Re: please revert bogus patch to vmscan.c

Linus Torvalds (torvalds@transmeta.com)
Mon, 29 Oct 2001 15:44:23 -0800 (PST)


On Mon, 29 Oct 2001, Benjamin LaHaise wrote:
>
> Think of CPUs with tagged tlbs and lots of entries. Or even a system that
> only runs 1 threaded app. Easily triggerable. If people want to optimise
> it, great. But go for correctness first, please...

"easily triggerable"?

I doubt you'll find _any_ system where you can trigger it. Think about it:
in order to get the kind of VM pressure that causes page-outs, your TLB
pressure will be a lot higher than _any_ CPU I have ever heard about.

256 TLB entries is considered "a lot". Tagged or not, if the VM pressure
is so big that we need to swap, those 256 entries are a grain of sand in
sahara.

Linus

-
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/