Skip to content

Simplify and speed-up math.hypot() and math.dist() #102734

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
rhettinger merged 6 commits into from
Mar 15, 2023
Merged

Simplify and speed-up math.hypot() and math.dist() #102734

Changes from 1 commit
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
cd7cfc7
Move double and triple length helper functions to the top
rhettinger Mar 15, 2023
cd5fb29
Add more double length functions to support vector_norm
rhettinger Mar 15, 2023
8f3e73c
Use DoubleLength functions in hypot().
rhettinger Mar 15, 2023
58377df
Update comments
rhettinger Mar 15, 2023
47e878d
Remove unused code
rhettinger Mar 15, 2023
9a63c70
Merge branch 'main' into faster_hypot
rhettinger Mar 15, 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
Prev Previous commit
Next Next commit
Remove unused code
  • Loading branch information
@rhettinger
rhettinger committed Mar 15, 2023
commit 47e878d2f139076adfc587cbd0e86248582884c3
10 changes: 1 addition & 9 deletions Modules/mathmodule.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -174,15 +174,6 @@ dl_mul(double x, double y)
return (DoubleLength) {z, zz};
}

static DoubleLength
dl_fma(double x, double y, DoubleLength total)
{
/* Algorithm 5.10 with SumKVert for K=2 */
DoubleLength pr = dl_mul(x, y);
DoubleLength sm = dl_sum(total.hi, pr.hi);
return DoubleLength {sm.hi, pr.lo + sm.lo + total.lo};
}

#endif

typedef struct { double hi; double lo; double tiny; } TripleLength;
Expand Down Expand Up @@ -2535,6 +2526,7 @@ vector_norm(Py_ssize_t n, double *vec, double max, int found_nan)

pr = dl_mul(x, x);
assert(pr.hi <= 1.0);

sm = dl_fast_sum(csum, pr.hi);
csum = sm.hi;
frac1 += pr.lo;
Expand Down