Re: Please don't beat me up (was Re: Bugs and wishes in memory management area)

Ingo Molnar (
Wed, 27 Nov 1996 21:41:28 +0100 (MET)

On 27 Nov 1996, Kevin Buhr wrote:

> It's cost? (1) Even if an emergency never occurs, every "kfree" will
> need another check, but depending on the actual code your new
> allocator uses, perhaps we'd be able to get this check done for free;
> (2) It might break nasty virtual/physical address dependencies deep
> inside some device driver somewhere despite the existence of a
> GFP_VIRTOK flag; (3) It's ugly.

with double indirection pointers we could implement some kind of garbage
collector kernel thread(s). Kernel code has to 'get' and 'put' such
resources, to avoid reentrancy problems.

The benefit would be soft relocatable buffers. [this one could be made
caching aware too, based on access pattern detection in the get/put
logic]. And this could coexist with no-MMU features like the 4MB pages or
the Sparc bigpages or the same Cyrix stuff?

the operations on these buffers would be: allocate(),get(),put(),free().
For once-only buffers it wont do no good, but for persistant buffers it is
an option?

-- mingo