Kernel related (?) user space crash at ARM11 MPCore

[Catalin Marinas]

On Wed, 2009-09-23 at 13:51 +0100, Catalin Marinas wrote:
> On Wed, 2009-09-23 at 13:42 +0100, Russell King - ARM Linux wrote:
> > On Wed, Sep 23, 2009 at 02:12:23PM +0200, Mikael Pettersson wrote:
> > > OABI is obsolete legacy, so while it should (perhaps) be
> > > kept working, it should not be allowed to negatively affect
> > > the performance of pure EABI systems.
> > > 
> > > If option 3 is best for EABI then that's my vote.
> > 
> > However, Catalins benchmarks don't include testing my patch with EABI
> > executables with RIX disabled, so we really don't know how it would
> > behave.
> 
> I'll do it shortly, no problem. But since it didn't take VM_EXEC into
> account, I thought it wouldn't really matter. But I see where you come
> from - just for comparison with my workaround.

Forgot to mention - I don't think lmbench is too relevant to CoW
performance since the applications are relatively small. But that's the
problem with any benchmark application which was designed to be, well, a
benchmark application. We would need real-world applications but I don't
have time for further investigation here.

-- 
Catalin