headers
smb | 1 Feb 2008 03:05

kern/37933: very frequent crashes after suspend/resume

>Number:         37933
>Category:       kern
>Synopsis:       very frequent crashes after suspend/resume
>Confidential:   no
>Severity:       serious
>Priority:       high
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Fri Feb 01 02:05:01 +0000 2008
>Originator:     Steven M. Bellovin
>Release:        NetBSD 4.99.51
>Organization:
Department of Computer Science, Columbia University
>Environment:
	
	
System: NetBSD yellowstone.machshav.com 4.99.51 NetBSD 4.99.51 (YELLOWSTONE) #6: Wed Jan 30 20:05:24
EST 2008 smb <at> yellowstone.machshav.com:/usr/BUILD/obj/sys/arch/amd64/compile/YELLOWSTONE amd64
Architecture: x86_64
Machine: amd64
>Description:
	I see very frequent crashes on resume after an (ACPI) suspend.
	I can reproduce it almost invariably by booting single-user and
	doing
		sysctl -w machdep.sleep_state=3
	twice (on occasion three times).  It very frequently happens in
	normal use, too, when I'm in X.
	The panics are usually in uvm_fault or amap_wipeout.
(Continue reading)

Permalink | Reply | 1 comment |
headers
Jared D. McNeill | 1 Feb 2008 03:12
Picon

Re: kern/37933: very frequent crashes after suspend/resume

smb <at> cs.columbia.edu wrote:
> 	The panics are usually in uvm_fault or amap_wipeout.

How about a backtrace? sysctl -w ddb.commandonenter=tr
Permalink | Reply | 0 comments |
headers
Jared D. McNeill | 1 Feb 2008 04:35
Picon

Re: kern/37933: very frequent crashes after suspend/resume

The following reply was made to PR kern/37933; it has been noted by GNATS.

From: "Jared D. McNeill" <jmcneill <at> invisible.ca>
To: gnats-bugs <at> NetBSD.org
Cc: kern-bug-people <at> netbsd.org, gnats-admin <at> netbsd.org, 
 netbsd-bugs <at> netbsd.org
Subject: Re: kern/37933: very frequent crashes after suspend/resume
Date: Thu, 31 Jan 2008 21:12:35 -0500

 smb <at> cs.columbia.edu wrote:
 > 	The panics are usually in uvm_fault or amap_wipeout.

 How about a backtrace? sysctl -w ddb.commandonenter=tr
Permalink | Reply | 2 comments |
headers
Steven M. Bellovin | 1 Feb 2008 04:54

Re: kern/37933: very frequent crashes after suspend/resume

On Fri,  1 Feb 2008 03:35:03 +0000 (UTC)
"Jared D. McNeill" <jmcneill <at> invisible.ca> wrote:

> The following reply was made to PR kern/37933; it has been noted by
> GNATS.
> 
> From: "Jared D. McNeill" <jmcneill <at> invisible.ca>
> To: gnats-bugs <at> NetBSD.org
> Cc: kern-bug-people <at> netbsd.org, gnats-admin <at> netbsd.org, 
>  netbsd-bugs <at> netbsd.org
> Subject: Re: kern/37933: very frequent crashes after suspend/resume
> Date: Thu, 31 Jan 2008 21:12:35 -0500
> 
>  smb <at> cs.columbia.edu wrote:
>  > 	The panics are usually in uvm_fault or amap_wipeout.
>  
>  How about a backtrace? sysctl -w ddb.commandonenter=tr
>  
I've been unable to get anything meaningful, including any usable dumps
(or indeed, any dumps, most of the time).  On the occasions when I get
a command prompt (when in multiuser, I have ddb.onpanic=0 set -- and I
don't get dumps), I've tried 'reboot 0x101', and it just hangs.  bt
hasn't given anything useful, either.  But I'll try the command you
suggest and see if anything better happens.

		--Steve Bellovin, http://www.cs.columbia.edu/~smb
Permalink | Reply | 0 comments |
headers
Steven M. Bellovin | 1 Feb 2008 04:55

Re: kern/37933: very frequent crashes after suspend/resume

The following reply was made to PR kern/37933; it has been noted by GNATS.

From: "Steven M. Bellovin" <smb <at> cs.columbia.edu>
To: gnats-bugs <at> NetBSD.org
Cc: jmcneill <at> invisible.ca, kern-bug-people <at> NetBSD.org,
 gnats-admin <at> NetBSD.org, netbsd-bugs <at> NetBSD.org
Subject: Re: kern/37933: very frequent crashes after suspend/resume
Date: Fri, 1 Feb 2008 03:54:24 +0000

 On Fri,  1 Feb 2008 03:35:03 +0000 (UTC)
 "Jared D. McNeill" <jmcneill <at> invisible.ca> wrote:

 > The following reply was made to PR kern/37933; it has been noted by
 > GNATS.
 > 
 > From: "Jared D. McNeill" <jmcneill <at> invisible.ca>
 > To: gnats-bugs <at> NetBSD.org
 > Cc: kern-bug-people <at> netbsd.org, gnats-admin <at> netbsd.org, 
 >  netbsd-bugs <at> netbsd.org
 > Subject: Re: kern/37933: very frequent crashes after suspend/resume
 > Date: Thu, 31 Jan 2008 21:12:35 -0500
 > 
 >  smb <at> cs.columbia.edu wrote:
 >  > 	The panics are usually in uvm_fault or amap_wipeout.
 >  
 >  How about a backtrace? sysctl -w ddb.commandonenter=tr
 >  
 I've been unable to get anything meaningful, including any usable dumps
 (or indeed, any dumps, most of the time).  On the occasions when I get
 a command prompt (when in multiuser, I have ddb.onpanic=0 set -- and I
(Continue reading)

