Re: VM tuning through fault trace gathering [with actual code]

John Fremlin (vii@users.sourceforge.net)
25 Jun 2001 22:15:31 +0100


Rik van Riel <riel@conectiva.com.br> writes:

> On 25 Jun 2001, John Fremlin wrote:
>
> > Last year I had the idea of tracing the memory accesses of the
> > system to improve the VM - the traces could be used to test
> > algorithms in userspace. The difficulty is of course making all
> > memory accesses fault without destroying system performance.
>
> Sounds like a cool idea. One thing you should keep in mind though
> is to gather traces of the WHOLE SYSTEM and not of individual
> applications.

In the current patch all pagefaults are recorded from all sources. I'd
like to be able to catch read(2) and write(2) (buffer cache stuff) as
well but I don't know how . . . .

> There has to be a way to balance the eviction of pages from
> applications against those of other applications.

Of course! It is important not to regard each thread group as an
independent entity IMHO (had a big old argument about this).

[...]

-- 

http://ape.n3.net - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/