Re: [PATCH] dentry and inode cache hash algorithm performance changes.

["Jose R. Santos" <jrsantos@austin.ibm.com>]

On 04/30/04 15:18:32, Andrew Morton wrote:
> "Jose R. Santos" <jrsantos@austin.ibm.com> wrote:
> >
> > diff -Nru a/fs/inode.c b/fs/inode.c
> > --- a/fs/inode.c	Fri Apr 30 12:14:23 2004
> > +++ b/fs/inode.c	Fri Apr 30 12:14:23 2004
> > @@ -671,8 +671,9 @@
> >  
> >  static inline unsigned long hash(struct super_block *sb, unsigned long hashval)
> >  {
> > -	unsigned long tmp = hashval + ((unsigned long) sb / L1_CACHE_BYTES);
> > -	tmp = tmp + (tmp >> I_HASHBITS);
> > +	unsigned long tmp = (hashval + ((unsigned long) sb) ^ (GOLDEN_RATIO_PRIME + hashval)
> > +			/ L1_CACHE_BYTES);
> > +	tmp = tmp + ((tmp ^ GOLDEN_RATIO_PRIME) >> I_HASHBITS);
> >  	return tmp & I_HASHMASK;
> >  }
> >  
> 
> Are you sure about this?  It's doing:
> 
> 	tmp = hashval + sb ^ ((GOLDEN_RATIO_PRIME + hashval) / L1_CACHE_BYTES)
> 
> should it not be:
> 
> 	tmp = hashval + (sb ^ (GOLDEN_RATIO_PRIME + hashval)) / L1_CACHE_BYTES
> 
> ?

err... Wrote the patch to fast.  It should read

	tmp = (hashval * sb) ^ (GOLDEN_RATIO_PRIME + hashval) / L1_CACHE_BYTES

I screw up... I'll send a fixed patch in a while.

-JRS