Testing - kernel-2.6.9-1.715_FC3 panics on GUI Exit

[Jim Cornette]

I just did a fresh install of FC3 and then applied the updates available 
from the FC3 testing and updates repository to my computer. I had 
several problems but the kernel panic seems to be the greatest.

The problem seems to be triggered when exiting the GUI. When booting in 
runlevel 5 and logging out of X, the hourglass stays in the center of 
the screen and the caplocks light flashes.
When booted into runlevel 3 and closing down, the text on the screen 
displays a series of numbers and states that it has a panic condition 
and the cap-locks light is flashing.

Synaptics is a problem also with the mouse not functioning, if this 
could be a related issue. The mouse was working on the runlevel 3 test 
and not functioning for the runlevel 5 error.

Jim

-- 
When you're down and out, lift up your voice and shout, "I'M DOWN AND
OUT"!

[Dan Williams]

On Wed, 2004-12-15 at 07:01 -0500, Jim Cornette wrote:
>  The problem seems to be triggered when exiting the GUI. When booting
> in 
>  runlevel 5 and logging out of X, the hourglass stays in the center of 
>  the screen and the caplocks light flashes.
>  When booted into runlevel 3 and closing down, the text on the screen 
>  displays a series of numbers and states that it has a panic condition 
>  and the cap-locks light is flashing. 
Known issue and has been fixed already in later kernels than FC3 (.906
and later I believe), and at some point I assume it will be pushed down
to FC3.

Do you happen to have a ThinkPad T31, or a Radeon graphics card?

Does your panic backtrace look anything like this?
------------------
Nov  7 22:22:30 localhost kernel: Call Trace:
Nov  7 22:22:30 localhost kernel:  [<c0156447>] handle_mm_fault+0xe5/0x233
Nov  7 22:22:30 localhost kernel:  [<c01783c3>] path_lookup+0xff/0x12f
Nov  7 22:22:30 localhost kernel:  [<c0178aa2>] open_namei+0x99/0x581
Nov  7 22:22:30 localhost kernel:  [<c01662be>] filp_open+0x23/0x3c
Nov  7 22:22:30 localhost kernel:  [<c0309b32>] __cond_resched+0x14/0x3b
Nov  7 22:22:30 localhost kernel:  [<c01e132e>] direct_strncpy_from_user+0x3e/0x5d
Nov  7 22:22:30 localhost kernel:  [<c01667e7>] sys_open+0x31/0x7d
Nov  7 22:22:30 localhost kernel:  [<c030b3c3>] syscall_call+0x7/0xb
------------------

Dan

[Alexandre Strube]

Em Qua, 2004-12-15 às 09:36 -0500, Dan Williams escreveu:
>  > The problem seems to be triggered when exiting the GUI. When
> booting in 
>  > runlevel 5 and logging out of X, the hourglass stays in the center of 
>  > the screen and the caplocks light flashes.
>  > When booted into runlevel 3 and closing down, the text on the screen 
>  > displays a series of numbers and states that it has a panic condition 
>  > and the cap-locks light is flashing.
>  Known issue and has been fixed already in later kernels than FC3 (.906
>  and later I believe), and at some point I assume it will be pushed down
>  to FC3. 
Gosh, please. I've been using my fedora3 machines logged on every time
since I realized this. The session I'm writing this is up for 6 days.

>  Do you happen to have a ThinkPad T31, or a Radeon graphics card?
No. An FIC Motherboard with VIA chipset, integrated SavagePro DDR video
chip. All of them use amd chipsets.

>  Does your panic backtrace look anything like this? 
I don't know. How do I look at it?

      * 
-- 
[]s

Alexandre Ganso
500 Four Vermelha - Diretor Steel Goose Moto Group

[Dave Jones]

On Wed, Dec 15, 2004 at 09:36:47AM -0500, Dan Williams wrote:
>  On Wed, 2004-12-15 at 07:01 -0500, Jim Cornette wrote:
>  > The problem seems to be triggered when exiting the GUI. When booting in 
>  > runlevel 5 and logging out of X, the hourglass stays in the center of 
>  > the screen and the caplocks light flashes.
>  > When booted into runlevel 3 and closing down, the text on the screen 
>  > displays a series of numbers and states that it has a panic condition 
>  > and the cap-locks light is flashing.
>  
>  Known issue and has been fixed already in later kernels than FC3 (.906
>  and later I believe), and at some point I assume it will be pushed down
>  to FC3. 
You're getting your branches mixed up 8-)  For those viewing at home,
906 was the RHEL4 kernel, FC3's is still at 715.  For all intents and
purposes, the two kernels are the same though.

		Dave

[Jim Cornette]

Dan Williams wrote:
>  On Wed, 2004-12-15 at 07:01 -0500, Jim Cornette wrote:
>  
> >The problem seems to be triggered when exiting the GUI. When booting in 
> >runlevel 5 and logging out of X, the hourglass stays in the center of 
> >the screen and the caplocks light flashes.
> >When booted into runlevel 3 and closing down, the text on the screen 
> >displays a series of numbers and states that it has a panic condition 
> >and the cap-locks light is flashing.
>  
>  
>  Known issue and has been fixed already in later kernels than FC3 (.906
>  and later I believe), and at some point I assume it will be pushed down
>  to FC3.
>  
>  Do you happen to have a ThinkPad T31, or a Radeon graphics card? 
01:05.0 VGA compatible controller: ATI Technologies Inc Radeon Mobility U1


>  
>  Does your panic backtrace look anything like this?
>  ------------------
>  Nov  7 22:22:30 localhost kernel: Call Trace:
>  Nov  7 22:22:30 localhost kernel:  [<c0156447>] handle_mm_fault+0xe5/0x233
>  Nov  7 22:22:30 localhost kernel:  [<c01783c3>] path_lookup+0xff/0x12f
>  Nov  7 22:22:30 localhost kernel:  [<c0178aa2>] open_namei+0x99/0x581
>  Nov  7 22:22:30 localhost kernel:  [<c01662be>] filp_open+0x23/0x3c
>  Nov  7 22:22:30 localhost kernel:  [<c0309b32>] __cond_resched+0x14/0x3b
>  Nov  7 22:22:30 localhost kernel:  [<c01e132e>] direct_strncpy_from_user+0x3e/0x5d
>  Nov  7 22:22:30 localhost kernel:  [<c01667e7>] sys_open+0x31/0x7d
>  Nov  7 22:22:30 localhost kernel:  [<c030b3c3>] syscall_call+0x7/0xb
>  ------------------ 
  It seemed to have large blocks of numbers, like on earlier kernels 
with debugging code present.
I logged out of X, then ran poweroff from a terminal on last logout. I 
did not have a kernel panic on poweroff, just the acpi_poweroff problem.

Jim


>  
>  Dan
>

[Alan Cox]

On Wed, Dec 15, 2004 at 11:55:05PM -0500, Jim Cornette wrote:
>  >Do you happen to have a ThinkPad T31, or a Radeon graphics card?
>  
>  01:05.0 VGA compatible controller: ATI Technologies Inc Radeon Mobility U1 
Interesting. The kernel-list has a short thread on rmap handling problems
with a test patch from Andrea and that was also ATI radeon mobility.

[Jim Cornette]

Alan Cox wrote:
>  On Wed, Dec 15, 2004 at 11:55:05PM -0500, Jim Cornette wrote:
>  
> >>Do you happen to have a ThinkPad T31, or a Radeon graphics card?
> >
> >01:05.0 VGA compatible controller: ATI Technologies Inc Radeon Mobility U1
>  
>  
>  Interesting. The kernel-list has a short thread on rmap handling problems
>  with a test patch from Andrea and that was also ATI radeon mobility.
>   
The symptoms seem to be different with each logout from the GUI. I 
logged out yesterday and the computer exited to a continuous loop with a 
similar output that Dan referred to earlier.

