jihoonl [Wed, 17 Feb 2010 01:42:13 +0000 (01:42 +0000)]
print out recovery time in micro second
jihoonl [Wed, 17 Feb 2010 00:39:38 +0000 (00:39 +0000)]
remove debug messages
jihoonl [Wed, 17 Feb 2010 00:27:06 +0000 (00:27 +0000)]
no join!
jihoonl [Wed, 17 Feb 2010 00:24:44 +0000 (00:24 +0000)]
well..
jihoonl [Wed, 17 Feb 2010 00:14:42 +0000 (00:14 +0000)]
debug
jihoonl [Wed, 17 Feb 2010 00:14:12 +0000 (00:14 +0000)]
changes
bdemsky [Mon, 15 Feb 2010 08:07:35 +0000 (08:07 +0000)]
fix code organization
adash [Mon, 15 Feb 2010 07:44:04 +0000 (07:44 +0000)]
changes to improve commit intensive benchmarks and combine invalid messages
and commit messages for speedup
jihoonl [Sat, 13 Feb 2010 00:09:44 +0000 (00:09 +0000)]
change
jihoonl [Fri, 12 Feb 2010 23:12:52 +0000 (23:12 +0000)]
mlookup bug fix
jihoonl [Fri, 12 Feb 2010 23:11:23 +0000 (23:11 +0000)]
mlookup -> altmlookup
hkhang [Fri, 12 Feb 2010 00:35:17 +0000 (00:35 +0000)]
*** empty log message ***
jjenista [Thu, 11 Feb 2010 23:32:38 +0000 (23:32 +0000)]
reevaluating abstract garbage collection, for now leave code but don't use it
hkhang [Thu, 11 Feb 2010 18:33:42 +0000 (18:33 +0000)]
*** empty log message ***
hkhang [Thu, 11 Feb 2010 18:10:52 +0000 (18:10 +0000)]
*** empty log message ***
jzhou [Wed, 10 Feb 2010 17:22:25 +0000 (17:22 +0000)]
bug fixing in multicore gc and add profiling code for gc
adash [Wed, 10 Feb 2010 02:59:53 +0000 (02:59 +0000)]
guard code with sandbox flag to prevent other dsm stuff from breaking
adash [Wed, 10 Feb 2010 02:22:03 +0000 (02:22 +0000)]
add sandbox checks for excessive memory allocation and make all exponential
delay variables as thread locals
jjenista [Tue, 9 Feb 2010 22:49:09 +0000 (22:49 +0000)]
change analysis so nodes are added on demand, and abstractly garbage collected when not reachable from roots
adash [Tue, 9 Feb 2010 20:14:52 +0000 (20:14 +0000)]
binary exponential backoff for soft abort
adash [Tue, 9 Feb 2010 01:35:38 +0000 (01:35 +0000)]
missing checkin
hkhang [Fri, 5 Feb 2010 23:39:12 +0000 (23:39 +0000)]
*** empty log message ***
hkhang [Fri, 5 Feb 2010 23:16:59 +0000 (23:16 +0000)]
*** empty log message ***
adash [Fri, 5 Feb 2010 02:55:58 +0000 (02:55 +0000)]
enable dsm sandboxing
adash [Fri, 5 Feb 2010 02:50:47 +0000 (02:50 +0000)]
complete sandbox implementation
jjenista [Thu, 4 Feb 2010 23:57:11 +0000 (23:57 +0000)]
implementing
jzhou [Thu, 4 Feb 2010 18:39:24 +0000 (18:39 +0000)]
add instanceof into multicore version runtime
adash [Wed, 3 Feb 2010 02:50:59 +0000 (02:50 +0000)]
implementing sandbox
check-in all the manual prefetch changes
yeom [Wed, 3 Feb 2010 02:30:08 +0000 (02:30 +0000)]
new implementation for compiling conflict queues.
hkhang [Wed, 3 Feb 2010 01:26:05 +0000 (01:26 +0000)]
*** empty log message ***
jjenista [Tue, 2 Feb 2010 20:08:01 +0000 (20:08 +0000)]
implementing
jzhou [Tue, 2 Feb 2010 17:38:37 +0000 (17:38 +0000)]
bug fix in multicore gc
jihoonl [Tue, 2 Feb 2010 01:22:32 +0000 (01:22 +0000)]
for test
jihoonl [Tue, 2 Feb 2010 01:11:58 +0000 (01:11 +0000)]
change
jihoonl [Tue, 2 Feb 2010 00:35:24 +0000 (00:35 +0000)]
well..
