Commit message (Collapse) | Author | Age | ||
---|---|---|---|---|
... | ||||
| * | | Merged eigen/eigen into default | 2016-11-04 | ||
| |\ \ | |/ / |/| | | ||||
| * | | Removed the sycl include from Eigen/Core and moved it to ↵ | 2016-11-04 | ||
| | | | | | | | | | | | | Unsupported/Eigen/CXX11/Tensor; added TensorReduction for sycl (full reduction and partial reduction); added TensorReduction test case for sycl (full reduction and partial reduction); fixed the tile size on TensorSyclRun.h based on the device max work group size; | |||
* | | | Added tag 3.3-rc2 for changeset ba05572dcb385c752fc2c0729f05ccb9ad04d7bd | 2016-11-04 | ||
| | | | ||||
* | | | bump to 3.3-rc2 | 2016-11-04 | ||
| | | | ||||
* | | | Improved AVX512 configuration | 2016-11-03 | ||
| | | | ||||
* | | | Reenable the generation of dynamic blas libraries. | 2016-11-03 | ||
| | | | ||||
* | | | Improved AVX512 support | 2016-11-03 | ||
| | | | ||||
* | | | Merged eigen/eigen into default | 2016-11-03 | ||
|\ \ \ | ||||
| * | | | bug #1337: improve doc of homogeneous() and hnormalized() | 2016-11-03 | ||
| | | | | ||||
| * | | | bug #1330: Cholmod supports double precision only, so let's trigger a static ↵ | 2016-11-03 | ||
| | | | | | | | | | | | | | | | | assertion if the scalar type does not match this requirement. | |||
| * | | | Merged in benoitsteiner/opencl (pull request PR-244) | 2016-11-02 | ||
| |\| | | | | | | | | | | | | | | Disable vectorization on device only when compiling for sycl | |||
| | * | | Disable vectorization on device only when compiling for sycl | 2016-11-02 | ||
| | | | | ||||
| * | | | Don't call lgamma_r when compiling for an Apple device, since the function ↵ | 2016-11-02 | ||
| | | | | | | | | | | | | | | | | isn't available on MacOS | |||
| * | | | Pulled latest updates from trunk | 2016-11-02 | ||
| |\ \ \ | ||||
| | * | | | Special functions require math.h: make sure it is included. | 2016-11-02 | ||
| | | | | | ||||
| * | | | | bug #1004: improve accuracy of LinSpaced for abs(low) >> abs(high). | 2016-11-02 | ||
| | | | | | ||||
| * | | | | Add pinsertfirst function and implement pinsertlast for complex on SSE/AVX. | 2016-11-02 | ||
| | |/ / | |/| | | ||||
| * | | | Merged in benoitsteiner/opencl (pull request PR-243) | 2016-11-02 | ||
| |\ \ \ | | |/ / | |/| | | | | | | Fixed the ambiguity in callig make_tuple for sycl backend. | |||
| * | | | Merged eigen/eigen into default | 2016-11-01 | ||
| |\ \ \ | ||||
| * | | | | Don't attempts to use lgamma_r for CUDA devices. | 2016-11-01 | ||
| | | | | | | | | | | | | | | | | | | | | Fix type in lgamma_impl<double>. | |||
| | * | | | Gate the code that refers to cuda fp16 primitives more thoroughly | 2016-11-01 | ||
| | | | | | ||||
| | | * | | Fixed the ambiguity in callig make_tuple for sycl backend. | 2016-10-31 | ||
| | |/ / | ||||
| | * | | Worked around Visual Studio compilation errors | 2016-10-28 | ||
| | | | | ||||
| | * | | Sharded the tensor_image_patch test to help it run on low power devices | 2016-10-27 | ||
| | | | | ||||
| | * | | Fixed a few more compilation warnings | 2016-10-28 | ||
| | | | | ||||
| | * | | Fixed a compilation warning | 2016-10-28 | ||
| | | | | ||||
| | * | | Fixed compilation warning | 2016-10-28 | ||
| | | | | ||||
