Stack Exchange

harrism

Brisbane, Australia

markmark.net

Employee of NVIDIA Corporation. Opinions expressed on StackOverflow.com are my own.

Graphics / HPC / GPGPU / Game Physics

C++, CUDA, Python, Ruby

top accounts reputation activity favorites subscriptions
Top Questions
No questions with score of 5 or more
Top Answers
16
CUDA: How many concurrent threads in total?
stackoverflow.com

16
Error in cudaMemcpyToSymbol using CUDA 5
stackoverflow.com

13
CUDA and Classes
stackoverflow.com

10
CUDA, using memset(or fill or …) to set an array of float to max val possible
stackoverflow.com

10
Doing multiple matrix-matrix multiplications in one operation
stackoverflow.com

10
how can a __global__ function RETURN a value or BREAK out like C/C++ does
stackoverflow.com

8
How to get the cuda version?
stackoverflow.com

8
CUDA coalesced access to global memory
stackoverflow.com

8
Can I get CUDA Compute capability (version) in compile time by #define?
stackoverflow.com

7
What is the context switching mechanism in GPU?
stackoverflow.com

7
How does the CUDA warp scheduler issue 2 instructions at a time for a warp?
stackoverflow.com

7
Why has atomicAdd not been implemented for doubles?
stackoverflow.com

7
CUDA - why is warp based parallel reduction slower?
stackoverflow.com

7
PTX “bit bucket” registers
stackoverflow.com

6
CUDA, Sort 3 arrays that are already in GPU mem. using Thrust perhaps?
stackoverflow.com

6
Building GPL C program with CUDA module
stackoverflow.com

6
CUDA host and device using same __constant__ memory
stackoverflow.com

6
cuda memory alignment
stackoverflow.com

5
Cuda and OpenGL Interop
stackoverflow.com

5
How do I diagnose a CUDA launch failure due to being out of resources?
stackoverflow.com

5
The peak throughput of cuda Kernel on NVIDA GPU
stackoverflow.com

5
Minimum number of GPU threads to be effective
stackoverflow.com

5
Copying a struct containing pointers to CUDA device
stackoverflow.com

5
L1 cache persistance across CUDA kernels
stackoverflow.com

5
Expression must have pointer-to-object type
stackoverflow.com

5
Aren't NPP functions completely optimized?
stackoverflow.com

5
Why can't member variables be shared?
stackoverflow.com

5
Which cards and compute capabilities are required to fully utilize CUDA 5's features
stackoverflow.com

5
atomicCAS: behaviour within a block
stackoverflow.com