Re: panic: kmem_malloc; Should I increase some setting?

I have more info.

When untarring a bzip2 archive of 30MB the nr of wired mem in top goes up,  
until it  reaches 120MB and than it panics. (Normally it is around  
50/60MB.)
How can I debug this more or can others reproduce this?

Greetings,

Ronald.

On Wed, 30 Jun 2004 02:10:16 +0200, Ronald Klop  
<ronald-freebsd7_at_klop.yi.org> wrote:

> Hello,
>
> I get the following panics while untarring a 30MB archive on an md device
> create with:
> tmpmfs="YES"
> tmpsize="256m"
>
> There is plenty of disk space and plenty of swap on the time of the  
> panic.
> What info can I provide. The kernel will not make a kernel dump on
> /dev/ad1s1b, but I don't know why.
> My dmesg is attached.
>
> Greetings, Ronald.
>
>
>
>
> panic: kmem_malloc(16384): kmem_map too small: 86233088 total allocated
>
> syncing disks, buffers remaining... panic: kmem_malloc(4096): kmem_map  
> too
> small
> : 86233088 total allocated
> Uptime: 4m1s
> Dumping 255 MB
> ad1: timeout waiting for write DRQad1: timeout waiting for write DRQ
>
> Fatal trap 12: page fault while in kernel mode
> fault virtual address   = 0x0
> fault code              = supervisor read, page not present
> instruction pointer     = 0x8:0xc04a5653
> stack pointer           = 0x10:0xcbb8933c
> frame pointer           = 0x10:0xcbb89358
> code segment            = base 0x0, limit 0xfffff, type 0x1b
>                          = DPL 0, pres 1, def32 1, gran 1
> processor eflags        = interrupt enabled, resume, IOPL = 0
> current process         = 44 (syncer)
> trap number             = 12
> panic: page fault
> Uptime: 4m2s
> pfs_vncache_unload(): 1 entries remaining
> Automatic reboot in 15 seconds - press a key on the console to abort
> Rebooting...
>
>
>
>
> And I just got another one: (This one gives a M_WAITOK)
>
> panic: kmem_malloc(4096): kmem_map too small: 86233088 total allocated
>
> syncing disks, buffers remaining... panic: malloc(M_WAITOK) returned NULL
> Uptime: 3m33s
> Dumping 255 MB
>
>
> Fatal trap 12: page fault while in kernel mode
> fault virtual address   = 0x0
> fault code              = supervisor read, page not present
> instruction pointer     = 0x8:0xc04a5653
> stack pointer           = 0x10:0xcbb893c0
> frame pointer           = 0x10:0xcbb893dc
> code segment            = base 0x0, limit 0xfffff, type 0x1b
>                          = DPL 0, pres 1, def32 1, gran 1
> processor eflags        = interrupt enabled, resume, IOPL = 0
> current process         = 44 (syncer)
> trap number             = 12
> panic: page fault
> Uptime: 3m34s
> pfs_vncache_unload(): 1 entries remaining
> Automatic reboot in 15 seconds - press a key on the console to abort
> Rebooting...
>



-- 
  Ronald Klop
  Amsterdam, The Netherlands