| | * | | Added missing template parameters | 2016-10-28 | ||
| | | | | ||||
| * | | | Use threadsafe versions of lgamma and lgammaf if possible. | 2016-10-27 | ||
| |/ / | ||||
| * | | Workaround MSVC issue. | 2016-10-27 | ||
| | | | ||||
| * | | Merged in enricodetoma/eigen (pull request PR-241) | 2016-10-27 | ||
| |\ \ | | | | | | | | | | | | | Always enable /bigobj for tests to avoid a compile error in MSVC 2015 | |||
| * \ \ | Pulled latest updates from trunk | 2016-10-26 | ||
| |\ \ \ | ||||
| * | | | | Removed a template parameter for fixed sized tensors | 2016-10-26 | ||
| | | | | | ||||
| | * | | | Fix regression in X = (X*X.transpose())/s with X rectangular by deferring ↵ | 2016-10-26 | ||
| | | | | | | | | | | | | | | | | | | | | resizing of the destination after the creation of the evaluator of the source expression. | |||
| | | * | | Always enable /bigobj for tests to avoid a compile error in MSVC 2015 | 2016-10-26 | ||
| | | | | | ||||
| | * | | | add a generic EIGEN_HAS_CXX11 | 2016-10-26 | ||
| | |/ / | ||||
| | * / | Fix warning with ICC | 2016-10-26 | ||
| |/ / | ||||
| * | | Replaced tabs with spaces | 2016-10-25 | ||
| | | | ||||
| * | | Code cleanup | 2016-10-25 | ||
| | | | ||||
| * | | Fix ICC warnings | 2016-10-25 | ||
| | | | ||||
| * | | Add missing inline keywords | 2016-10-25 | ||
| | | | ||||
| * | | Fixed a typo | 2016-10-25 | ||
| | | | ||||
| * | | bug #1004: one more rewrite of LinSpaced for floating point numbers to ↵ | 2016-10-25 | ||
| | | | | | | | | | | | | | | | | | | | | | | | | guarantee both interpolation and monotonicity. This version simply does low+i*step plus a branch to return high if i==size-1. Vectorization is accomplished with a branch and the help of pinsertlast. Some quick benchmark revealed that the overhead is really marginal, even when filling small vectors. | |||
| * | | Add a pinsertlast function replacing the last entry of a packet by a scalar. | 2016-10-25 | ||
| | | | | | | | | | | | | (useful to vectorize LinSpaced) | |||
| * | | bug #1333: fix bad usage of const_cast_derived. Better use .data() for that ↵ | 2016-10-24 | ||
| | | | | | | | | | | | | purpose. | |||
| * | | Cleanup ArrayWrapper and MatrixWrapper by removing redundant accessors. | 2016-10-24 | ||
| | | | ||||
| * | | bug #1004: remove the inaccurate "sequential" path for LinSpaced, mark ↵ | 2016-10-24 | ||
| | | | | | | | | | | | | | | | | | | respective function as deprecated, and enforce strict interpolation of the higher range using a correction term. Now, even with floating point precision, both the 'low' and 'high' bounds are exactly reproduced at i=0 and i=size-1 respectively. | |||
| * | | Merged in benoitsteiner/opencl (pull request PR-238) | 2016-10-24 | ||
| |\ \ | | | | | | | | | | | | | Added support for OpenCL to the Tensor Module | |||
| * | | | bug #698: rewrite LinSpaced for integer scalar types to avoid overflow and ↵ | 2016-10-24 | ||
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | guarantee an even spacing when possible. Otherwise, the "high" bound is implicitly lowered to the largest value allowing for an even distribution. This changeset also disable vectorization for this integer path. | |||
| * | | | Add unit test for overflow in LinSpaced | 2016-10-24 | ||
| | | | |