LG violates gpl? take 2

[Kees Bakker]

On Mar 24, 2010, at 6:43 PM, Luca Olivetti wrote:

> En/na Luca Olivetti ha escrit:
>> En/na Benny Amorsen ha escrit:
> 
>>> If you do make clean and rebuild, do you get a different image? It's
>>> difficult to get bit-for-bit reproducable builds.

I've done it for KiSS DP-558. It can be done, but it takes a lot
of time to figure out exactly where the differences are.

>> No, I didn't try that (yet), but one thing is some bytes out of sync due to timestamping, another is 50% of differences.
> 
> I did now, and effectively the only difference is the timestamp:
> 
> $ cmp -lb vmlinuz.config-flash vmlinuz
> 2490567  64 4     65 5
> 2490577 124 T    127 W
> 2490578 165 u    145 e
> 2490579 145 e    144 d
> 2490586  63 3     64 4
> 2490589  67 7     70 8
> 2490591  64 4     63 3
> 2490592  62 2     63 3
> 2490594  63 3     65 5
> 2490595  62 2     64 4
> 3159419  64 4     65 5
> 3159429 124 T    127 W
> 3159430 165 u    145 e
> 3159431 145 e    144 d
> 3159438  63 3     64 4
> 3159441  67 7     70 8
> 3159443  64 4     63 3
> 3159444  62 2     63 3
> 3159446  63 3     65 5
> 3159447  62 2     64 4

Why don't you do a "strings" on both kernels and diff the output of that? That is
a simpler first step. And if that doesn't show any differences. It may be worthwhile
to disassemble the kernels and diff the output of that,

I went through the same process to show that the Linux kernel for KiSS DP558 did
not match the source that KiSS published. And to my knowledge they still don't.

If you ask me, the output of cmp is not helping a lot, other then saying "they are different".

> 
> hence I suspect that the sources I do not corresponds to the kernel running on my tv.
> 
> Bye
> -- 
> Luca
>