From wkwu@csie.nctu.edu.tw  Sun Sep 12 12:29:54 2004
Return-Path: <wkwu@csie.nctu.edu.tw>
Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125])
	by hub.freebsd.org (Postfix) with ESMTP id B6E2B16A4CE
	for <FreeBSD-gnats-submit@freebsd.org>; Sun, 12 Sep 2004 12:29:54 +0000 (GMT)
Received: from mailgate.csie.nctu.edu.tw (mailgate.csie.nctu.edu.tw [140.113.209.7])
	by mx1.FreeBSD.org (Postfix) with ESMTP id 192AB43D1D
	for <FreeBSD-gnats-submit@freebsd.org>; Sun, 12 Sep 2004 12:29:54 +0000 (GMT)
	(envelope-from wkwu@csie.nctu.edu.tw)
Received: from mailgate.csie.nctu.edu.tw (wkwu@localhost [127.0.0.1])
	by mailgate.csie.nctu.edu.tw (8.13.1/8.13.1) with ESMTP id i8CCTshl038451
	for <FreeBSD-gnats-submit@freebsd.org>; Sun, 12 Sep 2004 20:29:54 +0800 (CST)
Received: (from wkwu@localhost)
	by mailgate.csie.nctu.edu.tw (8.13.1/8.13.1/Submit) id i8CCTsAV038450;
	Sun, 12 Sep 2004 20:29:54 +0800 (CST)
Message-Id: <200409121229.i8CCTsAV038450@mailgate.csie.nctu.edu.tw>
Date: Sun, 12 Sep 2004 20:29:54 +0800 (CST)
From: Wei-Kai.Wu@csie.nctu.edu.tw
Reply-To: Wei-Kai Wu <wkwu.amd64@csie.nctu.edu.tw>
To: FreeBSD-gnats-submit@freebsd.org
Cc:
Subject: amd64 5.3-BETA4 crash when heavy load
X-Send-Pr-Version: 3.113
X-GNATS-Notify:

>Number:         71644
>Category:       amd64
>Synopsis:       [panic] amd64 5.3-BETA4 crash when heavy load
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    linimon
>State:          closed
>Quarter:        
>Keywords:       
>Date-Required:  
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Sun Sep 12 12:30:19 GMT 2004
>Closed-Date:    Mon Jun 04 04:43:47 GMT 2007
>Last-Modified:  Mon Jun 04 04:43:47 GMT 2007
>Originator:     Wei-Kai Wu
>Release:        FreeBSD 5.3-BETA4 amd64
>Organization:
>Environment:
FreeBSD AMD64 5.3-BETA4 FreeBSD 5.3-BETA4 #1: Sun Sep 12 23:56:00 CST 2004     root@:/usr/obj/usr/src/sys/GENERIC  amd64

MSI K8D Master-3FS with dual Opteron 242 and 8G RAM
AHA 29160 SCSI card
2 bge (BroadCom) on board
>Description:
We run a extremely heavy load service that crashes often with kernel panic
"pmap_invalidate_page". More than ten thousand processes ran simultaneously.
The hardware should be no problem because we test on many boards and different
memory.

How can I help you to debug? Thanks!
>How-To-Repeat:
>Fix:
>Release-Note:
>Audit-Trail:

From: Kris Kennaway <kris@obsecurity.org>
To: Wei-Kai Wu <wkwu.amd64@csie.nctu.edu.tw>
Cc: FreeBSD-gnats-submit@FreeBSD.org
Subject: Re: kern/71644: amd64 5.3-BETA4 crash when heavy load
Date: Fri, 1 Oct 2004 09:46:47 -0700

 --oLBj+sq0vYjzfsbl
 Content-Type: text/plain; charset=us-ascii
 Content-Disposition: inline
 Content-Transfer-Encoding: quoted-printable
 
 On Sun, Sep 12, 2004 at 08:29:54PM +0800, Wei-Kai.Wu@csie.nctu.edu.tw wrote:
 
 > >Description:
 > We run a extremely heavy load service that crashes often with kernel panic
 > "pmap_invalidate_page". More than ten thousand processes ran simultaneous=
 ly.
 > The hardware should be no problem because we test on many boards and diff=
 erent
 > memory.
 >=20
 > How can I help you to debug? Thanks!
 
 Please retry with RELENG_5 (or wait for beta7), and see the following
 URL for detailed instructions on how to obtain necessary debugging
 information:
 
 http://www.freebsd.org/doc/en_US.ISO8859-1/books/developers-handbook/kernel=
 debug.html
 
 Kris
 
 --oLBj+sq0vYjzfsbl
 Content-Type: application/pgp-signature
 Content-Disposition: inline
 
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v1.2.6 (FreeBSD)
 
 iD8DBQFBXYn3Wry0BWjoQKURApmEAJ4wVVj38iyPsXr/M5UQjkSRksEyNgCgipCn
 T3OVYvd0JmKakcEcqfbd1Yk=
 =YpOm
 -----END PGP SIGNATURE-----
 
 --oLBj+sq0vYjzfsbl--
State-Changed-From-To: open->feedback 
State-Changed-By: delphij 
State-Changed-When: Sun Oct 31 17:40:46 GMT 2004 
State-Changed-Why:  
Dear submitter, 

Would you please try out 5.3-RC2 and confirm whether the issue is 
fixed?  The recent SACK fixes are likely to fix your problem. 

http://www.freebsd.org/cgi/query-pr.cgi?pr=71644 

From: Wei-Kai Wu <wkwu@csie.nctu.edu.tw>
To: FreeBSD-gnats-submit@FreeBSD.org
Cc:  
Subject: Re: amd64/71644: amd64 5.3-BETA4 crash when heavy load
Date: Sun, 28 Nov 2004 10:49:56 +0800

 On Sun, Oct 31, 2004 at 05:41:55PM +0000, Xin LI wrote:
 > Would you please try out 5.3-RC2 and confirm whether the issue is
 > fixed?  The recent SACK fixes are likely to fix your problem.
 
 We tested 5.3-STABLE (2004/11/27).
 The system crashed once with message :
 kernel trap 12, kernel panic
 current process (swi1: net)
 
 We are testing options NET_WITH_GIANT now.

From: Wei-Kai Wu <wkwu@csie.nctu.edu.tw>
To: Xin LI <delphij@FreeBSD.org>
Cc: FreeBSD-gnats-submit@FreeBSD.org, freebsd-bugs@FreeBSD.org
Subject: Re: amd64/71644: amd64 5.3-BETA4 crash when heavy load
Date: Sun, 28 Nov 2004 19:48:15 +0800

 Still crashed! kernel panic with Fatal trap 12.

From: Wei-Kai Wu <wkwu@csie.nctu.edu.tw>
To: Xin LI <delphij@frontfree.net>
Cc: FreeBSD-gnats-submit@FreeBSD.org, freebsd-bugs@FreeBSD.org
Subject: Re: amd64/71644: amd64 5.3-BETA4 crash when heavy load
Date: Sun, 28 Nov 2004 22:35:48 +0800

 On Sun, Nov 28, 2004 at 09:30:31PM +0800, Xin LI wrote:
 > Will you please try the following patch:
 > 
 > 	http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/vm/vm_contig.c.diff?r1=1.40&r2=1.41
 > 
 > (The patch should apply correctly in RELENG_5, while some positions might
 > be incorrect, that's harmless.)
 
 Still crash! Thousands of processes ran in 3 minutes and system crashed.

