[Gluster-devel] Change in glusterfs[master]: build: Support for unit tests using Cmockery2

[Harshavardhana]

This has been really bothering me a bit as our queues are getting
bigger and bigger upstream to even get smallest of the patches to get
fixed quickly.

In these scenarios a decentralized regression testing could be made mandatory?

stage-1 tests - runs on the Author's laptop (i.e Linux) - git hook
perhaps which runs for each ./rfc.sh (reports build issues, other
apparent compilation problems, segfaults on init etc.)
    This could comprise of
    - smoke.sh
    - 'make -j16, make -j32' for parallel build test
    - Unittests

stage-2 tests - run on the initial review post.
    - build rpms EL5, EL6, FC20, future
    - mockbuild
    - ./tests/basic/*
    - any others?

stage-3 tests - run on the final Verification process.
    - full blown ./tests/bugs/*

Currently if you look at the regression test suite it getting bigger
and bigger (our overall time of regression test) completion. Just a
though since simple build failures, compilation failure and other
really simple bugs - upstream servers shouldn't be used. One can
leverage Author's laptop :-)

Don't know what you guys think?

On Wed, May 7, 2014 at 12:00 PM, Luis Pabon (Code Review)
<review at dev.gluster.org> wrote:
> Luis Pabon has posted comments on this change.
>
> Change subject: build: Support for unit tests using Cmockery2
> ......................................................................
>
>
> Patch Set 6:
>
> Good point, but unit tests take no more time to compile, and only take 0.55 secs to run all of them (at the moment).  Is this really an issue?
>
> --
> To view, visit http://review.gluster.org/7538
> To unsubscribe, visit http://review.gluster.org/settings
>
> Gerrit-MessageType: comment
> Gerrit-Change-Id: I1b36cb1f56fd10916f9bf535e8ad080a3358289f
> Gerrit-PatchSet: 6
> Gerrit-Project: glusterfs
> Gerrit-Branch: master
> Gerrit-Owner: Luis Pabon <lpabon at redhat.com>
> Gerrit-Reviewer: Gluster Build System <jenkins at build.gluster.com>
> Gerrit-Reviewer: Harshavardhana <harsha at harshavardhana.net>
> Gerrit-Reviewer: Jeff Darcy <jdarcy at redhat.com>
> Gerrit-Reviewer: Justin Clift <justin at gluster.org>
> Gerrit-Reviewer: Kaleb KEITHLEY <kkeithle at redhat.com>
> Gerrit-Reviewer: Luis Pabon <lpabon at redhat.com>
> Gerrit-Reviewer: Rajesh Joseph <rjoseph at redhat.com>
> Gerrit-Reviewer: Ravishankar N <ravishankar at redhat.com>
> Gerrit-Reviewer: Vijay Bellur <vbellur at redhat.com>
> Gerrit-HasComments: No



-- 
Religious confuse piety with mere ritual, the virtuous confuse
regulation with outcomes