Keywords: kernel compile failure missing wrong header incorrect
undeclared identifier

>> arch/x86/kernel/apic/apic.c:475:2: error: 'MSR_IA32_TSC_DEADLINE'
>> undeclared
>
> Possibly a corrupt kernel tree because that is what happened in the
> previous known case:
>
> http://www.spinics.net/lists/linux-tip-commits/msg18902.html
>
> MSR_IA32_TSC_DEADLINE is defined in arch/x86/include/uapi/asm/msr-
> index.h
>
>  Cheers,
>
>  Mike Shell

Thanks, Mike.  That led me to the right answer, and its similar origin.
I guess I had too many keywords in my search, I didn't turn that one up
myself.  But to satisfy myself I really could correct it (there are two
ways with my process), I had to figure out how it got that way.  I'm
posting this whole diagnosis to aid others once it gets "googed".

Not counting the one we get from the "sanitized headers", the i7 system
had two msi-index.h, the Conroe only one.  (In 3.8{,.1} some headers
were moved, including this one.)  So on the i7, to follow the way it was
made on the Conroe, I laid down the 3.5.2 tarball, and then the new
stuff from the tarball after patching up to 3.9.11, so now I have the
old and new both.  How did it go away during the patching process?
Working backward through the patch files, in the 3.8 patch the old one
got patched to /dev/null!  BWAHAHAHAHA!  No way was my process going to
pick that up, and it's never going to!  Cummon guys, you didn't have to
make it this hard.  So much for breaking backward compatibility!
-- 
Paul Rogers
[email protected]
Rogers' Second Law: "Everything you do communicates."
(I do not personally endorse any additions after this line. TANSTAAFL :-)

        

-- 
http://www.fastmail.com - The way an email service should be

-- 
http://lists.linuxfromscratch.org/listinfo/blfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Reply via email to