavx2) and AArch64 translate u== into signaling
comparison instructions when vectorization.
x86_64 translates into "*avx_maskcmpv4df3_comm" thus:
   vcmpeq_uspd  x(,%r13,8), %ymm0, %ymm0
While AArch64 implements UNEQ as !LTGT in "vec_cmp<mode><v_cmp_result>".

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]