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

[Andy Wingo]

Hey folks,

Thanks for the debugging and the patch, Mark!

On Wed 01 Jun 2011 01:18, Neil Jerram <address@hidden> writes:

> address@hidden (Ludovic Courtès) writes:
>
>> Mark H Weaver <address@hidden> skribas:
>>
>>> The main reason I haven't already pushed this patch is that there is a
>>> slight complication: when you register custom allocation functions for
>>> use by GMP, they get used for _all_ allocation, not just for digits.  In
>>> particular, they get used to allocate the block returned by mpz_get_str
>>> (when the first argument is NULL).  This means that if the user of
>>> mpz_get_str uses the standard "free" to deallocate this block, it will
>>> fail badly.  This could potentially affect programs which use libguile
>>> but also use the GMP functions directly.
>>
>> Yes, that's a problem, probably even be a showstopper for 2.0.  :-(
>>
>> What do others think?
>
> Provide an API to allow the trade-off to be decided by the application
> developer?

I think this only makes sense as an interim measure, like if we want to
change GMP to allocate only via GC_malloc in 2.2, then provide the
option in 2.0.  There has to be a better long-term solution though.

What if we get libgc to track total heap size (whether the user has
malloc, g_slice_alloc, or whatever), and start doing GC more frequently
if it sees the total heap size going up?

Andy
-- 
http://wingolog.org/