| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
|
|
|
|
Updating the interconnect simulator to properly check for dependencies
|
|
|
|
Fixing BankGroup Indexing Bug
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Interconnection assertion failing demystifying
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Improving GPU core model.
|
|
|
|
|
|
1- round robin inst issue for warp multiple schedulers
2- add sector mask in the memory request (to bused later for L2 sector cache)
3- Adding Fermi coalescer
4- Ensure different exen units are used in dual_issue mode
5- Report how many dual_issue happened
6- Adding oldest_first scheduler
|
|
Dev purdue integration
|
|
|
|
The way we detect the shell is not fool-proof and fails for jenkins, although we are using bash
|
|
|
|
Purdue's end
|
|
updating description of CUDA versions
|
|
added mention of a couple of recent changes
|
|
Changing the regressions to compile + run different GPGPU-Sim config + verify functional correctness
|
|
|
|
|
|
|
|
|
|
|
|
pbs_mom built in it.
|
|
|
|
|
|
|
|
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.
|
|
|
|
|
|
|
|
|
