nullfs stabilization I

Csaba Henk csaba.henk at creo.hu
Tue Jan 10 04:56:12 PST 2006


On 2006-01-09, Simon 'corecode' Schubert <corecode at xxxxxxxxxxxx> wrote:
> I know Matt wants a generic cache coherency layer, but until this is in 
> place, I propose attached patch to fix this 
> rename-race-and-file-vanishes bug in nullfs.  Furthermore, you can't 
> mount a nullfs from a nullfs as this will lead to a recursion stack 
> overflow double fault.  This has been fixed as well.
>
> I'm not sure if this single linked list is actually enough, because 
> renames on the lower layer won't trigger cache_inval's on the upper 
> layer at the moment.
>
> What do people think?

OK, I see now what's missing...

I have a vague idea of pushing all this stuff out to the VFS level, so
that OT1H we don't need to mess up nullfs OTOH the goodies you offer
will be accessible for other fs-es by cleverer mechanisms than
copy'n'paste.

(This might be close to the cache coherency layer as tought by Matt
and might as well be lightyears away from that -- I don't exactly know
what's in his mind and neither what's in mine.)

I don't promise anything but try to stay tuned :)

Csaba





More information about the Kernel mailing list