Removed allocations for transpose/adjoint - diagonal multiplications (#2538)

Author: mkhrenov

lib/cublas/linalg.jl

@@ -347,19 +347,27 @@ function LinearAlgebra.mul!(C::CuMatrix{T}, A::Diagonal{T,<:CuVector}, B::CuMatr
 end
 
 function LinearAlgebra.mul!(C::CuMatrix{T}, A::Transpose{T,<:CuMatrix}, B::Diagonal{T,<:CuVector}) where {T<:CublasFloat}
-    return dgmm!('R', CuMatrix(A), B.diag, C)
+    C .= A
+    C .*= transpose(B.diag)
+    return C
 end
 
 function LinearAlgebra.mul!(C::CuMatrix{T}, A::Diagonal{T,<:CuVector}, B::Transpose{T,<:CuMatrix}) where {T<:CublasFloat}
-    return dgmm!('L', CuMatrix(B), A.diag, C)
+    C .= B
+    C .*= A.diag
+    return C
 end
 
 function LinearAlgebra.mul!(C::CuMatrix{T}, A::Adjoint{T,<:CuMatrix}, B::Diagonal{T,<:CuVector}) where {T<:CublasFloat}
-    return dgmm!('R', CuMatrix(A), B.diag, C)
+    C .= A
+    C .*= transpose(B.diag)
+    return C
 end
 
 function LinearAlgebra.mul!(C::CuMatrix{T}, A::Diagonal{T,<:CuVector}, B::Adjoint{T,<:CuMatrix}) where {T<:CublasFloat}
-    return dgmm!('L', CuMatrix(B), A.diag, C)
+    C .= B
+    C .*= A.diag
+    return C
 end
 
 # symmetric mul!