Merge "remove sad x3,x8 specializations" into main

4 files changed, 0 insertions, 1166 deletions

diff --git a/vpx_dsp/x86/sad4d_avx2.c b/vpx_dsp/x86/sad4d_avx2.c
index 81f1a916f..399b67b3f 100644
--- a/vpx_dsp/x86/sad4d_avx2.c
+++ b/vpx_dsp/x86/sad4d_avx2.c
@@ -73,63 +73,6 @@ void vpx_sad32x32x4d_avx2(const uint8_t *src_ptr, int src_stride,
   calc_final_4(sums, sad_array);
 }
 
-void vpx_sad32x32x8_avx2(const uint8_t *src_ptr, int src_stride,
-                         const uint8_t *ref_ptr, int ref_stride,
-                         uint32_t sad_array[8]) {
-  int i;
-  __m256i sums[8];
-
-  sums[0] = _mm256_setzero_si256();
-  sums[1] = _mm256_setzero_si256();
-  sums[2] = _mm256_setzero_si256();
-  sums[3] = _mm256_setzero_si256();
-  sums[4] = _mm256_setzero_si256();
-  sums[5] = _mm256_setzero_si256();
-  sums[6] = _mm256_setzero_si256();
-  sums[7] = _mm256_setzero_si256();
-
-  for (i = 0; i < 32; i++) {
-    __m256i r[8];
-
-    // load src and all ref[]
-    const __m256i s = _mm256_load_si256((const __m256i *)src_ptr);
-    r[0] = _mm256_loadu_si256((const __m256i *)&ref_ptr[0]);
-    r[1] = _mm256_loadu_si256((const __m256i *)&ref_ptr[1]);
-    r[2] = _mm256_loadu_si256((const __m256i *)&ref_ptr[2]);
-    r[3] = _mm256_loadu_si256((const __m256i *)&ref_ptr[3]);
-    r[4] = _mm256_loadu_si256((const __m256i *)&ref_ptr[4]);
-    r[5] = _mm256_loadu_si256((const __m256i *)&ref_ptr[5]);
-    r[6] = _mm256_loadu_si256((const __m256i *)&ref_ptr[6]);
-    r[7] = _mm256_loadu_si256((const __m256i *)&ref_ptr[7]);
-
-    // sum of the absolute differences between every ref[] to src
-    r[0] = _mm256_sad_epu8(r[0], s);
-    r[1] = _mm256_sad_epu8(r[1], s);
-    r[2] = _mm256_sad_epu8(r[2], s);
-    r[3] = _mm256_sad_epu8(r[3], s);
-    r[4] = _mm256_sad_epu8(r[4], s);
-    r[5] = _mm256_sad_epu8(r[5], s);
-    r[6] = _mm256_sad_epu8(r[6], s);
-    r[7] = _mm256_sad_epu8(r[7], s);
-
-    // sum every ref[]
-    sums[0] = _mm256_add_epi32(sums[0], r[0]);
-    sums[1] = _mm256_add_epi32(sums[1], r[1]);
-    sums[2] = _mm256_add_epi32(sums[2], r[2]);
-    sums[3] = _mm256_add_epi32(sums[3], r[3]);
-    sums[4] = _mm256_add_epi32(sums[4], r[4]);
-    sums[5] = _mm256_add_epi32(sums[5], r[5]);
-    sums[6] = _mm256_add_epi32(sums[6], r[6]);
-    sums[7] = _mm256_add_epi32(sums[7], r[7]);
-
-    src_ptr += src_stride;
-    ref_ptr += ref_stride;
-  }
-
-  calc_final_4(sums, sad_array);
-  calc_final_4(sums + 4, sad_array + 4);
-}
-
 void vpx_sad64x64x4d_avx2(const uint8_t *src_ptr, int src_stride,
                           const uint8_t *const ref_array[4], int ref_stride,
                           uint32_t sad_array[4]) {
diff --git a/vpx_dsp/x86/sad_sse3.asm b/vpx_dsp/x86/sad_sse3.asm
deleted file mode 100644
index acbd2e4fa..000000000
--- a/vpx_dsp/x86/sad_sse3.asm
+++ /dev/null
@@ -1,376 +0,0 @@
-;
-;  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.
-;
-
-%include "vpx_ports/x86_abi_support.asm"
-
-%macro STACK_FRAME_CREATE_X3 0
-%if ABI_IS_32BIT
-  %define     src_ptr       rsi
-  %define     src_stride    rax
-  %define     ref_ptr       rdi
-  %define     ref_stride    rdx
-  %define     end_ptr       rcx
-  %define     ret_var       rbx
-  %define     result_ptr    arg(4)
-  %define     height        dword ptr arg(4)
-    push        rbp
-    mov         rbp,        rsp
-    push        rsi
-    push        rdi
-    push        rbx
-
-    mov         rsi,        arg(0)              ; src_ptr
-    mov         rdi,        arg(2)              ; ref_ptr
-
-    movsxd      rax,        dword ptr arg(1)    ; src_stride
-    movsxd      rdx,        dword ptr arg(3)    ; ref_stride
-%else
-  %if LIBVPX_YASM_WIN64
-    SAVE_XMM 7, u
-    %define     src_ptr     rcx
-    %define     src_stride  rdx
-    %define     ref_ptr     r8
-    %define     ref_stride  r9
-    %define     end_ptr     r10
-    %define     ret_var     r11
-    %define     result_ptr  [rsp+xmm_stack_space+8+4*8]
-    %define     height      dword ptr [rsp+xmm_stack_space+8+4*8]
-  %else
-    %define     src_ptr     rdi
-    %define     src_stride  rsi
-    %define     ref_ptr     rdx
-    %define     ref_stride  rcx
-    %define     end_ptr     r9
-    %define     ret_var     r10
-    %define     result_ptr  r8
-    %define     height      r8
-  %endif
-%endif
-
-%endmacro
-
-%macro STACK_FRAME_DESTROY_X3 0
-  %define     src_ptr
-  %define     src_stride
-  %define     ref_ptr
-  %define     ref_stride
-  %define     end_ptr
-  %define     ret_var
-  %define     result_ptr
-  %define     height
-
-%if ABI_IS_32BIT
-    pop         rbx
-    pop         rdi
-    pop         rsi
-    pop         rbp
-%else
-  %if LIBVPX_YASM_WIN64
-    RESTORE_XMM
-  %endif
-%endif
-    ret
-%endmacro
-
-%macro PROCESS_16X2X3 5
-%if %1==0
-        movdqa          xmm0,       XMMWORD PTR [%2]
