[PATCH] Accurate count of freed cells

[PATCH] Accurate count of freed cells

[Ludovic Courtès]

Hi,

The patch below fixes the way freed cells are counted in
`scm_i_sweep_card ()'.  Basically, without this patch, FREE_COUNT is
incremented regardless of whether the cell pointed to by SCMPTR was
already free or not.

Thanks,
Ludovic.


2005-12-20  Ludovic Courtès  <ludovic.courtes@laas.fr>

	* gc-card.c (scm_i_sweep_card): Only increment FREE_COUNT for
	cells that were really freed.


--- orig/libguile/gc-card.c
+++ mod/libguile/gc-card.c
@@ -255,10 +255,13 @@
 	  abort();
 	}
 
+      if (SCM_CELL_TYPE (scmptr) != scm_tc_free_cell)
+	/* Only account for cells that were really freed.  */
+	free_count ++;
+
       SCM_GC_SET_CELL_WORD (scmptr, 0, scm_tc_free_cell);	  
       SCM_SET_FREE_CELL_CDR (scmptr, PTR2SCM (*free_list));
       *free_list = scmptr;
-      free_count ++;
     }
 
   return free_count;


_______________________________________________
Guile-devel mailing list
Guile-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/guile-devel

Re: [PATCH] Accurate count of freed cells

[Han-Wen Nienhuys]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 557 bytes --]

In article <87y82fsybo.fsf@laas.fr>,
Ludovic Courtès <ludovic.courtes@laas.fr> wrote:
>Hi,
>
>The patch below fixes the way freed cells are counted in
>`scm_i_sweep_card ()'.  Basically, without this patch, FREE_COUNT is
>incremented regardless of whether the cell pointed to by SCMPTR was
>already free or not.

isn't it tighter (one compare less) to put a

  free_count --;

in the scm_tc_free_cell case? 





_______________________________________________
Guile-devel mailing list
Guile-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/guile-devel

Re: [PATCH] Accurate count of freed cells

[Ludovic Courtès]

hanwen@byrd.xs4all.nl (Han-Wen Nienhuys) writes:

> isn't it tighter (one compare less) to put a
>
>   free_count --;
>
> in the scm_tc_free_cell case? 

Maybe.  But is it such a big deal?  ;-)

Thanks,
Ludovic.


_______________________________________________
Guile-devel mailing list
Guile-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/guile-devel

Re: [PATCH] Accurate count of freed cells

[Ludovic Courtès]

Hi,

ludovic.courtes@laas.fr (Ludovic Courtès) writes:

> 2005-12-20  Ludovic Courtès  <ludovic.courtes@laas.fr>
>
> 	* gc-card.c (scm_i_sweep_card): Only increment FREE_COUNT for
> 	cells that were really freed.

Can someone commit this and/or comment on it?

Thanks,
Ludovic.


_______________________________________________
Guile-devel mailing list
Guile-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/guile-devel

Re: [PATCH] Accurate count of freed cells

[Han-Wen Nienhuys]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 516 bytes --]

In article <87y81eszea.fsf@laas.fr>,
Ludovic Courtès <ludovic.courtes@laas.fr> wrote:
>Hi,
>
>ludovic.courtes@laas.fr (Ludovic Courtès) writes:
>
>> 2005-12-20  Ludovic Courtès  <ludovic.courtes@laas.fr>
>>
>> 	* gc-card.c (scm_i_sweep_card): Only increment FREE_COUNT for
>> 	cells that were really freed.
>
>Can someone commit this and/or comment on it?
>

See CVS.




_______________________________________________
Guile-devel mailing list
Guile-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/guile-devel

Re: [PATCH] Accurate count of freed cells

[Ludovic Courtès]

hanwen@byrd.xs4all.nl (Han-Wen Nienhuys) writes:

> See CVS.

Thanks!

Ludovic.


_______________________________________________
Guile-devel mailing list
Guile-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/guile-devel