From sudakov@sibptus.tomsk.ru  Wed Apr  1 07:38:50 2009
Return-Path: <sudakov@sibptus.tomsk.ru>
Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34])
	by hub.freebsd.org (Postfix) with ESMTP id 13F4F1065702
	for <FreeBSD-gnats-submit@freebsd.org>; Wed,  1 Apr 2009 07:38:50 +0000 (UTC)
	(envelope-from sudakov@sibptus.tomsk.ru)
Received: from relay2.tomsk.ru (relay2.tomsk.ru [212.73.124.8])
	by mx1.freebsd.org (Postfix) with ESMTP id F252E8FC17
	for <FreeBSD-gnats-submit@freebsd.org>; Wed,  1 Apr 2009 07:38:48 +0000 (UTC)
	(envelope-from sudakov@sibptus.tomsk.ru)
Received: from admin.sibptus.tomsk.ru (account sudakov@sibptus.tomsk.ru [212.73.125.240] verified)
  by relay2.tomsk.ru (CommuniGate Pro SMTP 5.1.13)
  with ESMTPSA id 12222266 for FreeBSD-gnats-submit@freebsd.org; Wed, 01 Apr 2009 14:38:47 +0700
Received: (from sudakov@localhost)
	by admin.sibptus.tomsk.ru (8.13.6/8.13.6/Submit) id n317ckeY097029;
	Wed, 1 Apr 2009 14:38:46 +0700 (OMSST)
	(envelope-from sudakov)
Message-Id: <200904010738.n317ckeY097029@admin.sibptus.tomsk.ru>
Date: Wed, 1 Apr 2009 14:38:46 +0700 (OMSST)
From: Victor Sudakov <sudakov@sibptus.tomsk.ru>
Reply-To: Victor Sudakov <sudakov@sibptus.tomsk.ru>
To: FreeBSD-gnats-submit@freebsd.org
Cc:
Subject: make buildworld on RELENG_6_4 is broken on SMP machines
X-Send-Pr-Version: 3.113
X-GNATS-Notify:

>Number:         133264
>Category:       misc
>Synopsis:       [build] make buildworld on RELENG_6_4 is broken on SMP machines
>Confidential:   no
>Severity:       serious
>Priority:       high
>Responsible:    freebsd-bugs
>State:          closed
>Quarter:        
>Keywords:       
>Date-Required:  
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Wed Apr 01 07:40:01 UTC 2009
>Closed-Date:    Thu Apr 02 14:29:58 UTC 2009
>Last-Modified:  Tue Apr 14 01:50:01 UTC 2009
>Originator:     Victor Sudakov <sudakov@sibptus.tomsk.ru>
>Release:        FreeBSD 6.4-RELEASE-p1 i386
>Organization:
AO "Svyaztransneft", SibPTUS
>Environment:
System: FreeBSD virtbsd64.sibptus.tomsk.ru 6.4-RELEASE-p1 FreeBSD 6.4-RELEASE-p1 #0: Sun Dec 21 08:13:10 UTC 2008 root@i386-builder.daemonology.net:/usr/obj/usr/src/sys/SMP i386

>Description:
make buildworld of RELENG_6_4 is broken on SMP machines

>How-To-Repeat:
Install RELENG_6_4 on an SMP machine. Run "make -j4 buildworld".
The make process will soon abort with:

