charm.git
8 years agoremove SPECIFIC_IDLE to make lrtsNotifyIdle general
Yanhua Sun [Fri, 8 Feb 2013 05:29:31 +0000 (05:29 +0000)]
remove SPECIFIC_IDLE to make lrtsNotifyIdle general

8 years agomove init persistent to persisnt.c file
Yanhua Sun [Fri, 8 Feb 2013 04:29:20 +0000 (04:29 +0000)]
move init persistent to persisnt.c file

8 years ago(1) fix L2Atomic Queue bug (2) add spin L2AtomicQueue on idle
Yanhua Sun [Fri, 8 Feb 2013 03:15:17 +0000 (03:15 +0000)]
(1) fix L2Atomic Queue bug (2) add spin L2AtomicQueue on idle

8 years agoMerge branch 'charm' of charmgit:charm into charm
Yanhua Sun [Fri, 8 Feb 2013 03:12:23 +0000 (03:12 +0000)]
Merge branch 'charm' of charmgit:charm into charm

8 years agoadd debugLog for pami
Yanhua Sun [Fri, 8 Feb 2013 03:12:18 +0000 (03:12 +0000)]
add debugLog for pami

8 years agoReducer type "random" should produce a message with reduction type "random", rather...
Lukasz Wesolowski [Fri, 8 Feb 2013 00:08:41 +0000 (18:08 -0600)]
Reducer type "random" should produce a message with reduction type "random", rather than "invalid"

8 years agoexamples/bigsim/emulator/Makefile: Finish fixing parallel build by only copying charm...
Phil Miller [Thu, 7 Feb 2013 21:28:42 +0000 (21:28 +0000)]
examples/bigsim/emulator/Makefile: Finish fixing parallel build by only copying charmrun once

8 years agoexamples/bigsim/emulator/Makefile: Fix parallel compilation and clean up Make style
Phil Miller [Thu, 7 Feb 2013 17:23:23 +0000 (11:23 -0600)]
examples/bigsim/emulator/Makefile: Fix parallel compilation and clean up Make style

8 years agoAdding support for disk checkpoint to fast message-logging.
Esteban Meneses [Thu, 7 Feb 2013 16:49:38 +0000 (10:49 -0600)]
Adding support for disk checkpoint to fast message-logging.

8 years agoProtecting cmiqueue.h with #if
Nikhil Jain [Thu, 7 Feb 2013 07:26:47 +0000 (07:26 +0000)]
Protecting cmiqueue.h with #if

8 years agoMoving SPECIFIC_QUEUE out of machine.c where it should not occur.
Nikhil Jain [Thu, 7 Feb 2013 06:59:47 +0000 (06:59 +0000)]
Moving SPECIFIC_QUEUE out of machine.c where it should not occur.

8 years agoInserting missing #if to protect checkpointing code
Nikhil Jain [Wed, 6 Feb 2013 07:11:44 +0000 (07:11 +0000)]
Inserting missing #if to protect checkpointing code

8 years agoAdd simple wrappers for runtime control of Cray Balanced Injection
Eric Bohm [Tue, 5 Feb 2013 21:46:49 +0000 (15:46 -0600)]
Add simple wrappers for runtime control of Cray Balanced Injection

8 years agoModified Charm++ reductions to accept dummy contributions with no data
Lukasz Wesolowski [Tue, 5 Feb 2013 08:48:13 +0000 (02:48 -0600)]
Modified Charm++ reductions to accept dummy contributions with no data
from elements that do not have anything useful to contribute.
The dummy messages, submitted by calling contribute without input data
(e.g. contribute(), contribute(callbackName)), will be assigned
reducer type CkReduction::random. At the receiving end, all messages with
reducer type random will be deleted from the buffer of messages to be reduced,
except when no message with a valid reducer type other than "random" arrives,
in which case the first "random" message that arrives will be used
as the return for that step of the reduction.

In the process of implementing these changes, fixed a bug in
the smp version of array reductions, where placeholder messages were
not being deleted, and instead indefinitely kept in the buffer
of received messages.

