amd64/183397: Kernel panic at first incoming ssh

2013-10-28 Thread Torbjorn Granlund

>Number: 183397
>Category:   amd64
>Synopsis:   Kernel panic at first incoming ssh
>Confidential:   no
>Severity:   non-critical
>Priority:   low
>Responsible:freebsd-amd64
>State:  open
>Quarter:
>Keywords:   
>Date-Required:
>Class:  sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Mon Oct 28 13:40:00 UTC 2013
>Closed-Date:
>Last-Modified:
>Originator: Torbjorn Granlund
>Release:10.0 BETA2
>Organization:
KTH
>Environment:
FreeBSD foo.gmplib.org 10.0-BETA2 FreeBSD 10.0-BETA2 #0 r257166: Sat Oct 26 
19:23:22 UTC 2013r...@snap.freebsd.org:/usr/obj/usr/src/sys/GENERIC amd64
>Description:
Environment:

Hardware is an Intel Haswell with 8 GB quality RAM
Virtualised using Xen 4.2 with NetBSD Dom0

Fresh vanilla FreeBSD 10.0 BETA2 install.

At first ssh into the system, it gets a panic.  Some basic
networking works, including telnet to the ssh port.

I've tried giving 256MiB and 512MiB to the FreeBSD guest.  The panic
happend in both cases.

Here is the retyped console output:

xn_rxeof: WARNING: response is -1!
xn_rxeof: WARNING: response is -1!

Fatal trap 12: page fault whle in kernel mode
cpuid = 0: apic id = 00
fault virtual address   = 0x1
fault code  = supervisor read data, page not present
instruction pointer = 0x20:0x8079e010
stack pointer   = 0x28:0xfe001b2509a0
frame pointer   = 0x28:0xfe001b2598f0
code segment= base 0x0, limit 0xf, type 0x1b
= DPL 0, prec 1, long 1, def32 0, gran 1
processor eflags= interrupt enabled, resume, IOPL = 0
current process = 12 (irq771: xn0)
trap number = 12
panic: page fault
cpuid = 0
KDB: stack backtrace:
(got tired of copying data here, vnc console will not allow me to
cut-and-paste.  If this data is needed, I can make a screen dump.)
>How-To-Repeat:

>Fix:


>Release-Note:
>Audit-Trail:
>Unformatted:
___
freebsd-amd64@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-amd64
To unsubscribe, send any mail to "freebsd-amd64-unsubscr...@freebsd.org"


Re: amd64/183397: Kernel panic at first incoming ssh

2013-10-28 Thread Torbjorn Granlund
I suppose non-critical/low might not be appropriate if Xen-based
virtualisation is common for FreeBSD.  Then critical/high is more
suitable!

For what it is worth, the exact same panic happens for a x86-32 install
(as opposed to x86-64 in the PR) under the same Xen/NetBSD system.

I should also perhaps mention that the Xen/NetBSD system runs several
guests flawlessly, including 32-bit and 64-bit FreeBSD 9.2, NetBSD
6.1.2, Debian GNU/Linux and Debian GNU/kFreeBSD.

I fully realise that the bug might very well be on the Xen side.

-- 
Torbjörn
___
freebsd-amd64@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-amd64
To unsubscribe, send any mail to "freebsd-amd64-unsubscr...@freebsd.org"

Re: amd64/183397: Kernel panic at first incoming ssh

2013-10-28 Thread Torbjorn Granlund
The following reply was made to PR amd64/183397; it has been noted by GNATS.

From: Torbjorn Granlund 
To: freebsd-gnats-sub...@freebsd.org
Cc: freebsd-amd64@FreeBSD.org
Subject: Re: amd64/183397: Kernel panic at first incoming ssh
Date: Mon, 28 Oct 2013 17:29:48 +0100

 I suppose non-critical/low might not be appropriate if Xen-based
 virtualisation is common for FreeBSD.  Then critical/high is more
 suitable!
 
 For what it is worth, the exact same panic happens for a x86-32 install
 (as opposed to x86-64 in the PR) under the same Xen/NetBSD system.
 
 I should also perhaps mention that the Xen/NetBSD system runs several
 guests flawlessly, including 32-bit and 64-bit FreeBSD 9.2, NetBSD
 6.1.2, Debian GNU/Linux and Debian GNU/kFreeBSD.
 
