Skip to content

fix incorrect results in expm1(::Union{Float16, Float32}) #50989

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 2 commits into from
Aug 21, 2023
Merged

fix incorrect results in expm1(::Union{Float16, Float32}) #50989

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
b37bc9a
fix UB in `expm1(::Union{Float16, Float32})
oscardssmith Aug 20, 2023
b6aaa57
add test
oscardssmith Aug 20, 2023
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
4 changes: 2 additions & 2 deletions base/special/exp.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -460,7 +460,7 @@ function expm1(x::Float32)
end
x = Float64(x)
N_float = round(x*Ln2INV(Float64))
N = unsafe_trunc(UInt64, N_float)
N = unsafe_trunc(Int64, N_float)
Copy link
Member

Choose a reason for hiding this comment

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

I imagine MAX_EXP is much smaller than the maximum integer here? If so LGTM

Copy link
Member Author

Choose a reason for hiding this comment

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

yeah. N will end up as the exponent for the float so it will be in [-128,128]

r = muladd(N_float, Ln2(Float64), x)
hi = evalpoly(r, (1.0, .5, 0.16666667546642386, 0.041666183019487026,
0.008332997481506921, 0.0013966479175977883, 0.0002004037059220124))
Expand All @@ -477,7 +477,7 @@ function expm1(x::Float16)
return Float16(x*evalpoly(x, (1f0, .5f0, 0.16666628f0, 0.04166785f0, 0.008351848f0, 0.0013675707f0)))
end
N_float = round(x*Ln2INV(Float32))
N = unsafe_trunc(UInt32, N_float)
N = unsafe_trunc(Int32, N_float)
r = muladd(N_float, Ln2(Float32), x)
hi = evalpoly(r, (1f0, .5f0, 0.16666667f0, 0.041665863f0, 0.008333111f0, 0.0013981499f0, 0.00019983904f0))
small_part = r*hi
Expand Down
1 change: 1 addition & 0 deletions test/math.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -188,6 +188,7 @@ end
@test exp10(x) ≈ exp10(big(x))
@test exp2(x) ≈ exp2(big(x))
@test expm1(x) ≈ expm1(big(x))
@test expm1(T(-1.1)) ≈ expm1(big(T(-1.1)))
@test hypot(x,y) ≈ hypot(big(x),big(y))
@test hypot(x,x,y) ≈ hypot(hypot(big(x),big(x)),big(y))
@test hypot(x,x,y,y) ≈ hypot(hypot(big(x),big(x)),hypot(big(y),big(y)))
Expand Down