libvpx: Pull from upstream

source/libvpx/vp9/encoder/x86/vp9_quantize_ssse3_x86_64.asm

 ;
 ;  Copyright (c) 2010 The WebM project authors. All Rights Reserved.
 ;
 ;  Use of this source code is governed by a BSD-style license
 ;  that can be found in the LICENSE file in the root of the source
 ;  tree. An additional intellectual property rights grant can be found
 ;  in the file PATENTS.  All contributing project authors may
 ;  be found in the AUTHORS file in the root of the source tree.
 ;
 
@@ skipping 264 matching lines @@
 %endif
   pmullw                          m8, m3                   ; r4[i] = r3[i] * q
   punpckhqdq                      m3, m3
   pmullw                         m13, m3                   ; r4[i] = r3[i] * q
 %ifidn %1, fp_32x32
   psrlw                           m8, 1
   psrlw                          m13, 1
   psignw                          m8, m9
   psignw                         m13, m10
   psrlw                           m0, m3, 2
+%else
+  psrlw                           m0, m3, 1
 %endif
   mova            [r4q+ncoeffq*2+ 0], m8
   mova            [r4q+ncoeffq*2+16], m13
   pcmpeqw                         m8, m5                   ; m8 = c[i] == 0
   pcmpeqw                        m13, m5                   ; m13 = c[i] == 0
   mova                            m6, [  r5q+ncoeffq*2+ 0] ; m6 = scan[i]
   mova                           m11, [  r5q+ncoeffq*2+16] ; m11 = scan[i]
   psubw                           m6, m7                   ; m6 = scan[i] + 1
   psubw                          m11, m7                   ; m11 = scan[i] + 1
   pandn                           m8, m6                   ; m8 = max(eob)
   pandn                          m13, m11                  ; m13 = max(eob)
   pmaxsw                          m8, m13
   add                        ncoeffq, mmsize
   jz .accumulate_eob
 
 .ac_only_loop:
   mova                            m9, [  coeffq+ncoeffq*2+ 0] ; m9 = c[i]
   mova                           m10, [  coeffq+ncoeffq*2+16] ; m10 = c[i]
   pabsw                           m6, m9                   ; m6 = abs(m9)
   pabsw                          m11, m10                  ; m11 = abs(m10)
-%ifidn %1, fp_32x32
+
   pcmpgtw                         m7, m6,  m0
   pcmpgtw                        m12, m11, m0
   pmovmskb                       r6d, m7
   pmovmskb                       r2d, m12
 
   or                              r6, r2
   jz .skip_iter
-%endif
+
   pcmpeqw                         m7, m7
 
   paddsw                          m6, m1                   ; m6 += round
   paddsw                         m11, m1                   ; m11 += round
   pmulhw                         m14, m6, m2               ; m14 = m6*q>>16
   pmulhw                         m13, m11, m2              ; m13 = m11*q>>16
   psignw                         m14, m9                   ; m14 = reinsert sign
   psignw                         m13, m10                  ; m13 = reinsert sign
   mova            [r3q+ncoeffq*2+ 0], m14
   mova            [r3q+ncoeffq*2+16], m13
@@ skipping 17 matching lines @@
   mova                           m11, [  r5q+ncoeffq*2+16] ; m11 = scan[i]
   psubw                           m6, m7                   ; m6 = scan[i] + 1
   psubw                          m11, m7                   ; m11 = scan[i] + 1
   pandn                          m14, m6                   ; m14 = max(eob)
   pandn                          m13, m11                  ; m13 = max(eob)
   pmaxsw                          m8, m14
   pmaxsw                          m8, m13
   add                        ncoeffq, mmsize
   jl .ac_only_loop
 
-%ifidn %1, fp_32x32
   jmp .accumulate_eob
 .skip_iter:
   mova            [r3q+ncoeffq*2+ 0], m5
   mova            [r3q+ncoeffq*2+16], m5
   mova            [r4q+ncoeffq*2+ 0], m5
   mova            [r4q+ncoeffq*2+16], m5
   add                        ncoeffq, mmsize
   jl .ac_only_loop
-%endif
 
 .accumulate_eob:
   ; horizontally accumulate/max eobs and write into [eob] memory pointer
   mov                             r2, eobmp
   pshufd                          m7, m8, 0xe
   pmaxsw                          m8, m7
   pshuflw                         m7, m8, 0xe
   pmaxsw                          m8, m7
   pshuflw                         m7, m8, 0x1
   pmaxsw                          m8, m7
@@ skipping 19 matching lines @@
   mova            [r2q+ncoeffq*2+16], m7
   add                        ncoeffq, mmsize
   jl .blank_loop
   mov                     word [r3q], 0
   RET
 %endmacro
 
 INIT_XMM ssse3
 QUANTIZE_FP fp, 7
 QUANTIZE_FP fp_32x32, 7