8 years agoMerge branch 'charm' of charmgit:charm into charm
Xiang Ni [Tue, 5 Feb 2013 06:23:17 +0000 (00:23 -0600)]
Merge branch 'charm' of charmgit:charm into charm

8 years agoprotect the codes in net build
Xiang Ni [Tue, 5 Feb 2013 06:23:00 +0000 (00:23 -0600)]
protect the codes in net build

8 years agoCompilation with GCC does not like inline declarations of variables and throws
Nikhil Jain [Tue, 5 Feb 2013 05:22:06 +0000 (05:22 +0000)]
Compilation with GCC does not like inline declarations of variables and throws
warning on some of the things; fixing them

8 years agoRevert "Begining work on interoperatin in non-mpi layer"
Nikhil Jain [Tue, 5 Feb 2013 05:03:15 +0000 (05:03 +0000)]
Revert "Begining work on interoperatin in non-mpi layer"

This reverts commit 33b8aa64328a71627e09c9900f42f20847bb766c.

8 years agoReplace the #ifdef with #if in pingpong example
Nikhil Jain [Tue, 5 Feb 2013 04:48:12 +0000 (20:48 -0800)]
Replace the #ifdef with #if in pingpong example

8 years agoMerge branch 'charm' of charmgit:charm into charm
Nikhil Jain [Tue, 5 Feb 2013 04:22:06 +0000 (20:22 -0800)]
Merge branch 'charm' of charmgit:charm into charm

8 years agofix the CmiDirect compilation error
Yanhua Sun [Tue, 5 Feb 2013 02:43:27 +0000 (20:43 -0600)]
fix the CmiDirect compilation error

8 years agoBug Fix - Correct variable name for BGP TopoManager
Nikhil Jain [Mon, 4 Feb 2013 10:06:57 +0000 (04:06 -0600)]
Bug Fix - Correct variable name for BGP TopoManager

8 years agoBug Fix - intialize the subdir to null if num partitions is 1
Nikhil Jain [Mon, 4 Feb 2013 09:10:09 +0000 (03:10 -0600)]
Bug Fix - intialize the subdir to null if num partitions is 1

8 years agoUse defensive names for variables defined in macros, to avoid conflicts like self...
Phil Miller [Mon, 4 Feb 2013 07:34:20 +0000 (01:34 -0600)]
Use defensive names for variables defined in macros, to avoid conflicts like self-assignment

8 years agoReverting addition of barrier to MPI layer - not all threads come to this stage,
Nikhil Jain [Mon, 4 Feb 2013 06:16:16 +0000 (00:16 -0600)]
Reverting addition of barrier to MPI layer - not all threads come to this stage,
and hence a CmiBarrier is bound to hang

8 years agoprotection for the previous changes
Xiang Ni [Mon, 4 Feb 2013 01:38:32 +0000 (19:38 -0600)]
protection for the previous changes

8 years agoremove the unnecessary conditional
Xiang Ni [Fri, 1 Feb 2013 23:27:37 +0000 (17:27 -0600)]
remove the unnecessary conditional

8 years agomake sure the consistence of the spare pe pool when there multiple partitions
Xiang Ni [Fri, 1 Feb 2013 22:56:48 +0000 (16:56 -0600)]
make sure the consistence of the spare pe pool when there multiple partitions

8 years agoFixing ManytoMany and ListSend for correctness when used with partitons in BGQ
Nikhil Jain [Fri, 1 Feb 2013 07:00:58 +0000 (07:00 +0000)]
Fixing ManytoMany and ListSend for correctness when used with partitons in BGQ

8 years agoFixing bug in BGQTorus.h - incorrect variable name, and setting one more
Nikhil Jain [Fri, 1 Feb 2013 07:00:09 +0000 (07:00 +0000)]
Fixing bug in BGQTorus.h - incorrect variable name, and setting one more
variable

8 years agoMaking correction in CmiDirect of Gemini to work correctly with partitions; when
Nikhil Jain [Fri, 1 Feb 2013 06:44:16 +0000 (22:44 -0800)]
Making correction in CmiDirect of Gemini to work correctly with partitions; when
CmiDirect's implementation is fixed, this should work correctly

8 years agoMaking changes to projection for handling partitions - earlier scheme used to
Nikhil Jain [Fri, 1 Feb 2013 06:26:13 +0000 (00:26 -0600)]
Making changes to projection for handling partitions - earlier scheme used to
dump directories in all cases (even if the code is not compiled in tracemode
projections); also skipping directory if only 1 partition

8 years agochkpt: Factor out file naming and opening to a separate function
Phil Miller [Fri, 1 Feb 2013 05:31:14 +0000 (23:31 -0600)]
chkpt: Factor out file naming and opening to a separate function

This makes partition-handling easier, makes error checking and reporting
uniform, and will make later changes in path structure much easier.

