[FFmpeg-devel] [PATCH v3] lavc/aarch64: hevc_add_res add 12bit variants

[J. Dekker]

On 16 Aug 2022, at 14:46, Martin Storsjö wrote:

> On Tue, 16 Aug 2022, J. Dekker wrote:
>
>> hevc_add_res_4x4_12_c: 46.0
>> hevc_add_res_4x4_12_neon: 18.7
>> hevc_add_res_8x8_12_c: 194.7
>> hevc_add_res_8x8_12_neon: 25.2
>> hevc_add_res_16x16_12_c: 716.0
>> hevc_add_res_16x16_12_neon: 69.7
>> hevc_add_res_32x32_12_c: 3820.7
>> hevc_add_res_32x32_12_neon: 261.0
>>
>> Signed-off-by: J. Dekker <jdek at itanimul.li>
>> ---
>>
>> libavcodec/aarch64/hevcdsp_idct_neon.S    | 156 ++++++++++++----------
>> libavcodec/aarch64/hevcdsp_init_aarch64.c |  34 ++---
>> 2 files changed, 105 insertions(+), 85 deletions(-)
>
> Thanks, this version seems fine to me.
>
>> diff --git a/libavcodec/aarch64/hevcdsp_init_aarch64.c b/libavcodec/aarch64/hevcdsp_init_aarch64.c
>> index 9cbe983870..b6d5efb77f 100644
>> --- a/libavcodec/aarch64/hevcdsp_init_aarch64.c
>> +++ b/libavcodec/aarch64/hevcdsp_init_aarch64.c
>> @@ -25,22 +25,18 @@
>> #include "libavutil/aarch64/cpu.h"
>> #include "libavcodec/hevcdsp.h"
>>
>> -void ff_hevc_add_residual_4x4_8_neon(uint8_t *_dst, const int16_t *coeffs,
>> -                                     ptrdiff_t stride);
>> +void ff_hevc_add_residual_4x4_8_neon(uint8_t *_dst, const int16_t *coeffs, ptrdiff_t stride);
>
> The joined forms of these lines end up a bit long, while they previously did fit below the 80 column soft-limit, so IMO I'd prefer to keep them wrapped - but it's not a big deal. (I guess it made more sense to join the lines before the 'const' was added.)
>
> // Martin

Pushed with these changes (entire set now).

Thanks,
-- 
jd