On Fri, 2005-06-03 at 09:22 +0100, Simon Marlow wrote:
> On 02 June 2005 16:48, Duncan Coutts wrote:
> 
> > I have a program which runs to completion on windows in about 90Mb of
> > heap space. Ie the following works ok:
> > 
> > prog.exe +RTS -M90m -RTS
> > 
> > However if I use the -H option to reserve a bigger initial heap size:
> > 
> > prog.exe +RTS -H60m -M90m -RTS
> > 
> > then the program fails part way through with heap exhaustion. This
> > seems odd given what the -H and -M options are supposed to do.
> > 
> > What particularly makes me think this is a bug is that the following
> > also fails:
> > 
> > prog.exe +RTS -H730m -M750m -RTS
> > 
> > and it fails ever sooner if I use:
> > 
> > prog.exe +RTS -H750m -M750m -RTS
> 
> Hi Duncan - could you send us output from the above run with +RTS
> -Sstderr?

So for that last run "prog.exe +RTS -H750m -M750m -RTS" we get

E:\dcoutts\MSYS1.0\home\dcoutts\build\gtk2hs-0.9.7.91\tools\c2hs\c2hsLocal.exe
-C-Ie:/dcoutts/GTK/include/glib-2.0 -C-Ie:/dcoutts/GTK/lib/glib-2.0/include
-C-Ie:/dcoutts/GTK/include/gtk-2.0 -C-Ie:/dcoutts/GTK/lib/gtk-2.0/include
-C-Ie:/dcoutts/GTK/include/pango-1.0 -C-Ie:/dcoutts/GTK/include/atk-1.0
--cppopts=-include "gtk2hs-config.h" --precomp=gtk/gtk.precomp gtk/gtk.h +RTS
-H750m -M750m -Sstderr 
    Alloc    Collect    Live    GC    GC     TOT     TOT  Page Flts
    bytes     bytes     bytes  user  elap    user    elap
   267224    262144     20008  0.01  0.03    0.01    0.03    0    0   (Gen:  1)
576910260 576921600  32643288  0.73  0.75    4.13    4.75    0    0   (Gen:  0)
170915076 684748800  34673400  0.69  0.78    5.89    6.64    0    0   (Gen:  0)
 67135828 691081216  35737220  0.59  0.59    7.25    8.00    0    0   (Gen:  1)
Heap exhausted;
Current maximum heap size is 749998080 bytes (715 Mb);
use `+RTS -M<size>' to increase it.

Let me know if you want the same output from other runs or any other output.

Duncan

_______________________________________________
Glasgow-haskell-bugs mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs

Reply via email to