Messages in this thread | | | Date | Thu, 19 Jan 2006 15:00:39 +0100 | From | Nick Piggin <> | Subject | Re: [patch 0/4] mm: de-skew page refcount |
| |
On Wed, Jan 18, 2006 at 11:27:13AM -0800, Linus Torvalds wrote: > > > On Wed, 18 Jan 2006, Nick Piggin wrote: > > > > > So I disagree with this patch series. It has real downsides. There's a > > > reason we have the offset. > > > > Yes, there is a reason, I detailed it in the changelog and got rid of it. > > And I'm not applying it. I'd be crazy to replace good code by code that is > objectively _worse_. >
And you're not? Damn.
> The fact that you _document_ that it's worse doesn't make it any better. > > The places that you improve (in the other patches) seem to have nothing at > all to do with the counter skew issue, so I don't see the point. >
You know, I believe you're right. I needed the de-skewing patch for something unrelated and it seemed that it opened the possibility for the following optimisations (ie. because we no longer touch a page after its refcount goes to zero).
But actually it doesn't matter that we might touch page_count, only that we not clear PageLRU. So the enabler is simply moving the TestClearPageLRU after the get_page_testone.
So I'll respin the patches without the de-skewing and the series will become much smaller and neater.
> So let me repeat: WHY DID YOU MAKE THE CODE WORSE? >
You've never bothered me about that until now...
Thanks for the feedback!
Nick - 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/
| |