From mboxrd@z Thu Jan 1 00:00:00 1970 Received: by wa-out-1112.google.com with SMTP id j37so1857321waf.22 for ; Wed, 03 Dec 2008 07:12:15 -0800 (PST) Message-ID: <2f11576a0812030712t1131c9d2x4dd0fd32eafa66ae@mail.gmail.com> Date: Thu, 4 Dec 2008 00:12:15 +0900 From: "KOSAKI Motohiro" Subject: Re: [PATCH] vmscan: improve reclaim throuput to bail out patch In-Reply-To: <49368DAF.9060206@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <49316CAF.2010006@redhat.com> <20081130150849.8140.KOSAKI.MOTOHIRO@jp.fujitsu.com> <20081203140419.1D44.KOSAKI.MOTOHIRO@jp.fujitsu.com> <49368DAF.9060206@redhat.com> Sender: owner-linux-mm@kvack.org Return-Path: To: Rik van Riel Cc: akpm@linux-foundation.org, linux-mm@kvack.org, linux-kernel@vger.kernel.org, mel@csn.ul.ie List-ID: >> I evaluate rvr bailout and skip-freeing patch in this week conteniously. >> I'd like to dump first output here. >> >> >> >> Rik, could you please review following? >> == >> vmscan bail out patch move nr_reclaimed variable to struct scan_control. >> Unfortunately, indirect access can easily happen cache miss. >> More unfortunately, Some architecture (e.g. ia64) don't access global >> variable so fast. > > That is amazing. Especially considering that the scan_control > is a local variable on the stack. Ahhhhh, I did want to write "indirect access(or likes global variables)", but my brain was sucked. sorry. I'll post description fixed version soon. thanks. >> if heavy memory pressure happend, that's ok. >> cache miss already plenty. it is not observable. >> >> but, if memory pressure is lite, performance degression is obserbable. > >> about 4-5% degression. >> >> Then, this patch introduce temporal local variable. > >> OK. the degression is disappeared. > > I can't argue with the numbers, though :) > > Maybe all the scanning we do ends up evicting the cache lines > with the scan_control struct in it from the fast part of the > CPU cache? Yeah, I think so. > >> Signed-off-by: KOSAKI Motohiro > > Acked-by: Rik van Riel -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org