Lynne
ace42cf581
x86/tx_float: add 15xN PFA FFT AVX SIMD ...
~4x faster than the C version.
The shuffles in the 15pt dim1 are seriously expensive. Not happy with it,
but I'm contempt.
Can be easily converted to pure AVX by removing all vpermpd/vpermps
instructions.
2022-09-23 12:35:27 +02:00
2022-05-02 15:01:15 +05:30
2022-09-03 16:19:40 -07:00
2022-05-10 10:03:30 +02:00
2022-07-30 11:42:06 +02:00
2022-06-20 22:56:50 +02:00
2022-06-29 14:03:36 +08:00
2022-09-22 11:52:16 -07:00
2022-04-06 09:22:31 +02:00
2022-08-08 16:20:58 +02:00
2022-04-21 12:28:05 +05:30
2022-06-13 22:42:55 +02:00
2022-09-05 20:32:12 +02:00
2022-09-18 19:43:42 +02:00
2022-09-16 10:02:19 +02:00
2022-08-28 22:06:45 +02:00
2022-06-16 21:28:20 +02:00
2022-04-19 13:04:42 +02:00
2022-06-29 14:03:32 +08:00
2022-09-13 16:50:43 -03:00
2022-06-09 23:57:20 +02:00
2022-09-23 12:35:27 +02:00
2022-08-18 17:10:42 +02:00
ace42cf581