Re: systematic panic on an SMP machine for 5.1-Release

2003-07-09 Thread Thierry Herbelot
[taking one random Terry mail to get back to the list] Hello, I have attached some more info about the machine : * script.crash : simple gdb output from the vmcore (with symbols) - one surprising point - for me - is the message : ---Can't read userspace from dump, or kernel process--- the core h

Re: systematic panic on an SMP machine for 5.1-Release

2003-07-09 Thread Terry Lambert
Steve Kargl wrote: > > PS : is this an indication of bug in the p-III or in the chipset ? > > (ISTR these options could be used to get around unnamed errata of the p-IV) > > You'll need to search the mailing list archive for vague ramblings > by Terry Lambert about these option, large memory machi

Re: systematic panic on an SMP machine for 5.1-Release

2003-07-09 Thread Terry Lambert
Kevin Oberman wrote: > > PS : is this an indication of bug in the p-III or in the chipset ? > > (ISTR these options could be used to get around unnamed errata of > > the p-IV) > > Terry has never been specific (due to non-disclosure), but adding > options DISABLE_PSE to the kernel of my P4 system

Re: systematic panic on an SMP machine for 5.1-Release

2003-07-08 Thread Bosko Milekic
On Tue, Jul 08, 2003 at 09:47:23PM +0200, Thierry Herbelot wrote: > Le Tuesday 08 July 2003 21:30, Steve Kargl a ?crit : > > On Tue, Jul 08, 2003 at 09:09:20PM +0200, Thierry Herbelot wrote: > > > TfH > > > > > > PS : as I said, the box has a 2GB RAM, so moving around the vmcore is not > > > eas

Re: systematic panic on an SMP machine for 5.1-Release

2003-07-08 Thread Steve Kargl
On Tue, Jul 08, 2003 at 09:47:23PM +0200, Thierry Herbelot wrote: > Le Tuesday 08 July 2003 21:30, Steve Kargl a ?crit : > > On Tue, Jul 08, 2003 at 09:09:20PM +0200, Thierry Herbelot wrote: > > > TfH > > > > > > PS : as I said, the box has a 2GB RAM, so moving around the vmcore is not > > > easy

Re: systematic panic on an SMP machine for 5.1-Release

2003-07-08 Thread Kevin Oberman
> From: Thierry Herbelot <[EMAIL PROTECTED]> > Date: Tue, 8 Jul 2003 21:47:23 +0200 > Sender: [EMAIL PROTECTED] > > PS : is this an indication of bug in the p-III or in the chipset ? > (ISTR these options could be used to get around unnamed errata of > the p-IV) Terry has never been specific (due

Re: systematic panic on an SMP machine for 5.1-Release

2003-07-08 Thread Thierry Herbelot
Le Tuesday 08 July 2003 21:30, Steve Kargl a écrit : > On Tue, Jul 08, 2003 at 09:09:20PM +0200, Thierry Herbelot wrote: > > TfH > > > > PS : as I said, the box has a 2GB RAM, so moving around the vmcore is not > > easy > > Add "options DISABLE_PSE" and "options DISABLE_PG_G" > to your kernel c

Re: systematic panic on an SMP machine for 5.1-Release

2003-07-08 Thread Steve Kargl
On Tue, Jul 08, 2003 at 09:09:20PM +0200, Thierry Herbelot wrote: > > TfH > > PS : as I said, the box has a 2GB RAM, so moving around the vmcore is not easy > Add "options DISABLE_PSE" and "options DISABLE_PG_G" to your kernel configuration. Report back if you still have a panic. -- St

Re: systematic panic on an SMP machine for 5.1-Release

2003-07-08 Thread Thierry Herbelot
Le Tuesday 08 July 2003 20:44, Doug White a écrit : > On Tue, 8 Jul 2003, Thierry Herbelot wrote: > > [I just see that this message is at least lacking a "dmesg" output : I'll > > send it tomorrow.] > > A crashdump will be inifintely more useful. > > > #1 0xc030aeb8 in boot () > > #2 0xc030b21f i

Re: systematic panic on an SMP machine for 5.1-Release

2003-07-08 Thread Doug White
On Tue, 8 Jul 2003, Thierry Herbelot wrote: > [I just see that this message is at least lacking a "dmesg" output : I'll send > it tomorrow.] A crashdump will be inifintely more useful. > #1 0xc030aeb8 in boot () > #2 0xc030b21f in panic () > #3 0xc0164142 in db_panic () Note no location info