Boost logo

Boost-Build :

From: Jürgen Hunold (hunold+lists.Boost_at_[hidden])
Date: 2004-02-01 13:56:11


Hi Volodya !

On Friday 30 January 2004 15:22, Vladimir Prus wrote:
> I've just comitted code which computes proper library order for gcc.
> The implementation is quite simple and supposedly inefficient, but on
> my project the performance did not suffer.

Well, today all hell broke loose again. It is "inefficient". ;-((
My problem:
Bjam startup time for RailSys build went up from 90 seconds to 22
minutes (!). That means building 6 full scale executables and some
tools from toplevel Jamfile.
startup time for my biggest single target exploded from 60 seconds to 5
minutes.

Question: How can I disable library ordering ?

Information: My "libraries" are quite "unordered" and have *lots* of
circular dependencies.
For other targets, with much better design (free of legacy), the impact
is much lower, but still measurable.

> I'd appreciate some testing of this new code. I attach the notes
> telling how it's supposed to work.

I can run more tests at work, providing more profiling data when needed.
Just drop a note ;-)

Yours,

Jürgen

-- 
* Dipl.-Math. Jürgen Hunold ! Institut für Verkehrswesen, Eisenbahnbau
* voice: ++49 511 762-2529 ! und -betrieb, Universität Hannover
* fax : ++49 511 762-3001 ! Appelstrasse 9a, D-30167 Hannover
* hunold_at_[hidden] ! www.ive.uni-hannover.de
 

Boost-Build list run by bdawes at acm.org, david.abrahams at rcn.com, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk