From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from e28smtp03.in.ibm.com (e28smtp03.in.ibm.com [122.248.162.3]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by lists.ozlabs.org (Postfix) with ESMTPS id DAAD31A057D for ; Tue, 3 Feb 2015 03:22:21 +1100 (AEDT) Received: from /spool/local by e28smtp03.in.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Mon, 2 Feb 2015 21:52:19 +0530 Received: from d28relay05.in.ibm.com (d28relay05.in.ibm.com [9.184.220.62]) by d28dlp03.in.ibm.com (Postfix) with ESMTP id 5080C125805C for ; Mon, 2 Feb 2015 21:53:17 +0530 (IST) Received: from d28av02.in.ibm.com (d28av02.in.ibm.com [9.184.220.64]) by d28relay05.in.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP id t12GLvX453805124 for ; Mon, 2 Feb 2015 21:51:57 +0530 Received: from d28av02.in.ibm.com (localhost [127.0.0.1]) by d28av02.in.ibm.com (8.14.4/8.14.4/NCO v10.0 AVout) with ESMTP id t12GLvSK018116 for ; Mon, 2 Feb 2015 21:51:57 +0530 From: "Aneesh Kumar K.V" To: Anton Blanchard , Michael Ellerman Subject: Re: [PATCH] powerpc/mm: Add trace point for tracking hash pte fault In-Reply-To: <20150202212607.7585dedb@kryten> 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> <20150202212607.7585dedb@kryten> Date: Mon, 02 Feb 2015 21:51:55 +0530 Message-ID: <87vbjk707w.fsf@linux.vnet.ibm.com> MIME-Version: 1.0 Content-Type: text/plain Cc: paulus@samba.org, linuxppc-dev@lists.ozlabs.org List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Anton Blanchard writes: > Hi, > >> > ebizzy with -S 30 -t 1 -P gave >> > 13627 records/s -> Without patch >> > 13546 records/s -> With patch with tracepoint disabled >> >> OK. So that's about -0.6%. Are we happy with that? I'm not sure. >> >> Can you do a few more runs and see if that's a stable result. > > Surprisingly large. Is CONFIG_JUMP_LABEL enabled? That should reduce the > tracepoint to just a nop. yes. We do use jump label. I also verified that looking at .s #APP # 23 "./arch/powerpc/include/asm/jump_label.h" 1 1: nop .pushsection __jump_table, "aw" .llong 1b, .L201, __tracepoint_hash_fault+8 #, .popsection # 0 "" 2 ..... -aneesh