Re: Laptop w/ dark screen & no (working) keyboard

On Dec 30, 2008, at 9:09, David Wolfskill <david_at_catwhisker.org> wrote:

> On Tue, Dec 09, 2008 at 03:22:09PM -0800, David Wolfskill wrote:
>> After a reboot -- which exhibited the same symptoms [dark screen & no
>> (working) keyboard] initially:
>> ...
>
> OK; this has been intermittently ongoing since I reported it.  I had
> thought that there might be some relationship with the "interrupt  
> storm"
> I saw with a Cisco 350 wireless card, but this morning, after building
> CURRENT fresh and rebooting without that card, but using a serial
> console, the "dark screen & no (working) keyboard" appears to be solid
> and reproducible.
>
> On the serial console (mind, this is on a laptop, so disabling the
> internal keyboard is something that requires invasive surgery, as  
> far as
> I know), I see:
>
> login: lock order reversal:
> 1st 0xd8cfd530 bufwait (bufwait) _at_ /usr/src/sys/kern/vfs_bio.c:2443
> 2nd 0xc5165a00 dirhash (dirhash) _at_ /usr/src/sys/ufs/ufs/ 
> ufs_dirhash.c:263
> KDB: stack backtrace:
> db_trace_self_wrapper(c0bb5f22,c4bf2898,c0850065,4,c0bb1422,...) at  
> db_trace_sel
> f_wrapper+0x26
> kdb_backtrace(4,c0bb1422,c4d23810,c4d26e80,c4bf28f4,...) at  
> kdb_backtrace+0x29
> _witness_debugger(c0bb8c0c,c5165a00,c0bd944d,c4d26e80,c0bd90e6,...)  
> at _witness_
> debugger+0x25
> witness_checkorder(c5165a00,9,c0bd90e6,107,0,...) at  
> witness_checkorder+0x839
> _sx_xlock(c5165a00,0,c0bd90e6,107,d93bb1d8,...) at _sx_xlock+0x85
> ufsdirhash_acquire(0,e,c5131000,d8cfd4d0,d93bb1d8,...) at  
> ufsdirhash_acquire+0x3
> 5
> ufsdirhash_remove(c5216e10,d93bb1d8,1d8,c4bf2984,c4bf2980,...) at  
> ufsdirhash_rem
> ove+0x14
> ufs_dirremove(c521096c,c570ea50,500940c,0,0,...) at ufs_dirremove+0xe5
> ufs_rename(c4bf2c1c,c4bf2c1c,c4bf2bcc,c4bf2b7c,c4bf2bcc,...) at  
> ufs_rename+0xbe3
>
> VOP_RENAME_APV(c0cbdc40,c4bf2c1c,101,0,5009410,...) at VOP_RENAME_APV 
> +0xa5
> kern_renameat(c5558000,ffffff9c,81e00e0,ffffff9c,81e0100,...) at  
> kern_renameat+0
> x2b7
> kern_rename(c5558000,81e00e0,81e0100,0,c4bf2d2c,...) at kern_rename 
> +0x36
> rename(c5558000,c4bf2cf8,8,c0bb9477,c0c986c0,...) at rename+0x29
> syscall(c4bf2d38) at syscall+0x2a3
> Xint0x80_syscall() at Xint0x80_syscall+0x20
> --- syscall (128, FreeBSD ELF32, rename), eip = 0x283c9483, esp =  
> 0xbfbfe8cc, eb
> p = 0xbfbfe988 ---
>
> then the laptop's screen blanks, and on the serial console, I see:
>
> drm0: <ATI Radeon Lf RV250 Mobility 9000 M9 / FireMV 2400 PCI> on  
> vgapci0
> vgapci0: child drm0 requested pci_enable_busmaster
> info: [drm] AGP at 0xe8000000 64MB
> info: [drm] Initialized radeon 1.29.0 20080613
> vgapci0: Reserved 0x10000 bytes for rid 0x18 type 3 at 0xfcff0000
> vgapci0: Reserved 0x8000000 bytes for rid 0x10 type 3 at 0xe0000000
>
> and the screen, rather than flickering & coming to life in graphics
> mode, stays dark.
>
> Hmm... I just saw (on the serial console):
>
> ACPI Exception (utmutex-0376): AE_TIME, Thread 186AD could not  
> acquire Mutex [0]
> [20070320]
> ACPI Error (exutils-0180): Could not acquire AML Interpreter mutex  
> [20070320]
> ACPI Error (utmutex-0421): Mutex [0] is not acquired, cannot release  
> [20070320]
> ACPI Error (exutils-0250): Could not release AML Interpreter mutex  
> [20070320]
>
> and the screen flickered and finally(!) -- after about 20 minutes
> -- came to life and things look "normal" now.
>
> During the time that the screen stayed dark, I couldn't login via
> the serial console, either; now that the screen is lit up, I have
> a login: prompt that works (on the serial console).
>
> The one thing  I can do when the screen is dark is hit Ctl+Alt+Esc on
> the keyboard & break into the debugger on the serial console (well, I
> can use the keyboard, but with a dark screen, it's hard to be sure  
> what
> I type, let alone what the responses are).
>
> Is there any information I can obtain & provide that might be useful  
> in
> debugging this?
>
> The machine is running:
>
> d252(8.0-C)[1] uname -v
> FreeBSD 8.0-CURRENT #908: Tue Dec 30 07:48:07 PST 2008     root_at_g1-37.catwhisker
> .org:/common/S4/obj/usr/src/sys/CANARY
> d252(8.0-C)[2]
>
> Sources were last updated:
> CVSup begin from cvsup2.freebsd.org at Tue Dec 30 03:37:33 PST 2008
> CVSup ended from cvsup2.freebsd.org at Tue Dec 30 03:44:27 PST 2008
>
> Thanks.
>
> Peace,
> david

Hi David,
     The LOR's you reported are known.
Cheers,
-Garrett