Re: panic of the day: pmap_enter: no pv entries available (amd64)
From: Christopher A. Kantarjiev (cak_at_dimebank.com)
Date: 07/27/05
- Previous message: Christopher A. Kantarjiev: "panic of the day: pmap_enter: no pv entries available"
- In reply to: Christopher A. Kantarjiev: "panic of the day: pmap_enter: no pv entries available"
- Next in thread: Christopher A. Kantarjiev: "Re: panic of the day: pmap_enter: no pv entries available (amd64)"
- Reply: Christopher A. Kantarjiev: "Re: panic of the day: pmap_enter: no pv entries available (amd64)"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Tue, 26 Jul 2005 21:21:49 -0700
Christopher A. Kantarjiev wrote:
> This is on amd64, 2 dual-core Opteron CPUs with 16GB of RAM, running
> 2.0.2_STABLE. This is the first time we've really exercised all the
> memory (mostly using mmap to move large files in/out of memory).
>
> It's not immediately obvious how to increase the size of the PV pool; it
> appears that pmap_alloc_pv() should be doing this dynamically as needed,
> and I guess it's failing.
It's been a long time since I looked at the uvm code in detail, but it sure
looks like I'm out of kvm address space. So I did
RCS file: RCS/vmparam.h,v
retrieving revision 1.1
diff -r1.1 vmparam.h
102c102
< #define VM_MAX_KERNEL_ADDRESS 0xffff800100000000
--- > #define VM_MAX_KERNEL_ADDRESS 0xffff800200000000 and we'll see what happens next - I'm ... reassured by the fact that it at least booted!
- Previous message: Christopher A. Kantarjiev: "panic of the day: pmap_enter: no pv entries available"
- In reply to: Christopher A. Kantarjiev: "panic of the day: pmap_enter: no pv entries available"
- Next in thread: Christopher A. Kantarjiev: "Re: panic of the day: pmap_enter: no pv entries available (amd64)"
- Reply: Christopher A. Kantarjiev: "Re: panic of the day: pmap_enter: no pv entries available (amd64)"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|