diff options
| author | Tim Rogers <[email protected]> | 2025-02-18 10:29:07 -0500 |
|---|---|---|
| committer | GitHub <[email protected]> | 2025-02-18 15:29:07 +0000 |
| commit | 287fe0aa17ec1ac5bdd595ca6e89aa97464bcac7 (patch) | |
| tree | cf52817b95b99f61cf494af5536e32bf6d2889e3 /libcuda/cuda_runtime_api.cc | |
| parent | 7934dfea8a6f44f592866414189970ef9ff994d6 (diff) | |
A bunch of maintenance fixes, the largest of which is getting the PTX simulation to work with CUDA 12. (#95)
* Fixing the formatter to always use a consistent format and running it on the codebase
* Update linux-so-version.txt
* Update Makefile
* A couple of unnecessary files that are lingering around
* Support CUDA 12
* Getting the PTX simulations to work with CUDA 12. The issue is that ptxas added more information (number of barriers and compile time). We have to parse these or lexx/yacc fail.
* Update ptxinfo.l
debug MACRO was ineffective
* Update gpgpusim_check.cmake
Update to make the CUDA version print a warning, not an error and updating the print to be more reflective of what the actual problem is.
Diffstat (limited to 'libcuda/cuda_runtime_api.cc')
| -rw-r--r-- | libcuda/cuda_runtime_api.cc | 9 |
1 files changed, 4 insertions, 5 deletions
diff --git a/libcuda/cuda_runtime_api.cc b/libcuda/cuda_runtime_api.cc index 3574fc1..502b941 100644 --- a/libcuda/cuda_runtime_api.cc +++ b/libcuda/cuda_runtime_api.cc @@ -3046,8 +3046,7 @@ __host__ cudaError_t CUDARTAPI cudaEventSynchronize(cudaEvent_t event) { printf("GPGPU-Sim API: cudaEventSynchronize ** waiting for event\n"); fflush(stdout); CUevent_st *e = (CUevent_st *)event; - while (!e->done()) - ; + while (!e->done()); printf("GPGPU-Sim API: cudaEventSynchronize ** event detected\n"); fflush(stdout); return g_last_cudaError = cudaSuccess; @@ -3171,7 +3170,7 @@ __host__ cudaError_t CUDARTAPI cudaGetExportTable( * * *******************************************************************************/ -//#include "../../cuobjdump_to_ptxplus/cuobjdump_parser.h" +// #include "../../cuobjdump_to_ptxplus/cuobjdump_parser.h" // extracts all ptx files from binary and dumps into // prog_name.unique_no.sm_<>.ptx files @@ -4068,9 +4067,9 @@ __host__ cudaError_t CUDARTAPI cudaDeviceSetLimit(enum cudaLimit limit, return g_last_cudaError = cudaSuccess; } -//#if CUDART_VERSION >= 9000 +// #if CUDART_VERSION >= 9000 //__host__ cudaError_t cudaFuncSetAttribute ( const void* func, enum -// cudaFuncAttribute attr, int value ) { +// cudaFuncAttribute attr, int value ) { // ignore this Attribute for now, and the default is that carveout = // cudaSharedmemCarveoutDefault; // (-1) |