The last few lines in my Xorg.0.log.old file contained this excerpt in 
last two lines. I don't recall seeing any entries like this before.

(II) RADEON(0): [drm] removed 1 reserved context for kernel
(II) RADEON(0): [drm] unmapping 8192 bytes of SAREA 0x0e9c1000 at 0xf2e78000

Jim

-- 
I allow the world to live as it chooses, and I allow myself to live as I
choose.

[Davide Rossetti]

Dan Williams wrote:

> On Wed, 2004-12-15 at 07:01 -0500, Jim Cornette wrote:
>   
>
> >The problem seems to be triggered when exiting the GUI. When booting in 
> >runlevel 5 and logging out of X, the hourglass stays in the center of 
> >the screen and the caplocks light flashes.
> >When booted into runlevel 3 and closing down, the text on the screen 
> >displays a series of numbers and states that it has a panic condition 
> >and the cap-locks light is flashing.
> >    
> >
>
> Known issue and has been fixed already in later kernels than FC3 (.906
> and later I believe), and at some point I assume it will be pushed down
> to FC3.
>
> Do you happen to have a ThinkPad T31, or a Radeon graphics card?
>
> Does your panic backtrace look anything like this?
> ------------------
> Nov  7 22:22:30 localhost kernel: Call Trace:
> Nov  7 22:22:30 localhost kernel:  [<c0156447>] handle_mm_fault+0xe5/0x233
> Nov  7 22:22:30 localhost kernel:  [<c01783c3>] path_lookup+0xff/0x12f
> Nov  7 22:22:30 localhost kernel:  [<c0178aa2>] open_namei+0x99/0x581
> Nov  7 22:22:30 localhost kernel:  [<c01662be>] filp_open+0x23/0x3c
> Nov  7 22:22:30 localhost kernel:  [<c0309b32>] __cond_resched+0x14/0x3b
> Nov  7 22:22:30 localhost kernel:  [<c01e132e>] direct_strncpy_from_user+0x3e/0x5d
> Nov  7 22:22:30 localhost kernel:  [<c01667e7>] sys_open+0x31/0x7d
> Nov  7 22:22:30 localhost kernel:  [<c030b3c3>] syscall_call+0x7/0xb
> ------------------
>
>   
>  
I have an Intel 830 ASUS laptop... and keep on having problems with 
1.715 kernels... better that previous kernels (as in 
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=142240). before it 
oops'd at gdm restart. now it locks down at shutdown time with spin_lock 
msgs in console... sysrq- is not active so at next boot I loose the 
dump. let me know if I can be of any help

[Alexandre Strube]

Em Qua, 2004-12-29 às 10:14 +0100, Davide Rossetti escreveu:

>  >>The problem seems to be triggered when exiting the GUI. When
> booting in 
>  >>runlevel 5 and logging out of X, the hourglass stays in the center of 
>  >>the screen and the caplocks light flashes. 
(...)
>  I have an Intel 830 ASUS laptop... and keep on having problems with 
>  1.715 kernels... better that previous kernels (as in 
>  https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=142240). before it 
>  oops'd at gdm restart. now it locks down at shutdown time with spin_lock 
>  msgs in console... sysrq- is not active so at next boot I loose the 
>  dump. let me know if I can be of any help 
Am I correct that we already are at 2.6.9-1.698_FC3? Did you tried it?
I'm having weird behaviour with sound on fc3 kernels anyway, but that's
another thread and another list :-)

[Davide Rossetti]

Alexandre Strube wrote:

> >I have an Intel 830 ASUS laptop... and keep on having problems
> with 
> >1.715 kernels... better that previous kernels (as in 
> >https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=142240). before it 
> >oops'd at gdm restart. now it locks down at shutdown time with spin_lock 
> >msgs in console... sysrq- is not active so at next boot I loose the 
> >dump. let me know if I can be of any help
> >    
> >
>
> Am I correct that we already are at 2.6.9-1.698_FC3? Did you tried it?
>   
>  
2.6.9-1.715_FC3 is the last test kernel. and it's faulty... they say 
because it has 4G/4G option off
2.6.9-1.681_FC3 is the last update kernel. and it does well...

