1. CUDA Toolkit Major Components
This section provides an overview of the major components of the CUDA Toolkit and points to their locations after installation.
- Compiler
- The CUDA-C and CUDA-C++ compiler, nvcc, is found in the bin/ directory. It is built on top of the NVVM optimizer, which is itself built on top of the LLVM compiler infrastructure. Developers who want to target NVVM directly can do so using the Compiler SDK, which is available in the nvvm/ directory.
- Tools
- The following development tools are available in the bin/ directory (except
for Nsight Visual Studio Edition (VSE) which is installed as a plug-in to
Microsoft Visual Studio).
- IDEs: nsight (Linux, Mac), Nsight VSE (Windows)
- Debuggers: cuda-memcheck, cuda-gdb (Linux, Mac), Nsight VSE (Windows)
- Profilers: nvprof, nvvp, Nsight VSE (Windows)
- Utilities: cuobjdump, nvdisasm
- Libraries
- The scientific and utility libraries listed below are available in the lib/
directory (DLLs on Windows are in bin/), and their interfaces
are available in the include/ directory.
- cublas (BLAS)
- cublas_device (BLAS Kernel Interface)
- cuda_occupancy (Kernel Occupancy Calculation [header file implementation])
- cudadevrt (CUDA Device Runtime)
- cudart (CUDA Runtime)
- cufft (Fast Fourier Transform [FFT])
- cupti (Profiling Tools Interface)
- curand (Random Number Generation)
- cusolver (Dense and Sparse Direct Linear Solvers and Eigen Solvers)
- cusparse (Sparse Matrix)
- npp (NVIDIA Performance Primitives [image and signal processing])
- nvblas ("Drop-in" BLAS)
- nvcuvid (CUDA Video Decoder [Windows, Linux])
- thrust (Parallel Algorithm Library [header file implementation])
- CUDA Samples
- Code samples that illustrate how to use various CUDA and library APIs are available in the samples/ directory on Linux and Mac, and are installed to C:\ProgramData\NVIDIA Corporation\CUDA Samples on Windows. On Linux and Mac, the samples/ directory is read-only and the samples must be copied to another location if they are to be modified. Further instructions can be found in the Getting Started Guides for Linux and Mac.
- Documentation
- The most current version of these release notes can be found online at http://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html.
- Documentation can be found in PDF form in the doc/pdf/ directory, or in HTML form at doc/html/index.html and online at http://docs.nvidia.com/cuda/index.html.
- Other
- The Open64 source files are controlled under terms of the GPL license. Current and previously released versions are located at ftp://download.nvidia.com/CUDAOpen64/.
- The CUDA-GDB source files are controlled under terms of the GPL license.
- The source code for CUDA-GDB that shipped with CUDA 5.5 and subsequent versions is located at https://github.com/NVIDIA/cuda-gdb.
- The source code for CUDA-GDB that shipped with CUDA 5.0 and previous versions is located at ftp://download.nvidia.com/CUDAOpen64/.
2. New Features
2.1. General CUDA
- Added a method to the CUDA Driver API, cuDevicePrimaryCtxRetain(), that allows a program to create (or to access if it already exists) the same CUDA context for a GPU device as the one used by the CUDART (CUDA Runtime API) library. This context is referred to as the primary context, and this new method allows for sharing the primary context between CUDART and other threads, which can reduce the performance overhead of creating and maintaining multiple contexts per device.
- Unified the device enumeration for CUDA, NVML, and related tools. Variable CUDA_DEVICE_ORDER can have a value of FASTEST_FIRST (default) or PCI_BUS_ID.
- Instrumented NVML (NVIDA Management Library) and the CUDA driver to ignore GPUs that have been made inaccessible via cgroups (control groups). This enables schedulers that rely on cgroups to enforce device access restrictions for their jobs. Job schedulers wanting to use cgroups for device restriction need CUDA and NVML to handle those restrictions in a graceful way.
- Implimented Multi-process Server (MPS), which enables concurrent execution of GPU tasks from multiple CPUs within a single node. It allows setting multiple MPI ranks in a single node.
- The Windows and Mac OS X installers are now also available as network installers. A network installer is much smaller than the traditional local installer and downloads only the components selected for installation.
2.2. CUDA Tools
2.2.1. CUDA Compiler
- Added support for GCC 4.9.
