- No category
advertisement
NVIDIA CUDA Toolkit v4.0 Release Notes of the newest one. CUDA project files previously specified the include paths to be
$(CUDA_PATH)\include
. To address this, SDK sample projects now specify either
$(CudaToolkitIncludeDir)
or
$(CudaToolkitDir)\include
.
3. Individual SDK solutions from VS2005, VS2008, VS2010 do not build properly.
Each SDK sample solution may depend on
cutil
,
shrUtils
, or
oclUtils
libraries which are also part of the SDK. In order to build with the proper dependencies, developers needed to open the
release_vs200?.sln
solution file for all dependencies to work. The individual SDK sample solutions for CUDA, CUDALibraries, and OpenCL now include dependencies from individual solution files.
‣
In some cases, Visual Profiler global memory derived statistics and hints may be incorrect. If the kernel has local memory accesses, the derived statistics-
global memory excess load %
and
global memory excess store %
can yield incorrect results. This is because the L2 throughput that is used to calculate these values include local memory accesses too. As a result, the hints which use these statistics are incorrect as well since the excess loads given by this formula are caused due to the local memory accesses (in addition to possibly uncoalesced memory access pattern)
‣
In a multi-gpu setup, when compute mode is set to
compute prohibited
for some GPUs, the Visual Profiler cannot profile a CUDA runtime application; Visual
Profiler reports an error and profiling data is not shown.
‣ CudaHostRegister()
is not supported in RHEL4. Please refer to the NVIDIA
CUDA C Programming Guide for details on
CudaHostRegister()
6.3.4. More Information
For more information and help with CUDA, please visit http://www.nvidia.com/cuda .
6.4. List of Important Files
bin/nvcc Command line compiler include/
cuda.h CUDA driver API header
cudaGL.h CUDA OpenGL interop header for driver API
cudaVDPAU.h CUDA VDPAU interop header for driver API
(Linux only)
cuda_gl_interop.h CUDA OpenGL interop header for toolkit API
(Linux only)
cuda_vdpau_interop.h CUDA VDPAU interop header for toolkit API
(Linux only)
cudaD3D9.h CUDA DirectX 9 interop header (Windows only)
cudaD3D10.h CUDA DirectX 10 interop header (Windows only)
cudaD3D11.h CUDA Directx 11 interop header (Windows only)
cufft.h CUFFT API header
cublas.h CUBLAS API header
cusparse.h CUSPARSE API header
curand.h CURAND API header
curand_kernel.h CURAND device API header
thrust/* Thrust Headers
npp.h NPP API Header
nvcuvid.h CUDA Video Decoder header (Windows and Linux)
cuviddec.h CUDA Video Decoder header (Windows and Linux)
www.nvidia.com
NVIDIA CUDA Toolkit v5.5 for POWER8 RN-06722-001 _v5.5 for POWER8 | 71
advertisement
Related manuals
advertisement
Table of contents
- 9 Chapter 1. NVIDIA CUDA Toolkit v5.5 for POWER8 Release Notes
- 9 1.1. Release Overview
- 9 1.2. Errata
- 9 1.2.1. CUDA Tools
- 10 1.2.2. CUDA Samples
- 10 1.3. Supported NVIDIA Hardware