-        lddqu           xmm5,       XMMWORD PTR [%3]
-        lddqu           xmm6,       XMMWORD PTR [%3+1]
-        lddqu           xmm7,       XMMWORD PTR [%3+2]
-
-        psadbw          xmm5,       xmm0
-        psadbw          xmm6,       xmm0
-        psadbw          xmm7,       xmm0
-%else
-        movdqa          xmm0,       XMMWORD PTR [%2]
-        lddqu           xmm1,       XMMWORD PTR [%3]
-        lddqu           xmm2,       XMMWORD PTR [%3+1]
-        lddqu           xmm3,       XMMWORD PTR [%3+2]
-
-        psadbw          xmm1,       xmm0
-        psadbw          xmm2,       xmm0
-        psadbw          xmm3,       xmm0
-
-        paddw           xmm5,       xmm1
-        paddw           xmm6,       xmm2
-        paddw           xmm7,       xmm3
-%endif
-        movdqa          xmm0,       XMMWORD PTR [%2+%4]
-        lddqu           xmm1,       XMMWORD PTR [%3+%5]
-        lddqu           xmm2,       XMMWORD PTR [%3+%5+1]
-        lddqu           xmm3,       XMMWORD PTR [%3+%5+2]
-
-%if %1==0 || %1==1
-        lea             %2,         [%2+%4*2]
-        lea             %3,         [%3+%5*2]
-%endif
-
-        psadbw          xmm1,       xmm0
-        psadbw          xmm2,       xmm0
-        psadbw          xmm3,       xmm0
-
-        paddw           xmm5,       xmm1
-        paddw           xmm6,       xmm2
-        paddw           xmm7,       xmm3
-%endmacro
-
-%macro PROCESS_8X2X3 5
-%if %1==0
-        movq            mm0,       QWORD PTR [%2]
-        movq            mm5,       QWORD PTR [%3]
-        movq            mm6,       QWORD PTR [%3+1]
-        movq            mm7,       QWORD PTR [%3+2]
-
-        psadbw          mm5,       mm0
-        psadbw          mm6,       mm0
-        psadbw          mm7,       mm0
-%else
-        movq            mm0,       QWORD PTR [%2]
-        movq            mm1,       QWORD PTR [%3]
-        movq            mm2,       QWORD PTR [%3+1]
-        movq            mm3,       QWORD PTR [%3+2]
-
-        psadbw          mm1,       mm0
-        psadbw          mm2,       mm0
-        psadbw          mm3,       mm0
-
-        paddw           mm5,       mm1
-        paddw           mm6,       mm2
-        paddw           mm7,       mm3
-%endif
-        movq            mm0,       QWORD PTR [%2+%4]
-        movq            mm1,       QWORD PTR [%3+%5]
-        movq            mm2,       QWORD PTR [%3+%5+1]
-        movq            mm3,       QWORD PTR [%3+%5+2]
-
-%if %1==0 || %1==1
-        lea             %2,        [%2+%4*2]
-        lea             %3,        [%3+%5*2]
-%endif
-
-        psadbw          mm1,       mm0
-        psadbw          mm2,       mm0
-        psadbw          mm3,       mm0
-
-        paddw           mm5,       mm1
-        paddw           mm6,       mm2
-        paddw           mm7,       mm3
-%endmacro
-
-SECTION .text
-
-;void int vpx_sad16x16x3_sse3(
-;    unsigned char *src_ptr,
-;    int  src_stride,
-;    unsigned char *ref_ptr,
-;    int  ref_stride,
-;    int  *results)
-globalsym(vpx_sad16x16x3_sse3)
-sym(vpx_sad16x16x3_sse3):
-
-    STACK_FRAME_CREATE_X3
-
-        PROCESS_16X2X3 0, src_ptr, ref_ptr, src_stride, ref_stride
-        PROCESS_16X2X3 1, src_ptr, ref_ptr, src_stride, ref_stride
-        PROCESS_16X2X3 1, src_ptr, ref_ptr, src_stride, ref_stride
-        PROCESS_16X2X3 1, src_ptr, ref_ptr, src_stride, ref_stride
-        PROCESS_16X2X3 1, src_ptr, ref_ptr, src_stride, ref_stride
-        PROCESS_16X2X3 1, src_ptr, ref_ptr, src_stride, ref_stride
-        PROCESS_16X2X3 1, src_ptr, ref_ptr, src_stride, ref_stride
-        PROCESS_16X2X3 2, src_ptr, ref_ptr, src_stride, ref_stride
-
-        mov             rcx,        result_ptr
-
-        movq            xmm0,       xmm5
-        psrldq          xmm5,       8
-
-        paddw           xmm0,       xmm5
-        movd            [rcx],      xmm0
-;-
-        movq            xmm0,       xmm6
-        psrldq          xmm6,       8
-
-        paddw           xmm0,       xmm6
-        movd            [rcx+4],    xmm0
-;-
-        movq            xmm0,       xmm7
-        psrldq          xmm7,       8
-
-        paddw           xmm0,       xmm7
-        movd            [rcx+8],    xmm0
-
-    STACK_FRAME_DESTROY_X3
-
-;void int vpx_sad16x8x3_sse3(
-;    unsigned char *src_ptr,
-;    int  src_stride,
-;    unsigned char *ref_ptr,
-;    int  ref_stride,
-;    int  *results)
-globalsym(vpx_sad16x8x3_sse3)
-sym(vpx_sad16x8x3_sse3):
-
-    STACK_FRAME_CREATE_X3
-
-        PROCESS_16X2X3 0, src_ptr, ref_ptr, src_stride, ref_stride
-        PROCESS_16X2X3 1, src_ptr, ref_ptr, src_stride, ref_stride
-        PROCESS_16X2X3 1, src_ptr, ref_ptr, src_stride, ref_stride
-        PROCESS_16X2X3 2, src_ptr, ref_ptr, src_stride, ref_stride
-
-        mov             rcx,        result_ptr
