Re: Problems Configuring (C Compiler cannot produce executables)

2012-08-23 Thread Miles Bader
Russ Allbery r...@stanford.edu writes:
 Also, you should generally not add -Wall -Wextra to the configure flags,
 and instead add it after configure completes, since many of the tricks
 configure has to use will result in warnings when you turn on all the
 compiler warnings, which can confuse configure.

How can that confuse configure?

AFAICT, configure seems quite unconcerned with warnings during
configuration.

Thanks,

-miles

-- 
Is it true that nothing can be known?  If so how do we know this?  -Woody Allen

___
Autoconf mailing list
Autoconf@gnu.org
https://lists.gnu.org/mailman/listinfo/autoconf


Re: Problems Configuring (C Compiler cannot produce executables)

2012-08-23 Thread Russ Allbery
Miles Bader mi...@gnu.org writes:
 Russ Allbery r...@stanford.edu writes:

 Also, you should generally not add -Wall -Wextra to the configure
 flags, and instead add it after configure completes, since many of the
 tricks configure has to use will result in warnings when you turn on
 all the compiler warnings, which can confuse configure.

 How can that confuse configure?

 AFAICT, configure seems quite unconcerned with warnings during
 configuration.

I may be misremembering previous discussions here, and the fact that we do
indeed seem to pass -Wall into configure all the time without any trouble
makes me think that I am misremembering, but I thought there were some
checks where (due largely to various broken vendor compilers) configure
had to analyze the compiler output to figure out if things went wrong.

It's possible that I'm conflating this discussion with cases where people
use -Werror, which has more obvious issues.

-- 
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/

___
Autoconf mailing list
Autoconf@gnu.org
https://lists.gnu.org/mailman/listinfo/autoconf


Re: [autoconf] Re: Problems Configuring (C Compiler cannot produce executables)

2012-08-23 Thread suzuki toshiya
It is not autoconf's responsibility, and I think the testing code
in configure should not depend with -Wall and -Wextra, but some
software packages may have the configure checking if any warnings
are issued /or not. I remember, the configure in pixman had ever
checked the warning.

Regards,
mpsuzuki

Miles Bader wrote:
 Russ Allbery r...@stanford.edu writes:
 Also, you should generally not add -Wall -Wextra to the configure flags,
 and instead add it after configure completes, since many of the tricks
 configure has to use will result in warnings when you turn on all the
 compiler warnings, which can confuse configure.
 
 How can that confuse configure?
 
 AFAICT, configure seems quite unconcerned with warnings during
 configuration.
 
 Thanks,
 
 -miles
 


___
Autoconf mailing list
Autoconf@gnu.org
https://lists.gnu.org/mailman/listinfo/autoconf


Re: Problems Configuring (C Compiler cannot produce executables)

2012-08-23 Thread Ralf Corsepius

On 08/23/2012 09:14 AM, Miles Bader wrote:

Russ Allbery r...@stanford.edu writes:

Also, you should generally not add -Wall -Wextra to the configure flags,
and instead add it after configure completes, since many of the tricks
configure has to use will result in warnings when you turn on all the
compiler warnings, which can confuse configure.


How can that confuse configure?


No idea. The working priciples of standard autoconf checks are based on 
evaluating compiler errors only and to ignore warnings[1], therefore 
-Wall -Wextra must not desturb by definition.


However, adding -Werror to CFLAGS is dangerous, because this will raise 
GCC warnings to errors, which will cause autoconf to become confused and 
to produce bogus results.


Ralf

[1] There exist (non-standard) autoconf checks which are based on 
evaluating compiler warnings. If properly written, these also should not 
be affected by -Wall -Wextra, ... if they are, these checks need to be 
considered broken ;)



___
Autoconf mailing list
Autoconf@gnu.org
https://lists.gnu.org/mailman/listinfo/autoconf


Re: [autoconf] Problems Configuring (C Compiler cannot produce executables)

2012-08-23 Thread Jeffrey Walton
On Wed, Aug 22, 2012 at 7:21 PM, Mike Frysinger vap...@gentoo.org wrote:
 On Wednesday 22 August 2012 18:17:37 Jeffrey Walton wrote:
 The posture would have saved a number of folks from, for example,
 Pidgin's latest rounds of Critical Vulnerabilities (memory corruption
 and code execution). No-exec stacks and heaps would have reduced
 many/most to an annoying UI problem (a call to abort()).

 bad example: pidgin doesn't require execstacks (i'm not sure it ever has), so
 that would have made 0 difference.
I think Pidgin is a perfect example (humbly):
http://www.pidgin.im/news/security/ and
http://www.securityfocus.com/archive/105/515814.

If Pidgin does not require NX stacks and heaps, why was it running
with them? Not only did Pidgin not observe an SDLC on Linux, it did
not do so on Windows either. Pidgin for Windows (2.10.6) - Missing
DEP and ASLR, http://developer.pidgin.im/ticket/15209,

Linux might not have an SDLC, but Microsoft certainly does.

 no one does exec-heaps by default ... the
 code itself has to explicitly do this, and there's nothing the toolchain could
 have done to stop that (not that pidgin enables exec on memory returned by
 malloc afaik).  only a kernel patch (such as PaX) which explicitly denies
 mprotect calls that try to enable exec  write bits simultaneously would have
 prevented this scenario proactively.
Its unfortunate that most Linux do not do no-exec heaps. Gentoo is an
exception with its PaX security.