 I fully realise that the bug might very well be on the Xen side.
 
 --=20
 Torbj=C3=B6rn
___
freebsd-amd64@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-amd64
To unsubscribe, send any mail to "freebsd-amd64-unsubscr...@freebsd.org"


Re: amd64/183397: Kernel panic at first incoming ssh

2013-10-31 Thread Torbjorn Granlund
  The contents of the stack trace would indeed be very valuable.

These are screendumps (sorry, no text version!) of the last few lines of
the amd64 panic first, and then the x86-32 panic:



fbsd64-oh-no-ssh.bz2
Description: Binary data


fbsd32-oh-no-ssh.bz2
Description: Binary data

The image file(s) and the xen config file(s) can be made available on
request.  The image files might be just a bit to large for email
attachments, so I'd put at the gmplib ftp server.

My installs where completely standard, since the systems proved unusable
before I had a chance of customising them.  Also the NetBSD 6.1.2 / Xen
4.2 system is quite standard.

It might provide some useful unformation to note that a similar amd64
install works fine under kvm (Debian 7.2); the x86-32 install misbehaves
badly there, but it passes the "ssh test".  I suspect kvm, given the
observed state of the BSDs there: http://gmplib.org/~tege/virt.html

-- 
Torbjörn
___
freebsd-amd64@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-amd64
To unsubscribe, send any mail to "freebsd-amd64-unsubscr...@freebsd.org"

Re: amd64/183397: Kernel panic at first incoming ssh

2013-10-31 Thread Torbjorn Granlund
The following reply was made to PR amd64/183397; it has been noted by GNATS.

From: Torbjorn Granlund 
To: John Baldwin 
Cc: freebsd-amd64@freebsd.org, freebsd-gnats-sub...@freebsd.org
Subject: Re: amd64/183397: Kernel panic at first incoming ssh
Date: Thu, 31 Oct 2013 22:10:31 +0100

 --=-=-=
 
   The contents of the stack trace would indeed be very valuable.
 
 These are screendumps (sorry, no text version!) of the last few lines of
 the amd64 panic first, and then the x86-32 panic:
 
 
 --=-=-=
 Content-Type: application/octet-stream
 Content-Disposition: attachment; filename=fbsd64-oh-no-ssh.bz2
 Content-Transfer-Encoding: base64
 
