<div dir="ltr"><div><div dir="ltr" class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><span style="color:rgb(0,0,0);font-family:arial;font-size:14px">All the patches of this series have been pushed to the master branch.<br></span><b>__________________________</b></div><div><b>Karam Singh</b></div><div><b>Ph.D. IIT Guwahati</b></div><div><font size="1">Senior Software (Video Coding) Engineer  </font></div><div><font size="1">Mobile: +91 8011279030</font></div><div><font size="1">Block 9A, 6th floor, DLF Cyber City</font></div><div><font size="1">Manapakkam, Chennai 600 089</font></div></div></div></div><br></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Tue, Aug 13, 2024 at 2:44 AM Hari Limaye <<a href="mailto:hari.limaye@arm.com" target="_blank">hari.limaye@arm.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Hi,<br>
<br>
This series optimises the AArch64 Neon implementations of quant and<br>
nquant primitives.<br>
<br>
The series has no dependencies on any other on-list patch sets, except<br>
for patch:<br>
    [PATCH 1/4] AArch64: Remove SVE assembly implementation of quant<br>
which is based on refactoring changes to aarch64/asm-primitives.cpp made<br>
in the SAD patch series:<br>
    <a href="https://mailman.videolan.org/pipermail/x265-devel/2024-May/013707.html" rel="noreferrer" target="_blank">https://mailman.videolan.org/pipermail/x265-devel/2024-May/013707.html</a><br>
<br>
Relative performance observed compared to the existing Neon<br>
implementation:<br>
<br>
quant_neon:<br>
<br>
  Neoverse N1: 1.57x<br>
  Neoverse V1: 1.59x<br>
  Neoverse N2: 1.54x<br>
  Neoverse V2: 1.59x<br>
<br>
nquant_neon:<br>
<br>
  Neoverse N1: 1.79x<br>
  Neoverse V1: 1.77x<br>
  Neoverse N2: 1.70x<br>
  Neoverse V2: 1.73x<br>
<br>
Many thanks,<br>
<br>
Hari<br>
<br>
Hari Limaye (4):<br>
  AArch64: Remove SVE assembly implementation of quant<br>
  AArch64: Optimise quant_neon<br>
  Test: Update values used in check_nquant_primitive<br>
  AArch64: Optimise nquant_neon<br>
<br>
 source/common/aarch64/asm-primitives.cpp |   3 -<br>
 source/common/aarch64/fun-decls.h        |   2 -<br>
 source/common/aarch64/pixel-util-sve.S   |  57 -------------<br>
 source/common/aarch64/pixel-util.S       | 103 ++++++++++++-----------<br>
 source/test/mbdstharness.cpp             |  10 ++-<br>
 5 files changed, 61 insertions(+), 114 deletions(-)<br>
<br>
-- <br>
2.42.1<br>
<br>
_______________________________________________<br>
x265-devel mailing list<br>
<a href="mailto:x265-devel@videolan.org" target="_blank">x265-devel@videolan.org</a><br>
<a href="https://mailman.videolan.org/listinfo/x265-devel" rel="noreferrer" target="_blank">https://mailman.videolan.org/listinfo/x265-devel</a><br>
</blockquote></div>