From: Wei-Kai Wu <wkwu@csie.nctu.edu.tw>
To: Xin LI <delphij@frontfree.net>
Cc: FreeBSD-gnats-submit@FreeBSD.org, freebsd-bugs@FreeBSD.org
Subject: Re: amd64/71644: amd64 5.3-BETA4 crash when heavy load
Date: Mon, 29 Nov 2004 01:07:26 +0800

 On Sun, Nov 28, 2004 at 11:16:39PM +0800, Xin LI wrote:
 > Hmm...   Is it possible to obtain a backtrace (bt full under kgdb) and post
 > it?
 > Additionally I suggest that you post the same thing to -stable@ and cc to
 > rwatson@, along with your dmesg.boot (verberose perferred).
 
 In fact, each time the system crashes, I have to fsck the disk
 for an hour. If you can tell me how to modify the code to umount
 the disk before kernel panic, I would like to try again and again.

From: Wei-Kai Wu <wkwu@csie.nctu.edu.tw>
To: Xin LI <delphij@frontfree.net>, freebsd-bugs@FreeBSD.org
Cc: freebsd-stable@FreeBSD.org, rwatson@FreeBSD.org
Subject: Re: amd64/71644: amd64 5.3-BETA4 crash when heavy load
Date: Thu, 2 Dec 2004 12:29:04 +0800

 On Sun, Nov 28, 2004 at 11:16:39PM +0800, Xin LI wrote:
 > Hmm...   Is it possible to obtain a backtrace (bt full under kgdb) and post
 > it?
 > Additionally I suggest that you post the same thing to -stable@ and cc to
 > rwatson@, along with your dmesg.boot (verberose perferred).
 
 Here is the panic messages: (If this is not what you want,
 could you tell me what I shall do? thanks!)
 
 Fatal trap 12: page fault while in kernel mode
 cpuid = 3; apic id = 03
 fault virtual address = 0x18
 fault code            = supervisor read, page not present
 instruction pointer   = 0x8:0xffffffff8026ece0
 stack pointer         = 0x10:0xffffffffd2d6f8b0
 frame pointer         = 0x10:0xffffffffd2d6f920
 code segment          = base 0x0, limit 0xfffff, type 0x1b
                       = DPL 0, pres 1, long 1, def32 0, gran 1
 processor eflags      = resume, IOPL = 0
 current process       = 46 (swi1: net)
 [thread 100025]
 Stopped at m_copym+0x40:    incl %ebp
 
 db> trace
 m_copym() at m_copym+0x40
 tcp_output() at tcp_output+0xdf1
 tcp_input() at tcp_input+0x1d2b
 netisr_processqueue() at netisr_processqueue+0xd2
 swi_net() at swi_net+0x13c
 ithread_loop() at ithread_loop+0x1b8
 fork_exit() at fork_exit+0xaa
 fork_trampoline() at fork_trampoline+0xe
 --- trap 0, rip = 0, rsp = 0xffffffffffd2d6fd00, rbp = 0 ---
State-Changed-From-To: feedback->open 
State-Changed-By: linimon 
State-Changed-When: Sun Oct 23 18:36:24 GMT 2005 
State-Changed-Why:  
Feedback was received quite some time ago. 

http://www.freebsd.org/cgi/query-pr.cgi?pr=71644 
State-Changed-From-To: open->feedback 
State-Changed-By: linimon 
State-Changed-When: Sat Jun 2 19:28:20 UTC 2007 
State-Changed-Why:  
Is this still a problem with recent versions of FreeBSD? 


Responsible-Changed-From-To: freebsd-amd64->linimon 
Responsible-Changed-By: linimon 
Responsible-Changed-When: Sat Jun 2 19:28:20 UTC 2007 
Responsible-Changed-Why:  

http://www.freebsd.org/cgi/query-pr.cgi?pr=71644 
State-Changed-From-To: feedback->closed 
State-Changed-By: linimon 
State-Changed-When: Mon Jun 4 04:43:23 UTC 2007 
State-Changed-Why:  
Submitter says this can be closed. 

http://www.freebsd.org/cgi/query-pr.cgi?pr=71644 
>Unformatted:
