| Age | Commit message (Collapse) | Author |
|
1. Adds support for using cuobjdump for both ptx and ptxplus execution. This has been tested with CUDA 4.0
. Ptxplus is no longer supported through decuda/decuda_to_ptxplus
2. Adds support for converting the SASS output by cuobjdump to ptxplus. This has been tested with CUDA 4.0
. The old path that extracts ptx from cubin files is still preserved
3. Adds a bank group model. (WARNING: memory config has changed, please adapt yours). To disable the bank groups model, set nbkgrp to 1 and tCCDL and tRTPL to 0
Diff the configuration files to learn about how to use those new options.
Merging
//depot/gpgpu_sim_research/fermi-test/distribution/...
to //depot/gpgpu_sim_research/fermi/distribution/...
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 12023]
|
|
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 11898]
|
|
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 10963]
|
|
Does not support sm_20
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 10951]
|
|
for constants to not appear in some cases.
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 10931]
|
|
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 10903]
|
|
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 10902]
|
|
the work of CL10697 which undone the changes to one file (stream_manager.cc). Here I undo the changes for 4 make files and more importantly the setup_enviroment file where I uncommented the include path of openCL as the case before CL10695, and I kept Tim changes to the CUDA path done later in CL10723 which he made originally to undo the work I did but didn't uncommet the openCL path which I do here.
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 10730]
|
|
cudasim.
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 10695]
|
|
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 10478]
|
|
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 9749]
|
|
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 8316]
|
|
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 8289]
|
|
except it reads and stores from two adjacent registers instead of a single 64 bit register. ".ff64" instructions are now printed in decuda_to_ptxplus. Support in the simulator for ".ff64" has been added but it is untested.
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 8278]
|
|
to decuda_to_ptxplus to support this.
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 8229]
|
|
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 8119]
|
|
Moved getDecuda into distribution as well.
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7943]
|
|
Edited baseline_ptxplus.config so the ptxplus regression runs properly.
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7942]
|