Bump version again.

By having skipped CI on the previous tagged commit, we didn't get documentation.
[skip tests]

## CUDA v2.2.0

[Diff since v2.1.0](JuliaGPU/CUDA.jl@v2.1.0...v2.2.0)


**Closed issues:**
- cudnn missing after downloading artifact (JuliaGPU#521)
- Downloading artifact: CUDA110 when using DiffEqFlux (JuliaGPU#542)

**Merged pull requests:**
- Update manifest (JuliaGPU#520) (@github-actions[bot])
- Try out Buildkite. (JuliaGPU#522) (@maleadt)
- Update manifest (JuliaGPU#529) (@github-actions[bot])
- Support for / Upgrade to CUDA 11.1 update 1. (JuliaGPU#530) (@maleadt)
- Fix and test svd! (JuliaGPU#531) (@maleadt)
- Move more CI to Buildkite. (JuliaGPU#532) (@maleadt)
- Use type symbols to generate wrapper methods (JuliaGPU#534) (@cqql)
- Fully move to Buildkite. (JuliaGPU#537) (@maleadt)
- Add unit_diag option for sv2! functions (JuliaGPU#540) (@amontoison)
- Documentation fixes (JuliaGPU#543) (@maleadt)

## CUDA v2.1.0

[Diff since v2.0.2](JuliaGPU/CUDA.jl@v2.0.2...v2.1.0)


**Closed issues:**
- CUDNN convolution with Float16 always returns zeros  (JuliaGPU#92)
- axp(b)y! and mul! (scalar multiplication) with mixed argument types (JuliaGPU#144)
- Dispatching to generic matmul instead of CUBLAS (JuliaGPU#164)
- Support for Ints and Float16? (JuliaGPU#165)
- Subarrays/views support (JuliaGPU#172)
- Easy way to pick among multiple GPUs  (JuliaGPU#174)
- More prominently document JULIA_CUDA_USE_BINARYBUILDER (JuliaGPU#204)
- ERROR_COOPERATIVE_LAUNCH_TOO_LARGE during tests (JuliaGPU#247)
- Pkg.test error for cutensor test on Windows (JuliaGPU#422)
- Runtime build improvements (JuliaGPU#456)
- Fusing Wrappers (JuliaGPU#467)
- Could not find nvToolsExt (libnvToolsExt.dylib.1.0 or libnvToolsExt.dylib.1) in /Users/imac/.julia/artifacts/b502baf54095dff4a69fd6aba8667124583f6929/lib (JuliaGPU#482)
- mapreduce assumes commutative op (JuliaGPU#484)
- SubArray Broadcast Bug in 2.0 (JuliaGPU#488)
- Nested SubArray Scalar Indexing (JuliaGPU#490)
- Sparse matrix * view(vector) regression in 2.0 (JuliaGPU#493)
- Error transforming a reshaped 0-dimentional GPU array to a CPU array (JuliaGPU#494)
- test cuda FAILURE (JuliaGPU#496)
- Reshaped CuArray is not DenseCuArray (JuliaGPU#511)
- assignment failure when using array slicing. (JuliaGPU#516)

**Merged pull requests:**
- Use the correct CUDNN scaling parameter type. (JuliaGPU#454) (@maleadt)
- Fix versioned dylib discovery. (JuliaGPU#486) (@maleadt)
- Move inv from GPUArrays. (JuliaGPU#487) (@maleadt)
- Use dense array types in sparse wrappers. (JuliaGPU#495) (@maleadt)
- Update manifest (JuliaGPU#497) (@github-actions[bot])
- Revert array wrapper union changes (JuliaGPU#498) (@maleadt)
- Clean-up pointer field. (JuliaGPU#499) (@maleadt)
- mapreduce: change iteration for compatibility with non-commutative operators. (JuliaGPU#500) (@maleadt)
- Use versioned libcuda (JuliaGPU#502) (@maleadt)
- Dynamically choose versioned libcuda (JuliaGPU#503) (@mustafaquraish)
- Update multigpu.md (JuliaGPU#504) (@efmanu)
- Upgrade artifacts for CUDA 11 compatibility. (JuliaGPU#506) (@maleadt)
- Update dependencies. (JuliaGPU#507) (@maleadt)
- Convert unsigned short ints to Cint for printf. (JuliaGPU#508) (@maleadt)
- Update manifest (JuliaGPU#510) (@github-actions[bot])
- Fix reshape with missing dimensions. (JuliaGPU#512) (@maleadt)
- Don't return a pointer from 'alias'. (JuliaGPU#513) (@maleadt)
- Add some docs (JuliaGPU#514) (@maleadt)
- Fix CUDNN-optimized activation broadcasts (JuliaGPU#515) (@maleadt)
- Fix cooperative launch test. (JuliaGPU#517) (@maleadt)
- Fixes for Windows (JuliaGPU#518) (@maleadt)
- CUTENSOR fixes on Windows (JuliaGPU#519) (@maleadt)

## CUDA v2.0.2

[Diff since v2.0.1](JuliaGPU/CUDA.jl@v2.0.1...v2.0.2)


**Closed issues:**
- cu() behavior for complex floating point numbers (JuliaGPU#91)
- Error when following example on using multiple GPUs on multiple processes (JuliaGPU#468)
- MacOS without nvidia GPU is trying to download CUDA111 on julia nightly (JuliaGPU#469)
- Drop BinaryProvider? (JuliaGPU#474)
- Latest version of master doesn't work on Windows (JuliaGPU#477)
- `sum(CUDA.rand(3,3))` broken (JuliaGPU#480)
- copyto!() between cpu and gpu with subarrays (JuliaGPU#491)

**Merged pull requests:**
- Adapt to GPUCompiler changes. (JuliaGPU#458) (@maleadt)
- Fix initialization of global state (JuliaGPU#471) (@maleadt)
- Remove 'view' implementation. (JuliaGPU#472) (@maleadt)
- Workaround new artifact"" eagerness that prevents loading on unsupported platforms (JuliaGPU#473) (@ianshmean)
- Remove BinaryProvider dep. (JuliaGPU#475) (@maleadt)
- typo:  libcuda.dll -> libcuda.so on Linux (JuliaGPU#476) (@Alexander-Barth)
- NFC array simplifications. (JuliaGPU#481) (@maleadt)
- Update manifest (JuliaGPU#485) (@github-actions[bot])
- Convert AbstractArray{ComplexF64} to CuArray{ComplexF32} by default (JuliaGPU#489) (@pabloferz)

## CUDA v2.0.1

[Diff since v2.0.0](JuliaGPU/CUDA.jl@v2.0.0...v2.0.1)


**Closed issues:**
- Can't update (JuliaGPU#462)

**Merged pull requests:**
- Remove duplicate comment (JuliaGPU#464) (@blegat)
- Add functionality to precompile the runtime library. (JuliaGPU#465) (@maleadt)
- Update manifest (JuliaGPU#470) (@github-actions[bot])

## CUDA v2.0.0

[Diff since v1.3.3](JuliaGPU/CUDA.jl@v1.3.3...v2.0.0)


**Closed issues:**
- Test failure during threading tests (JuliaGPU#15)
- Bad allocations in memory pool after device_reset! (JuliaGPU#16)
- CuArrays can lose Blas on reshaped views (JuliaGPU#78)
- allowscalar performance (JuliaGPU#87)
- Indexing with a CuArrays causes a 'scalar indexing disallowed' error from checkbounds (JuliaGPU#90)
- 5-arg mul! for CUSPARSE (JuliaGPU#98)
- copyto!(Device, Host) uses scalar iteration in case of type mismatch (JuliaGPU#105)
- Array primitives broken for CUSPARSE arrays (JuliaGPU#113)
- SplittingPool: CPU allocations (JuliaGPU#117)
- error while concatenating to an empty CuArray (JuliaGPU#139)
- Showing sparse arrays goes wrong (JuliaGPU#146)
- Improve test coverage (JuliaGPU#147)
- CuArrays allocates a lot of memory on the default GPU (JuliaGPU#153)
- [Feature Request] Indexing CuArray with CuArray (JuliaGPU#155)
- Reshaping CuArray throws error during backpropagation (JuliaGPU#162)
- Match syntax and APIs against Julia 1.0 standard libraries (JuliaGPU#163)
- CURAND_STATUS_PREEXISTING_FAILURE when setting seed multiple times. (JuliaGPU#212)
- RFC: converts `SparseMatrixCSC` to `CuSparseMatrixCSR` via `cu` by default (JuliaGPU#216)
- Add a CuSparseMatrixCOO type (JuliaGPU#220)
- Test runner stumbles over path separators (JuliaGPU#236)
- Error: Invalid bitcode signature when loading CUDA.jl after precompilation  (JuliaGPU#293)
- Atomic operations only work on global memory (JuliaGPU#311)
- Performance: cudnn algorithm selection (JuliaGPU#318)
- CUSPARSE is broken in CUDA.jl 1.2 (JuliaGPU#322)
- Device-side broadcast regression on 1.5 (JuliaGPU#350)
- API for fast math-like mode (JuliaGPU#354)
- CUDA 11.0 Update 1: cublasSetWorkspace (JuliaGPU#365)
- Can't precompile CUDA.jl on Kubuntu 20.04 (JuliaGPU#396)
- CuPtr should be Ptr in cudnnGetDropoutDescriptor (JuliaGPU#397)
- CUDA throws OOM error when initializing API on multiple devices (JuliaGPU#398)
- Cannot launch kernel with > 5 args using Dynamic Parallelism (JuliaGPU#401)
- Reverse performance regression (JuliaGPU#410)
- Tag for LLVM 3? (JuliaGPU#412)
- CUDA not working  (JuliaGPU#415)
- `StatsBase.transform` fails on `CuArray` (JuliaGPU#426)
- Further unification of `CUBLAS.axpy!` and `LinearAlgebra.BLAS.axpy!` (JuliaGPU#432)
- size(range), length(range) and range[end] fail inside CUDA kernels (JuliaGPU#434)
- InitError: Cannot use memory pool 'binned' when CUDA.jl was precompiled for memory pool 'split'. (JuliaGPU#446)
- Missing dispatch for matrix multiplication with views? (JuliaGPU#448)
- New version not available yet? (JuliaGPU#452)
- using CUDA or CUArray, output: UndefVarError: AddrSpacePtr not defined (JuliaGPU#457)
- Unable to upgrade to the latest version (JuliaGPU#459)

**Merged pull requests:**
- Performance improvements by calling cuDNN API (JuliaGPU#321) (@gartangh)
- Use ccall wrapper for correct pointer type conversions (JuliaGPU#392) (@maleadt)
- Simplify Statistics.var and fix dims=tuple. (JuliaGPU#393) (@maleadt)
- Adapt to GPUArrays test change. (JuliaGPU#394) (@maleadt)
- Default to per-thread stream semantics (JuliaGPU#395) (@maleadt)
- Add a missing context argument for stateless codegen. (JuliaGPU#399) (@maleadt)
- Keep track of package latency timings. (JuliaGPU#400) (@maleadt)
- Update manifest (JuliaGPU#402) (@github-actions[bot])
- Latency improvements (JuliaGPU#403) (@maleadt)
- Fix bounds checking with GPU views. (JuliaGPU#404) (@maleadt)
- Force specialization for dynamic_cudacall to support more arguments. (JuliaGPU#407) (@maleadt)
- Fix some wrong pointer types in the CUDNN headers. (JuliaGPU#408) (@maleadt)
- Refactor CUSPARSE (JuliaGPU#409) (@maleadt)
- Fix typo (JuliaGPU#411) (@yixingfu)
- Update manifest (JuliaGPU#413) (@github-actions[bot])
- Simplify library wrappers by introducing a CUDA Ref (JuliaGPU#414) (@maleadt)
- Simplify and update wrappers (JuliaGPU#416) (@maleadt)
- GEMM improvements (JuliaGPU#417) (@maleadt)
- CompatHelper: add new compat entry for "BFloat16s" at version "0.1" (JuliaGPU#418) (@github-actions[bot])
- add CuSparseMatrixCOO (JuliaGPU#421) (@marius311)
- Update manifest (JuliaGPU#423) (@github-actions[bot])
- Global math mode for easy use of lower-precision functionality (JuliaGPU#424) (@maleadt)
- Improve init error message (JuliaGPU#425) (@maleadt)
- CUBLAS: wrap rot! to implement rotate! and reflect! (JuliaGPU#427) (@maleadt)
- CUFFT-related optimizations (JuliaGPU#428) (@maleadt)
- Fix reverse/view regression (JuliaGPU#429) (@maleadt)
- Update packages (JuliaGPU#433) (@maleadt)
- Introduce StridedCuArray (JuliaGPU#435) (@maleadt)
- Retry curandGenerateSeeds when OOM. (JuliaGPU#436) (@maleadt)
- Introduce DenseCuArray union (JuliaGPU#437) (@maleadt)
- Array simplifications (JuliaGPU#438) (@maleadt)
- Fix and test reverse on wrapped array. (JuliaGPU#439) (@maleadt)
- Fixes after recent array wrapper changes (JuliaGPU#441) (@maleadt)
- Adapt to GPUArrays changes. (JuliaGPU#442) (@maleadt)
- Provide CUBLAS with a pool-backed workspace. (JuliaGPU#443) (@maleadt)
- Fix finalization of copied arrays. (JuliaGPU#444) (@maleadt)
- Support for/Add CUDA 11.1 (JuliaGPU#445) (@maleadt)
- Update manifest (JuliaGPU#449) (@github-actions[bot])
- Allow use of strided vectors with mul! (gemv! and gemm!) (JuliaGPU#450) (@maleadt)
- Have convert call CuSparseArray's constructors. (JuliaGPU#451) (@maleadt)

## CUDA v1.3.3

[Diff since v1.3.2](JuliaGPU/CUDA.jl@v1.3.2...v1.3.3)


**Closed issues:**
- Type changing Array conversions give error when allowscalar(false) (JuliaGPU#344)
- getindex(::CuArray, ::Adjoint, ::Colon) fails (JuliaGPU#345)
- View with array indices causes memory copy before broadcast (JuliaGPU#384)
- Regression with Julia 1.5 (JuliaGPU#390)

**Merged pull requests:**
- Replace DevicePtr with Core.LLVMPtr. (JuliaGPU#199) (@maleadt)
- Make sure view indices reside on the GPU too. (JuliaGPU#388) (@maleadt)
- CompatHelper: Update DataStructures to v0.18 (JuliaGPU#389) (@ChrisRackauckas)

## CUDA v1.3.2

[Diff since v1.3.1](JuliaGPU/CUDA.jl@v1.3.1...v1.3.2)


**Closed issues:**
- LLVM WMMA errors  (JuliaGPU#380)

**Merged pull requests:**
- Fix handling of tests to skip. (JuliaGPU#386) (@maleadt)
- Update manifest (JuliaGPU#387) (@github-actions[bot])

## CUDA v1.3.1

[Diff since v1.3.0](JuliaGPU/CUDA.jl@v1.3.0...v1.3.1)


**Closed issues:**
- Element-wise conversion fails (JuliaGPU#378)
- atomic_min fails for Int32 in global CuDeviceArrays (JuliaGPU#379)
- Segmentation fault from @cuprint on char (JuliaGPU#381)
- error in versioninfo(), name not defined (JuliaGPU#385)

**Merged pull requests:**
- Fix docs (JuliaGPU#330) (@maleadt)
- Wrap cusparseSpMV (JuliaGPU#351) (@marius311)
- specify Cchar rather than char in the doc for @cuprint (JuliaGPU#382) (@MasonProtter)
- Adapt to LLVM.jl changes for stateless codegen. (JuliaGPU#383) (@maleadt)

## CUDA v1.3.0

[Diff since v1.2.1](JuliaGPU/CUDA.jl@v1.2.1...v1.3.0)


**Closed issues:**
- Trouble with the @. macro (JuliaGPU#346)
- NVMLError: Not Supported (code 3) (JuliaGPU#348)
- Nvidia Xavier devices: exception thrown during kernel execution on device Xavier (JuliaGPU#349)
- Could not load CUTENSOR artifact dll on Windows 10 (JuliaGPU#355)
- CuTextureArray for 3D array (JuliaGPU#357)
- Bug in julia 1.5.0 I have CUDA 11.0 installed in Ubuntu 18.04 (JuliaGPU#360)
- Callback-based logging (JuliaGPU#366)
- Artifact download timeout (JuliaGPU#369)
- `sum!` accumulates when called multiple times (JuliaGPU#370)
- nvprof does not detect kernel launches (JuliaGPU#371)
- KernelError: passing and using non-bitstype argument (JuliaGPU#372)
- CUDA.jl fails to find libcudadevrt.a due on a cluster install with multi-arch target (JuliaGPU#376)

**Merged pull requests:**
- Make the memory allocator context-aware (JuliaGPU#253) (@maleadt)
- Update manifest (JuliaGPU#347) (@github-actions[bot])
- Guard against unsupported NVML usage in the test runner. (JuliaGPU#352) (@maleadt)
- Bump CUDNN to v8.0.2 (JuliaGPU#353) (@maleadt)
- Rework thread state management (JuliaGPU#356) (@maleadt)
- Update manifest (JuliaGPU#358) (@github-actions[bot])
- Memory allocator simplifications (JuliaGPU#361) (@maleadt)
- Deduplicate code from memory pools (JuliaGPU#362) (@maleadt)
- Fix show of ArrayBuffer. (JuliaGPU#363) (@maleadt)
- Clean-up the Buffer interface. (JuliaGPU#364) (@maleadt)
- Use callback APIs to get library debug logs. (JuliaGPU#367) (@maleadt)
- Allow selecting the memcheck tool. (JuliaGPU#368) (@maleadt)
- Update GPUArrays. (JuliaGPU#373) (@maleadt)
- Update to CUDA 11.0 update 1 (JuliaGPU#374) (@maleadt)
- Number and iterate devices in versioninfo() following CUDA. (JuliaGPU#375) (@maleadt)
- Reinstate support for Julia 1.3 (JuliaGPU#377) (@maleadt)