IRC.git
13 years agoa new field has been added to the standard object layout, shifting the access for oid
jjenista [Wed, 20 Apr 2011 23:46:41 +0000 (23:46 +0000)]
a new field has been added to the standard object layout, shifting the access for oid

13 years agochanges to work with imports...
bdemsky [Wed, 20 Apr 2011 22:54:59 +0000 (22:54 +0000)]
changes to work with imports...

13 years agochanges to make sure that i don't step on stephen's work on imports...
bdemsky [Wed, 20 Apr 2011 22:39:55 +0000 (22:39 +0000)]
changes to make sure that i don't step on stephen's work on imports...

13 years agoenables the labeled_statements. The scope of a label declared by a labled statement...
yeom [Wed, 20 Apr 2011 18:14:38 +0000 (18:14 +0000)]
enables the labeled_statements. The scope of a label declared by a labled statement is the block node following the labeled statement. Note that it might be a different way when Sonny implements continue/break with labels

13 years agoleft out of last checkin
stephey [Wed, 20 Apr 2011 10:01:05 +0000 (10:01 +0000)]
left out of last checkin

13 years agoWork in progress for import packages. In this versio ONLY SINGLE IMPORTS and package...
stephey [Wed, 20 Apr 2011 09:59:54 +0000 (09:59 +0000)]
Work in progress for import packages. In this versio ONLY SINGLE IMPORTS and package declarations at a level OTHER THAN THE MAIN FILE work at the moment. A class Foo in package bar.tar would be internally renamed to bar___________tar___________Foo and that's basically how it works at the moment. Multi-imports (import on demand) to comeIR/ClassDescriptor.java

13 years agobug fix for multicore gc version
jzhou [Wed, 20 Apr 2011 01:29:10 +0000 (01:29 +0000)]
bug fix for multicore gc version

13 years agochanges.
yeom [Tue, 19 Apr 2011 18:07:23 +0000 (18:07 +0000)]
changes.

13 years agoRemove __thread as it does not work on Tilera and use newflag for Task version
jzhou [Tue, 19 Apr 2011 00:46:07 +0000 (00:46 +0000)]
Remove __thread as it does not work on Tilera and use newflag for Task version

13 years agoadd enw file
bdemsky [Tue, 19 Apr 2011 00:27:43 +0000 (00:27 +0000)]
add enw file

13 years agohandle the missing case in the flow down analysis: for a method invocation, check...
yeom [Mon, 18 Apr 2011 23:51:04 +0000 (23:51 +0000)]
handle the missing case in the flow down analysis: for a method invocation, check if ordering relations among caller's method arguments are relatively preserved in callee's arguments

13 years agoadded TODO comments to BuildIR to implement import and packages
limw [Mon, 18 Apr 2011 23:31:12 +0000 (23:31 +0000)]
added TODO comments to BuildIR to implement import and packages

13 years agofix for ssJava: realized that annotation should not be a part of hash code.
yeom [Mon, 18 Apr 2011 22:24:08 +0000 (22:24 +0000)]
fix for ssJava: realized that annotation should not be a part of hash code.

13 years agobug fixes: printing out linenum & source file name in the conflict graph.
yeom [Mon, 18 Apr 2011 22:22:41 +0000 (22:22 +0000)]
bug fixes: printing out linenum & source file name in the conflict graph.

13 years agomake callgraph an interface
bdemsky [Mon, 18 Apr 2011 21:40:31 +0000 (21:40 +0000)]
make callgraph an interface

13 years agobug fix - during call site transfer func we temporarily double nodes per allocation...
jjenista [Mon, 18 Apr 2011 20:43:56 +0000 (20:43 +0000)]
bug fix - during call site transfer func we temporarily double nodes per allocation site where nodes from the callee are called shadow nodes, and nodes that were in the caller before the call are named normal nodes.  During the transfer we move the shadow newest, which are the newest, into the caller and merge normal nodes down to make room.  The bug is that during the merge of normal nodes in this procedure there shoudl be a corresponding aging of reachability states, so when global sweep occurred after the node merging process some states look imprecise and get swept away

13 years agoextend grammar for parameter annotation. forgot to handle this case in the past
yeom [Mon, 18 Apr 2011 19:10:48 +0000 (19:10 +0000)]
extend grammar for parameter annotation. forgot to handle this case in the past

13 years agono need to build these classes unless they are actually accessed
bdemsky [Mon, 18 Apr 2011 18:46:57 +0000 (18:46 +0000)]
no need to build these classes unless they are actually accessed

13 years agoreorganize.
yeom [Mon, 18 Apr 2011 17:55:41 +0000 (17:55 +0000)]
reorganize.

13 years agothis bug is not really a bug, a variable gets optimized out and the points-to and...
jjenista [Mon, 18 Apr 2011 16:44:12 +0000 (16:44 +0000)]
this bug is not really a bug, a variable gets optimized out and the points-to and task in-set is fine, just has a different name

13 years ago*** empty log message ***
bdemsky [Sun, 17 Apr 2011 08:29:16 +0000 (08:29 +0000)]
*** empty log message ***

13 years agochanges
bdemsky [Sun, 17 Apr 2011 08:23:44 +0000 (08:23 +0000)]
changes

13 years agochanges
bdemsky [Sun, 17 Apr 2011 08:13:37 +0000 (08:13 +0000)]
changes

13 years agochange
bdemsky [Sat, 16 Apr 2011 08:12:37 +0000 (08:12 +0000)]
change

13 years agoadd keyword for task allocation...
bdemsky [Sat, 16 Apr 2011 08:06:46 +0000 (08:06 +0000)]
add keyword for task allocation...

13 years agochange
bdemsky [Sat, 16 Apr 2011 07:51:03 +0000 (07:51 +0000)]
change

13 years agochange
bdemsky [Sat, 16 Apr 2011 07:42:01 +0000 (07:42 +0000)]
change

13 years agochanges
bdemsky [Sat, 16 Apr 2011 07:29:52 +0000 (07:29 +0000)]
changes

13 years agoremove comment
bdemsky [Sat, 16 Apr 2011 06:18:53 +0000 (06:18 +0000)]
remove comment

13 years agochanges
bdemsky [Sat, 16 Apr 2011 06:16:59 +0000 (06:16 +0000)]
changes

13 years agomore bug fixes
bdemsky [Sat, 16 Apr 2011 05:30:13 +0000 (05:30 +0000)]
more bug fixes

13 years agofix buildscript
bdemsky [Sat, 16 Apr 2011 05:16:30 +0000 (05:16 +0000)]
fix buildscript

13 years agofixes for oooJava gc
yeom [Sat, 16 Apr 2011 00:14:36 +0000 (00:14 +0000)]
fixes for oooJava gc

13 years agoadd enqueue to garbage.h
bdemsky [Sat, 16 Apr 2011 00:10:35 +0000 (00:10 +0000)]
add enqueue to garbage.h

13 years agochnges
bdemsky [Sat, 16 Apr 2011 00:09:24 +0000 (00:09 +0000)]
chnges

13 years agomore bug fixes
bdemsky [Fri, 15 Apr 2011 23:55:56 +0000 (23:55 +0000)]
more bug fixes

13 years agochanges
bdemsky [Fri, 15 Apr 2011 23:53:49 +0000 (23:53 +0000)]
changes

13 years agobug fixes for tasks...
bdemsky [Fri, 15 Apr 2011 23:53:34 +0000 (23:53 +0000)]
bug fixes for tasks...

13 years agoChanges regarding the lock system. Still bugs and the TASK version code does not...
jzhou [Fri, 15 Apr 2011 23:43:32 +0000 (23:43 +0000)]
Changes regarding the lock system. Still bugs and the TASK version code does not work now

13 years agobug fixes
bdemsky [Fri, 15 Apr 2011 23:27:59 +0000 (23:27 +0000)]
bug fixes

13 years agosplit garbage collector into three files...
bdemsky [Fri, 15 Apr 2011 23:20:10 +0000 (23:20 +0000)]
split garbage collector into three files...

13 years agosimplify the gc code...
bdemsky [Fri, 15 Apr 2011 22:39:24 +0000 (22:39 +0000)]
simplify the gc code...

13 years agobug...wonder how this used to work??
bdemsky [Fri, 15 Apr 2011 22:38:58 +0000 (22:38 +0000)]
bug...wonder how this used to work??

13 years agochanges.
yeom [Fri, 15 Apr 2011 22:36:17 +0000 (22:36 +0000)]
changes.

13 years agomore bug fixes
bdemsky [Fri, 15 Apr 2011 21:50:06 +0000 (21:50 +0000)]
more bug fixes

13 years agochanges
bdemsky [Fri, 15 Apr 2011 07:56:55 +0000 (07:56 +0000)]
changes

13 years agocode to generate calls into JNI native code...
bdemsky [Fri, 15 Apr 2011 07:50:22 +0000 (07:50 +0000)]
code to generate calls into JNI native code...

13 years agochanges
bdemsky [Fri, 15 Apr 2011 04:51:48 +0000 (04:51 +0000)]
changes

13 years agomore changes
bdemsky [Fri, 15 Apr 2011 00:40:28 +0000 (00:40 +0000)]
more changes

13 years agomore
bdemsky [Fri, 15 Apr 2011 00:21:47 +0000 (00:21 +0000)]
more

13 years agoChange the implementation of classobj
jzhou [Fri, 15 Apr 2011 00:21:15 +0000 (00:21 +0000)]
Change the implementation of classobj

13 years agochanges
bdemsky [Fri, 15 Apr 2011 00:17:55 +0000 (00:17 +0000)]
changes

13 years agotemplates towards jni
bdemsky [Thu, 14 Apr 2011 23:57:09 +0000 (23:57 +0000)]
templates towards jni

13 years agoextends the grammar to have a way to define a new type of location that is allowed...
yeom [Thu, 14 Apr 2011 21:20:14 +0000 (21:20 +0000)]
extends the grammar to have a way to define a new type of location that is allowed to have spinning values

13 years agomove hashcode into header...part of prep to switch class libraries
bdemsky [Thu, 14 Apr 2011 18:53:46 +0000 (18:53 +0000)]
move hashcode into header...part of prep to switch class libraries

13 years agoadd hashcode field
bdemsky [Thu, 14 Apr 2011 18:46:25 +0000 (18:46 +0000)]
add hashcode field

13 years agochange
bdemsky [Thu, 14 Apr 2011 06:57:11 +0000 (06:57 +0000)]
change

13 years agochanges
bdemsky [Thu, 14 Apr 2011 06:51:33 +0000 (06:51 +0000)]
changes

13 years agomake compiler work on mac...
bdemsky [Wed, 13 Apr 2011 08:03:44 +0000 (08:03 +0000)]
make compiler work on mac...

13 years agomove up initializethreads
bdemsky [Wed, 13 Apr 2011 02:11:49 +0000 (02:11 +0000)]
move up initializethreads

13 years agomissing changes
bdemsky [Wed, 13 Apr 2011 01:51:56 +0000 (01:51 +0000)]
missing changes

13 years agomake work on mac
bdemsky [Wed, 13 Apr 2011 01:51:41 +0000 (01:51 +0000)]
make work on mac

13 years agochange
bdemsky [Wed, 13 Apr 2011 00:53:06 +0000 (00:53 +0000)]
change

13 years agochanges.
yeom [Wed, 13 Apr 2011 00:27:34 +0000 (00:27 +0000)]
changes.

13 years agobug fix for optimizations`
bdemsky [Tue, 12 Apr 2011 23:51:41 +0000 (23:51 +0000)]
bug fix for optimizations`

13 years agochanges.
yeom [Tue, 12 Apr 2011 02:49:03 +0000 (02:49 +0000)]
changes.

13 years agothread id should be an int
bdemsky [Mon, 11 Apr 2011 22:17:29 +0000 (22:17 +0000)]
thread id should be an int

13 years agobug
bdemsky [Mon, 11 Apr 2011 21:38:30 +0000 (21:38 +0000)]
bug

13 years agochanges
bdemsky [Mon, 11 Apr 2011 21:31:25 +0000 (21:31 +0000)]
changes

13 years agoremove lockcount
bdemsky [Mon, 11 Apr 2011 21:17:37 +0000 (21:17 +0000)]
remove lockcount

13 years agochanges.
yeom [Mon, 11 Apr 2011 18:30:31 +0000 (18:30 +0000)]
changes.

13 years agogood for a little more speedup...
bdemsky [Mon, 11 Apr 2011 09:28:00 +0000 (09:28 +0000)]
good for a little more speedup...

13 years agoimprove specjbb marks...rework locks a little...
bdemsky [Mon, 11 Apr 2011 09:14:58 +0000 (09:14 +0000)]
improve specjbb marks...rework locks a little...

13 years agobug fixes
bdemsky [Mon, 11 Apr 2011 08:34:55 +0000 (08:34 +0000)]
bug fixes

13 years agochange
bdemsky [Mon, 11 Apr 2011 07:47:48 +0000 (07:47 +0000)]
change

13 years agochanges
bdemsky [Mon, 11 Apr 2011 07:42:26 +0000 (07:42 +0000)]
changes

13 years agobug fixes
bdemsky [Sun, 10 Apr 2011 22:29:41 +0000 (22:29 +0000)]
bug fixes

13 years agobug fix
bdemsky [Sun, 10 Apr 2011 06:02:29 +0000 (06:02 +0000)]
bug fix

13 years agoget specjbb compiling again...get rid of annoying warnings...still left with large...
bdemsky [Sat, 9 Apr 2011 10:45:10 +0000 (10:45 +0000)]
get specjbb compiling again...get rid of annoying warnings...still left with large long warning..unclear of what to do..

13 years agocheckin to only generate C for callable methods...
bdemsky [Sat, 9 Apr 2011 10:27:07 +0000 (10:27 +0000)]
checkin to only generate C for callable methods...
use javacallgraph when appropriate
get rid of some annoying warnings

13 years agohaving better speedup
yeom [Sat, 9 Apr 2011 00:24:29 +0000 (00:24 +0000)]
having better speedup

13 years agoadjust run configurations
yeom [Fri, 8 Apr 2011 23:34:46 +0000 (23:34 +0000)]
adjust run configurations

13 years agochange...
bdemsky [Fri, 8 Apr 2011 22:59:14 +0000 (22:59 +0000)]
change...

13 years agoobligatory ascii art
jjenista [Fri, 8 Apr 2011 22:44:11 +0000 (22:44 +0000)]
obligatory ascii art

13 years agodetect possibly evil tasks and do the right thing if a possibly evil execution patter...
jjenista [Fri, 8 Apr 2011 22:38:01 +0000 (22:38 +0000)]
detect possibly evil tasks and do the right thing if a possibly evil execution pattern is detected

13 years agoan example of an evil task
jjenista [Fri, 8 Apr 2011 22:37:08 +0000 (22:37 +0000)]
an example of an evil task

13 years agocommit the temporarily abandoned version of Delaunay refinement
jjenista [Fri, 8 Apr 2011 22:35:26 +0000 (22:35 +0000)]
commit the temporarily abandoned version of Delaunay refinement

13 years agohaving working version of both single and rcrpointer. currently it's working fine...
yeom [Fri, 8 Apr 2011 22:31:08 +0000 (22:31 +0000)]
having working version of both single and rcrpointer. currently it's working fine with original benchmark size 64k points which seems to be realllllllllly small for our env)

13 years agoAdded the Generic N-Tuple capability. Main File is MultiHash which uses Tuples interf...
linsw [Fri, 8 Apr 2011 19:24:31 +0000 (19:24 +0000)]
Added the Generic N-Tuple capability. Main File is MultiHash which uses Tuples interface file and Tuple to get the job done. OptimizedView is a single view of one of the tables.

13 years agohaving a verification routine.
yeom [Fri, 8 Apr 2011 18:41:25 +0000 (18:41 +0000)]
having a verification routine.

13 years agofix: handle the case that TASK doesn't have any heap conflicts
yeom [Fri, 8 Apr 2011 17:17:06 +0000 (17:17 +0000)]
fix: handle the case that TASK doesn't have any heap conflicts

13 years agochanges
stephey [Fri, 8 Apr 2011 06:33:01 +0000 (06:33 +0000)]
changes

13 years agochanged number of RCR workers to matchooo workers. The new speed up is either 13x...
stephey [Fri, 8 Apr 2011 06:31:48 +0000 (06:31 +0000)]
changed number of RCR workers to matchooo workers. The new speed up is either 13x or 6x...... 13x is running the single version but for some odd reason, it garbage collects... the 6x comes from running the rcrpointer version with depth of 0 which SHOULD call no traverser and spawn no sese (other than the parent).

I think it could be faster still, but a bug in the conflict graph is preventing me from adding more seses to speed things up.

13 years agovoronoi benchmark. On dc-11, the computation part runs 3x faster with a computation...
stephey [Fri, 8 Apr 2011 05:39:01 +0000 (05:39 +0000)]
voronoi benchmark. On dc-11, the computation part runs 3x faster with a computation depth of 3.

It could probably do better but for some reason, when I add another sese (see Vertex.java line 115 where it's commented out), RuntimeConflictResolver.java crashes on line 506 (cannot find a conflict graph for a given sese).

13 years agofixes -nostalltr flag: if it is valid to prune heap examiners, turning off both of...
yeom [Fri, 8 Apr 2011 03:45:29 +0000 (03:45 +0000)]
fixes -nostalltr flag: if it is valid to prune heap examiners, turning off both of task & stallsite examiners

13 years agoEnable 1,2,4,8,16,32,50 cores' execution
jzhou [Fri, 8 Apr 2011 03:42:44 +0000 (03:42 +0000)]
Enable 1,2,4,8,16,32,50 cores' execution

13 years agoTODO: move record pool to a thread-local thing and never deallocate it. Issue is...
jjenista [Thu, 7 Apr 2011 02:33:53 +0000 (02:33 +0000)]
TODO: move record pool to a thread-local thing and never deallocate it.  Issue is that a child can retire, then its parent, then deallocate the child's entire mem pool... then the child's traverser finally gets around to the record

13 years agobug fix, at task exit we are looking for sources that cause virtual read of soemthing...
jjenista [Thu, 7 Apr 2011 02:32:00 +0000 (02:32 +0000)]
bug fix, at task exit we are looking for sources that cause virtual read of soemthing the task wrote, don't just look at parent and siblings, but also the parent's parent and the parent's sibling, etc.

13 years agochange on the labyrinth run script: making memory interleaved on all cpus as like...
yeom [Thu, 7 Apr 2011 02:24:50 +0000 (02:24 +0000)]
change on the labyrinth run script: making memory interleaved on all cpus as like other runscripts

13 years agohack seems to help a little...power isn't going to have good performance...
bdemsky [Thu, 7 Apr 2011 01:21:33 +0000 (01:21 +0000)]
hack seems to help a little...power isn't going to have good performance...