aboutsummaryrefslogtreecommitdiffhomepage
path: root/unsupported/Eigen
Commit message (Collapse)AuthorAge
* Add a EIGEN_NO_CUDA option, and introduce EIGEN_CUDACC and EIGEN_CUDA_ARCH ↵Gravatar Gael Guennebaud2017-07-17
| | | | aliases
* Code cleanupGravatar Benoit Steiner2017-07-10
|
* Fixed syntax errors generated by xcodeGravatar Benoit Steiner2017-07-09
|
* Avoid relying on cxx11 features when possible.Gravatar Benoit Steiner2017-07-08
|
* Merged in benoitsteiner/opencl (pull request PR-323)Gravatar Benoit Steiner2017-07-07
|\ | | | | | | Improved support for OpenCL
* \ Merged in hughperkins/eigen/add-endif-labels-TensorReductionCuda.h (pull ↵Gravatar Benoit Steiner2017-07-07
|\ \ | | | | | | | | | | | | | | | request PR-315) Add labels to #ifdef, in TensorReductionCuda.h
* | | Merged in tntnatbry/eigen (pull request PR-319)Gravatar Benoit Steiner2017-07-07
| | | | | | | | | | | | Tensor Trace op
* | | Improved the randomness of the tensor random generatorGravatar Benoit Steiner2017-07-06
| | |
* | | Fixed compilation warningGravatar Benoit Steiner2017-07-06
| | |
| | * Merged in mehdi_goli/upstr_benoit/TensorSYCLImageVolumePatchFixed (pull ↵Gravatar Benoit Steiner2017-07-06
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | request PR-14) Applying Benoit's comment for Fixing ImageVolumePatch. * Applying Benoit's comment for Fixing ImageVolumePatch. Fixing conflict on cmake file. * Fixing dealocation of the memory in ImagePatch test for SYCL. * Fixing the automerge issue.
| | * Merged in mehdi_goli/opencl/DataDependancy (pull request PR-10)Gravatar Benoit Steiner2017-06-28
| |/ |/| | | | | | | | | | | | | | | | | DataDependancy * Wrapping data type to the pointer class for sycl in non-terminal nodes; not having that breaks Tensorflow Conv2d code. * Applying Ronnan's Comments. * Applying benoit's comments
* | Merged in benoitsteiner/opencl (pull request PR-318)Gravatar Benoit Steiner2017-06-13
|\ \ | | | | | | | | | Improved support for OpenCL
| | * Add labels to #ifdef, in TensorReductionCuda.hGravatar Hugh Perkins2017-06-06
| | |
| * | Merged in mehdi_goli/opencl/SYCLAlignAllocator (pull request PR-7)Gravatar Benoit Steiner2017-05-26
| |\ \ | | | | | | | | | | | | Fixing SYCL alignment issue required by TensorFlow.
| * \ \ Merged eigen/eigen into defaultGravatar Benoit Steiner2017-05-26
| |\ \ \ | | | |/ | | |/|
| | | * Applying Ronnan's comments.Gravatar Mehdi Goli2017-05-26
| | | |
| | | * Applying Benoit's comment;removing dead code.Gravatar Mehdi Goli2017-05-25
| | | |
| * | | Restore misplaced commentGravatar a-doumoulakis2017-05-24
| | | |
| * | | Merge changed from upstreamGravatar a-doumoulakis2017-05-24
| |\ \ \
* | | | | Specializing numeric_limits For AutoDiffScalarGravatar Mmanu Chaturvedi2017-05-23
| | | | |
* | | | | Fix compilation of matrix log with Map as inputGravatar Gael Guennebaud2017-06-07
| |_|/ / |/| | |
| | | * Fixing SYCL alignment issue required by TensorFlow.Gravatar Mehdi Goli2017-05-22
| | | |
| | * | Fixing suported device list.Gravatar Mehdi Goli2017-05-22
| | |/
| * / Add support for triSYCLGravatar a-doumoulakis2017-05-05
| |/ | | | | | | | | | | Eigen is now able to use triSYCL with EIGEN_SYCL_TRISYCL and TRISYCL_INCLUDE_DIR options Fix contraction kernel with correct nd_item dimension
* / Use scalar_sum_op and scalar_quotient_op instead of operator+ and operator/ ↵Gravatar RJ Ryan2017-04-14
|/ | | | | | | | | | in MeanReducer. Improves support for std::complex types when compiling for CUDA. Expands on e2e9cdd16970914cf0a892fea5e7c4402b3ede41 and 2bda1b0d93fb627d0c500ec48b20302d44c32cb7 .
