ffmpeg/libavcodec
Rémi Denis-Courmont 2d514f5d48 lavc/flacdsp: do not assume maximum R-V VL
This loop correctly assumes that VLMAX=16 (4x128-bit vectors
with 32-bit elements) and 32 >= pred_order > 16. We need to alternate
between VL=16 and VL=t2=pred_order-16 elements to add up to pred_order.

The current code requests AVL=a2=pred_order elements. In QEMU and on
thte K230 hardware, this sets VL=16 as we need. But the specification
merely guarantees that we get: ceil(AVL / 2) <= VL <= VLMAX. For
instance, if pred_order equals 27, we could end up with VL=14 or VL=15
instead of VL=16. So instead, request literally VLMAX=16.

(cherry picked from commit f8837465875205207bd281ecad9e4b9a12638c7e)
Signed-off-by: Rémi Denis-Courmont <remi@remlab.net>
2024-05-25 10:34:08 +03:00
..
2024-03-15 12:51:15 +01:00
2024-02-27 00:11:21 +01:00
2024-02-27 00:11:21 +01:00
2024-03-25 08:54:18 +01:00
2024-03-15 12:51:15 +01:00
2024-03-15 12:51:15 +01:00
2024-02-12 22:00:03 +02:00
2024-02-12 22:00:03 +02:00
2024-05-22 22:04:29 +02:00
2024-03-15 12:51:15 +01:00
2024-03-15 12:51:15 +01:00
2024-03-15 12:51:15 +01:00
2024-03-15 12:51:15 +01:00
2024-03-04 17:45:00 +01:00
2024-05-22 22:04:40 +02:00
2024-05-22 22:04:39 +02:00
2024-03-10 13:51:37 +01:00
2024-05-22 22:04:30 +02:00
2024-03-15 12:51:15 +01:00
2024-03-15 12:51:15 +01:00
2024-02-29 10:30:16 +01:00
2024-02-20 17:31:40 +11:00
2024-01-25 11:07:28 -05:00
2024-02-21 20:08:25 +02:00
2024-02-21 20:08:25 +02:00
2024-03-12 14:37:49 +01:00
2024-03-15 12:51:15 +01:00
2024-03-15 12:51:15 +01:00
2024-03-15 12:51:15 +01:00
2024-03-26 23:19:49 +01:00
2024-03-12 14:37:49 +01:00
2024-03-15 12:51:15 +01:00
2024-03-27 01:04:53 +01:00
2024-02-03 00:26:17 +01:00
2024-05-22 22:04:41 +02:00
2024-05-25 01:43:05 +02:00
2024-02-17 14:45:49 +02:00
2024-02-17 14:45:49 +02:00