From adb22e7ae3bd65e96c4f4bcf7f011532979ab9a7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Matthias=20R=C3=A4ncker?= <theonetruecamper@gmx.de>
Date: Sat, 2 Feb 2013 22:13:14 +0100
Subject: [PATCH] add x32 abi support
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit


Signed-off-by: Matthias Räncker <theonetruecamper@gmx.de>
---
 configure                          | 12 +++++-
 libavcodec/x86/dsputil_yasm.asm    | 40 +++++++++----------
 libavcodec/x86/fft_mmx.asm         | 30 +++++---------
 libavcodec/x86/fmtconvert.asm      | 32 +++++++--------
 libavcodec/x86/h264_idct.asm       | 18 ++++-----
 libavcodec/x86/h264_idct_10bit.asm |  6 +--
 libavcodec/x86/mlpdsp.c            |  4 +-
 libavutil/x86/cpu.c                |  4 +-
 libavutil/x86/x86inc.asm           | 34 +++++++++++++++-
 libavutil/x86_cpu.h                | 20 +++++++++-
 libswscale/x86/output.asm          |  4 +-
 libswscale/x86/swscale_template.c  | 80 +++++++++++++++++++-------------------
 12 files changed, 164 insertions(+), 120 deletions(-)

diff --git a/configure b/configure
index 2d17ce0..21c5645 100755
--- a/configure
+++ b/configure
@@ -1023,6 +1023,8 @@ ARCH_LIST='
     x86
     x86_32
     x86_64
+    x86_64_x32
+    x86_64_x64
 '
 
 ARCH_EXT_LIST='
@@ -2398,9 +2400,14 @@ case "$arch" in
     x86)
         subarch="x86_32"
         check_cc <<EOF && subarch="x86_64"
-        int test[(int)sizeof(char*) - 7];
+        #ifndef __x86_64__
+        int test[-1];
+        #endif
 EOF
         if test "$subarch" = "x86_64"; then
+            check_cc <<EOF && subarch="x86_64 x86_64_x64" || subarch="x86_64 x86_64_x32"
+            int test[(int)sizeof(char*) - 7];
+EOF
             spic=$shared
         fi
     ;;
@@ -2770,7 +2777,8 @@ EOF
 
     if ! disabled_any asm mmx yasm; then
         if check_cmd $yasmexe --version; then
-            enabled x86_64 && yasm_extra="-m amd64"
+            enabled x86_64_x64 && yasm_extra="-m amd64"
+            enabled x86_64_x32 && yasm_extra="-m x32"
             yasm_debug="-g dwarf2"
         elif check_cmd nasm -v; then
             yasmexe=nasm
diff --git a/libavcodec/x86/dsputil_yasm.asm b/libavcodec/x86/dsputil_yasm.asm
index 4d2fb6a..57c8f45 100644
--- a/libavcodec/x86/dsputil_yasm.asm
+++ b/libavcodec/x86/dsputil_yasm.asm
@@ -505,13 +505,13 @@ cglobal emu_edge_core, 6, 7, 1
 %else
 %define w_reg r6
 cglobal emu_edge_core, 2, 7, 0
-    mov         r4, r4m         ; end_y
-    mov         r5, r5m         ; block_h
+    mov        r4p, r4m         ; end_y
+    mov        r5p, r5m         ; block_h
 %endif
 
     ; start with vertical extend (top/bottom) and body pixel copy
-    mov      w_reg, r7m
-    sub      w_reg, r6m         ; w = start_x - end_x
+    mov preg(w_reg), r7m
+    sub preg(w_reg), r6m        ; w = start_x - end_x
     sub         r5, r4
 %ifdef ARCH_X86_64
     sub         r4, r3
@@ -534,7 +534,7 @@ cglobal emu_edge_core, 2, 7, 0
 .v_extend_end:
 
