Danek Duvall wrote:
On Thu, Mar 26, 2009 at 01:44:30PM -0500, Shawn Walker wrote:
I could re-work the structure slightly so that this information is only
recorded for entries that are "discovered" when checking installed packages
and are no longer in the publisher's catalog. That should limit the impact
to catalog entries that are for installed packages for a removed publisher.
I think that would help substantially, though obviously the code gets a bit
more complex that way. It'd be nice to know what the working set increase
is from gate bits to your current bits, and then to this.
The size of my catalog.pkl with gate bits is:
7,339,725 bytes (roughly 7.0M)
...with the first webrev's bits:
7,469,202 bytes (roughly 7.2M)
Which is about 1.76% larger (126K).
Changing pub to a dict from a tuple yields this improvement (I've tried
several variations of other data structures and all were much larger):
7,274,766 bytes (roughly 6.9M)
Which is about .885% *smaller* (63K) than the gate bits and 2.6% (190K)
smaller than my previous webrev :)
updated webrev:
http://cr.opensolaris.org/~swalker/pkg-6856-2/
diff from last webrev:
http://cr.opensolaris.org/~swalker/pkg-6856-2/v1-v2.patch
Change Summary:
* reduced size of catalog cache
* simplified publisher code in inventory
* clarified comments in filter_multimatch
* dropped filtering of preferred publisher multiple matches
* added test cases for bug 6654, since the fix for 6856 fixes the same issue
Cheers,
--
Shawn Walker
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
