In article <199606131617.SAA28682@i17linuxb.ists.pwr.wroc.pl>, Marek
Michalkiewicz <marekm@i17linuxb.ists.pwr.wroc.pl> writes:
> Thomas Koenig:
>> Now that Linux 2.0 is out, it's time for another round of Request For
>> Features.
> Good. A few more ideas:
> * Secure TCP sequence numbers (the random driver has some code
> for that which doesn't seem to be used anywhere at the moment)\
Look again. This is implemented in 2.0. See linux/drivers/char/random.c,
function secure_tcp_sequence_number().
> * Deal with out of virtual memory situations more gracefully -
> at least, reserve a few last VM pages to root; maybe add VM
> per user (not per process) limits? Putting processes which
> fail to allocate memory to sleep might still be better than
> killing them as long as there is a way to prevent deadlocks.
Killing is better if we get to the point where we cannot allocate
memory, but we might want to implement complete process swapping as a
way to avoid getting into that state. Per-user resource limits would
be good.
> * Allocate ftape DMA buffers (96KB!) on first open (free on last
> close) instead of during driver initialization; needs fixing
> the memory allocation code first so that it doesn't cause so
> much fragmentation (and failing to allocate large contiguous
> areas even though there are still lots of free pages)
Impossible. DMA *needs* contiguous memory, and how you allocate
memory is largely irrelevant --- you simply don't have control over
how the application will access that memory, so you can't control
which order pages get freed int. Any method which can guarantee large
contiguous requests is going waste as much memory as the DMA buffers
would be occupying themselves.
> * Swap files over NFS (for diskless workstations)
Yes.
Cheers,
Stephen.
--- Stephen Tweedie <sct@dcs.ed.ac.uk> Department of Computer Science, Edinburgh University, Scotland.