[x265] [PATCH 00/10] AArch64: Add Neon optimisations of interp functions

[chen]

Need use X265_NS:: 




Of course, constant table may merge into one instance.




2025-03-11 11:25:13，"Pavan Tarun Chakka Venkata" <pavan.tarun at multicorewareinc.com> 

Hi,

Ran smoke test on 10 patches and getting these errors in multilib

/home/ubuntu/x265_git/source/common/aarch64/filter-prim.cpp:2253:27: error: use of undeclared identifier 'x265'
 2253 |         vabs_s16(vld1_s16(x265::g_chromaFilter[coeffIdx])));
      |                           ^
/home/ubuntu/x265_git/source/common/aarch64/filter-prim.cpp:2646:51: error: use of undeclared identifier 'x265'
 2646 |         vreinterpretq_u16_s16(vabsq_s16(vld1q_s16(x265::g_lumaFilter[coeffIdx])));
      |                                                   ^
/home/ubuntu/x265_git/source/common/aarch64/filter-prim.cpp:2843:27: error: use of undeclared identifier 'x265'
 2843 |         vabs_s16(vld1_s16(x265::g_chromaFilter[coeffIdx])));
      |                           ^
/home/ubuntu/x265_git/source/common/aarch64/filter-prim.cpp:3237:51: error: use of undeclared identifier 'x265'
 3237 |         vreinterpretq_u16_s16(vabsq_s16(vld1q_s16(x265::g_lumaFilter[coeffIdx])));
      |                                                   ^
/home/ubuntu/x265_git/source/common/aarch64/filter-prim.cpp:3305:27: error: use of undeclared identifier 'x265'
 3305 |         vabs_s16(vld1_s16(x265::g_chromaFilter[coeffIdx])));
      |                           ^
/home/ubuntu/x265_git/source/common/aarch64/filter-prim.cpp:3470:51: error: use of undeclared identifier 'x265'
 3470 |         vreinterpretq_u16_s16(vabsq_s16(vld1q_s16(x265::g_lumaFilter[coeffIdx])));
      |                                                   ^
/home/ubuntu/x265_git/source/common/aarch64/filter-prim.cpp:3643:27: error: use of undeclared identifier 'x265'
 3643 |         vabs_s16(vld1_s16(x265::g_chromaFilter[coeffIdx])));
      |                           ^
/home/ubuntu/x265_git/source/common/aarch64/filter-prim.cpp:3803:51: error: use of undeclared identifier 'x265'
 3803 |         vreinterpretq_u16_s16(vabsq_s16(vld1q_s16(x265::g_lumaFilter[coeffIdx])));
      |                                                   ^
/home/ubuntu/x265_git/source/common/aarch64/filter-prim.cpp:4058:33: error: use of undeclared identifier 'x265'
 4058 |     int16x4_t filter = vld1_s16(x265::g_chromaFilter[coeffIdx]);
      |                                 ^
/home/ubuntu/x265_git/source/common/aarch64/filter-prim.cpp:4354:40: error: use of undeclared identifier 'x265'
 4354 |     const int16x8_t filter = vld1q_s16(x265::g_lumaFilter[coeffIdx]);
      |       

Thanks,
Pavan Tarun


From: x265-devel <x265-devel-bounces at videolan.org> on behalf of Jonathan Wright <Jonathan.Wright at arm.com>
Sent: 10 March 2025 04:44 PM
To: x265-devel at videolan.org <x265-devel at videolan.org>
Cc: nd <nd at arm.com>
Subject: Re: [x265] [PATCH 00/10] AArch64: Add Neon optimisations of interp functions
 
Hi,


Could we get these patches merged to the master branch please? 


(They provide rather a lot of uplift for 10-bit encoding, 11-44% measured on an AWS Graviton 4, depending on encoding speed preset.)


Thanks,
Jonathan
From: x265-devel <x265-devel-bounces at videolan.org> on behalf of Gerda Zsejke More <gerdazsejke.more at arm.com>
Sent: 21 February 2025 16:04
To: x265-devel at videolan.org <x265-devel at videolan.org>
Cc: nd <nd at arm.com>
Subject: [x265] [PATCH 00/10] AArch64: Add Neon optimisations of interp functions
 
Hi,

This patch series adds Neon intrinsic optimisations of interp functions.

Many thanks,
Gerda

Gerda Zsejke More (10):
  AArch64: Optimise HBD interp_horiz_pp_neon
  AArch64: Optimise HBD interp_horiz_pp_neon for 10-bit
  AArch64: Optimise HBD interp_horiz_ps_neon
  AArch64: Optimise HBD interp_horiz_ps_neon for 10-bit
  AArch64: Optimise HBD interp_vert_pp_neon
  AArch64: Optimise HBD interp_vert_ps_neon
  AArch64: Optimise SBD interp_vert_ss/sp
  AArch64: Optimise HBD interp_vert_sp_neon
  AArch64: Enable existing interp_vert_ss impl for HBD
  AArch64: Clean up interp function definitions

 source/common/aarch64/filter-prim.cpp | 3561 ++++++++++++++++++++-----
 source/common/aarch64/mem-neon.h      |  154 +-
 2 files changed, 2992 insertions(+), 723 deletions(-)

--
2.39.5 (Apple Git-154)

_______________________________________________
x265-devel mailing list
x265-devel at videolan.org
https://mailman.videolan.org/listinfo/x265-devel

IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x265-devel/attachments/20250311/2e3ddb70/attachment-0001.htm>