Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 30 Nov 2004 13:30:06 -0800
From:      Steve Kargl <sgk@troutmask.apl.washington.edu>
To:        "Ketrien I. Saihr-Kenchedra" <ketrien@error404.nls.net>
Cc:        freebsd-amd64@freebsd.org
Subject:   Re: kernel panic with greater that 8 GB of memory
Message-ID:  <20041130213006.GA1079@troutmask.apl.washington.edu>
In-Reply-To: <20041130194227.GA32604@troutmask.apl.washington.edu>
References:  <20041129211341.GA26548@troutmask.apl.washington.edu> <20041129233211.GA27426@troutmask.apl.washington.edu> <41AC6FF8.40501@freebsd.org> <20041130183555.GA32237@troutmask.apl.washington.edu> <20041130142652.A59122@bahre.achedra.org> <20041130194227.GA32604@troutmask.apl.washington.edu>

next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Nov 30, 2004 at 11:42:27AM -0800, Steve Kargl wrote:
> On Tue, Nov 30, 2004 at 02:35:30PM -0500, Ketrien I. Saihr-Kenchedra wrote:
> 
> > More importantly, which BIOS version? You have to run the 2.03 BIOS on 
> > the K8SPro if you're running 16GB. Make sure that ccNUMA support is off
> > in the BIOS as well - this was a never-ending source of fun on both the
> > S2882 and S4882's. I can't tell you exactly why, but having ccNUMA on
> > with the S4882 and S288x's is a never-ending font of panic()s. But the
> > 2.01 and 2.02 BIOSes on the S2882 are exceptionally miserable.
> > At least it's not an S4882 1.02:
> > "DDR 400MHz Memmory Detected"
> 
> The BIOS is at 2.03.  I haven't tried turning off the ccNUMA support.
> I'll give it a shot in a moment.
> 

I disabled the only thing in the BIOS that mentions NUMA
(under Advanced ACPI Configuration).  FreeBSD now sees
the 16 GB, but she still panics.  A verbose boot shows

real memory = 17179869184 (16384 MB)
Physical memory chuck(s):
0x0000000000001000 - 0000000000009bfff, 634880 bytes (155 pages)
0x000000000082d000 - 000000000fbfeffff, 4219219968 bytes (1030083 pages)
0x0000000100000000 - 000000003e1fcffff, 12381388800 bytes (3022800 pages)
panic: vm_page_insert: page already inserted
cpuid = 0
KDB: stack backtrace:
kdb_backtrace() at kdb_backtrace+0x37
panic() at panic+0x1d1
vm_page_insert() at vm_page_insert+0x4e
vm_page_alloc() at vm_page_alloc+0x315
vm_page_grab() at vm_page_grab+0x141
kmem_alloc() at kmem_alloc+0x10c
vm_ksubmap_init() at vm_ksubmap_init+0x7d
cpu_startup() at cpu_startup+0xbc
mi_statup() at mi_startup+0xb6
btext() at btext+0x2c




-- 
Steve



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20041130213006.GA1079>