summaryrefslogtreecommitdiff
path: root/libopencl/opencl_runtime_api.cc
AgeCommit message (Collapse)Author
2020-04-07Merge remote-tracking branch 'localpub/dev' into devtgrogers
2019-10-17libopencl: Calculate a valid offset in bind_args()Roy Spliet
v2: Adhere to alignment requirements. v3: Small style fix. Signed-off-by: Roy Spliet <[email protected]>
2019-10-17libopencl: Perform PDOM analysisRoy Spliet
Signed-off-by: Roy Spliet <[email protected]>
2019-10-17libopencl: Stub clGetProgramBuildInfo()Roy Spliet
Assumes everything is ok, even if it isn't. Allows for some Rodinia benchmarks to fail slightly later. v2: Use available case macros. Signed-off-by: Roy Spliet <[email protected]>
2019-09-12Remove GPGPUsim_ctx_ptr()Mengchi Zhang
Signed-off-by: Mengchi Zhang <[email protected]>
2019-07-10Move g_globals and g_constantsMengchi Zhang
Signed-off-by: Mengchi Zhang <[email protected]>
2019-07-09Move g_ptx_sim_modeMengchi Zhang
Signed-off-by: Mengchi Zhang <[email protected]>
2019-07-09Move g_ptx_kernel_countMengchi Zhang
Signed-off-by: Mengchi Zhang <[email protected]>
2019-07-08Move cp_countMengchi Zhang
Signed-off-by: Mengchi Zhang <[email protected]>
2019-07-07g_keep_intermediate_filesMengchi Zhang
Signed-off-by: Mengchi Zhang <[email protected]>
2019-07-01Move g_debug_ir_generation and GPGPUSim_InitMengchi Zhang
Signed-off-by: Mengchi Zhang <[email protected]>
2019-06-12Integrate ptxinfo into gpgpu_contextMengchi Zhang
Signed-off-by: Mengchi Zhang <[email protected]>
2019-06-12Move some function from ptx_loader to gpgpu_contextMengchi Zhang
Signed-off-by: Mengchi Zhang <[email protected]>
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.
2016-10-31update opencl_runtime similar to cuda_runtimeAhmed ElTantawy
2016-05-15updating CHANGESAhmed ElTantawy
2014-08-14Forgot a file in CL 16503Tim Rogers
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 16505]
2014-08-14fixing broken opencl regression from last regression (when the CreateKernel ↵Ahmed El-Shafiey
does not specify error pointer) [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 15970]
2014-08-14Two small fixes in the OpenCL APIAhmed El-Shafiey
Review ID: 27001 [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 15969]
2014-08-14Removing a slew of code still compiled with gcc and the need for a bunch of ↵Tim Rogers
external C linkage [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 15912]
2014-08-14Fix for bug 53 and 54. Now clGetDeviceInfo() is returning CL_DEVICE_TYPE ↵Wilson Fung
property in the proper size. Also added code to determine the workgroup size automatically. [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 15798]
2014-08-14Fix for bug 51 and 55: Declaration of clCreateContextFromType() is updated ↵Wilson Fung
to match OpenCL 1.1 and later. The API call now accepts device types besides CL_DEVICE_TYPE_GPU. [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 15796]
2014-08-14 Fixing a slew of compiler warningsTim Rogers
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 15218]
2014-08-14Apply patch from Kito Cheng to update libopencl for checking NULL error code ↵Wilson Fung
pointer. [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 15043]
2014-08-14OpenCL newer diver fixHadi Jooybar
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 14554]
2014-08-14Integrated in CL14390.Inderpreet Singh
Fixed clCreateProgramWithSource to accept NULL or 0 as string length parameter Fixed replacement of printf modifiers when compiling OpenCL code (e.g. '%f') [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 14391]
2014-08-14Fixing bad memory references in opencl based on some disconnect between what ↵Tim Rogers
the nvidia opencl wraper expects and what our opencl_runtime_api provides. This was causing us to print grabage in the dumped ptx files generated by calls to oclPtxLog and valgrind complains about it. [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 13837]
2014-08-14Integration from tm-311 branch. Adding support for atomic operation with ↵Wilson Fung
generic memory space. Also adding a define to allow support for OpenCL 1.0 commands that are deprecated in OpenCL 1.1. [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 13722]
2014-08-14Fixing a bug in the opencl prebuilt ptx flow.Tim Rogers
If a kernel made modidifcations to the .cl code at runtime (like changed the block size or some other constants) then recompiled itself the saved_embedded_ptx would just get overwritten. Same on using the prebuilt ptx file - I always wanted to load the same ptx file even though there should have been more than one. [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 13217]
2014-08-14Adding stub function implementation to get EA_Ropa running on fermi branch.Wilson Fung
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 11311]
2014-08-14using rsync because of ssh having issues with concurreny copyingTim Rogers
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 11065]
2014-08-14Fixing the copy directoriesTim Rogers
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 11053]
2014-08-14Adding support to change the directory on the remote opencl compilationTim Rogers
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 11052]
2014-08-14Warnings cleanupTim Rogers
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 10753]
2014-08-14Fix for Bug 109 - memory alignment should be 256 bytes.Inderpreet Singh
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 10090]
