[Mono-dev] Status of Mono Garbage Collector?

[Atsushi Eno]

Paolo Molaro wrote:
> On 03/01/06 Hubert FONGARNAND wrote:
>> I'd want to know more about the statut of this new GC. I think, that
>> mono will be usable (on a desktop) (try to use beagle...) only if
>> there's a good precise GC... 
> 
> You're completely wrong. First, if there is a leak it is most likely not
> the GC's fault. Of the many excessive memory usage cases I debugged the
> issue was:
> *) most of the time a bug in some p/invoke binding (40%)
> *) very often the user or a library call made 100 allocations when just
> one was needed (40%)
> *) sometines a leak inside mono (10%)
> *) excessive memory retention by the GC (10%)

I think some guys dream too much performance boost from the
runtime improvements.

Even after having about half of a month for optimizing System.Xml
and 1.5x - 2x performance improvement (it is after several
optimization attempts done before by Ben, myself and others),
I still have/had very few things to blame on the runtime - even
enabling inline / copyprop / consprop / deadce helps only about
10% improvement, and I don't think it could improve performance
much more.

It might be nice if Hubert post actual code that uncovers slowness
that (you claim) caused by the GC, and then everyone would happen
to know which is bad (it's still useful if we could find the
performance problem inside the class libraries).

BTW If there are such people who can improve class library
performance including sys.xml better than now, they are
highly welcomed ;-)

Atsushi Eno