I think the usual suggestion at this point would be to create a short, self-contained, complete test case, that reproduces the issue, and file a bug.
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| nvcc error : 'ptxas' died due to signal 11 (Invalid memory reference) | 8 | 5120 | March 12, 2014 | |
| compile error using CUDA 2.0 'cudafe' died due to signal 11 | 5 | 15332 | September 11, 2008 | |
| gcc passing compiler options to nvcc release 8.0, V8.0.26 - cudafe died signal 11 | 9 | 1802 | September 29, 2016 | |
| compiler crash | 0 | 2089 | July 17, 2010 | |
| nvcc error | 1 | 3622 | January 10, 2013 | |
| CUDA 1.1 Bug - Compiler crash (ptxas) w/repro | 16 | 8649 | May 19, 2008 | |
| Debugging problem nvcc -G gives error messages | 6 | 7532 | September 18, 2008 | |
| How to debug "Invalid memory reference" while generating linker | 17 | 5269 | August 21, 2014 | |
| Assertion failure at line 1440 of ../../be/cg/cgem | 5 | 4036 | August 27, 2007 | |
| cuda-gdb and CMake build instructions for cuda-gdb | 18 | 43939 | December 31, 2009 |