HI !
doing a series of static code analysis on 2.6.34-rcX each of them with
defconfig (x86 32bit) - the strangeness is that even though the config
file is the same for rc3...5 the number of files actually compiled
is very different. (linux-2.6 git tree)
2.6.34-rc5 1695
2.6.34-rc4 3404
2.6.34-rc3 5109
I diffed the .configs - rc4/rc5 only shows
3,4c3,4
< # Linux kernel version: 2.6.34-rc4
< # Thu Apr 22 11:37:43 2010
---
> # Linux kernel version: 2.6.34-rc5
> # Thu Apr 22 10:44:02 2010
why then this large difference in files being compiled ?
any hint apprecieated.
thx!
hofrat
On Thu, 22 Apr 2010 18:07:23 +0200 Nicholas Mc Guire wrote:
>
> HI !
>
> doing a series of static code analysis on 2.6.34-rcX each of them with
> defconfig (x86 32bit) - the strangeness is that even though the config
> file is the same for rc3...5 the number of files actually compiled
> is very different. (linux-2.6 git tree)
>
> 2.6.34-rc5 1695
> 2.6.34-rc4 3404
> 2.6.34-rc3 5109
>
> I diffed the .configs - rc4/rc5 only shows
>
> 3,4c3,4
> < # Linux kernel version: 2.6.34-rc4
> < # Thu Apr 22 11:37:43 2010
> ---
> > # Linux kernel version: 2.6.34-rc5
> > # Thu Apr 22 10:44:02 2010
>
> why then this large difference in files being compiled ?
>
> any hint apprecieated.
Just diff the build logs to see what is happening.
Or use 'make V=2' so that make will tell you why it's building each piece.
---
~Randy
On Thu, Apr 22, 2010 at 06:07:23PM +0200, Nicholas Mc Guire wrote:
>
>HI !
>
> doing a series of static code analysis on 2.6.34-rcX each of them with
> defconfig (x86 32bit) - the strangeness is that even though the config
> file is the same for rc3...5 the number of files actually compiled
> is very different. (linux-2.6 git tree)
>
> 2.6.34-rc5 1695
> 2.6.34-rc4 3404
> 2.6.34-rc3 5109
>
> I diffed the .configs - rc4/rc5 only shows
>
>3,4c3,4
>< # Linux kernel version: 2.6.34-rc4
>< # Thu Apr 22 11:37:43 2010
>---
>> # Linux kernel version: 2.6.34-rc5
>> # Thu Apr 22 10:44:02 2010
>
> why then this large difference in files being compiled ?
>
> any hint apprecieated.
>
Hmm, interesting.
I guess it would be due to some dependences.
Use the way that you used to find the number of compiled files to
find which files are compiled, make a sorted list and do a diff.
Thanks.