panic: vm_page_remove(): page not found in hash

[Miklos Niedermayer]

Hi,

I've found some short threads on $subject in the archives, but
unfortunately I still don't have a solution.  I have two completely
different machines:

- PII-400 with 256 MB of RAM, xl0 NIC, and an IDE hard drive
- Athlon-1200, Asus/Sis motherboard, 256 MB, onboard rl0, Quantum 30GB IDE
  drive

The first machine started to act like if it's RAM was going to die
(strange reboots), so i quickly set up another machine, dump & restored
the things, etc.  At the first make buildworld it rebooted (in fact, it
paniced, but i wasn't able to see that on an SSH terminal).  My kernel is
almost-GENERIC (MAXUSERS set to 64, NMBCLUSTERS to 8192 -tried to leave it
undefined as well-), sysctl.conf:

kern.ps_showallprocs=0
net.inet.icmp.log_redirect=1
jail.set_hostname_allowed=0
net.inet.tcp.blackhole=2
net.inet.udp.blackhole=1

I also use IPFilter with lots of ipfmeta-generated rules if that counts.
A make -j4 buildworld almost certainly panics the system, a simple kernel
build works, fortunately.  I've also made some tests regarding those
IDE corruption problems on some AMD motherboards/chipsets but my system
seems to be immunic.  (I've put 4 IDE disks into the machine and stressed
them for hours)

If anyone wants to take a look at my crash dumps etc. in detail, i'll be
glad to make them available.

This GDB was configured as "i386-unknown-freebsd"...
IdlePTD 4853760
initial pcb at 3e7ba0
panicstr: vm_page_remove(): page not found in hash
panic messages:
---
panic: vm_page_remove(): page not found in hash
syncing disks... 158 15 3 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
giving up on 1 buffers
Uptime: 13h49m16s

dumping to dev #ad/0x20001, offset 1589400
dump ata0: resetting devices .. done
<dump blah blah removed>

---
#0  dumpsys () at ../../kern/kern_shutdown.c:473
473             if (dumping++) {
(kgdb) where
#0  dumpsys () at ../../kern/kern_shutdown.c:473
#1  0xc01bbe9b in boot (howto=256) at ../../kern/kern_shutdown.c:313
#2  0xc01bc268 in poweroff_wait (junk=0xc0388ae0, howto=-1061484440)
    at ../../kern/kern_shutdown.c:581
#3  0xc02d7f0b in vm_page_remove (m=0xc0bb0868) at ../../vm/vm_page.c:460
#4  0xc02d8548 in vm_page_free_toq (m=0xc0bb0868) at ../../vm/vm_page.c:1090
#5  0xc02d6952 in vm_object_terminate (object=0xcdfa2cc0)
    at ../../vm/vm_page.h:527
#6  0xc02d681e in vm_object_deallocate (object=0xcdfa2cc0)
    at ../../vm/vm_object.c:387
#7  0xc02d3deb in vm_map_entry_delete (map=0xcddf9400, entry=0xcde71fc0)
    at ../../vm/vm_map.c:1823
#8  0xc02d3f6d in vm_map_delete (map=0xcddf9400, start=0, end=3217031168)
    at ../../vm/vm_map.c:1926
#9  0xc02d3ffa in vm_map_remove (map=0xcddf9400, start=0, end=3217031168)
    at ../../vm/vm_map.c:1951
#10 0xc01b45ec in exit1 (p=0xcde0b0c0, rv=0) at ../../kern/kern_exit.c:217
#11 0xc01b43cc in exit1 (p=0xcde0b0c0, rv=-1071908839)
    at ../../kern/kern_exit.c:103
#12 0xc032ea49 in syscall2 (frame={tf_fs = 47, tf_es = 47, tf_ds = 47,
      tf_edi = 136225832, tf_esi = 0, tf_ebp = -1077939620,
      tf_isp = -840220716, tf_ebx = -1, tf_edx = 135829604, tf_ecx = 30,
      tf_eax = 1, tf_trapno = 0, tf_err = 2, tf_eip = 135787856, tf_cs = 31,
      tf_eflags = 643, tf_esp = -1077939664, tf_ss = 47})
    at ../../i386/i386/trap.c:1155
#13 0xc0320445 in Xint0x80_syscall ()
Cannot access memory at address 0xbfbff25c.
(kgdb)




                                         ______  o _. __
                                         / / / (_(_(__(_)  @ bsd.hu