public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "lewissbaker.opensource at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug c++/95615] New: [coroutines] Coroutine frame and promise is leaked if exception thrown from promise.initial_suspend()
Date: Tue, 09 Jun 2020 18:35:25 +0000	[thread overview]
Message-ID: <bug-95615-4@http.gcc.gnu.org/bugzilla/> (raw)

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95615

            Bug ID: 95615
           Summary: [coroutines] Coroutine frame and promise is leaked if
                    exception thrown from promise.initial_suspend()
           Product: gcc
           Version: 10.1.1
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: lewissbaker.opensource at gmail dot com
  Target Milestone: ---

If an exception is thrown from any of the following:
- promise_type constructor
- promise.get_return_object()
- promise.initial_suspend()
- initSuspendAwaitable.await_ready()
- initSuspendAwaitable.await_suspend()

Then I believe the compiler is required to automatically
destroy the promise (if constructor completed successfully),
destroy the return return-object (if get_return_object() completed
successfully) and free the coroutine frame before letting the
exception propagate to the caller.

However, it seems that this cleanup logic is not currently
begin called by GCC in these situations.

For example, see https://godbolt.org/z/kQWjpF which shows the
behaviour when an exception is thrown from promise.initial_suspend()'s
await_suspend() method (other variants are commented out in the code).


The wording described in [dcl.fct.def.coroutine] p5 in conjunction with p11
seems to indicate that the promise object should be destroyed as the
exception propagates out of the coroutine.

             reply	other threads:[~2020-06-09 18:35 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-06-09 18:35 lewissbaker.opensource at gmail dot com [this message]
2020-06-09 18:56 ` [Bug c++/95615] " iains at gcc dot gnu.org
2020-07-23  6:52 ` rguenth at gcc dot gnu.org
2021-02-15 14:00 ` iains at gcc dot gnu.org
2021-02-16  8:50 ` iains at gcc dot gnu.org
2021-03-05 16:59 ` cvs-commit at gcc dot gnu.org
2021-03-22 22:04 ` cvs-commit at gcc dot gnu.org
2021-03-24 12:37 ` iains 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-95615-4@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).