Martin Hořeňovský
d99eb8bec8
Optimize 64x64 extended multiplication implementation
...
Now we use intrinsics when possible, and fallback to optimized
implementation in portable C++. The difference is about 4x when
we can use intrinsics and about 2x when we cannot.
This should speed up our Lemire's algorithm implementation nicely.
2024-04-03 13:28:25 +02:00
..
2023-02-06 15:29:01 +01:00
2023-08-07 22:07:31 +02:00
2022-10-28 11:30:15 +02:00
2022-10-28 11:30:15 +02:00
2022-10-28 11:30:15 +02:00
2022-10-28 11:30:15 +02:00
2024-03-01 21:24:45 +01:00
2024-01-14 21:15:02 +01:00
2024-03-01 21:24:45 +01:00
2024-04-03 13:28:25 +02:00
2023-09-08 10:04:31 +02:00
2023-11-14 23:35:22 +01:00
2022-10-28 11:30:15 +02:00
2023-01-29 10:14:20 +01:00
2024-04-03 13:27:10 +02:00
2024-03-01 21:24:45 +01:00
2022-10-28 11:30:15 +02:00
2022-10-28 11:30:15 +02:00
2023-10-28 21:35:03 +02:00
2024-03-01 21:24:45 +01:00
2023-05-20 21:13:48 +02:00
2022-10-28 11:30:15 +02:00
2024-03-01 21:24:45 +01:00
2022-10-28 11:30:15 +02:00
2024-03-01 21:24:45 +01:00
2022-10-28 11:30:15 +02:00
2022-11-04 19:24:44 +01:00
2022-10-28 11:30:15 +02:00
2022-10-28 11:30:15 +02:00