Consider the following code doing a simple computation: function test_vectorized(a, a_old) @. a[:] = a_old[:] * 2 end function test_vectorized2(a, a_old) @. a = a_old * 2 end function test_devectorized(a, a_old) nx = size(a, 1) @inbounds @simd for i = 1:nx a[i] = a_old[i…

@views function test_vectorized3(a, a_old) @. a[:] = a_old[:] * 2 end julia> @time test_vectorized3(a2, a); 0.126813 seconds (5 allocations: 208 bytes) I think views will also work for your stencils.

Performance drop for explicit indexing in fused loops

General Usage

dpsanders June 26, 2017, 1:38pm 5

Don’t benchmark in global scope.

CSV Reading (rewrite in C?)

Topic		Replies	Views
Loop over multi-dimensional array optimization Performance question	9	1132	August 15, 2019
Julia 1.0, tight-binding benchmark and array slices Performance	9	1939	September 22, 2018
Broadcast vs. scalar loop, can Julia vectorize better? Internals & Design	8	1958	February 15, 2020
Performance of views vs. explicit index referencing Performance	2	512	July 7, 2020
Migrating code from Matlab: Accessing arrays without allocating Performance	20	1993	November 19, 2018

Performance drop for explicit indexing in fused loops

Related topics