jzhou [Fri, 10 Apr 2009 18:44:31 +0000 (18:44 +0000)]
small changes
jjenista [Fri, 10 Apr 2009 18:25:09 +0000 (18:25 +0000)]
stable, not fully functional
bdemsky [Fri, 10 Apr 2009 08:52:48 +0000 (08:52 +0000)]
change our vector remove method to be like java
bdemsky [Fri, 10 Apr 2009 06:29:30 +0000 (06:29 +0000)]
64 bit support in case we want it...it appears that 64 bit binares are slower
bdemsky [Fri, 10 Apr 2009 06:16:02 +0000 (06:16 +0000)]
somehow we gained two implementations of some methods...this of course caused some problems
jjenista [Fri, 10 Apr 2009 03:45:24 +0000 (03:45 +0000)]
structure for manipulating triples
jjenista [Thu, 9 Apr 2009 23:16:52 +0000 (23:16 +0000)]
quick capture of some stable structures
jjenista [Thu, 9 Apr 2009 21:20:30 +0000 (21:20 +0000)]
more new mlp stuff
bdemsky [Thu, 9 Apr 2009 20:06:04 +0000 (20:06 +0000)]
change
jjenista [Thu, 9 Apr 2009 16:30:59 +0000 (16:30 +0000)]
stable introduction of mlp analysis, infrastructure for variable analysis in place
jjenista [Thu, 9 Apr 2009 16:26:12 +0000 (16:26 +0000)]
when inserting FlatExit, consider if task body's block node-node pair ends with a null node
bdemsky [Thu, 9 Apr 2009 08:20:51 +0000 (08:20 +0000)]
fix bugs
bdemsky [Thu, 9 Apr 2009 07:01:04 +0000 (07:01 +0000)]
fix same bug in DSTM system
bdemsky [Thu, 9 Apr 2009 06:58:34 +0000 (06:58 +0000)]
fix memory leak
bdemsky [Thu, 9 Apr 2009 06:38:16 +0000 (06:38 +0000)]
bug fixes
adash [Thu, 9 Apr 2009 00:05:16 +0000 (00:05 +0000)]
fix array reading bug
bdemsky [Thu, 9 Apr 2009 00:02:23 +0000 (00:02 +0000)]
annoying gc bug
adash [Wed, 8 Apr 2009 22:33:15 +0000 (22:33 +0000)]
bug fix: Was never starting a transaction in the run method but now it does
bdemsky [Wed, 8 Apr 2009 19:34:25 +0000 (19:34 +0000)]
bug fix
bdemsky [Wed, 8 Apr 2009 18:27:11 +0000 (18:27 +0000)]
change
bdemsky [Wed, 8 Apr 2009 09:41:46 +0000 (09:41 +0000)]
allocate memory in large blocks
bdemsky [Wed, 8 Apr 2009 08:51:09 +0000 (08:51 +0000)]
hack stuff not to put new objects in lookup table...trying to make object creation inside of transactions even cheaper
bdemsky [Wed, 8 Apr 2009 02:45:47 +0000 (02:45 +0000)]
large number of bug corrections
bdemsky [Wed, 8 Apr 2009 01:24:42 +0000 (01:24 +0000)]
program to test debug our system on
bdemsky [Wed, 8 Apr 2009 01:20:51 +0000 (01:20 +0000)]
bug fixes
bdemsky [Wed, 8 Apr 2009 00:02:39 +0000 (00:02 +0000)]
collect statistics
bdemsky [Tue, 7 Apr 2009 23:46:37 +0000 (23:46 +0000)]
add support for garbage collection statistics
bdemsky [Tue, 7 Apr 2009 23:21:34 +0000 (23:21 +0000)]
code changes
bdemsky [Tue, 7 Apr 2009 22:35:37 +0000 (22:35 +0000)]
bug fix
bdemsky [Tue, 7 Apr 2009 22:18:50 +0000 (22:18 +0000)]
fix for alokika's bug
adash [Tue, 7 Apr 2009 22:09:47 +0000 (22:09 +0000)]
changes to makefile for debugging
some compile errors still
adash [Tue, 7 Apr 2009 20:28:33 +0000 (20:28 +0000)]
change of option for makefile
adash [Tue, 7 Apr 2009 20:26:36 +0000 (20:26 +0000)]
changes to Test benchmark
jjenista [Tue, 7 Apr 2009 16:24:45 +0000 (16:24 +0000)]
