[PATCH 3/4] Introduce syscall fault injection feature

[Dmitry V. Levin]

Hi,

On Sat, Aug 20, 2016 at 02:44:55AM +0200, Nahim El Atmani wrote:
> On Fri, 29 Jul 2016 17:57:33 +0300, Dmitry V. Levin wrote:
> > On Fri, Jul 29, 2016 at 03:06:50PM +0200, Gabriel Laskar wrote:
> > > On Fri, 29 Jul 2016 04:04:02 +0300, Dmitry V. Levin wrote:
> > > > On Wed, Jul 27, 2016 at 08:41:16PM +0200, Nahim El Atmani wrote:
> > [...]
> > > > >   new configure flag have been created (--enable-fault-injection)
> > > > > to keep the build clean on other architectures.  
> > > > 
> > > > Please don't make this option configurable.  It complicates the code
> > > > (== causes mistakes) and gives very little in return.
> > > 
> > > The goal here was to enable the feature only for architecture where
> > > this was really tested, mainly in order to validate the model, and
> > > after, enable it without an option, when properly tested on all
> > > architectures.
> > 
> > If these arch-specific functions are difficult to implement,
> > let's re-use the approach taken by arch_sigreturn.
> > 
> > However, all arch specific details are well known so I think
> > this is not going to be as difficult as arch_sigreturn.
> 
> I'm currently passing tests on MIPS as well as x86 32/64bits. This is not
> complicated but getting the test environment for each architecture is time
> consumming. Stay tuned!

Well, the mips kernel does not implement substitution of syscall numbers,
so if your test passes on mips, it means that the test has failed to detect
that fault injection actually does not work there.

Anyway, do you have any interest in the subject yet?


-- 
ldv
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.strace.io/pipermail/strace-devel/attachments/20161115/0e80848b/attachment.bin>