Re: Regression (gdm no longer shuts down) - 2.4.24.x and 2.6.25

[Ken Moffat]

On Wed, Mar 26, 2008 at 04:19:34PM +0100, Jean Delvare wrote:
> Hi Ken,
> 
> On Tue, 25 Mar 2008 20:07:14 +0000, Ken Moffat wrote:
> > Hi,
> > 
> >  on one of my boxes, I've got a problem with gdm and kernels newer
> > than 2.6.24 (tested on 2.6.24.2, 2.6.24.4).  If I try to restart or
> > shut down from gdm, the window disappears but the X background remains
> > and the box stays in runlevel 5 until I switch to a tty and shut it
> > down (as root) or give it a 3-fingered salute to reboot.
> 
> I have to admit that I am very skeptical. The i2c-viapro driver deals
> with the motherboard's SMBus. It's not related in any way to gdm nor
> drm, so I just can't see how it could cause the problem you report.
> What gave you the idea to try unloading this driver?
 Your skepticism seems reasonable.  I tried this after I went back
to the "bad" 2.6.24.4 (to see if there was anything in the X log)
only to discover it was now shutting sown ok.  I knew I had
forgotten to change the extraversion when I reverted the patch, so
I figured it must be something in the modules (i.e. the good ones
overwrote the bad ones - dunno if they all load or not).

 After going back to 2.6.24.2 where I'd first seen this, I had a
look at what was loaded - only r8169, w83627hf with hwmon_vid, and
i2c_viapro.  I tried hwmon_vid and w83627hf but it didn't help.
Then I tried i2c_viapro and it seemed to fix it.
> 
> Do you have I2C or SMBus devices connected to the SMBus on that machine?
> If you don't know, i2cdetect should tell.

 After modprobing i2c-dev I get
root@bluesbreaker /home/ken #i2cdetect -l
i2c-0	i2c       	monid                           	I2C
adapter
i2c-1	i2c       	dvi                             	I2C
adapter
i2c-2	i2c       	vga                             	I2C
adapter
i2c-3	i2c       	crt2                            	I2C
adapter
i2c-4	smbus     	SMBus Via Pro adapter at 0400
SMBus adapter
 [ nothing shows on 1 ]
root@bluesbreaker /home/ken #i2cdetect 2
WARNING! This program can confuse your I2C bus, cause data loss and
worse!
I will probe file /dev/i2c-2.
I will probe address range 0x03-0x77.
Continue? [Y/n] 
     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f
00:          -- -- -- -- -- -- -- -- -- -- -- -- -- 
10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
50: 50 -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
70: -- -- -- -- -- -- -- --                         
 [ othing shows on 3 ]
root@bluesbreaker /home/ken #i2cdetect 4
WARNING! This program can confuse your I2C bus, cause data loss and
worse!
I will probe file /dev/i2c-4.
I will probe address range 0x03-0x77.
Continue? [Y/n] 
     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f
00:          -- -- -- -- -- -- -- -- -- -- -- -- -- 
10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 2f 
30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
50: 50 51 -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
60: -- -- -- -- -- -- -- -- -- 69 -- -- -- -- -- -- 
70: -- -- -- -- -- -- -- --                         

> 
> You say that your 32-bit and 64-bit configs are very different. It
> might be interesting to get them to match as much as possible. This
> could either point at a 64-bit specific problem, or isolate a kernel
> configuration option which causes the problem.
> 
 I'll have a look at that.

[...]
> 
> This brings a question: how many times (out of how many tries) did you
> manage to fix your 2.6.24.x kernel with "rmmod i2c-viapro"? If it's
> just 1 our of 1 try, it could be just luck? Did you try to rmmod any
> other driver?
> 
 I think it was 2 or 3 times, but as I already said, my notes say I
managed to shutdown 2.6.25-rc6-git8 once, but I can't replicate
that.

 Funnily enough, earlier today I was using what I thought was a
fixed 2.6.24.4 kernel (with the drm patches reverted) and gdm hung
again.  Maybe it is something else entirely.  All I know for certain
is that I didn't see the problem with vanilla 2.6.24.

Ken
-- 
das eine Mal als Tragödie, das andere Mal als Farce
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/