===> lib/bind/lwres (clean)
rm -f lwres.3.gz lwres_buffer.3.gz lwres_config.3.gz lwres_context.3.gz lwres_gabn.3.gz lwres_gai_strerror.3.gz lwres_getaddrinfo.3.gz lwres_gethostent.3.gz lwres_getipnode.3.gz lwres_getnameinfo.3.gz lwres_getrrsetbyname.3.gz lwres_gnba.3.gz lwres_hstrerror.3.gz lwres_inetntop.3.gz lwres_noop.3.gz lwres_packet.3.gz lwres_resutil.3.gz lwres.3.cat.gz lwres_buffer.3.cat.gz lwres_config.3.cat.gz lwres_context.3.cat.gz lwres_gabn.3.cat.gz lwres_gai_strerror.3.cat.gz lwres_getaddrinfo.3.cat.gz lwres_gethostent.3.cat.gz lwres_getipnode.3.cat.gz lwres_getnameinfo.3.cat.gz lwres_getrrsetbyname.3.cat.gz lwres_gnba.3.cat.gz lwres_hstrerror.3.cat.gz lwres_inetntop.3.cat.gz lwres_noop.3.cat.gz lwres_packet.3.cat.gz lwres_resutil.3.cat.gz
rm -f a.out context.o gai_strerror.o getaddrinfo.o gethost.o getipnode.o getnameinfo.o getrrset.o herror.o lwbuffer.o lwconfig.o lwpacket.o lwresutil.o lwres_gabn.o lwres_gnba.o lwres_grbn.o lwres_noop.o lwinetaton.o lwinetpton.o lwinetntop.o print.o context.o.tmp gai_strerror.o.tmp getaddrinfo.o.tmp gethost.o.tmp getipnode.o.tmp getnameinfo.o.tmp getrrset.o.tmp herror.o.tmp lwbuffer.o.tmp lwconfig.o.tmp lwpacket.o.tmp lwresutil.o.tmp lwres_gabn.o.tmp lwres_gnba.o.tmp lwres_grbn.o.tmp lwres_noop.o.tmp lwinetaton.o.tmp lwinetpton.o.tmp lwinetntop.o.tmp print.o.tmp
rm -f context.po gai_strerror.po getaddrinfo.po gethost.po getipnode.po getnameinfo.po getrrset.po herror.po lwbuffer.po lwconfig.po lwpacket.po lwresutil.po lwres_gabn.po lwres_gnba.po lwres_grbn.po lwres_noop.po lwinetaton.po lwinetpton.po lwinetntop.po print.po  context.po.tmp gai_strerror.po.tmp getaddrinfo.po.tmp gethost.po.tmp getipnode.po.tmp getnameinfo.po.tmp getrrset.po.tmp herror.po.tmp lwbuffer.po.tmp lwconfig.po.tmp lwpacket.po.tmp lwresutil.po.tmp lwres_gabn.po.tmp lwres_gnba.po.tmp lwres_grbn.po.tmp lwres_noop.po.tmp lwinetaton.po.tmp lwinetpton.po.tmp lwinetntop.po.tmp print.po.tmp
rm -f context.So gai_strerror.So getaddrinfo.So gethost.So getipnode.So getnameinfo.So getrrset.So herror.So lwbuffer.So lwconfig.So lwpacket.So lwresutil.So lwres_gabn.So lwres_gnba.So lwres_grbn.So lwres_noop.So lwinetaton.So lwinetpton.So lwinetntop.So print.So context.so gai_strerror.so getaddrinfo.so gethost.so getipnode.so getnameinfo.so getrrset.so herror.so lwbuffer.so lwconfig.so lwpacket.so lwresutil.so lwres_gabn.so lwres_gnba.so lwres_grbn.so lwres_noop.so lwinetaton.so lwinetpton.so lwinetntop.so print.so context.So.tmp gai_strerror.So.tmp getaddrinfo.So.tmp gethost.So.tmp getipnode.So.tmp getnameinfo.So.tmp getrrset.So.tmp herror.So.tmp lwbuffer.So.tmp lwconfig.So.tmp lwpacket.So.tmp lwresutil.So.tmp lwres_gabn.So.tmp lwres_gnba.So.tmp lwres_grbn.So.tmp lwres_noop.So.tmp lwinetaton.So.tmp lwinetpton.So.tmp lwinetntop.So.tmp print.So.tmp
rm -f liblwres.so
rm -f liblwres.so.* liblwres.so
rm -f liblwres.a liblwres_p.a liblwres.so.11
1 error
*** Error code 2
1 error
*** Error code 2
1 error
17.852u 23.780s 0:33.68 123.6%	1009+742k 0+4693io 7pf+0w
virtbsd64#

>Fix:
None known to me. However, RELENG_6 does not seem to have the problem.

