Re: Kernel Panic 2.6.14-git (pictures)

Re: Kernel Panic 2.6.14-git (pictures)

[Alejandro Bonilla]

Jesper Juhl wrote:

>On 11/9/05, Alejandro Bonilla Beeche <abonilla@linuxwireless.org> wrote:
>
>>Hi,
>>
>>    I have an IBM T42 and as always I test several kernel images to test
>>the git tree.
>>
>>Last one I did 2005-11-05 17:17 kernel-image-2.6.14_T42.v1.8_i386.deb
>>works perfectly, but the new one with same config 2005-11-09 11:47
>>kernel-image-2.6.14_T42.v1.9_i386.deb gives me a kernel panic on boot.
>>
>
>Where's the panic message?

Hi,

I collected some pictures, I have kept updating the kernel and still has the
same problem. This is not related to hard disk I think cause actually it even
loads ipw2200 and other modules and half the way done gives this problem.

I have taken some crappy pictures of the panic, I hope they help.

http://linuxwireless.org/kernel

.Alejandro

--
Open WebMail Project (http://openwebmail.org)

Re: Kernel Panic 2.6.14-git (pictures)

[Andrew Morton]

"Alejandro Bonilla" <abonilla@linuxwireless.org> wrote:
>
> I have taken some crappy pictures of the panic, I hope they help.
> 
> http://linuxwireless.org/kernel

Yes, photos of the screen work very nicely, thanks.

Hi, Robert ;)

Re: Kernel Panic 2.6.14-git (pictures)

[Dmitry Torokhov]

On Thursday 10 November 2005 20:55, Andrew Morton wrote:
> "Alejandro Bonilla" <abonilla@linuxwireless.org> wrote:
> >
> > I have taken some crappy pictures of the panic, I hope they help.
> > 
> > http://linuxwireless.org/kernel
> 
> Yes, photos of the screen work very nicely, thanks.
> 
> Hi, Robert ;)
>

Hmm, I though the following made it into Linus's tree...

-- 
Dmitry

Subject: Convert hdaps driver to dynamic input_dev allocation

Input: convert hdaps to dynamic input_dev allocation.

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---

 drivers/hwmon/hdaps.c |   41 +++++++++++++++++++++++------------------
 1 files changed, 23 insertions(+), 18 deletions(-)

Index: work/drivers/hwmon/hdaps.c
===================================================================
--- work.orig/drivers/hwmon/hdaps.c
+++ work/drivers/hwmon/hdaps.c
@@ -60,9 +60,11 @@
 
 #define HDAPS_POLL_PERIOD	(HZ/20)	/* poll for input every 1/20s */
 #define HDAPS_INPUT_FUZZ	4	/* input event threshold */
+#define HDAPS_INPUT_FLAT	4
 
 static struct timer_list hdaps_timer;
 static struct platform_device *pdev;
+static struct input_dev *hdaps_idev;
 static unsigned int hdaps_invert;
 static u8 km_activity;
 static int rest_x;
@@ -311,18 +313,6 @@ static struct device_driver hdaps_driver
 	.resume = hdaps_resume
 };
 
-/* Input class stuff */
-
-static struct input_dev hdaps_idev = {
-	.name = "hdaps",
-	.evbit = { BIT(EV_ABS) },
-	.absbit = { BIT(ABS_X) | BIT(ABS_Y) },
-	.absmin  = { [ABS_X] = -256, [ABS_Y] = -256 },
-	.absmax  = { [ABS_X] = 256, [ABS_Y] = 256 },
-	.absfuzz = { [ABS_X] = HDAPS_INPUT_FUZZ, [ABS_Y] = HDAPS_INPUT_FUZZ },
-	.absflat = { [ABS_X] = HDAPS_INPUT_FUZZ, [ABS_Y] = HDAPS_INPUT_FUZZ },
-};
-
 /*
  * hdaps_calibrate - Set our "resting" values.  Callers must hold hdaps_sem.
  */
@@ -344,9 +334,9 @@ static void hdaps_mousedev_poll(unsigned
 	if (__hdaps_read_pair(HDAPS_PORT_XPOS, HDAPS_PORT_YPOS, &x, &y))
 		goto out;
 