     ; horizontal extend (left/right)
-    mov      w_reg, r6m         ; start_x
+    mov preg(w_reg), r6m        ; start_x
     sub         r0, w_reg
 %ifdef ARCH_X86_64
     mov         r3, r0          ; backup of buf+block_h*linesize
@@ -568,8 +568,8 @@ cglobal emu_edge_core, 2, 7, 0
     mov         r0, r0m
     mov         r5, r5m
 %endif
-    mov      w_reg, r7m         ; end_x
-    mov         r1, r8m         ; block_w
+    mov preg(w_reg), r7m        ; end_x
+    mov        r1p, r8m         ; block_w
     mov         r4, r1
     sub         r1, w_reg
     jz .h_extend_end            ; if (end_x == block_w) goto h_extend_end
@@ -750,7 +750,7 @@ ALIGN 128
     READ_NUM_BYTES  top,    %%n              ; read bytes
 .emuedge_extend_top_ %+ %%n %+ _loop:        ; do {
     WRITE_NUM_BYTES top,    %%n              ;   write bytes
-    add            r0 , r2                   ;   dst += linesize
+    add           r0p , r2p                  ;   dst += linesize
 %ifdef ARCH_X86_64
     dec            r3d
 %else ; ARCH_X86_32
@@ -762,19 +762,19 @@ ALIGN 128
 .emuedge_copy_body_ %+ %%n %+ _loop:         ; do {
     READ_NUM_BYTES  body,   %%n              ;   read bytes
     WRITE_NUM_BYTES body,   %%n              ;   write bytes
-    add            r0 , r2                   ;   dst += linesize
-    add            r1 , r2                   ;   src += linesize
+    add           r0p , r2p                  ;   dst += linesize
+    add           r1p , r2p                  ;   src += linesize
     dec            r4d
     jnz .emuedge_copy_body_ %+ %%n %+ _loop  ; } while (--end_y)
 
     ; copy bottom pixels
     test           r5 , r5                   ; if (!block_h)
     jz .emuedge_v_extend_end_ %+ %%n         ;   goto end
-    sub            r1 , r2                   ; src -= linesize
+    sub           r1p , r2p                  ; src -= linesize
     READ_NUM_BYTES  bottom, %%n              ; read bytes
 .emuedge_extend_bottom_ %+ %%n %+ _loop:     ; do {
     WRITE_NUM_BYTES bottom, %%n              ;   write bytes
-    add            r0 , r2                   ;   dst += linesize
+    add           r0p , r2p                  ;   dst += linesize
     dec            r5d
     jnz .emuedge_extend_bottom_ %+ %%n %+ _loop ; } while (--block_h)
 
@@ -836,7 +836,7 @@ ALIGN 128
 %rep 11
 ALIGN 64
 .emuedge_extend_left_ %+ %%n:          ; do {
-    sub         r0, r2                 ;   dst -= linesize
+    sub        r0p, r2p                ;   dst -= linesize
     READ_V_PIXEL  %%n, [r0+r1]         ;   read pixels
     WRITE_V_PIXEL %%n, r0              ;   write pixels
     dec         r5
@@ -857,7 +857,7 @@ ALIGN 64
 ALIGN 64
 .emuedge_extend_right_ %+ %%n:          ; do {
 %ifdef ARCH_X86_64
-    sub        r3, r2                   ;   dst -= linesize
+    sub       r3p, r2p                  ;   dst -= linesize
     READ_V_PIXEL  %%n, [r3+w_reg-1]     ;   read pixels
     WRITE_V_PIXEL %%n, r3+r4-%%n        ;   write pixels
     dec       r11
@@ -907,7 +907,7 @@ ALIGN 64
 
 %macro V_COPY_ROW 2
 %ifidn %1, bottom
-    sub         r1, linesize
+    sub        r1p, linesize
 %endif
 .%1_copy_loop:
     xor    cnt_reg, cnt_reg
@@ -917,7 +917,7 @@ ALIGN 64
 %else ; sse
     V_COPY_NPX %1, xmm0, movups, 16, 0xFFFFFFF0
 %ifdef ARCH_X86_64
-%define linesize r2
+%define linesize r2p
     V_COPY_NPX %1, rax , mov,     8
 %else ; ARCH_X86_32
 %define linesize r2m
@@ -929,9 +929,9 @@ ALIGN 64
     V_COPY_NPX %1, vall, mov,     1
     mov      w_reg, cnt_reg
 %ifidn %1, body
-    add         r1, linesize
+    add        r1p, linesize
 %endif
-    add         r0, linesize
+    add        r0p, linesize
     dec         %2
     jnz .%1_copy_loop
 %endmacro
@@ -978,7 +978,7 @@ ALIGN 64
 .slow_left_extend_loop:
 ; r0=buf+block_h*linesize,r2=linesize,r6(64)/r3(32)=val,r5=block_h,r4=cntr,r10/r6=start_x
     mov         r4, 8
-    sub         r0, linesize
+    sub        r0p, linesize
     READ_V_PIXEL 8, [r0+w_reg]
 .left_extend_8px_loop:
     movq [r0+r4-8], mm0
@@ -1014,7 +1014,7 @@ ALIGN 64
 %define bh_reg r5
 %endif
     lea         r1, [r4-8]
-    sub    buf_reg, linesize
+    sub preg(buf_reg), linesize
     READ_V_PIXEL 8, [buf_reg+w_reg-1]
 .right_extend_8px_loop:
     movq [buf_reg+r1], mm0
diff --git a/libavcodec/x86/fft_mmx.asm b/libavcodec/x86/fft_mmx.asm
index d6d07b8..6b9b9c2 100644
--- a/libavcodec/x86/fft_mmx.asm
+++ b/libavcodec/x86/fft_mmx.asm
@@ -30,21 +30,15 @@
 
