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]
Date:	Wed, 28 Apr 2010 10:09:02 -0400
From:	Mathieu Desnoyers <mathieu.desnoyers@...icios.com>
To:	Tony Breeds <tony@...eyournoodle.com>,
	linux-kernel@...r.kernel.org, dank@...el.com
Subject: Re: [PATCH 00/11] jump label v7

* Tony Breeds (tony@...eyournoodle.com) wrote:
> On Tue, Apr 27, 2010 at 09:49:41PM -0400, Mathieu Desnoyers wrote:
> 
> > The crosstool package from Dan Kegel did a good job for this. Not sure
> > it's currently maintained though. It was a very useful project, it's a
> > shame if it does not live on. It would be good to have up-to-date and
> > tested compilers for various architectures available on kernel.org,
> > ideally with access to a package that helps building compilers for
> > various architectures.
> 
> Trimmed the CC' as this is a bit of a tangent.
> 
> Crosstool did a good job but seems to be less current than we'd like for the
> kernel[1].

Ouch. The build failure ratio is surprisingly high.

>            When using crosstool in the past I found a lot of the complexity
> was in *libc to the toolcahins I've built don';t have a libc so they're only
> helpful for kernel (or similar) builds.

So maybe we could do something more specialized for kernel needs, e.g. a
kcrosstool ? It could be a "simplified" crosstool.

Personally, I built a set of cross-compilers with crosstool 0.43 a while
ago, but must now do a lot of trial and error to find new correct compiler
and toolchain versions for each architecture. It would be good to have a
kcrosstool git repository that would:

- Have scripts and documentation specialized for Linux kernel build.
- Contain an updated list of "working" and "non-working" toolchain
  versions for each arch.
- Could be easily updated; people would just have to report
  success/failure of their config so these could be added to the git
  repository. The build could probably prepare a report automatically to
  lessen the reporter job (I'm even thinking about a scheme that would
  give the option to report automatically).
- Optionally contain documentation on how to set up a cron job to
  automatically fetch a git tree and do a distributed multi-architecture
  build each day.

I think have this kind of updated tool would be of great value for
testing.

Thanks,

Mathieu

> 
> In terms of "testing", I build a kernel for each toolchain but I don't boot
> them and in the past I think blackfin said that what I had wouldn't actually be
> runable.  Last time I looked the wasn't a non-intel cross-toolchain for blackfin.
> 
> I shoudl also mention that a subset of these compilers are used to build test
> linux-next daily.
> 
> Yours Tony
> 
> [1] http://www.kegel.com/crosstool/crosstool-0.43/buildlogs/

-- 
Mathieu Desnoyers
Operating System Efficiency R&D Consultant
EfficiOS Inc.
http://www.efficios.com
--
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