4.8 STABLE page fault under heavy swapping

[Jonathan]

kernel: FreeBSD 4.8-STABLE #0: Mon Aug 25 19:26:01 CDT 2003
compiler: gcc version 2.95.4 20020320 [FreeBSD]
computer:  IBM ThinkPad 600E

Hi:

 Both my desktop and laptop are crashing/rebooting when running portupgrade 
while at the same time using X. This never used to happen. To find out what 
was going on I ran several heavy computational sessions while at the same 
time portsdb -Uu was executing. Here is the backtrace from kdbg:


(kgdb) where
#0  dumpsys () at /usr/src/sys/kern/kern_shutdown.c:487
#1  0xc01abda7 in boot (howto=260) at /usr/src/sys/kern/kern_shutdown.c:316
#2  0xc01ac1e5 in panic (fmt=0xc031e8c4 "from debugger") at 
/usr/src/sys/kern/kern_shutdown.c:595
#3  0xc0134f0d in db_panic (addr=-898601789, have_addr=0, count=-1, 
modif=0xca706a34 "") at /usr/src/sys/ddb/db_command.c:435
#4  0xc0134ead in db_command (last_cmdp=0xc03703c0, cmd_table=0xc0370200, 
aux_cmd_tablep=0xc03acdf8) at /usr/src/sys/ddb/db_command.c:333
#5  0xc0134f72 in db_command_loop () at /usr/src/sys/ddb/db_command.c:457
#6  0xc01370af in db_trap (type=12, code=0) at /usr/src/sys/ddb/db_trap.c:71
#7  0xc02e57c2 in kdb_trap (type=12, code=0, regs=0xca706b64) at 
/usr/src/sys/i386/i386/db_interface.c:158
#8  0xc02f500c in trap_fatal (frame=0xca706b64, eva=14) at 
/usr/src/sys/i386/i386/trap.c:969
#9  0xc02f4a1a in trap (frame={tf_fs = 16, tf_es = 16, tf_ds = 16, tf_edi = 0, 
tf_esi = -1069833940, tf_ebp = -898602080,
      tf_isp = -898602096, tf_ebx = -898455944, tf_edx = -901929760, tf_ecx = 
1, tf_eax = 14, tf_trapno = 12, tf_err = 2,
      tf_eip = -898601789, tf_cs = 8, tf_eflags = 66118, tf_esp = -1071020156, 
tf_ss = -898455944}) at /usr/src/sys/i386/i386/trap.c:636
#10 0xca706cc3 in ?? ()
#11 0xc0298784 in vm_fault (map=Cannot access memory at address 0x1024e.
) at /usr/src/sys/vm/vm_fault.c:292
Cannot access memory at address 0x10246.
(kgdb)

 I performed a same stress test on Debian Linux and it worked flawlessly. No 
crashes. The FreeBSD crashes everytime. The computer in question is dual boot 
(Debian / FreeBSD) so both systems are using identical hardware.

 What I want to know is this happening because of the code merged from 
current? I read somwhere that they were going to try and make the accessible 
memory greater than 2 Gb. 

 Yes I know this is a newbie forum. I am a newbie and I posted to the stable 
list and got no answer. I did see many e-mails with the title "Ok, are all 
the panics fixed now?". That worries me. I want to know if they are putting 
code from CURRENT into the 4.8 STABLE?



							Kind regards
							Jonathan