8 years agoAdding a new flag for partition support, and making the change accordingly in
Nikhil Jain [Fri, 1 Feb 2013 04:57:54 +0000 (22:57 -0600)]
Adding a new flag for partition support, and making the change accordingly in
converse.h

8 years agoFixing a bug in BGPTorus - private variables names were incorrect
Nikhil Jain [Thu, 31 Jan 2013 18:36:21 +0000 (12:36 -0600)]
Fixing a bug in BGPTorus - private variables names were incorrect

8 years agoAdding fix for DCMF many to many for multi partition run; I was not able to test
Nikhil Jain [Wed, 30 Jan 2013 11:00:30 +0000 (11:00 +0000)]
Adding fix for DCMF many to many for multi partition run; I was not able to test
the correctness because the only use case (in NAMD) seems to be broken

8 years agoFor Cray machine > XT3, providing correct information even when multiple
Nikhil Jain [Wed, 30 Jan 2013 10:50:40 +0000 (04:50 -0600)]
For Cray machine > XT3, providing correct information even when multiple
partitions are present is straightforward, and hence I am using it. For XT3, I
leave it as it - after getting opinions from people, I plan to retire the
support for XT3

8 years agoModifying the TopoManager related files for BG machines - if more than one
Nikhil Jain [Wed, 30 Jan 2013 10:44:24 +0000 (04:44 -0600)]
Modifying the TopoManager related files for BG machines - if more than one
partitions are present, return a linear topology for now

8 years agoClubbing partition related function into a pound define protected by LRTS, and
Nikhil Jain [Wed, 30 Jan 2013 10:09:25 +0000 (04:09 -0600)]
Clubbing partition related function into a pound define protected by LRTS, and
providing dummy implementations to be used in non-LRTS cases

8 years agoForgot a colon in the previous commit
Nikhil Jain [Wed, 30 Jan 2013 10:03:24 +0000 (04:03 -0600)]
Forgot a colon in the previous commit

8 years agoAdding a CMK pound define for layer that make use of LRTS interface;
Nikhil Jain [Wed, 30 Jan 2013 09:45:33 +0000 (03:45 -0600)]
Adding a CMK pound define for layer that make use of LRTS interface;

8 years agoAdding a barrier before exit in MPI layer; this has been done to unavoid trouble
Nikhil Jain [Wed, 30 Jan 2013 09:42:32 +0000 (03:42 -0600)]
Adding a barrier before exit in MPI layer; this has been done to unavoid trouble
incase some partitions wants to exit before others

8 years agochanges for disk checkpointing to write into different files in each partition
Xiang Ni [Wed, 30 Jan 2013 07:23:13 +0000 (01:23 -0600)]
changes for disk checkpointing to write into different files in each partition

8 years agocreat different directories for different replicas
YanhuaSun [Thu, 31 Jan 2013 00:36:59 +0000 (18:36 -0600)]
creat different directories for different replicas

8 years agoThis instance got missed last time Yanhua and I made replica compatible with
Nikhil Jain [Mon, 28 Jan 2013 18:12:37 +0000 (10:12 -0800)]
This instance got missed last time Yanhua and I made replica compatible with
persistent - fixing it now

8 years agoReplacing replica with partition
Nikhil Jain [Mon, 28 Jan 2013 09:00:01 +0000 (01:00 -0800)]
Replacing replica with partition

8 years agoAdding option for using replica as an alternate to partition
Nikhil Jain [Mon, 28 Jan 2013 08:54:47 +0000 (00:54 -0800)]
Adding option for using replica as an alternate to partition

8 years agoRemoving the local usage of CmiAssert which is bound to fail in replicated
Nikhil Jain [Fri, 1 Feb 2013 09:50:54 +0000 (03:50 -0600)]
Removing the local usage of CmiAssert which is bound to fail in replicated
environment

Conflicts:
src/arch/gemini_gni/machine-persistent.c

8 years agoFix for VC++ - it does not like variable declarations after other code
Nikhil Jain [Thu, 24 Jan 2013 07:18:24 +0000 (01:18 -0600)]
Fix for VC++ - it does not like variable declarations after other code

8 years agoHandle persistent comm in Gemini correctly:
Nikhil Jain [Tue, 22 Jan 2013 05:59:58 +0000 (23:59 -0600)]
Handle persistent comm in Gemini correctly:

8 years agoAdd support for +stdout; Change nomenclature of functions used by partitions
Nikhil Jain [Fri, 1 Feb 2013 09:46:54 +0000 (03:46 -0600)]
Add support for +stdout; Change nomenclature of functions used by partitions

Conflicts:
src/arch/util/machine-common-core.c

8 years agoadding extern
Xiang Ni [Tue, 4 Dec 2012 18:48:56 +0000 (12:48 -0600)]
adding extern

8 years agoAdding converse level functions for sending messages to remote replicas
Nikhil Jain [Mon, 29 Oct 2012 09:37:36 +0000 (04:37 -0500)]
Adding converse level functions for sending messages to remote replicas

8 years agoRemove usage of functions not meant to be exposed
Nikhil Jain [Fri, 1 Feb 2013 09:45:08 +0000 (03:45 -0600)]
Remove usage of functions not meant to be exposed

8 years agoFirst commmit to enable partition logic - added the concepts of localrank, and
Nikhil Jain [Fri, 1 Feb 2013 09:36:23 +0000 (03:36 -0600)]
First commmit to enable partition logic - added the concepts of localrank, and
global ranks; modifying LRTS to receive destNode; modifying converse to pass
global dest rank; trying to take care of smp, may not work yet; not addressing
optimization such as pxshm, xpmem, persistent - too much for one go; attempt is
to get it working for mpi layer, and then see how much effort is needed to other
layers - if I have planned it correct, the effort should be minimal; finally
only trying to get things working for layers that support Lrts

Conflicts:
src/arch/mpi/machine.c
src/arch/util/machine-common-core.c

8 years agoBug fix - in an earlier commit, the name of variable got changed from CMK_SMP to
Nikhil Jain [Mon, 4 Feb 2013 05:54:35 +0000 (23:54 -0600)]
Bug fix - in an earlier commit, the name of variable got changed from CMK_SMP to
CMK_SMK and this caused issues while running in SMP mode

8 years agoBegining work on interoperatin in non-mpi layer
Nikhil Jain [Sun, 3 Feb 2013 03:32:59 +0000 (19:32 -0800)]
Begining work on interoperatin in non-mpi layer

8 years agoAdding a time sharing example for Interoperation
Nikhil Jain [Sun, 3 Feb 2013 02:44:42 +0000 (18:44 -0800)]
Adding a time sharing example for Interoperation

8 years agoAdding async as a build target for pami build on BGQ
Nikhil Jain [Fri, 1 Feb 2013 07:43:41 +0000 (07:43 +0000)]
Adding async as a build target for pami build on BGQ

8 years agoMerge branch 'pami-lrts-ready' of charmgit:charm into charm
Nikhil Jain [Fri, 1 Feb 2013 03:29:36 +0000 (03:29 +0000)]
Merge branch 'pami-lrts-ready' of charmgit:charm into charm

8 years agofix a bug of message size alignment in persistent messages
Yanhua Sun [Fri, 1 Feb 2013 02:30:44 +0000 (21:30 -0500)]
fix a bug of message size alignment in persistent messages

8 years agofix the size of memory for L2Atomic queue
Nikhil Jain [Wed, 30 Jan 2013 20:19:54 +0000 (20:19 +0000)]
fix the size of memory for L2Atomic queue

8 years agoreserve one slot in L2Atomic Queue for node queue
Nikhil Jain [Wed, 30 Jan 2013 16:30:07 +0000 (16:30 +0000)]
reserve one slot in L2Atomic Queue for node queue

8 years agoDisabling LRTS specific List send till I can find a test case to verfiy their
Nikhil Jain [Wed, 30 Jan 2013 15:57:29 +0000 (15:57 +0000)]
Disabling LRTS specific List send till I can find a test case to verfiy their
correctness

8 years agoRemove the partition related code;
Nikhil Jain [Wed, 30 Jan 2013 15:49:11 +0000 (15:49 +0000)]
Remove the partition related code;

8 years agoAdding PAMI specific List Send functions
Nikhil Jain [Wed, 30 Jan 2013 12:27:29 +0000 (12:27 +0000)]
Adding PAMI specific List Send functions

8 years agoAdding async progress based implementation a build target
Nikhil Jain [Wed, 30 Jan 2013 11:47:45 +0000 (11:47 +0000)]
Adding async progress based implementation a build target

8 years agoReverting a change made by Yanhua in the previous commit - one cannot call
Nikhil Jain [Wed, 30 Jan 2013 10:27:01 +0000 (10:27 +0000)]
Reverting a change made by Yanhua in the previous commit - one cannot call
CmiBarrier after the context has been destroyed.

8 years agoadd L2Atomic Support, it compiles
Nikhil Jain [Wed, 30 Jan 2013 07:17:28 +0000 (07:17 +0000)]
add L2Atomic Support, it compiles

8 years agoBug Fix - I think targeting context to be random was a bad idea. Multiple
Nikhil Jain [Mon, 28 Jan 2013 14:19:42 +0000 (14:19 +0000)]
Bug Fix - I think targeting context to be random was a bad idea. Multiple
threads can be pushing to the same PE if the context is randomly chosen

8 years agoSome code clean up; bug fix in LrtsSendFunc - it is passed PE, not rank; but
Nikhil Jain [Mon, 28 Jan 2013 13:23:31 +0000 (13:23 +0000)]
Some code clean up; bug fix in LrtsSendFunc - it is passed PE, not rank; but
there is still a bug somewhere, looking for it

8 years agoadd CMIQueue macro to expand LRTS queue to use common pcqueue or not
Yanhua Sun [Mon, 28 Jan 2013 00:03:34 +0000 (00:03 +0000)]
add CMIQueue macro to expand LRTS queue to use common pcqueue or not

8 years agoFix the way exit happens
Nikhil Jain [Mon, 28 Jan 2013 10:27:18 +0000 (10:27 +0000)]
Fix the way exit happens

8 years agoadd persistent message files and changes for PAMI BGQ
Yanhua Sun [Fri, 25 Jan 2013 16:45:10 +0000 (16:45 +0000)]
add persistent message files and changes for PAMI BGQ

8 years agoBug fixes, changes to make pamilrts work - two version work now...
Nikhil Jain [Sat, 26 Jan 2013 10:53:41 +0000 (10:53 +0000)]
Bug fixes, changes to make pamilrts work - two version work now...

non smp and smp without xlc - as of now xlc is not supported because cldb is not
being supported using xlc

8 years agoChange from a pointer to actual data - bug fix
Nikhil Jain [Fri, 1 Feb 2013 00:09:28 +0000 (18:09 -0600)]
Change from a pointer to actual data - bug fix

8 years agoModifying charmrun to behave corectly if launched from within a script
Nikhil Jain [Fri, 1 Feb 2013 00:08:29 +0000 (18:08 -0600)]
Modifying charmrun to behave corectly if launched from within a script

8 years agoAdding a new target pamilrts-bluegeneq
Nikhil Jain [Thu, 24 Jan 2013 09:01:23 +0000 (09:01 +0000)]
Adding a new target pamilrts-bluegeneq

8 years agoFirst commit for porting PAMI to lrts
Nikhil Jain [Wed, 9 Jan 2013 08:45:41 +0000 (08:45 +0000)]
First commit for porting PAMI to lrts

8 years agobug fix in get-based persistent scheme on Gemini
Yanhua Sun [Fri, 1 Feb 2013 00:19:31 +0000 (19:19 -0500)]
bug fix in get-based persistent scheme on Gemini

8 years agoRevert "minor"
Yanhua Sun [Thu, 31 Jan 2013 19:44:03 +0000 (13:44 -0600)]
Revert "minor"

This reverts commit 638c8e6afb9093b4c223e1518da235db023a38de.

8 years agominor
Yanhua Sun [Thu, 31 Jan 2013 19:39:27 +0000 (13:39 -0600)]
minor

8 years agobluegenq: Correct const error on revised CmiMemcpy macro
Phil Miller [Thu, 31 Jan 2013 06:12:57 +0000 (06:12 +0000)]
bluegenq: Correct const error on revised CmiMemcpy macro

8 years agobluegenq: Conform CmiMemcpy macro to best practices
Phil Miller [Thu, 31 Jan 2013 06:05:03 +0000 (00:05 -0600)]
bluegenq: Conform CmiMemcpy macro to best practices

8 years agoDrop support for XT3 in TopoManager
Nikhil Jain [Thu, 31 Jan 2013 02:04:20 +0000 (20:04 -0600)]
Drop support for XT3 in TopoManager

8 years agotests: sdag: add the proper target for running the 'case' statement test
Jonathan Lifflander [Wed, 30 Jan 2013 22:55:42 +0000 (16:55 -0600)]
tests: sdag: add the proper target for running the 'case' statement test

8 years agoFixing bug in failure detection in message-logging.
Esteban Meneses [Wed, 30 Jan 2013 22:37:15 +0000 (16:37 -0600)]
Fixing bug in failure detection in message-logging.

8 years ago__sync version of atomic increment and decrement were switched incorrectly. fixed
Gengbin Zheng [Wed, 30 Jan 2013 17:48:44 +0000 (11:48 -0600)]
__sync version of atomic increment and decrement were switched incorrectly. fixed

8 years agomove debugLog creating code to common code
Yanhua Sun [Wed, 30 Jan 2013 17:18:44 +0000 (11:18 -0600)]
move debugLog creating code to common code

8 years agoMerge branch 'charm' of charmgit:charm into charm
Yanhua Sun [Wed, 30 Jan 2013 05:33:51 +0000 (23:33 -0600)]
Merge branch 'charm' of charmgit:charm into charm

8 years agominor, add barrier in lrtsexit
Yanhua Sun [Wed, 30 Jan 2013 05:33:44 +0000 (23:33 -0600)]
minor, add barrier in lrtsexit

8 years agoCharj: Sdag bug fixed
Bilge Acun [Wed, 30 Jan 2013 04:15:48 +0000 (22:15 -0600)]
Charj: Sdag bug fixed
it was caused by a wrong string template grammer

8 years agoin-mem chkpt: initialize pup variable
Michael Robson [Wed, 30 Jan 2013 00:46:17 +0000 (18:46 -0600)]
in-mem chkpt: initialize pup variable

Fix cppcheck 1.57 error
[src/ck-core/ckmemcheckpoint.C:1089]: (error) Uninitialized variable: numElements

8 years agoloc array map: initialize error checked variable
Michael Robson [Wed, 30 Jan 2013 00:42:46 +0000 (18:42 -0600)]
loc array map: initialize error checked variable

Fix cppcheck 1.57 error
[src/ck-core/cklocation.C:460]: (error) Uninitialized variable: flati

8 years agochkpt: intialize several pup variables
Michael Robson [Wed, 30 Jan 2013 00:36:06 +0000 (18:36 -0600)]
chkpt: intialize several pup variables

Fix cppcheck 1.57 errors
[src/ck-core/ckcheckpoint.C:183]: (error) Uninitialized variable: _numReadonlies
[src/ck-core/ckcheckpoint.C:235]: (error) Uninitialized variable: n
[src/ck-core/ckcheckpoint.C:395]: (error) Uninitialized variable: numGroups
[src/ck-core/ckcheckpoint.C:454]: (error) Uninitialized variable: numNodeGroups
[src/ck-core/ckcheckpoint.C:511]: (error) Uninitialized variable: numElements
[src/ck-core/ckcheckpoint.C:241]: (error) Uninitialized variable: chare_type

8 years agomsg replay: initalize pup variable
Michael Robson [Wed, 30 Jan 2013 00:26:36 +0000 (18:26 -0600)]
msg replay: initalize pup variable

Fix cppcheck 1.57 errors
[src/ck-core/ck.C:2404]: (error) Uninitialized variable: num_moves
[src/ck-core/ck.C:2405]: (error) Uninitialized variable: num_moves

8 years agock-core: intialize pup variable
Michael Robson [Wed, 30 Jan 2013 00:03:14 +0000 (18:03 -0600)]
ck-core: intialize pup variable

Fix cppcheck 1.52 error
[src/ck-core/ck.C:366]: (error) Uninitialized variable: migCtor

8 years agovirt char: initialize pup variable
Michael Robson [Tue, 29 Jan 2013 23:53:30 +0000 (17:53 -0600)]
virt char: initialize pup variable

Fix cppcheck 1.52 error
[src/ck-core/ck.h:118]: (error) Uninitialized variable: s

8 years agotrace-tau: Don't subscript the event array by -1
Phil Miller [Wed, 30 Jan 2013 00:36:34 +0000 (18:36 -0600)]
trace-tau: Don't subscript the event array by -1

8 years agodocs: Define all term macros with a trailing xspace, so that we don't need silly...
Phil Miller [Tue, 29 Jan 2013 20:27:01 +0000 (14:27 -0600)]
docs: Define all term macros with a trailing xspace, so that we don't need silly slashes or braces