Skip to content
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

Jump to bottom

Rebase: Performance regression of scalar randn() between Julia 1.4 and 1.5 #39319

Merged
merged 3 commits into from
Jan 19, 2021
Merged

Rebase: Performance regression of scalar randn() between Julia 1.4 and 1.5 #39319

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
f1a30b7
definition of randn for scalars reverted to 1.4
lungben Aug 27, 2020
e95b38b
Added comments why this change is done
lungben Aug 28, 2020
c4fe755
Update normal.jl
KristofferC Sep 12, 2020
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
21 changes: 20 additions & 1 deletion stdlib/Random/src/normal.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,7 +35,26 @@ julia> randn(rng, ComplexF32, (2, 3))
0.611224+1.56403im 0.355204-0.365563im 0.0905552+1.31012im
```
"""
@inline randn(rng::AbstractRNG=default_rng()) = _randn(rng, rand(rng, UInt52Raw()))
@inline function randn(rng::AbstractRNG=default_rng())
#=
When defining
`@inline randn(rng::AbstractRNG=default_rng()) = _randn(rng, rand(rng, UInt52Raw()))`
the function call to `_randn` is currently not inlined, resulting in slightly worse
performance for scalar random normal numbers than repeating the code of `_randn`
inside the following function.
=#
@inbounds begin
r = rand(rng, UInt52Raw())

# the following code is identical to the one in `_randn(rng::AbstractRNG, r::UInt64)`
r &= 0x000fffffffffffff
rabs = Int64(r>>1) # One bit for the sign
idx = rabs & 0xFF
x = ifelse(r % Bool, -rabs, rabs)*wi[idx+1]
rabs < ki[idx+1] && return x # 99.3% of the time we return here 1st try
return randn_unlikely(rng, idx, rabs, x)
end
end

@inline function _randn(rng::AbstractRNG, r::UInt64)
@inbounds begin
Expand Down