 QlpoOTFBWSZTWaYCcu4AKjd/4CH+8AAA
 DeHOD6X0+yTKhSiEBdmEUFKJSFFsBlE7YkoSFANJBWdEid3OlMUdAaULNDQiZSeptKn7
 QalPyntKNMU/SnlPTCT1D0IZBiHqAGTQANGJgmEaekaGE9QYAgAADEYQDT1GQA0DJpoyaGmmmIZM
 RiaGgwSmRE0EQCNCZTEYE9TU9QNpkCMmRkBptRoDEANAAaAADTRoAGmg0ABoDI0AAAeoB6jQAAAG
 QGgCAAAJgACYADQmAADQAAAjRhGAmAA0ABoA0AAR6mTAAEwACYABCpIkzU0yFPUeoDNQ
 aPUAaGjEAADTIBoAGgyNAxNAPSAAABoMgMmgG0gaABppoAGQDQA0AZAANAiVKVNBo00Bo0GE00aa
 YQ0aaNMmBMmmIxMRkZGQZBoaYQZAZDTIaZMjJpoGmBMCNMATJkADJkwgNGhkGEyaPfvJVFBJqISa
 BTUNEn6p7VGgD9UBo0AAAZAaAADQAAAGgAAANAAHIaQREG4KLE1SBkqAqSKo
 kdc4vAGSCKhSqCSwgAJCSgoksCKhDCIiQkorIEgBAEANKg0oK0rDAiEhIKEEKLKyqkiQCw1ISyVB
 ASQk6VhFhhCYVpEqoGpkkYEgpJCSWIICCAhlgIKCohggmSJgKUgmEiAIlklmkoBpCCGShihSZWSi
 IoJiWkgkmQiGICSmYhIIVpIJIilFoJkKASkKUgoKgaAIgmoaWkiCYgqpKCkYgopqKiZiAKaUoKJo
 ilEmCahSJVGmqQJlKkgqkpKaQmIqSlBYgaCZKCioqBiBApUpRpQKBUppWhJKhClpUapEJlCmlBmU
 gkZiYAmWRkgkqkiKGiCoYmkJlYCEppIigIKiKWZCEaEgkImpJomCJpYqEiFpCkQiIkBopmpqIWKg
 pSZQppRamAiGlSmKiCUiQGmihaIAgqWAaAaSokqJqq9uATgiEoCBUaEpEKFiFDWlEUxKlIRCU0UQ
 S0g0UVMg0BUSpFLAUCFGY1JVQXNjCHZpRVB8pCaUqqgBqyqoOxkQBTqUoAKOoO1xo+wd9yABFy6Z
 gcSVa+FRUHElCdBxhBUBptGAwQgAJTSxKCiUNRAioU01QiIlFLVKK0NCUUgBFStADUwCJBKANU0U
 CIUNMSChRREKLQVQlKqUFDSlALTRVKUxNAUVVAVTVCVRRSLQRFCEVFCsUSDSU00I0jRTQlJRVAUE
 VMRVAUVVAUlES0NEQMUQ1TEFMyUTBVFFBRTMVRQUzCVUQBUVARNUtURRJFSVM1VUzFDUVLREFNUU
 k1DRNQUzBRMVNBBFTQVFQEyFVEBVRCUIKgNTBqQWMCiCCGzAAJBFEdtGsgoAIVQLSqTA0ggg54AV
 BMJtGBBF12A2yYJhGJZMYxjHKOPBScDMm26dDPt9wpbjTU8NRwVCp4Oqq6bhKy5VVbXV9hc0BBQk
 NE+KIioyOkJKUlpianJ6gogFIApqnyVVZXWFlaW1xdXl9gYWJjZGUCzM7QBaWprbG1ub3BxcnN0d
 XZ3eHl6+f0e/r9nx9fn/7/iAgYKDhIOFhvzDxETFRcZGx0fISMlJgpSVBS0vMTM1Nzk7PT4MHQUN
 FR0lLTU4SoqQlVV1lbXV9hY2VnaWttb3FzdXd5e31/gYOFh4mLjY+Rk5WXmZv6/YUL/M7P/uho6W
 nqautr7Gztbe5u73+3+Dh4uPk5ebn6Onq6++/53//f/aLOrrGj0mjAVQdJonIM5oZIqsyhFM2ZsE
 VAc2Bmwc2ZsUABM2BmwDNmbARRM2LmwzZmxEVDNgZsDNhmxRETNjmwDNhmwFXNi5sE0+8qZ/KFME
 AplMKYAUxAaYUymUxBEplMplM6WKANMKYGdOTl0UMkRCmLTApmdOQihl+iGTTGmUxUXyfewySmFM
 UpgANMSmLr9OQNMQWmB4XvIZNMKYDTAplMCmIUxphTCmIeG76GRTAphTFKYUxpjTBXwn8YZJTKZs
 MPVU5KFMphTGmKtMphTKYI0zRRKYHne6hkFM8VBHwOxwyFpmcKZoR0PIwyUpg0xphTENp0yGSUym
