PANIC in kmem_alloc when loading many modules

From: Jun Su <junsu_at_m-net.arbornet.org>
Date: Thu, 15 Jan 2004 08:07:21 -0500 (EST)
Hi All,

I got a panic when I load many modules. The list of modules is:
kernel (ether.1, sysushm.1 sysvsem.1 sysvmsg.1 iosa.1 ppbus.1 pci.1)
if_fxp
miibus
if_wi
wlan
rc4
pccard
agp
cbb
exca
if_ln
snd_ich
snd_pcm
ums
usb
umass
cam

If I disable any of module, the system can boot ok. From the calling stack, 
it is obversious there is a recursive call in kmem_malloc. If you need any
more information, please feel free to let me know.

After I enable UMA_DEBUG, UMA_DEBUG_ALLOC, UMA_DEBUG_ALLOC_1, I got the following 
messages. (I typed by my hand. :-))
==============================
Other BIOS signatures found:
allocating one item from 1024(0xc0c1e000)
INERNAL: allocating one item from 16 bucket
slab_zalloc: allocating a new slab from 1024
INTERNAL: allocating one item from UMA Slabs
Allocating one item from KMAP Entry
INTERNAL : Allocating one item from 128 bucket
slab_zalloc: Allocating a new slab for KMAP ENTRY
PANIC: _mtx_lock_sleep: recuresed on non-recursive mutex system map

cpuid = 0;
Stack backtrace:
panic
_mtx_lock_sleep
_mtx_lock_flags
_vm_map_lock
kmem_malloc(c0c330a0, 1000, 101, c0821ad4, c055f8c4)
page_alloc
startup_alloc
slab_zalloc
uma_zone_slab
uma_zalloc_bucket
uma_zalloc_arg
vm_map_entry_create
vm_map_insert
kmem_malloc(c0c330a0, 1000, 2, c0821c74, c055f608)
page_alloc
slab_zalloc
uma_zone_slab
uma_zalloc_bucket
uma_zalloc_arg
malloc
hashinit
procinit
proc0_init
mi_startup

db>
Received on Thu Jan 15 2004 - 04:04:24 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:38 UTC