Re: Panic on boot after svn update

2012-07-29 Thread Garrett Cooper
On Sat, Jul 28, 2012 at 10:15 PM, Greg 'groggy' Lehey g...@freebsd.org wrote:
 On Sunday, 29 July 2012 at  0:53:55 -0400, David J. Weller-Fahy wrote:
 So, I recently updated and encountered a panic on boot which is
 reproducible, and wanted to see if anyone's encountered this before I
 file a PR.  I found a problem in (I think) recent changes to the e1000
 driver.  I'm running FreeBSD 10-CURRENT as a VirtualBox guest.

 #v+
 FreeBSD fork-pooh 10.0-CURRENT FreeBSD 10.0-CURRENT #0 r238764: Sat Jul 28 
 17:21:47 EDT 2012 root@fork-pooh:/usr/obj/usr/src/sys/GENERIC  amd64
 #v-

 I have the Adapter Type set to, Intel PRO/1000 MT Desktop (82540EM), and 
 the
 following card is detected by pciconf.
 ...
 Updating motd:.
 Starting ntpd.
 panic: _mtx_lock_sleep: recursed on non-recursive mutex em0 @ 
 /usr/src/sys/dev/e1000/if_lem.c:881

 aolMe too/aol The panic message is identical, and I'm also running
 in VirtualBox.  My version string (from strings on the kernel) is:

 FreeBSD 10.0-CURRENT #4: Sat Jul 28 09:45:10 EST 2012
 r...@swamp.lemis.com:/usr/obj/src/FreeBSD/svn/head/sys/GENERIC

 Note that this is a different EST (UTC+10).

 I have a dump, but I can't get much sense out of it:

 kgdb: kvm_read: invalid address (0x354540a)
 #0  0x in ?? ()

 I'm currently rebuilding the system, but it looks as if that won't
 help much.  One interesting point is that the first panic happened
 after installing the new image (from yesterday's sources) while I was
 trying to reboot with the old kernel, dating back to

 FreeBSD swamp.lemis.com 10.0-CURRENT FreeBSD 10.0-CURRENT #3: Sun May 13 
 14:34:43 EST 2012 
 r...@swamp.lemis.com:/usr/obj/src/FreeBSD/svn/head/sys/GENERIC  i386

See this thread:
http://lists.freebsd.org/pipermail/freebsd-current/2012-July/035593.html
.
-Garrett
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org


Re: Panic on boot after svn update

2012-07-29 Thread David J. Weller-Fahy
* Garrett Cooper yaneg...@gmail.com [2012-07-29 02:34 -0400]:
 See this thread:
 http://lists.freebsd.org/pipermail/freebsd-current/2012-July/035593.html

Huh - apparently my SA was not at its highest yesterday... Thanks for
the heads up!

-- 
dave [ please don't CC me ]


pgp9V1FHEuPMz.pgp
Description: PGP signature


Panic on boot after svn update

2012-07-28 Thread David J. Weller-Fahy
So, I recently updated and encountered a panic on boot which is
reproducible, and wanted to see if anyone's encountered this before I
file a PR.  I found a problem in (I think) recent changes to the e1000
driver.  I'm running FreeBSD 10-CURRENT as a VirtualBox guest.

#v+
FreeBSD fork-pooh 10.0-CURRENT FreeBSD 10.0-CURRENT #0 r238764: Sat Jul 28 
17:21:47 EDT 2012 root@fork-pooh:/usr/obj/usr/src/sys/GENERIC  amd64
#v-

I have the Adapter Type set to, Intel PRO/1000 MT Desktop (82540EM), and the
following card is detected by pciconf.

#v+
dave@fork-pooh:~$ sudo pciconf -l -bcev
...
em0@pci0:0:3:0: class=0x02 card=0x001e8086 chip=0x100e8086 rev=0x02 hdr=0x00
vendor = 'Intel Corporation'
device = '82540EM Gigabit Ethernet Controller'
class  = network
subclass   = ethernet
bar   [10] = type Memory, range 32, base 0xf000, size 131072, enabled
bar   [18] = type I/O Port, range 32, base 0xd010, size  8, enabled
cap 01[dc] = powerspec 2  supports D0 D3  current D0
cap 07[e4] = PCI-X supports 2048 burst read, 1 split transaction
...
#v-

With revision 238764 the system boots and connects to the network with no
problem.  However, if I update to 238770 (as for why 238770, last revision
which touched if_lem.c [1])...

[1]: http://svnweb.freebsd.org/base/head/sys/dev/e1000/if_lem.c?view=log

#v+
dave@fork-pooh:/usr/src$ sudo svn up -r 238770
Password:
Updating '.':
Usys/netinet/ip_carp.c
Usys/dev/e1000/if_lem.c
Usys/dev/usb/serial/uplcom.c
Usys/dev/usb/usbdevs
Updated to revision 238770.
dave@fork-pooh:/usr/src$ cd /usr/obj/
dave@fork-pooh:/usr/obj$ sudo chflags -R noschg *
dave@fork-pooh:/usr/obj$ sudo rm -rf *
dave@fork-pooh:/usr/obj$ cd /usr/src/
dave@fork-pooh:/usr/src$ sudo make buildkernel
dave@fork-pooh:/usr/src$ make installkernel
dave@fork-pooh:/usr/src$ sudo shutdown -r now
...
#v-

Upon reboot, I see...

#v+
...
Updating motd:.
Starting ntpd.
panic: _mtx_lock_sleep: recursed on non-recursive mutex em0 @
/usr/src/sys/dev/e1000/if_lem.c:881

cpuid = 0
KDB: enter: panic
[ thread pid 12 tid 100025 ]
Stopped at  kdb_enter+0x3b: movq$0,0x9c4752(%rip)
#v-

After changing the Network Adapter (in VirtualBox) to PCnet-PCI II
(Am79C970A) the system boots up correctly (on the same revision), and the
network is functional.

What other information would be useful in figuring out what's wrong?

Regards,
-- 
dave [ please don't CC me ]


pgpLtbpRUtYlz.pgp
Description: PGP signature


Re: Panic on boot after svn update

2012-07-28 Thread Greg 'groggy' Lehey
On Sunday, 29 July 2012 at  0:53:55 -0400, David J. Weller-Fahy wrote:
 So, I recently updated and encountered a panic on boot which is
 reproducible, and wanted to see if anyone's encountered this before I
 file a PR.  I found a problem in (I think) recent changes to the e1000
 driver.  I'm running FreeBSD 10-CURRENT as a VirtualBox guest.

 #v+
 FreeBSD fork-pooh 10.0-CURRENT FreeBSD 10.0-CURRENT #0 r238764: Sat Jul 28 
 17:21:47 EDT 2012 root@fork-pooh:/usr/obj/usr/src/sys/GENERIC  amd64
 #v-

 I have the Adapter Type set to, Intel PRO/1000 MT Desktop (82540EM), and the
 following card is detected by pciconf.
 ...
 Updating motd:.
 Starting ntpd.
 panic: _mtx_lock_sleep: recursed on non-recursive mutex em0 @ 
 /usr/src/sys/dev/e1000/if_lem.c:881

aolMe too/aol The panic message is identical, and I'm also running
in VirtualBox.  My version string (from strings on the kernel) is:

FreeBSD 10.0-CURRENT #4: Sat Jul 28 09:45:10 EST 2012
r...@swamp.lemis.com:/usr/obj/src/FreeBSD/svn/head/sys/GENERIC

Note that this is a different EST (UTC+10).

I have a dump, but I can't get much sense out of it:

kgdb: kvm_read: invalid address (0x354540a)
#0  0x in ?? ()

I'm currently rebuilding the system, but it looks as if that won't
help much.  One interesting point is that the first panic happened
after installing the new image (from yesterday's sources) while I was
trying to reboot with the old kernel, dating back to

FreeBSD swamp.lemis.com 10.0-CURRENT FreeBSD 10.0-CURRENT #3: Sun May 13 
14:34:43 EST 2012 
r...@swamp.lemis.com:/usr/obj/src/FreeBSD/svn/head/sys/GENERIC  i386

Greg
--
Sent from my desktop computer.
Finger g...@freebsd.org for PGP public key.
See complete headers for address and phone numbers.
This message is digitally signed.  If your Microsoft MUA reports
problems, please read http://tinyurl.com/broken-mua


pgprHxhQYsUWK.pgp
Description: PGP signature