jihoonl [Tue, 2 Feb 2010 00:13:58 +0000 (00:13 +0000)]
data file generator
yeom [Mon, 1 Feb 2010 21:29:49 +0000 (21:29 +0000)]
keep last snapshot for benchmark before moving to new queue impl.
jihoonl [Mon, 1 Feb 2010 09:32:27 +0000 (09:32 +0000)]
change
jihoonl [Sun, 31 Jan 2010 10:04:47 +0000 (10:04 +0000)]
change
jihoonl [Sun, 31 Jan 2010 10:02:20 +0000 (10:02 +0000)]
printrecoverystat in work.java
jihoonl [Sun, 31 Jan 2010 09:53:29 +0000 (09:53 +0000)]
remove recoverystats
bdemsky [Sun, 31 Jan 2010 04:40:03 +0000 (04:40 +0000)]
fix the 64 bit clean bug with mlp
yeom [Fri, 29 Jan 2010 17:26:07 +0000 (17:26 +0000)]
changes on work scheduler.
jzhou [Fri, 29 Jan 2010 17:17:05 +0000 (17:17 +0000)]
bug fixing in multicore gc
yeom [Fri, 29 Jan 2010 01:58:51 +0000 (01:58 +0000)]
changes.
jjenista [Fri, 29 Jan 2010 01:29:40 +0000 (01:29 +0000)]
make mlp related makefiles use a relative buildscript so everyone can use easily, and added rules to raytracer to make 32bit versions easily
jjenista [Fri, 29 Jan 2010 01:23:09 +0000 (01:23 +0000)]
a tiny test exposing a bug, we think might be memory conflict queue related
jjenista [Fri, 29 Jan 2010 01:22:35 +0000 (01:22 +0000)]
realized there were two print statements in regression testin different SESEs, breaks IO guarantees so chenged to one
yeom [Fri, 29 Jan 2010 00:18:06 +0000 (00:18 +0000)]
kmeans benchmark.
jjenista [Thu, 28 Jan 2010 23:55:59 +0000 (23:55 +0000)]
bug fix for ready-static-dynamic variable classification, which rippled into other bug fixes. Tested on my tiny test, but the regression test still does not match up: there are more bugs out there
adash [Thu, 28 Jan 2010 23:03:22 +0000 (23:03 +0000)]
changes to all manual prefetch benchmarks
yeom [Thu, 28 Jan 2010 02:03:06 +0000 (02:03 +0000)]
changes. get rid of unnecessary iteration over the queue.
yeom [Thu, 28 Jan 2010 00:42:52 +0000 (00:42 +0000)]
changes.
jjenista [Thu, 28 Jan 2010 00:32:26 +0000 (00:32 +0000)]
alert user if allocating an SESE record fails
yeom [Thu, 28 Jan 2010 00:06:04 +0000 (00:06 +0000)]
changes.
yeom [Wed, 27 Jan 2010 18:19:52 +0000 (18:19 +0000)]
changes, use atomic operation for managing dependency counter.
hkhang [Wed, 27 Jan 2010 01:40:20 +0000 (01:40 +0000)]
*** empty log message ***
adash [Tue, 26 Jan 2010 23:30:10 +0000 (23:30 +0000)]
changes
adash [Tue, 26 Jan 2010 00:56:03 +0000 (00:56 +0000)]
changes to fft manual prefetch transalation
adash [Mon, 25 Jan 2010 20:01:21 +0000 (20:01 +0000)]
add manual prefetches to the SOR benchmark
jihoonl [Sat, 23 Jan 2010 06:45:18 +0000 (06:45 +0000)]
change
jihoonl [Sat, 23 Jan 2010 06:44:36 +0000 (06:44 +0000)]
changes for single machine run
yeom [Fri, 22 Jan 2010 17:57:15 +0000 (17:57 +0000)]
changes. add MLP condition in the place of memory allocation.
adash [Fri, 22 Jan 2010 00:49:14 +0000 (00:49 +0000)]
more changes to reduce number of remote reads for manual prefetch
yeom [Fri, 22 Jan 2010 00:44:27 +0000 (00:44 +0000)]
changes.
