Date: Fri, 27 Jan 2006 17:45:15 -0500 From: John Baldwin <jhb@freebsd.org> To: src-committers@freebsd.org Cc: cvs-src@freebsd.org, cvs-all@freebsd.org Subject: Re: cvs commit: src/sys/i386/i386 trap.c src/sys/amd64/amd64 trap.c Message-ID: <200601271745.17186.jhb@freebsd.org> In-Reply-To: <200601272222.k0RMMAoR062022@repoman.freebsd.org> References: <200601272222.k0RMMAoR062022@repoman.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Friday 27 January 2006 17:22, John Baldwin wrote:
> jhb 2006-01-27 22:22:10 UTC
>
> FreeBSD src repository
>
> Modified files:
> sys/i386/i386 trap.c
> sys/amd64/amd64 trap.c
> Log:
> Call WITNESS_CHECK() in the page fault handler and immediately assume it
> is a fatal fault if we are holding any non-sleepable locks. This should
> cut down on the number of bogus LORs we currently get when the kernel
> panics due to a NULL (or bogus) pointer dereference that goes wandering
> off into the VM system which tries to acquire locks and then kicks off
> the spurious LORs. This should probably be ported to all the archs at
> some point.
>
> Tested on: i386
Sample output for a fault with a mutex held:
32 fault with a mutex held
# sysctl debug.crash.test=32
debug.crash.testc: r0as -> h: fa0ult
with a mutex held
Kern# el page fault with the following non-sleepable locks held:
exclusive sleep mutex test r = 0 (0xc4c318d4) locked @ crash.c:95
KDB: stack backtrace:
kdb_backtrace(c0811c3c,e00dabe4,1,c4774780,c4777678) at kdb_backtrace+0x2e
witness_warn(5,0,c0773810,c4774780,0) at witness_warn+0x1d3
trap(c0540008,c4c30028,28,c4c2f910,21) at trap+0x13e
calltrap() at calltrap+0x5
--- trap 0xc, eip = 0xc4c2e1db, esp = 0xe00dacb0, ebp = 0xe00dacc4 ---
fault_with_lock(c4c304e9,c4c2ffbf,c4c2ffd7,255,0) at fault_with_lock+0x2b
crash_thread(0,e00dad38,c074f249,30e,c4774780) at crash_thread+0xdd
fork_exit(c4c2f910,0,e00dad38) at fork_exit+0x117
fork_trampoline() at fork_trampoline+0x8
--- trap 0x1, eip = 0, esp = 0xe00dad6c, ebp = 0 ---
Fatal trap 12: page fault while in kernel mode
cpuid = 3; apic id = 03
fault virtual address = 0x0
fault code = supervisor read, page not present
instruction pointer = 0x20:0xc4c2e1db
stack pointer = 0x28:0xe00dacb0
frame pointer = 0x28:0xe00dacc4
code segment = base 0x0, limit 0xfffff, type 0x1b
= DPL 0, pres 1, def32 1, gran 1
processor eflags = interrupt enabled, resume, IOPL = 0
current process = 1325 (crash)
[thread pid 1325 tid 100087 ]
Stopped at fault_with_lock+0x2b: movsbl 0,%ebx
db>
--
John Baldwin <jhb@FreeBSD.org> <>< http://www.FreeBSD.org/~jhb/
"Power Users Use the Power to Serve" = http://www.FreeBSD.org
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200601271745.17186.jhb>