charm.git
13 years agouse a new puper to pup int8 type.
Gengbin Zheng [Fri, 28 Mar 2008 17:47:22 +0000 (17:47 +0000)]
use a new puper to pup int8 type.

13 years agoa complete change to promot int to long to overcome the intmax limit.
Gengbin Zheng [Fri, 28 Mar 2008 16:23:41 +0000 (16:23 +0000)]
a complete change to promot int to long to overcome the intmax limit.

13 years agono eltsComp
Pritish Jetley [Fri, 28 Mar 2008 15:19:27 +0000 (15:19 +0000)]
no eltsComp

13 years agochanged the return type of CmiIsomallocLength to long.
Gengbin Zheng [Fri, 28 Mar 2008 14:08:00 +0000 (14:08 +0000)]
changed the return type of CmiIsomallocLength to long.
Something forgot to check in yesterday.

13 years agono redundant computation
Pritish Jetley [Fri, 28 Mar 2008 08:39:52 +0000 (08:39 +0000)]
no redundant computation

13 years agofixed bug. all elements being touched now.
Pritish Jetley [Fri, 28 Mar 2008 07:59:13 +0000 (07:59 +0000)]
fixed bug. all elements being touched now.
next item - remove redundant computations and replace with copies.

13 years ago*** empty log message ***
Pritish Jetley [Fri, 28 Mar 2008 05:50:35 +0000 (05:50 +0000)]
*** empty log message ***

13 years ago*** empty log message ***
Pritish Jetley [Fri, 28 Mar 2008 05:49:06 +0000 (05:49 +0000)]
*** empty log message ***

13 years agoadded eltsComp
Pritish Jetley [Fri, 28 Mar 2008 05:46:31 +0000 (05:46 +0000)]
added eltsComp

13 years agoiterations++ was in the wrong place
Pritish Jetley [Fri, 28 Mar 2008 05:28:37 +0000 (05:28 +0000)]
iterations++ was in the wrong place

13 years agoremove the constraint that the number of slots on each processor can not exceed the...
Gengbin Zheng [Fri, 28 Mar 2008 02:52:06 +0000 (02:52 +0000)]
remove the constraint that the number of slots on each processor can not exceed the intMax. This is done by prompting relevant int type to long.

13 years ago*** empty log message ***
Pritish Jetley [Thu, 27 Mar 2008 23:28:32 +0000 (23:28 +0000)]
*** empty log message ***

13 years agoAdded initialization and cleanup for bulk adjacency attributes in add/remove
Aaron Becker [Thu, 27 Mar 2008 23:04:17 +0000 (23:04 +0000)]
Added initialization and cleanup for bulk adjacency attributes in add/remove
element

13 years ago*** empty log message ***
Pritish Jetley [Thu, 27 Mar 2008 22:22:54 +0000 (22:22 +0000)]
*** empty log message ***

13 years agofixed a bug that when calling an operation on a remote processor (grab or free),...
Gengbin Zheng [Thu, 27 Mar 2008 22:00:08 +0000 (22:00 +0000)]
fixed a bug that when calling an operation on a remote processor (grab or free), it passed wrong nslots parameter.

13 years agofixed Filippo's check in. Get rid of preloading of conv-config.sh, which changed...
Gengbin Zheng [Thu, 27 Mar 2008 16:31:05 +0000 (16:31 +0000)]
fixed Filippo's check in. Get rid of preloading of conv-config.sh, which changed the value of $@ on some platform for example IBM SP.

