Submitter-Id: net
Originator:MadCoder
Organization:
Confidential: no
Synopsis: Error of compilation of GCJ-3.0 (debian gcj-3.0.4-5
Severity: critical
Priority: medium
Category: java
Class: wrong-code
Release: 3.0.4 (Debian testing/unstable)
Environment:
Package: gcc-4.4
Version: 4.4.3-4
Severity: grave
Since gcc-4.4 version 4.4.3-4 (and yes -5 is still affected), gcc miscompiles
__builtin_expect when no optimization is set (at least).
Test case:
int foo(int t) {
if (__builtin_expect(t 0x100, 0))
return 0;
that pops up at the -O0 level that never shows up with any
other gcc release. And I deeply trust the mentioned code to be correct.
The code in question uses a lot of gcc __builtin_* functions if that
helps (ctz, clz, bswap among other).
On Thu, Apr 01, 2010 at 01:38:20AM +0200, Pierre Habouzit wrote
it on the work repository, and lto ICEs the compiler. Plus I
get 2 other independant ICEs that I've not had time to reduce (hence the
lack of bug report yet).
Though this happens with the gcc-4.5 in unstable, I've not tried with
the one from experimental yet.
--
·O· Pierre Habouzit
··O
,--add-needed). But clearly, boost is wrong here, sadly.
--
·O· Pierre Habouzit
··Omadco...@debian.org
OOOhttp://www.madism.org
--
To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org
On Mon, Aug 23, 2010 at 02:02:35PM +0200, Matthias Klose wrote:
On 23.08.2010 13:21, Pierre Habouzit wrote:
On Mon, Aug 23, 2010 at 04:05:32AM +0200, Matthias Klose wrote:
On 21.08.2010 14:56, Julien Cristau wrote:
On Fri, Aug 20, 2010 at 19:33:12 +0200, Arthur Loiret wrote:
Now
On Mon, Aug 23, 2010 at 02:05:25PM +0200, Matthias Klose wrote:
On 23.08.2010 13:30, Pierre Habouzit wrote:
On Mon, Aug 23, 2010 at 01:21:04PM +0200, Pierre Habouzit wrote:
On Mon, Aug 23, 2010 at 04:05:32AM +0200, Matthias Klose wrote:
On 21.08.2010 14:56, Julien Cristau wrote:
On Fri, Aug
Package: gcc-4.5
Version: 4.5.1-2
Severity: normal
$ cat a.c
#pragma GCC optimize(-O3)
int main(void)
{
return 0;
}
$ gcc-4.5 -o /dev/null -c -O2 -flto a.c
a.c:6:1: sorry, unimplemented: gimple bytecode streams do not support the
optimization attribute
--
To UNSUBSCRIBE, email to
On Mon, Aug 23, 2010 at 02:05:25PM +0200, Matthias Klose wrote:
On 23.08.2010 13:30, Pierre Habouzit wrote:
On Mon, Aug 23, 2010 at 01:21:04PM +0200, Pierre Habouzit wrote:
On Mon, Aug 23, 2010 at 04:05:32AM +0200, Matthias Klose wrote:
On 21.08.2010 14:56, Julien Cristau wrote:
On Fri, Aug
Package: gcc-4.5
Version: 4.5.1-2
Severity: important
Forwarded: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44802
This is upstream http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44802
In other words, as soon as you use visibility and commodity archives,
you can't use LTO. Which makes it worthless
Package: gcc-snapshot
Version: 20070422-1
Severity: important
sample code:
=
#include stdio.h
#include stdint.h
#include stdlib.h
static int bar(void *p)
{
return (int)(intptr_t)p;
}
static int foo(const void *p)
{
return bar((void *)p);
}
int
On Fri, May 11, 2007 at 04:46:28PM +0200, Martin Michlmayr wrote:
* Pierre Habouzit [EMAIL PROTECTED] [2007-05-11 16:34]:
does not work:
/usr/lib/gcc-snapshot/bin/gcc -O2 -o a.o a.c
a.c: In function 'main':
a.c:12: warning: passing argument 1 of 'bar' discards qualifiers from
On Wed, May 16, 2007 at 03:33:50PM +0200, Martin Michlmayr wrote:
* Pierre Habouzit [EMAIL PROTECTED] [2007-05-11 17:51]:
btree.c: In function 'fbtree_fetch':
btree.c:988: internal compiler error: in expand_builtin_expect, at
builtins.c:5089
Please submit a full bug report,
I've
Package: gdc-4.1
Version: 0.24-4.1.2-16
Severity: grave
Justification: renders package unusable
$ cat test.d
import std.c.stdio;
int main()
{
FILE *files[1] = [stdin];
return 0;
}
$ gdc -o /dev/null test.d
Package: gdc-4.1
Version: 0.25-4.1.2-17
Severity: grave
Justification: renders package unusable
The following code triggers the issue:
--
import std.c.process;
int main(char args[][]) {
execl(/bin/sh, sh, -c, ls, null);
printf(%m\n);
on the stack, which confuses the C.
