Issue 1282: darcs get is way slow in 2.2 (at least in some setups) - Darcs bug tracker

Title	darcs get is way slow in 2.2 (at least in some setups)
Priority	bug	Status	given-up
Milestone		Resolved in
Superseder		Nosy List	darcs-devel, dmitry.kurochkin, kowey, mornfall, thorkilnaur
Assigned To	mornfall	Topics	Performance, Regression

Created on 2008-12-24.22:53:52 by mornfall, last changed 2010-07-07.14:28:47 by mornfall.

Messages
msg6877 (view)	Author: mornfall	Date: 2008-12-24.22:53:49
\|\| darcs-2.0.2 \| darcs-2.1 \| darcs-2.2+ get full \|\| 3.5s 2M \| 2.2s 2M \| 7.9s 3M !! get full \|\| 18.7s 11M \| 12.7s 10M \| 57.3s 9M !! get full \|\| 16.2s 19M \| 12.6s 18M \| 41.4s 19M !! These are three different medium to large repos. Something went wrong between 2.1 and current 2.2, since the get times are waay off. I suspect a cache- related (or to be more exact, sources-related) problem. Will investigate.
msg10480 (view)	Author: kowey	Date: 2010-03-23.23:42:01
Do we know more about what happened now?
msg11694 (view)	Author: mornfall	Date: 2010-07-07.14:28:45
Whatever this was, it got fixxored at some point. The current darcs instances outperform darcs-2.0.2 on full get easily (whatever that means... it could be also compilation flags or GHC version, I am not saying the code itself changed significantly). Anyway, go away issue.

History
Date	User	Action	Args
2008-12-24 22:53:52	mornfall	create
2009-08-06 15:18:20	kowey	set	topic: + Performance nosy: kowey, simon, thorkilnaur, dmitry.kurochkin, mornfall
2009-08-06 15:18:38	kowey	set	topic: + Regression nosy: kowey, simon, thorkilnaur, dmitry.kurochkin, mornfall
2009-08-15 08:53:19	kowey	set	status: unknown -> needs-reproduction nosy: kowey, simon, thorkilnaur, dmitry.kurochkin, mornfall
2009-08-25 17:36:53	admin	set	nosy: + darcs-devel, - simon
2009-08-27 14:24:19	admin	set	nosy: kowey, darcs-devel, thorkilnaur, dmitry.kurochkin, mornfall
2010-03-23 23:42:03	kowey	set	priority: urgent -> bug messages: + msg10480
2010-07-07 14:28:47	mornfall	set	status: needs-reproduction -> given-up messages: + msg11694