Re: kernel lockup on 6.0 beta2 on cardbus insertion on Dell Inspiron 9100 laptop

[David Young <dyoung%pobox.com@localhost>]

On Thu, Jun 07, 2012 at 05:54:52PM +0200, Martin Husemann wrote:
> On Thu, Jun 07, 2012 at 10:10:00AM -0500, David Young wrote:
> > On Thu, Jun 07, 2012 at 03:29:59AM -0700, John Refling wrote:
> > > 
> > > bar_tag=da525b1c
> > > bar_size=131072
> > > len=2048
> > > bar_memh=db230000
> > > cis_ptr=da525b20
> > > tuples=0
> 
> > It looks to me like the CIS is not inside of the address window, because
> > cis_ptr > bar_size.
> 
> John, where did you print that value? Note that cis_ptr is modified inside
> the function, and this value kinda looks like you printed it earlier.
> 
> > Maybe skip the bus_space_read_region_1(), aprint_error(), clean up and
> > return with error from cardbus_read_tuples() cis_ptr >= bar_size?
> 
> That certainly should be done, however, I wonder how the complete CIS
> looks like on this card.
> 
> John: the "512-hack" meant your modified kernel that allowed you booting, i.e.
> instead of the MIN()-term, just read 512 bytes. The way to use PCMCIA_CISDEBUG
> is: compile a kernel with options PCMCIACISDEBUG, then boot -d the kernel,
> in ddb do:
> 
>   w pcmciacis_debug=1
>   c
> 
> and then get the CIS info from dmesg. David, is there an easier method to
> check the CIS?

Not that I know of, no.

Dave

-- 
David Young
dyoung%pobox.com@localhost    Urbana, IL    (217) 721-9981