IRC.git
15 years agochanges to support logging
bdemsky [Thu, 24 Sep 2009 06:25:04 +0000 (06:25 +0000)]
changes to support logging

15 years agoinitial commit for method effects analysis.
yeom [Thu, 24 Sep 2009 00:22:17 +0000 (00:22 +0000)]
initial commit for method effects analysis.
it captures implementation for last demo.

15 years agonew parameter decomposition, can't chain results yet
jjenista [Wed, 23 Sep 2009 23:09:00 +0000 (23:09 +0000)]
new parameter decomposition, can't chain results yet

15 years agoget some more improvements...now at 12x speedup for remote array accesses
bdemsky [Wed, 23 Sep 2009 09:36:19 +0000 (09:36 +0000)]
get some more improvements...now at 12x speedup for remote array accesses

15 years agobatch the sends
bdemsky [Wed, 23 Sep 2009 07:59:07 +0000 (07:59 +0000)]
batch the sends

15 years agoadd "count of Prefetch response" into stats
adash [Wed, 23 Sep 2009 01:20:55 +0000 (01:20 +0000)]
add "count of Prefetch response" into stats
small inline changes that yeild some benefit
A good number to play around is the size of the prefetch
threshold that initiates a garbage collection.

15 years agosaving the working version..
jihoonl [Tue, 22 Sep 2009 21:04:48 +0000 (21:04 +0000)]
saving the working version..

15 years agoadd buffered send method
bdemsky [Tue, 22 Sep 2009 20:42:31 +0000 (20:42 +0000)]
add buffered send method

15 years agoa common example to compare with shape analyis
jjenista [Tue, 22 Sep 2009 20:33:27 +0000 (20:33 +0000)]
a common example to compare with shape analyis

15 years agomore bug fixes
bdemsky [Tue, 22 Sep 2009 19:28:35 +0000 (19:28 +0000)]
more bug fixes

15 years agobug fix
bdemsky [Tue, 22 Sep 2009 19:15:11 +0000 (19:15 +0000)]
bug fix

15 years agoalmost forgot
bdemsky [Tue, 22 Sep 2009 11:37:47 +0000 (11:37 +0000)]
almost forgot

15 years agochanges
bdemsky [Tue, 22 Sep 2009 10:34:04 +0000 (10:34 +0000)]
changes

15 years ago recovery
jihoonl [Tue, 22 Sep 2009 00:19:08 +0000 (00:19 +0000)]
 recovery

15 years ago recovery.
jihoonl [Tue, 22 Sep 2009 00:14:48 +0000 (00:14 +0000)]
 recovery.
 Thread.java got a new static start method

15 years ago simple benchmark for recovery
jihoonl [Mon, 21 Sep 2009 23:18:38 +0000 (23:18 +0000)]
 simple benchmark for recovery

15 years agochange
bdemsky [Mon, 21 Sep 2009 22:23:31 +0000 (22:23 +0000)]
change

15 years agochange
bdemsky [Mon, 21 Sep 2009 10:16:44 +0000 (10:16 +0000)]
change

15 years agoan interface for supporting decomposition of disjointness parameter results
jjenista [Fri, 18 Sep 2009 20:01:25 +0000 (20:01 +0000)]
an interface for supporting decomposition of disjointness parameter results

15 years agofix bug
bdemsky [Fri, 18 Sep 2009 04:32:39 +0000 (04:32 +0000)]
fix bug

15 years agoadd SESETree
bdemsky [Fri, 18 Sep 2009 04:32:27 +0000 (04:32 +0000)]
add SESETree

15 years agoonly look at things reachable from main
bdemsky [Fri, 18 Sep 2009 03:28:11 +0000 (03:28 +0000)]
only look at things reachable from main

15 years agonew dsm web crawler benchmark
hkhang [Thu, 17 Sep 2009 23:04:22 +0000 (23:04 +0000)]
new dsm web crawler benchmark

15 years agoexplicit null heap region changes rolled back
jjenista [Thu, 17 Sep 2009 22:08:35 +0000 (22:08 +0000)]
explicit null heap region changes rolled back

15 years agocode to do plots of transaction events..
bdemsky [Thu, 17 Sep 2009 00:38:10 +0000 (00:38 +0000)]
code to do plots of transaction events..

15 years agoadd code to support outputting gnuplot compatible graphs
bdemsky [Thu, 17 Sep 2009 00:04:47 +0000 (00:04 +0000)]
add code to support outputting gnuplot compatible graphs

15 years agoallocated regions have class fields initialized to null region, but when I try to...
jjenista [Tue, 15 Sep 2009 22:06:10 +0000 (22:06 +0000)]
allocated regions have class fields initialized to null region, but when I try to add array initialization the same way a completely unrelated assert fails, got to find the problem before null regions are done

15 years agochange
bdemsky [Tue, 15 Sep 2009 21:50:20 +0000 (21:50 +0000)]
change

15 years agomore null region fixes, but I think method results can still map outgoing edges to...
jjenista [Tue, 15 Sep 2009 21:15:46 +0000 (21:15 +0000)]
more null region fixes, but I think method results can still map outgoing edges to the null field

15 years agosimulator now models two locking modes:
bdemsky [Tue, 15 Sep 2009 19:40:21 +0000 (19:40 +0000)]
simulator now models two locking modes:
lock -- lazy conflict detection
lockcommit -- kill threads as soon as committing thread conflicts with them

15 years agoadded support for null heap region, doesn't work perfectly because null region gets...
jjenista [Tue, 15 Sep 2009 17:57:58 +0000 (17:57 +0000)]
added support for null heap region, doesn't work perfectly because null region gets outgoing edges

15 years agochange
bdemsky [Mon, 14 Sep 2009 23:59:22 +0000 (23:59 +0000)]
change

15 years agoadd support for alokika's locking scheme to simulator
bdemsky [Mon, 14 Sep 2009 23:59:06 +0000 (23:59 +0000)]
add support for alokika's locking scheme to simulator

15 years agofix for x = y.f statements copying by type and ignoring field name
jjenista [Mon, 14 Sep 2009 22:25:47 +0000 (22:25 +0000)]
fix for x = y.f statements copying by type and ignoring field name

15 years agofixed some bugs
bdemsky [Mon, 14 Sep 2009 08:41:16 +0000 (08:41 +0000)]
fixed some bugs
support more scheduling policies

15 years agoTransaction simulation framework. The idea is to basically give us a way to explore...
bdemsky [Mon, 14 Sep 2009 05:12:03 +0000 (05:12 +0000)]
Transaction simulation framework.  The idea is to basically give us a way to explore possible benefits of various contention management approaches.

15 years agoadded example files for cpu pinning and clock skew detection
adash [Fri, 11 Sep 2009 22:28:58 +0000 (22:28 +0000)]
added example files for cpu pinning and clock skew detection

15 years agochanges to get clocks ticks, add stmlog option to log results in file
adash [Fri, 11 Sep 2009 22:27:44 +0000 (22:27 +0000)]
changes to get clocks ticks, add stmlog option to log results in file

15 years agofixed
bdemsky [Fri, 11 Sep 2009 05:54:01 +0000 (05:54 +0000)]
fixed

15 years agobug fix
adash [Tue, 8 Sep 2009 23:40:40 +0000 (23:40 +0000)]
bug fix

15 years agoobj support in MLP is in good shape, code gets generated but answers wont agree until...
jjenista [Tue, 8 Sep 2009 22:53:51 +0000 (22:53 +0000)]
obj support in MLP is in good shape, code gets generated but answers wont agree until mem conflict component works

15 years agobug fixes in multicore gc
jzhou [Fri, 4 Sep 2009 00:54:38 +0000 (00:54 +0000)]
bug fixes in multicore gc

15 years agoFixed variable dynamic source bookkeeping bug, regression test runs for single and...
jjenista [Fri, 4 Sep 2009 00:09:10 +0000 (00:09 +0000)]
Fixed variable dynamic source bookkeeping bug, regression test runs for single and parallelized binaries agree

15 years agobug fixes in multicore gc
jzhou [Thu, 3 Sep 2009 00:18:36 +0000 (00:18 +0000)]
bug fixes in multicore gc

15 years agoupdate email list
bdemsky [Tue, 1 Sep 2009 19:30:48 +0000 (19:30 +0000)]
update email list

15 years agochanges to read clockticks from file
adash [Tue, 1 Sep 2009 01:43:50 +0000 (01:43 +0000)]
changes to read clockticks from file

15 years agoVariable result table pruning is buggy, remove for now, also adding improved tests...
jjenista [Mon, 31 Aug 2009 23:21:49 +0000 (23:21 +0000)]
Variable result table pruning is buggy, remove for now, also adding improved tests.  However, for larger test cases inconsistencies appear

15 years agochanges for modifying the microbenchmarks
adash [Sat, 29 Aug 2009 00:28:28 +0000 (00:28 +0000)]
changes for modifying the microbenchmarks
plot changes
add improved debug statement
measure time lapses using the "timestamp counter" processor register,rdtsc assembly instruction

15 years agobug fixes in multicore gc
jzhou [Sat, 29 Aug 2009 00:12:52 +0000 (00:12 +0000)]
bug fixes in multicore gc

15 years agoLots of bug fixes, stable system, parallelized binary computes incorrect answer
jjenista [Fri, 28 Aug 2009 23:11:28 +0000 (23:11 +0000)]
Lots of bug fixes, stable system, parallelized binary computes incorrect answer

