This is the mail archive of the
newlib@sourceware.org
mailing list for the newlib project.
Re: Branch creation policy?
Mark Mitchell wrote:
Jeff Johnston wrote:
Mark Mitchell wrote:
Is there any objection to creating newlib branches under the same
policies that have been established for Binutils and GDB?
What is the reason behind this request?
The same as for any other case where a branch is appropriate: sometimes
it's useful to do development work which is either not yet ready for
inclusion on mainline, which is in some way unsuitable for mainline, or
for which the patch submission process will take more time than is
presently available.
I was hoping you had something more specific.
I do not wish to start a collection of permanent branches for various
groups out there. My concern is this leads to checking in of
questionable changes, and a lack of effort in properly submitting
changes back into the mainline. I won't quote examples of this in other
projects, but I'm sure you know them. I am not sure why you feel this
is necessary considering the fact that newlib is fairly liberal in patch
acceptance. Is there a patch submitted by you that you felt was
unfairly reviewed?
I am fine with temporary branches created for any of the following reasons:
1. you wish to add a complex piece of functionality and want
to collaborate with others (e.g. adding robust arm-linux support)
2. you have added a risky piece of code to newlib and you would like
others to help you test it (e.g. rewriting the conversion
routines) and verify it does not affect platforms you don't
have access to.
Emphasis on temporary.
-- Jeff J.
CodeSourcery could of course maintain its own private branches in its
own repository, but that is less convenient for us, and deprives other
people of convenient access to our changes.
-- Jeff J.