From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Fri, 27 Sep 2002 21:36:55 -0700 From: William Lee Irwin III Subject: Re: 2.5.38-mm3 Message-ID: <20020928043655.GU3530@holomorphy.com> References: <20020927152833.D25021@in.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Description: brief message Content-Disposition: inline In-Reply-To: Sender: owner-linux-mm@kvack.org Return-Path: To: Zwane Mwaikambo Cc: Dipankar Sarma , Andrew Morton , lkml , "linux-mm@kvack.org" List-ID: > On Fri, 27 Sep 2002, Dipankar Sarma wrote: >> The counts are off by one. >> With a UP kernel, I see that fget() cost is negligible. >> So it is most likely the atomic operations for rwlock acquisition/release >> in fget() that is adding to its cost. Unless of course my sampling >> is too less. On Sat, Sep 28, 2002 at 12:35:30AM -0400, Zwane Mwaikambo wrote: > Mine is a UP box not an SMP kernel, although preempt is enabled; > 0xc013d370 : push %ebx > 0xc013d371 : mov %eax,%ecx > 0xc013d373 : mov $0xffffe000,%edx > 0xc013d378 : and %esp,%edx > 0xc013d37a : incl 0x4(%edx) Do you have instruction-level profiles to show where the cost is on UP? Thanks, Bill -- 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/