15 years agobug fixes in multicore gc
jzhou [Fri, 28 Aug 2009 01:35:30 +0000 (01:35 +0000)]
bug fixes in multicore gc

15 years agocheck latency of a lock release and lock acquire
adash [Thu, 27 Aug 2009 23:13:48 +0000 (23:13 +0000)]
check latency of a lock release and lock acquire

15 years agobug fixes in multicore gc
jzhou [Thu, 27 Aug 2009 00:26:22 +0000 (00:26 +0000)]
bug fixes in multicore gc

15 years agoestimate thread context switching overhead
adash [Wed, 26 Aug 2009 01:01:13 +0000 (01:01 +0000)]
estimate thread context switching overhead

15 years agobug fixes in multicore gc
jzhou [Wed, 26 Aug 2009 00:10:57 +0000 (00:10 +0000)]
bug fixes in multicore gc

15 years agomethod call support works for small programs, bigger ones hang
jjenista [Tue, 25 Aug 2009 21:32:41 +0000 (21:32 +0000)]
method call support works for small programs, bigger ones hang

15 years agoStill haven't worked all the bugs out of MLP's support for method calls, but stable...
jjenista [Tue, 25 Aug 2009 18:11:59 +0000 (18:11 +0000)]
Still haven't worked all the bugs out of MLP's support for method calls, but stable at least

15 years agobug fix in multicore gc
jzhou [Tue, 25 Aug 2009 01:26:55 +0000 (01:26 +0000)]
bug fix in multicore gc

15 years agoregression updates
jjenista [Mon, 24 Aug 2009 16:50:27 +0000 (16:50 +0000)]
regression updates

15 years agoA regression test to make sure new changes don't break the system
jjenista [Mon, 24 Aug 2009 16:34:32 +0000 (16:34 +0000)]
A regression test to make sure new changes don't break the system

15 years agolots of bug fixes, partial solution towards allowing method calls, stable version...
jjenista [Thu, 20 Aug 2009 17:39:47 +0000 (17:39 +0000)]
lots of bug fixes, partial solution towards allowing method calls, stable version that runs many primitive-only programs

15 years agofinish gc codes and fix some bug
jzhou [Thu, 20 Aug 2009 01:29:31 +0000 (01:29 +0000)]
finish gc codes and fix some bug

15 years agomissing makefile
adash [Wed, 19 Aug 2009 23:44:56 +0000 (23:44 +0000)]
missing makefile

15 years agoa better workscheduler--still has a deficiency because it does not dynamically create...
jjenista [Tue, 18 Aug 2009 18:28:08 +0000 (18:28 +0000)]
a better workscheduler--still has a deficiency because it does not dynamically create more worker threads as threads stall on one another, so some programs could exhaust the pool and all threads will hang

15 years agofound one reason for system hang--there is still at least one other problem that...
jjenista [Tue, 18 Aug 2009 17:32:54 +0000 (17:32 +0000)]
found one reason for system hang--there is still at least one other problem that sometimes causes it, now much less frequently

15 years agochanges on compact phase
jzhou [Tue, 18 Aug 2009 02:16:01 +0000 (02:16 +0000)]
changes on compact phase

15 years agobug fixes
jjenista [Mon, 17 Aug 2009 20:30:04 +0000 (20:30 +0000)]
bug fixes

15 years agobug fixes, use pointers no problem, generate temps properly
jjenista [Fri, 14 Aug 2009 23:40:45 +0000 (23:40 +0000)]
bug fixes, use pointers no problem, generate temps properly

15 years agobug fix, track num running children and wait before exiting
jjenista [Fri, 14 Aug 2009 21:10:40 +0000 (21:10 +0000)]
bug fix, track num running children and wait before exiting

15 years agoa bunch of dynamic var bugs
jjenista [Fri, 14 Aug 2009 21:03:10 +0000 (21:03 +0000)]
a bunch of dynamic var bugs

15 years agosese's wait for children to exit before exiting
jjenista [Thu, 13 Aug 2009 23:38:58 +0000 (23:38 +0000)]
sese's wait for children to exit before exiting

15 years agoprimitive dynamic variables are working
jjenista [Thu, 13 Aug 2009 19:16:41 +0000 (19:16 +0000)]
primitive dynamic variables are working

15 years agotracked down more bugs in not-avail and variable analyses
jjenista [Wed, 12 Aug 2009 17:58:16 +0000 (17:58 +0000)]
tracked down more bugs in not-avail and variable analyses

15 years agoBug fixes in use of liveness analysis, bug fix in not-available analysis, something...
jjenista [Wed, 12 Aug 2009 00:09:43 +0000 (00:09 +0000)]
Bug fixes in use of liveness analysis, bug fix in not-available analysis, something still wrong with computation of dynamic stalls

15 years agosmall changes and bug fix for getting stats
adash [Tue, 11 Aug 2009 00:14:58 +0000 (00:14 +0000)]
small changes and bug fix for getting stats

15 years agoWorking towards dynamic source variables in MLP, not quite ready
jjenista [Mon, 10 Aug 2009 18:45:09 +0000 (18:45 +0000)]
Working towards dynamic source variables in MLP, not quite ready

15 years agochanges, large objs processing not finished yet
jzhou [Sat, 8 Aug 2009 00:15:03 +0000 (00:15 +0000)]
changes, large objs processing not finished yet

15 years agofinish most parts, need to fix large objs handling and memory allocation(add free...
jzhou [Fri, 7 Aug 2009 00:13:05 +0000 (00:13 +0000)]
finish most parts, need to fix large objs handling and memory allocation(add free memory space list)

15 years agoadd gnuplot file
adash [Thu, 6 Aug 2009 01:45:50 +0000 (01:45 +0000)]
add gnuplot file

15 years agoadd run script - needs few changes later
adash [Thu, 6 Aug 2009 01:44:47 +0000 (01:44 +0000)]
add run script - needs few changes later

15 years agosmall changes
adash [Thu, 6 Aug 2009 01:43:45 +0000 (01:43 +0000)]
small changes

15 years agoadd compact phase code for multicore gc, not finished yet
jzhou [Thu, 6 Aug 2009 00:20:26 +0000 (00:20 +0000)]
add compact phase code for multicore gc, not finished yet

15 years agofully realized system for primitives that displays parent-child stalling and sibling...
jjenista [Thu, 6 Aug 2009 00:00:07 +0000 (00:00 +0000)]
fully realized system for primitives that displays parent-child stalling and sibling dependence

15 years agosimple microbenchmark to study locking
adash [Wed, 5 Aug 2009 00:49:31 +0000 (00:49 +0000)]
simple microbenchmark to study locking

15 years agoadd load balance module for multicore gc, fix message handling and memory allocation...
jzhou [Wed, 5 Aug 2009 00:22:33 +0000 (00:22 +0000)]
add load balance module for multicore gc, fix message handling and memory allocation in multicore gc

15 years agobig update for forwarding lists and dependency counts, stable compile but seg faults
jjenista [Wed, 5 Aug 2009 00:12:39 +0000 (00:12 +0000)]
big update for forwarding lists and dependency counts, stable compile but seg faults

15 years agoadded peek functions
jjenista [Wed, 5 Aug 2009 00:12:01 +0000 (00:12 +0000)]
added peek functions

15 years agoprimitives passed to a child and then all available stuff copied back at one stall...
jjenista [Mon, 3 Aug 2009 18:10:02 +0000 (18:10 +0000)]
primitives passed to a child and then all available stuff copied back at one stall, successfully

15 years agotake navid off list
bdemsky [Sat, 1 Aug 2009 02:30:52 +0000 (02:30 +0000)]
take navid off list
fixup options for new server

15 years agoget emails working again
bdemsky [Sat, 1 Aug 2009 02:27:34 +0000 (02:27 +0000)]
get emails working again

15 years agochange
bdemsky [Sat, 1 Aug 2009 01:22:12 +0000 (01:22 +0000)]
change

15 years agochanges for multicore GC. Chang the lock mechanism for Tilera and messages for GC.
jzhou [Sat, 1 Aug 2009 00:28:03 +0000 (00:28 +0000)]
changes for multicore GC. Chang the lock mechanism for Tilera and messages for GC.

15 years agostill isn't getting the right answer, but one variable gets into a child and back...
jjenista [Fri, 31 Jul 2009 22:31:20 +0000 (22:31 +0000)]
still isn't getting the right answer, but one variable gets into a child and back into parent correctly--also has some thread race condition problem, apparently

15 years agomore changes
jzhou [Fri, 31 Jul 2009 00:45:25 +0000 (00:45 +0000)]
more changes

15 years agoinitial multicore gargbage collection, not finish yet
jzhou [Fri, 31 Jul 2009 00:37:36 +0000 (00:37 +0000)]
initial multicore gargbage collection, not finish yet

15 years agostall mechanism working, but data passing between SESE's incorrect
jjenista [Thu, 30 Jul 2009 22:21:11 +0000 (22:21 +0000)]
stall mechanism working, but data passing between SESE's incorrect

15 years ago1. runtime changes for a simple probability model to decide when objects are risky
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

15 years agoAn implementation of parent/child stalling that compiles but hangs during execution...
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

15 years agobug fixes
jjenista [Tue, 28 Jul 2009 23:34:55 +0000 (23:34 +0000)]
bug fixes

15 years agonewly realized mlp successfully issues and runs root SESE as a work unit
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