https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112440
Jonathan Wakely <redi at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Last reconfirmed| |2023-11-08
Status|UNCONFIRMED |NEW
Ever confirmed|0 |1
--- Comment #1 from Jonathan Wakely <redi at gcc dot gnu.org> ---
(In reply to Antony Polukhin from comment #0)
> Consider the example:
>
> #include <string>
> void test1(std::size_t summ) {
> std::string result;
> result.resize(summ);
>
> if (result.size() > summ) {
> __builtin_abort();
> }
> }
>
> The resulting assembly contains `call abort` and code to check the string
> size: https://godbolt.org/z/zcj3Pc3G8
>
> Looks like this is due to char* aliasing with string internals, switching to
> std::u8string removes the `call abort` related assembly:
> https://godbolt.org/z/a6bKaqqn5
Yes, the problem of std::string's data member aliasing itself has been
discussed in bugzilla before. It's presumably also an issue for
std::vector<char>.
I think we discussed a new attribute which would say that none of the members
of *this refer to *this, but that wouldn't be correct for the SSO std::string,
where it's data pointer can point to its internal buffer. Maybe it would work
if it meant that the pointer doesn't alias anything of a different type in
*this. That would allow the char* std::string::_M_dataplus._M_p to point to the
char array std::string::_M_local_buf, but would tell the compiler it never
points to _M_string_length or _M_allocated_capacity.
> I've failed to come up with a generic solution, but looks like adding
> __builtin_unreachable() to the end of basic_string::resize and
> basic_string::reserve helps: https://godbolt.org/z/vWcjqGK94
>
>
> P.S.: such hints help to shorten the assembly for reserve+append*n cases
> https://godbolt.org/z/nsEGsWdP3 , https://godbolt.org/z/qMf4b7dd8 ,
> https://godbolt.org/z/1r6dd6d5M which are quire common
That seems worth doing, thanks for the suggestion.