13 years agofixed previous checkin errors
Filippo Gioachin [Thu, 27 Mar 2008 00:46:28 +0000 (00:46 +0000)]
fixed previous checkin errors

13 years agoInclude cmidirect.h and ckdirect.h in include directory of build
Sayantan Chakravorty [Wed, 26 Mar 2008 23:20:12 +0000 (23:20 +0000)]
Include cmidirect.h and ckdirect.h in include directory of build

13 years agoCmiDirect/CkDirect changes
Pritish Jetley [Wed, 26 Mar 2008 22:44:11 +0000 (22:44 +0000)]
CmiDirect/CkDirect changes

13 years ago*** empty log message ***
Pritish Jetley [Wed, 26 Mar 2008 22:25:57 +0000 (22:25 +0000)]
*** empty log message ***

13 years ago3d versions, with and without CkDirect
Pritish Jetley [Wed, 26 Mar 2008 22:23:00 +0000 (22:23 +0000)]
3d versions, with and without CkDirect

13 years agoforgot check for machines without python
Filippo Gioachin [Wed, 26 Mar 2008 21:15:01 +0000 (21:15 +0000)]
forgot check for machines without python

13 years ago*** empty log message ***
Filippo Gioachin [Wed, 26 Mar 2008 18:31:26 +0000 (18:31 +0000)]
*** empty log message ***

13 years ago3d stencil computation; no copying of send/recv buffers; quite a pain in the rear...
Pritish Jetley [Wed, 26 Mar 2008 07:08:58 +0000 (07:08 +0000)]
3d stencil computation; no copying of send/recv buffers; quite a pain in the rear to write,
even more so to understand.

13 years agoAdded flag "-debug" to charmc for usage with charmdebug.
Filippo Gioachin [Wed, 26 Mar 2008 03:33:19 +0000 (03:33 +0000)]
Added flag "-debug" to charmc for usage with charmdebug.
Some work done on CmiReduce to make it possible to have concurrent reductions.
Added memory tagging by chare ID in charmdebug.
Fixed breakpointing with charmdebug: now each entry point is defined by its sequential index in the EP table (before a string match was performed).
Added stepping: delivery of a single message in the queue. Also, all the system entry methods are currently always executed, and never get buffered by charmdebug (they can still breakpoint if the user so chooses).
Fixed race condition in python interface. To do that the interface itself has been modified, in particular with regard to high-level python. Python is now more flexible and it allows any chare collection to be "python" and receive requests. The replies are reduced to a PythonCCS mainchare which will handle the socket.
The python directory has now another module (charmdebug-python) for usage by charmdebug introspection code.

13 years agoAdded flag "-debug" to charmc for usage with charmdebug.
Filippo Gioachin [Wed, 26 Mar 2008 03:21:29 +0000 (03:21 +0000)]
Added flag "-debug" to charmc for usage with charmdebug.
Some work done on CmiReduce to make it possible to have concurrent reductions.
Added memory tagging by chare ID in charmdebug.
Fixed breakpointing with charmdebug: now each entry point is defined by its sequential index in the EP table (before a string match was performed).
Added stepping: delivery of a single message in the queue. Also, all the system entry methods are currently always executed, and never get buffered by charmdebug (they can still breakpoint if the user so chooses).
Fixed race condition in python interface. To do that the interface itself has been modified, in particular with regard to high-level python. Python is now more flexible and it allows any chare collection to be "python" and receive requests. The replies are reduced to a PythonCCS mainchare which will handle the socket.
The python directory has now another module (charmdebug-python) for usage by charmdebug introspection code.

13 years agoAdded copy adaptAdj functionality. Also if my previous commit didn't work,
Aaron Becker [Tue, 25 Mar 2008 23:16:53 +0000 (23:16 +0000)]
Added copy adaptAdj functionality. Also if my previous commit didn't work,
fixed edge adjacency creation bug

13 years agoclean up
Pritish Jetley [Mon, 24 Mar 2008 04:35:28 +0000 (04:35 +0000)]
clean up

13 years agofixed bugs
Pritish Jetley [Mon, 24 Mar 2008 04:34:26 +0000 (04:34 +0000)]
fixed bugs

13 years agobug fixes ... answers are now correct
Abhinav Bhatele [Mon, 24 Mar 2008 02:25:08 +0000 (02:25 +0000)]
bug fixes ... answers are now correct

13 years agoIncreased timeout value for ibverbs
Sayantan Chakravorty [Sun, 23 Mar 2008 21:36:24 +0000 (21:36 +0000)]
Increased timeout value for ibverbs

