[lkml]   [2001]   [Dec]   [7]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: 2GB process crashing on 2.4.14
On Fri, Dec 07, 2001 at 01:48:00PM +0000, Alan Cox wrote:
> > > Most probably the process is running out of address space to allocate from.
> > > There is 3Gb of available space.
> >
> > That would be from 0x00000000 to 0xC0000000, Right?
> Correct (0xBFFFFFFF)
> > > binary, some your libraries. Getting above 3Gb/process on x86 is very hairy
> > > with a bad performance hit
> >
> > So if I was hitting this limit then I should see no / very few gaps, in the
> > /proc/<pid>/maps. Is that true?
> Providing the memory allocator it is using is sufficiently smart

Where "it" is the app?

OK, well looking at the maps output, there seems to be three distinct

1) from 0x00000000 to 0x01c6a000 (30MB-ish) are mappings of the executable.

2) from 0xbca9a000 to 0xbfffffff (56MB-ish) are the libs, plus a few other
areas, which I've assumed are stack, and scratch areas for the libs.

3) a single mapping, (was 1.1GB-ish in the map output I attached) which
starts at the end of section 1, and is continually growing, and which I
can see has no reason to stop until it gets to the start of section 2
(some 3GB - 86MB later).

Now admittedly, it's possible that some of the other mappings may grow by a
factor of 20 to suddenly eat up 1GB of address space, but I doubt it. So I'm
not buying the address space idea at the moment. That said, I'm not going to
discount it and will keep a log of what happens on the mappings while this
process is running, just in case something really wacky like that happens.

Paul Sargent
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
More majordomo info at
Please read the FAQ at

 \ /
  Last update: 2005-03-22 13:13    [W:0.049 / U:10.088 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site