[Gnu-arch-users] Re: [PATCH] arch speedups on big trees

[Chris Mason]

On Wed, 2004-01-07 at 01:09, Miles Bader wrote:
> Charles Duffy <address@hidden> writes:
> > Depends on where you are. On AFS, for instance, stat() calls aren't
> > cached on the client, so they're potentially much slower (!) than
> > read()s; for that reason, checking the inode signatures is exceedingly
> > slow there, and so this optimization will be a very substantial gain.
> 
> Does AFS not special-case stats of files that are in the cache?!?
> 
> I've been using tla in NFS, btw, and one huge horrible lose there is that
> computing a changeset seems to do, for every file, `diff a b > ,tmp',
> where ",tmp" is _in the current tree_.  Creating a temporary file then
> deleting it is ... slow.
> 
That part is actually easy.  arch_make_changeset creates the tmpfile,
but could instead push that responsibility to arch_invoke_diff.  It
would only be created/unlinked if the diff func reported changes.

-chris