-	input_report_abs(&hdaps_idev, ABS_X, x - rest_x);
-	input_report_abs(&hdaps_idev, ABS_Y, y - rest_y);
-	input_sync(&hdaps_idev);
+	input_report_abs(hdaps_idev, ABS_X, x - rest_x);
+	input_report_abs(hdaps_idev, ABS_Y, y - rest_y);
+	input_sync(hdaps_idev);
 
 	mod_timer(&hdaps_timer, jiffies + HDAPS_POLL_PERIOD);
 
@@ -566,12 +556,25 @@ static int __init hdaps_init(void)
 	if (ret)
 		goto out_device;
 
+	hdaps_idev = input_allocate_device();
+	if (!hdaps_idev) {
+		ret = -ENOMEM;
+		goto out_group;
+	}
+
 	/* initial calibrate for the input device */
 	hdaps_calibrate();
 
 	/* initialize the input class */
-	hdaps_idev.dev = &pdev->dev;
-	input_register_device(&hdaps_idev);
+	hdaps_idev->name = "hdaps";
+	hdaps_idev->cdev.dev = &pdev->dev;
+	hdaps_idev->evbit[0] = BIT(EV_ABS);
+	input_set_abs_params(hdaps_idev, ABS_X,
+			-256, 256, HDAPS_INPUT_FUZZ, HDAPS_INPUT_FLAT);
+	input_set_abs_params(hdaps_idev, ABS_X,
+			-256, 256, HDAPS_INPUT_FUZZ, HDAPS_INPUT_FLAT);
+
+	input_register_device(hdaps_idev);
 
 	/* start up our timer for the input device */
 	init_timer(&hdaps_timer);
@@ -582,6 +585,8 @@ static int __init hdaps_init(void)
 	printk(KERN_INFO "hdaps: driver successfully loaded.\n");
 	return 0;
 
+out_group:
+	sysfs_remove_group(&pdev->dev.kobj, &hdaps_attribute_group);
 out_device:
 	platform_device_unregister(pdev);
 out_driver:
@@ -596,7 +601,7 @@ out:
 static void __exit hdaps_exit(void)
 {
 	del_timer_sync(&hdaps_timer);
-	input_unregister_device(&hdaps_idev);
+	input_unregister_device(hdaps_idev);
 	sysfs_remove_group(&pdev->dev.kobj, &hdaps_attribute_group);
 	platform_device_unregister(pdev);
 	driver_unregister(&hdaps_driver);

Re: Kernel Panic 2.6.14-git (pictures)

[Robert Love]

On Thu, 2005-11-10 at 17:55 -0800, Andrew Morton wrote:

> Yes, photos of the screen work very nicely, thanks.

Excellent!

> Hi, Robert ;)

Andrew.  ;-)

Alejandro - there is Dmitry Torokhov that fixed this and I thought it
went in Linus's tree (it was in Greg's tree and he pushed it over a week
ago).

Dmitry?

	Robert Love

Re: Kernel Panic 2.6.14-git (pictures)

[Alejandro Bonilla Beeche]

Robert Love wrote:

>On Thu, 2005-11-10 at 17:55 -0800, Andrew Morton wrote:
>
>  
>
>>Yes, photos of the screen work very nicely, thanks.
>>    
>>
>
>Excellent!
>
>  
>
>>Hi, Robert ;)
>>    
>>
>
>Andrew.  ;-)
>
>Alejandro - there is Dmitry Torokhov that fixed this and I thought it
>went in Linus's tree (it was in Greg's tree and he pushed it over a week
>ago).
>
>Dmitry?
>  
>
LOL, everyone is pointing to someone else... ;-)

Linus, can you please merge Dmitry's patch? ;o

.Alejandro

>	Robert Love
>
>
>
>  
>

Re: Kernel Panic 2.6.14-git (pictures)

[Dmitry Torokhov]