13 years agoChanged timeout value
Sayantan Chakravorty [Sun, 23 Mar 2008 21:14:40 +0000 (21:14 +0000)]
Changed timeout value

13 years agoanother bug :)
Abhinav Bhatele [Sun, 23 Mar 2008 07:56:07 +0000 (07:56 +0000)]
another bug :)

13 years agoa bug fixed
Abhinav Bhatele [Sun, 23 Mar 2008 07:50:30 +0000 (07:50 +0000)]
a bug fixed

13 years agocomm of C fixed
Abhinav Bhatele [Sun, 23 Mar 2008 07:07:36 +0000 (07:07 +0000)]
comm of C fixed

13 years agomemset
Pritish Jetley [Sun, 23 Mar 2008 06:29:21 +0000 (06:29 +0000)]
memset

13 years agocannon's algorithms
Pritish Jetley [Sun, 23 Mar 2008 06:07:08 +0000 (06:07 +0000)]
cannon's algorithms

13 years agoSMP machine.c with optimized spanning tree broadcast.
Sameer Kumar [Fri, 21 Mar 2008 14:35:26 +0000 (14:35 +0000)]
SMP machine.c with optimized spanning tree broadcast.

13 years agoremove pcqueue lock for AMD64.
Gengbin Zheng [Wed, 19 Mar 2008 22:40:47 +0000 (22:40 +0000)]
remove pcqueue lock for AMD64.
This is because AMD64 offers a slightly stronger memory-consistency model, in that it does not reorder a store ahead of a load. Probably we can do away without using lock and fence.
But THIS NEEDS MORE TESTING.

13 years agotopology code checked in
Abhinav Bhatele [Wed, 19 Mar 2008 19:25:10 +0000 (19:25 +0000)]
topology code checked in

13 years agoskip netfem for multicore version.
Gengbin Zheng [Wed, 19 Mar 2008 18:00:49 +0000 (18:00 +0000)]
skip netfem for multicore version.

13 years agofor multicore/smp, copy argc and argv after some of them which are specific to machin...
Gengbin Zheng [Wed, 19 Mar 2008 14:07:29 +0000 (14:07 +0000)]
for multicore/smp, copy argc and argv after some of them which are specific to machine.c are consumed.
This change is TENTATIVE, should be tested on ALL charm versions!

13 years agosecond version
Abhinav Bhatele [Wed, 19 Mar 2008 00:28:55 +0000 (00:28 +0000)]
second version

13 years agoModified adjacency dumper to put a | separator between the lists of
Terry L. Wilmarth [Tue, 18 Mar 2008 20:19:13 +0000 (20:19 +0000)]
Modified adjacency dumper to put a | separator between the lists of
elements that are adjacent to each edge for better readability.

13 years agoremoved static keyword from user handle definition
Pritish Jetley [Tue, 18 Mar 2008 20:15:27 +0000 (20:15 +0000)]
removed static keyword from user handle definition

13 years agolookupEdgeAdaptAdjacencies was getting the int table instead of the char
Terry L. Wilmarth [Tue, 18 Mar 2008 20:01:52 +0000 (20:01 +0000)]
lookupEdgeAdaptAdjacencies was getting the int table instead of the char
table where the data seems to have been stored.

13 years agoThere is some flaw in an assert in adapt_adj's dump adjacencies function
Terry L. Wilmarth [Tue, 18 Mar 2008 19:52:54 +0000 (19:52 +0000)]
There is some flaw in an assert in adapt_adj's dump adjacencies function
which I have remedied.