 FMCmeL8bDJ1TxMMimIbVtEMmmFMKYhTBpjTKYBTFpjTKYjTKYUymLTCmBTKYlMCmUymDTGmaMtyt
 iWwtjbG2JbG2BbLZbG2GmW5NsS2BbLZbFLYWwtlsQLYWyxAQgABwyvuNDQLYhbLYFsLYBbC2Nsth
 bLYNsthbBthbLZbAthbC2FsC2Wy2NsW2FsthbALY2y2JbAthbG2DbLZozDkYYuGOGYZhhhiYYGGY
 ZhiYY4ZhmGOGGGYZhhhmGYZhiYZhhhgYYmGGGhhmGYYJhmGIYZhimGKCoDhhhmGY
 Y6QUVAQkBQUVAkQRB3bEQUkxEoC0gKUgCUgghCiQIAFCqg/vvpaDAiomaEQVoB3Uc1CoB/yREM0i
 IFCFKBu9PAIrp+c9BkmZh3/S91ZABiD2kUCupBQhSae3wuVQiaJknJJDZJOD+WyQ9BxQQs53Zl+2
 ZD8jX83HABq4wgbqVNlKBrwB9qBzwh8SQRzSomnGtIidgkNM8eS7UK/1ORk4Isppf4ahSfZ/ts/l
 EzdeHo0kz6NRvMTdURWiTNwoxmnhEp1hErJBkYJUTDRMEyJhsUUtm2/7BjOTwz/b/POZ9OcjYOyD
 AlMpk2pugN3hZRxZmtBcQifriZ8lyVDLshSKZ4JQ8lEXFDSRS4UAJ70v1iaNK7OKFr+zno21zd72
 kixg+AdJq1CUsekfOihHZWnos4eJuRqAoIhEEVX4Po7wQBlBCOoojul6fsdOKoI4vBkkDeITYyLC
 PQ762tw8jg0x8/oWlMnwUdlVfW12c/uBJC/E+onA1ZQGRURkGxvVlLIRjIojFPb24warWvDWe78E
 USDAMb2EUFk/qmi3riBIJyJDV1KmkGkI6GoSImTaFXOM8TccVXhHSpZO839Ff4ioyA8OXnzv8tjT
 H5mzyVi0sEjcijM+e7nxfYVKvgPFTEVLfk3DDx7tB6Hor9VNIW8LEpUURZvRQklIxkKwUQA1CMPO
 XHjd7PEeEV4XsXpXDtsJ0FGEM3jfQlxKqRmH5C1uEJPFf4yYtdBPAspZ5E0eKvmdhZeBcR6A+H3q
 3GCAJz7QeEXCUzCUIDJbyh9iXGQp0lmcxJIoy1N4qgKIpA9HkJKJBoIytQT5N1baYcIaHYQoBptt
 ZIyK8kPKoKOoRjPFoInRDyFrWJ8HfMarWLColmKaUKJ5vU2k4Cx5OdUdB+aUnz/PMmTGNN7/a6Yz
 yY0VTUig0NDnx3TnjMF3SopIS4a59swaC5cWk8sxQgUezNl7SpjSQUm4h0TnyE7WepgMKBpkuWwa
 cWWeVCLXtFIYpdFEUWAtIcG4MYM3j2V5d1vS5jLB5fj/JxeI4kAXhPVy3scHB0ZJp7MbQhmBEsKc
 0mkoaiOHPimLCgR0+mMkzWvHx8nHfRLKGI22npcZJxtM062OrUadbRiiEcr4ptk5GuZO7iUSCMJh
 msThOve3wiCCTUqVW1wUjVoGY8CWJQ+AbVhI5Mw7YziZyeDjeaGihDHcotWWjvR329ckzwZzTxHW
 cBFmTXWq1RTFFgUtbDQMcC0XNjbTz7/XieJmyUyDta1CbEfYZbhGyKQ2+3cEmPcnEWEUQcg+AULd
 nbVu3JySTnBDgajMeZLLsZk7x/O7rEb2e1Lf0m+4uszmschyowu06LjTAZIDjbgfFwIZK9G4te3c
 1HJCLRjM5iQO24s9xg3F2NqHEaSFoHCMIOgiuZ7SbcOdKKy4+FZV+QvOeF4AzD8PVrtso3pXsj53
 zu/uDlv2YjEEPORVVSOoJEJCBiaEIUMEVR1ag2QUqHDC7gUhRVQogLBYwPK3IhmYiGqxjFNWxQC5
 AyNXNG8CWWMsR4nGKqEmEoM5JiUZiLIJjKFFEnoG472wpqkfQPsN+2ZQKlrD1ot8dUhSBJ9JkHUR
 IqFEoFhh8TuaVwWosbdmd89iTgxZ5jVEsKaoSgGETrdgimt6qyUxE3a1ooVMlMBE0lQZBmZnOJM4
 zbAwNpSOqsR0VTGCMMJQqVyoKrW3GUGZWxsAsUhZshTwknt7ZYoSwV0M+EUFo0ZpCxg9SK0GhXbM
 MQoWQOM5JcKdRPESMWIqJueNU28

