[PATCH 3 of 3] merge: delay writing the mergestate during applyupdates

[Matt Mackall]

On Mon, 2010-08-23 at 09:05 +0200, Peter Arrenbrecht wrote:
> On Thu, Aug 19, 2010 at 12:20 AM, Matt Mackall <mpm at selenic.com> wrote:
> > On Sun, 2010-07-11 at 11:49 +0900, Nicolas Dumazet wrote:
> >> On Fri, 18 Jun 2010 09:36:47 +0200
> >> Peter Arrenbrecht <peter.arrenbrecht at gmail.com> wrote:
> >>
> >> > # HG changeset patch
> >> > # User Peter Arrenbrecht <peter.arrenbrecht at gmail.com>
> >> > # Date 1276845982 -7200
> >> > merge: delay writing the mergestate during applyupdates
> >> >
> >> > This speeds up merges of lots of files considerably.
> >>
> >> So it seems that patch 1 and 2 of this serie were pushed, but not patch #3.
> >>
> >> Anything wrong in the idea of delaying writes?
> >
> > No, I just wasn't thrilled about the implementation. We should probably
> > just change things so that an explicit commit() is required, and add a
> > dirty flag.
> 
> Fine by me. Has this already been done or should I pick it up again?

Please have at it.

-- 
Mathematics is the supreme nostalgia of our time.