lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <48208F4F.1030405@de.ibm.com>
Date:	Tue, 06 May 2008 19:03:11 +0200
From:	Peter Oberparleiter <peter.oberparleiter@...ibm.com>
To:	Sam Ravnborg <sam@...nborg.org>
CC:	ltp-list@...ts.sourceforge.net,
	Andrew Morton <akpm@...ux-foundation.org>,
	ltp-coverage@...ts.sourceforge.net, linux-kernel@...r.kernel.org
Subject: Re: [RFC PATCH 3/6] kbuild: convert include and	source paths

Sam Ravnborg wrote:
> On Mon, May 05, 2008 at 05:24:26PM +0200, Peter Oberparleiter wrote:
>> From: Peter Oberparleiter <peter.oberparleiter@...ibm.com>
>> 
>> Modify kbuild to convert relative include and source paths to absolute
>> form. Also change the module versioning mechanism to alter object file
>> names only after compiling.
>> 
>> Required by the gcov profiling infrastructure: source paths are
>> referenced by the compiled object files. Using relative paths or
>> object file names which are different from the source name would
>> prevent the gcov tool from finding the corresponding source files.
> 
> For the relative path issue do you really fix this at the right place?
> I would say that teaching gcov tool from where to look would be better.

Actually it is gcc that would need to be changed to only include
absolute paths in compiled object files. While I haven't led that
particular discussion with gcc developers yet, I could imagine that they
would argue that such an option already exists: by specifying only
absolute paths during compilation. Also, any dependency on a future
version of gcc would, in my opinion, greatly reduce the overall
usefulness of this gcov kernel infrastructure.

Would it be acceptable to isolate these changes to kbuild by
if-statements, i.e. use absolute paths only of CONFIG_GCOV is
specified (as Jan Engelhardt suggested)?

>> Also change the module versioning mechanism to alter object file
>> names only after compiling.
> Can I get this as a separate patch please.
> It may be general useful and makes sense to look at alone.

Ok, I will do that.


Regards,
  Peter
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