2014-08-14Initial modifications to the opencl runtime API to allow the AMD sample ↵Tayler Hetherington
applications to run. Some of these modifications are temporary or incomplete and will be modified as necessary. [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 9942]
2014-08-14Integration change. Bringing in the remote opencl changes from mem_divergenceTim Rogers
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 9764]
2010-12-28- parameter memory and active threads now part of kernel_info_t:Tor Aamodt
Parameters are finalized at kernel launch, which means the contents of parameter memory are initialized. Kernel arguement names have a fixed order, hence same address should be assigned on subsequent kernel launches of same kernel in other streams provided the data size param_t::size of arguments for each kernel launch is identical (an assertion has been added to check this is true). - passing regression [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 8303]
2010-12-28- Checkpointing new support for concurrent kernel execution (CUDA only, not ↵Tor Aamodt
OpenCL) This changelist adds full support for streams supported by a new class, stream_manager and enables concurrent execution of kernels from different streams. - fast_regression.sh fails for simpleMultiCopy, simpleStreams (other tests passing) ** Known issues ** - Kernel parameter passing is not done correctly for concurrent kernel execution (somehow concurrentKernels is not affected by this): the parameters are stored inside function_info, which is shared among parallel kernel launches so that the values passed into the launch are likely to get overwritten if multiple grids are launched in parallel streams. - Statistics are printed out whenever the simulation thread runs out of cuda commands (doesn't make sense to print out when a kernel ends during concurrent kernel execution). This will probably require further tweaking so as to be more compatible with data collection scripts. [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 8302]
2010-10-241. adding top level configuration class and making shader and memory ↵Tor Aamodt
configuration components of this class. 2. clock memory pipeline no. subwarp times for each shader clock and increase rob-size for texture cache (trying to improve correlation, currently at 0.9218) 3. start to modify shader stats to add back features for visualizer (warp divergence distribution kind of working again) passing cuda 3.1 regression and ptxplus correlation tests [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7909]
2010-10-18Re-designed cache model:Tor Aamodt
- read only cache model with integrated mshrs (no L1D, yet); new cache interface should be easily extendable to support texture cache with latency fifo and separate tag/data arrays, though this is not yet added (currently tags and data arrays are not decoupled for texture) - new partition model using the above removes all old MSHRs, L1D etc... passing CUDA 3.1 regression [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7875]
2010-10-161. enable OpenCL to PTX compilation on remote host without NFSTor Aamodt
2. hopefully one-size-fits-all setup_environment script [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7866]
2010-10-09fixes problem with remote OpenCL PTX generationTor Aamodt
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7843]
2010-10-09integrate CL 7836 (running regressions on cluster)Tor Aamodt
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7841]
2010-10-09refactoring: moving texture reference mappings into gpgpu_t Tor Aamodt
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7835]
2010-10-081. refactoring cuda api code for loading PTX (removing external PTX loading ↵Tor Aamodt
entirely) 2. some bug fixes for warp_inst_t 3. creating a new class, gpgpu_t, which contains the functional "memory" state visible to all threads running on a GPU (doing this as part of my continuing effort to hunt down and eradicate every global variable that is not the top level "the gpu") 4. other misc. changes Almost passing CUDA 3.1 regression? oclHistogram keeps failing under torque, but does not fail when run on the command line from the same directory. [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7827]
2010-10-01integrating recent changes from fermi-test into fermiTor Aamodt
(i'll use "fermi" for more disruptive changes to the pipeline model such as updating the MSHRs and getting rid of the warp tracker, ripping out DWF, etc...) [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7805]
2010-08-23- oclMatrixMul passing (added support for dynamically allocating shared ↵Tor Aamodt
memory on kernel launch for OpenCL) - added some debugging printfs to oclMatrixMul [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7255]
2010-08-23- add '-keep' option to keep intermediate files (used for OpenCL only right now)Tor Aamodt
- remove redundant initialization code of options (which confuses new users) - make bool option variables bool type - remove some more extern decls [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7247]
2010-08-12- allow multiple clCreateBuffer wth same host_ptr (undefined behavior in ↵Tor Aamodt
OpenCL spec, but required for oclTranspose). - undo earlier fix that wasn't a fix [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7194]
generated by cgit v1.3 (git 2.53.0) at 2026-07-10 22:23:57 +0000