Re: diffcore-rename performance mode

From: Junio C Hamano <gitster@pobox.com>
Date: Sep 18, 2007, 09:17 UTC
Message-ID: <7v8x74wcxe.fsf@gitster.siamese.dyndns.org>
In-Reply-To: <20070918090105.GA11854@coredump.intra.peff.net>

Jeff King <peff@peff.net> writes:

Show 14 quoted lines

> On Tue, Sep 18, 2007 at 01:58:17AM -0700, Junio C Hamano wrote:
>
>> > I thought we were holding counts of hashes, in which case there _is_ no
>> > overflow.
>> 
>> The raw hashval (the fingerprint recorded in struct spanhash) is
>> further reduced and used as an index into spahash_top.data[].
>> So more than one hashval can try to sit in the same slot in
>> spanhash_top.data[] array.
>
> Right, that's sort of what I was hinting at in the original message. Can
> we just make the hash table big enough to use the fingerprint hashes
> directly? It's going to use a bit more memory, but lookups should be
> very fast. I'll try to experiment and get some numbers.

Thanks -- I vaguely recall large hash was disastrous for me (trashed cache), but that was on a different hardware, different time.