- 10 1.4. Supported Operating Systems
- 10 1.4.1. Linux
- 10 1.5. New Features
- 10 1.5.1. CUDA Tools
- 10 1.5.1.1. CUDA Compiler
- 11 Chapter 2. NVIDIA CUDA Toolkit v5.5 Release Notes
- 11 2.1. Errata
- 11 2.1.1. General CUDA
- 12 2.1.2. CUDA Libraries
- 12 2.1.2.1. CUBLAS
- 12 2.1.2.2. CUFFT
- 16 2.1.3. CUDA Samples
- 16 2.1.4. CUDA Tools
- 18 2.2. Documentation
- 18 2.3. List of Important Files
- 18 2.3.1. Core Files
- 19 2.3.2. Windows lib Files
- 19 2.3.3. Linux lib Files
- 20 2.3.4. Mac OS X lib Files
- 20 2.4. Supported NVIDIA Hardware
- 20 2.5. Supported Operating Systems
- 20 2.5.1. Windows
- 21 2.5.2. Linux
- 21 2.5.3. Mac OS X
- 21 2.6. Installation Notes
- 21 2.6.1. Windows
- 22 2.6.2. Linux
- 22 2.7. Deprecated Features
- 23 2.8. New Features
- 23 2.8.1. General CUDA
- 24 2.8.2. CUDA Libraries
- 24 2.8.2.1. CUBLAS
- 24 2.8.2.2. CUFFT
- 24 2.8.2.3. CURAND
- 24 2.8.2.4. CUSPARSE
- 25 2.8.2.5. Thrust
- 25 2.8.3. CUDA Tools
- 25 2.8.3.1. CUDA Compiler
- 25 2.8.3.2. CUDA-GDB
- 26 2.8.3.3. CUDA-MEMCHECK
- 26 2.8.3.4. CUDA Profiler
- 27 2.8.3.5. Debugger API
- 27 2.8.3.6. Nsight Eclipse Edition
- 28 2.8.3.7. NVIDIA Visual Profiler
- 28 2.9. Performance Improvements
- 28 2.9.1. CUDA Libraries
- 28 2.9.1.1. CUBLAS
- 28 2.9.1.2. Math
- 28 2.10. Resolved Issues
- 29 2.11. Known Issues
- 29 2.12. Source Code for Open64 and CUDA-GDB
- 29 2.13. More Information
- 30 Chapter 3. NVIDIA CUDA Toolkit v5.0 Release Notes
- 30 3.1. Errata
- 30 3.1.1. Known Issues
- 30 3.1.1.1. General CUDA
- 31 3.1.1.2. CUDA Libraries
- 31 3.1.1.3. CUDA Tools
- 32 3.2. Documentation
- 32 3.3. List of Important Files
- 32 3.3.1. Core Files
- 33 3.3.2. Windows lib Files
- 33 3.3.3. Linux lib Files
- 33 3.3.4. Mac OS X lib Files
- 34 3.4. Supported NVIDIA Hardware
- 34 3.5. Supported Operating Systems
- 34 3.5.1. Windows
- 34 3.5.2. Linux
- 35 3.5.3. Mac OS X
- 35 3.6. Installation Notes
- 35 3.6.1. Windows
- 35 3.6.2. Linux
- 36 3.7. New Features
- 36 3.7.1. General CUDA
- 37 3.7.1.1. Linux
- 38 3.7.2. CUDA Libraries
- 38 3.7.2.1. CUBLAS
- 38 3.7.2.2. CURAND
- 38 3.7.2.3. CUSPARSE
- 39 3.7.2.4. Math
- 40 3.7.2.5. NPP
- 40 3.7.3. CUDA Tools
- 40 3.7.3.1. CUDA Compiler
- 41 3.7.3.2. CUDA-GDB
- 41 3.7.3.3. CUDA-MEMCHECK
- 41 3.7.3.4. NVIDIA Nsight Eclipse Edition
- 41 3.7.3.5. NVIDIA Visual Profiler, Command Line Profiler
- 42 3.8. Performance Improvements
- 42 3.8.1. CUDA Libraries
- 42 3.8.1.1. CUBLAS
- 42 3.8.1.2. CURAND
- 42 3.8.1.3. Math
- 42 3.9. Resolved Issues
- 43 3.9.1. General CUDA
- 43 3.9.2. CUDA Libraries
- 43 3.9.2.1. CURAND
- 43 3.9.2.2. CUSPARSE
- 44 3.9.2.3. NPP
- 44 3.9.2.4. Thrust
- 44 3.9.3. CUDA Tools
- 44 3.9.3.1. CUDA Compiler
- 44 3.9.3.2. CUDA Occupancy Calculator
- 45 3.10. Known Issues
- 45 3.10.1. General CUDA
- 45 3.10.1.1. Linux, Mac OS
- 46 3.10.1.2. Windows
- 46 3.10.2. CUDA Libraries
- 46 3.10.2.1. NPP
- 46 3.10.3. CUDA Tools
- 46 3.10.3.1. CUDA Compiler
- 47 3.10.3.2. NVIDIA Visual Profiler, Command Line Profiler