more mlp infrastructure, stable
bdemsky [Tue, 7 Apr 2009 07:39:09 +0000 (07:39 +0000)]
changes to garbage collect STM
bdemsky [Tue, 7 Apr 2009 06:43:05 +0000 (06:43 +0000)]
fix compile
bdemsky [Tue, 7 Apr 2009 06:39:06 +0000 (06:39 +0000)]
memory stuff
bdemsky [Tue, 7 Apr 2009 05:06:39 +0000 (05:06 +0000)]
fix loop bug
adash [Tue, 7 Apr 2009 02:21:56 +0000 (02:21 +0000)]
STM benchmark ported
problems compiling with flattenLoopNode error
bdemsky [Tue, 7 Apr 2009 01:17:04 +0000 (01:17 +0000)]
STM testcase works...plenty of bugs still
bdemsky [Mon, 6 Apr 2009 23:19:48 +0000 (23:19 +0000)]
number of changes
bdemsky [Mon, 6 Apr 2009 22:48:15 +0000 (22:48 +0000)]
closer to compiling
jjenista [Mon, 6 Apr 2009 21:49:54 +0000 (21:49 +0000)]
some commented out code transplanted from BuildFlat that will be the basis for this analysis as a separate pass
jjenista [Mon, 6 Apr 2009 21:48:08 +0000 (21:48 +0000)]
Move mlp and sese stuff into a separate compiler pass
jjenista [Mon, 6 Apr 2009 21:41:31 +0000 (21:41 +0000)]
use a clearer sese name when viewed in built code and other print outs
bdemsky [Mon, 6 Apr 2009 20:37:24 +0000 (20:37 +0000)]
mistake
bdemsky [Mon, 6 Apr 2009 20:35:53 +0000 (20:35 +0000)]
get rid of compile errors
bdemsky [Mon, 6 Apr 2009 19:59:01 +0000 (19:59 +0000)]
oops
bdemsky [Mon, 6 Apr 2009 19:57:23 +0000 (19:57 +0000)]
get rid of error message about needtocollect
adash [Mon, 6 Apr 2009 18:36:19 +0000 (18:36 +0000)]
more changes (get rid of getnewOID and use TRANSREAD macro)
bdemsky [Mon, 6 Apr 2009 17:42:07 +0000 (17:42 +0000)]
another idea
bdemsky [Mon, 6 Apr 2009 07:00:30 +0000 (07:00 +0000)]
STM versions
bdemsky [Mon, 6 Apr 2009 06:58:59 +0000 (06:58 +0000)]
get compiler side of STM working
bdemsky [Sun, 5 Apr 2009 04:52:04 +0000 (04:52 +0000)]
new write barrier class
fixed many bugs
adash [Fri, 3 Apr 2009 22:29:03 +0000 (22:29 +0000)]
added options for STM
adash [Fri, 3 Apr 2009 22:28:10 +0000 (22:28 +0000)]
new files for STM
bdemsky [Fri, 3 Apr 2009 22:15:38 +0000 (22:15 +0000)]
changes
adash [Fri, 3 Apr 2009 22:08:28 +0000 (22:08 +0000)]
small changes to buildscript to add runtime check for STM
adash [Fri, 3 Apr 2009 22:06:15 +0000 (22:06 +0000)]
test file for STM
bdemsky [Fri, 3 Apr 2009 20:40:20 +0000 (20:40 +0000)]
support for STM
bdemsky [Fri, 3 Apr 2009 20:19:37 +0000 (20:19 +0000)]
state change
bdemsky [Fri, 3 Apr 2009 20:14:38 +0000 (20:14 +0000)]
updates
bdemsky [Fri, 3 Apr 2009 19:54:21 +0000 (19:54 +0000)]
code
bdemsky [Fri, 3 Apr 2009 09:16:14 +0000 (09:16 +0000)]
new makefile changes
bdemsky [Fri, 3 Apr 2009 09:15:15 +0000 (09:15 +0000)]
changes
bdemsky [Fri, 3 Apr 2009 09:06:12 +0000 (09:06 +0000)]
lots of bugs
jjenista [Thu, 2 Apr 2009 21:05:13 +0000 (21:05 +0000)]
loop doesn't work, function calls don't work
jjenista [Thu, 2 Apr 2009 20:53:54 +0000 (20:53 +0000)]
a setjmp implementation of single-threaded mlp API
bdemsky [Thu, 2 Apr 2009 06:32:55 +0000 (06:32 +0000)]
a whole bunch of optimizations...should be useful for transactions