13 years agoa better fix for the reduction group/nodegroup initialization contention
Gengbin Zheng [Tue, 18 Mar 2008 16:23:52 +0000 (16:23 +0000)]
a better fix for the reduction group/nodegroup initialization contention

13 years agofirst version
Abhinav Bhatele [Tue, 18 Mar 2008 05:53:01 +0000 (05:53 +0000)]
first version

13 years agoChanged the timeout value for queue pairs
Sayantan Chakravorty [Tue, 18 Mar 2008 02:12:26 +0000 (02:12 +0000)]
Changed the timeout value for queue pairs

13 years agocreating a group is a two step processors, first a ckreduction nodegroup, and then...
Gengbin Zheng [Mon, 17 Mar 2008 22:02:31 +0000 (22:02 +0000)]
creating a group is a two step processors, first a ckreduction nodegroup, and then the group itself. This can cause some race condition because one needs to know the other.
I fixed this by sending the reduction group ID in the ckNew message when creating a group.

13 years agoincreased stack size
Gengbin Zheng [Mon, 17 Mar 2008 19:35:11 +0000 (19:35 +0000)]
increased stack size

13 years agofixed a race condition in SMP and bigsim. use a smp lock to protect Bnv initialization.
Gengbin Zheng [Mon, 17 Mar 2008 18:59:51 +0000 (18:59 +0000)]
fixed a race condition in SMP and bigsim. use a smp lock to protect Bnv initialization.

13 years agoa better fix to the previous checkin
Gengbin Zheng [Fri, 14 Mar 2008 20:09:09 +0000 (20:09 +0000)]
a better fix to the previous checkin

13 years agoresize loadbalancers (a ckvec) instead of calling growatleast
Gengbin Zheng [Fri, 14 Mar 2008 19:49:23 +0000 (19:49 +0000)]
resize loadbalancers (a ckvec) instead of calling growatleast

13 years agoGot rid of absoft for net-linux-ppc, replaced with xlf.
Terry L. Wilmarth [Thu, 13 Mar 2008 22:01:09 +0000 (22:01 +0000)]
Got rid of absoft for net-linux-ppc, replaced with xlf.

13 years agoturn off DISPLAY name mangling. Allow ++display to override DISPLAY
Eric Bohm [Thu, 13 Mar 2008 21:45:34 +0000 (21:45 +0000)]
turn off DISPLAY name mangling.  Allow ++display to override DISPLAY

13 years agoAdded the "SAVE" type for variables in Fortran that are "global"
Celso Mendes [Wed, 12 Mar 2008 21:53:27 +0000 (21:53 +0000)]
Added the "SAVE" type for variables in Fortran that are "global"

13 years agofixed printf doubles.
Gengbin Zheng [Tue, 11 Mar 2008 20:08:15 +0000 (20:08 +0000)]
fixed printf doubles.

13 years agoAdding optimized rzv send. To enable compile with -DOPT_RZV.
Sameer Kumar [Tue, 11 Mar 2008 18:54:52 +0000 (18:54 +0000)]
Adding optimized rzv send.  To enable compile with -DOPT_RZV.

13 years agoa demo for Sanjay's class
Gengbin Zheng [Tue, 11 Mar 2008 18:28:54 +0000 (18:28 +0000)]
a demo for Sanjay's class

13 years agofixed a bug related multiple dimensions in fortran mode
Gengbin Zheng [Tue, 11 Mar 2008 18:26:45 +0000 (18:26 +0000)]
fixed a bug related multiple dimensions in fortran mode

13 years agofixed a case when parameter list is void.
Gengbin Zheng [Tue, 11 Mar 2008 18:22:33 +0000 (18:22 +0000)]
fixed a case when parameter list is void.

13 years agoadded CmiWalltimer(), etc
Gengbin Zheng [Tue, 11 Mar 2008 18:21:58 +0000 (18:21 +0000)]
added CmiWalltimer(), etc

13 years agoupdated.
Gengbin Zheng [Tue, 11 Mar 2008 03:43:04 +0000 (03:43 +0000)]
updated.

13 years agotest load balancing under fortran
Gengbin Zheng [Tue, 11 Mar 2008 03:42:22 +0000 (03:42 +0000)]
test load balancing under fortran

13 years agoin fortran mode, generate atSync and resumeFromSync calls.
Gengbin Zheng [Tue, 11 Mar 2008 03:40:42 +0000 (03:40 +0000)]
in fortran mode, generate atSync and resumeFromSync calls.

