Re: [rvm-research] GSoC 2013 Call for Jikes RVM project proposals

Hi,

On 21.03.2013 13:05, Eliot Moss wrote:
> I thought we more or less had baseline working and the main gap was lack
> of opt.  Of course accomplishing good merges is important here, and if the
> project moves forward, I would prioritize merging baseline (but probably
> working in various support pieces first).
> 
> Are you aware of problem areas when trying to merge?

I'm not aware of any problem areas for merges for GSoC projects. If
students configure their development environment correctly (e.g. setup
of checkstyle and tests) and use the usual best practices for
development (e.g. commits that contain only one conceptual change),
there shouldn't be any major problems. However, the diffs may become
quite large. It may make sense to think about a good way to review the
changes beforehand.

In the case of the merge of the existing patch set for the opt compiler,
the main problem for me is that the patch set is organized according to
sub-areas of the RVM and not grouped into commits. This makes it very
hard for me to tease the changes apart because I'm not familiar with all
areas of the RVM. It's also not easy to determine why a change was made
and if it is correct. Similar problems occur with the OpenJDK patch set.


Kind regards,

Erik Brangs

------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar