
The clusterpush.pl script allows you to run a cluster regression using
code that has not been committed to svn.  It's supposed to be a
replacement for clusterpush.sh.

Usage:

  clusterpush.pl [type [user]]

type gs, pcl, xps, and/or ls (standing for language switch).  If you do
not specify a type clusterpush.pl will guess what you want to do based
on the current directory. You can specify a combination of types by
enclosing them in quotes (e.g. clusterpush.pl "gs pcl").  You can also
specify bmpcmp as the type to generate bitmaps that compare the results
of the last clusterpush with the head code (see below for details).

user is one of alex, chris, henrys, hintak, ken, marcos, masaki, mvrhel,
ray, or tor.  If you don't specify 'user' clusterpush.pl uses the value
of $USER.  This makes sense on Linux and Mac OS X and probably cygwin,
but I'm not sure what happens when run under Windows.  If you prefer a
different default username than the one listed above please let me know.

You need to run clusterpush.pl from the top level directory (the same
directory from where you normally run 'make').  Clusterpush prints and
error if you try to run it from a different directory.

Results will be sent via email after the run is completed, which takes
about 25 minuts, presuming the cluster isn't busy.  Your code is compared
to the most recent baseline, so for meaningful results you will want
to sync your code with head if recent changes to head have resulted in
regression differences.

The first time you run clusterpush.pl it will take a really long time
to rsync your source to the master machine and then from the master to
the nodes.  After that things will go much quicker (unless, like me,
you accidently leave giant .ppm files in the current directory).

bmpcmp mode:

Starting with r10876 the local cluster can now generate bitmaps that
show differences between clusterpush jobs and trunk.  

To use the feature run clusterpush.pl with the option bmpcmp (i.e.
"toolbin/localcluster/clusterpush.pl bmpcmp").  This will queue a job
that looks at the most recent none-bmpcmp clusterpush job you ran and
generates bitmaps of the differences for the first 1000 files listed
in the report.  It puts these files on the internets in the location
<http://www.ghostscript.com/~regression/$USER> and then sends you
an email telling you that it's done.  Since some of the test files
are customer confidential this directory is password protected, I've
emailed out passwords, contact me if you have misplaced your password.
BTW, I consider the fact that can look at output from other users, e.g.
<http://www.ghostscript.com/~regression/marcos>, a feature.

I'll add an option to clusterpush.pl real soon now (tm) that will allow
you specify a regression test and a bitmap comparison in one step,
but for now you can queue both by calling clusterpush.pl twice:

 clusterpush.pl gs ; sleep 60 ; clusterpush.pl bmpcmp

(the sleep 60 is required because the cluster master only checks for
requests once a minute, and the second clusterpush request overwrites
the first if it is run before the check happens).
