[pvrusb2] Tracking down a problem? pt2

[Mike Isely]

On Fri, 27 Apr 2007, G Mc.Pherson wrote:

> Hi Mike,
> 
>     Just to keep you upto date, I spent last night playing with the 
> current driver and have not been able to force the problem at all. The 
> driver seems to be rock solid. Not that I expected the driver to match, 
> but I did an md5sum on Ubuntu's driver and the one I compiled and did 
> get totally different values. So I think it's safe to say that Ubuntu's 
> copy was probably an older one.

Calculating an md5sum of the .ko file won't tell you much, because 
information about the compilation environment (e.g. the specific kernel 
version string) is compiled into the object.  So there's no way the 
comparison would have matched.  Also, the driver sources are very 
different between what is in the kernel and what is in the standalone 
driver.  The standalone driver has various bits of extra logic to detect 
and adapt to the run-time module environment (but this is not so 
important for newer kernels).  The in-kernel driver removes all that 
since far better assumptions can be made about how it is being run.

The best way to compare is to run 'uname -r' on your system and then 
grab the kernel sources for the kernel you are running (using whatever 
method the distro implements for grabbing a source tree).  Then grab 
corresponding vanilla sources (based on the uname -r output) and compare 
the drivers/media/video/pvrusb2 directory from both.


> 
>     Oh and as a passing interest, Ubuntu 7.04 (Feisty) restricted kernel 
> modules package includes the firmware dumps for both the 24xxx and 29xxx 
> versions, so the device works out of the box.
> 

Nice.

  -Mike


-- 
                        |         Mike Isely          |     PGP fingerprint
     Spammers Die!!     |                             | 03 54 43 4D 75 E5 CC 92
                        |   isely @ pobox (dot) com   | 71 16 01 E2 B5 F5 C1 E8
                        |                             |