-
-        movq            xmm0,       xmm5
-        psrldq          xmm5,       8
-
-        paddw           xmm0,       xmm5
-        movd            [rcx],      xmm0
-;-
-        movq            xmm0,       xmm6
-        psrldq          xmm6,       8
-
-        paddw           xmm0,       xmm6
-        movd            [rcx+4],    xmm0
-;-
-        movq            xmm0,       xmm7
-        psrldq          xmm7,       8
-
-        paddw           xmm0,       xmm7
-        movd            [rcx+8],    xmm0
-
-    STACK_FRAME_DESTROY_X3
-
-;void int vpx_sad8x16x3_sse3(
-;    unsigned char *src_ptr,
-;    int  src_stride,
-;    unsigned char *ref_ptr,
-;    int  ref_stride,
-;    int  *results)
-globalsym(vpx_sad8x16x3_sse3)
-sym(vpx_sad8x16x3_sse3):
-
-    STACK_FRAME_CREATE_X3
-
-        PROCESS_8X2X3 0, src_ptr, ref_ptr, src_stride, ref_stride
-        PROCESS_8X2X3 1, src_ptr, ref_ptr, src_stride, ref_stride
-        PROCESS_8X2X3 1, src_ptr, ref_ptr, src_stride, ref_stride
-        PROCESS_8X2X3 1, src_ptr, ref_ptr, src_stride, ref_stride
-        PROCESS_8X2X3 1, src_ptr, ref_ptr, src_stride, ref_stride
-        PROCESS_8X2X3 1, src_ptr, ref_ptr, src_stride, ref_stride
-        PROCESS_8X2X3 1, src_ptr, ref_ptr, src_stride, ref_stride
-        PROCESS_8X2X3 2, src_ptr, ref_ptr, src_stride, ref_stride
-
-        mov             rcx,        result_ptr
-
-        punpckldq       mm5,        mm6
-
-        movq            [rcx],      mm5
-        movd            [rcx+8],    mm7
-
-    STACK_FRAME_DESTROY_X3
-
-;void int vpx_sad8x8x3_sse3(
-;    unsigned char *src_ptr,
-;    int  src_stride,
-;    unsigned char *ref_ptr,
-;    int  ref_stride,
-;    int  *results)
-globalsym(vpx_sad8x8x3_sse3)
-sym(vpx_sad8x8x3_sse3):
-
-    STACK_FRAME_CREATE_X3
-
-        PROCESS_8X2X3 0, src_ptr, ref_ptr, src_stride, ref_stride
-        PROCESS_8X2X3 1, src_ptr, ref_ptr, src_stride, ref_stride
-        PROCESS_8X2X3 1, src_ptr, ref_ptr, src_stride, ref_stride
-        PROCESS_8X2X3 2, src_ptr, ref_ptr, src_stride, ref_stride
-
-        mov             rcx,        result_ptr
-
-        punpckldq       mm5,        mm6
-
-        movq            [rcx],      mm5
-        movd            [rcx+8],    mm7
-
-    STACK_FRAME_DESTROY_X3
-
-;void int vpx_sad4x4x3_sse3(
-;    unsigned char *src_ptr,
-;    int  src_stride,
-;    unsigned char *ref_ptr,
-;    int  ref_stride,
-;    int  *results)
-globalsym(vpx_sad4x4x3_sse3)
-sym(vpx_sad4x4x3_sse3):
-
-    STACK_FRAME_CREATE_X3
-
-        movd            mm0,        DWORD PTR [src_ptr]
-        movd            mm1,        DWORD PTR [ref_ptr]
-
-        movd            mm2,        DWORD PTR [src_ptr+src_stride]
-        movd            mm3,        DWORD PTR [ref_ptr+ref_stride]
-
-        punpcklbw       mm0,        mm2
-        punpcklbw       mm1,        mm3
-
-        movd            mm4,        DWORD PTR [ref_ptr+1]
-        movd            mm5,        DWORD PTR [ref_ptr+2]
-
-        movd            mm2,        DWORD PTR [ref_ptr+ref_stride+1]
-        movd            mm3,        DWORD PTR [ref_ptr+ref_stride+2]
-
-        psadbw          mm1,        mm0
-
-        punpcklbw       mm4,        mm2
-        punpcklbw       mm5,        mm3
-
-        psadbw          mm4,        mm0
-        psadbw          mm5,        mm0
-
-        lea             src_ptr,    [src_ptr+src_stride*2]
-        lea             ref_ptr,    [ref_ptr+ref_stride*2]
-
-        movd            mm0,        DWORD PTR [src_ptr]
-        movd            mm2,        DWORD PTR [ref_ptr]
-
-        movd            mm3,        DWORD PTR [src_ptr+src_stride]
-        movd            mm6,        DWORD PTR [ref_ptr+ref_stride]
-
-        punpcklbw       mm0,        mm3
-        punpcklbw       mm2,        mm6
-
-        movd            mm3,        DWORD PTR [ref_ptr+1]
-        movd            mm7,        DWORD PTR [ref_ptr+2]
-
-        psadbw          mm2,        mm0
-
-        paddw           mm1,        mm2
-
-        movd            mm2,        DWORD PTR [ref_ptr+ref_stride+1]
-        movd            mm6,        DWORD PTR [ref_ptr+ref_stride+2]
-
-        punpcklbw       mm3,        mm2
-        punpcklbw       mm7,        mm6
-
-        psadbw          mm3,        mm0
-        psadbw          mm7,        mm0
-
-        paddw           mm3,        mm4
-        paddw           mm7,        mm5
-
-        mov             rcx,        result_ptr
-
-        punpckldq       mm1,        mm3
-
-        movq            [rcx],      mm1
-        movd            [rcx+8],    mm7
-
-    STACK_FRAME_DESTROY_X3
diff --git a/vpx_dsp/x86/sad_sse4.asm b/vpx_dsp/x86/sad_sse4.asm
deleted file mode 100644
index 0818ed5f0..000000000
--- a/vpx_dsp/x86/sad_sse4.asm
+++ /dev/null
@@ -1,361 +0,0 @@
-;
-;  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.
-;
-
-
-%include "vpx_ports/x86_abi_support.asm"
-
-%macro PROCESS_16X2X8 1
-%if %1
-        movdqa          xmm0,       XMMWORD PTR [rsi]
-        movq            xmm1,       MMWORD PTR [rdi]
-        movq            xmm3,       MMWORD PTR [rdi+8]
-        movq            xmm2,       MMWORD PTR [rdi+16]
-        punpcklqdq      xmm1,       xmm3
-        punpcklqdq      xmm3,       xmm2
-
-        movdqa          xmm2,       xmm1
-        mpsadbw         xmm1,       xmm0,  0x0
-        mpsadbw         xmm2,       xmm0,  0x5
-
-        psrldq          xmm0,       8
-
-        movdqa          xmm4,       xmm3
-        mpsadbw         xmm3,       xmm0,  0x0
-        mpsadbw         xmm4,       xmm0,  0x5
-
-        paddw           xmm1,       xmm2
-        paddw           xmm1,       xmm3
-        paddw           xmm1,       xmm4
-%else
-        movdqa          xmm0,       XMMWORD PTR [rsi]
-        movq            xmm5,       MMWORD PTR [rdi]
-        movq            xmm3,       MMWORD PTR [rdi+8]
-        movq            xmm2,       MMWORD PTR [rdi+16]
-        punpcklqdq      xmm5,       xmm3
-        punpcklqdq      xmm3,       xmm2
-
-        movdqa          xmm2,       xmm5
-        mpsadbw         xmm5,       xmm0,  0x0
-        mpsadbw         xmm2,       xmm0,  0x5
-
-        psrldq          xmm0,       8
-
-        movdqa          xmm4,       xmm3
-        mpsadbw         xmm3,       xmm0,  0x0
-        mpsadbw         xmm4,       xmm0,  0x5
-
-        paddw           xmm5,       xmm2
-        paddw           xmm5,       xmm3
-        paddw           xmm5,       xmm4
-
-        paddw           xmm1,       xmm5
-%endif
-        movdqa          xmm0,       XMMWORD PTR [rsi + rax]
-        movq            xmm5,       MMWORD PTR [rdi+ rdx]
-        movq            xmm3,       MMWORD PTR [rdi+ rdx+8]
-        movq            xmm2,       MMWORD PTR [rdi+ rdx+16]
-        punpcklqdq      xmm5,       xmm3
-        punpcklqdq      xmm3,       xmm2
-
-        lea             rsi,        [rsi+rax*2]
-        lea             rdi,        [rdi+rdx*2]
-
-        movdqa          xmm2,       xmm5
-        mpsadbw         xmm5,       xmm0,  0x0
-        mpsadbw         xmm2,       xmm0,  0x5
-
-        psrldq          xmm0,       8
-        movdqa          xmm4,       xmm3
-        mpsadbw         xmm3,       xmm0,  0x0
-        mpsadbw         xmm4,       xmm0,  0x5
-
-        paddw           xmm5,       xmm2
-        paddw           xmm5,       xmm3
-        paddw           xmm5,       xmm4
-
-        paddw           xmm1,       xmm5
-%endmacro
-
-%macro PROCESS_8X2X8 1
-%if %1
-        movq            xmm0,       MMWORD PTR [rsi]
-        movq            xmm1,       MMWORD PTR [rdi]
-        movq            xmm3,       MMWORD PTR [rdi+8]
-        punpcklqdq      xmm1,       xmm3
-
-        movdqa          xmm2,       xmm1
-        mpsadbw         xmm1,       xmm0,  0x0
-        mpsadbw         xmm2,       xmm0,  0x5
-        paddw           xmm1,       xmm2
-%else
-        movq            xmm0,       MMWORD PTR [rsi]
-        movq            xmm5,       MMWORD PTR [rdi]
-        movq            xmm3,       MMWORD PTR [rdi+8]
-        punpcklqdq      xmm5,       xmm3
-
-        movdqa          xmm2,       xmm5
-        mpsadbw         xmm5,       xmm0,  0x0
-        mpsadbw         xmm2,       xmm0,  0x5
-        paddw           xmm5,       xmm2
-
-        paddw           xmm1,       xmm5
-%endif
-        movq            xmm0,       MMWORD PTR [rsi + rax]
-        movq            xmm5,       MMWORD PTR [rdi+ rdx]
-        movq            xmm3,       MMWORD PTR [rdi+ rdx+8]
-        punpcklqdq      xmm5,       xmm3
-
-        lea             rsi,        [rsi+rax*2]
-        lea             rdi,        [rdi+rdx*2]
-
-        movdqa          xmm2,       xmm5
-        mpsadbw         xmm5,       xmm0,  0x0
-        mpsadbw         xmm2,       xmm0,  0x5
-        paddw           xmm5,       xmm2
-
-        paddw           xmm1,       xmm5
-%endmacro
-
-%macro PROCESS_4X2X8 1
-%if %1
-        movd            xmm0,       [rsi]
-        movq            xmm1,       MMWORD PTR [rdi]
-        movq            xmm3,       MMWORD PTR [rdi+8]
-        punpcklqdq      xmm1,       xmm3
-
-        mpsadbw         xmm1,       xmm0,  0x0
-%else
-        movd            xmm0,       [rsi]
-        movq            xmm5,       MMWORD PTR [rdi]
-        movq            xmm3,       MMWORD PTR [rdi+8]
-        punpcklqdq      xmm5,       xmm3
-
-        mpsadbw         xmm5,       xmm0,  0x0
-
-        paddw           xmm1,       xmm5
-%endif
-        movd            xmm0,       [rsi + rax]
-        movq            xmm5,       MMWORD PTR [rdi+ rdx]
-        movq            xmm3,       MMWORD PTR [rdi+ rdx+8]
-        punpcklqdq      xmm5,       xmm3
-
-        lea             rsi,        [rsi+rax*2]
-        lea             rdi,        [rdi+rdx*2]
-
-        mpsadbw         xmm5,       xmm0,  0x0
-
-        paddw           xmm1,       xmm5
