Re: Hashing and directories

[torvalds@transmeta.com (Linus Torvalds)]

In article <20010307013729.A7184@pcep-jamie.cern.ch>,
Jamie Lokier  <lk@tantalophile.demon.co.uk> wrote:
>Pavel Machek wrote:
>> > the space allowed for arguments is not a userland issue, it is a kernel
>> > limit defined by MAX_ARG_PAGES in binfmts.h, so one could tweak it if one
>> > wanted to without breaking any userland.
>> 
>> Which is exactly what I done on my system. 2MB for command line is
>> very nice.
>
>Mine too (until recently).  A proper patch to remove the limit, and copy
>the args into swappable memory as they go (to avoid DoS) would be nicer,
>but a quick change to MAX_ARG_PAGES seemed so much easier :-)

You can't just change MAX_ARG_PAGES. The space for the array is
allocated on the stack, and you'll just overflow the stack if you start
upping it a lot.

The long-term solution for this is to create the new VM space for the
new process early, and add it to the list of mm_struct's that the
swapper knows about, and then just get rid of the pages[MAX_ARG_PAGES]
array completely and instead just populate the new VM directly.  That
way the destination is swappable etc, and you can also remove the
"put_dirty_page()" loop later on, as the pages will already be in their
right places. 

It's definitely not a one-liner, but if somebody really feels strongly
about this, then I can tell already that the above is the only way to do
it sanely.

		Linus