- Added support for the C++11 language dialect.
- On Mac OS X, libc++ is supported with XCode 5.x. Command-line option -Xcompiler -stdlib=libstdc++ is no longer needed when invoking NVCC. Instead, NVCC uses the default library that Clang chooses on Mac OS X. Users are still able to choose between libc++ and libstdc++ by passing -Xcompiler -stdlib=libc++ or -Xcompiler -stdlib=libstdc++ to NVCC.
- The Runtime Compilation library (nvrtc) provides an API to compile CUDA-C++ device source code at runtime. The resulting compiled PTX can be launched on a GPU using the CUDA Driver API. More details can be found in the libNVRTC User Guide.
2.2.2. CUDA-GDB
- Starting with CUDA 7.0, GPU core dumps can read by CUDA-GDB with the target cudacore ${gpucoredump} and target core ${cpucoredump} ${gpucoredump} commands.
- Enabled CUDA applications to generate a GPU core dump when an exception is hit on the GPU. The feature is supported on Windows, Mac OS X, and Linux desktops. (Android, L4T, and Vibrante support may come in the future.) On Windows, this feature is only supported in TCC mode. On Unix-like OSs (Linux, OS X, etc.), a CPU core dump is generated along with a GPU core dump.
2.2.5. Nsight Eclipse Edition
- Cross compiling to the Power8 target architecture using the GNU tool-chain is now supported within the Nsight IDE.
2.2.6. NVIDIA Visual Profiler
- With GPU PC Sampling, which is supported for devices with compute capability 5.2, the Visual Profiler shows stall causes for each source and assembly line. This helps in pinpointing latency bottlenecks in a GPU kernel at the source level.
2.3. CUDA Libraries
2.3.1. cuBLAS Library
- The batched LU solver cublas{T}getrsBatched routine has been added to cuBLAS. It takes the output of the batched factorization routines cublas{T}getrfBatched to compute the solution given the provided batch of right-hand-side matrices.
- A license is no longer required in order to use cuBLAS-XT with more than two GPUs.
2.3.2. cuFFT Library
- For CUDA 7.0, support for callback routines, invoked when cuFFT loads and/or stores data, no longer requires an evaluation license file.
- For CUDA 7.0, cuFFT multiple-GPU execution is supported on up to four GPUs, except for single 1D complex-to-complex transforms, which are supported on two or four GPUs.
- In CUDA 7.0, transform execution may be distributed to four GPUs with the same CUDA architecture. In addition, multiple GPU support for two or four GPUs is no longer constrained to GPUs on a single board. Use of this functionality requires a free evaluation license file, which is available to registered developers via the cuFFT developer page.
- For CUDA 7.0, single complex-to-complex 2D and 3D transforms with dimensions that can be factored into primes less than or equal to 127 are supported on multiple GPUs. Single complex-to-complex 1D transforms on multiple GPUs continue to be limited to sizes that are powers of 2.
2.3.3. cuSOLVER Library
- CUDA 7.0 introduces cuSOLVER, a new library that is a collection of routines to solve linear systems and Eigen problems. It includes dense and sparse linear solvers and sparse refactorization.
- Enabled offloading dense linear algebra calls to the GPUs in a sparse direct solver.
2.3.4. cuSPARSE Library
- Added a new cusparse<t>csrgemm2() routine optimized for small matrices and operations C = a*A*B + b*D, where A, B, and D are CSR matrices.
- Added graph coloring.
2.3.5. CUDA Math Library
- Support for 3D and 4D Euclidean norm and 3D Euclidean reciprocal norm has been added to the math library.
2.3.6. Thrust Library
- Thrust version 1.8.0 introduces support for algorithm invocation from CUDA __device__ code, support for CUDA streams, and algorithm performance improvements. Users may now invoke Thrust algorithms from CUDA __device__ code, providing a parallel algorithms library to CUDA programmers authoring custom kernels as well as allowing Thrust programmers to nest their algorithm calls within functors. The thrust::seq execution policy allows users to require sequential algorithm execution in the calling thread and makes a sequential algorithms library available to individual CUDA threads. The .on(stream) syntax allows users to request a CUDA stream for kernels launched during algorithm execution. Finally, new CUDA algorithm implementations provide substantial performance improvements.
2.4. CUDA Samples
- The CUDA Samples makefile x86_64=1 and ARMv7=1 options have been deprecated. Please use TARGET_ARCH to set the targeted build architecture instead. The CUDA Samples makefile GCC option has been deprecated. Please use HOST_COMPILER to set the host compiler instead.
3. Unsupported Features
- Support for 32-bit x86 Linux Systems
- The CUDA Toolkit and CUDA Driver no longer support developing and running CUDA and OpenCL Applications on 32-bit x86 Linux operating systems.
- Note 1: Developing and running 32-bit applications on 64-bit (x86_64) Linux operating systems is still supported, but that functionality is marked as deprecated and may be dropped in a future release. 64-bit applications are not impacted.
- Note 2: This notice applies to x86 architectures only; 32-bit application support on the ARM architecture remains officially supported.
- Red Hat Enterprise Linux 5 and CentOS 5
- CUDA no longer supports the RHEL 5 and CentOS 5 Linux distributions. Please note that RHEL 6, RHEL 7, CentOS 6, and CentOS 7 are all supported.
- CUDA Toolkit and CUDA Driver Support for Tesla Architecture
- The CUDA Toolkit and CUDA Driver no longer supports the sm_10, sm_11, sm_12, and sm_13 architectures. As a consequence, CU_TARGET_COMPUTE_1x enum values have been removed from the CUDA headers.
- Certain CUDA Features on 32-bit and 32-on-64-bit Windows Systems
- The CUDA Toolkit no longer supports 32-bit Windows operating systems. Furthermore, the Windows CUDA Driver no longer supports Tesla and Quadro products on 32-bit Windows operating systems. Additionally, on 64-bit Windows operating systems, the following features are no longer supported by the CUDA driver or CUDA toolkit:
-
- Running 32-bit applications on Tesla and Quadro products
- Using the Thrust library from 32-bit applications
- 32-bit versions of the CUDA Toolkit scientific libraries, including cuBLAS, cuSPARSE, cuFFT, cuRAND, and NPP
- 32-bit versions of the CUDA samples
- Note the above list doesn't impact any 64-bit components on Windows.
- Using gcc as a Host Compiler on Mac OS X
- On Mac OS X platforms, nvcc no longer supports the gcc toolchain for compiling host code, including the GNU stdlibc++ standard C++ library. Developers should use the clang/llvm toolchain for compiling host code instead; nvcc does this by default in CUDA 7.0 on supported Mac OS X platforms.
4. Deprecated Features
The following features are deprecated in the current release of the CUDA software. The features still work in the current release, but their documentation may have been removed, and they will become officially unsupported in a future release. We recommend that developers employ alternative solutions to these features in their software.
- Developing and Running 32-bit CUDA and OpenCL Applications on x86 Linux Platforms
-
Support for developing and running 32-bit CUDA and OpenCL applications on 64-bit x86 Linux platforms is deprecated.
- CUDA Samples Makefile x86_64=1 and ARMv7=1 Options
-
The CUDA Samples Makefile x86_64=1 and ARMv7=1 options have been deprecated. Please use TARGET_ARCH to set the targeted build architecture instead. The CUDA Samples Makefile GCC option has been deprecated. Please use HOST_COMPILER to set the host compiler instead.
- Header File sobol_direction_vectors.h.
-
The sobol_direction_vectors.h header file is deprecated. This file allowed developers to employ the cuRAND device API with sobol distributions. However, since it is large and takes significant amounts of time and RAM to compile, this file is deprecated in favor of the curandGetDirectionVectors{32,64}() and curandGetScrambleConstants{32,64}() functions. These functions return a memory pointer to the direction vectors that are precompiled into the cuRAND library, and developers should use this pointer to copy the vectors to the GPU device memory.
5. Performance Improvements
5.1. CUDA Libraries
5.1.1. cuFFT Library
- For CUDA 7.0, a new mode has been added for copying single 1D transform input from the host to the GPU. This mode redistributes the inputs as required by the first computation phase of the algorithm and eliminates the overhead of this data redistribution from the execution phase.
- In CUDA 7.0, new composite-sized FFT kernels have been added for many sizes which can be factored into small primes. Composite sizes up to 256 are nearly completely represented, and there is some coverage for sizes up to 1920. This was done to reduce the number of kernel invocations done on the host and will result in significant speed improvements for many composite sizes. Note that these composite-size kernels are not used in combination with the cuFFT callback feature.
