From: Ramsay Jones <ramsay@ramsayjones.plus.com>
Date: Fri, 28 Oct 2016 23:59:14 GMT
Subject: Re: [PATCHv2 00/36] Revamp the attr subsystem!
Message-ID: <588e34f9-015f-82e1-019b-c83571801e7d@ramsayjones.plus.com>
In-Reply-To: <20161028185502.8789-1-sbeller@google.com>



On 28/10/16 19:54, Stefan Beller wrote:
> previous discussion at https://public-inbox.org/git/20161022233225.8883-1-sbeller@google.com
> 
> This implements the discarded series':
> jc/attr
> jc/attr-more
> sb/pathspec-label
> sb/submodule-default-paths
> 
> This includes
> * The fixes for windows
> * Junios latest suggestion to use git_attr_check_initv instead of
>   alloc/append.
> 
> * I implemented the thread safe attr API in patch 27 (attr: convert to new threadsafe API)
> * patch 28 (attr: keep attr stack for each check) makes it actually possible
>   to run in a multithreaded environment.
> * I added a test for the multithreaded when it is introduced in patch 32
>   (pathspec: allow querying for attributes) as well as a test to disallow
>   multiple "attr"s in a pathspec.

By the end of this series, 'git_attr_counted()' and 'git_attr()' are
both file local symbols and can be marked with static. (I gave up the
search for which actual patch should change the symbols to static).

Also, 'git_attr()' ends up with a single caller, so maybe inline that
call?

I was about to have a moan about PTHREAD_MUTEX_INITIALIZER, since it
causes sparse to issue some warnings, but I see that you have decided
not to use it. So, phew! ;-)

ATB,
Ramsay Jones