* Merged in benoitsteiner/opencl (pull request PR-309)Gravatar Benoit Steiner2017-04-05
|\ | | | | | | OpenCL improvements
| * Deleted empty line of codeGravatar Benoit Steiner2017-04-04
| |
| * Guard sycl specific code under a EIGEN_USE_SYCL ifdefGravatar Benoit Steiner2017-04-04
| |
| * Code cleanupGravatar Benoit Steiner2017-04-04
| |
| * Guard the sycl specific code with EIGEN_USE_SYCLGravatar Benoit Steiner2017-04-04
| |
| * Guard the sycl specific code with a #ifdef EIGEN_USE_SYCLGravatar Benoit Steiner2017-04-04
| |
| * iGate the sycl specific code under a EIGEN_USE_SYCL defineGravatar Benoit Steiner2017-04-04
| |
| * Fixed compilation error when sycl is enabled.Gravatar Benoit Steiner2017-04-04
| |
* | fix typos in the Tensor readmeGravatar Gagan Goel2017-03-31
| |
| * Restored code compatibility with compilers that dont support c++11Gravatar Benoit Steiner2017-03-31
| | | | | | | | Gated more sycl code under #ifdef sycl
| * Restore the old constructors to retain compatibility with non c++11 compilers.Gravatar Benoit Steiner2017-03-31
| |
| * Gate the sycl specific code under #ifdef syclGravatar Benoit Steiner2017-03-31
| |
| * Fixing TensorArgMaxSycl.h; Removing warning related to the hardcoded type of ↵Gravatar Mehdi Goli2017-03-28
| | | | | | | | dims to be int in Argmax.
| * Introduces align allocator for SYCL bufferGravatar Luke Iwanski2017-03-20
| |
| * Merged eigen/eigen into defaultGravatar Benoit Steiner2017-03-15
| |\ | |/ |/|
* | Silenced compilation warningGravatar Benoit Steiner2017-03-15
| |
| * Fixes bug in get_sycl_supported_devices() that was reporting unsupported ↵Gravatar Luke Iwanski2017-03-15
| | | | | | | | Intel CPU on AMD platform - causing timeouts in that configuration
* | Made the reduction code compile with cuda-clangGravatar Benoit Steiner2017-03-14
| |
| * Adding synchronisation to convolution kernel for sycl backend.Gravatar Mehdi Goli2017-03-13
| |
* | Get rid of Init().Gravatar Rasmus Munk Larsen2017-03-10
| |
* | Use C++11 ctor forwarding to simplify code a bit.Gravatar Rasmus Munk Larsen2017-03-10
| |
* | Make the non-blocking threadpool more flexible and less wasteful of CPU ↵Gravatar Rasmus Munk Larsen2017-03-09
| | | | | | | | | | | | | | | | | | | | | | | | cycles for high-latency use-cases. * Adds a hint to ThreadPool allowing us to turn off spin waiting. Currently each reader and record yielder op in a graph creates a threadpool with a thread that spins for 1000 iterations through the work stealing loop before yielding. This is wasteful for such ops that process I/O. * This also changes the number of iterations through the steal loop to be inversely proportional to the number of threads. Since the time of each iteration is proportional to the number of threads, this yields roughly a constant spin time. * Implement a separate worker loop for the num_threads == 1 case since there is no point in going through the expensive steal loop. Moreover, since Steal() calls PopBack() on the victim queues it might reverse the order in which ops are executed, compared to the order in which they are scheduled, which is usually counter-productive for the types of I/O workloads the single thread pools tend to be used for. * Store num_threads in a member variable for simplicity and to avoid a data race between the thread creation loop and worker threads calling threads_.size().
| * Use name to distinguish name instead of the vendorGravatar Luke Iwanski2017-03-08
| |
* | bug #1401: fix compilation of "cond ? x : -x" with x an AutoDiffScalarGravatar Gael Guennebaud2017-03-08
| |
| * Adding sycl Benchmarks.Gravatar Mehdi Goli2017-03-08
| |
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-03 12:20:53 +0000