A test malloc program makes 64-Bit FC-4 dying

[Reuben D. Budiardja]

On Thursday 03 November 2005 18:22, Dave Mitchell wrote:
> On Thu, Nov 03, 2005 at 04:51:17PM -0500, Reuben D. Budiardja wrote:
> > That's probably too, but my question is: why does this program quickly
> > terminate on other platform (RHEL 4/3, FC-4 32-bit) without grinding
> > those machines to an almost halt, but not on FC-4 64-bit ? All machines
> > have 1GB RAM.
>
> It's not how much RAM, but how much swap disk they have that probably
> matters.  It will fill the available RAM nearly instantly, then grind away
> till the swap disk is full too, and then die.

Well, that's the thing. All machines have about the same swap (1GB), but on 
other platform I got the "Memory exhaustion complete" message from the 
program pretty quickly without the machine grind away and using swap.

> > Secondly, why cant I limit the memory usage using ulimit (it seems to
> > have no effect) ?
>
> You probably need to be using ulimit -d rather than -m

That doesn't work.

Thanks.
RDB