Added new api and fixed type errors in cuStateVec

[LeeQY1996]

Added new api and fixed type errors in cuStateVec

[github-actions]

Your PR requires formatting changes to meet the project's style guidelines.
Please consider running Runic (git runic master) to apply these changes.

Click here to view the suggested changes.

diff --git a/lib/custatevec/src/libcustatevec.jl b/lib/custatevec/src/libcustatevec.jl
index 62b72e654..a7037cb29 100644
--- a/lib/custatevec/src/libcustatevec.jl
+++ b/lib/custatevec/src/libcustatevec.jl
@@ -1000,7 +1000,7 @@ end
                                                                                     nIndexBits::UInt32,
                                                                                     nSVs::UInt32,
                                                                                     svStride::custatevecIndex_t,
-                                                                                    matrices::PtrOrCuPtr{Cvoid},
+        matrices::PtrOrCuPtr{Cvoid},
                                                                                     matrixDataType::cudaDataType_t,
                                                                                     layout::custatevecMatrixLayout_t,
                                                                                     nMatrices::UInt32,
@@ -1018,20 +1018,20 @@ end
                                                       extraWorkspaceSizeInBytes)
     initialize_context()
     @gcsafe_ccall libcustatevec.custatevecComputeExpectationBatched(handle::custatevecHandle_t,
-                                                                    batchedSv::CuPtr{Cvoid},
+        batchedSv::CuPtr{Cvoid},
                                                                     svDataType::cudaDataType_t,
                                                                     nIndexBits::UInt32,
                                                                     nSVs::UInt32,
                                                                     svStride::custatevecIndex_t,
-                                                                    expectationValues::PtrOrCuPtr{ComplexF64},
-                                                                    matrices::PtrOrCuPtr{Cvoid},
+        expectationValues::PtrOrCuPtr{ComplexF64},
+        matrices::PtrOrCuPtr{Cvoid},
                                                                     matrixDataType::cudaDataType_t,
                                                                     layout::custatevecMatrixLayout_t,
                                                                     nMatrices::UInt32,
                                                                     basisBits::Ptr{Int32},
                                                                     nBasisBits::UInt32,
                                                                     computeType::custatevecComputeType_t,
-                                                                    extraWorkspace::CuPtr{Cvoid},
+        extraWorkspace::CuPtr{Cvoid},
                                                                     extraWorkspaceSizeInBytes::Csize_t)::custatevecStatus_t
 end
 
diff --git a/lib/custatevec/src/statevec.jl b/lib/custatevec/src/statevec.jl
index 722680d8e..c51fd6ad8 100644
--- a/lib/custatevec/src/statevec.jl
+++ b/lib/custatevec/src/statevec.jl
@@ -107,10 +107,10 @@ function batchMeasure!(sv::CuStateVec, bitordering::Vector{<:Integer}, randnum::
     return sv, bitstring
 end
 
-function measureBatched!(sv::CuStateVec, n_svs::Int, bitordering::Vector{<:Integer},randnums::Vector{<:Integer}, collapse::custatevecCollapseOp_t=CUSTATEVEC_COLLAPSE_NONE)
+function measureBatched!(sv::CuStateVec, n_svs::Int, bitordering::Vector{<:Integer}, randnums::Vector{<:Integer}, collapse::custatevecCollapseOp_t = CUSTATEVEC_COLLAPSE_NONE)
     all(0.0 .<= randnums .< 1.0) && length(randnums) == n_svs || throw(ArgumentError("randnums must have length $nsvs and all elements must be in the interval [0, 1)."))
-    bitStrings = zeros(Int32, length(bitordering)*n_svs)
-    sv_stride    = div(length(sv.data), n_svs)
+    bitStrings = zeros(Int32, length(bitordering) * n_svs)
+    sv_stride = div(length(sv.data), n_svs)
     n_index_bits = Int(log2(div(length(sv.data), n_svs)))
     custatevecMeasureBatched(handle(), sv.data, eltype(sv), n_index_bits, n_svs, sv_stride, bitStrings, convert(Vector{Int32}, bitordering), length(bitstring), randnums, collapse)
     return sv, bitstrings
@@ -138,14 +138,14 @@ function expectation(sv::CuStateVec, matrix::Union{Matrix, CuMatrix}, basis_bits
 end
 
 function expectationBatched(sv::CuStateVec, n_svs::Int, matrix::Union{CuVector, Vector}, n_matrices::Int, basis_bits::Vector{<:Integer})
-    sv_stride    = div(length(sv.data), n_svs)
+    sv_stride = div(length(sv.data), n_svs)
     n_index_bits = Int(log2(div(length(sv.data), n_svs)))
     function bufferSize()
         out = Ref{Csize_t}()
         custatevecComputeExpectationBatchedGetWorkspaceSize(handle(), eltype(sv), n_index_bits, n_svs, sv_stride, matrix, eltype(matrix), CUSTATEVEC_MATRIX_LAYOUT_COL, n_matrices, length(basis_bits), compute_type(eltype(sv), eltype(matrix)), out)
-        out[]
+        return out[]
     end
-    expVals = zeros(ComplexF64, n_matrices*n_svs)
+    expVals = zeros(ComplexF64, n_matrices * n_svs)
     with_workspace(handle().cache, bufferSize) do buffer
         custatevecComputeExpectationBatched(handle(), sv.data, eltype(sv), n_index_bits, n_svs, sv_stride, expVals, matrix, eltype(matrix), CUSTATEVEC_MATRIX_LAYOUT_COL, n_matrices, convert(Vector{Int32}, basis_bits), length(basis_bits), compute_type(eltype(sv), eltype(matrix)), buffer, sizeof(buffer))
     end

[codecov]

Codecov Report

Attention: Patch coverage is 0% with 18 lines in your changes missing coverage. Please review.

