IRC.git
9 years agochanges.
yeom [Wed, 8 Aug 2012 09:16:35 +0000 (09:16 +0000)]
changes.

9 years agochanges.
yeom [Tue, 7 Aug 2012 16:25:40 +0000 (16:25 +0000)]
changes.

9 years agochanges.
yeom [Tue, 7 Aug 2012 01:19:59 +0000 (01:19 +0000)]
changes.

9 years agochanges.
yeom [Mon, 6 Aug 2012 08:36:34 +0000 (08:36 +0000)]
changes.

9 years agochanges.
yeom [Mon, 6 Aug 2012 00:34:22 +0000 (00:34 +0000)]
changes.

9 years agomore changes.
yeom [Sat, 4 Aug 2012 18:49:18 +0000 (18:49 +0000)]
more changes.

9 years agochanges.
yeom [Sat, 4 Aug 2012 02:09:12 +0000 (02:09 +0000)]
changes.

9 years agofixed another problem of the flow-down check. If an assignment statement has OP+EQ...
yeom [Thu, 2 Aug 2012 21:06:46 +0000 (21:06 +0000)]
fixed another problem of the flow-down check. If an assignment statement has OP+EQ operator(eg,+=,*=), calculate GLB of LHS & RHS and then assign it as the location type of RHS. + annotation fixes according to fix.

9 years agoadd PCLOC annotations. all three benchmarks are type-checked now.
yeom [Thu, 2 Aug 2012 00:48:32 +0000 (00:48 +0000)]
add PCLOC annotations. all three benchmarks are type-checked now.

9 years agoAllow to declare a composite location for the initial program counter location.
yeom [Wed, 1 Aug 2012 22:50:30 +0000 (22:50 +0000)]
Allow to declare a composite location for the initial program counter location.

The developer can declare the program counter location as a composite location that  starts with the location type of the current object 'this'. It provides more specific constraints on value flows between fields in the current object.

9 years agoimplemented PCLOC annotation.
yeom [Wed, 1 Aug 2012 18:28:54 +0000 (18:28 +0000)]
implemented PCLOC annotation.

9 years agofix.
yeom [Wed, 1 Aug 2012 06:47:43 +0000 (06:47 +0000)]
fix.

9 years agochanges on the inference engine.
yeom [Wed, 1 Aug 2012 00:45:29 +0000 (00:45 +0000)]
changes on the inference engine.

9 years agoadd source code that does not have location annotations.
yeom [Tue, 31 Jul 2012 00:34:02 +0000 (00:34 +0000)]
add source code that does not have location annotations.

9 years agochanges.
yeom [Mon, 30 Jul 2012 17:42:23 +0000 (17:42 +0000)]
changes.

9 years agochanges.
yeom [Fri, 13 Jul 2012 19:09:33 +0000 (19:09 +0000)]
changes.

9 years agoa bunch of fixes.
yeom [Sat, 7 Jul 2012 02:08:11 +0000 (02:08 +0000)]
a bunch of fixes.

9 years agoimplemented a fixed point based interprocedural analysis that analyzes flow-graphs...
yeom [Fri, 29 Jun 2012 00:34:27 +0000 (00:34 +0000)]
implemented a fixed point based interprocedural analysis that analyzes flow-graphs and construct a lattice for each method.

9 years agoInitialize class objects first as in the staic blocks there might be references to...
jzhou [Tue, 5 Jun 2012 02:23:17 +0000 (02:23 +0000)]
Initialize class objects first as in the staic blocks there might be references to the class objects

9 years agoremove Input file related stuff from the compiler and runtime system
jzhou [Fri, 1 Jun 2012 03:01:32 +0000 (03:01 +0000)]
remove Input file related stuff from the compiler and runtime system

9 years agokeep a copy of my current dissertation example to investiaget further and another...
jjenista [Thu, 31 May 2012 02:09:01 +0000 (02:09 +0000)]
keep a copy of my current dissertation example to investiaget further and another one to just fix and use

9 years agoFix bugs: 1) reorder the initialization of static fields and the execution of static...
jzhou [Thu, 31 May 2012 00:17:54 +0000 (00:17 +0000)]
Fix bugs: 1) reorder the initialization of static fields and the execution of static blocks to make sure that a class' static fields/blocks are initialized/executed after those of its super classes' 2) fix compile problem of FileInputStream.nativeOpen()

9 years agoBug fix, the compressed version of input files can work now
jzhou [Sat, 19 May 2012 04:17:29 +0000 (04:17 +0000)]
Bug fix, the compressed version of input files can work now

9 years agoAdd compressed version of input files, still have bugs
jzhou [Fri, 18 May 2012 02:46:40 +0000 (02:46 +0000)]
Add compressed version of input files, still have bugs

9 years agoEnable the nextDoublt() call for reading input files
jzhou [Thu, 17 May 2012 03:01:58 +0000 (03:01 +0000)]
Enable the nextDoublt() call for reading input files

9 years agoChanges for reading input files in MGC version
jzhou [Tue, 15 May 2012 00:27:08 +0000 (00:27 +0000)]
Changes for reading input files in MGC version

9 years agochanges for reading input files
jzhou [Thu, 10 May 2012 03:51:35 +0000 (03:51 +0000)]
changes for reading input files

9 years agochanges for reading input files
jzhou [Wed, 9 May 2012 04:05:09 +0000 (04:05 +0000)]
changes for reading input files

9 years agoChanges for reading input files in MGC version
jzhou [Tue, 8 May 2012 00:46:32 +0000 (00:46 +0000)]
Changes for reading input files in MGC version

9 years agomore implementation for the inference: propagating relations of callee parameters...
yeom [Mon, 7 May 2012 00:09:43 +0000 (00:09 +0000)]
more implementation for the inference: propagating relations of callee parameters to the caller's lattice

9 years agorun-doj-validation-test.sh: a single script that performs validation tests for all...
yeom [Fri, 4 May 2012 02:04:30 +0000 (02:04 +0000)]
run-doj-validation-test.sh: a single script that performs validation tests for all DOJ benchmarks.

9 years agoFix the bug in the variable analysis: When we define nested tasks, the task liveness...
yeom [Thu, 3 May 2012 17:33:35 +0000 (17:33 +0000)]
Fix the bug in the variable analysis: When we define nested tasks, the task liveness analysis computes a wrong inset/outset. The new analysis computes liveness for each task and keeps the local view of liveness rather than doing the global liveness computation and taking  special care with nested task enter/exit nodes.

9 years agoFix the thread safe problem
jzhou [Thu, 3 May 2012 17:31:29 +0000 (17:31 +0000)]
Fix the thread safe problem

9 years agoFix the thread safe problem
jzhou [Thu, 3 May 2012 17:18:22 +0000 (17:18 +0000)]
Fix the thread safe problem

9 years agoChanges for galois
jzhou [Thu, 3 May 2012 01:31:23 +0000 (01:31 +0000)]
Changes for galois

9 years agoFix: need to keep the number of allocation site for the heap examiner. Somehow the...
yeom [Thu, 3 May 2012 00:38:20 +0000 (00:38 +0000)]
Fix: need to keep the number of allocation site for the heap examiner. Somehow the code that assigns the alloc# to the corresponding variable was commented out. Toggle the comment. Now the Barneshut works fine.

9 years agofixes to remove compilation warnings. now DOJ works fine with the Voronoi.
yeom [Wed, 2 May 2012 17:34:29 +0000 (17:34 +0000)]
fixes to remove compilation warnings. now DOJ works fine with the Voronoi.

9 years agofix: the return value declarations were missing in some method declarations in the...
yeom [Wed, 2 May 2012 17:33:38 +0000 (17:33 +0000)]
fix: the return value declarations were missing in some method declarations in the LinkedList class. so curious about how it could work well without having a compilation error.

9 years agobug fix: The result of the pointer analysis is broken when it tries to do something...
yeom [Wed, 2 May 2012 00:32:02 +0000 (00:32 +0000)]
bug fix: The result of the pointer analysis is broken when it tries to do something for a flat literal node. Since the modeling of a flat literal node is only required for the additional runtime check which verifies the allocation site of objects pointed to(+ it has not yet fully implemented), no problem to turn it off. Turn back to the same way of the OOPSLA DOJ version.

9 years agochanges on the SJava inference
yeom [Mon, 30 Apr 2012 17:16:35 +0000 (17:16 +0000)]
changes on the SJava inference

9 years agomore changes.
yeom [Fri, 20 Apr 2012 00:21:04 +0000 (00:21 +0000)]
more changes.

9 years agochanges.
yeom [Thu, 19 Apr 2012 00:48:21 +0000 (00:48 +0000)]
changes.

9 years agochanges.
yeom [Mon, 16 Apr 2012 19:17:09 +0000 (19:17 +0000)]
changes.

9 years agostart implementing basic approach
yeom [Tue, 10 Apr 2012 22:48:07 +0000 (22:48 +0000)]
start implementing basic approach

9 years agothe parallel implementation doesn't achieve much because there isn't enough work...
jjenista [Mon, 26 Mar 2012 23:19:38 +0000 (23:19 +0000)]
the parallel implementation doesn't achieve much because there isn't enough work to do

9 years agochanges for galois
jzhou [Fri, 23 Mar 2012 02:08:20 +0000 (02:08 +0000)]
changes for galois

9 years agochanges on inference
yeom [Thu, 22 Mar 2012 18:44:29 +0000 (18:44 +0000)]
changes on inference

9 years agoadd a new test case for inference
yeom [Thu, 22 Mar 2012 18:40:04 +0000 (18:40 +0000)]
add a new test case for inference

9 years agochanges and data/eval files for new evaluations
yeom [Wed, 21 Mar 2012 22:12:21 +0000 (22:12 +0000)]
changes and data/eval files for new evaluations

9 years agoImplement Thread.getThreadLocals() and Thread.currentThread() in MGC version
jzhou [Wed, 21 Mar 2012 12:17:41 +0000 (12:17 +0000)]
Implement Thread.getThreadLocals() and Thread.currentThread() in MGC version

9 years agoraw numbers from error injection evaluations
yeom [Fri, 9 Mar 2012 02:59:02 +0000 (02:59 +0000)]
raw numbers from error injection evaluations

9 years agoraw evaluation numbers for the mp3decoder
yeom [Wed, 7 Mar 2012 18:45:57 +0000 (18:45 +0000)]
raw evaluation numbers for the mp3decoder

9 years agoadd batch-mode script
yeom [Tue, 6 Mar 2012 02:27:32 +0000 (02:27 +0000)]
add batch-mode script

9 years agochange example for now...
bdemsky [Fri, 24 Feb 2012 22:30:21 +0000 (22:30 +0000)]
change example for now...

9 years agoadd a simplifed bh.
yeom [Fri, 24 Feb 2012 22:09:05 +0000 (22:09 +0000)]
add a simplifed bh.

9 years agochanges.
yeom [Fri, 17 Feb 2012 20:57:14 +0000 (20:57 +0000)]
changes.

9 years agoadd new compilation flag -ssjavainfer for the location inference.
yeom [Fri, 17 Feb 2012 20:12:54 +0000 (20:12 +0000)]
add new compilation flag -ssjavainfer for the location inference.

9 years agosetup some interfaces & skeleton codes for Adrian's project
yeom [Fri, 17 Feb 2012 19:59:03 +0000 (19:59 +0000)]
setup some interfaces & skeleton codes for Adrian's project

9 years agonew version of experiments
jjenista [Fri, 27 Jan 2012 18:00:48 +0000 (18:00 +0000)]
new version of experiments

9 years agomy laptop is a billion versions of bash ahead of the server, doh
jjenista [Fri, 27 Jan 2012 02:15:32 +0000 (02:15 +0000)]
my laptop is a billion versions of bash ahead of the server, doh

9 years agolittle twiddles for kmeans investigation
jjenista [Fri, 27 Jan 2012 01:14:50 +0000 (01:14 +0000)]
little twiddles for kmeans investigation

9 years agoset up to count graph elements over every final graph for every analyzed method
jjenista [Wed, 25 Jan 2012 20:42:03 +0000 (20:42 +0000)]
set up to count graph elements over every final graph for every analyzed method

9 years agoline numbers for allocation sites not propagating to flat nodes
jjenista [Wed, 25 Jan 2012 00:27:11 +0000 (00:27 +0000)]
line numbers for allocation sites not propagating to flat nodes

9 years agodidnt get what I wanted before
jjenista [Tue, 24 Jan 2012 21:49:29 +0000 (21:49 +0000)]
didnt get what I wanted before

9 years agobits to help investigate node count changing
jjenista [Tue, 24 Jan 2012 17:52:19 +0000 (17:52 +0000)]
bits to help investigate node count changing

10 years agochanges to MGC classlibrary
jzhou [Fri, 20 Jan 2012 01:54:03 +0000 (01:54 +0000)]
changes to MGC classlibrary

10 years agorunning new experiments
jjenista [Fri, 20 Jan 2012 01:42:22 +0000 (01:42 +0000)]
running new experiments

10 years agohave another way to assign a THIS LOCATION to method: using @THISLOC annotation on...
yeom [Fri, 20 Jan 2012 00:30:49 +0000 (00:30 +0000)]
have another way to assign a THIS LOCATION to method: using @THISLOC annotation on the method declaration as the example section of ssjava paper.

10 years agochanges to MGC class library
jzhou [Wed, 18 Jan 2012 01:17:20 +0000 (01:17 +0000)]
changes to MGC class library

10 years agopredicates are not feasible to flip on and off. The entire splicing mechanism for...
jjenista [Tue, 17 Jan 2012 23:24:05 +0000 (23:24 +0000)]
predicates are not feasible to flip on and off.  The entire splicing mechanism for interprocedural analysis uses predicates of callee elements to match fuse points in the caller context.  This is critical because all the caller context heap references and variables out of the callee context can commonly reference different nodes after callee analysis.  When you remove predicates and bring the entire callee context into the caller, the current system *correctly* has no fuse points for the callee elements and wouldn't reference them.

10 years agofix up summarize-per-class mode
jjenista [Tue, 17 Jan 2012 22:58:14 +0000 (22:58 +0000)]
fix up summarize-per-class mode

10 years agoget set up for experiment: test impact of disabling components
jjenista [Tue, 17 Jan 2012 19:48:51 +0000 (19:48 +0000)]
get set up for experiment: test impact of disabling components

10 years agodefreach experiment
jjenista [Mon, 9 Jan 2012 19:10:15 +0000 (19:10 +0000)]
defreach experiment

10 years agoconvergence test for defreach
jjenista [Mon, 9 Jan 2012 19:09:11 +0000 (19:09 +0000)]
convergence test for defreach

10 years agothis little example only parallelizes for oooj with defreach
jjenista [Fri, 6 Jan 2012 23:41:12 +0000 (23:41 +0000)]
this little example only parallelizes for oooj with defreach

10 years agoadd System.nanoTimes() support in MGC version
jzhou [Fri, 6 Jan 2012 22:21:16 +0000 (22:21 +0000)]
add System.nanoTimes() support in MGC version

10 years agoget this dir too
jjenista [Fri, 6 Jan 2012 21:44:49 +0000 (21:44 +0000)]
get this dir too

10 years agonot working out like I wanted
jjenista [Fri, 6 Jan 2012 19:02:29 +0000 (19:02 +0000)]
not working out like I wanted

10 years agoa useful example that needs defreach
jjenista [Fri, 6 Jan 2012 18:14:26 +0000 (18:14 +0000)]
a useful example that needs defreach

10 years agotweaks for running definite reach
jjenista [Fri, 6 Jan 2012 18:13:30 +0000 (18:13 +0000)]
tweaks for running definite reach

10 years agoremove the '$' in the C code as tilera does not allow it
jzhou [Thu, 5 Jan 2012 01:26:50 +0000 (01:26 +0000)]
remove the '$' in the C code as tilera does not allow it

10 years agoremove the '$' in the C code as tilera does not allow it
jzhou [Wed, 4 Jan 2012 01:25:14 +0000 (01:25 +0000)]
remove the '$' in the C code as tilera does not allow it

10 years agoaccidentally changed this makefile, roll it back
jzhou [Fri, 30 Dec 2011 00:21:56 +0000 (00:21 +0000)]
accidentally changed this makefile, roll it back

10 years agochanges to classes
jzhou [Fri, 30 Dec 2011 00:18:30 +0000 (00:18 +0000)]
changes to classes

10 years agochanges to MGC class library
jzhou [Wed, 28 Dec 2011 00:48:54 +0000 (00:48 +0000)]
changes to MGC class library

10 years agoAnother bug fix for inner class code: when trying to find a field in the surrounding...
jzhou [Wed, 21 Dec 2011 02:00:30 +0000 (02:00 +0000)]
Another bug fix for inner class code: when trying to find a field in the surrounding classes, the this was wrongly constructred.