13 years agofixed a bug in initialization of indexed type constructor
Gengbin Zheng [Mon, 10 Mar 2008 22:03:30 +0000 (22:03 +0000)]
fixed a bug in initialization of indexed type constructor

13 years agominor format change
Gengbin Zheng [Thu, 6 Mar 2008 22:27:31 +0000 (22:27 +0000)]
minor format change

13 years agoFixed ibverbs layer bug that showed up while using ++batch
Sayantan Chakravorty [Wed, 5 Mar 2008 22:40:06 +0000 (22:40 +0000)]
Fixed ibverbs layer bug that showed up while using ++batch

13 years agoReduced the number of buffers to 10000 from 100000
Sayantan Chakravorty [Wed, 5 Mar 2008 22:39:13 +0000 (22:39 +0000)]
Reduced the number of buffers to 10000 from 100000

13 years agoCmiObjIds changed to 4 ints (used to be 3 ints). Change the bglog version number...
Gengbin Zheng [Wed, 27 Feb 2008 22:43:19 +0000 (22:43 +0000)]
CmiObjIds changed to 4 ints (used to be 3 ints). Change the bglog version number so that it can read old logs.

13 years agoSendHypercube and SendBroadcastTree require copy of messages rather than messages
Sayantan Chakravorty [Wed, 27 Feb 2008 02:41:11 +0000 (02:41 +0000)]
SendHypercube and SendBroadcastTree require copy of messages rather than messages

13 years agoIncreased the packet size for better performance
Sayantan Chakravorty [Sat, 23 Feb 2008 08:22:17 +0000 (08:22 +0000)]
Increased the packet size for better performance

13 years agoNew description of coloring.
Isaac Dooley [Thu, 21 Feb 2008 22:17:08 +0000 (22:17 +0000)]
New description of coloring.

13 years agoSome more locking for the smp version of ibverbs
Sayantan Chakravorty [Thu, 21 Feb 2008 22:15:15 +0000 (22:15 +0000)]
Some more locking for the smp version of ibverbs

13 years agoAdded memlock to calls to memory allocation routines
Sayantan Chakravorty [Thu, 21 Feb 2008 21:13:09 +0000 (21:13 +0000)]
Added memlock to calls to memory allocation routines
in the smp mode

13 years agoRemoved some inconsistencies in conv-mach
Sayantan Chakravorty [Mon, 18 Feb 2008 00:15:34 +0000 (00:15 +0000)]
Removed some inconsistencies in conv-mach

13 years agoChecked in code for statistics collection
Sayantan Chakravorty [Mon, 18 Feb 2008 00:14:26 +0000 (00:14 +0000)]
Checked in code for statistics collection

13 years agoDealt with the case when the number of processors is not a multiple of nodesize
Sayantan Chakravorty [Fri, 15 Feb 2008 20:17:59 +0000 (20:17 +0000)]
Dealt with the case when the number of processors is not a multiple of nodesize

13 years agoMade the broadcast routines use the pxshm routines if they need to
Sayantan Chakravorty [Fri, 15 Feb 2008 00:40:59 +0000 (00:40 +0000)]
Made the broadcast routines use the pxshm routines if they need to

13 years agoReduced the number of calls to ibv_reg_mr in startup
Sayantan Chakravorty [Thu, 14 Feb 2008 20:52:09 +0000 (20:52 +0000)]
Reduced the number of calls to ibv_reg_mr in startup

13 years agoFixed a strange bug in ibverbs..calling the ibv_poll_cq function with an array of...
Sayantan Chakravorty [Thu, 14 Feb 2008 20:07:56 +0000 (20:07 +0000)]
Fixed a strange bug in ibverbs..calling the ibv_poll_cq function with an array of length
longer than 256 causes stack corruption leading to memory problems.