I believe that D should do the implicit cast in that case, as it'll
break a _lot_ of programs using C APIs and variadic arguments in very
subtle ways.
--
·O· Pierre Habouzit
··O[EMAIL PROTECTED]
OOO
the other members correct me if I'm wrong.
--
·O· Pierre Habouzit
··O[EMAIL PROTECTED]
OOOhttp://www.madism.org
pgppVczXzW84q.pgp
Description: PGP signature
On Sun, Mar 23, 2008 at 07:23:10PM +, Florian Weimer wrote:
* Pierre Habouzit:
Isn't it risky for partial upgrades from etch ? Shouldn't we wait for
lenny+1 to revert this ?
I second that, please don't revert the patch until lenny+1. FWIW I
believe the release team as a whole
package,
which is not needed for an upgrade. Are you aware of others?
Yes, any package that runs memset/memmove in a signal handler. I
happened to do that on software at work, those are not open source, but
I would fail to see why we should not support them properly.
--
·O· Pierre Habouzit
··O
On Sun, Mar 23, 2008 at 07:41:59PM +, Pierre Habouzit wrote:
On Sun, Mar 23, 2008 at 04:54:29PM +, Matthias Klose wrote:
On Sun, Mar 23, 2008 at 07:15:11AM +, Mike Hommey wrote:
Isn't it risky for partial upgrades from etch ? Shouldn't we wait for
lenny+1 to revert
code developed on Debian would need
porting to work on other distributions. I believe Debian to be an
excellent development platform, and we should not lag behind wrt stable
development tools.
[1] http://gcc.gnu.org/ml/gcc-patches/2008-03/msg00417.html
--
·O· Pierre Habouzit
··O
toolchain right now.
--
·O· Pierre Habouzit
··O[EMAIL PROTECTED]
OOOhttp://www.madism.org
pgpao6lJq8rmi.pgp
Description: PGP signature
On Tue, Mar 25, 2008 at 08:39:01AM +, Bastian Blank wrote:
On Tue, Mar 25, 2008 at 09:26:51AM +0100, Pierre Habouzit wrote:
Problem is that memcpy/memmove/memset probably generate rep stos; in
the end, I believe memset/memcpy/memmove to be async signal safe, and
those are inlined
Package: gcc-4.6
Version: 4.6.0-6
Severity: important
gcc-4.6 has a regression wrt flattening (__attribute__((flatten))) that
makes it disregard the fact that some functions aren't inlineable.
This causes build failures if you call functions with va_args (this is
PR#48731 upstream).
It also
On Mon, May 02, 2011 at 11:28:45AM +0200, Matthias Klose wrote:
On 05/02/2011 10:35 AM, Pierre Habouzit wrote:
Package: gcc-4.6
Version: 4.6.0-6
Severity: important
The bug is fixed upstream please backport it, it's r172963 in the
gcc-4.6 branch.
4.6.0-6 is based on r173059. what
to be redesigned, else we'll have broken links every
other week or so. This results in build failures like #364959.
--
·O· Pierre Habouzit
··O[EMAIL PROTECTED]
OOOhttp://www.madism.org
Le Jeu 4 Mai 2006 09:01, Dererk a écrit :
I wonder if that means the bug was set as resolved/fixed...
I'm afraid I still having the same bug...
Waiting for news...
it means that the bug is *maybe* fixed upstream. meaning in the next
version.
--
·O· Pierre Habouzit
··O
.
There is no problems, only solutions.
best regards,
[1] http://lists.debian.org/debian-devel/2006/05/msg00078.html
and the rest of the thread.
--
·O· Pierre Habouzit
··O[EMAIL PROTECTED]
OOOhttp
permit users to go in one click to the right remote
bug report. That achieves IMHO the best compromise.
Don't hesitate to contact [EMAIL PROTECTED] to ask for
improvements !
Cheers,
[1] http://lists.debian.org/debian-devel/2006/05/msg00144.html
--
·O· Pierre Habouzit
··O
29 matches
Mail list logo