Permalink | Reply | 0 comments |
headers
gcw | 1 Feb 2008 05:05
Picon

kern/37934: Interrupt type ugen devices crash kernel on access close

>Number:         37934
>Category:       kern
>Synopsis:       Extra clfree() crashes kernel in ugenclose() for interrupt ugen devices
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Fri Feb 01 04:05:00 +0000 2008
>Originator:     Geoff C. Wing
>Release:        NetBSD 4.99.52
>Organization:
>Environment:
System: NetBSD g.primenet.com.au 4.99.52 NetBSD 4.99.52 (G) #0: Fri Feb 1 14:46:50 EST 2008
gcw <at> g.primenet.com.au:/usr/netbsd/src/sys/arch/i386/compile/G i386
Architecture: i386
Machine: i386
>Description:
	Only interrupt transfer type USB generic device attachments
	call clalloc(), however on close it correctly calls clfree()
	in sys/dev/usb/ugen.c:536 then incorrectly calls it again
	at line 556.  This probably also affects isochronous type
	transfers since they do not clalloc().
>How-To-Repeat:
	Plug in something which uses this - I ran apcupsd with an APC UPS -
	then stop the program.  See computer crash.
>Fix:
(Continue reading)

Permalink | Reply | 0 comments |
headers
Steven M. Bellovin | 1 Feb 2008 05:20

Re: kern/37933: very frequent crashes after suspend/resume

On Fri,  1 Feb 2008 03:35:03 +0000 (UTC)
"Jared D. McNeill" <jmcneill <at> invisible.ca> wrote:

> The following reply was made to PR kern/37933; it has been noted by
> GNATS.
> 
> From: "Jared D. McNeill" <jmcneill <at> invisible.ca>
> To: gnats-bugs <at> NetBSD.org
> Cc: kern-bug-people <at> netbsd.org, gnats-admin <at> netbsd.org, 
>  netbsd-bugs <at> netbsd.org
> Subject: Re: kern/37933: very frequent crashes after suspend/resume
> Date: Thu, 31 Jan 2008 21:12:35 -0500
> 
>  smb <at> cs.columbia.edu wrote:
>  > 	The panics are usually in uvm_fault or amap_wipeout.
>  
>  How about a backtrace? sysctl -w ddb.commandonenter=tr
>  

kernel:protection fault trap, code=0
Stopped in pid 8.1 (sysctl) at amap_wipeout+0x82:  movl 0(%r13),%esi

Beyond that:

uvm_unmap_detach()+0x45
uvmspace_free()+0xe4
exit1()+0x236
sys_exit()+0x67
syscall()+0xa9
(Continue reading)

Permalink | Reply | 0 comments |
headers
Steven M. Bellovin | 1 Feb 2008 05:25

Re: kern/37933: very frequent crashes after suspend/resume

The following reply was made to PR kern/37933; it has been noted by GNATS.

From: "Steven M. Bellovin" <smb <at> cs.columbia.edu>
To: gnats-bugs <at> NetBSD.org
Cc: jmcneill <at> invisible.ca, kern-bug-people <at> NetBSD.org, gnats-admin <at> NetBSD.org,
        netbsd-bugs <at> NetBSD.org
Subject: Re: kern/37933: very frequent crashes after suspend/resume
Date: Thu, 31 Jan 2008 23:20:46 -0500

 On Fri,  1 Feb 2008 03:35:03 +0000 (UTC)
 "Jared D. McNeill" <jmcneill <at> invisible.ca> wrote:

 > The following reply was made to PR kern/37933; it has been noted by
 > GNATS.
 > 
 > From: "Jared D. McNeill" <jmcneill <at> invisible.ca>
 > To: gnats-bugs <at> NetBSD.org
 > Cc: kern-bug-people <at> netbsd.org, gnats-admin <at> netbsd.org, 
 >  netbsd-bugs <at> netbsd.org
 > Subject: Re: kern/37933: very frequent crashes after suspend/resume
 > Date: Thu, 31 Jan 2008 21:12:35 -0500
 > 
 >  smb <at> cs.columbia.edu wrote:
 >  > 	The panics are usually in uvm_fault or amap_wipeout.
 >  
 >  How about a backtrace? sysctl -w ddb.commandonenter=tr
 >  

 
 kernel:protection fault trap, code=0
(Continue reading)

Permalink | Reply | 0 comments |
headers
Geoff Wing | 1 Feb 2008 05:30
Picon
Favicon

Re: kern/37375: missing __KERNEL_RCSID tags in sys/dist/ipf

The following reply was made to PR kern/37375; it has been noted by GNATS.

From: Geoff Wing <gcw <at> pobox.com>
To: NetBSD GNATS <gnats-bugs <at> netbsd.org>
Cc: 
Subject: Re: kern/37375: missing __KERNEL_RCSID tags in sys/dist/ipf
Date: Fri, 1 Feb 2008 15:26:30 +1100

 Fixed.  Please close PR.

 Thank you,
 Geoff
Permalink | Reply | 0 comments |
headers
lukem | 1 Feb 2008 05:44
Picon

Re: kern/37375 (missing __KERNEL_RCSID tags in sys/dist/ipf)

Synopsis: missing __KERNEL_RCSID tags in sys/dist/ipf

State-Changed-From-To: open->closed
State-Changed-By: lukem <at> narn.netbsd.org
State-Changed-When: Fri, 01 Feb 2008 04:44:05 +0000
State-Changed-Why:
Submitted says the problem is fixed.
Infact, I think I fixed it, without realising this PR was here ...
Permalink | Reply | 0 comments |

Gmane