[x265] [PATCH 218 of 307] x86: disable similar perfoamnce kernels - need more analysis

[mythreyi at multicorewareinc.com]

# HG changeset patch
# User Praveen Tiwari <praveen at multicorewareinc.com>
# Date 1512035958 -19800
#      Thu Nov 30 15:29:18 2017 +0530
# Node ID d7af8d747bffacafa5dfe8f4d513bbd09314ad63
# Parent  aff686238f2a30ab42b0e2ad296be54e77179531
x86: disable similar perfoamnce kernels - need more analysis

Currently these kernels performance are similar to AVX2 version, we need a to improve
them further to ebable it. Probably a Vtune analysis will help here.

diff -r aff686238f2a -r d7af8d747bff source/common/x86/asm-primitives.cpp
--- a/source/common/x86/asm-primitives.cpp	Wed Nov 29 20:42:07 2017 -0800
+++ b/source/common/x86/asm-primitives.cpp	Thu Nov 30 15:29:18 2017 +0530
@@ -5004,8 +5004,13 @@
         p.pu[LUMA_64x32].luma_vpp = PFX(interp_8tap_vert_pp_64x32_avx512);
         p.pu[LUMA_64x16].luma_vpp = PFX(interp_8tap_vert_pp_64x16_avx512);
         p.cu[BLOCK_8x8].dct    = PFX(dct8_avx512);
-        p.cu[BLOCK_16x16].dct  = PFX(dct16_avx512);
-        p.cu[BLOCK_32x32].dct  = PFX(dct32_avx512);
+
+        /* TODO: Currently these kernels performance are similar to AVX2 version, we need a to improve them further to ebable
+         * it. Probably a Vtune analysis will help here.
+
+         * p.cu[BLOCK_16x16].dct  = PFX(dct16_avx512);
+         * p.cu[BLOCK_32x32].dct  = PFX(dct32_avx512); */
+
         p.cu[BLOCK_8x8].idct   = PFX(idct8_avx512);
         p.cu[BLOCK_16x16].idct = PFX(idct16_avx512);
         p.cu[BLOCK_32x32].idct = PFX(idct32_avx512);