Arpad Panyik
1e9d2006ae
AArch64: Use better block copy8
The vector copy is only necessary for 16-byte blocks on AArch64.
Decompression uplifts on a Neoverse V2 system, using Zstd-1.5.8
compiled with "-O3 -march=armv8.2-a+sve2":
Clang-19 Clang-20 GCC-14 GCC-15
1#silesia.tar: +0.316% +0.865% +0.025% +0.096%
2#silesia.tar: +0.689% +1.374% +0.027% +0.065%
3#silesia.tar: +0.811% +1.654% +0.034% +0.033%
4#silesia.tar: +0.912% +1.755% +0.027% +0.042%
5#silesia.tar: +0.995% +1.826% +0.062% +0.094%
6#silesia.tar: +0.976% +1.777% +0.065% +0.104%
7#silesia.tar: +0.910% +1.738% +0.077% +0.110%
2025-06-20 17:05:41 +00:00
..
2023-06-14 20:07:18 -07:00
2025-02-26 07:41:34 -08:00
2025-01-24 03:03:21 +02:00
2025-06-11 12:14:22 +00:00
2024-12-04 06:47:32 -08:00
2023-11-17 09:54:10 -08:00
2024-12-18 09:56:50 -08:00
2024-05-15 10:47:50 +08:00
2024-12-20 10:37:00 -08:00
2024-12-19 16:00:11 -08:00
2024-06-20 17:46:16 -04:00
2025-01-19 18:05:57 -08:00
2024-12-19 16:00:11 -08:00
2025-01-21 02:27:39 +02:00
2024-01-15 11:16:46 -08:00
2024-12-19 16:00:11 -08:00
2025-03-07 10:31:19 -05:00
2023-06-20 08:34:26 -07:00
2024-12-19 16:00:11 -08:00
2023-11-16 16:19:25 -08:00
2025-01-21 02:27:39 +02:00
2025-03-07 10:31:19 -05:00
2024-07-29 17:13:58 -07:00
2025-06-20 17:05:41 +00:00
2025-01-07 10:06:08 -08:00