>Release-Note:
>Audit-Trail:

From: John Baldwin <jhb@FreeBSD.org>
To: bug-followup@FreeBSD.org, sudakov@sibptus.tomsk.ru
Cc:  
Subject: Re: misc/133264: [build] make buildworld on RELENG_6_4 is broken on SMP machines
Date: Thu, 2 Apr 2009 08:44:57 -0400

 I suspect you have a similar issue here in that you have a "dirty" source 
 tree.  If that is the case, a build without -j4 should also fail and should 
 show you where you have object files lying around.
 
 -- 
 John Baldwin
State-Changed-From-To: open->closed 
State-Changed-By: jhb 
State-Changed-When: Thu Apr 2 14:29:28 UTC 2009 
State-Changed-Why:  
This is most likely caused by a "dirty" source tree.  If that turns out not 
to be the case then this can be reopened. 

http://www.freebsd.org/cgi/query-pr.cgi?pr=133264 

From: Victor Sudakov <sudakov@sibptus.tomsk.ru>
To: John Baldwin <jhb@FreeBSD.org>
Cc: bug-followup@FreeBSD.org
Subject: Re: misc/133264: [build] make buildworld on RELENG_6_4 is broken on SMP machines
Date: Mon, 6 Apr 2009 08:40:17 +0700

 John Baldwin wrote:
 > tree.  If that is the case, a build without -j4 should also fail and should 
 > show you where you have object files lying around.
 
 Building on VMWare ESXi 3.5.0 Update 3 (FreeBSD as a guest OS) still
 fails occasionally if more than 1 CPU is allocated to the virtual
 machine. The failures are due to processes like sh, sed or cc1 dupming
 core on signal 11 during the build.
 
 I am using a pristine source tree now.
 
 The problem seems to be SMP related because enabling only 1 virtual
 CPU removes the problem.
 
 Should I open a new PR?
 
 -- 
 Victor Sudakov,  VAS4-RIPE, VAS47-RIPN
 sip:sudakov@sibptus.tomsk.ru

From: John Baldwin <jhb@freebsd.org>
To: Victor Sudakov <sudakov@sibptus.tomsk.ru>
Cc: bug-followup@freebsd.org
Subject: Re: misc/133264: [build] make buildworld on RELENG_6_4 is broken on SMP machines
Date: Mon, 6 Apr 2009 12:02:44 -0400

 On Sunday 05 April 2009 9:40:17 pm Victor Sudakov wrote:
 > John Baldwin wrote:
 > > tree.  If that is the case, a build without -j4 should also fail and 
 should 
 > > show you where you have object files lying around.
 > 
 > Building on VMWare ESXi 3.5.0 Update 3 (FreeBSD as a guest OS) still
 > fails occasionally if more than 1 CPU is allocated to the virtual
 > machine. The failures are due to processes like sh, sed or cc1 dupming
 > core on signal 11 during the build.
 > 
 > I am using a pristine source tree now.
 > 
 > The problem seems to be SMP related because enabling only 1 virtual
 > CPU removes the problem.
 > 
 > Should I open a new PR?
 
 Yes, this seems to be a different problem. :(  Do you see this building other 
 source trees as well?  (e.g. RELENG_6 or RELENG_7).
 
 -- 
 John Baldwin

From: Victor Sudakov <sudakov@sibptus.tomsk.ru>
To: John Baldwin <jhb@freebsd.org>
Cc: bug-followup@freebsd.org
Subject: Re: misc/133264: [build] make buildworld on RELENG_6_4 is broken on SMP machines
Date: Tue, 7 Apr 2009 10:12:51 +0700

 John Baldwin wrote:
 > > > tree.  If that is the case, a build without -j4 should also fail and 
 > should 
 > > > show you where you have object files lying around.
 > > 
 > > Building on VMWare ESXi 3.5.0 Update 3 (FreeBSD as a guest OS) still
 > > fails occasionally if more than 1 CPU is allocated to the virtual
 > > machine. The failures are due to processes like sh, sed or cc1 dupming
 > > core on signal 11 during the build.
 > > 
 > > I am using a pristine source tree now.
 > > 
 > > The problem seems to be SMP related because enabling only 1 virtual
 > > CPU removes the problem.
 > > 
 > > Should I open a new PR?
 > 
 > Yes, this seems to be a different problem. :(  Do you see this building other 
 > source trees as well?  (e.g. RELENG_6 or RELENG_7).
 
