Date: Sat, 8 Jul 1995 11:40:25 -0700 From: Matt Dillon <root@best.com> To: dima@blob.best.net Cc: bugs@freebsd.org Subject: Shell crash dump 10:45 a.m. sat (obvious problem) Message-ID: <199507081840.LAA25657@blob.best.net>
index | next in thread | raw e-mail
The crash dump shows:
#0 boot (arghowto=256) at ../../i386/i386/machdep.c:866
#1 0xf0110543 in panic (fmt=0xf01a1ece "page fault")
at ../../kern/subr_prf.c:128
#2 0xf01a297e in trap_fatal (frame=0xefbffcfc) at ../../i386/i386/trap.c:688
#3 0xf01a24f0 in trap_pfault (frame=0xefbffcfc, usermode=0)
at ../../i386/i386/trap.c:610
#4 0xf01a21b7 in trap (frame={tf_es = 16, tf_ds = 16, tf_edi = 188480,
tf_esi = 249920, tf_ebp = -272630088, tf_isp = -272630492,
tf_ebx = -559038242, tf_edx = -162992128, tf_ecx = 0, tf_eax = 0,
tf_trapno = 12, tf_err = 0, tf_eip = -267329968, tf_cs = 8,
tf_eflags = 66182, tf_esp = 61440, tf_ss = 0})
at ../../i386/i386/trap.c:290
#5 0xf019b461 in calltrap ()
#6 0xf010d96c in kern_sysctl (name=0xefbfff2c, namelen=3, oldp=0xf000,
oldlenp=0xefbfff24, newp=0x0, newlen=0, p=0xf269f900)
at ../../kern/kern_sysctl.c:275
#7 0xf010d64a in __sysctl (p=0xf269f900, uap=0xefbfff94, retval=0xefbfff8c)
at ../../kern/kern_sysctl.c:149
#8 0xf01a2b67 in syscall (frame={tf_es = 39, tf_ds = 39, tf_edi = -272639240,
...
However, the actual crash occured on line 634 of kern/kern_sysctl.c:
(kgdb) frame 2
#2 0xf01a297e in trap_fatal (frame=0xefbffcfc) at ../../i386/i386/trap.c:688
../../i386/i386/trap.c:688: No such file or directory.
(kgdb) print eva
$9 = -268981392
(kgdb) x/i eva
0xeff7ab70: Cannot access memory at address 0xeff7ab70.
(kgdb) print frame->tf_eip
$10 = -267329968
(kgdb) x/i frame->tf_eip
0xf010de50 <sysctl_doproc+100>: cmpb $0x1,0x2c(%ebx)
(kgdb)
I believe the problem is that the copyout() in the loop pagefaults and the process
being scanned goes away in the mean time, so when we come back p is invalid.
-Matt
help
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199507081840.LAA25657>