Re: [PATCH] Configure GMP to use GC allocation functions, remove bignum finalizers

[Andy Wingo]

Greets,

On Tue 29 Nov 2011 14:18, address@hidden (Ludovic Courtès) writes:

> Andy Wingo <address@hidden> skribis:
>
>> On Tue 29 Nov 2011 12:06, address@hidden (Ludovic Courtès) writes:
>>
>>> Could you run the stuff under gc-benchmarks/ with and without the
>>> heuristic, as in [0]?
>>>
>>> [0] http://thread.gmane.org/gmane.lisp.guile.devel/7803
>
> IIRC the “Guile” lines below correspond to whatever guile is in $PATH.
> Do you know what that is in your case?

I fixed the environment to make it so that it was something post-2.0.3,
but before these heuristics.

> The funny thing is that the initial FSD (which is what is shown here)
> has an impact, although one might think the initial value doesn’t matter
> much since adjust_gc_something changes it anyway.

The FSD set in the env. var is the minimum FSD, so it does matter.

> The other interesting thing is that it seems to be noticeably better,
> even on pure Scheme code–i.e., code that only does GC_malloc, not
> malloc.

It could be that having a better baseline memory usage affects the
specific benchmarks being run.  Dunno tho.  Growing the image size can
also happen due to GC allocations though, so perhaps increasing
allocation in those times does also make sense.

Cheers,

Andy
-- 
http://wingolog.org/