Skip to content

Replace memset with constexpr fill_n in bigint::align and fix memset count to account for sizeof T #4471

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
vitaut merged 4 commits into from
Jun 23, 2025
Merged

Replace memset with constexpr fill_n in bigint::align and fix memset count to account for sizeof T #4471

Changes from 3 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
5ffa586
Use fill_n in place of memset in bigint::align to respect constexpr
Tomek-Stolarczyk Jun 18, 2025
90a59f5
Since memset operates on (unsigned char), we should multiply the coun…
Tomek-Stolarczyk Jun 18, 2025
d174022
Remove the sizeof(T) from fill_n taking char and replace with a stati…
Tomek-Stolarczyk Jun 21, 2025
7a9392b
Fix some nit comments and format errors to split the static assert ac…
Tomek-Stolarczyk Jun 22, 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
3 changes: 2 additions & 1 deletion include/fmt/format.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -553,6 +553,7 @@ FMT_CONSTEXPR auto fill_n(OutputIt out, Size count, const T& value)
template <typename T, typename Size>
FMT_CONSTEXPR20 auto fill_n(T* out, Size count, char value) -> T* {
if (is_constant_evaluated()) return fill_n<T*, Size, T>(out, count, value);
static_assert(1 == sizeof(T), "sizeof(T) must be 1 to use char for initialization");
std::memset(out, value, to_unsigned(count));
return out + count;
}
Expand Down Expand Up @@ -2799,7 +2800,7 @@ class bigint {
bigits_.resize(to_unsigned(num_bigits + exp_difference));
for (int i = num_bigits - 1, j = i + exp_difference; i >= 0; --i, --j)
bigits_[j] = bigits_[i];
memset(bigits_.data(), 0, to_unsigned(exp_difference) * sizeof(bigit));
fill_n(bigits_.data(), to_unsigned(exp_difference), 0U);
Copy link
Contributor

@vitaut vitaut Jun 22, 2025

Choose a reason for hiding this comment

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

Why U?

Copy link
Contributor Author

@Tomek-Stolarczyk Tomek-Stolarczyk Jun 22, 2025

Choose a reason for hiding this comment

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

bigit is a uint32_t so bigits_.data() returns a uint32_t*. 0U is the unsigned literal for 0. I could use fill_n(bigits_.data(), to_unsigned(exp_difference), to_unsigned(0)); if that's better but thought using the literal U would convey the same meaning. LMK if you prefer I change it to something else. The 0 is by default signed so I got a signed/unsigned warning.

exp_ -= exp_difference;
}

Expand Down
Loading