[Bug lto/63409] [5 Regression] FAIL: g++.dg/lto/pr63270 cp_lto_pr63270_0.o-cp_lto_pr63270_1.o link, -flto -O2 -Wno-odr -fno-linker-plugin

["ubizjak at gmail dot com" <gcc-bugzilla@gcc.gnu.org>]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63409

--- Comment #9 from Uroš Bizjak <ubizjak at gmail dot com> ---
(In reply to Martin Liška from comment #8)
> Created attachment 33653 [details]
> Fix patch2

Yes, this patch works for me.
>From gcc-bugs-return-463421-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Tue Oct 07 07:23:25 2014
Return-Path: <gcc-bugs-return-463421-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 19618 invoked by alias); 7 Oct 2014 07:23:25 -0000
Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc-bugs.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc-bugs/>
List-Post: <mailto:gcc-bugs@gcc.gnu.org>
List-Help: <mailto:gcc-bugs-help@gcc.gnu.org>
Sender: gcc-bugs-owner@gcc.gnu.org
Delivered-To: mailing list gcc-bugs@gcc.gnu.org
Received: (qmail 19554 invoked by uid 55); 7 Oct 2014 07:23:19 -0000
From: "rguenther at suse dot de" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug lto/61886] [4.8/4.9/5 Regression] LTO breaks fread with _FORTIFY_SOURCE=2
Date: Tue, 07 Oct 2014 07:23:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: lto
X-Bugzilla-Version: 4.9.1
X-Bugzilla-Keywords: diagnostic, lto, wrong-code
X-Bugzilla-Severity: normal
X-Bugzilla-Who: rguenther at suse dot de
X-Bugzilla-Status: ASSIGNED
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: hubicka at gcc dot gnu.org
X-Bugzilla-Target-Milestone: 4.8.4
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields:
Message-ID: <bug-61886-4-uibUm7mdxP@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-61886-4@http.gcc.gnu.org/bugzilla/>
References: <bug-61886-4@http.gcc.gnu.org/bugzilla/>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 7bit
X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/
Auto-Submitted: auto-generated
MIME-Version: 1.0
X-SW-Source: 2014-10/txt/msg00442.txt.bz2
Content-length: 1146

https://gcc.gnu.org/bugzilla/show_bug.cgi?ida886

--- Comment #19 from rguenther at suse dot de <rguenther at suse dot de> ---
On Mon, 6 Oct 2014, hubicka at ucw dot cz wrote:

> https://gcc.gnu.org/bugzilla/show_bug.cgi?ida886
>
> --- Comment #11 from Jan Hubicka <hubicka at ucw dot cz> ---
> Hi,
> this patch implements the lowring.  Each call with warn attribute triggers code
> in cgraphunit that inserts call to bulitin_warning/error that is output at
> expansion time.
>
> Do we have way to define bulitin that is not user accessible?
>
> Also we do not have way to define LOOPING_CONST bulitin, so I am simply forcing
> the flag in cgraphunit.c that is somewhat ugly.

But this also means that when attaching the attribute to pure/const
functions which result becomes unused and thus the call gets DCEd
will still emit the warning/error.  Similar if the function gets
inlined (formerly no warning/error).

I think you need an alternate lowering for the first case at least,
like passing through the return value.  Not sure how to deal with the 2nd
case.

That said, I don't think this is the way to go to implement the
attribute.