Skip to content

[libc][stdfix] Implement fxdivi functions (rdivi) #154914

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 14 commits into from
Oct 8, 2025
Merged

[libc][stdfix] Implement fxdivi functions (rdivi) #154914

Changes from 1 commit
Commits
Show all changes
14 commits
Select commit Hold shift + click to select a range
f54208e
[libc][stdfix] Implement fxdivi functions
bojle Aug 22, 2025
040e833
[libc][stdfix] add a couple more tests and error info for divi
bojle Aug 30, 2025
71d283b
[libc] add a few tests to handles edge cases with FRACT_{MIN/MAX}
bojle Sep 5, 2025
cfe5122
[libc] fix formatting for rdivi.cpp
bojle Sep 5, 2025
c614888
[libc] comments, remove extra iterations, and misc changes
bojle Sep 13, 2025
40cf04a
[libc] formatting
bojle Sep 25, 2025
744463c
[libc] INT_MIN handling
bojle Sep 26, 2025
f5ff751
[libc] formatting
bojle Sep 26, 2025
31bbc8b
[libc] multiply and divide by 2 when INT_MIN
bojle Sep 26, 2025
b631411
[libc] specially handle po2 cases
bojle Oct 5, 2025
73a93c3
[libc] add special handling for po2
bojle Oct 8, 2025
303dc94
[libc] use abs from algorithm.h
bojle Oct 8, 2025
07e1de8
[libc] add doc, fix func name case
bojle Oct 8, 2025
c2b4b99
[libc] TotalBits -> TOTAL_BITS
bojle Oct 8, 2025
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
17 changes: 12 additions & 5 deletions libc/src/__support/fixed_point/fx_bits.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -281,11 +281,18 @@ template <typename XType> LIBC_INLINE constexpr XType divi(int n, int d) {
// to quadratic convergence. So,
// E1 = (0.059)^2 = 0.0034
long accum val = nrstep(d_scaled_val, initial_approx);
// E2 = 0.0000121
val = nrstep(d_scaled_val, val);
// E3 = 1.468e−10
val = nrstep(d_scaled_val, val);

auto isPowerOfTwo = [](int n) { return (n > 0) && ((n & (n - 1)) == 0); };
// Division with a power of 2 would generally be expected to be
// exact, we handle this by specially treating po2 cases and having
// extra iterations for them.
if (FXRep<XType>::FRACTION_LEN > 8 || isPowerOfTwo(cpp::abs(d))) {
// E2 = 0.0000121
val = nrstep(d_scaled_val, val);
if (FXRep<XType>::FRACTION_LEN > 16 || isPowerOfTwo(cpp::abs(d))) {
// E3 = 1.468e−10
val = nrstep(d_scaled_val, val);
}
}
long accum res = n_scaled_val * val;

if (result_is_negative) {
Expand Down
Loading