10 years agoFix some inner class related bugs: 1) whenever a class is added into the class list...
jzhou [Tue, 20 Dec 2011 00:15:20 +0000 (00:15 +0000)]
Fix some inner class related bugs: 1) whenever a class is added into the class list of state, it should check its inner class list to do addOuterClassParam() and addOuterClassReferences() operations, previously a lot of this checking is missing 2) previously the ith lexically enclosing instance for a deeply nested inner class was wrongly passed in to the inner class' constructor

10 years agotest case for the ownership checking
yeom [Fri, 16 Dec 2011 19:58:12 +0000 (19:58 +0000)]
test case for the ownership checking

10 years agochanges to MGC class library
jzhou [Fri, 16 Dec 2011 01:42:49 +0000 (01:42 +0000)]
changes to MGC class library

10 years agoFinally, all benchmarks pass the definitely written analysis
yeom [Thu, 15 Dec 2011 21:57:19 +0000 (21:57 +0000)]
Finally, all benchmarks pass the definitely written analysis

10 years agoNew shared loc analysis found a shared location that is not overwritten completely...
yeom [Thu, 15 Dec 2011 21:32:43 +0000 (21:32 +0000)]
New shared loc analysis found a shared location that is not overwritten completely in the synthesis filter. Decoder adds input samples to the Synthesis Filter in circular fashion, which means it needs to keep the last position of the previous loop of the decode to the circular queue. However, the value of start index does not matter because the synthesis filter resumes a normal behavior, as long as data is added to the synthesis filter in a row. Therefore, a workaround solution is that BitStream, which is trusted to ss, assigns the index of header that is incremented by one, and then the decoder overwrites the location of the last position in the Synthesis Filter using the header index. Even the header index value is corrupted, the synthesis filter will resume because BitStream is going to providing following headers.

10 years agochanges to MGC class library
jzhou [Thu, 15 Dec 2011 02:07:29 +0000 (02:07 +0000)]
changes to MGC class library

10 years agoassociate assignment nodes that writes to the same height location with corresponding...
yeom [Thu, 15 Dec 2011 01:27:15 +0000 (01:27 +0000)]
associate assignment nodes that writes to the same height location with corresponding flat nodes in order to check  definitely written rules easily.

10 years agofix mp3decoder: clear memory locations that are not overwritten.
yeom [Thu, 15 Dec 2011 01:21:45 +0000 (01:21 +0000)]
fix mp3decoder: clear memory locations that are not overwritten.

10 years agochanges to MGC class library
jzhou [Wed, 14 Dec 2011 01:38:54 +0000 (01:38 +0000)]
changes to MGC class library

10 years agochanges.
yeom [Tue, 13 Dec 2011 01:43:14 +0000 (01:43 +0000)]
changes.

10 years agoChanges to MGC class library and some missing unit test files for inner class
jzhou [Tue, 13 Dec 2011 01:02:51 +0000 (01:02 +0000)]
Changes to MGC class library and some missing unit test files for inner class

10 years agofix another bug: callee propagated wrong write effects associated with its local...
yeom [Mon, 12 Dec 2011 19:51:01 +0000 (19:51 +0000)]
fix another bug: callee propagated wrong write effects associated with its local vars to the caller's may-write set.

10 years agomore changes, but still broken with mp3decoder
yeom [Mon, 12 Dec 2011 01:18:55 +0000 (01:18 +0000)]
more changes, but still broken with mp3decoder

10 years agoChanges to MGC class library and fix a bug regarding nested inline class declaration
jzhou [Sat, 10 Dec 2011 01:24:01 +0000 (01:24 +0000)]
Changes to MGC class library and fix a bug regarding nested inline class declaration