[2.6.10-rc1] Segmentation fault in program "X"

[2.6.10-rc1] Segmentation fault in program "X"

[Thomas Meyer]

Hello.

X doesn't work under 2.6.10-rc1. i'm using the framebuffer X server. 
Kernel 2.6.9 works. How could that be?

with kind regards
Thomas

Re: [2.6.10-rc1] Segmentation fault in program "X"

[Denis Vlasenko]

On Sunday 24 October 2004 11:38, Thomas Meyer wrote:
> Hello.
> 
> X doesn't work under 2.6.10-rc1. i'm using the framebuffer X server. 
> Kernel 2.6.9 works. How could that be?

Details?
--
vda

Re: [2.6.10-rc1] Segmentation fault in program "X"

[Thomas Meyer]

Denis Vlasenko wrote:
> On Sunday 24 October 2004 11:38, Thomas Meyer wrote:
> 
>>Hello.
>>
>>X doesn't work under 2.6.10-rc1. i'm using the framebuffer X server. 
>>Kernel 2.6.9 works. How could that be?
> 
> 
> Details?
> --
> vda
> 
> 

Hi.

Signal SIGSEGV happens while doing sys function
"ioctl(5, FBIOBLANK <unfinished ...>"

seems to be some changes between 2.6.9 and 2.6.10-rc1 in file "fbmem.c"

with kind regards
Thomas

Re: [2.6.10-rc1] Segmentation fault in program "X"

[Linus Torvalds]

On Sun, 24 Oct 2004, Thomas Meyer wrote:
> 
> Denis Vlasenko wrote:
> > On Sunday 24 October 2004 11:38, Thomas Meyer wrote:
> > 
> >>Hello.
> >>
> >>X doesn't work under 2.6.10-rc1. i'm using the framebuffer X server. 
> >>Kernel 2.6.9 works. How could that be?
> > 
> > 
> > Details?
> > --
> > vda
> > 
> > 
> 
> Hi.
> 
> Signal SIGSEGV happens while doing sys function
> "ioctl(5, FBIOBLANK <unfinished ...>"
> 
> seems to be some changes between 2.6.9 and 2.6.10-rc1 in file "fbmem.c"

Do you have radeon hardware? Is there any oops in your logs?

It definitely sounds like it's due to this one:

	ChangeSet@1.1988.69.186, 2004-10-19 08:09:17-07:00, benh@kernel.crashing.org
	  [PATCH] rework radeonfb blanking
	  
	  This patch cleans up some old cruft in the manipulation of the LVDS
	  interface registers and fixes the blanking code to work with various DVI
	  flat panels.
	  
	  Since this is all very sensitive stuff, I'm posting the patch here for
	  testing before submitting it upstream, though Andrew is welcome to put it
	  in -mm.
	  
	  It also fix some problems with getting the right PLL setup on recent Mac
	  laptops, replacing the old hard coded list of values with cleaner code that
	  "probes" the PLL setup done by the firmware.
	  
	  Signed-off-by: Andrew Morton <akpm@osdl.org>
	  Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Ben, any suggestions for Thomas? 

		Linus

Re: [2.6.10-rc1] Segmentation fault in program "X"

[Benjamin Herrenschmidt]

On Mon, 2004-10-25 at 06:24, Linus Torvalds wrote:

> > Signal SIGSEGV happens while doing sys function
> > "ioctl(5, FBIOBLANK <unfinished ...>"
> > 
> > seems to be some changes between 2.6.9 and 2.6.10-rc1 in file "fbmem.c"
> 
> Do you have radeon hardware? Is there any oops in your logs?

A Oops log would be useful...

I don't see anything, are you sure he is using radeonfb ? Look at the
fix posted by Tony Dapalas today fixing a possible Oops on blank for
fbdev's that have no blank() callback ...

Ben.