> I'm having weird behaviour with sound on fc3 kernels anyway, but
> that's
> another thread and another list :-)
>   
>  
me too... after install my intel i8x0 mobile chipset audio was not 
auto-detected.
regards

[Alexandre Strube]

Em Qui, 2004-12-30 às 14:48 +0100, Davide Rossetti escreveu:
>  2.6.9-1.715_FC3 is the last test kernel. and it's faulty... they
> say 
>  because it has 4G/4G option off
>  2.6.9-1.681_FC3 is the last update kernel. and it does well... 
Could you please explain it to me? You said that -715 is the last test
kernel, what does it mean? I was counting just as integers, higher is
newer :-)


>  >I'm having weird behaviour with sound on fc3 kernels anyway,
> but that's
>  >another thread and another list :-)
>  me too... after install my intel i8x0 mobile chipset audio was not 
>  auto-detected. 
I have SEVERAL machines upgraded from fedora 1 and 2. All of them
(intel, via and sis chipsets) cannot open volume control anymore, and
have no sound.

What should I do?

[Jim Cornette]

Davide Rossetti wrote:
>  Dan Williams wrote:
>  
> > On Wed, 2004-12-15 at 07:01 -0500, Jim Cornette wrote:
> >  
> >
> >> The problem seems to be triggered when exiting the GUI. When booting 
> >> in runlevel 5 and logging out of X, the hourglass stays in the center 
> >> of the screen and the caplocks light flashes.
> >> When booted into runlevel 3 and closing down, the text on the screen 
> >> displays a series of numbers and states that it has a panic condition 
> >> and the cap-locks light is flashing.
> >>   
> >
> >
> > Known issue and has been fixed already in later kernels than FC3 (.906
> > and later I believe), and at some point I assume it will be pushed down
> > to FC3.
> >
> > Do you happen to have a ThinkPad T31, or a Radeon graphics card?
> >
> > Does your panic backtrace look anything like this?
> > ------------------
> > Nov  7 22:22:30 localhost kernel: Call Trace:
> > Nov  7 22:22:30 localhost kernel:  [<c0156447>] 
> > handle_mm_fault+0xe5/0x233
> > Nov  7 22:22:30 localhost kernel:  [<c01783c3>] path_lookup+0xff/0x12f
> > Nov  7 22:22:30 localhost kernel:  [<c0178aa2>] open_namei+0x99/0x581
> > Nov  7 22:22:30 localhost kernel:  [<c01662be>] filp_open+0x23/0x3c
> > Nov  7 22:22:30 localhost kernel:  [<c0309b32>] __cond_resched+0x14/0x3b
> > Nov  7 22:22:30 localhost kernel:  [<c01e132e>] 
> > direct_strncpy_from_user+0x3e/0x5d
> > Nov  7 22:22:30 localhost kernel:  [<c01667e7>] sys_open+0x31/0x7d
> > Nov  7 22:22:30 localhost kernel:  [<c030b3c3>] syscall_call+0x7/0xb
> > ------------------
> >
> >  
> >
>  I have an Intel 830 ASUS laptop... and keep on having problems with 
>  1.715 kernels... better that previous kernels (as in 
>  https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=142240). before it 
>  oops'd at gdm restart. now it locks down at shutdown time with spin_lock 
>  msgs in console... sysrq- is not active so at next boot I loose the 
>  dump. let me know if I can be of any help
>   
I still had the problem with later than 715 kernels and am using kernel 
2.6.9-1.681_FC3. I get a lockup sometimes when ctl-alt-Fn to a terminal 
and then back to the GUI. The shutdown is alright in most cases.