On Friday 11 November 2005 01:22, Alejandro Bonilla Beeche wrote:
> Robert Love wrote:
> 
> >On Thu, 2005-11-10 at 17:55 -0800, Andrew Morton wrote:
> >
> >  
> >
> >>Yes, photos of the screen work very nicely, thanks.
> >>    
> >>
> >
> >Excellent!
> >
> >  
> >
> >>Hi, Robert ;)
> >>    
> >>
> >
> >Andrew.  ;-)
> >
> >Alejandro - there is Dmitry Torokhov that fixed this and I thought it
> >went in Linus's tree (it was in Greg's tree and he pushed it over a week
> >ago).
> >
> >Dmitry?
> >  
> >
> LOL, everyone is pointing to someone else... ;-)
>

Heh ;)

Alejandro, didn't you have another issue with pcspeaker driver in
conjunction with PNP? Did my patch moving inpu core into a separate
directory and registering it early help?

 
> Linus, can you please merge Dmitry's patch? ;o
> 

Thank you for testing it!

-- 
Dmitry

Re: Kernel Panic 2.6.14-git (pictures)

[Alejandro Bonilla Beeche]

Dmitry Torokhov wrote:

>On Friday 11 November 2005 01:22, Alejandro Bonilla Beeche wrote:
>  
>
>>Robert Love wrote:
>>
>>    
>>
>>>On Thu, 2005-11-10 at 17:55 -0800, Andrew Morton wrote:
>>>
>>> 
>>>
>>>      
>>>
>>>>Yes, photos of the screen work very nicely, thanks.
>>>>   
>>>>
>>>>        
>>>>
>>>Excellent!
>>>
>>> 
>>>
>>>      
>>>
>>>>Hi, Robert ;)
>>>>   
>>>>
>>>>        
>>>>
>>>Andrew.  ;-)
>>>
>>>Alejandro - there is Dmitry Torokhov that fixed this and I thought it
>>>went in Linus's tree (it was in Greg's tree and he pushed it over a week
>>>ago).
>>>
>>>Dmitry?
>>> 
>>>
>>>      
>>>
>>LOL, everyone is pointing to someone else... ;-)
>>
>>    
>>
>
>Heh ;)
>
>Alejandro, didn't you have another issue with pcspeaker driver in
>conjunction with PNP? Did my patch moving inpu core into a separate
>directory and registering it early help?
>  
>
That wasn't me. ;-| It has always been about this hdaps module I 
think... now it became a panic, before was just an oops.

> 
>  
>
>>Linus, can you please merge Dmitry's patch? ;o
>>
>>    
>>
>
>Thank you for testing it!
>  
>
Sure, I gotta help in something! ;-)

.Alejandro

Re: Kernel Panic 2.6.14-git (pictures)

[Diego Calleja]

El Thu, 10 Nov 2005 17:55:22 -0800,
Andrew Morton <akpm@osdl.org> escribió:

> Yes, photos of the screen work very nicely, thanks.

So, it may be aceptable as "official method"?

--- stable/Documentation/oops-tracing.txt.old	2005-11-11 11:54:01.000000000 +0100
+++ stable/Documentation/oops-tracing.txt	2005-11-11 11:59:42.000000000 +0100
@@ -30,7 +30,9 @@ the disk is not available then you have 
 
 (1) Hand copy the text from the screen and type it in after the machine
     has restarted.  Messy but it is the only option if you have not
-    planned for a crash.
+    planned for a crash. Alternatively, you can take a picture of
+    the screen with a digital camera - not nice, but better than
+    nothing.
 
 (2) Boot with a serial console (see Documentation/serial-console.txt),
     run a null modem to a second machine and capture the output there

Re: Kernel Panic 2.6.14-git (pictures)

[Alejandro Bonilla]

Dmitry Torokhov wrote:
>Heh ;)
>
>Alejandro, didn't you have another issue with pcspeaker driver in
>conjunction with PNP? Did my patch moving inpu core into a separate
>directory and registering it early help?
>
> 
>
>>Linus, can you please merge Dmitry's patch? ;o
>>
>
>Thank you for testing it!

Dmitry,

It works perfectly now. Thanks for the patch and for merging.

.Alejandro

>