Re: amd64/183397: Kernel panic at first incoming ssh

2013-10-31 Thread Torbjorn Granlund
Those attachments didn't come out too pretty at the web PR interface.

I put the screen dumps in png format here:

http://gmplib.org/~tege/fbsd32-oh-no-ssh.png
http://gmplib.org/~tege/fbsd64-oh-no-ssh.png

-- 
Torbjörn
___
freebsd-amd64@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-amd64
To unsubscribe, send any mail to "freebsd-amd64-unsubscr...@freebsd.org"

Re: amd64/183397: Kernel panic at first incoming ssh

2013-10-31 Thread Torbjorn Granlund
The following reply was made to PR amd64/183397; it has been noted by GNATS.

From: Torbjorn Granlund 
To: John Baldwin 
Cc: freebsd-amd64@freebsd.org,  freebsd-gnats-sub...@freebsd.org
Subject: Re: amd64/183397: Kernel panic at first incoming ssh
Date: Fri, 01 Nov 2013 00:42:19 +0100

 Those attachments didn't come out too pretty at the web PR interface.
 
 I put the screen dumps in png format here:
 
 http://gmplib.org/~tege/fbsd32-oh-no-ssh.png
 http://gmplib.org/~tege/fbsd64-oh-no-ssh.png
 
 --=20
 Torbj=C3=B6rn
___
freebsd-amd64@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-amd64
To unsubscribe, send any mail to "freebsd-amd64-unsubscr...@freebsd.org"


Re: amd64/183397: Kernel panic at first incoming ssh

2013-11-01 Thread Torbjorn Granlund
John Baldwin  writes:

  Can you fire up gdb against your 64-bit kernel file (e.g. gdb 
  /boot/kernel/kernel) and do 'l *xn_intr+0x7d'?

I'm afraid my ignorance of how to debug the kernel will show itself
here.

I did this:

1. booted the system.
2. logged in as root on the (vnc) console.
3. issued the command "gdb /boot/kernel/kernel"
4. Issued the above command and got this printout:

(gdb) l *xn_intr+0x7d
0x8079fb7d is in xn_intr (atomic.h:161).
156 atomic.h: No such file ot directory.
in atomic.h
(gdb)

This looks somewhat unsatisfactory to me...

If it would be beneficial to the freebsd project, I could bring this
NetBSD+Xen Haswell system to a public IP address and set up a temp
account.  I'd then would like a public ssh key which I can download from
*.freebsd.org.  But if that is not convenient for you, I'll certainly
follow through with any tests you will ask me to perform.

-- 
Torbjörn
___
freebsd-amd64@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-amd64
To unsubscribe, send any mail to "freebsd-amd64-unsubscr...@freebsd.org"

Re: amd64/183397: Kernel panic at first incoming ssh

2013-11-01 Thread Torbjorn Granlund
John Baldwin  writes:

  Hummm, I assume you can't get a crashdump when this happens?

Fortunately, you assume wrong.  :-)
I put a dump at http://gmplib.org/~tege/crash.tar.lz.

This is the entire /var/crash directory after the crash.

(While networking isn't working, making copying of the crash dump over
xn difficult, loop mounting the disk image on another FreeBSD machine
was a doddle.)

-- 
Torbjörn
___
freebsd-amd64@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-amd64
To unsubscribe, send any mail to "freebsd-amd64-unsubscr...@freebsd.org"

Re: amd64/183397: Kernel panic at first incoming ssh

2013-11-08 Thread Torbjorn Granlund
This bug is still present in BETA3.

I am quite stuck now wrt my GNU MP development and FreeBSD.  The m4 eval
bug I reported in the spring of 2012 is in all 4 subsequent 8.x and 9.x
releases which makes GMP disastrously miscompiled on BMI2 systems
(e.g. Haswell).  And FreeBSD 10 (tried A4, A5, B1, B2, B3) doesn't work
well enough on my Haswell system to allow incoming or outgoing ssh
connections without panicking.  (Curiously enough, I can transfer files
into the system without causing any kernel panic; I would have thought
ftp and ssh looked the same from the kernel's perspective.)

A home-brew kernel without HVM tolerates ssh connections.

(I'll save time and will not test any more FreeBSD 10 pre-releases unless
I get wind that this problem is solved.)

-- 
Torbjörn
___
freebsd-amd64@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-amd64
To unsubscribe, send any mail to "freebsd-amd64-unsubscr...@freebsd.org"