NetBSD Problem Report #53308

From gson@gson.org  Wed May 23 13:05:53 2018
Return-Path: <gson@gson.org>
Received: from mail.netbsd.org (mail.netbsd.org [199.233.217.200])
	(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
	(Client CN "mail.NetBSD.org", Issuer "mail.NetBSD.org CA" (not verified))
	by mollari.NetBSD.org (Postfix) with ESMTPS id EEBBB7A1C8
	for <gnats-bugs@gnats.NetBSD.org>; Wed, 23 May 2018 13:05:53 +0000 (UTC)
Message-Id: <20180523130548.1B82698B6A5@guava.gson.org>
Date: Wed, 23 May 2018 16:05:48 +0300 (EEST)
From: gson@gson.org (Andreas Gustafsson)
Reply-To: gson@gson.org (Andreas Gustafsson)
To: gnats-bugs@NetBSD.org
Subject: evbarm-earmv7hf performance regression under qemu
X-Send-Pr-Version: 3.95

>Number:         53308
>Category:       port-evbarm
>Synopsis:       evbarm-earmv7hf performance regression under qemu
>Confidential:   no
>Severity:       serious
>Priority:       high
>Responsible:    jmcneill
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Wed May 23 13:10:01 +0000 2018
>Closed-Date:    Tue Sep 18 05:16:42 +0000 2018
>Last-Modified:  Tue Sep 18 05:16:42 +0000 2018
>Originator:     Andreas Gustafsson
>Release:        NetBSD-current, source date >= 2018.05.14.17.15.54
>Organization:
>Environment:
System: NetBSD
Architecture: arm
Machine: evbarm
>Description:

The evbarm-earmv7hf ATF tests on the TNF testbed have typically
completed in about 24 hours (unless they hit PR 53303), but their
performance recently worsened such that they no longer complete within
the testbed timeout of 40 hours.

The problem started with the following commits:

  2018.05.14.17.15.54 joerg src/sys/arch/aarch64/include/armreg.h 1.10
  2018.05.14.17.15.54 joerg src/sys/arch/arm/cortex/gtmr.c 1.27
  2018.05.14.17.15.54 joerg src/sys/arch/arm/include/armreg.h 1.121

Perhaps qemu is slow enough that each gtmr read returns a different
value?

>How-To-Repeat:

Inspect the test logs at

  http://releng.netbsd.org/b5reports/evbarm-earmv7hf/commits-2018.05.html#2018.05.14.17.15.54

>Fix:

-- 
Andreas Gustafsson, gson@gson.org

>Release-Note:

>Audit-Trail:

Responsible-Changed-From-To: port-evbarm-maintainer->joerg
Responsible-Changed-By: gson@NetBSD.org
Responsible-Changed-When: Wed, 23 May 2018 13:11:32 +0000
Responsible-Changed-Why:
Over to committer


From: Andreas Gustafsson <gson@NetBSD.org>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: port-evbarm/53308 (evbarm-earmv7hf performance regression under qemu)
Date: Tue, 29 May 2018 10:23:55 +0300

 Joerg wrote on port-arm:
 > You can check your system with "vmstat -e | grep gtmr":
 > 
 > armgtmr0 CNTV_TVAL write retry max       8    0 misc
 > armgtmr0 CNTV_CVAL read retry max        1    0 misc
 > armgtmr0 CNTVCT read retry max           5    0 misc
 > armgtmr0 CNTPCT read retry max           6    0 misc

 armv7# vmstat -e | grep gtmr
 armgtmr0 CNTV_TVAL write retry max     201    0 misc
 armgtmr0 CNTV_CVAL read retry max        1    0 misc
 armgtmr0 CNTVCT read retry max         200    0 misc
 armgtmr0 CNTPCT read retry max         200    0 misc

 This is from qemu 2.11.1 with the following command line:

 qemu-system-arm -m \
     128 \
     -drive \
     file=/bracket/evbarm-earmv7hf/test/2018.05.27.17.16.39/anita/wd0.img,format=raw,media=disk,snapshot=on,if=sd \
     -nographic \
     -drive \
     file=/bracket/evbarm-earmv7hf/test/2018.05.27.17.16.39/anita/tests-results.img,format=raw,media=disk,snapshot=off \
     -tb-size \
     128 \
     -M \
     vexpress-a15 \
     -kernel \
     /bracket/evbarm-earmv7hf/test/2018.05.27.17.16.39/anita/netbsd-VEXPRESS_A15.ub \
     -append \
     root=ld0a \
     -dtb \
     /usr/pkg/share/dtb/arm/vexpress-v2p-ca15-tc1.dtb

 -- 
 Andreas Gustafsson, gson@NetBSD.org

From: Martin Husemann <martin@duskware.de>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: port-evbarm/53308 (evbarm-earmv7hf performance regression under
 qemu)
Date: Thu, 30 Aug 2018 17:06:53 +0200

 Here is output from another real hardware (sunxi a20 cubietruck):

  > vmstat -e | grep gtmr
 armgtmr0 CNTV_TVAL write retry max                           5    0 misc
 armgtmr0 CNTV_CVAL read retry max                            1    0 misc
 armgtmr0 CNTVCT read retry max                             149    0 misc

 Martin

Responsible-Changed-From-To: joerg->jmcneill
Responsible-Changed-By: jmcneill@NetBSD.org
Responsible-Changed-When: Sun, 16 Sep 2018 13:12:04 +0000
Responsible-Changed-Why:
take


State-Changed-From-To: open->feedback
State-Changed-By: jmcneill@NetBSD.org
State-Changed-When: Sun, 16 Sep 2018 13:49:39 +0000
State-Changed-Why:
Fixed in -current, please test again.


State-Changed-From-To: feedback->closed
State-Changed-By: gson@NetBSD.org
State-Changed-When: Tue, 18 Sep 2018 05:16:42 +0000
State-Changed-Why:
The testbed has now completed a test run for the first time since May.
Thanks jmcneill!


>Unformatted:

Home	PR Database Search