Backport changes to 3.7 - #2949

Merged

9prady9 merged 115 commits into

arrayfire:v3.7from

umar456:backport_372

Jun 27, 2020

Commits on Jun 26, 2020

Remove cuda_thrust_sort_by_key static dependency

9prady9
authored and
umar456
committed
8d528e4
Use static cufft,cublas,cusolver and cusolver on Unix

9prady9
authored and
umar456
committed
022a1c1
Remove unsed header from wrap cpu kernel

9prady9
authored and
umar456
committed
6e2fa13
Refactor cpu confidence cc to use ParamIterator

9prady9
authored and
umar456
committed
a932788
Fix byteToString where the byte value is > a petabyte
umar456
committed
d978cec
Fix warning in boost stacktrace on newer gcc compilers
umar456
committed
fd4bf32
Update forge_visualization.md

pauljurczak
authored and
umar456
committed
627d16b
Use boost env var on linux github ci jobs

9prady9
authored and
umar456
committed
dd60c3d
Windows github action ci job for CPU backend

9prady9
authored and
umar456
committed
9f5c57a
Avoid print_info as ctest post command for non-ninja win generators

9prady9
authored and
umar456
committed
b1fb0ec
Fix the af_get_memory_pressure_threshold by assigning value parameter
umar456
committed
6e1ee89
Fix documentation to mem step size and clean up memory manager test

jacobkahn
authored and
umar456
committed
7fa7607
Fix constant mem declaration in CUDA morph kernel (#2835)

9prady9
authored and
umar456
committed
6da3045
Make cudnn dependency optional (#2836)

syurkevi
and
umar456
committed
48ba063
Fix lu, rank and qr handling of empty arrays and check for nullptr
umar456
committed
65d4f53
Renamed a few variables in the default alloc and unlock funcitons
umar456
committed
2dfaa37
Minor refactor in median. Add one and two element tests
umar456
committed
22b3524
Fixed formatting issue in test/memory.cpp
umar456
committed
bbab0ca
Remove MKL_ThreadingLibrary from required var. Sequential doesn have one
umar456
committed
549ed65
Fix pinned memory manager check, was testing the function pointer
umar456
committed
56a976c
Apply clang-tidy suggestions to all backends (#2839)
umar456
committed
034cd5d
Escape % character in windows install instructions

9prady9
authored and
umar456
committed
c6c8656
Correct lib path suffix for linux install instructions

9prady9
authored and
umar456
committed
285820d
Remove constexpr not supported by VS2015 (#2850)

cschreib-ibex
authored and
umar456
committed
0565724
Fix dereference of memory_info iterator before check
umar456
committed
e4b1b22
Use double to calculate mean in random engine uniform tests if avialable
umar456
committed
d9bf968
Prevent the optimizations in the MeanOp on cpu.
umar456
committed
047798b
Fix the MatrixMultiplyBatch test so that we are testing the result
umar456
committed
6cacbbb
Remove unnecessary tile from var. Use arith output parameter instead
umar456
committed
83695dc
Address all warnings with -Wall flags in GCC 9.3
umar456
committed
dcdf795
Enable the -Wall flags if the compiler supports it
umar456
committed
3e6efa9
Speed up CPU transpose
umar456
committed
81e9390
Optimize join using memcpy
umar456
committed
9784b10
Remove unnecessary instantiations of join in all backends
umar456
committed
7e1262e
Use builtin ocl work group scan when available in reduceByKey

9prady9
authored and
umar456
committed
f441e2b
Use persistent boost env var in windows github ci job

9prady9
authored and
umar456
committed
15dfb6d
Remove faulty cpu::BinOp struct implementation

9prady9
authored and
umar456
committed
e0e4f91
Use backend agnostic flip at af_flip entry

9prady9
authored and
umar456
committed
baf25cf
Return input if pad output dims match input

9prady9
authored and
umar456
committed
5d34baf
Remove kernel size limit for b8 inputs of erode/dilate

9prady9
authored and
umar456
committed
715de7c
Add minval and maxval for half in OpenCL and CPU backends
umar456
committed
bdcc8a2
Pass USE_(DOUBLE,HALF) definitions to OpenCL build in a uniform way
umar456
committed
bd37d26
Adjust JIT heuristics for the Intel GPU on the OpenCL backend
umar456
committed
107c376
Auto cache compiled CUDA kernels on disk to speed up compilation (#2848)

cschreib-ibex
authored and
umar456
committed
12fc7d5
Fix gfor third format type in gfor tutorial

9prady9
authored and
umar456
committed
696a020
Add tests for gfor-loop with non-unit step sequence

9prady9
authored and
umar456
committed
a6bf7fb
Fix unused var warning by moving it to relevant build arm(#if)

9prady9
authored and
umar456
committed
3c5b90e
Add OpenCL show build log info to debugging docs page

9prady9
authored and
umar456
committed
63a9001
Clean packed dims calculation in cpu fftconvolve

9prady9
authored and
umar456
committed
ae778fa
Remove unnecessary template instantiations for fftconvolve

9prady9
authored and
umar456
committed
678fbe9
Create a test library to speed up test compilation
umar456
committed
e0e2476
Remove unnecessary specilization for getArray for half. Other warnings
umar456
committed
a6b6035
Remove AF_TEST_WITH_MTX_FILES check for mmio project build

9prady9
authored and
umar456
committed
48b4d11
Fix mtx tests macro check in sparse_arith test

9prady9
authored and
umar456
committed
5a9780c
Refactor padArray to an apt name, reshape

9prady9
authored and
umar456
committed
dac7ad0
Refactor qr,solve to use padArrayBorders

9prady9
authored and
umar456
committed
d9de264
Handle zero padding case in padArrayBorders

