From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from e28smtp09.in.ibm.com (e28smtp09.in.ibm.com [122.248.162.9]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by lists.ozlabs.org (Postfix) with ESMTPS id C31CC1A0895 for ; Thu, 2 Apr 2015 19:44:26 +1100 (AEDT) Received: from /spool/local by e28smtp09.in.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Thu, 2 Apr 2015 14:14:23 +0530 Received: from d28relay01.in.ibm.com (d28relay01.in.ibm.com [9.184.220.58]) by d28dlp03.in.ibm.com (Postfix) with ESMTP id 369BE1258044 for ; Thu, 2 Apr 2015 14:16:03 +0530 (IST) Received: from d28av01.in.ibm.com (d28av01.in.ibm.com [9.184.220.63]) by d28relay01.in.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP id t328iIQe51511376 for ; Thu, 2 Apr 2015 14:14:19 +0530 Received: from d28av01.in.ibm.com (localhost [127.0.0.1]) by d28av01.in.ibm.com (8.14.4/8.14.4/NCO v10.0 AVout) with ESMTP id t328iH7Q003611 for ; Thu, 2 Apr 2015 14:14:18 +0530 From: "Aneesh Kumar K.V" To: Michael Ellerman , Anton Blanchard Subject: Re: [PATCH] powerpc/mm: Add trace point for tracking hash pte fault In-Reply-To: <87y4og70nu.fsf@linux.vnet.ibm.com> References: <1421753750-17135-1-git-send-email-aneesh.kumar@linux.vnet.ibm.com> <1421809627.4900.1.camel@ellerman.id.au> <87d268v7xm.fsf@linux.vnet.ibm.com> <1422425467.11009.2.camel@ellerman.id.au> <87y4og70nu.fsf@linux.vnet.ibm.com> Date: Thu, 02 Apr 2015 14:14:18 +0530 Message-ID: <87ego2c42l.fsf@linux.vnet.ibm.com> MIME-Version: 1.0 Content-Type: text/plain Cc: linuxppc-dev@lists.ozlabs.org, paulus@samba.org List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , "Aneesh Kumar K.V" writes: > Michael Ellerman writes: > .... .... > With patch: > sys: 0m11.3258 > > ie, a -0.7% impact > > If that impact is high we could possibly put that tracepoint within #ifdef > CONFIG_DEBUG_VM ? Since the ebizzy runs results were not stable, I did a micro benchmark to measure this and I noticed that results observed are within the run variance of the test. I made sure we don't have context-switches between the runs. If I try to get large number of page-faults, we end up with context switches. for ex: We get without patch -------------------------------- root@qemu-pr-host trace-fault]# bash run Performance counter stats for './a.out 3000 300': 643 page-faults # 0.089 M/sec 7.236562 task-clock (msec) # 0.928 CPUs utilized 2,179,213 stalled-cycles-frontend # 0.00% frontend cycles idle 17,174,367 stalled-cycles-backend # 0.00% backend cycles idle 0 context-switches # 0.000 K/sec 0.007794658 seconds time elapsed [root@qemu-pr-host trace-fault]# And with-patch: --------------- [root@qemu-pr-host trace-fault]# bash run Performance counter stats for './a.out 3000 300': 643 page-faults # 0.089 M/sec 7.233746 task-clock (msec) # 0.921 CPUs utilized 0 context-switches # 0.000 K/sec 0.007854876 seconds time elapsed Performance counter stats for './a.out 3000 300': 643 page-faults # 0.087 M/sec 649 powerpc:hash_fault # 0.087 M/sec 7.430376 task-clock (msec) # 0.938 CPUs utilized 2,347,174 stalled-cycles-frontend # 0.00% frontend cycles idle 17,524,282 stalled-cycles-backend # 0.00% backend cycles idle 0 context-switches # 0.000 K/sec 0.007920284 seconds time elapsed [root@qemu-pr-host trace-fault]#