summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2018-07-25Merging conlicted travis line. Just use the new one.tgrogers
2018-07-23Getting the regressions to pass by explicitly setting the CUDA version. The ↵tgrogers
image now contains 2 CUDA versions and the default one is CUDA 9.1
2018-06-30Explicitly setting the env vars in the docker calltgrogers
2018-06-30Explicitly calling out CUDA in the 480 build tootgrogers
2018-06-30oops - you should set this if it does not exist, not the other way around...tgrogers
2018-06-30oopstgrogers
2018-06-30Explicitly saying what CUDA to usetgrogers
2018-06-30use CUDA 9.1tgrogers
2018-06-30Adding in a check to not blow away the PTXAS variable if you already had it settgrogers
2018-06-30oops - need a yml list of envstgrogers
2018-06-30Adding the 9.2 option to the PASCAL configtgrogers
2018-06-30Cleaning up the setup env message a bit and having the simulator properly ↵tgrogers
dump to stderr on error
2018-06-30Preventing bad git commit numbers from printing when you run setup_env from ↵tgrogers
a dir that is not the gpgpusim_distr dir
2018-06-30changing the flow s.t. you cannot compile the simulator with a different ↵tgrogers
version of CUDA from the app.
2018-06-29Fixing the PTXPLUS + new CUDA execution case.tgrogers
2018-06-29Allowing modern GPU configrations to properly run PTXPLUS.tgrogers
There are several problems with this: 1) To get proper occupancy, based on register file usage, you must querry ptxas with the sm version that corresponds to the register usage in the config file. To enable this, a new config option has been added that determines what sm version you pass to ptxas to compute reg-usage. This configuration option is always required in the gpgpusim.config file 2) If you are running PTXPLUS with a modern card (i.e. volta/pascal), you need ptxas from CUDA 9.1. However since PTXPLUS only supports sm_13 - you need a version of CUDA where cudaobjectdump supports sm_1x. This ended at CUDA 5 - and PTXPLUS requires CUDA 4.2. Therefore, to run PTXPLUS on a modern card, you need CUDA 4.2 + modern CUDA installed. To fascilitate this, a new envronment varaible is added and the setup envrionment script prints an appropraite warning if you are using a newer CUDA. We have tried to make this as fail-proof as possible - and die appropraitely when something is wrong.
2018-06-29Whitespace edits - removing tabs - replacing with spacestgrogers
2018-06-29Adding a print to indicate which PTX version we are actually usingtgrogers
2018-06-22Merge pull request #72 from tgrogers/devTim Rogers
Getting the simulator to compile with older versions of g++
2018-05-04Getting the simulator to compile with older versions of g++tgrogers
2018-04-11Merge pull request #65 from tgrogers/devTor Aamodt
Bunch of changes made to 3.x that help with data collection
2018-04-11This build status line was broken when there were cached filestgrogers
2018-04-11Merge remote-tracking branch 'upstream/dev' into devtgrogers
2018-04-05Nice to know when the simulaton thread actually gets the exit call under ↵tgrogers
normal operation. This helps immensely in a clustered environmrnt where jobs randomly fail
2018-04-05fixing gpu-tot-cycle bugMahmoud
2018-04-05adding new stats counterMahmoud
2018-04-05adding new counters for parition level para and L2 BWMahmoud
2018-04-05GPGPU-Sim should always print when the sim thread exits - if this doesn't ↵tgrogers
happen, something bad has happened
2018-04-02Adding in Mahmoud's changes to the print cache statstgrogers
2018-03-31Getting rid of our constant, annoying prints. Running workloads of any size ↵tgrogers
causes ridiculous output file sizes
2018-03-30When CDP was added - we started printing garbage stats after every memcpy. I ↵tgrogers
am stopping this by gaurding against it in the same way we do for updating the stat counts (checking if we have executed any cycles before printing). I am not sure what this does to CDP stat prints - as I am not sure what CDP kernels so to sim_cycles.... However EVERY other app except CDP is spewing junk to the output and it really messes up any per-kernel stat collection scripts.
2018-03-28Merge pull request #64 from aamodt/devTor Aamodt
fix compile error
2018-03-28Fixing a minor typo in an old configtgrogers
2018-03-28(attempt to) fix compile error in last checkinTor Aamodt
2018-03-28Merge remote-tracking branch 'upstream/dev' into devTor Aamodt
2018-03-28fix compile errors on Ubuntu LTS 16.04Tor Aamodt
2018-03-26Merge pull request #63 from pigrew/dev-printfFixgpgpu-sim
Don't directly pass strings to printf (beacuse they wouldn't be escap…
2018-03-26Merge pull request #62 from pigrew/dev-multiArchBreakpointgpgpu-sim
Provide portable (non-x86) breakpoint method which should work on all…
2018-03-26Merge pull request #61 from pigrew/dev-gitignoreUpdategpgpu-sim
Dev gitignore update
2018-03-26Merge pull request #60 from pigrew/dev-ParserFixgpgpu-sim
Dev parser fix
2018-03-26Don't directly pass strings to printf (beacuse they wouldn't be escapped). ↵Nathan Conrad
Clang gives warnings about this.
2018-03-26Provide portable (non-x86) breakpoint method which should work on all ↵Nathan Conrad
linuxes. Tested on PowerPC.
2018-03-26Ignore generated docs in gitignoreNathan J Conrad
2018-03-26Add vim swap files to gitignoreNathan J Conrad
2018-03-26Remove duplicate token in PTX parserNathan Conrad
2018-03-26.call instruction may have an empty argument listNathan Conrad
2018-03-25Merge pull request #59 from tgrogers/devgpgpu-sim
Apps compiled with newer versions of CUDA want the version number embedded in the so
2018-03-24The new CUDAs complain about version info not being embedded into the so ↵tgrogers
file. Having gpgpu-sim ebed some version info to stop the error complaints
2018-03-22Merge pull request #58 from tgrogers/devgpgpu-sim
Integrating some setup stuff and proper support or a maximum number of instructions (which was previously broken)
2018-03-22Fixing the break limit bugMahmoud
generated by cgit v1.3 (git 2.53.0) at 2026-07-11 18:33:18 +0000