X^W is fine, too. Nothing is befuddling since the defenses are there -
all a programmer has to do is ask for them.

 in fact, very little to no packages request an executable stack by default.
 binary-only packages tend to be the only ones nowadays that do, and that's
 usually because the people producing the pkgs have broken code.
All code has bugs. Its the reason we should be running with full defenses.

Jeff

___
Autoconf mailing list
Autoconf@gnu.org
https://lists.gnu.org/mailman/listinfo/autoconf


Re: [autoconf] Problems Configuring (C Compiler cannot produce executables)

2012-08-23 Thread Jeffrey Walton
On Wed, Aug 22, 2012 at 9:41 PM, Mike Frysinger vap...@gentoo.org wrote:
 On Wednesday 22 August 2012 18:28:52 Russ Allbery wrote:
 special exceptions.  Being able to turn off executable stack as at least
 another easily-accessible option is an interesting idea, and I may raise
 that on debian-devel.  (Although it can be a little hard to predict which
 packages need that.

 it's trivial to locate:
 readelf -lW /bin/bash | grep GNU_STACK
 if it's set as RWE, that's bad.  if the ELF lacks a GNU_STACK, that's bad.
I believe there is more to it. The PT_GNU_STACK marking must be
present *and* have a size of 0. See Ian Lance Taylor's blog
http://www.airs.com/blog/archives/120.

Trivia: How does one audit a Gentoo binary for no-exec heap
compliance. Hint: its not readelf because Gentoo did not modify the
utility to dump PaX flags. (I'm asking because it took me some time to
discover the information).

Jeff

___
Autoconf mailing list
Autoconf@gnu.org
https://lists.gnu.org/mailman/listinfo/autoconf


Re: [autoconf] Problems Configuring (C Compiler cannot produce executables)

2012-08-23 Thread Jeffrey Walton
On Thu, Aug 23, 2012 at 9:13 AM, Jeffrey Walton noloa...@gmail.com wrote:
 On Wed, Aug 22, 2012 at 9:41 PM, Mike Frysinger vap...@gentoo.org wrote:
 On Wednesday 22 August 2012 18:28:52 Russ Allbery wrote:
 special exceptions.  Being able to turn off executable stack as at least
 another easily-accessible option is an interesting idea, and I may raise
 that on debian-devel.  (Although it can be a little hard to predict which
 packages need that.

 it's trivial to locate:
 readelf -lW /bin/bash | grep GNU_STACK
 if it's set as RWE, that's bad.  if the ELF lacks a GNU_STACK, that's bad.
 I believe there is more to it. The PT_GNU_STACK marking must be
 present *and* have a size of 0. See Ian Lance Taylor's blog
Bad reference (my apologies). Try Ian's blog on Executable Stacks
instead http://www.airs.com/blog/archives/518.

Jeff

___
Autoconf mailing list
Autoconf@gnu.org
https://lists.gnu.org/mailman/listinfo/autoconf


Re: Add more check in autoreconf

2012-08-23 Thread Javier Jardón
On 17 August 2012 05:17, Eric Blake ebl...@redhat.com wrote:

 If you are willing to write patches to autoreconf to incorporate new
 tools into the common autotools, then we will gladly review them.  But I

Patch to check for gtk-doc sent:

Also, if someone can give some guideance about the best why to fix
this problem with upstream gettext: [2]

Thanks!

[1] http://lists.gnu.org/archive/html/autoconf-patches/2012-08/msg2.html
[2] http://lists.gnu.org/archive/html/bug-gettext/2012-08/msg3.html
-- 
Javier Jardón Cabezas

___
Autoconf mailing list
Autoconf@gnu.org
https://lists.gnu.org/mailman/listinfo/autoconf


Re: Problems Configuring (C Compiler cannot produce executables)

2012-08-23 Thread Russ Allbery
Ralf Corsepius rc040...@freenet.de writes:

 No idea. The working priciples of standard autoconf checks are based on
 evaluating compiler errors only and to ignore warnings[1], therefore -Wall
 -Wextra must not desturb by definition.

 However, adding -Werror to CFLAGS is dangerous, because this will raise
 GCC warnings to errors, which will cause autoconf to become confused and
 to produce bogus results.

 Ralf

 [1] There exist (non-standard) autoconf checks which are based on
 evaluating compiler warnings. If properly written, these also should not
 be affected by -Wall -Wextra, ... if they are, these checks need to be
 considered broken ;)

I've clearly just misremembered, then.  Apologies for the noise.

-- 
Russ Allbery (r...@stanford.edu) http://www.eyrie.org/~eagle/

___
Autoconf mailing list
Autoconf@gnu.org
https://lists.gnu.org/mailman/listinfo/autoconf