Skip to content

[libc][math][c23] Fix undefined behavior in expxf16.h #112734

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 1 commit into from
Oct 17, 2024
Merged

[libc][math][c23] Fix undefined behavior in expxf16.h #112734

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
8 changes: 4 additions & 4 deletions libc/src/math/generic/expxf16.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -108,8 +108,8 @@ LIBC_INLINE ExpRangeReduction exp2_range_reduction(float16 x) {
float xf = x;
float kf = fputil::nearest_integer(xf * 0x1.0p+3f);
int x_hi_mid = static_cast<int>(kf);
int x_hi = x_hi_mid >> 3;
int x_mid = x_hi_mid & 0x7;
unsigned x_hi = static_cast<unsigned>(x_hi_mid) >> 3;
unsigned x_mid = static_cast<unsigned>(x_hi_mid) & 0x7;
Comment on lines 109 to +112
Copy link
Member Author

@overmighty overmighty Oct 17, 2024 •
edited
Loading

Choose a reason for hiding this comment

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

Casting kf (potentially negative float) to unsigned directly would also be UB: https://eel.is/c++draft/conv.fpint#1. Hence why I left x_hi_mid as int.

We really only need x_hi to be unsigned. I made x_mid unsigned too for consistency.

// lo = x - (hi + mid) = round(x * 2^3) * (-2^(-3)) + x
float lo = fputil::multiply_add(kf, -0x1.0p-3f, xf);

Expand Down Expand Up @@ -155,8 +155,8 @@ LIBC_INLINE ExpRangeReduction exp10_range_reduction(float16 x) {
float xf = x;
float kf = fputil::nearest_integer(xf * (LOG2F_10 * 0x1.0p+3f));
int x_hi_mid = static_cast<int>(kf);
int x_hi = x_hi_mid >> 3;
int x_mid = x_hi_mid & 0x7;
unsigned x_hi = static_cast<unsigned>(x_hi_mid) >> 3;
unsigned x_mid = static_cast<unsigned>(x_hi_mid) & 0x7;
// lo = x - (hi + mid) = round(x * 2^3 * log2(10)) * log10(2) * (-2^(-3)) + x
float lo = fputil::multiply_add(kf, LOG10F_2 * -0x1.0p-3f, xf);

Expand Down
Loading