Re: [RESEND] [PATCH] PPC64: First in the series of patches implementing

[Joseph Myers <joseph at codesourcery dot com>]

On Thu, 28 Feb 2019, Tulio Magno Quites Machado Filho wrote:

> 1. Guarantee the new code does not export neither a new ABI nor a new API.

Right now it does export an ABI (meaning both the library symbols, and the 
declarations in bits/math-vector.h that are expected to be interpreted by 
a future compiler as meaning, in accordance with the ABI document that 
needs to be written, that those library symbols are available - 
bits/math-vector.h being installed regardless of whether the library is 
built).

> 2. A description on how to enable the code in order to test it without compiler
>    support.  A public branch/repository is ideal.

Sure, a branch is appropriate; the issues with lack of compiler support 
are issues for having the support on master.  Then, maybe a wiki page that 
points to the glibc branch, the GCC branch (once available), the current 
work on the ABI, etc., so people can find all the relevant pieces.

When a feature crosses multiple toolchain components (or involves the 
kernel as well as glibc, etc.), close cooperation between the people 
developing the features in those different components can be necessary - 
which means patch submissions for each component should include detailed 
information about the status of the work for the other components (and 
pointers to patches or such a wiki page, or CC the person working on the 
other components and ask them to describe the status, etc.).

> 3. Review the patch in order to guarantee it does match the community
>    standards.

This process is already ongoing.  (But if the series ends up including 
sincos it will be especially important to look closely at compiler 
interactions there, given the issues that arose on x86_64 - the addition 
of sincos will need to come with a careful explanation of what the ABI is, 
which may or may not be the same as that used for x86_64, of why that's 
the right ABI for POWER, and of how the header declarations result in that 
ABI being used.)

-- 
Joseph S. Myers
joseph@codesourcery.com