Changelog History
-
v1.2.10616 Changes
September 05, 2019๐ fix #85
๐ extended support for NPP -
v1.2.10607 Changes
August 29, 2019- ๐ NPP partial support
- ๐ minor bug fixes
-
v1.2.10484 Changes
May 16, 2019- CUDA 10.1
- Visual 2019
- more generic functions
- ๐ bug fixes
- ๐ clang 8 (better AVX512 support)
-
v1.2.10341 Changes
February 21, 2019๐ New in this release :
- [experimental] generic functions
- [experimental] cooperative_groups (no multi device sync)
- [CUDA only] pass virtual function as argument (ldvirtftn opcode)
- [CUDA only] complete support of is/as keywords (requires type conversion support in hybridizer options)
- ๐ documentation website
- โ add SingleStaticAssignment attribute - for types which must be assigned immediately after declaration, such as thread_block.
- more options for script version of hybridizer (relocatable device code, additional libraries for jitter, intrinsics, etc)
๐ Bug fixes:
- ๐ fix atomicExpr C# implementation
- ICustomMarshaller didn't marshal memory back to host
- removal of hybdelegate, replaced by hybaction and hybfunc everywhere
- static methods of generic types can be marshalled
-
v1.2.10271 Changes
January 11, 2019๐ Minor release -- mostly bugfixes.
- ๐ better cufft support
- ๐ multiple bugfixes about delegates and marshalling
- ๐ bugfix in the windows service (renew license was hanging forever for some unknown reason)
- ๐ move to c++14 as default standard (required because of "auto return type" feature, which we use for delegates)
๐ There is a bug (#71) in the CUDA 10.0 samples. It's now fixed in the master branch.
-
v1.2.10227 Changes
December 20, 2018- ๐ Func<> improvements : support of ldvirtftn : we can now pass a virtual function as parameter
- AtomicExpr<> : invoke any Func<T, T, T> where T is int/long/float/double
Example : AtomicExpr.apply(ref x, y, (u, v) => Math.Max(u, v)) - Generic Resident array for value types.
- ๐ cusparse and curand for cuda 9.1, 9.2 and 10.0
- value types constructors (they must be decorated with [Kernel])
- ๐ builtin support for double.MinValue/MaxValue/Epsilon and NegativeZero
- initonly and literal static fields are automatically [HybConstant]
- โ warnings!
-
v1.2.10129 Changes
November 21, 2018๐ New features:
- ๐ Support of
Func<>
(up to four generic arguments) - ๐ Support of passing static functions, or
Func<>
as argument. - ๐ Basic support of
as
andis
keywords - ๐ง Linux improvements
- hybridizer-cuda can work in any directory
๐ MSI installer logs errors and exceptions into C:\Program Files\Altimesh\Hybridizer Essentials\
- ๐ Support of
-
v1.2.10108 Changes
November 16, 2018๐ New features:
- ๐ Support of Func<,,> (up to four generic arguments)
- ๐ Support of passing static functions as argument.
- ๐ง Linux improvements
- hybridizer-cuda can work in any directory
๐ MSI installer logs errors and exceptions into C:\Program Files\Altimesh\Hybridizer Essentials\
-
v1.2.10021 Changes
October 30, 2018๐ CUDA 10 basic support :
no cudaGraph_t
๐ no grid-wide synchronization๐ better error reporting
๐ more early exit on unsupported code paths.
๐ many bug fixes about generic code
๐ bug fixes about Func<>
-
v1.1.9833 Changes
September 17, 2018- ๐ AVX/AVX2/AVX512 support!
๐ We introduced support of AVX* instruction sets for Intel/AMD CPUs. You can keep the same C# code and target your CPU instead of your NVIDIA GPU. All the code and binaries are deployed, but you'll need a special license to use them. For now, those license are reserved to special customers. You can read more about that in our blog post. This is a very large feature, so it's still in beta with restricted access.
Commercial licenses
You can purchase commercial licenses from our website๐ป Command line tool
๐ง Our command line tool has been massively improved. It's been renamed hybcc.exe on windows, and just hybridizer on linux. It now supports:
- project files : hybridizer MyProj.csproj -run
- all target flavors : hybridizer Main.cs -flavor [CUDA]/AVX/AVX2/AVX512 -run
- openmp for CPU targets : hybridizer Main.cs -flavor AVX -fopenmp -run
verbose mode
Notebooks
๐ We added new notebooks and completed existing. See them in our github!
- Minor improvements
- fix license query in Visual Studio : #55
- move files around in the installer to get a cleaner installation.
- completed cuda types (cudaError_t)
- a few more bug fixes.