[arch-general] file-5.14 - change in how information is read from vmlinuz?

[Oon-Ee Ng]

On Tue, Apr 2, 2013 at 11:28 PM, Kevin Vesga <31337h4ck3r at gmail.com> wrote:
> It's a bug:
>
> http://mx.gw.com/pipermail/file/2013/001158.html

Thanks, so its a priority thing in the magic.mgc I guess.

On Tue, Apr 2, 2013 at 11:06 PM, Chris “Kwpolska” Warrick
<kwpolska at gmail.com> wrote:
> On Tue, Apr 2, 2013 at 4:28 PM, Oon-Ee Ng <ngoonee.talk at gmail.com> wrote:
>> Hi, I previously used the output from `file /boot/*` on the various
>> vmlinuz files in /boot/* to detect what kernels were installed on a
>> machine. As of file 5.14 this no longer works[1].
>>
>> I'm not sure what's changed, perhaps the magic.mgc file? When I copy
>> the magic.mgc file from previous (5.13) version of the file package I
>> get the 'expected' output[2].
>>
>> Not sure if this is a bug or simply a change in behaviour.
>>
>> [1] - /boot/vmlinuz-linux: x86 boot sector
>>
>> [2] - /boot/vmlinuz-linux: Linux kernel x86 boot executable bzImage,
>> version 3.8.5-1-ARCH (tobias at T-POWA-LX) #1 SMP PREEMPT Fri Mar 29
>> 19:18, RO-rootFS, swap_dev 0x3, Normal VGA
>
> file -k gives better results[1].  I believe that happens because “x86
> boot sector” has a higher priority or something like that in the new
> version.
>
> [1]: /boot/vmlinuz-linux: x86 boot sector\012- Linux kernel x86 boot
> executable bzImage, version 3.8.5-1-ARCH (tobias at T-POWA-LX) #1 SMP
> PREEMPT Fri Mar 29 19:18, RO-rootFS, swap_dev 0x3, Normal VGA PE32+
> executable (EFI application) x86-64 (stripped to external PDB), for MS
> Windows

This is useful, should prevent such a 'bug' regarding priority ever
affecting my script again, thanks =)