9prady9
authored and
umar456
committed
3cc6ea4
Cleanup opencl backend header inclusions

9prady9
authored and
umar456
committed
ca33ba1
Fix constness of operator* in ParamIterator
umar456
committed
49617de
Fix warning in older versions of boost stacktrace
umar456
committed
c9a7aef
Prefer downloaded boost compute over system version. Set min version
umar456
committed
29ae570
Update CLBlast version to 1.5.1
umar456
committed
352f032
Remove deprecated variable from doxygen mk file
umar456
committed
a53a238
Set current device to array device before releasing array
umar456
committed
168e9a0
Add boost to the doxygen ci jobs to pass minimum requirements
umar456
committed
8631371
Remove CUDA backend BinOP default implementation

9prady9
authored and
umar456
committed
3c5fb16
Use kernel source with instance name for hashing

9prady9
authored and
umar456
committed
f2b8f09
Fix error in GCC 6.1 because of a noexcept move constructor in AfError
umar456
committed
69857e7
Fix overflow warning. Make pinverse and cholesky tests SERIAL
umar456
committed
69604e9
Fix error in GCC 8.3: cl_float* -> float* using static_cast is invalid
umar456
committed
24c33d0
Add an ArrayFire conanfile.py that pulls from the linux binary installer (#2875)

jacobkahn
authored and
umar456
committed
3d3daaa
CMake support to link against static Intel MKL (#2877)

9prady9
authored and
umar456
committed
7c0a07d
Use read/write buffer in reduce by key instead of fill buffer (#2884)

9prady9
authored and
umar456
committed
48a7952
Fix input ndims validation in fast,orb,sift

9prady9
authored and
umar456
committed
668b590
Simplify/Merge CUDA and OpenCL kernel caching API

9prady9
authored and
umar456
committed
f0ab7f3
remove placeholder pooling docs

syurkevi
authored and
umar456
committed
0f24a7b
Refactor kernel wrappers to use new caching API (#2890)

9prady9
authored and
umar456
committed
e8f7511
Refactor common::Node and remove cpu::jit::Node
umar456
committed
5696e3f
More clang-tidy fixes.
umar456
committed
a16c58a
Move Binary and Transform to src/backend/common
umar456
committed
dd4b0af
Merge all jit caches into a single one

9prady9
authored and
umar456
committed
cf8d4a7
Refactor loadKernel fn name as loadKernelFromDisk

9prady9
authored and
umar456
committed
74b0296
Enable ccache based compilation when it is available (#2893)

9prady9
authored and
umar456
committed
3a68dec
Fix hardcoded include paths in arrayfire_test library

9prady9
authored and
umar456
committed
1e4592c
Fix leak of the device, context and queue OpenCLHPP objects
umar456
committed
da3c8e8
Refactor kernel cache to store CUmodule/cl_program

9prady9
authored and
umar456
committed
689015f
Change kernel cache to mutex protected static storage

9prady9
authored and
umar456
committed
f87611f
Reduce unnecessary instantiations across backends

9prady9
authored and
umar456
committed
6ac9511
Enable non-type template parameters at cpu/kernel level

9prady9
authored and
umar456
committed
0da111d
enqueueWriteBuffer asynchronously in vision kernels (#2910)

9prady9
authored and
umar456
committed
f73d497
Add out of memory test using custom memory manager
umar456
committed
42feefb
Return cl_mem instead of cl::Buffer from nativeAlloc
umar456
committed
9f142e1
Fix leak in OpenCL Indexing
umar456
committed
b67b92b
Update OpenCL interop page so they discuss deleting of memory
umar456
committed
a69bb96
Add missing ndims arg check in indexing fns

9prady9
authored and
umar456
committed
d493ad1
Fix libs for the CUDA 9 Toolkit. Remove rdc and dlink flags
umar456
committed
3078334
Fix several error messages when compiling against CUDA 9.0
umar456
committed
4384b21
Fix CUSOLVER_CHECK error message
umar456
committed
5c28176
Fix several warnings with older compilers
umar456
committed
52aa2ef
Fix ccache launch scripts to use sh compatible syntax

9prady9
authored and
umar456
committed
d3818cc
Adds PR template (#2929)

syurkevi
and
umar456
committed
e9f6d8b
adds missing WITH_CUDNN guard for cudnn.hpp

syurkevi
authored and
umar456
committed
d18a9c1
Split pack expansion to work around a possible bug in VS 2015

9prady9
authored and
umar456
committed
2ba510c
Use cxx_relaxed_constexpr check to define AF_CONSTEXPR

9prady9
authored and
umar456
committed
d68c7e7
Add /Zc:__cplusplus to MSVC CUDA builds. constexpr on NVRTC compiles
umar456
committed
4266829
Create issue templates (#2928)

committed
8fe1ad0
Remove obsolete OSX specific patch in CLBlast external project

9prady9
authored and
umar456
committed
298c207
Workaround for bug in Apple's OpenCL, a missing definition

9prady9
authored and
umar456
committed
0b97f43
Increase minimum required CUDA toolkit version to build

9prady9
authored and
umar456
committed
c886688
Fix several errors when compiling on OSX
umar456
committed
a55933d
Add static asserts and move constructors for several classes
umar456
committed
4dc7b7c
Use descriptor based cusparse API for sparse blas fns

9prady9
authored and
umar456
committed
e5d1542
Changes to support build with CUDA 11

9prady9
authored and
umar456
committed
1fd0e0b
Cautionary notes about default random engine handle management

9prady9
authored and
umar456
committed
45410db

Commits on Jun 27, 2020

Update version to 3.7.2 and add release notes
umar456
committed
a6c72ff