From: Brent Roman <brent@...>
Date: 2008-01-12T01:09:20+09:00
Subject: Re: 1.8.6 gc.c thoughts


Roger,

Can't help you with the HEAP_SLOTS issue,
but I did post about the strange equation for malloc_increase and
malloc_limit
a while back:

http://www.nabble.com/-PATCH--Limit-Ruby-1.8.6-process-size-to13713127.html#a13713127

I looked to me like an attempt to lessen the frequency of
GC mark/sweep passes.  It was entirely too successful at that :-)
Does anyone recall what could cause malloc_limit to decrease?

- brent



Roger Pack wrote:
> 
> 
> On another point, I have a question on this line of code, run at the end
> of
> garbage collection:
> if (malloc_increase > malloc_limit) {
>     malloc_limit += (malloc_increase - malloc_limit) * (double)live /
> (live
> + freed); // this line
>     if (malloc_limit < GC_MALLOC_LIMIT) malloc_limit = GC_MALLOC_LIMIT;
>  }
> 
> I haven't checked this, but it seems to me that It seems to me that
> (malloc_increase - malloc_limit) will always be a very small number (?)
> which may not be what was expected.  I could be wrong.  In reality I think
> what it should best do is
> malloc_limit = GC_MALLOC_LIMIT *
> percent_of_recent_allocated_memory_that_was_freed // allow the
> malloc_limit
> to change dynamically, maybe with a fixed maximum.  I could code it up.
> 
> Anyway just bouncing back some thoughts.
> Ruby rox.
> -- 
> -Roger Pack
> 
> 

-- 
View this message in context: http://www.nabble.com/1.8.6-gc.c-thoughts-tp13491288p14759517.html
Sent from the ruby-core mailing list archive at Nabble.com.