Skip to content
Toggle navigation
P
Projects
G
Groups
S
Snippets
Help
las3_pub
/
predictable_parallel_patterns
This project
Loading...
Sign in
Toggle navigation
Go to a project
Project
Repository
Issues
0
Merge Requests
0
Pipelines
Wiki
Members
Activity
Graph
Charts
Create a new issue
Jobs
Commits
Issue Boards
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Charts
c25e61343f5ed2d3acabb033efa4305f91ad86ce
Switch branch/tag
You can move around the graph by using the arrow keys.
Begin with the selected commit
Created with Raphaël 2.2.0
9
Feb
4
3
Aug
31
Jul
22
4
30
Jun
28
24
23
22
13
11
9
8
4
3
29
May
27
26
19
18
17
16
15
24
Apr
23
17
15
14
6
31
Mar
30
26
25
24
18
13
23
Feb
18
17
9
5
3
30
Jan
29
27
26
24
23
22
21
20
13
10
4
20
Dec
5
4
2
29
Nov
27
25
19
7
6
5
2
Oct
1
30
Sep
16
2
30
Aug
2
1
31
Jul
30
29
24
22
19
11
10
9
3
28
Jun
24
20
18
17
14
12
11
10
7
6
5
4
3
28
May
16
14
13
10
7
6
3
2
30
Apr
18
17
16
15
12
11
10
9
8
5
4
3
2
1
29
Mar
27
26
25
22
21
20
19
18
15
changed markdown syntax highlighting
master
master
Added Odroid XU4 as another supported board
Fix top-pointer help (potentially not lock free without the fix).
Add range library to mentioned lincences.
Add licence for external projects.
Rework README
Remove notes fol PLS_v1.
Remove unused data structures (old work stealing deques).
Fix compiler warnings.
Remove old data-flow code from PLS_v1.
Add profiling information to benchmarks and matrix_then_fft benchmark.
Minor optimizations to the thread sleeping mechanism.
Minor adjustment to coordinating thread spawns (catch seems to have problems with the other implementation).
Update: add spawn_and_sync to final version for benchmarks.
Final version ready for benchmarking.
Add support for spawn-and-sync statement for not wasting steals.
Stable version of Re-Work resource-trading implementation to fix race condition.
WIP: Re-Work resource-trading implementation to fix race condition.
WIP: Re-Work resource-trading implementation to fix race condition.
Final preparations for single-app benchmark runs.
Allow flexible problem sizes and periodic execution for benchmarks.
Allow run-time size options for all benchmarks.
Add pre-allocation to memory measures.
Add memory measures to benchmarks.
Add standard divide and conquer matrix multiplication test for comparison.
WIP: Start work on divide and conquer example with temporary buffers.
Add support for serial tail calls.
Add prototype support for other, strain_local resources.
Add scheduler_active flag and PLS_SERIAL_ELUSION compile option.
Touch all stack memory in scheduler.
Sync benchmark runner.
Add proper build flavour system and fix some worker-sleep issues.
Add first, experimental version of 'sleep on empty flag'.
Finish up DAG profiler.
Split profiling code into cpp files.
Add output for basic stats: T_1, T_inf, T_wall, memory and basic graph drawing.
Add simple profiling code that records dynamic executed DAG.
Remove EasyProfiler references.
Refine setup process for banana pi
Simplify: use raw pointers instead of optional wrappers on deque.