public inbox for glibc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "slyfox at inbox dot ru" <sourceware-bugzilla@sourceware.org>
To: glibc-bugs@sourceware.org
Subject: [Bug dynamic-link/25680] ifuncmain9picstatic and ifuncmain9picstatic crash in IFUNC resolver due to stack canary (--enable-stack-protector=all)
Date: Sun, 15 Mar 2020 12:18:35 +0000	[thread overview]
Message-ID: <bug-25680-131-0b8yFVYfG0@http.sourceware.org/bugzilla/> (raw)
In-Reply-To: <bug-25680-131@http.sourceware.org/bugzilla/>

https://sourceware.org/bugzilla/show_bug.cgi?id=25680

--- Comment #3 from Sergei Trofimovich <slyfox at inbox dot ru> ---
(In reply to Florian Weimer from comment #2)
> You should not use --enable-stack-protector=all, but
> --enable-stack-protector=strong.  “all” is very unlikely to add additional
> protection. If you do that, the IFUNC resolver will not be instrumented
> (because it has no local addressable variables, so no stack buffer overflow
> is possible), so the bug goes away.
> 
> I'm not sure if we can or should order the startup sequence for statically
> linked binaries differently, so I'm not closing this bug.

Oh, that's surprising to hear. Gentoo will have to switch the default from =all
to =strong then.

Does it mean --enable-stack-protector=all is not really a supported option for
glibc? If' it's not I can hack a patch that complains loudly about the option
being experimental.

My main worry is that resolver in ifuncmain9picstatic's source code only
happens to be a part of glibc. It could have been an external program. Any
external program might decide to enable -fstack-protector-all and will get
non-working resolver as a result. Is there a doc that describes this and other
limitations imposed on IFUNC resolves by glibc? I wonder what else we break by
accident.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

  parent reply	other threads:[~2020-03-15 12:18 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-03-15 10:15 [Bug dynamic-link/25680] New: " slyfox at inbox dot ru
2020-03-15 10:18 ` [Bug dynamic-link/25680] " slyfox at inbox dot ru
2020-03-15 10:39 ` fw at deneb dot enyo.de
2020-03-15 12:18 ` slyfox at inbox dot ru [this message]
2020-03-15 12:25 ` fw at deneb dot enyo.de
2020-04-05 19:17 ` nick.alcock at oracle dot com
2020-04-05 19:18 ` nick.alcock at oracle dot com
2020-04-05 19:20 ` nick.alcock at oracle dot com
2020-06-25 22:37 ` davidhughes205 at gmail dot com
2021-03-15 16:25 ` cvs-commit at gcc dot gnu.org
2021-03-15 16:36 ` siddhesh at sourceware dot org
2021-03-15 16:37 ` siddhesh at sourceware dot org
2024-06-12 23:22 ` sam at gentoo dot org
2024-06-12 23:22 ` sam at gentoo dot 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-25680-131-0b8yFVYfG0@http.sourceware.org/bugzilla/ \
    --to=sourceware-bugzilla@sourceware.org \
    --cc=glibc-bugs@sourceware.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).