-%endmacro
-
-%macro WRITE_AS_INTS 0
-    mov             rdi,        arg(4)           ;Results
-    pxor            xmm0, xmm0
-    movdqa          xmm2, xmm1
-    punpcklwd       xmm1, xmm0
-    punpckhwd       xmm2, xmm0
-
-    movdqa          [rdi],    xmm1
-    movdqa          [rdi + 16],    xmm2
-%endmacro
-
-SECTION .text
-
-;void vpx_sad16x16x8_sse4_1(
-;    const unsigned char *src_ptr,
-;    int  src_stride,
-;    const unsigned char *ref_ptr,
-;    int  ref_stride,
-;    unsigned short *sad_array);
-globalsym(vpx_sad16x16x8_sse4_1)
-sym(vpx_sad16x16x8_sse4_1):
-    push        rbp
-    mov         rbp, rsp
-    SHADOW_ARGS_TO_STACK 5
-    push        rsi
-    push        rdi
-    ; end prolog
-
-    mov             rsi,        arg(0)           ;src_ptr
-    mov             rdi,        arg(2)           ;ref_ptr
-
-    movsxd          rax,        dword ptr arg(1) ;src_stride
-    movsxd          rdx,        dword ptr arg(3) ;ref_stride
-
-    PROCESS_16X2X8 1
-    PROCESS_16X2X8 0
-    PROCESS_16X2X8 0
-    PROCESS_16X2X8 0
-    PROCESS_16X2X8 0
-    PROCESS_16X2X8 0
-    PROCESS_16X2X8 0
-    PROCESS_16X2X8 0
-
-    WRITE_AS_INTS
-
-    ; begin epilog
-    pop         rdi
-    pop         rsi
-    UNSHADOW_ARGS
-    pop         rbp
-    ret
-
-
-;void vpx_sad16x8x8_sse4_1(
-;    const unsigned char *src_ptr,
-;    int  src_stride,
-;    const unsigned char *ref_ptr,
-;    int  ref_stride,
-;    unsigned short *sad_array
-;);
-globalsym(vpx_sad16x8x8_sse4_1)
-sym(vpx_sad16x8x8_sse4_1):
-    push        rbp
-    mov         rbp, rsp
-    SHADOW_ARGS_TO_STACK 5
-    push        rsi
-    push        rdi
-    ; end prolog
-
-    mov             rsi,        arg(0)           ;src_ptr
-    mov             rdi,        arg(2)           ;ref_ptr
-
-    movsxd          rax,        dword ptr arg(1) ;src_stride
-    movsxd          rdx,        dword ptr arg(3) ;ref_stride
-
-    PROCESS_16X2X8 1
-    PROCESS_16X2X8 0
-    PROCESS_16X2X8 0
-    PROCESS_16X2X8 0
-
-    WRITE_AS_INTS
-
-    ; begin epilog
-    pop         rdi
-    pop         rsi
-    UNSHADOW_ARGS
-    pop         rbp
-    ret
-
-
-;void vpx_sad8x8x8_sse4_1(
-;    const unsigned char *src_ptr,
-;    int  src_stride,
-;    const unsigned char *ref_ptr,
-;    int  ref_stride,
-;    unsigned short *sad_array
-;);
-globalsym(vpx_sad8x8x8_sse4_1)
-sym(vpx_sad8x8x8_sse4_1):
-    push        rbp
-    mov         rbp, rsp
-    SHADOW_ARGS_TO_STACK 5
-    push        rsi
-    push        rdi
-    ; end prolog
-
-    mov             rsi,        arg(0)           ;src_ptr
-    mov             rdi,        arg(2)           ;ref_ptr
-
-    movsxd          rax,        dword ptr arg(1) ;src_stride
-    movsxd          rdx,        dword ptr arg(3) ;ref_stride
-
-    PROCESS_8X2X8 1
-    PROCESS_8X2X8 0
-    PROCESS_8X2X8 0
-    PROCESS_8X2X8 0
-
-    WRITE_AS_INTS
-
-    ; begin epilog
-    pop         rdi
-    pop         rsi
-    UNSHADOW_ARGS
-    pop         rbp
-    ret
-
-
-;void vpx_sad8x16x8_sse4_1(
-;    const unsigned char *src_ptr,
-;    int  src_stride,
-;    const unsigned char *ref_ptr,
-;    int  ref_stride,
-;    unsigned short *sad_array
-;);
-globalsym(vpx_sad8x16x8_sse4_1)
-sym(vpx_sad8x16x8_sse4_1):
-    push        rbp
-    mov         rbp, rsp
-    SHADOW_ARGS_TO_STACK 5
-    push        rsi
-    push        rdi
-    ; end prolog
-
-    mov             rsi,        arg(0)           ;src_ptr
-    mov             rdi,        arg(2)           ;ref_ptr
-
-    movsxd          rax,        dword ptr arg(1) ;src_stride
-    movsxd          rdx,        dword ptr arg(3) ;ref_stride
-
-    PROCESS_8X2X8 1
-    PROCESS_8X2X8 0
-    PROCESS_8X2X8 0
-    PROCESS_8X2X8 0
-    PROCESS_8X2X8 0
-    PROCESS_8X2X8 0
-    PROCESS_8X2X8 0
-    PROCESS_8X2X8 0
-
-    WRITE_AS_INTS
-
-    ; begin epilog
-    pop         rdi
-    pop         rsi
-    UNSHADOW_ARGS
-    pop         rbp
-    ret
-
-
-;void vpx_sad4x4x8_sse4_1(
-;    const unsigned char *src_ptr,
-;    int  src_stride,
-;    const unsigned char *ref_ptr,
-;    int  ref_stride,
-;    unsigned short *sad_array
-;);
-globalsym(vpx_sad4x4x8_sse4_1)
-sym(vpx_sad4x4x8_sse4_1):
-    push        rbp
-    mov         rbp, rsp
-    SHADOW_ARGS_TO_STACK 5
-    push        rsi
-    push        rdi
-    ; end prolog
-
-    mov             rsi,        arg(0)           ;src_ptr
-    mov             rdi,        arg(2)           ;ref_ptr
-
-    movsxd          rax,        dword ptr arg(1) ;src_stride
-    movsxd          rdx,        dword ptr arg(3) ;ref_stride
-
-    PROCESS_4X2X8 1
-    PROCESS_4X2X8 0
-
-    WRITE_AS_INTS
-
-    ; begin epilog
-    pop         rdi
-    pop         rsi
-    UNSHADOW_ARGS
-    pop         rbp
-    ret
-
-
-
-
diff --git a/vpx_dsp/x86/sad_ssse3.asm b/vpx_dsp/x86/sad_ssse3.asm
deleted file mode 100644
index a5bc6d730..000000000
--- a/vpx_dsp/x86/sad_ssse3.asm
+++ /dev/null
@@ -1,372 +0,0 @@
-;
-;  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.
