Re: glibc 2.23 --- Starting soft/slush freeze

["Carlos O'Donell" <carlos at redhat dot com>]

On 01/12/2016 08:37 AM, Stephen Gallagher wrote:
> On 01/11/2016 03:52 PM, Adhemerval Zanella wrote:
>> Hi all,
> 
>> At stated in a previous message we are now in soft/slushy freeze
>> mode. Please do not commit new features other than the ones already
>> reviewed. If your new feature still needs review and was not listed
>> in the releases wiki as a blocker feature [2] it is unfortunately
>> too late. Please defer to glibc 2.23 when it opens.
> 
>> Documentation changes and bug fixes are OK.
> 
>> [1] https://sourceware.org/ml/libc-alpha/2016-01/msg00052.html [2]
>> https://sourceware.org/glibc/wiki/Release/2.23
> 
> I've had the GroupMerging patch under review for some weeks now and
> the only feedback I've been getting have been nitpicks about the code
> style. If at all possible, I'd like to get this into 2.23; I see no
> one arguing that there are any issues with the approach.
> 

I second this request to get Group Merging into 2.23.

It solves a real problem with existing deployed systems.

It doesn't add any ABI/API to the shared library, only an extra
"[SUCCESS=merge]" specifier for /etc/nsswitch.conf, and therefore
rollout will be gradual and incremental with the opportunity to
revert if we have any future issues (which I don't expect).

I commit to writing an automated regression test for this using
the __nss_configure_lookup API (already in glibc) to alter the
behaviour of group lookups and to use two custom NSS modules to
test the code in a "black box" manner, observing merging and
failed merge scenarios. The automated testing can go in later
or as part of the bug fixing by adding a new test.

At present the code has been reviewed by:
- Mike Frysinger
- Andreas Schwab
- Carlos O'Donell (many times during internal review).

So far only nit picks.

Cheers,
Carlos.