 RELENG_7_1 is not affected.
 RELENG_6 is not affected.
 
 -- 
 Victor Sudakov,  VAS4-RIPE, VAS47-RIPN
 sip:sudakov@sibptus.tomsk.ru

From: John Baldwin <jhb@freebsd.org>
To: Victor Sudakov <sudakov@sibptus.tomsk.ru>
Cc: bug-followup@freebsd.org
Subject: Re: misc/133264: [build] make buildworld on RELENG_6_4 is broken on SMP machines
Date: Tue, 7 Apr 2009 09:32:35 -0400

 On Monday 06 April 2009 11:12:51 pm Victor Sudakov wrote:
 > John Baldwin wrote:
 > > > > tree.  If that is the case, a build without -j4 should also fail and 
 > > should 
 > > > > show you where you have object files lying around.
 > > > 
 > > > Building on VMWare ESXi 3.5.0 Update 3 (FreeBSD as a guest OS) still
 > > > fails occasionally if more than 1 CPU is allocated to the virtual
 > > > machine. The failures are due to processes like sh, sed or cc1 dupming
 > > > core on signal 11 during the build.
 > > > 
 > > > I am using a pristine source tree now.
 > > > 
 > > > The problem seems to be SMP related because enabling only 1 virtual
 > > > CPU removes the problem.
 > > > 
 > > > Should I open a new PR?
 > > 
 > > Yes, this seems to be a different problem. :(  Do you see this building 
 other 
 > > source trees as well?  (e.g. RELENG_6 or RELENG_7).
 > 
 > RELENG_7_1 is not affected.
 > RELENG_6 is not affected.
 
 This is all while running the same kernel?
 
 -- 
 John Baldwin

From: Victor Sudakov <sudakov@sibptus.tomsk.ru>
To: John Baldwin <jhb@freebsd.org>
Cc: bug-followup@freebsd.org
Subject: Re: misc/133264: [build] make buildworld on RELENG_6_4 is broken on SMP machines
Date: Wed, 8 Apr 2009 12:30:02 +0700

 John Baldwin wrote:
 > > > > > tree.  If that is the case, a build without -j4 should also fail and 
 > > > should 
 > > > > > show you where you have object files lying around.
 > > > > 
 > > > > Building on VMWare ESXi 3.5.0 Update 3 (FreeBSD as a guest OS) still
 > > > > fails occasionally if more than 1 CPU is allocated to the virtual
 > > > > machine. The failures are due to processes like sh, sed or cc1 dupming
 > > > > core on signal 11 during the build.
 > > > > 
 > > > > I am using a pristine source tree now.
 > > > > 
 > > > > The problem seems to be SMP related because enabling only 1 virtual
 > > > > CPU removes the problem.
 > > > > 
 > > > > Should I open a new PR?
 > > > 
 > > > Yes, this seems to be a different problem. :(  Do you see this building other 
 > > > source trees as well?  (e.g. RELENG_6 or RELENG_7).
 > > 
 > > RELENG_7_1 is not affected.
 > > RELENG_6 is not affected.
 > 
 > This is all while running the same kernel?
 
 Each its own kernel.
 
 On the 6.4-RELEASE SMP kernel, any large compilation will crash
 eventually with signal 11. Using "-j4" or higher seems to aggravate
 the problem.
 
 -- 
 Victor Sudakov,  VAS4-RIPE, VAS47-RIPN
 sip:sudakov@sibptus.tomsk.ru

From: John Baldwin <jhb@freebsd.org>
To: Victor Sudakov <sudakov@sibptus.tomsk.ru>
Cc: bug-followup@freebsd.org
Subject: Re: misc/133264: [build] make buildworld on RELENG_6_4 is broken on SMP machines
Date: Mon, 13 Apr 2009 13:21:56 -0400