-;
-
-
-%include "vpx_ports/x86_abi_support.asm"
-
-%macro PROCESS_16X2X3 1
-%if %1
-        movdqa          xmm0,       XMMWORD PTR [rsi]
-        lddqu           xmm5,       XMMWORD PTR [rdi]
-        lddqu           xmm6,       XMMWORD PTR [rdi+1]
-        lddqu           xmm7,       XMMWORD PTR [rdi+2]
-
-        psadbw          xmm5,       xmm0
-        psadbw          xmm6,       xmm0
-        psadbw          xmm7,       xmm0
-%else
-        movdqa          xmm0,       XMMWORD PTR [rsi]
-        lddqu           xmm1,       XMMWORD PTR [rdi]
-        lddqu           xmm2,       XMMWORD PTR [rdi+1]
-        lddqu           xmm3,       XMMWORD PTR [rdi+2]
-
-        psadbw          xmm1,       xmm0
-        psadbw          xmm2,       xmm0
-        psadbw          xmm3,       xmm0
-
-        paddw           xmm5,       xmm1
-        paddw           xmm6,       xmm2
-        paddw           xmm7,       xmm3
-%endif
-        movdqa          xmm0,       XMMWORD PTR [rsi+rax]
-        lddqu           xmm1,       XMMWORD PTR [rdi+rdx]
-        lddqu           xmm2,       XMMWORD PTR [rdi+rdx+1]
-        lddqu           xmm3,       XMMWORD PTR [rdi+rdx+2]
-
-        lea             rsi,        [rsi+rax*2]
-        lea             rdi,        [rdi+rdx*2]
-
-        psadbw          xmm1,       xmm0
-        psadbw          xmm2,       xmm0
-        psadbw          xmm3,       xmm0
-
-        paddw           xmm5,       xmm1
-        paddw           xmm6,       xmm2
-        paddw           xmm7,       xmm3
-%endmacro
-
-%macro PROCESS_16X2X3_OFFSET 2
-%if %1
-        movdqa          xmm0,       XMMWORD PTR [rsi]
-        movdqa          xmm4,       XMMWORD PTR [rdi]
-        movdqa          xmm7,       XMMWORD PTR [rdi+16]
-
-        movdqa          xmm5,       xmm7
-        palignr         xmm5,       xmm4,       %2
-
-        movdqa          xmm6,       xmm7
-        palignr         xmm6,       xmm4,       (%2+1)
-
-        palignr         xmm7,       xmm4,       (%2+2)
-
-        psadbw          xmm5,       xmm0
-        psadbw          xmm6,       xmm0
-        psadbw          xmm7,       xmm0
-%else
-        movdqa          xmm0,       XMMWORD PTR [rsi]
-        movdqa          xmm4,       XMMWORD PTR [rdi]
-        movdqa          xmm3,       XMMWORD PTR [rdi+16]
-
-        movdqa          xmm1,       xmm3
-        palignr         xmm1,       xmm4,       %2
-
-        movdqa          xmm2,       xmm3
-        palignr         xmm2,       xmm4,       (%2+1)
-
-        palignr         xmm3,       xmm4,       (%2+2)
-
-        psadbw          xmm1,       xmm0
-        psadbw          xmm2,       xmm0
-        psadbw          xmm3,       xmm0
-
-        paddw           xmm5,       xmm1
-        paddw           xmm6,       xmm2
-        paddw           xmm7,       xmm3
-%endif
-        movdqa          xmm0,       XMMWORD PTR [rsi+rax]
-        movdqa          xmm4,       XMMWORD PTR [rdi+rdx]
-        movdqa          xmm3,       XMMWORD PTR [rdi+rdx+16]
-
-        movdqa          xmm1,       xmm3
-        palignr         xmm1,       xmm4,       %2
-
-        movdqa          xmm2,       xmm3
-        palignr         xmm2,       xmm4,       (%2+1)
-
-        palignr         xmm3,       xmm4,       (%2+2)
-
-        lea             rsi,        [rsi+rax*2]
-        lea             rdi,        [rdi+rdx*2]
-
-        psadbw          xmm1,       xmm0
-        psadbw          xmm2,       xmm0
-        psadbw          xmm3,       xmm0
-
-        paddw           xmm5,       xmm1
-        paddw           xmm6,       xmm2
-        paddw           xmm7,       xmm3
-%endmacro
-
-%macro PROCESS_16X16X3_OFFSET 2
-%2_aligned_by_%1:
-
-        sub             rdi,        %1
-
-        PROCESS_16X2X3_OFFSET 1, %1
-        PROCESS_16X2X3_OFFSET 0, %1
-        PROCESS_16X2X3_OFFSET 0, %1
-        PROCESS_16X2X3_OFFSET 0, %1
-        PROCESS_16X2X3_OFFSET 0, %1
-        PROCESS_16X2X3_OFFSET 0, %1
-        PROCESS_16X2X3_OFFSET 0, %1
-        PROCESS_16X2X3_OFFSET 0, %1
-
-        jmp             %2_store_off
-
-%endmacro
-
-%macro PROCESS_16X8X3_OFFSET 2
-%2_aligned_by_%1:
-
-        sub             rdi,        %1
-
-        PROCESS_16X2X3_OFFSET 1, %1
-        PROCESS_16X2X3_OFFSET 0, %1
-        PROCESS_16X2X3_OFFSET 0, %1
-        PROCESS_16X2X3_OFFSET 0, %1
-
-        jmp             %2_store_off
-
-%endmacro
-
-SECTION .text
-
-;void int vpx_sad16x16x3_ssse3(
