Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Faster implementation of sum #5235

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 7 commits into from
Dec 27, 2013
Merged

Faster implementation of sum #5235

Changes from 1 commit
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
ba8881c
A faster sum implementation (using four-way accumulation)
lindahua Dec 20, 2013
32a1b4e
better explanation of the micro-optimization of sum_seq
lindahua Dec 20, 2013
a1507d5
A faster sum implementation (using four-way accumulation)
lindahua Dec 20, 2013
d3c1634
better explanation of the micro-optimization of sum_seq
lindahua Dec 20, 2013
4c03d2d
modified sum based on Steven's comments.
lindahua Dec 25, 2013
a63a93e
Merge branch 'dh/fastsum' of https://github.com/lindahua/julia into d…
lindahua Dec 25, 2013
17bc08b
move @inbounds to the beginning of the if-else block (for sum_seq)
lindahua Dec 27, 2013
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
move @inbounds to the beginning of the if-else block (for sum_seq)
  • Loading branch information
@lindahua
lindahua committed Dec 27, 2013
commit 17bc08b1aaa378ef4d8efb513f56d89e7ecaa8ec
24 changes: 12 additions & 12 deletions base/abstractarray.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1441,11 +1441,11 @@ prod(A::AbstractArray{Bool}, region) =
# a fast implementation of sum in sequential order (from left to right)
function sum_seq{T}(a::AbstractArray{T}, ifirst::Int, ilast::Int)

if ifirst + 3 >= ilast # a has at most four elements
@inbounds if ifirst + 3 >= ilast # a has at most four elements
s = zero(T)
i = ifirst
while i <= ilast
@inbounds s += a[i]
s += a[i]
i += 1
end
return s
Expand All @@ -1457,24 +1457,24 @@ function sum_seq{T}(a::AbstractArray{T}, ifirst::Int, ilast::Int)
# into four-way accumulation. Benchmark shows
# that this results in about 2x speed-up.

@inbounds s1 = a[ifirst]
@inbounds s2 = a[ifirst + 1]
@inbounds s3 = a[ifirst + 2]
@inbounds s4 = a[ifirst + 3]
s1 = a[ifirst]
s2 = a[ifirst + 1]
s3 = a[ifirst + 2]
s4 = a[ifirst + 3]

i = ifirst + 4
il = ilast - 3
while i <= il
@inbounds s1 += a[i]
@inbounds s2 += a[i+1]
@inbounds s3 += a[i+2]
@inbounds s4 += a[i+3]
s1 += a[i]
s2 += a[i+1]
s3 += a[i+2]
s4 += a[i+3]
i += 4
end
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Doesn't it work if you put @inbounds on the whole function or block of code?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Not around the whole function, but before the if should work.


while i <= ilast
@inbounds s1 += a[i]
i += 1
s1 += a[i]
i += 1
end

return s1 + s2 + s3 + s4
Expand Down