ArrayFire: a general purpose GPU library.
The source code with sub-modules can be downloaded directly from the following link:
http://arrayfire.com/arrayfire_source/arrayfire-full-3.6.4.tar.bz2
The source code with sub-modules can be downloaded directly from the following link:
http://arrayfire.com/arrayfire_source/arrayfire-full-3.6.3.tar.bz2
cblas_Xgemm_batched
#2206AF_JIT_KERNEL_TRACE
environment variable #2404void*
pointers are now allowed as arguments to af::array::write()
#2367af::array::array_proxy
move assignment operator #2479af::write
tests #2380void*
memory allocations to use unsigned char type #2459af_device_array
API #2396Special thanks to our contributors: Alessandro Bessi zhihaoy Jacob Khan William Tambellini
The source code with sub-modules can be downloaded directly from the following link:
http://arrayfire.com/arrayfire_source/arrayfire-full-3.6.2.tar.bz2
cond
argument in select() [#2243]dim4::ndims
. [#2289]af::array
destructor [#2319]af_draw_hist
[#2230]af::array
for indexing [#2311]unwrap
documentation [#2301]wrap
documentation [#2320]accum
documentation [#2298]tile
documentation [#2293]approx*
indexing in documentation [#2287]lookup
examples [#2288]af*
library and dependencies directory changed to lib64
[#2186]Thank you to our contributors: Jacob Kahn Vardan Akopian
The source code for this release can be downloaded here: http://arrayfire.com/arrayfire_source/arrayfire-full-3.6.1.tar.bz2
AF_TRACE
environment variable [#2169][#2216]af::seq
for indexing [#2153]The source code with submodules can be downloaded directly from the following link: http://arrayfire.com/arrayfire_source/arrayfire-full-3.6.0.tar.bz2
topk()
function. 1
anisotropicDiffusion()
.Documentation 3.anisotropicDiffusion()
. Documentation 3.topk()
function, which returns the top k elements along a given dimension of the input. Documentation. 4
select()
and shift()
functions for CUDA and OpenCL backends. 1
af_colormap
struct to match forge's definition. 6
af::erfc
function for cumulative normal distribution.9.shfl_down()
warnings with CUDA 9. 1
lib
directory has arch(64) suffix.3
MemoryManager::nativeFree
6
vectorization.md
. 1
AF_API_VERSION
value in Doxygen config file. 2
-T
flag like so cmake -G "Visual Studio 15 2017 Win64" -T v140
.
canny_opencl, fft_opencl, gen_assign_opencl, homography_opencl, reduce_opencl, scan_by_key_opencl, solve_dense_opencl, sparse_arith_opencl, sparse_convert_opencl, where_opencl
Special thanks to our contributors: Adrien F. Vincent, Cedric Nugteren, Felix, Filip Matzner, HoneyPatouceul, Patrick Lavin, Ralf Stubner, William Tambellini
The source code with submodules can be downloaded directly from the following link: http://arrayfire.com/arrayfire_source/arrayfire-full-3.5.1.tar.bz2
Installer CUDA Version: 8.0 (Required) Installer OpenCL Version: 1.2 (Minimum)
af::unwrap()
function's arguments. 1
af::histogram()
on CUDA and
OpenCL kernels. 1
af::matmul()
which occured when its RHS argument was an
indexed vector. 1
af::replace
so that it is now copy-on-write. 1
clEnqueueReadBuffer
bug on OSX.
1
The source code with submodules can be downloaded directly from the following link: http://arrayfire.com/arrayfire_source/arrayfire-full-3.5.0.tar.bz2
Installer CUDA Version: 8.0 (Required) Installer OpenCL Version: 1.2 (Minimum)
AF_CANNY_THRESHOLD_AUTO_OTSU
af_get_allocated_bytes()
. 1
c32
,c64
) multiplication in OpenCL convolution kernels. 1
cuda::findPlan()
. 1
gcc
> 5.x for CUDA. 1
README.md
to improve readability and formatting. 1
README.md
to mention Julia and Nim wrappers. 1
docs/pages/install.md
. 1
37x.xx
contains a bug which causes fftconvolve_opencl
to fail. Upgrade or downgrade to a different version of the driver to avoid this failure.threading_cuda
,qr_dense_opencl
, solve_dense_opencl
.lu_dense_{cpu,opencl}
, solve_dense_{cpu,opencl}
, inverse_dense_{cpu,opencl}
.fft_large_cuda
and svd_dense_cuda
.cholesky_dense_opencl
and scan_by_key_opencl
.The source code with submodules can be downloaded directly from the following link: http://arrayfire.com/arrayfire_source/arrayfire-full-3.4.2.tar.bz2
Installer CUDA Version: 8.0 (Required) Installer OpenCL Version: 1.2 (Minimum)
This release supports CUDA 6.5 and higher. The next ArrayFire release will support CUDA 7.0 and higher, dropping support for CUDA 6.5. Reasons for no longer supporting CUDA 6.5 include:
As a result, the older Jetson TK1 / Tegra K1 will no longer be supported in the next ArrayFire release. The newer Jetson TX1 / Tegra X1 will continue to have full capability with ArrayFire.
boost::scoped_array
instead of boost::scoped_ptr
when managing
array resources.
1
const
references
to avoid unnecessary copying of ArrayInfo
objects.
1
cholesky.cpp
example.
1
The source code with submodules can be downloaded directly from the following link: http://arrayfire.com/arrayfire_source/arrayfire-full-3.4.1.tar.bz2
Installer CUDA Version: 8.0 (Required) Installer OpenCL Version: 1.2 (Minimum)
aarch64
detection on Jetson TX1 64-bit OS.
1
af_set_fft_plan_cache_size
in unified backend.
1
-arch=sm_30
for PTX compilation as default.
The source code with submodules can be downloaded directly from the following link: http://arrayfire.com/arrayfire_source/arrayfire-full-3.4.0.tar.bz2
Installer CUDA Version: 7.5 (Required) Installer OpenCL Version: 1.2 (Minimum)
bin2cpp
.
1
AF_OPENCL_CPU_OFFLOAD=0
.The following functions have been deprecated and may be modified or removed permanently from future versions of ArrayFire.
Certain CUDA functions are known to be broken on Tegra K1. The following ArrayFire tests are currently failing: