Re: [RFC] - Some notions that I would like comments on

Jamie Lokier (lkd@tantalophile.demon.co.uk)
Mon, 12 Jul 1999 17:48:36 +0200


Stefan Monnier wrote:
> > This involves finding parallelism in the kernel and getting an
> > otherwise idle processor to handle one of the parallel chunks of
> > work.
>
> Hopeless. The cost of inter-CPU communication is way too high to try
> to share such fine-grained tasks.

I reckon a spare CPU could happily

- pre-zero some pages, preferably spread over the range of page colours.
- move pages around to make space for multi-page allocations.
- checksum pages to find duplicates and merge them in the background

> > Compressive paging
> > Before writing a page out to swap, compress it.

> > IPC through page remapping
>
> Slow. Only worth it if the message is big enough (last time I heard, the
> cut-over point was around 8KB).

Uusually the overhead of TLB flushes or the need for inter-CPU TLB
flush messages is given as the reason.

But there are cases when those can be shown to be not necessary.
Like when you read a page, and the pte is not marked "accessed" yet.
No TLB flush required.

Similarly when you write a page and its marked "dirty" but not
"accessed" -- I'm not sure if this combination (written but not read) is
flagged by CPUs though.

-- Jamie

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/