 %include "x86inc.asm"
 
-%ifdef ARCH_X86_64
-%define pointer resq
-%else
-%define pointer resd
-%endif
-
 struc FFTContext
     .nbits:    resd 1
     .reverse:  resd 1
-    .revtab:   pointer 1
-    .tmpbuf:   pointer 1
+    .revtab:   resp 1
+    .tmpbuf:   resp 1
     .mdctsize: resd 1
     .mdctbits: resd 1
-    .tcos:     pointer 1
-    .tsin:     pointer 1
+    .tcos:     resp 1
+    .tsin:     resp 1
 endstruc
 
 SECTION_RODATA
@@ -73,12 +67,6 @@ cextern cos_ %+ i
 %assign i i<<1
 %endrep
 
-%ifdef ARCH_X86_64
-    %define pointer dq
-%else
-    %define pointer dd
-%endif
-
 %macro IF0 1+
 %endmacro
 %macro IF1 1+
@@ -584,7 +572,7 @@ DECL_PASS pass_interleave_3dn, PASS_BIG 0
 
 %macro FFT_DISPATCH 2; clobbers 5 GPRs, 8 XMMs
     lea r2, [dispatch_tab%1]
-    mov r2, [r2 + (%2q-2)*gprsize]
+    mov r2p, [r2 + (%2q-2)*ptrsize]
 %ifdef PIC
     lea r3, [$$]
     add r2, r3
@@ -624,7 +612,7 @@ fft %+ n %+ %3%2:
 %undef n
 
 align 8
-dispatch_tab%3%2: pointer list_of_fft
+dispatch_tab%3%2: dp list_of_fft
 
 section .text
 
@@ -765,8 +753,8 @@ cglobal imdct_half%1, 3,7,8; FFTContext *s, FFTSample *output, const FFTSample *
     mov   r3d, [r0+FFTContext.mdctsize]
     add   r2, r3
     shr   r3, 1
-    mov   rtcos, [r0+FFTContext.tcos]
-    mov   rtsin, [r0+FFTContext.tsin]
+    mov   preg(rtcos), [r0+FFTContext.tcos]
+    mov   preg(rtsin), [r0+FFTContext.tsin]
     add   rtcos, r3
     add   rtsin, r3
 %ifndef ARCH_X86_64
@@ -774,7 +762,7 @@ cglobal imdct_half%1, 3,7,8; FFTContext *s, FFTSample *output, const FFTSample *
     push  rtsin
 %endif
     shr   r3, 1
-    mov   rrevtab, [r0+FFTContext.revtab]
+    mov   preg(rrevtab), [r0+FFTContext.revtab]
     add   rrevtab, r3
 %ifndef ARCH_X86_64
     push  rrevtab
diff --git a/libavcodec/x86/fmtconvert.asm b/libavcodec/x86/fmtconvert.asm
index d621089..32dc500 100644
--- a/libavcodec/x86/fmtconvert.asm
+++ b/libavcodec/x86/fmtconvert.asm
@@ -122,8 +122,8 @@ FLOAT_TO_INT16 3dnow, 0
 %macro FLOAT_TO_INT16_INTERLEAVE2 1
 cglobal float_to_int16_interleave2_%1, 3,4,2, dst, src0, src1, len
     lea      lenq, [4*r2q]
-    mov     src1q, [src0q+gprsize]
-    mov     src0q, [src0q]
+    mov     src1p, [src0q+ptrsize]
+    mov     src0p, [src0q]
     add      dstq, lenq
     add     src0q, lenq
     add     src1q, lenq
@@ -186,12 +186,12 @@ cglobal float_to_int16_interleave6_%1, 2,7,0, dst, src, src1, src2, src3, src4,
 %else
     %define lend dword r2m
 %endif
-    mov src1q, [srcq+1*gprsize]
-    mov src2q, [srcq+2*gprsize]
-    mov src3q, [srcq+3*gprsize]
-    mov src4q, [srcq+4*gprsize]
-    mov src5q, [srcq+5*gprsize]
-    mov srcq,  [srcq]
+    mov src1p, [srcq+1*ptrsize]
+    mov src2p, [srcq+2*ptrsize]
+    mov src3p, [srcq+3*ptrsize]
+    mov src4p, [srcq+4*ptrsize]
+    mov src5p, [srcq+5*ptrsize]
+    mov srcp,  [srcq]
     sub src1q, srcq
     sub src2q, srcq
     sub src3q, srcq
@@ -247,12 +247,12 @@ cglobal float_interleave6_%1, 2,7,%2, dst, src, src1, src2, src3, src4, src5
 %else
     %define lend dword r2m
 %endif
-    mov    src1q, [srcq+1*gprsize]
-    mov    src2q, [srcq+2*gprsize]
-    mov    src3q, [srcq+3*gprsize]
-    mov    src4q, [srcq+4*gprsize]
-    mov    src5q, [srcq+5*gprsize]
-    mov     srcq, [srcq]
+    mov    src1p, [srcq+1*ptrsize]
+    mov    src2p, [srcq+2*ptrsize]
+    mov    src3p, [srcq+3*ptrsize]
+    mov    src4p, [srcq+4*ptrsize]
+    mov    src5p, [srcq+5*ptrsize]
+    mov     srcp, [srcq]
     sub    src1q, srcq
     sub    src2q, srcq
     sub    src3q, srcq
@@ -325,8 +325,8 @@ FLOAT_INTERLEAVE6 sse, 7
 
