jzhou [Fri, 31 Jul 2009 00:37:36 +0000 (00:37 +0000)]
initial multicore gargbage collection, not finish yet
jjenista [Thu, 30 Jul 2009 22:21:11 +0000 (22:21 +0000)]
stall mechanism working, but data passing between SESE's incorrect
adash [Thu, 30 Jul 2009 22:18:14 +0000 (22:18 +0000)]
1. runtime changes for a simple probability model to decide when objects are risky
2. Keep track of average size of transactions that accessed objects, causing the
transaction to abort. Keep average size info per each object type
jjenista [Thu, 30 Jul 2009 22:06:27 +0000 (22:06 +0000)]
An implementation of parent/child stalling that compiles but hangs during execution. Committing this stable version before tinkering to fix the problem
jjenista [Tue, 28 Jul 2009 23:34:55 +0000 (23:34 +0000)]
bug fixes
jjenista [Tue, 28 Jul 2009 23:12:58 +0000 (23:12 +0000)]
newly realized mlp successfully issues and runs root SESE as a work unit
jjenista [Tue, 28 Jul 2009 00:01:01 +0000 (00:01 +0000)]
mlp system changes
jjenista [Mon, 27 Jul 2009 17:32:59 +0000 (17:32 +0000)]
simplified work schedule, will implement work-stealing later
jjenista [Fri, 24 Jul 2009 22:55:55 +0000 (22:55 +0000)]
changes
jjenista [Fri, 24 Jul 2009 21:41:56 +0000 (21:41 +0000)]
working up mlp system
adash [Fri, 24 Jul 2009 00:20:46 +0000 (00:20 +0000)]
now supports locking of only those objects that were first accessed in a transaction
leading to a transaction abort. locking in read sets are not supported yet.
jjenista [Thu, 23 Jul 2009 22:45:31 +0000 (22:45 +0000)]
ensure MLP system always uses precise garbage collection
adash [Thu, 23 Jul 2009 01:36:56 +0000 (01:36 +0000)]
initial changes to getting sequence id/age for objects as accessed in a transaction cache
bdemsky [Wed, 22 Jul 2009 21:20:04 +0000 (21:20 +0000)]
changes
jjenista [Mon, 20 Jul 2009 22:10:46 +0000 (22:10 +0000)]
mlp compile wasn't pulling in its runtime files anymore
adash [Mon, 20 Jul 2009 22:09:34 +0000 (22:09 +0000)]
changes to Array benchmark
bdemsky [Mon, 20 Jul 2009 22:02:30 +0000 (22:02 +0000)]
bug fixes
adash [Mon, 20 Jul 2009 22:00:24 +0000 (22:00 +0000)]
changes in benchmarks for the PPoPP paper
jzhou [Mon, 20 Jul 2009 17:16:42 +0000 (17:16 +0000)]
fix bugs in memory model in multi-core version
jihoonl [Thu, 16 Jul 2009 22:09:58 +0000 (22:09 +0000)]
Intruder using byte[]
bdemsky [Thu, 16 Jul 2009 11:47:46 +0000 (11:47 +0000)]
last changes
bdemsky [Thu, 16 Jul 2009 10:49:53 +0000 (10:49 +0000)]
bug fix
bdemsky [Thu, 16 Jul 2009 10:14:11 +0000 (10:14 +0000)]
last changes to build
bdemsky [Thu, 16 Jul 2009 10:12:18 +0000 (10:12 +0000)]
changes
bdemsky [Thu, 16 Jul 2009 10:00:58 +0000 (10:00 +0000)]
try to get another bm for ppopp paper
jzhou [Tue, 14 Jul 2009 17:01:50 +0000 (17:01 +0000)]
KMeans bench for multi-core version
bdemsky [Tue, 14 Jul 2009 02:51:26 +0000 (02:51 +0000)]
changes to fix alignment issues
bdemsky [Tue, 14 Jul 2009 00:13:25 +0000 (00:13 +0000)]
change to fix bug
bdemsky [Mon, 13 Jul 2009 00:37:56 +0000 (00:37 +0000)]
benchmarks as used for paper
bdemsky [Mon, 13 Jul 2009 00:37:07 +0000 (00:37 +0000)]
runtime support for read only TRANSREADS
bdemsky [Mon, 13 Jul 2009 00:36:03 +0000 (00:36 +0000)]
bug in casting code
bdemsky [Mon, 13 Jul 2009 00:35:47 +0000 (00:35 +0000)]
bug in cast code
bdemsky [Mon, 13 Jul 2009 00:35:22 +0000 (00:35 +0000)]
changes to get last benchmark working well
bdemsky [Sun, 12 Jul 2009 23:01:51 +0000 (23:01 +0000)]
changes to support wrapper types
shtucker [Fri, 10 Jul 2009 23:24:52 +0000 (23:24 +0000)]
*** empty log message ***
bdemsky [Fri, 10 Jul 2009 01:37:49 +0000 (01:37 +0000)]
Changes:
1) Bug fix
2) make char arrays -> byte arrays to match c code
3) add timing code
jihoonl [Fri, 10 Jul 2009 01:03:49 +0000 (01:03 +0000)]
fixed
bdemsky [Thu, 9 Jul 2009 20:55:22 +0000 (20:55 +0000)]
changes
bdemsky [Thu, 9 Jul 2009 08:31:19 +0000 (08:31 +0000)]
changes
bdemsky [Thu, 9 Jul 2009 02:51:57 +0000 (02:51 +0000)]
change files to match C version
jihoonl [Wed, 8 Jul 2009 22:01:30 +0000 (22:01 +0000)]
Critical mistake was in Stream.java
allocating string problem also been removed after fixing the mistake
jihoonl [Wed, 8 Jul 2009 19:26:48 +0000 (19:26 +0000)]
changes
bdemsky [Wed, 8 Jul 2009 00:29:15 +0000 (00:29 +0000)]
check in my changes
jihoonl [Wed, 8 Jul 2009 00:02:47 +0000 (00:02 +0000)]
Intruder final
jihoonl [Tue, 7 Jul 2009 23:56:43 +0000 (23:56 +0000)]
intruder
jihoonl [Tue, 7 Jul 2009 23:48:41 +0000 (23:48 +0000)]
:
jihoonl [Tue, 7 Jul 2009 23:47:59 +0000 (23:47 +0000)]
*** empty log message ***
jihoonl [Tue, 7 Jul 2009 23:44:58 +0000 (23:44 +0000)]
Working version
to test
./Intruder.bin -a 10 -l 32 -n 100000 -s 1
jihoonl [Sat, 4 Jul 2009 00:16:43 +0000 (00:16 +0000)]
change. not compiling yet.
bdemsky [Fri, 3 Jul 2009 22:46:47 +0000 (22:46 +0000)]
changes
bdemsky [Fri, 3 Jul 2009 22:39:34 +0000 (22:39 +0000)]
check change in
bdemsky [Fri, 3 Jul 2009 06:18:54 +0000 (06:18 +0000)]
changes
jzhou [Thu, 2 Jul 2009 23:35:24 +0000 (23:35 +0000)]
Changes of benchmarks regarding tilera version
jzhou [Thu, 2 Jul 2009 23:34:16 +0000 (23:34 +0000)]
fix a scheduling simulator bug, basically change all time related parameters to type long
bdemsky [Thu, 2 Jul 2009 22:30:21 +0000 (22:30 +0000)]
changes
bdemsky [Thu, 2 Jul 2009 04:07:34 +0000 (04:07 +0000)]
add native copy method
bdemsky [Thu, 2 Jul 2009 04:07:17 +0000 (04:07 +0000)]
bug fix
bdemsky [Thu, 2 Jul 2009 04:03:53 +0000 (04:03 +0000)]
make 3d java version...uses 3d arrays...
bdemsky [Wed, 1 Jul 2009 07:05:52 +0000 (07:05 +0000)]
change
bdemsky [Wed, 1 Jul 2009 05:25:49 +0000 (05:25 +0000)]
new option
bdemsky [Wed, 1 Jul 2009 01:54:24 +0000 (01:54 +0000)]
make setup for benchmarks
bdemsky [Wed, 1 Jul 2009 01:52:36 +0000 (01:52 +0000)]
new make setup
jihoonl [Wed, 1 Jul 2009 00:56:41 +0000 (00:56 +0000)]
Intruder!!
