comp.lang.ada
 help / color / mirror / Atom feed
* Re: Memory Access for Large Array
       [not found] <1000493387.173635@news.drenet.dnd.ca>
@ 2001-09-14 20:54 ` those who know me have no need of my name
  2001-09-15 18:01 ` Gerhard Häring
  1 sibling, 0 replies; 3+ messages in thread
From: those who know me have no need of my name @ 2001-09-14 20:54 UTC (permalink / raw)


<1000493387.173635@news.drenet.dnd.ca> divulged:

>I have a Sun Solaris with 1 Gbytes of RAM and lots of SAWP space. 

>With 1 Gbytes of RAM, I should be able to have an array size of up to
>16000 unknown. However, I was able to go up to 2000 only which uses
>only 16 Mbytes of memory. No matter what we do, we seem to be limited
>to 16 Mbytes of memory access. We tried the same test program with
>FORTRAN, and we did not have the same limit. 

what happens when you try exceeding 2000?  what message is printed, and by
what agent?

i'd guess that 16meg is the soft limit for some resource, and fortran
either does not use that resource or it's run-time requests an increase
(perhaps to the hard limit).  you should probably: a) also ask in a solaris
related forum, e.g., comp.unix.solaris, and b) truss each process (fortran
and ada) to see if and how resource limits are being manipulated.

-- 
okay, have a sig then



^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Memory Access for Large Array
       [not found] <1000493387.173635@news.drenet.dnd.ca>
  2001-09-14 20:54 ` Memory Access for Large Array those who know me have no need of my name
@ 2001-09-15 18:01 ` Gerhard Häring
  2001-09-18  3:50   ` those who know me have no need of my name
  1 sibling, 1 reply; 3+ messages in thread
From: Gerhard Häring @ 2001-09-15 18:01 UTC (permalink / raw)


On Fri, 14 Sep 2001 14:49:47 -0400, Sfoo <senglee.foo@dreo.dnd.ca> wrote:
>I have a Sun Solaris with 1 Gbytes of RAM and lots of SAWP space. I am =
>compiling my Ada program with GNU Ada95 compiler. I wrote a small test =
>program which define a large 2 dimensional array as follow:

I couldn't get around that limit either. I use Linux, and my "ulimit -s" shows
I have unlimited stack size. I somehow passing -stack <some-large-value> to the
GNU linker might do the trick, but I didn't suceed in tricking gnatlink to pass
this option to the linker and the linker still understanding it.

If anybody knows a solution (I've read rumours that GNAT 3.14 has similar
problems fixed), then I'll be glad to hear, too.

Gerhard
-- 
mail:   gerhard <at> bigfoot <dot> de       registered Linux user #64239
web:    http://www.cs.fhm.edu/~ifw00065/    public key at homepage
public key fingerprint: DEC1 1D02 5743 1159 CD20  A4B6 7B22 6575 86AB 43C0
reduce(lambda x,y:x+y,map(lambda x:chr(ord(x)^42),tuple('zS^BED\nX_FOY\x0b')))



^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Memory Access for Large Array
  2001-09-15 18:01 ` Gerhard Häring
@ 2001-09-18  3:50   ` those who know me have no need of my name
  0 siblings, 0 replies; 3+ messages in thread
From: those who know me have no need of my name @ 2001-09-18  3:50 UTC (permalink / raw)


<slrn9q75rl.gu.gerhard.nospam@lilith.hqd-internal> divulged:
>On Fri, 14 Sep 2001 14:49:47 -0400, Sfoo <senglee.foo@dreo.dnd.ca> wrote:

>>I have a Sun Solaris with 1 Gbytes of RAM and lots of SAWP space. I am =
>>compiling my Ada program with GNU Ada95 compiler. I wrote a small test =
>>program which define a large 2 dimensional array as follow:
>
>I couldn't get around that limit either. I use Linux, and my "ulimit -s" shows
>I have unlimited stack size. 

just because the limit is "unlimited" doesn't meant that there is no upper
bound, bizarre as that might sound to you.  (you don't have infinite memory
installed, right?)  the limits you can inspect and change are
administrative policy limits.  there remain platform related or imposed
upper bounds.  for better clues as to why linux acts as it does you should
ask in a linux forum, e.g., comp.os.linux.admin.

-- 
okay, have a sig then



^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2001-09-18  3:50 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
     [not found] <1000493387.173635@news.drenet.dnd.ca>
2001-09-14 20:54 ` Memory Access for Large Array those who know me have no need of my name
2001-09-15 18:01 ` Gerhard Häring
2001-09-18  3:50   ` those who know me have no need of my name

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox