Re: [RFC][PATCH] fail kernel compilation in case of unresolved symbols

[Kirill Korotaev <dev@sw.ru>]

Adrian Bunk wrote:
> On Tue, Sep 05, 2006 at 05:31:59PM +0200, Sam Ravnborg wrote:
> 
>>On Tue, Sep 05, 2006 at 05:47:25PM +0400, Kirill Korotaev wrote:
>>
>>>At stage 2 modpost utility is used to check modules.
>>>In case of unresolved symbols modpost only prints warning.
>>>
>>>IMHO it is a good idea to fail compilation process in case of
>>>unresolved symbols, since usually such errors are left unnoticed,
>>>but kernel modules are broken.
>>
>>The primary reason why we do not fail in this case is that building
>>external modules often result in unresolved symbols at modpost time.
>>
>>And there is many legitime uses of external modules that we shall support.

> Is there a way we can get this only for building the kernel itself?
> In this case an unresolved symbol is a real bug that should cause an 
> abort of the compilation.
IMHO for kernel linking will fail...

Don't you consider the kernel to be broken if suddenly one of your modules
began to have unresolved symbols?

> I'm often doing compile tests for the kernel, and the current warnings 
> are too easy to miss.
exactly. and I'm pretty sure, that vendors have the same problem.

Thanks,
Kirill