bdemsky [Tue, 30 Jun 2009 23:43:14 +0000 (23:43 +0000)]
add bug fixes to two classes to get KMeans to compile
bdemsky [Tue, 30 Jun 2009 19:32:36 +0000 (19:32 +0000)]
really annoying bug fixed...my benchmarks run now
adash [Tue, 30 Jun 2009 06:02:38 +0000 (06:02 +0000)]
test
bdemsky [Tue, 30 Jun 2009 03:47:40 +0000 (03:47 +0000)]
bug fixes
bdemsky [Tue, 30 Jun 2009 03:47:31 +0000 (03:47 +0000)]
more analysis flexibility
bdemsky [Tue, 30 Jun 2009 03:47:16 +0000 (03:47 +0000)]
locking bug fix
bdemsky [Mon, 29 Jun 2009 23:19:39 +0000 (23:19 +0000)]
file to build STMLabyrinth
bdemsky [Mon, 29 Jun 2009 23:18:48 +0000 (23:18 +0000)]
changes
jihoonl [Mon, 29 Jun 2009 19:44:49 +0000 (19:44 +0000)]
*** empty log message ***
jihoonl [Mon, 29 Jun 2009 19:44:24 +0000 (19:44 +0000)]
Labyrinth works
bdemsky [Mon, 29 Jun 2009 10:30:31 +0000 (10:30 +0000)]
allow turning off this analysisi
bdemsky [Mon, 29 Jun 2009 10:29:12 +0000 (10:29 +0000)]
bug fixes
bdemsky [Mon, 29 Jun 2009 10:26:59 +0000 (10:26 +0000)]
turn off write barrier
bdemsky [Mon, 29 Jun 2009 10:25:44 +0000 (10:25 +0000)]
major change in the way GC works..hope it doesn't break stuff
bdemsky [Mon, 29 Jun 2009 10:24:41 +0000 (10:24 +0000)]
add 32 bit option for compiling 32 bit binaries under 64 bit linux
bdemsky [Mon, 29 Jun 2009 10:20:49 +0000 (10:20 +0000)]
clean up code to just use method that returns positive integers
bdemsky [Mon, 29 Jun 2009 10:19:35 +0000 (10:19 +0000)]
add new method to return positive integers only
bdemsky [Mon, 29 Jun 2009 09:18:33 +0000 (09:18 +0000)]
bug fixes
adash [Mon, 29 Jun 2009 05:41:55 +0000 (05:41 +0000)]
Add new benchmark. Main still incomplete and missing few other files
bdemsky [Sun, 28 Jun 2009 08:17:30 +0000 (08:17 +0000)]
changes
bdemsky [Sat, 27 Jun 2009 23:31:44 +0000 (23:31 +0000)]
make runtime 64 bit clean again
bdemsky [Sat, 27 Jun 2009 23:22:28 +0000 (23:22 +0000)]
missing check ins
bdemsky [Sat, 27 Jun 2009 00:04:15 +0000 (00:04 +0000)]
improve code generation
bdemsky [Fri, 26 Jun 2009 23:46:13 +0000 (23:46 +0000)]
changes
bdemsky [Fri, 26 Jun 2009 20:01:21 +0000 (20:01 +0000)]
bug fixes
bdemsky [Fri, 26 Jun 2009 19:39:08 +0000 (19:39 +0000)]
lots of bug fixes
bdemsky [Fri, 26 Jun 2009 10:29:18 +0000 (10:29 +0000)]
small code changes...didn't make up much of the difference with the C version
bdemsky [Fri, 26 Jun 2009 09:18:18 +0000 (09:18 +0000)]
clean up iterator stuff...
bdemsky [Fri, 26 Jun 2009 06:28:27 +0000 (06:28 +0000)]
optimizations
bdemsky [Fri, 26 Jun 2009 05:02:52 +0000 (05:02 +0000)]
changes
bdemsky [Fri, 26 Jun 2009 01:08:18 +0000 (01:08 +0000)]
fix inliner
bdemsky [Fri, 26 Jun 2009 01:06:28 +0000 (01:06 +0000)]
don't inline native methods
adash [Fri, 26 Jun 2009 00:58:26 +0000 (00:58 +0000)]
changes to account for all objects(read + modified) when counting the
total number of objects that can cause a single transaction to abort
bdemsky [Fri, 26 Jun 2009 00:45:46 +0000 (00:45 +0000)]
bug fix
bdemsky [Fri, 26 Jun 2009 00:39:45 +0000 (00:39 +0000)]
fix bugs
bdemsky [Fri, 26 Jun 2009 00:29:54 +0000 (00:29 +0000)]
missing file
bdemsky [Fri, 26 Jun 2009 00:29:08 +0000 (00:29 +0000)]
first version of inliner
projects / IRC.git / log