>Hi folks,
>
>I needed to reimplement fs/inode.c due to some reasons I will explain later.
>Here are the main features of the new implementation:
>
> - dramatically faster
> - allows more concurrency
> - intended to be MP-safe (not tested)
I tried it. Needs some adjustment in kernel/ksyms.c to compile.
I can't tell if it is actually faster since the first thing I noticed
was that linux couldn't set execute permission on files anymore.
I did a complete recompile of the kernel and the only other kernel patch
I have in /linux/fs is the one for joliet isofs.
Alberto