 On Wednesday 08 April 2009 1:30:02 am Victor Sudakov wrote:
 > John Baldwin wrote:
 > > > > > > tree.  If that is the case, a build without -j4 should also fail 
 and 
 > > > > should 
 > > > > > > show you where you have object files lying around.
 > > > > > 
 > > > > > Building on VMWare ESXi 3.5.0 Update 3 (FreeBSD as a guest OS) still
 > > > > > fails occasionally if more than 1 CPU is allocated to the virtual
 > > > > > machine. The failures are due to processes like sh, sed or cc1 
 dupming
 > > > > > core on signal 11 during the build.
 > > > > > 
 > > > > > I am using a pristine source tree now.
 > > > > > 
 > > > > > The problem seems to be SMP related because enabling only 1 virtual
 > > > > > CPU removes the problem.
 > > > > > 
 > > > > > Should I open a new PR?
 > > > > 
 > > > > Yes, this seems to be a different problem. :(  Do you see this 
 building other 
 > > > > source trees as well?  (e.g. RELENG_6 or RELENG_7).
 > > > 
 > > > RELENG_7_1 is not affected.
 > > > RELENG_6 is not affected.
 > > 
 > > This is all while running the same kernel?
 > 
 > Each its own kernel.
 > 
 > On the 6.4-RELEASE SMP kernel, any large compilation will crash
 > eventually with signal 11. Using "-j4" or higher seems to aggravate
 > the problem.
 
 Hmm, ok.  I'm not really sure why you are seeing this.  I don't know of 
 anything in RELENG_6 that isn't in 6.4 that would explain that.  Are you sure 
 your hardware is ok?
 
 -- 
 John Baldwin

From: Victor Sudakov <sudakov@sibptus.tomsk.ru>
To: John Baldwin <jhb@freebsd.org>
Cc: bug-followup@freebsd.org
Subject: Re: misc/133264: [build] make buildworld on RELENG_6_4 is broken on SMP machines
Date: Tue, 14 Apr 2009 08:47:09 +0700

 John Baldwin wrote:
 > > > > > > > tree.  If that is the case, a build without -j4 should also fail 
 > and 
 > > > > > should 
 > > > > > > > show you where you have object files lying around.
 > > > > > > 
 > > > > > > Building on VMWare ESXi 3.5.0 Update 3 (FreeBSD as a guest OS) still
 > > > > > > fails occasionally if more than 1 CPU is allocated to the virtual
 > > > > > > machine. The failures are due to processes like sh, sed or cc1 
 > dupming
 > > > > > > core on signal 11 during the build.
 > > > > > > 
 > > > > > > I am using a pristine source tree now.
 > > > > > > 
 > > > > > > The problem seems to be SMP related because enabling only 1 virtual
 > > > > > > CPU removes the problem.
 > > > > > > 
 > > > > > > Should I open a new PR?
 > > > > > 
 > > > > > Yes, this seems to be a different problem. :(  Do you see this 
 > building other 
 > > > > > source trees as well?  (e.g. RELENG_6 or RELENG_7).
 > > > > 
 > > > > RELENG_7_1 is not affected.
 > > > > RELENG_6 is not affected.
 > > > 
 > > > This is all while running the same kernel?
 > > 
 > > Each its own kernel.
 > > 
 > > On the 6.4-RELEASE SMP kernel, any large compilation will crash
 > > eventually with signal 11. Using "-j4" or higher seems to aggravate
 > > the problem.
 > 
 > Hmm, ok.  I'm not really sure why you are seeing this.  I don't know of 
 > anything in RELENG_6 that isn't in 6.4 that would explain that.  Are you sure 
 > your hardware is ok?
 
 As I stated above, my hardware is virtualized, it is a VMWare ESXi
 3.5.0 Update 3. Of course I am not sure it is ok (i.e. compatible with
 FreeBSD). However RELENG_7_1 compiles fine on it.
 
 I have had various problems with FreeBSD on different virtual
 platforms. For example, FreeBSD on Microsoft VPC has timing issues
 (the "microuptime went backwards" message). On Sun's VirtualBox,
 FreeBSD processes hang mysteriously with the obscure 
 "sigreturn: eflags = 0x80286" kernel message. Maybe FreeBSD does not
 feel like being a guest OS.
 
 
 -- 
 Victor Sudakov,  VAS4-RIPE, VAS47-RIPN
 sip:sudakov@sibptus.tomsk.ru
>Unformatted:
