Lines of Code
[root]/CCA/Components/Schedulers
| Author | Changes | Lines of Code | Lines per Change |
|---|---|---|---|
| Totals | 353 (100.0%) | 5364 (100.0%) | 15.1 |
| ahumphrey | 101 (28.6%) | 3002 (56.0%) | 29.7 |
| allen | 128 (36.3%) | 1106 (20.6%) | 8.6 |
| mcconnell | 4 (1.1%) | 655 (12.2%) | 163.7 |
| jholmen | 34 (9.6%) | 406 (7.6%) | 11.9 |
| harman | 15 (4.2%) | 82 (1.5%) | 5.4 |
| dav | 5 (1.4%) | 65 (1.2%) | 13.0 |
| jas | 63 (17.8%) | 47 (0.9%) | 0.7 |
| derekhar | 1 (0.3%) | 1 (0.0%) | 1.0 |
| bpeterson | 2 (0.6%) | 0 (0.0%) | 0.0 |
clean up of the loop and removed of the need for the map::at call
0 lines of code changed in 2 files:
Additionally report ghost cell requirements for ForgenVariable debug info.
0 lines of code changed in 2 files:
Changed variable name, to be consistent with previous commit, 59061
1 lines of code changed in 1 file:
clean up for PIDX
5 lines of code changed in 3 files:
Generalize the concept of a task graph having tasks with "distal" halo requirements, meaning the number, order, and appearance of TGs with distal requires is arbitrary. This is a much more robust way to determine distal neighborhood creation, etc. Prior to this, the code was quite brittle, depending on the distal TG being last in the order.
23 lines of code changed in 7 files:
* Fix MPI buffer truncation error in threaded RMCRT Examples tests.
* Some small cleanup in UnifiedScheduler (USE_PACKING is defined in MPIScheduler, not needed in Unified).
14 lines of code changed in 3 files:
* Fix RMCRT memory bloat.
* Move from referenceing SHRT_MAX to "req->m_num_ghost_cells >= MAX_HALO_DEPTH"
* Fix unitialized vars.
3 lines of code changed in 1 file:
removed redundant code and cleaned up formating
90 lines of code changed in 1 file:
added support to handle sole vars similar to reduction vars. Both being global vars
112 lines of code changed in 3 files:
added support to handle sole vars similar to reduction vars. Both being global vars
49 lines of code changed in 1 file:
addTask()
- added bulletproofing when temporal scheduling is used. tg_num must be < m_num_task_graphs.
7 lines of code changed in 1 file:
added support to handle sole vars similar to reduction vars. Both being global vars
2 lines of code changed in 1 file:
* Brace initializers for POD and pointer_type class members.
* cbegin() and/or cend() for const_iterators, especially for methods that return a const_iterator, e.g., Level::const_patch_iterator Level::patchesBegin() const.
* Use prefix ++ operator (e.g., ++iter) for more complicated types, avoiding tmp creation.
* Prefer using const when possible.
* Use auto keyword, leaving correct type deduction to the compiler.
69 lines of code changed in 5 files:
Cleanup under the hood as the most recent TG work/project moves forward. Mostly cosmetic changes for readability.
28 lines of code changed in 3 files:
added bulletproofing to the map:;at method usage
67 lines of code changed in 1 file:
Fix more inaccurate error mesgs in DWDatabase.
2 lines of code changed in 1 file:
Fix accuracy of error message when putting foreign vars into DW.
Reorder appearance of methods in OnDemandDW to match header... send/recvMPI methods adjacent in impl file.
A few minor cosmetic changes elswhere.
119 lines of code changed in 3 files:
Massive cleanup and refactoring in OnDemandDW. This source code was diffuclut to read, and I'm needing to make significant changes in the MPI engine to support optimizations in TG dependency analysis (in the context of RMCRT radiation with threaded schedulers).
1454 lines of code changed in 5 files:
added DOUT for run a task acually runs
9 lines of code changed in 3 files:
Update and revive DW task access checking and RunningTaskInfo code.
Fixed a few other typos and formatting in these areas while under the hood.
96 lines of code changed in 4 files:
added perpatch to the list
4 lines of code changed in 1 file:
Infrastructure restructuring and cleanup while starting last thesis-related task graph work.
Big things that I will always change when I see them.... and I saw much of this recently:
1.) ALWAYS use {} for 1-line conditionals.
if (<predicate>) {
// do stuff
}
2.) Use C++ static _cast over C-style casts, static_cast<int>(myvar) is checked at compile time, C casts are not and can fail at runtime.
3.) use "nullptr" over "0 or NULL" in the context of pointers and their assignment. Using NULL or 0 can be ambigous, e.g., compiler converting to non-pointer type 'int' from NULL (nullptr can't be assigned to an integral type). nullptr is a keyword and usage/intent is clear. From the standard: "The pointer literal is the keyword nullptr. It is an rvalue of type std::nullptr_t.".
4.) Things like Dout/Debugstreams that are specific to a compilation unit should be placed into an unnamed namespace (which is a utility to make an identifier translation unit local). The effect is almost identical to decalring something static.
5.) Use meanigful names, a bit lengthy is OK if it helps clarity. Remember, the compiler doesn't care about your names (values just go into registers), but the people maintaining your code do.. Code is meant to be read and maintained.
6.) Whitespace is our friend, but 80 column lines went away long ago.....
243 lines of code changed in 4 files:
* Added the GPUIDs SCI_DEBUG environment variable.
This environment variable reports information helpful for uniquely identifying GPUs when using multiple MPI processes and GPUs per node.
26 lines of code changed in 1 file:
Specialize std::hash structure and inject into the std namespace so that VarLabelMatl<DomainType> can be used as a key in std::unordered_map.
Removed older version that publicly inherited from std::unary_function, which is deprecated as of C++11.
11 lines of code changed in 1 file:
Remove all traces of sci_hashmap. With Uintah being fully transitioned to C++11, these old TR1, GNU_HASHMAP and other extension hacks are unnecessary, confusing and error prone. We just use std::unordered_map, etc.
Generate new configure script.
35 lines of code changed in 2 files:
removed the intial value and set the initial count to zero
6 lines of code changed in 1 file:
Added comments regarding the side affect of writing out of sync output/checkpoint files
85 lines of code changed in 3 files:
* Cleaned up TaskDBG use in Kokkos-OpenMP Scheduler.
0 lines of code changed in 4 files:
* Bulletproof Kokkos-OpenMP Scheduler TaskDBG for non-Kokkos builds.
17 lines of code changed in 1 file:
* Added support for the TaskDBG SCI_DEBUG environment variable within the Kokkos-OpenMP Scheduler.
12 lines of code changed in 1 file:
Remove PAPI support.
This direct support is no longer necessary, as almost all profiling tools sit on top of PAPI. No sense in us maintaining code that's avialble through professional tools.
0 lines of code changed in 1 file:
Complete the removal of RunLengthEncoder.
2 lines of code changed in 1 file:
* Remove RunLengthEncoder and supporting classes, etc. This is no longer used. Removal will help clarity regarding Uintah::Variable functionality and features. This change has neccesitated explicit inclusion of "unistd.h" in a few spots, most specifically - MPM, for getpid() calls.
* Remove build system support for IPM and MAGMA libs, which have never been used. MAGAMA in fact, may ultimately hender Kokkos GPU asynchrony and performance.
* Generate new configure script with enable_option_checking bulletproofing.
* Delete unused header files.
* A few other cosmetic changes while under the hood, formatting, white space, etc.
2 lines of code changed in 1 file:
variable and comment clean up to refer to the material manager rather than the state
18 lines of code changed in 16 files:
addition logic for nullptr material pointer and cleanup of unsigned warnings
2 lines of code changed in 2 files:
removal of shared state and introduction of a material manager
29 lines of code changed in 14 files:
made the recompute and abort reduction vars
49 lines of code changed in 5 files:
This fixes the bug that is causing GPU trunk nightly regression test failures. I cut down the amonut of threads per kernel, for some machines running in debug mode, 1024 threads per block for this particular kernel is too much
0 lines of code changed in 2 files:
change from restart to recompute time step so to differentiate from restarting a time step from a checkpoint
32 lines of code changed in 6 files:
This fixes an issue with (older/non-compliant?) file systems with respect to deleting directories.
For many years We had a 'hack' in place for AIX file systems to use 'stat' instead of the cheaper
"file->d_type" check. It turns out that we should (have been) use the hack anytime that file->d_type
is DT_UNKNOWN, not just under AIX. This should fix the problem that some people were seeing on some
file systems where (checkpoint) diretories were not being deleted when they should have been.
M CCA/Components/DataArchiver/DataArchiver.h
M CCA/Components/DataArchiver/DataArchiver.cc
- Const'ify a number of method params that are not changed in the functions (and thus should be declared const).
- Add some white space for readability.
- Actually print out warning messages when old checkpoint directories fail to be deleted.
M CCA/Components/Schedulers/OnDemandDataWarehouse.cc
M CCA/Components/SimulationController/SimulationController.cc
- Purely cosmetic: Add some white space for readability.
M CCA/Ports/Output.h
- Match updated interface (ie, use of const) for DataArchiver.
- Add some white space for readability.
M Core/OS/Dir.h
M Core/OS/Dir.cc
- Add 'const' due to addition of const above.
- Add some white space for readability.
- Fix the check that we use to determine if a 'file' is a directory (see above).
M Core/OS/Dir_boost.cc
- Add const, white space, and {} for if statement.
3 lines of code changed in 1 file:
(52 more)