 %macro FLOAT_INTERLEAVE2 2
 cglobal float_interleave2_%1, 3,4,%2, dst, src, len, src1
-    mov     src1q, [srcq+gprsize]
-    mov      srcq, [srcq        ]
+    mov     src1p, [srcq+ptrsize]
+    mov      srcp, [srcq        ]
     sub     src1q, srcq
 .loop
     MOVPS      m0, [srcq             ]
diff --git a/libavcodec/x86/h264_idct.asm b/libavcodec/x86/h264_idct.asm
index 37c2c90..f44d76a 100644
--- a/libavcodec/x86/h264_idct.asm
+++ b/libavcodec/x86/h264_idct.asm
@@ -624,9 +624,9 @@ cglobal h264_idct_add8_8_mmx, 5, 7, 0
     mov          r5, 32
     add          r2, 384
 %ifdef ARCH_X86_64
-    add         r10, gprsize
+    add         r10, ptrsize
 %else
-    add        r0mp, gprsize
+    add        r0mp, ptrsize
 %endif
     call         h264_idct_add8_mmx_plane
     RET
@@ -639,7 +639,7 @@ h264_idct_add8_mmx2_plane
     jz .try_dc
 %ifdef ARCH_X86_64
     mov         r0d, dword [r1+r5*4]
-    add          r0, [r10]
+    add         r0p, [r10]
 %else
     mov          r0, r1m ; XXX r1m here is actually r0m of the calling func
     mov          r0, [r0]
@@ -687,9 +687,9 @@ cglobal h264_idct_add8_8_mmx2, 5, 7, 0
     mov          r5, 32
     add          r2, 384
 %ifdef ARCH_X86_64
-    add         r10, gprsize
+    add         r10, ptrsize
 %else
-    add        r0mp, gprsize
+    add        r0mp, ptrsize
 %endif
     call h264_idct_add8_mmx2_plane
     RET
@@ -819,7 +819,7 @@ cglobal h264_idct_add16intra_8_sse2, 5, 7, 8
     jz .try%1dc
 %ifdef ARCH_X86_64
     mov        r0d, dword [r1+(%1&1)*8+64*(1+(%1>>1))]
-    add         r0, [r10]
+    add        r0p, [r10]
 %else
     mov         r0, r0m
     mov         r0, [r0]
@@ -833,7 +833,7 @@ cglobal h264_idct_add16intra_8_sse2, 5, 7, 8
     jz .cycle%1end
 %ifdef ARCH_X86_64
     mov        r0d, dword [r1+(%1&1)*8+64*(1+(%1>>1))]
-    add         r0, [r10]
+    add        r0p, [r10]
 %else
     mov         r0, r0m
     mov         r0, [r0]
@@ -858,9 +858,9 @@ cglobal h264_idct_add8_8_sse2, 5, 7, 8
     add8_sse2_cycle 0, 0x34
     add8_sse2_cycle 1, 0x3c
 %ifdef ARCH_X86_64
-    add         r10, gprsize
+    add         r10, ptrsize
 %else
-    add        r0mp, gprsize
+    add        r0mp, ptrsize
 %endif
     add8_sse2_cycle 2, 0x5c
     add8_sse2_cycle 3, 0x64
diff --git a/libavcodec/x86/h264_idct_10bit.asm b/libavcodec/x86/h264_idct_10bit.asm
index 54636a9..fc26e18 100644
--- a/libavcodec/x86/h264_idct_10bit.asm
+++ b/libavcodec/x86/h264_idct_10bit.asm
@@ -320,15 +320,15 @@ cglobal h264_idct_add8_10_%1,5,7
     mov r10, r0
 %endif
     add      r2, 1024
-    mov      r0, [r0]
+    mov     r0p, [r0]
     ADD16_OP_INTRA %1, 16, 4+ 6*8
     ADD16_OP_INTRA %1, 18, 4+ 7*8
     add      r2, 1024-128*2
 %ifdef ARCH_X86_64
-    mov      r0, [r10+gprsize]
+    mov     r0p, [r10+ptrsize]
 %else
     mov      r0, r0m
-    mov      r0, [r0+gprsize]
+    mov      r0, [r0+ptrsize]
 %endif
     ADD16_OP_INTRA %1, 32, 4+11*8
     ADD16_OP_INTRA %1, 34, 4+12*8
diff --git a/libavcodec/x86/mlpdsp.c b/libavcodec/x86/mlpdsp.c
index 400855d..75c3db1 100644
--- a/libavcodec/x86/mlpdsp.c
+++ b/libavcodec/x86/mlpdsp.c
@@ -156,8 +156,8 @@ static void mlp_filter_channel_x86(int32_t *state, const int32_t *coeff,
           /* 2*/"+r"(sample_buffer),
 #if ARCH_X86_64
           /* 3*/"+r"(blocksize)
-        : /* 4*/"r"((x86_reg)mask), /* 5*/"r"(firjump),
-          /* 6*/"r"(iirjump)      , /* 7*/"c"(filter_shift)
+        : /* 4*/"r"((x86_native_reg)mask), /* 5*/"r"((x86_native_reg)(uintptr_t)firjump),
+          /* 6*/"r"((x86_native_reg)(uintptr_t)iirjump), /* 7*/"c"(filter_shift)
         , /* 8*/"r"((int64_t)coeff[0])
         , /* 9*/"r"((int64_t)coeff[1])
         , /*10*/"r"((int64_t)coeff[2])
diff --git a/libavutil/x86/cpu.c b/libavutil/x86/cpu.c
index 2424fe4..dad80d5 100644
--- a/libavutil/x86/cpu.c
+++ b/libavutil/x86/cpu.c
@@ -28,9 +28,9 @@
 /* ebx saving is necessary for PIC. gcc seems unable to see it alone */
 #define cpuid(index,eax,ebx,ecx,edx)\
     __asm__ volatile\
-        ("mov %%"REG_b", %%"REG_S"\n\t"\
+        ("mov %%"REG_rb", %%"REG_rS"\n\t"\
          "cpuid\n\t"\
-         "xchg %%"REG_b", %%"REG_S\
+         "xchg %%"REG_rb", %%"REG_rS\
          : "=a" (eax), "=S" (ebx),\
            "=c" (ecx), "=d" (edx)\
          : "0" (index));
diff --git a/libavutil/x86/x86inc.asm b/libavutil/x86/x86inc.asm
index 6941c1a..a25662b 100644
--- a/libavutil/x86/x86inc.asm
+++ b/libavutil/x86/x86inc.asm
@@ -120,12 +120,17 @@
 
 ; registers:
 ; rN and rNq are the native-size register holding function argument N
-; rNd, rNw, rNb are dword, word, and byte size
+; rNp, rNd, rNw, rNb are pointer, dword, word, and byte size
 ; rNm is the original location of arg N (a register or on the stack), dword
 ; rNmp is native size
 
