From owner-freebsd-bugs@FreeBSD.ORG Fri Mar 20 01:04:21 2015 Return-Path: Delivered-To: freebsd-bugs@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [8.8.178.115]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id 2DCA54AE for ; Fri, 20 Mar 2015 01:04:21 +0000 (UTC) Received: from kenobi.freebsd.org (kenobi.freebsd.org [IPv6:2001:1900:2254:206a::16:76]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 0F3EF665 for ; Fri, 20 Mar 2015 01:04:21 +0000 (UTC) Received: from bugs.freebsd.org ([127.0.1.118]) by kenobi.freebsd.org (8.14.9/8.14.9) with ESMTP id t2K14KEv015934 for ; Fri, 20 Mar 2015 01:04:20 GMT (envelope-from bugzilla-noreply@freebsd.org) From: bugzilla-noreply@freebsd.org To: freebsd-bugs@FreeBSD.org Subject: [Bug 198723] [hwpmc] pmcstat -p (process counting) gets corrupted even if pinned to a CPU Date: Fri, 20 Mar 2015 01:04:21 +0000 X-Bugzilla-Reason: AssignedTo X-Bugzilla-Type: new X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: Base System X-Bugzilla-Component: kern X-Bugzilla-Version: 11.0-CURRENT X-Bugzilla-Keywords: X-Bugzilla-Severity: Affects Only Me X-Bugzilla-Who: adrian@freebsd.org X-Bugzilla-Status: New X-Bugzilla-Priority: --- X-Bugzilla-Assigned-To: freebsd-bugs@FreeBSD.org X-Bugzilla-Target-Milestone: --- X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: bug_id short_desc product version rep_platform op_sys bug_status bug_severity priority component assigned_to reporter Message-ID: Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit X-Bugzilla-URL: https://bugs.freebsd.org/bugzilla/ Auto-Submitted: auto-generated MIME-Version: 1.0 X-BeenThere: freebsd-bugs@freebsd.org X-Mailman-Version: 2.1.18-1 Precedence: list List-Id: Bug reports List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 20 Mar 2015 01:04:21 -0000 https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=198723 Bug ID: 198723 Summary: [hwpmc] pmcstat -p (process counting) gets corrupted even if pinned to a CPU Product: Base System Version: 11.0-CURRENT Hardware: Any OS: Any Status: New Severity: Affects Only Me Priority: --- Component: kern Assignee: freebsd-bugs@FreeBSD.org Reporter: adrian@freebsd.org Here's a fun one: cpuset -l 2 pmcstat -w 1 -p CPU_CLK_UNHALTED_CORE -p PARTIAL_RAT_STALLS.SLOW_LEA_WINDOW -p PARTIAL_RAT_STALLS.MUL_SINGLE_UOP ./himenobmtxpa M 384 mimax = 128 mjmax = 128 mkmax = 256 imax = 127 jmax = 127 kmax =255 Start rehearsal measurement process. Measure the performance in 3 times. MFLOPS: 1487.222542 time(s): 0.271122 1.733593e-03 Now, start the actual measurement process. The loop will be excuted in 663 times This will take about one minute. Wait for a while # p/CPU_CLK_UNHALTED_CORE p/PARTIAL_RAT_STALLS.SLOW_LEA_WINDOW p/PARTIAL_RAT_STALLS.MUL_SINGLE_UOP 2391336164 2808870 468916620 2391116254 2841466 580442618 2391286584 2855097 580087732 2391228957 2900509 578770832 2391232010 2840303 281471262184436 2391175692 2841502 580400260 2391261851 2844140 580443655 2391267208 2845348 18446462599313378116 2391258793 2844923 580426258 2391283394 2844559 580358142 2391251898 2844617 580418628 2391287371 2844894 281471262213505 2391220843 2844253 580344542 2391210632 2844542 580359758 2391286696 2844145 18446462599313346923 2391222766 2843902 580428539 2391301558 2845050 580449769 2391220296 2845291 580429033 2391229903 2843954 281471262221908 2391151757 2844754 580463504 2391264515 2844853 580580231 2390984664 2846816 580756378 2391363113 2848440 18446462599314405744 2391327705 2844575 580393026 2391327491 2845494 580487580 2391338337 2844307 580304875 2391360320 2881127 281471261156617 2391356197 2880822 579204685 2391272958 2845046 580382413 2391316177 2843457 18446462599313098960 2391336593 2845009 580494044 2391314220 2844990 580520104 2391321150 2844087 580434830 2391344912 2843830 281471262041250 2391335165 2844355 580422841 ^Croot@bruce:/home/adrian # Notice how some of these event counters are huge? they shouldn't be; they're cycle counters. -- You are receiving this mail because: You are the assignee for the bug.