public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "de34 at live dot cn" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug libstdc++/105995] New: QoI: constexpr basic_string variable must use all of its SSO buffer
Date: Thu, 16 Jun 2022 07:13:57 +0000	[thread overview]
Message-ID: <bug-105995-4@http.gcc.gnu.org/bugzilla/> (raw)

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105995

            Bug ID: 105995
           Summary: QoI: constexpr basic_string variable must use all of
                    its SSO buffer
           Product: gcc
           Version: 12.1.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: libstdc++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: de34 at live dot cn
  Target Milestone: ---

The current implementation in libstdc++ only supports constexpr
std::basic_string variable of static storage duration which uses all of its SSO
buffer:

#include <string>

constexpr std::string str15{"0123456789abcde"};
// constexpr std::string str14{"0123456789abcd"}; // error: should we support
this?
// constexpr std::string str16{"0123456789abcdef"}; // error: unsupportable

constexpr std::u16string u16str7{u"0123456"};
constexpr std::u32string u32str3{U"012"};

int main()
{
    // constexpr std::string strdyn{"0123456789abcde"}; // error: unsupportable
}

Compiler Explorer links:
https://gcc.godbolt.org/z/cvbY1hK3G
https://gcc.godbolt.org/z/634888Y96

I think constexpr string variable that requires heap allocation is currently
unimplementable, and constexpr string variable of dynamic storage duration
can't be supported by libstdc++-v3 without ABI breakage now.

However, should we allow shorter strings to be stored into constexpr
basic_string variables? We can do this by always fully initializing the SSO
buffer (if it's used) during constant evluation.

             reply	other threads:[~2022-06-16  7:13 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-06-16  7:13 de34 at live dot cn [this message]
2022-06-16  7:19 ` [Bug libstdc++/105995] " pinskia at gcc dot gnu.org
2022-06-16  7:41 ` de34 at live dot cn
2022-06-16 11:08 ` redi at gcc dot gnu.org
2022-06-16 11:14 ` redi at gcc dot gnu.org
2022-06-16 19:20 ` cvs-commit at gcc dot gnu.org
2022-06-16 19:21 ` redi at gcc dot gnu.org
2022-08-03 13:47 ` cvs-commit at gcc dot gnu.org
2022-08-03 13:49 ` redi at gcc dot gnu.org

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=bug-105995-4@http.gcc.gnu.org/bugzilla/ \
    --to=gcc-bugzilla@gcc.gnu.org \
    --cc=gcc-bugs@gcc.gnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).