Re: [Monotone-devel] Re: times to load various things from the database

[Derek Scherger]

On Sun, Jan 18, 2009 at 4:35 AM, Nathaniel Smith <address@hidden> wrote:

On Sun, Jan 18, 2009 at 3:14 AM, Thomas Moschny <address@hidden> wrote:
> Derek Scherger wrote:
>> - after a reconstruction path is loaded it is scanned in forward order
>> looking for a cached roster to start from. if one is found, the
>> reconstruction path is truncated at that point and reconstruction will
>> start from the closest cached roster.
>
> Somehow I always thought the code that searches for the reconstruction
> patch would do this already: not only stop at base rosters but also at
> cached rosters - seems I was wrong.

Err, yeah, it definitely should be doing that!  And if it isn't, it
should be fixed to do that, rather than papering it over like Derek
described...

Looking at this a bit further the reconstruction path code calls is_base on the roster_reconstruction_graph which does check to see if the cache currently contains a roster to start from, so trimming the reconstruction path is probably a no-op.

I have run timings on a few different things, the testsuite, db check and full pulls, with and without this change and  it doesn't seem to make a lot of difference in general.  I suppose db regenerate_caches might be another interesting test case although that's probably pretty similar to the client side pull.

Cheers,
Derek