-;    unsigned char *src_ptr,
-;    int  src_stride,
-;    unsigned char *ref_ptr,
-;    int  ref_stride,
-;    int  *results)
-globalsym(vpx_sad16x16x3_ssse3)
-sym(vpx_sad16x16x3_ssse3):
-    push        rbp
-    mov         rbp, rsp
-    SHADOW_ARGS_TO_STACK 5
-    SAVE_XMM 7
-    push        rsi
-    push        rdi
-    push        rcx
-    ; end prolog
-
-        mov             rsi,        arg(0) ;src_ptr
-        mov             rdi,        arg(2) ;ref_ptr
-
-        mov             rdx,        0xf
-        and             rdx,        rdi
-
-        jmp .vpx_sad16x16x3_ssse3_skiptable
-.vpx_sad16x16x3_ssse3_jumptable:
-        dd .vpx_sad16x16x3_ssse3_aligned_by_0  - .vpx_sad16x16x3_ssse3_do_jump
-        dd .vpx_sad16x16x3_ssse3_aligned_by_1  - .vpx_sad16x16x3_ssse3_do_jump
-        dd .vpx_sad16x16x3_ssse3_aligned_by_2  - .vpx_sad16x16x3_ssse3_do_jump
-        dd .vpx_sad16x16x3_ssse3_aligned_by_3  - .vpx_sad16x16x3_ssse3_do_jump
-        dd .vpx_sad16x16x3_ssse3_aligned_by_4  - .vpx_sad16x16x3_ssse3_do_jump
-        dd .vpx_sad16x16x3_ssse3_aligned_by_5  - .vpx_sad16x16x3_ssse3_do_jump
-        dd .vpx_sad16x16x3_ssse3_aligned_by_6  - .vpx_sad16x16x3_ssse3_do_jump
-        dd .vpx_sad16x16x3_ssse3_aligned_by_7  - .vpx_sad16x16x3_ssse3_do_jump
-        dd .vpx_sad16x16x3_ssse3_aligned_by_8  - .vpx_sad16x16x3_ssse3_do_jump
-        dd .vpx_sad16x16x3_ssse3_aligned_by_9  - .vpx_sad16x16x3_ssse3_do_jump
-        dd .vpx_sad16x16x3_ssse3_aligned_by_10 - .vpx_sad16x16x3_ssse3_do_jump
-        dd .vpx_sad16x16x3_ssse3_aligned_by_11 - .vpx_sad16x16x3_ssse3_do_jump
-        dd .vpx_sad16x16x3_ssse3_aligned_by_12 - .vpx_sad16x16x3_ssse3_do_jump
-        dd .vpx_sad16x16x3_ssse3_aligned_by_13 - .vpx_sad16x16x3_ssse3_do_jump
-        dd .vpx_sad16x16x3_ssse3_aligned_by_14 - .vpx_sad16x16x3_ssse3_do_jump
-        dd .vpx_sad16x16x3_ssse3_aligned_by_15 - .vpx_sad16x16x3_ssse3_do_jump
-.vpx_sad16x16x3_ssse3_skiptable:
-
-        call .vpx_sad16x16x3_ssse3_do_jump
-.vpx_sad16x16x3_ssse3_do_jump:
-        pop             rcx                         ; get the address of do_jump
-        mov             rax,  .vpx_sad16x16x3_ssse3_jumptable - .vpx_sad16x16x3_ssse3_do_jump
-        add             rax,  rcx  ; get the absolute address of vpx_sad16x16x3_ssse3_jumptable
-
-        movsxd          rax,  dword [rax + 4*rdx]   ; get the 32 bit offset from the jumptable
-        add             rcx,        rax
-
-        movsxd          rax,        dword ptr arg(1) ;src_stride
-        movsxd          rdx,        dword ptr arg(3) ;ref_stride
-
-        jmp             rcx
-
-        PROCESS_16X16X3_OFFSET 0,  .vpx_sad16x16x3_ssse3
-        PROCESS_16X16X3_OFFSET 1,  .vpx_sad16x16x3_ssse3
-        PROCESS_16X16X3_OFFSET 2,  .vpx_sad16x16x3_ssse3
-        PROCESS_16X16X3_OFFSET 3,  .vpx_sad16x16x3_ssse3
-        PROCESS_16X16X3_OFFSET 4,  .vpx_sad16x16x3_ssse3
-        PROCESS_16X16X3_OFFSET 5,  .vpx_sad16x16x3_ssse3
-        PROCESS_16X16X3_OFFSET 6,  .vpx_sad16x16x3_ssse3
-        PROCESS_16X16X3_OFFSET 7,  .vpx_sad16x16x3_ssse3
-        PROCESS_16X16X3_OFFSET 8,  .vpx_sad16x16x3_ssse3
-        PROCESS_16X16X3_OFFSET 9,  .vpx_sad16x16x3_ssse3
-        PROCESS_16X16X3_OFFSET 10, .vpx_sad16x16x3_ssse3
-        PROCESS_16X16X3_OFFSET 11, .vpx_sad16x16x3_ssse3
-        PROCESS_16X16X3_OFFSET 12, .vpx_sad16x16x3_ssse3
-        PROCESS_16X16X3_OFFSET 13, .vpx_sad16x16x3_ssse3
-        PROCESS_16X16X3_OFFSET 14, .vpx_sad16x16x3_ssse3
-
-.vpx_sad16x16x3_ssse3_aligned_by_15:
-        PROCESS_16X2X3 1
-        PROCESS_16X2X3 0
-        PROCESS_16X2X3 0
-        PROCESS_16X2X3 0
-        PROCESS_16X2X3 0
-        PROCESS_16X2X3 0
-        PROCESS_16X2X3 0
-        PROCESS_16X2X3 0
-
-.vpx_sad16x16x3_ssse3_store_off:
-        mov             rdi,        arg(4) ;Results
-
-        movq            xmm0,       xmm5
-        psrldq          xmm5,       8
-
-        paddw           xmm0,       xmm5
-        movd            [rdi],      xmm0
-;-
-        movq            xmm0,       xmm6
-        psrldq          xmm6,       8
-
-        paddw           xmm0,       xmm6
-        movd            [rdi+4],    xmm0
-;-
-        movq            xmm0,       xmm7
-        psrldq          xmm7,       8
-
-        paddw           xmm0,       xmm7
-        movd            [rdi+8],    xmm0
-
-    ; begin epilog
-    pop         rcx
-    pop         rdi