yeom [Thu, 21 Jan 2010 06:38:35 +0000 (06:38 +0000)]
bug fix, converting method effects into sese effects.
yeom [Thu, 21 Jan 2010 00:27:51 +0000 (00:27 +0000)]
changes.
adash [Wed, 20 Jan 2010 23:53:42 +0000 (23:53 +0000)]
final manual prefetch changes for MatrixMultiply
yeom [Wed, 20 Jan 2010 23:50:08 +0000 (23:50 +0000)]
add
jjenista [Wed, 20 Jan 2010 23:42:40 +0000 (23:42 +0000)]
bug fix: do dynamic tracking when a variable's source transitions from static OR ready to dynamic
jjenista [Wed, 20 Jan 2010 19:11:45 +0000 (19:11 +0000)]
bug fix for copying locals to out-set
jihoonl [Wed, 20 Jan 2010 09:18:58 +0000 (09:18 +0000)]
change
yeom [Wed, 20 Jan 2010 05:50:11 +0000 (05:50 +0000)]
changes.
jjenista [Wed, 20 Jan 2010 05:43:36 +0000 (05:43 +0000)]
fixed a bug in not-available set calculation for MLP
jjenista [Wed, 20 Jan 2010 05:42:57 +0000 (05:42 +0000)]
added a benchmark for disjointness reporting
jjenista [Wed, 20 Jan 2010 00:37:25 +0000 (00:37 +0000)]
finding mlp analysis bugs
jihoonl [Tue, 19 Jan 2010 23:34:48 +0000 (23:34 +0000)]
script to test
jihoonl [Tue, 19 Jan 2010 22:58:22 +0000 (22:58 +0000)]
change
yeom [Tue, 19 Jan 2010 22:50:12 +0000 (22:50 +0000)]
changes.
jzhou [Tue, 19 Jan 2010 22:43:09 +0000 (22:43 +0000)]
add codes for randomly generating core group mappings on cores
yeom [Mon, 18 Jan 2010 19:56:23 +0000 (19:56 +0000)]
changes.
jihoonl [Mon, 18 Jan 2010 09:37:43 +0000 (09:37 +0000)]
chnage
jihoonl [Mon, 18 Jan 2010 09:37:21 +0000 (09:37 +0000)]
change
jihoonl [Sun, 17 Jan 2010 10:26:40 +0000 (10:26 +0000)]
change
hkhang [Sun, 17 Jan 2010 04:38:17 +0000 (04:38 +0000)]
*** empty log message ***
hkhang [Sun, 17 Jan 2010 03:39:13 +0000 (03:39 +0000)]
*** empty log message ***
jihoonl [Sat, 16 Jan 2010 10:43:59 +0000 (10:43 +0000)]
change
jihoonl [Fri, 15 Jan 2010 01:13:58 +0000 (01:13 +0000)]
change
jihoonl [Fri, 15 Jan 2010 01:12:56 +0000 (01:12 +0000)]
-recoverystats
yeom [Wed, 13 Jan 2010 19:09:05 +0000 (19:09 +0000)]
changes in SESE lock scheme using clique covering.
jzhou [Tue, 12 Jan 2010 01:12:00 +0000 (01:12 +0000)]
change in the multicore version runtime. now it is able to choose which cores to execute the program and which cores to run gc
jihoonl [Mon, 11 Jan 2010 10:55:52 +0000 (10:55 +0000)]
changes
jihoonl [Mon, 11 Jan 2010 10:55:36 +0000 (10:55 +0000)]
removed error msg
jihoonl [Mon, 11 Jan 2010 10:54:50 +0000 (10:54 +0000)]
change
jihoonl [Mon, 11 Jan 2010 10:54:37 +0000 (10:54 +0000)]
-recoverystats
jihoonl [Mon, 11 Jan 2010 10:54:20 +0000 (10:54 +0000)]
-recoverystats flag
bdemsky [Mon, 11 Jan 2010 08:14:02 +0000 (08:14 +0000)]
add new contention manager
adash [Mon, 11 Jan 2010 07:35:47 +0000 (07:35 +0000)]
changes for manual prefetch in matrix multiply
adash [Mon, 11 Jan 2010 06:16:29 +0000 (06:16 +0000)]
changes
bdemsky [Sun, 10 Jan 2010 07:50:33 +0000 (07:50 +0000)]
changes
projects / IRC.git / log