Project coverage is 88.63%. Comparing base (430b7d6) to head (1207c36).
Report is 10 commits behind head on master.

Files with missing lines	Patch %	Lines
lib/custatevec/src/statevec.jl	0.00%	18 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #2728   +/-   ##
=======================================
  Coverage   88.62%   88.63%           
=======================================
  Files         153      153           
  Lines       13156    13174   +18     
=======================================
+ Hits        11660    11677   +17     
- Misses       1496     1497    +1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[github-actions]

CUDA.jl Benchmarks

Benchmark suite	Current: 1207c36	Previous: 430b7d6	Ratio
latency/precompile	45868224655.5 ns	45833846958 ns	1.00
latency/ttfp	6631324465.5 ns	6614791754 ns	1.00
latency/import	3171755571 ns	3161117471.5 ns	1.00
integration/volumerhs	9622766 ns	9611888 ns	1.00
integration/byval/slices=1	147117 ns	146919 ns	1.00
integration/byval/slices=3	425071 ns	425277.5 ns	1.00
integration/byval/reference	144896 ns	145019 ns	1.00
integration/byval/slices=2	286006 ns	286025 ns	1.00
integration/cudadevrt	103444.5 ns	103516 ns	1.00
kernel/indexing	14184.5 ns	14168.5 ns	1.00
kernel/indexing_checked	14834 ns	15018 ns	0.99
kernel/occupancy	715.0915492957746 ns	746.9731543624162 ns	0.96
kernel/launch	2131.2 ns	2173.4444444444443 ns	0.98
kernel/rand	14790 ns	14748 ns	1.00
array/reverse/1d	19865 ns	20211 ns	0.98
array/reverse/2d	25633 ns	25621 ns	1.00
array/reverse/1d_inplace	10667 ns	11311 ns	0.94
array/reverse/2d_inplace	12172 ns	13122 ns	0.93
array/copy	21321 ns	21481 ns	0.99
array/iteration/findall/int	159149 ns	159934.5 ns	1.00
array/iteration/findall/bool	139797 ns	140354 ns	1.00
array/iteration/findfirst/int	153977 ns	162145 ns	0.95
array/iteration/findfirst/bool	154701 ns	163586 ns	0.95
array/iteration/scalar	71235.5 ns	74124 ns	0.96
array/iteration/logical	216741 ns	218118.5 ns	0.99
array/iteration/findmin/1d	41778 ns	42265 ns	0.99
array/iteration/findmin/2d	94170 ns	94863 ns	0.99
array/reductions/reduce/1d	36252 ns	36895 ns	0.98
array/reductions/reduce/2d	40764 ns	41510 ns	0.98
array/reductions/mapreduce/1d	34041 ns	34847 ns	0.98
array/reductions/mapreduce/2d	40801 ns	50343 ns	0.81
array/broadcast	20916 ns	21159 ns	0.99
array/copyto!/gpu_to_gpu	13588 ns	11835 ns	1.15
array/copyto!/cpu_to_gpu	210253 ns	210538 ns	1.00
array/copyto!/gpu_to_cpu	243469 ns	243720 ns	1.00
array/accumulate/1d	109533 ns	109687 ns	1.00
array/accumulate/2d	79832 ns	80651 ns	0.99
array/construct	1274.8 ns	1283.7 ns	0.99
array/random/randn/Float32	49820.5 ns	45696 ns	1.09
array/random/randn!/Float32	26745 ns	27045 ns	0.99
array/random/rand!/Int64	27053 ns	27352 ns	0.99
array/random/rand!/Float32	8759 ns	8850 ns	0.99
array/random/rand/Int64	38262 ns	30500 ns	1.25
array/random/rand/Float32	13158 ns	13331 ns	0.99
array/permutedims/4d	62107 ns	62035 ns	1.00
array/permutedims/2d	55838 ns	55848 ns	1.00
array/permutedims/3d	56424 ns	56692 ns	1.00
array/sorting/1d	2776270 ns	2758536 ns	1.01
array/sorting/by	3368691.5 ns	3369076 ns	1.00
array/sorting/2d	1085447.5 ns	1078977 ns	1.01
cuda/synchronization/stream/auto	1066.7272727272727 ns	1033.4 ns	1.03
cuda/synchronization/stream/nonblocking	6479.6 ns	6533.8 ns	0.99
cuda/synchronization/stream/blocking	842 ns	857.2446808510638 ns	0.98
cuda/synchronization/context/auto	1204.6 ns	1184 ns	1.02
cuda/synchronization/context/nonblocking	6709 ns	6691.700000000001 ns	1.00
cuda/synchronization/context/blocking	935.3333333333334 ns	945.1481481481482 ns	0.99

This comment was automatically generated by workflow using github-action-benchmark.

[kshyatt]

These changes need to be replicated in the res/wrap logic too

[kshyatt]

Can we add some tests for these new functions?