This is the mail archive of the guile@sources.redhat.com mailing list for the Guile project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: OK, what about some resolution (Re: GUILE's GC - why we struggling to solve already solved problems?)


bernard URBAN <Bernard.Urban@meteo.fr> writes:

> Sorry to be a bit provocative, but a lot of discussion on this list
> appears hair-splitting: why change a GC which works fine ?

Good question.  And I wonder why people are discussion this without
actually posting benchmark results.  Is it really so difficult to
compile and execute guile and bigloo with profiling enabled and then
post gc results?  

Note that even if you find that boehm's (gen.) gc is as fast as
guile's, it still doesn't make sense to me to use a malloc()
replacement instead of precise garbage collector in a scheme
*interpreter*.  Bigloo compiles Scheme to C code and in this special
case using boehm's gc is a natural choice.

The it's a problem that guile takes over main: Take a look at Maciej's
patch.


Jost

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]