adash [Wed, 1 Apr 2009 21:25:33 +0000 (21:25 +0000)]
add new option -singleTM for committing transactions on a single machine
jjenista [Tue, 31 Mar 2009 20:56:27 +0000 (20:56 +0000)]
Added mlp runtime skeleton, flags to build it, syntax option to name sese blocks
jjenista [Mon, 30 Mar 2009 21:58:07 +0000 (21:58 +0000)]
a new tiny test and better SESE node printing
bdemsky [Mon, 30 Mar 2009 18:52:17 +0000 (18:52 +0000)]
changes
bdemsky [Fri, 27 Mar 2009 09:02:25 +0000 (09:02 +0000)]
loop analysis stuff
bdemsky [Fri, 27 Mar 2009 01:32:11 +0000 (01:32 +0000)]
DefUse analysis
bdemsky [Fri, 27 Mar 2009 00:59:36 +0000 (00:59 +0000)]
changes
bdemsky [Fri, 27 Mar 2009 00:26:40 +0000 (00:26 +0000)]
hack loop optimizations
jjenista [Tue, 24 Mar 2009 06:43:47 +0000 (06:43 +0000)]
new benchmark results
jjenista [Tue, 24 Mar 2009 06:03:25 +0000 (06:03 +0000)]
keep alpha source states during node propagation
bdemsky [Tue, 24 Mar 2009 06:01:55 +0000 (06:01 +0000)]
print out line count
jjenista [Tue, 24 Mar 2009 05:39:27 +0000 (05:39 +0000)]
edge matching bug fix and add alpha source states back
bdemsky [Tue, 24 Mar 2009 04:37:13 +0000 (04:37 +0000)]
cache lots of results instead of recomputing them
jjenista [Tue, 24 Mar 2009 02:30:07 +0000 (02:30 +0000)]
fix bug when mapping callee edges into caller, gotta look at super classes for possible match
bdemsky [Tue, 24 Mar 2009 01:35:30 +0000 (01:35 +0000)]
speedup
jjenista [Tue, 24 Mar 2009 01:05:19 +0000 (01:05 +0000)]
improve strong update conditions for store, add strong updates from callee effects and clean up commented debugging code
bdemsky [Tue, 24 Mar 2009 00:39:14 +0000 (00:39 +0000)]
improve performance
jjenista [Mon, 23 Mar 2009 23:35:06 +0000 (23:35 +0000)]
fix edge propagation to union effects
bdemsky [Mon, 23 Mar 2009 21:36:06 +0000 (21:36 +0000)]
changes
jjenista [Mon, 23 Mar 2009 20:27:40 +0000 (20:27 +0000)]
bug fix: changed contains() to containsKey() where erroneous
jjenista [Mon, 23 Mar 2009 19:57:58 +0000 (19:57 +0000)]
don't do strong updates on array element references
bdemsky [Mon, 23 Mar 2009 19:44:45 +0000 (19:44 +0000)]
small test
bdemsky [Mon, 23 Mar 2009 19:33:47 +0000 (19:33 +0000)]
even simpler
jjenista [Mon, 23 Mar 2009 19:02:30 +0000 (19:02 +0000)]
don't add immutable or primitive parameters to alias contexts
bdemsky [Mon, 23 Mar 2009 18:17:03 +0000 (18:17 +0000)]
evil hack...need to fix arity scheme to match paper and remove this
jjenista [Mon, 23 Mar 2009 00:16:37 +0000 (00:16 +0000)]
commenting out debugging
jjenista [Sun, 22 Mar 2009 19:36:28 +0000 (19:36 +0000)]
merging lab changes with home changes
jjenista [Sun, 22 Mar 2009 19:28:01 +0000 (19:28 +0000)]
added smaller version of directo for something in between tiny and full tests of analysis
jjenista [Sun, 22 Mar 2009 19:25:23 +0000 (19:25 +0000)]
just leave commented becuase it is a useful little message to flip on sometimes
jjenista [Sun, 22 Mar 2009 19:24:46 +0000 (19:24 +0000)]
see array dimensions when writing out
jjenista [Sun, 22 Mar 2009 17:15:55 +0000 (17:15 +0000)]
changes and debug