If we do split in that manner , then the recurrence relation will be :
T(n) = T(9n / 10) + T(n / 10) + O(n)
where the first branch is of the size 9n/10 and second one is n/10 so branches have nodes in 9 : 1 ratio as mentioned in the question..
And O(n) as we know for partitioning algortihm..
So solving the above using recursion tree approach , we find that the cost at each level of tree which is given is n here ..Let us see how..
At 1st level , the cost = n
At 2nd level , the cost = 9n / 10 + n / 10 = n
So cost will remain same at all levels..
So time complexity = Summation of cost of all levels
= O(n * log10/9 n) if we take longer branch [T(9n / 10)] thus getting the upper bound
= Ω(n * log10 n) which is the shorter branch
But anyway base of logarithm does not matter as it is only a matter of constant..
Hence B) is the correct answer..