Re: panic: vm_proc_new: upage allocation failed (5.6-BETA6)

On Thu, Sep 30, 2004 at 02:38:22PM -0400, Kevin M. Dulzo wrote:
> PAE still panics on boot, GENERIC still works flawlessly save 4G of RAM.
> 
> (kgdb) where
> #0  kdb_enter (msg=???) at /usr/src/sys/kern/subr_kdb.c:241
> #1  0xc039e3c7 in panic (fmt=0xc0564ae2 "vm_proc_new: upage allocation failed")
>     at /usr/src/sys/kern/kern_shutdown.c:537
> #2  0xc04c1a68 in vm_proc_new (p=0xca86ba80) at /usr/src/sys/vm/vm_glue.c:255
> #3  0xc0396add in proc_init (mem=0xca86ba80, size=448, flags=259)
>     at /usr/src/sys/kern/kern_proc.c:182
> #4  0xc04d33f5 in slab_zalloc (zone=0xc0a232c0, wait=259)
>     at /usr/src/sys/vm/uma_core.c:839
> #5  0xc04d4a18 in uma_zone_slab (zone=0xc0a232c0, flags=3)
>     at /usr/src/sys/vm/uma_core.c:1959
> #6  0xc04d4c2c in uma_zalloc_bucket (zone=0xc0a232c0, flags=3)
>     at /usr/src/sys/vm/uma_core.c:2062
> #7  0xc04d48c2 in uma_zalloc_arg (zone=0xc0a232c0, udata=0x0, flags=2)
>     at /usr/src/sys/vm/uma_core.c:1886
> #8  0xc038765f in fork1 (td=0xc05d9fc0, flags=131124, pages=0,
>     procp=0xc087fd50) at uma.h:274
> #9  0xc038dafd in kthread_create (func=0xc04d14bc <vm_pageout>, arg=0x0,
>     newpp=0xc05f2d98, flags=0, pages=0, fmt=0xc0536614 "%s")
>     at /usr/src/sys/kern/kern_kthread.c:86
> #10 0xc038daa3 in kproc_start (udata=0xc05c7840)
>     at /usr/src/sys/kern/kern_kthread.c:58
> #11 0xc03761f2 in mi_startup () at /usr/src/sys/kern/init_main.c:210
> #12 0xc024113e in begin () at /usr/src/sys/i386/i386/locore.s:348

Can you try doing "show map kernel_map" (or the address for kernel_map
if that doesn't work) in DDB when it panics, or translating that function
into gdb-speak and getting the output?  The address space seems to be
getting fatally full.

-- 
Brian Fundakowski Feldman                           \'[ FreeBSD ]''''''''''\
  <> green_at_FreeBSD.org                               \  The Power to Serve! \
 Opinions expressed are my own.                       \,,,,,,,,,,,,,,,,,,,,,,\