Therefore the WC_LIST_SIZE was reduced to 128

13 years ago*** empty log message ***
Abhinav Bhatele [Wed, 13 Feb 2008 07:41:21 +0000 (07:41 +0000)]
*** empty log message ***

13 years ago*** empty log message ***
Abhinav Bhatele [Wed, 13 Feb 2008 07:06:26 +0000 (07:06 +0000)]
*** empty log message ***

13 years ago*** empty log message ***
Abhinav Bhatele [Wed, 13 Feb 2008 06:21:28 +0000 (06:21 +0000)]
*** empty log message ***

13 years agofigures were not showing up in the html manual
Abhinav Bhatele [Mon, 11 Feb 2008 05:50:41 +0000 (05:50 +0000)]
figures were not showing up in the html manual

13 years agofixed a non-existent pointer to the charm distrib
Abhinav Bhatele [Mon, 11 Feb 2008 05:48:10 +0000 (05:48 +0000)]
fixed a non-existent pointer to the charm distrib

13 years agoDocumentation error: fixed error in parameters for a commlib call.
Isaac Dooley [Sun, 10 Feb 2008 21:29:00 +0000 (21:29 +0000)]
Documentation error: fixed error in parameters for a commlib call.

13 years agoFixed some compilation errors because the Microsoft C compiler doesn't fully comply...
Chao Mei [Fri, 8 Feb 2008 22:40:53 +0000 (22:40 +0000)]
Fixed some compilation errors because the Microsoft C compiler doesn't fully comply with the C99 standard. Particularly, the variable has to be declared at the beginning of the function

13 years agoupdate information for building charm++ on a x64 machine with Microsoft
Chao Mei [Fri, 8 Feb 2008 07:22:30 +0000 (07:22 +0000)]
update information for building charm++ on a x64 machine with Microsoft
Visual Studio Express.

13 years ago1. updated the settings with the latest microsoft visual studio express edition ...
Chao Mei [Fri, 8 Feb 2008 07:12:59 +0000 (07:12 +0000)]
1. updated the settings with the latest microsoft visual studio express edition (2008).

2. Add the full path to the likner, otherwise the linker in cygwin takes precedence over the one of Microsoft

13 years agochanges to accomodate SMP topology mapping
Abhinav Bhatele [Thu, 7 Feb 2008 20:09:40 +0000 (20:09 +0000)]
changes to accomodate SMP topology mapping

13 years agosome changes for the fake torus case
Abhinav Bhatele [Mon, 4 Feb 2008 23:55:50 +0000 (23:55 +0000)]
some changes for the fake torus case

13 years agoAdding a new type of projections log entry for memory usage. The new log entries...
Isaac Dooley [Mon, 4 Feb 2008 22:20:47 +0000 (22:20 +0000)]
Adding a new type of projections log entry for memory usage. The new log entries are currently only generated by a user call similar to the user event
tracing.

13 years agoMinor changes to makefile(parallel make will work better) and to one of the IndexAttr...
Isaac Dooley [Mon, 4 Feb 2008 02:44:36 +0000 (02:44 +0000)]
Minor changes to makefile(parallel make will work better) and to one of the IndexAttributes

13 years agoAdding more functions for the new program
Isaac Dooley [Mon, 4 Feb 2008 02:43:20 +0000 (02:43 +0000)]
Adding more functions for the new program

13 years agoA new branch of ParFUM-TOPS for the 3d cohesive version. I don't want to break the...
Isaac Dooley [Sat, 2 Feb 2008 22:22:39 +0000 (22:22 +0000)]
A new branch of ParFUM-TOPS for the 3d cohesive version. I don't want to break the existing library accidentally.

13 years agoAdded a new adjacency accessor for PTOPS
Isaac Dooley [Sat, 2 Feb 2008 21:52:12 +0000 (21:52 +0000)]
Added a new adjacency accessor for PTOPS

13 years agoAdding A new projections trace log entry. Users can now specify an integer to be...
Isaac Dooley [Thu, 31 Jan 2008 18:31:12 +0000 (18:31 +0000)]
Adding A new projections trace log entry. Users can now specify an integer to be associated with each entry method, for
example, a timestep number. Projections can interpret this value for the entry methods.