Re: [edk2-devel] [PATCH 0/4] Arm, ArmPlatform, Crypto, Embedded: list internal headers in [Sources]

["Laszlo Ersek" <lersek@redhat.com>]

On 07/23/19 14:19, Leif Lindholm wrote:
> On Tue, Jul 23, 2019 at 01:54:54PM +0200, Laszlo Ersek wrote:
>>>> I wasn't annoyed at the feature itself -- if it helps developers catch
>>>> unlisted headers as soon as incomplete INF files are introduced, then
>>>> it's not a bad feature IMO.
>>>
>>> I agree that the optional nature of whether to list local .h files or
>>> not in the .inf was suboptimal.
>>
>> Hmm, has that ever been officially optional?
>>
>> (The INF spec chapter at
>> <https://edk2-docs.gitbooks.io/edk-ii-inf-specification/content/2_inf_overview/25_[sources]_section.html>
>> doesn't seem to mention header files at all. Thus, I can imagine both
>> "mandatory to list headers" by omission, and "optional to list headers"
>> by omission...)
> 
> Yeah, indeed.
> 
>>> I am just not pleased with the issue
>>> bringing this to the fore is caused by the new caching feature using a
>>> different mechanism for tracking header file dependencies than the
>>> primary build process.
>>
>> Ugh... that's a lot of statements compressed into a single sentence. Can
>> you please break it down for me? (Yes, I remember the mailing list
>> reference you posted earlier, that discussion was too divergent for me.)
> 
> The inclusion of .h files in .inf is not necessary for determining
> build-time dependencies on the Makefile level.
> 
> Thus, the warnings come out of a different and unrelated level of the
> build system, related to the recent build cache features. Which means
> we're checking header file build dependencies through two different
> mechanisms at two different points of the build.

OK, thanks. Now I understand your point. It hasn't been clear to me that
listing the module-internal headers isn't actually necessary for getting
the build dependencies right.

> Or I have fundamentally misunderstood what is going on. Which is also
> possible. In which case we're maintaining our own header file
> dependency tracking infrastructure, despite us in the end relying on
> generating Makefiles. Which also feels less than ideal.

Right.

Thanks
Laszlo