- 48 3.11. Source Code for Open64 and CUDA-GDB
- 48 3.12. More Information
- 49 Chapter 4. NVIDIA CUDA Toolkit v4.2 Release Notes
- 49 4.1. Errata
- 49 4.1.1. Known Issues
- 49 4.2. Release Highlights
- 50 4.3. Documentation
- 50 4.4. List of Important Files
- 51 4.4.1. Windows lib Files
- 51 4.4.2. Linux lib Files
- 51 4.4.3. Mac OS X lib Files
- 51 4.5. Supported NVIDIA Hardware
- 51 4.6. Supported Operating Systems
- 51 4.6.1. Windows
- 52 4.6.2. Linux
- 53 4.6.3. Mac OS X
- 53 4.7. Installation Notes
- 53 4.7.1. Windows
- 53 4.7.2. Linux
- 54 4.8. New Features
- 54 4.9. Resolved Issues
- 55 4.10. Known Issues
- 55 4.10.1. Windows
- 55 4.10.2. Linux & Mac
- 56 4.10.3. Mac
- 56 4.10.4. Visual Profiler and Command Line Profiler
- 57 4.11. Source Code for Open64 and CUDA-GDB
- 57 4.12. More Information
- 58 Chapter 5. NVIDIA CUDA Toolkit v4.1 Release Notes
- 58 5.1. Release Highlights
- 59 5.2. Documentation
- 59 5.3. List of Important Files
- 60 5.3.1. Windows lib Files
- 60 5.3.2. Linux lib Files
- 60 5.3.3. Mac OS X lib Files
- 60 5.4. Supported NVIDIA Hardware
- 61 5.5. Supported Operating Systems
- 61 5.5.1. Windows
- 61 5.5.2. Linux
- 62 5.5.3. Mac OS X
- 62 5.6. Installation Notes
- 62 5.6.1. Windows
- 62 5.6.2. Linux
- 63 5.7. Upgrading from Previous CUDA Toolkit
- 63 5.7.1. Vista, Server 2008 and Windows 7 Related
- 64 5.7.2. Linux and Mac
- 64 5.7.3. Mac Related
- 64 5.8. CUDA Toolkit Known Issues
- 64 5.8.1. SDK Related
- 65 5.8.2. Visual Profiler and Command Line Profiler
- 67 5.8.3. CUDA-MEMCHECK
- 67 5.9. New Features in CUDA Release
- 67 5.9.1. CUDA Runtime
- 67 5.9.2. Compiler Related
- 68 5.9.3. CUDA Libraries
- 70 5.9.4. CUDA Driver
- 71 5.10. Performance Improvements in CUDA Release
- 72 5.11. Resolved Issues
- 74 5.12. Source Code for Open64 and CUDA-GDB
- 74 5.13. More Information
- 75 5.14. Acknowledgements
- 76 Chapter 6. NVIDIA CUDA Toolkit v4.0 Release Notes
- 76 6.1. Release Highlights
- 77 6.2. Documentation
- 77 6.3. Errata for Windows, Linux, and Mac OS X
- 77 6.3.1. Linux
- 77 6.3.2. Resolved Issues
- 77 6.3.3. Known Issues
- 79 6.3.4. More Information
- 79 6.4. List of Important Files
- 80 6.4.1. Windows lib Files
- 80 6.4.2. Linux lib Files
- 80 6.4.3. Mac OS X lib Files
- 80 6.5. Supported NVIDIA Hardware
- 81 6.6. Supported Operating Systems for Windows, Linux, and Mac OS X
- 81 6.6.1. Windows
- 81 6.6.2. Linux
- 82 6.6.3. Mac OS X
- 82 6.7. Installation Notes
- 82 6.7.1. Windows
- 82 6.7.2. Linux
- 83 6.8. Upgrading from Previous CUDA Toolkit
- 83 6.9. Notes on New Features and Performance Improvements
- 83 6.9.1. CUDA Driver Features
- 87 6.9.2. CUDA Compiler Features
- 88 6.9.3. CUDA Libraries Features
- 91 6.9.4. CUDA Libraries Performance
- 92 6.10. Known Issues
- 94 6.10.1. Vista, Server 2008 and Windows 7 Related
- 94 6.10.2. XP, Vista, Server 2008 and Windows 7 Related
- 95 6.10.3. XP Related
- 95 6.10.4. Linux Only
- 96 6.10.5. Linux and Mac
- 96 6.10.6. Mac Only
- 97 6.11. Resolved Issues