public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* Large process & 2.4.X
@ 2003-05-28 14:14 jlnance
  0 siblings, 0 replies; only message in thread
From: jlnance @ 2003-05-28 14:14 UTC (permalink / raw)
  To: linux-kernel

Hello All,
    I have found a set of instructions for compiling a kernel which
allows processes to access about an extra 500M of memory.  I need
this functionality since I run large processes which sometimes die
because they can not allocate more then 3G of memory.
    Before I go too far down this road, I want to bounce these
instructions off this list to make sure this is the best way to
get what I am after and also to see what the downsides are.

    What I am susposed to do is very simple:

1) Modify page.h such that __PAGE_OFFSET is changed from 0xC0000000 to
   0xE0000000.

2) Modify processor.h such that TASK_UNMAPPED_BASE is changed from
   (TASK_SIZE/3) to (TASK_SIZE/16)

3) Modify vmlinux.lds similar to step 1

Then the kernel is built with CONFIG_HIGHMEM4G=y (the machines have 4G of
ram).

The part that has me concerned is that I am not sure why I need to change
__PAGE_OFFSET if CONFIG_HIGHMEM4G is on.  I am not sure how these options
interact with each other.  Also, what am I loosing by making __PAGE_OFFSET
this large?  I believe that TASK_UNMAPPED_BASE is just moving the stack
around so that it does not get in the way of malloc().  Is this correct?

Thanks,

Jim

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2003-05-28 14:12 UTC | newest]

Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2003-05-28 14:14 Large process & 2.4.X jlnance

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