Re: StGit: kha/{stable,safe,experimental} updated

From: Karl Hasselström <kha@treskal.com>
Date: Aug 6, 2008, 10:18 UTC
Message-ID: <20080806101813.GC18336@diana.vm.bytemark.co.uk>
In-Reply-To: <b0943d9e0808051433i667f6a05iff1a5220e37602e7@mail.gmail.com>

On 2008-08-05 22:33:57 +0100, Catalin Marinas wrote:

Show 8 quoted lines

> 2008/8/1 Karl Hasselström <kha@treskal.com>:
>
> > Heh. It's _always_ your main worry. But rightly so, since mistakes
> > could be costly. Thanks for spending time on this with me.
>
> I'm usually worried about performance but will give it a try with a
> Linux kernel and real patches. Have you done any tests to compare it
> with my master branch?

Yes. I posted the benchmark results a while back -- as I recall, the performance was about the same for large uncommit and rebase operations.

Also, I'm currently investigating some possible optimizations. Will post new bechmarks afterwards, if I get something working.

> Would we even need to prune the stack history? It might get pretty
> large after about 1-2 years of usage.

Yes, it's growing without bound, so eventually one would have to prune it.

-- 
Karl Hasselström, kha@treskal.com
      www.treskal.com/kalle