Jim

-- 
Hideously disfigured by an ancient Indian curse?

		WE CAN HELP!

Call [phone] for an immediate appointment.

[Alexandre Strube]

Em Qua, 2004-12-29 às 07:04 -0500, Jim Cornette escreveu:

>  I still had the problem with later than 715 kernels and am using
> kernel 
>  2.6.9-1.681_FC3. I get a lockup sometimes when ctl-alt-Fn to a terminal 
>  and then back to the GUI. The shutdown is alright in most cases. 
Did you tried 698? Changing to kernel 698 and changing from integrated
agp savage 8x to a old pci-voodoo4 solved it to me. Don't know which one
of those solved, I did it at the same time.

[Jim Cornette]

Alexandre Strube wrote:
>  Em Qua, 2004-12-29 às 07:04 -0500, Jim Cornette escreveu:
>  
>  
> >I still had the problem with later than 715 kernels and am using kernel 
> >2.6.9-1.681_FC3. I get a lockup sometimes when ctl-alt-Fn to a terminal 
> >and then back to the GUI. The shutdown is alright in most cases.
>  
>  
>  Did you tried 698? Changing to kernel 698 and changing from integrated
>  agp savage 8x to a old pci-voodoo4 solved it to me. Don't know which one
>  of those solved, I did it at the same time.
>   
I downloaded and installed it through firefox just now. I was surprised 
that clicking on the file would launch system-config-packages and have 
automated download followed by an installation. The installation process 
was rather slow, but a welcome addition. I hope it improves for speed 
and functionality for FC4.

Thanks! Since this was a clean install, I missed the introduction of 
kernel-2.6.9-1.698_FC3 in testing. The 715 version, which crashes this 
computer on shutdown, was already in the testing repository. 698 was 
obscured by the newer kernel version being there also.

I'll see if this works without putting the computer into a panic state.

Jim

-- 
Double your disk space - delete Windows!

    -- Albert Dorofeev

[Jim Cornette]

Jim Cornette wrote:
>  Alexandre Strube wrote:
>  
> > Em Qua, 2004-12-29 às 07:04 -0500, Jim Cornette escreveu:
> >
> >
> >> I still had the problem with later than 715 kernels and am using 
> >> kernel 2.6.9-1.681_FC3. I get a lockup sometimes when ctl-alt-Fn to a 
> >> terminal and then back to the GUI. The shutdown is alright in most 
> >> cases.
> >
> >
> >
> > Did you tried 698? Changing to kernel 698 and changing from integrated
> > agp savage 8x to a old pci-voodoo4 solved it to me. Don't know which one
> > of those solved, I did it at the same time.
> > 
The computer shutdown with the 698 versioned kernel similar to the way 
with the 681 kernel. Hopefully the occasional lockup that happens when 
switching to a virtual terminal disappears with the 698 kernel. My 
concern now is with later kernels causing the kernel panic with the 
radeon graphics controller.

Jim

-- 
Walk softly and carry a megawatt laser.

[Alexandre Strube]

Em Qua, 2004-12-29 às 23:41 -0500, Jim Cornette escreveu:
>  >> Did you tried 698? Changing to kernel 698 and changing from
> integrated
>  >> agp savage 8x to a old pci-voodoo4 solved it to me. Don't know which one
>  >> of those solved, I did it at the same time.
>  The computer shutdown with the 698 versioned kernel similar to the way 
>  with the 681 kernel. Hopefully the occasional lockup that happens when 
>  switching to a virtual terminal disappears with the 698 kernel. My 
>  concern now is with later kernels causing the kernel panic with the 
>  radeon graphics controller. 
698 still has the issue of logging out to gdm.