-    pop         rsi
-    RESTORE_XMM
-    UNSHADOW_ARGS
-    pop         rbp
-    ret
-
-;void int vpx_sad16x8x3_ssse3(
-;    unsigned char *src_ptr,
-;    int  src_stride,
-;    unsigned char *ref_ptr,
-;    int  ref_stride,
-;    int  *results)
-globalsym(vpx_sad16x8x3_ssse3)
-sym(vpx_sad16x8x3_ssse3):
-    push        rbp
-    mov         rbp, rsp
-    SHADOW_ARGS_TO_STACK 5
-    SAVE_XMM 7
-    push        rsi
-    push        rdi
-    push        rcx
-    ; end prolog
-
-        mov             rsi,        arg(0) ;src_ptr
-        mov             rdi,        arg(2) ;ref_ptr
-
-        mov             rdx,        0xf
-        and             rdx,        rdi
-
-        jmp .vpx_sad16x8x3_ssse3_skiptable
-.vpx_sad16x8x3_ssse3_jumptable:
-        dd .vpx_sad16x8x3_ssse3_aligned_by_0  - .vpx_sad16x8x3_ssse3_do_jump
-        dd .vpx_sad16x8x3_ssse3_aligned_by_1  - .vpx_sad16x8x3_ssse3_do_jump
-        dd .vpx_sad16x8x3_ssse3_aligned_by_2  - .vpx_sad16x8x3_ssse3_do_jump
-        dd .vpx_sad16x8x3_ssse3_aligned_by_3  - .vpx_sad16x8x3_ssse3_do_jump
-        dd .vpx_sad16x8x3_ssse3_aligned_by_4  - .vpx_sad16x8x3_ssse3_do_jump
-        dd .vpx_sad16x8x3_ssse3_aligned_by_5  - .vpx_sad16x8x3_ssse3_do_jump
-        dd .vpx_sad16x8x3_ssse3_aligned_by_6  - .vpx_sad16x8x3_ssse3_do_jump
-        dd .vpx_sad16x8x3_ssse3_aligned_by_7  - .vpx_sad16x8x3_ssse3_do_jump
-        dd .vpx_sad16x8x3_ssse3_aligned_by_8  - .vpx_sad16x8x3_ssse3_do_jump
-        dd .vpx_sad16x8x3_ssse3_aligned_by_9  - .vpx_sad16x8x3_ssse3_do_jump
-        dd .vpx_sad16x8x3_ssse3_aligned_by_10 - .vpx_sad16x8x3_ssse3_do_jump
-        dd .vpx_sad16x8x3_ssse3_aligned_by_11 - .vpx_sad16x8x3_ssse3_do_jump
-        dd .vpx_sad16x8x3_ssse3_aligned_by_12 - .vpx_sad16x8x3_ssse3_do_jump
-        dd .vpx_sad16x8x3_ssse3_aligned_by_13 - .vpx_sad16x8x3_ssse3_do_jump
-        dd .vpx_sad16x8x3_ssse3_aligned_by_14 - .vpx_sad16x8x3_ssse3_do_jump
-        dd .vpx_sad16x8x3_ssse3_aligned_by_15 - .vpx_sad16x8x3_ssse3_do_jump
-.vpx_sad16x8x3_ssse3_skiptable:
-
-        call .vpx_sad16x8x3_ssse3_do_jump
-.vpx_sad16x8x3_ssse3_do_jump:
-        pop             rcx                         ; get the address of do_jump
-        mov             rax,  .vpx_sad16x8x3_ssse3_jumptable - .vpx_sad16x8x3_ssse3_do_jump
-        add             rax,  rcx  ; get the absolute address of vpx_sad16x8x3_ssse3_jumptable
-
-        movsxd          rax,  dword [rax + 4*rdx]   ; get the 32 bit offset from the jumptable
-        add             rcx,        rax
-
-        movsxd          rax,        dword ptr arg(1) ;src_stride
-        movsxd          rdx,        dword ptr arg(3) ;ref_stride
-
-        jmp             rcx
-
-        PROCESS_16X8X3_OFFSET 0,  .vpx_sad16x8x3_ssse3
-        PROCESS_16X8X3_OFFSET 1,  .vpx_sad16x8x3_ssse3
-        PROCESS_16X8X3_OFFSET 2,  .vpx_sad16x8x3_ssse3
-        PROCESS_16X8X3_OFFSET 3,  .vpx_sad16x8x3_ssse3
-        PROCESS_16X8X3_OFFSET 4,  .vpx_sad16x8x3_ssse3
-        PROCESS_16X8X3_OFFSET 5,  .vpx_sad16x8x3_ssse3
-        PROCESS_16X8X3_OFFSET 6,  .vpx_sad16x8x3_ssse3
-        PROCESS_16X8X3_OFFSET 7,  .vpx_sad16x8x3_ssse3
-        PROCESS_16X8X3_OFFSET 8,  .vpx_sad16x8x3_ssse3
-        PROCESS_16X8X3_OFFSET 9,  .vpx_sad16x8x3_ssse3
-        PROCESS_16X8X3_OFFSET 10, .vpx_sad16x8x3_ssse3
-        PROCESS_16X8X3_OFFSET 11, .vpx_sad16x8x3_ssse3
-        PROCESS_16X8X3_OFFSET 12, .vpx_sad16x8x3_ssse3
-        PROCESS_16X8X3_OFFSET 13, .vpx_sad16x8x3_ssse3
-        PROCESS_16X8X3_OFFSET 14, .vpx_sad16x8x3_ssse3
-
-.vpx_sad16x8x3_ssse3_aligned_by_15:
-
-        PROCESS_16X2X3 1
-        PROCESS_16X2X3 0
-        PROCESS_16X2X3 0
-        PROCESS_16X2X3 0
-
-.vpx_sad16x8x3_ssse3_store_off:
-        mov             rdi,        arg(4) ;Results
-
-        movq            xmm0,       xmm5
-        psrldq          xmm5,       8
-
-        paddw           xmm0,       xmm5
-        movd            [rdi],      xmm0
-;-
-        movq            xmm0,       xmm6
-        psrldq          xmm6,       8
-
-        paddw           xmm0,       xmm6
-        movd            [rdi+4],    xmm0
-;-
-        movq            xmm0,       xmm7
-        psrldq          xmm7,       8
-
-        paddw           xmm0,       xmm7
-        movd            [rdi+8],    xmm0
-
-    ; begin epilog
-    pop         rcx
-    pop         rdi
-    pop         rsi
-    RESTORE_XMM
-    UNSHADOW_ARGS
-    pop         rbp
-    ret