https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118141
Sam James <sjames at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |sjames at gcc dot gnu.org
--- Comment #6 from Sam James <sjames at gcc dot gnu.org> ---
(In reply to Richard Yao from comment #3)
> (In reply to Andrew Pinski from comment #2)
> > for (int i = 0; i < 8; i++) {
> > input_vec[i] = ((unsigned int*) input)[i];
> > }
> >
> > float input[8] = {
> > 1.0f, 2.0f, 3.0f, 4.0f,
> > 5.0f, 6.0f, 7.8f, 8.0f
> > };
> >
> > unsigned short output[8];
> >
> > // Convert fp32 to bfloat16
> > convert_fp32_to_bfloat16(input, output);
> >
> >
> > You are violating C/C++ aliasing rules. Either use memcpy (with c++20, you
> > can use std::bit_cast), an union to do type punning or use
> > -fno-strict-aliasing.
> >
> > Adding -fno-strict-aliasing fixes the issue.
>
> Where is the strict aliasing rule violation? input is a void pointer. We are
> allowed to cast it to another pointer type and use it as per the C
> specification.
No, you cannot then access it via void*, you have to cast it to the original
type and use it (or a compatible type).
