public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "ndesaulniers at google dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug c/108548] New: gcc asm goto with outputs not implicitly volatile
Date: Wed, 25 Jan 2023 22:10:12 +0000	[thread overview]
Message-ID: <bug-108548-4@http.gcc.gnu.org/bugzilla/> (raw)

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108548

            Bug ID: 108548
           Summary: gcc asm goto with outputs not implicitly volatile
           Product: gcc
           Version: 12.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
          Assignee: unassigned at gcc dot gnu.org
          Reporter: ndesaulniers at google dot com
                CC: eli.friedman at gmail dot com, foom at fuhm dot net
  Target Milestone: ---

https://gcc.gnu.org/onlinedocs/gcc/Extended-Asm.html#Volatile mentions:

> GCC’s optimizers sometimes discard asm statements if they determine there is no need for the output variables. ... Using the volatile qualifier disables these optimizations. asm statements that have no output operands and asm goto statements, are implicitly volatile.

So what about asm goto statements that do have outputs, which GCC recently
gained support for?

Looks like I get different codegen targeting x86 at -O2 with:

```
int foo (void) {
    void *x;
    asm goto (
        "movq   %l1, %0\n\t"
        "jmp    *%0"
        :"=r"(x):::bar);
    return 0;
bar:
    return 1;
}
```
based on whether the asm goto statement is marked volatile or not.

Should asm goto statements with outputs be implicitly volatile (implying a bug
currently in GCC) or not (implying the documentation could perhaps be updated)?

             reply	other threads:[~2023-01-25 22:10 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-01-25 22:10 ndesaulniers at google dot com [this message]
2023-01-25 22:17 ` [Bug middle-end/108548] " pinskia 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-108548-4@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).