public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "johannes.kellner at wandelbots dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug c++/109470] New: unexpected const & behavior
Date: Tue, 11 Apr 2023 07:43:50 +0000	[thread overview]
Message-ID: <bug-109470-4@http.gcc.gnu.org/bugzilla/> (raw)

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109470

            Bug ID: 109470
           Summary: unexpected const & behavior
           Product: gcc
           Version: unknown
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: johannes.kellner at wandelbots dot com
  Target Milestone: ---

Hello, I am filing this ticket, as I had source code, where this funny Min
template created a seg fault when compiling release (optimized code).
I changed the this to an example - to check the behavior and the first to show
is this UNEXPECTED/INVALID? behavior:

#include <string>
#include <cassert>

template<typename TYPE>
const TYPE& Min(const TYPE& lhs, const TYPE& rhs)
{
   if (lhs < rhs) return lhs;
   return rhs;
}

int main()
{
   size_t lenght = 16;
   const int MAX = 32;

   const int& dst = Min(MAX, (int)lenght);
   assert(dst <= MAX);
   return dst;
}

Expected behavior: return 16

If compiled with gcc 12.2 and -O0: returns 16
If compiled with gcc 12.2 and -O1: returns 0  UNEXPECTED !!!

I assume, that something with the extension of lifetime, 
for the result of `(int)lenght` fails in the optimized code.

I found this unexpected behavior, because this Min function was used to
calculate the size for a memcpy. And if compiled with optimization, dst became
invalid (very large), just after the call to Min.

We tested with gcc10 but have the same problem with gcc12.

Please feel free to contact me for details.

             reply	other threads:[~2023-04-11  7:43 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-04-11  7:43 johannes.kellner at wandelbots dot com [this message]
2023-04-11  8:25 ` [Bug c++/109470] " xry111 at gcc dot gnu.org
2023-04-11  8:36 ` xry111 at gcc dot gnu.org
2023-04-11  9:48 ` johannes.kellner at wandelbots dot com
2023-04-11  9:52 ` pinskia at gcc dot gnu.org
2023-04-11  9:56 ` redi at gcc dot gnu.org
2023-04-11 10:23 ` johannes.kellner at wandelbots dot com
2023-04-11 10:36 ` jakub at gcc dot gnu.org
2023-04-11 10:37 ` 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-109470-4@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).