Re: process creation time increases linearly with shmem

[Rik van Riel <riel@redhat.com>]

On Fri, 26 Aug 2005, Ray Fucillo wrote:

> However, there is still a need that the child, once successfully forked, is
> operational reasonably quickly.  I suspect that Ross's idea of paging in
> everything after the first fault would not be optimal for us, because we'd
> still be talking about hundreds of ms of work done before the child does
> anything useful. 

Simply skipping the page table setup of MAP_SHARED regions
should be enough to fix this issue.

> It would still be far better than the behavior we have today because 
> that time would no longer be synchronous with the fork().

Filling in all the page table entries at the first fault to
a VMA doesn't make much sense, IMHO.

The reason I think this is that people have experimented
with prefaulting already resident pages at page fault time,
and those experiments have never shown a conclusive benefit.

Now, if doing such prefaulting for normal processes does not
show a benefit - why would it be beneficial to recently forked
processes with a huge SHM area ?

I suspect we would be better off without that extra complexity,
unless there is a demonstrated benefit to it.

-- 
All Rights Reversed
-
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/