GitLab

the others.

The python.org Python distribution has the executable at the top level, not in a bin subdir, with include below it.

file that doesn't exist in the tree.

Minor, but important cleanup operation.

Support for reloading garcias.

Currently this is a total hack; we simply rsync the stargate as it is freed
from the experiment.  However, until we unravel os_load dependancies between
nodes and subnodes (motes and stargates in this case), doing this through
the appropriate setup channels won't work.

The tbrsync script borrows much of its infrastructure from Rob's tbuisp
script.

Some more robot integration.

	* event/lib/event.h, event/lib/event.c: Add some
	event_notification creation functions that get used in
	event-sched.

	* event/sched/event-sched.c, event/sched/rpc.h,
	event/sched/rpc.cc: Sync the start of event time with the
	robots reaching their initial positions.  This is done by
	creating a master event-sequence that takes care of sending
	the SETDEST events and then starting the Simulator timeline.

	* mote/tbuisp.in: Use node_reboot instead of ssh'ing in.

	* robots/GNUmakefile.in: Don't build tbsetdest if ulsshxmlrpcpp is
	not available.

	* robots/emc/loclistener.in: Clear the destination values for a
	node when it reaches its destination.

	* robots/primotion/garcia-pilot.cc,
	robots/primotion/pilotClient.cc: Some cleanup and debugging.

	* robots/primotion/wheelManager.cc: Check the rear sensors for
	obstructions and then decide which way to pivot.

	* robots/rmcd/pilotConnection.h, robots/rmcd/pilotConnection.c,
	robots/rmcd/rclip.h: Even more tweaking.

	* robots/tbsetdest/tbsetdest.cc: Don't generate points that are
	outside the camera bounds or inside an obstacle.

	* robots/vmcd/visionTrack.c, robots/vmcd/vmcd.h,
	robots/vmcd/vmcd.c:  Add more debugging output.

	* tbsetup/os_setup.in: Removed the robot hack used when deciding
	which nodes to reconfig/reboot.  Added a robot hack to avoid
	rebooting a robot whose mote is being os_load'd, since it would
	interrupt tbuisp which does the reboot anyways.  Also fixed a
	small typo.

	* tbsetup/ns2ir/node.tcl, tbsetup/ns2ir/sim.tcl.in: Oops, forgot
	to convert degrees to radians.

	* tbsetup/ns2ir/topography.tcl: When checking for node destination
	points in obstacles, include the implicit exclusion zone.

	* tmcd/common/bootsubnodes: Add empty "mote" case.

	* tmcd/linux-sg/GNUmakefile.in: Make some of the /etc subdirs when
	doing the install.

	* tmcd/linux-sg/rc.stargate: Start garcia-pilot.

	* vis/floormap.in: Add options for showing the camera bounds,
	obstacle exclusion zones, and displaying vnames instead of pnames.

	* www/ledpipe.php3: Finally figured out how to use a socket
	instead of popening a perl script.

	* www/robotmap.php3: Add checkboxes for displaying/not displaying
	the camera bounds and obstacle exclusion zones.  Add a legend
	showing what actual vs. destination points are.  Pass pid/eid
	through to vis/floormap if it is given.

	* www/showexp.php3: Add a "Robot Map" link to experiments that
	have allocated garcias.

	* www/floormap/map_legend_node.gif,
	www/floormap/map_legend_node_dst.gif: Icons used by the robot map
	legend.

	* www/floormap/robots-4.jpg: Added an obstacle around the entryway
	so people are slightly less likely to trip over the robots.  Added
	a coordinate system legend to the top left corner.

	* www/tutorial/mobilewireless.php3: Add links to David's movie of
	the robot making its way around the pillar.

	* www/tutorial/robot_anim.gif: A nifty gifanim clip of the robot
	movie.

	* xmlrpc/emulabserver.py.in: Pull the camera data from the DB,
	instead of returning hardcoded stuff.

plain "make".  There seemed to be a number of missing dependencies that
prevented this from working (how did this ever work?)

Maybe by then someone will figure out the CRL stuff.

Was using $BOOTDIR which is different in the frisbee MFS than anywhere else.

oops - hook in rc.stargate.

interface to IP address assignment, and is trivially parsable.
This is less onerous then telling people to parse the output of
ifconfig with perl.

can avoid building robots/tbsetdest.

	tb-set-delay-capacity 1

Will override the default delay capacity as specfied in the node_types
table for each node type, and set it for all types when generating the
ptop file.

This is a big stick, to be used with caution, as it will effectively
double the number of nodes used for delay nodes (withing an experiment).

Stargate client-side boot support.

Similar to linux, but the sg linux distribution is based on debian, so some
things are a little different.

WARNING: if you do a client-install in a cross-compile environment
(with DESTDIR set), watch out for bogus symlinks in <root>/etc/rc*.d!
Have to fix these up by hand in this case.

Also note that the stargate rc script doesn't yet start the robot pilot
daemon - Tim said he needed to do finalize some stuff w.r.t. this, so
he would take care of modifying the rc script.

The ntp settings are still not ideal.  Even though the root fs is on a
journalling jffs filesys, its still not good to write to it regularly.  Need
to point the ntp drift file off into /var (ramfs)

ELABINELAB: allow SSLXMLRPC from inside to boss.  Needed for frisbee load
	of images.
ALL: allow through all ICMP for now.

Only place the last bit of the address in the reverse file, not the
entire IP address.  Still, this whole file will probably only work for
a /24 net...

Tweaks for updating the robotmap:

	* robots/emc/emcd.c: Only send updates to loclistener when the
	position actually changes.

	* robots/emc/loclistener.in: Listen for SETDEST events and update
	the destination position in the nodes table.  Also update the
	orientation column.

	* robots/emc/test_emcd.config, robots/emc/test_emcd.sh.in: Update
	for changes elsewhere.

a certain number of tries go to sleep for a while; the idea is that they
try to recover once the obstacle has moved.

local...)

make it mean 10 thousand. We were running overflowing a 32-bit int
when we had 214 of these things - now we can go to 214 thousand.

cameras more intelligently.  Changed a debug level in vmcd.c .

* If there is a destination for a rebot, display that as a shadow circle.
  Draw a light grey line from source to destination.

* If an orientation is in the location_info table, the current location
  circle gets a stick drawn in that direction.

* If the destination has an orientation in the nodes table, the shadow
  circle gets a stick drawn in that direction.

* I added a little bit of code so that labels are drawn on top if the
  orientation stick would draw through it. I'm sure that will interact with
  cropping badly if a node is right at the edge of the picture.