Re: 5.4-RELEASE lockups on amd64 SMP

2005-06-09 Thread Matthew Grooms
t now I have just been rebooting back to a UP kernel which has never shown any sign of problems. Matthew Grooms -Original Message- From: Grooms, Matthew Sent: Wed 6/8/2005 6:22 PM To: Max Laier Cc: Palle Girgensohn; Kris Kennaway; freebsd-stable@freebsd.org; [EMAIL PROTECTED]; [EMAIL PROTECT

Re: 5.4-RELEASE lockups on amd64 SMP

2005-06-08 Thread Max Laier
hew Grooms > > -Original Message- > From: Grooms, Matthew > Sent: Wed 6/8/2005 6:22 PM > To: Max Laier > Cc: Palle Girgensohn; Kris Kennaway; freebsd-stable@freebsd.org; > [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: RE: 5.4-RELEASE lockups on > amd64 SMP > > Ma

RE: 5.4-RELEASE lockups on amd64 SMP

2005-06-08 Thread Grooms, Matthew
to a UP kernel which has never shown any sign of problems. Matthew Grooms -Original Message- From: Grooms, Matthew Sent: Wed 6/8/2005 6:22 PM To: Max Laier Cc: Palle Girgensohn; Kris Kennaway; freebsd-stable@freebsd.org; [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: RE: 5.4-RELEASE l

RE: 5.4-RELEASE lockups on amd64 SMP

2005-06-08 Thread Grooms, Matthew
Matthew, can you try the attached diff. Available for 5 and CURRENT. I recall that this problem was seen before, strange that I didn't see the problem. Sounds familiar to you? Please try the patch and let me know if that helps. Thanks a lot. On Wednesday 08 June 2005 01:35, Matthew Grooms

Re: 5.4-RELEASE lockups on amd64 SMP

2005-06-08 Thread Matthew Grooms
Are these available for download somewhere? I am getting rejected chunks and I suspect it may be the mail client I am using. This does apply cleanly to 5.4-RELEASE correct? Matthew Grooms Network Engineer Seton Healthcare Network http://www.seton.net/ [EMAIL PROTECTED] (512) 324 9913 Max Laier

Re: 5.4-RELEASE lockups on amd64 SMP

2005-06-08 Thread Matthew Grooms
BTW : Had you tested pfsync between two SMP systems with decent traffic flow? It usually took about 3 days to hit the LOR but the panic shows up in about 10-20 minutes. Matthew Grooms Network Engineer Seton Healthcare Network http://www.seton.net/ [EMAIL PROTECTED] (512) 324 9913 Max Laier wro

Re: 5.4-RELEASE lockups on amd64 SMP

2005-06-08 Thread Max Laier
Matthew, can you try the attached diff. Available for 5 and CURRENT. I recall that this problem was seen before, strange that I didn't see the problem. Sounds familiar to you? Please try the patch and let me know if that helps. Thanks a lot. On Wednesday 08 June 2005 01:35, Matthew Grooms

Re: 5.4-RELEASE lockups on amd64 SMP

2005-06-07 Thread Matthew Grooms
Palle, Its a dell 2850 w/ Dual CPU, AMR controller and 6x em devices ( 2x on board and 4x Intel Pro 1000 MT ). If you still want the full dmesg output, reply and I will send it to you. I suspect the issues I am seeing are related to a SMP locking deficiency in pf/pfsync. I have a second iden

Re: 5.4-RELEASE lockups on amd64 SMP

2005-06-07 Thread Palle Girgensohn
--On måndag, juni 06, 2005 19.57.03 -0400 Kris Kennaway <[EMAIL PROTECTED]> wrote: On Mon, Jun 06, 2005 at 06:54:05PM -0500, Grooms, Matthew wrote: My appologies. With the debug options listed in my previous post ( should have read 5.4 not 5.3 ), I got a lock order reversal. After a while, it

RE: 5.4-RELEASE lockups on amd64 SMP

2005-06-07 Thread Grooms, Matthew
--- From: Grooms, Matthew Sent: Mon 6/6/2005 6:54 PM To: freebsd-stable@freebsd.org Subject: 5.4-RELEASE lockups on amd64 SMP My appologies. With the debug options listed in my previous post ( should have read 5.4 not 5.3 ), I got a lock order reversal. After a while, it paniced and spat

Re: 5.4-RELEASE lockups on amd64 SMP

2005-06-06 Thread Kris Kennaway
On Mon, Jun 06, 2005 at 06:54:05PM -0500, Grooms, Matthew wrote: > My appologies. With the debug options listed in my previous post ( should > have read 5.4 not 5.3 ), I got a lock order reversal. After a while, it > paniced and spat out this ... > > lock order reversal > 1st 0x80752ec0

5.4-RELEASE lockups on amd64 SMP

2005-06-06 Thread Grooms, Matthew
My appologies. With the debug options listed in my previous post ( should have read 5.4 not 5.3 ), I got a lock order reversal. After a while, it paniced and spat out this ... lock order reversal 1st 0x80752ec0 pf task mtx (pf task mtx) @ contrib/pf/net/if_pfsync.c:1621 2nd 0x80