From: Yi Luo <luoyi@google.com>
Date: Wed, 23 Mar 2016 18:30:39 +0000 (+0000)
Subject: Merge "Highbd fht4x4 SSE4.1 optimization for DCT_DCT mode - Setup function vp10_highb... 
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=deb33056d127f0c178f3415caa7032497ba14f61;p=libvpx

Merge "Highbd fht4x4 SSE4.1 optimization for DCT_DCT mode - Setup function vp10_highbd_fht4x4_sse4_1 for highbd SSE4.1   intrinsics optimization. - Wrote SSE4.1 functions: load_buffer_4x4(), write_buffer_4x4(),   and fdct4x4_sse4_1(). - Used logic right shift to avoid coeff memory write/read. - Turned on vp10_highbd_fht4x4_sse4_1 for DCT_DCT mode only. - Improved overall encoding performance >2.3% for 50 frames   sequence, park_joy_1080p_12.y4m, in which, --input-bit-depth=12,   --bit-depth=12, 50 frames. - Unit test passed." into nextgenv2
---

deb33056d127f0c178f3415caa7032497ba14f61
diff --cc vp10/encoder/hybrid_fwd_txfm.c
index 785fef088,58eb7483a..e9dd70be6
--- a/vp10/encoder/hybrid_fwd_txfm.c
+++ b/vp10/encoder/hybrid_fwd_txfm.c
@@@ -213,14 -227,17 +215,14 @@@ void vp10_highbd_fwd_txfm_4x4(const int
      case FLIPADST_FLIPADST:
      case ADST_FLIPADST:
      case FLIPADST_ADST:
-       vp10_highbd_fht4x4(src_diff, coeff, diff_stride, tx_type);
+       vp10_highbd_fht4x4_c(src_diff, coeff, diff_stride, tx_type);
        break;
 -    case DST_DST:
 -    case DCT_DST:
 -    case DST_DCT:
 -    case DST_ADST:
 -    case ADST_DST:
 -    case DST_FLIPADST:
 -    case FLIPADST_DST:
 -    case H_DCT:
      case V_DCT:
 +    case H_DCT:
 +    case V_ADST:
 +    case H_ADST:
 +    case V_FLIPADST:
 +    case H_FLIPADST:
        vp10_highbd_fht4x4_c(src_diff, coeff, diff_stride, tx_type);
        break;
      case IDTX: