weft - CPEN 511 Project - Weft: Improving SIMD Utilization through MIMD-like Co-issue and Thread Compaction

Age	Commit message (Collapse)	Author
2017-07-11	Update README	gpgpu-sim
	updating description of CUDA versions
2017-07-11	Update CHANGES	gpgpu-sim
	added mention of a couple of recent changes
2017-07-11	Merge pull request #38 from tgrogers/travis-mods	gpgpu-sim
	Changing the regressions to compile + run different GPGPU-Sim config + verify functional correctness
2017-07-11	Disbaling the maxwell config	tgrogers

2017-07-11	reducing the reundant text in the config defintions	tgrogers

2017-07-11	trvis tick?	tgrogers

2017-07-11	looks like i need to explicitly include docker??	tgrogers

2017-07-10	Attempting to parallelize the build across all the configs at once	tgrogers

2017-07-10	pointing a the latest version of the docker image that already has the new ↵	tgrogers
	pbs_mom built in it.
2017-07-10	Now that mom stays up, lets try this again...	tgrogers

2017-07-10	forgot to change dirs	tgrogers

2017-07-10	new build got us past that fail - now lets see if it can stay up...	tgrogers

2017-07-10	After diggin in the torque source code, it looks like it might be this posix ↵	tgrogers
	memlock stuff that is killing the travis install. I guess it is possible that the server disables this feature on whatever infrastructure they are using. I am trying to rebuild pbs_mom without the memlock and see how much further we get.
2017-07-10	a short in the dark...	tgrogers

2017-07-10	another test of pbs_mom	tgrogers

2017-07-10	More info in logs?	tgrogers

2017-07-10	Let's test mom by lunching it over and over	tgrogers

2017-07-09	is it hosts?	tgrogers

2017-07-09	lookimg at messages	tgrogers

2017-07-09	Adding another slash for the logvel 0 this time maybe...	tgrogers

2017-07-09	Adding an extra slash for the loglevel	tgrogers

2017-07-09	increasing mom lovel to see why it is continously failing	tgrogers

2017-07-09	again, try to restart	tgrogers

2017-07-09	Adding another check for the pbs_mom	tgrogers

2017-07-09	Debugging non-running mom	tgrogers

2017-07-09	new debugging commandline	tgrogers

2017-07-09	A test for what is wrong with torque	tgrogers

2017-07-09	What works, sort of...	tgrogers

2017-07-09	again with the parallel	tgrogers

2017-07-09	trying out parallel builds	tgrogers

2017-07-09	Call the monitor in verbose mode?	tgrogers

2017-07-09	giving runner permissions to the distributions directory	tgrogers

2017-07-09	specify rw for the mapped directory	tgrogers

2017-07-09	Magic line	tgrogers

2017-07-09	trying pwd	tgrogers

2017-07-09	Adding in the first pass at the actual regression commands	tgrogers

2017-07-09	wrong docker tag	tgrogers

2017-07-09	wrong docker tag	tgrogers

2017-07-09	What is going on?	tgrogers

2017-07-09	Adding in some mroe stuff to debug travis	tgrogers

2017-07-09	Commit to wake travis up?	tgrogers

2017-07-08	Adding in a test to see what happens when travis tries to build with docker.	tgrogers

2017-07-06	Adding the correct dependency for the detailed_version file. In order to ↵	tgrogers
	updatet the built number output when we run gpgpu-sim we need to recompile cuda-sim everytime the detailed_version has changed
2017-05-17	Merge pull request #37 from tgrogers/tim-embed-commit	Tim Rogers
	Embedding the gpgpu-sim version information (the commit number that was built) in the shared object executable.
2017-05-17	Adding in the embedded .so bit	tgrogers

2017-05-17	Getting rid of the branch since it gets weird, espeically in travis	tgrogers

2017-05-17	Fixing a bug with where the build version is made. Moveing it after the ↵	tgrogers
	right directories ect are defined
2017-05-17	Changing the version detection to be much more detailed. Now the git commit ↵	tgrogers
	# and branch will be embedded in the built executable and print out when gpgpu-sim runs
2017-05-09	Merge pull request #35 from echoedit/dev	Tim Rogers
	Fix next block addr to link predicate ret block to consecutive block
2017-05-09	Fix next block addr to link predicate ret block to consecutive block	Mengchi Zhang
	The block containing predicate ret instruction should add the consecutive block to its successor_ids set. next_addr should be assigned with current instruction address add instruction size instead of 1. Signed-off-by: Mengchi Zhang <[email protected]>