Skip to content
GitLab
Explore
Projects
Groups
Topics
Snippets
Projects
Groups
Topics
Snippets
/
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Sign in
Toggle navigation
Menu
ras-devel
FLAG
Graph
34fca3c8df1f7490e657ee4671060b229c2dcb1b
Select Git revision
Branches
6
cuda-improvements
fix-hangs
master
default
net-thread-improvements
pipeline-rework
transpose-in-net
You can move around the graph by using the arrow keys.
Begin with the selected commit
Created with Raphaël 2.2.0
22
Jul
26
Jun
25
20
11
8
4
31
May
28
23
16
15
22
Apr
10
Sep
23
Jul
13
24
Jan
18
12
2
Aug
1
28
Jul
18
12
26
May
25
12
10
3
2
28
Apr
26
24
21
20
18
17
14
13
12
3
29
Mar
27
24
23
22
21
20
17
13
12
10
9
2
28
Feb
27
9
7
6
2
31
Jan
30
19
4
Oct
6
Sep
11
Aug
27
Jul
25
22
20
18
15
14
12
11
7
30
Jun
24
23
20
17
16
14
10
9
7
6
27
May
24
4
29
Apr
21
Mar
1
29
Feb
24
19
10
4
1
8
Dec
19
Nov
17
4
Sep
25
Aug
21
19
18
13
Jul
24
Jun
18
17
10
18
Feb
Updating repo for software review.
master transpos…
master transpose-in-net
Moved backup files from src directory to backup_flag_files folder underthe account of mruzinda on FLAG HPCs.
Compiled code and fixed a few issues after merge. RTBF, CALCORR, and PFBCORR were run and work as they should. No stalls or bad blocks.
Merging changes made to include transpose logic in net thread and remove transpose thread from HASHPIPE pipeline. The file that should be called in place of flag_net_thread.c in the pipeline is flag_net_transpose_thread.c and instead of flag_correlator_thread.c, use flag_corr_transpose_thread.c. These are specified in Makefile.am.
Net thread improvements as well as additional files for testing.
cuda-improvements
cuda-improvements
Strip net thread to re work processing logic
pipeline-rework
pipeline-rework
Merge branch 'net-thread-improvements' into 'master'
Add comments explainging some of the new binfo variables
net-thread-impr…
net-thread-improvements
Rework of mcnt packet processing logic
Attempt to resole mcnt processing logic
Include blocking busywait methods for flag input databuf
Prevent corsave from hanging on quit
Prevent save thread from hanging on a quit
Make indentation consistent through code
Format and clean net thread a little before working on it
Modify install script
fix-hangs
fix-hangs
modify install script to accept alternate install prefix from environment
Writing to unmodified binary file in flag_beamsave_thread.c
fix debug misspelling in hashpipe.c
Merge branch 'cuda-improvements' into 'master'
Pinned memory, streams and events implemented.
Most recent version of code used on telescope with setVersion script and soft links for lib and src to the versions folder
This contains versions of the scalloping and no scalloping versions of the pfb. The version installed in this commit is the scalloping one. To install versions, use rsync -a and copy the appropriate directories (lib and src) over.
Install script with no scalloping fix. Used for packet_gen.m right now.
PFB scalloping fix code modifications and additions.
Final changes to branch.
Just updating everything. Change made to Makefile for cuda JIT issue. It's the first commented line of code.
Made changes to transpose and databuf.h. In transpose thread, I changed line 185 to ACQUIRE, and in flag_databuf.h, I changed VERBOSE to 1 and DEBUG to 1.
Just pushing the install executable.
Reverted back to RTBF code without GPU transpose implementation.
Merge branch 'master' of https://github.com/rallenblack/flag_gpu
Merge pull request #11 from mitchburnett/master
forgot to track the new beamform thread
Merge branch 'master' of https://github.com/rallenblack/flag_gpu
New and modified matlab files
Merge pull request #10 from mitchburnett/master
This commit was made at the end of me leaving GBO during the middle of the July/Aug commissioning. Many of these updates have improved the performance of the pipeline. Performance measured in the frequency of stalled/hanging banks and bad blocks. We tuned core assignments, input block counts and more. Decreasesd the amount of time accessed by shared memory. Prevented packets sneaking through at the end of scans in the net thread. Removed the cpu transpose for the beamformer and implemented on the GPU.
Fixes to pfb corr mode. The GPU needed to flush the buffer that stores the last filter length worth of data. Also added new clean up logic lat waits until the net_thread has gone to the idle state. Also added a method for most threads that looks at the amount of locked blocks extending a hashpipe function
Updated MATLAB code
Merge pull request #9 from mitchburnett/master
Menu
Explore
Projects
Groups
Topics
Snippets