6. Resolved Issues
6.1. General CUDA
- On openSUSE and SLES, X no longer fails to load if the CUDA Toolkit RPM packages are installed using relocation immediately following an installation of the cuda-drivers package.
- The Windows toolkit installation no longer fails if Visual Studio, Nvda.Launcher.exe, Nsight.Monitor.exe, or Nvda.CrashReporter.exe is running.
- The cuda and gpu-deployment-kit packages must be installed by separate executions of yum. See the Linux Getting Started Guide for more details.
- The CUDA reference manual now correctly describes the CUDA device pointer CUdeviceptr as an unsigned integer type whose size matches the size of a pointer on the target platform.
6.2. CUDA Tools
6.2.1. CUDA Compiler
- C++11 support added. The new nvcc flag -std=c++11 turns on C++11 features in the CUDA compiler as well as the host compiler and linker. The flag is supported by host compilers gcc >= 4.7 and clang. In addition, any C++11 features that are enabled by default by a supported host compiler are also allowed to be used in device code. Please see the CUDA Programming Guide for further details.
6.2.2. Nsight Eclipse Edition
- Starting with CUDA 7.0, to cross-compile a CUDA project, Nsight uses the default cross-compiler available on the host (or remote) operating system.
6.2.3. NVIDIA Visual Profiler
- In the Visual Profiler timeline, the colors of intervals on the Compute and Stream timelines are no longer incorrect. Also the timeline modes color by stream and color by process do work.
6.3. CUDA Libraries
6.3.1. cuSPARSE Library
-
CUDA 7.0 fixed bugs in the csr2csc() and bsr2bsc() routines that were in the CUDA 6.0 and 6.5 releases. As a consequence, csrsv(), csrsv2(), csrsm(), bsrsv2(), bsrsm2(), and csrgemm() now produce correct results when working with transpose (CUSPARSE_OPERATION_TRANSPOSE) or conjugate-transpose (CUSPARSE_OPERATION_CONJUGATE_TRANSPOSE) operations.
6.4. CUDA Samples
- To properly build the simpleCUFFT_callback sample, the -dc compiler flag no longer must be added to the compilation commands.
7. Known Issues
7.1. General CUDA
- If the Windows toolkit installation fails, it may be because Visual Studio, Nvda.Launcher.exe, Nsight.Monitor.exe, or Nvda.CrashReporter.exe is running. Make sure these programs are closed and try to install again.
- Peer access is disabled between two devices if either of them is in SLI mode.
7.2. CUDA Tools
7.2.1. CUDA Compiler
- On Mac OS X, when Clang is used as the host compiler, 32-bit target compilation is not supported. This is because the Clang compiler doesn't support the -malign-double switch that the NVCC compiler needs to properly align double-precision structure fields when compiling for a 32-bit target (GCC does support this switch). Note that GCC is the default host compiler used by NVCC on Mac OS X 10.8 and Clang is the default on Mac OS X 10.9.
Notices
Notice
ALL NVIDIA DESIGN SPECIFICATIONS, REFERENCE BOARDS, FILES, DRAWINGS, DIAGNOSTICS, LISTS, AND OTHER DOCUMENTS (TOGETHER AND SEPARATELY, "MATERIALS") ARE BEING PROVIDED "AS IS." NVIDIA MAKES NO WARRANTIES, EXPRESSED, IMPLIED, STATUTORY, OR OTHERWISE WITH RESPECT TO THE MATERIALS, AND EXPRESSLY DISCLAIMS ALL IMPLIED WARRANTIES OF NONINFRINGEMENT, MERCHANTABILITY, AND FITNESS FOR A PARTICULAR PURPOSE.
Information furnished is believed to be accurate and reliable. However, NVIDIA Corporation assumes no responsibility for the consequences of use of such information or for any infringement of patents or other rights of third parties that may result from its use. No license is granted by implication of otherwise under any patent rights of NVIDIA Corporation. Specifications mentioned in this publication are subject to change without notice. This publication supersedes and replaces all other information previously supplied. NVIDIA Corporation products are not authorized as critical components in life support devices or systems without express written approval of NVIDIA Corporation.