 %macro DECLARE_REG 6
     %define r%1q %2
+    %ifdef ARCH_X86_64_X64
+        %define r%1p %2
+    %else
+        %define r%1p %3
+    %endif
     %define r%1d %3
     %define r%1w %4
     %define r%1b %5
@@ -143,6 +148,12 @@
 %macro DECLARE_REG_SIZE 2
     %define r%1q r%1
     %define e%1q r%1
+    %ifdef ARCH_X86_64_X64
+        %define r%1p r%1
+    %else
+        %define r%1p e%1
+    %endif
+    %define e%1p e%1
     %define r%1d e%1
     %define e%1d e%1
     %define r%1w %1
@@ -176,6 +187,11 @@ DECLARE_REG_SIZE bp, bpl
 %macro DECLARE_REG_TMP_SIZE 0-*
     %rep %0
         %define t%1q t%1 %+ q
+        %ifdef ARCH_X86_64_X64
+            %define t%1p t%1
+        %else
+            %define t%1p t%1 %+ d
+        %endif
         %define t%1d t%1 %+ d
         %define t%1w t%1 %+ w
         %define t%1b t%1 %+ b
@@ -191,6 +207,20 @@ DECLARE_REG_TMP_SIZE 0,1,2,3,4,5,6,7,8,9
     %define gprsize 4
 %endif
 
+%ifdef ARCH_X86_64_X64
+    %define ptrsize 8
+    %define pword qword
+    %define dp dq
+    %define resp resq
+    %define preg(x) x
+%else
+    %define ptrsize 4
+    %define pword dword
+    %define dp dd
+    %define resp resd
+    %define preg(x) x %+ d
+%endif
+
 %macro PUSH 1
     push %1
     %assign stack_offset stack_offset+gprsize
@@ -238,6 +268,7 @@ DECLARE_REG_TMP_SIZE 0,1,2,3,4,5,6,7,8,9
         %assign %%i 0
         %rep n_arg_names
             CAT_UNDEF arg_name %+ %%i, q
+            CAT_UNDEF arg_name %+ %%i, p
             CAT_UNDEF arg_name %+ %%i, d
             CAT_UNDEF arg_name %+ %%i, w
             CAT_UNDEF arg_name %+ %%i, b
@@ -250,6 +281,7 @@ DECLARE_REG_TMP_SIZE 0,1,2,3,4,5,6,7,8,9
     %assign %%i 0
     %rep %0
         %xdefine %1q r %+ %%i %+ q
+        %xdefine %1p r %+ %%i %+ p
         %xdefine %1d r %+ %%i %+ d
         %xdefine %1w r %+ %%i %+ w
         %xdefine %1b r %+ %%i %+ b
diff --git a/libavutil/x86_cpu.h b/libavutil/x86_cpu.h
index f84eba6..50f2646 100644
--- a/libavutil/x86_cpu.h
+++ b/libavutil/x86_cpu.h
@@ -24,7 +24,7 @@
 #include <stdint.h>
 #include "config.h"
 
-#if ARCH_X86_64
+#if ARCH_X86_64_X64
 #    define OPSIZE "q"
 #    define REG_a "rax"
 #    define REG_b "rbx"
@@ -32,6 +32,7 @@
 #    define REG_d "rdx"
 #    define REG_D "rdi"
 #    define REG_S "rsi"
+#    define REG_8 "r8"
 #    define PTR_SIZE "8"
 typedef int64_t x86_reg;
 
@@ -43,8 +44,9 @@ typedef int64_t x86_reg;
 #    define REGc    rcx
 #    define REGd    rdx
 #    define REGSP   rsp
+#    define REG8    r8
 
-#elif ARCH_X86_32
+#elif ARCH_X86_32 || ARCH_X86_64_X32
 
 #    define OPSIZE "l"
 #    define REG_a "eax"
@@ -53,6 +55,7 @@ typedef int64_t x86_reg;
 #    define REG_d "edx"
 #    define REG_D "edi"
 #    define REG_S "esi"
+#    define REG_8 "r8d"
 #    define PTR_SIZE "4"
 typedef int32_t x86_reg;
 
@@ -64,10 +67,23 @@ typedef int32_t x86_reg;
 #    define REGc    ecx
 #    define REGd    edx
 #    define REGSP   esp
+#    define REG8    r8d
 #else
 typedef int x86_reg;
 #endif
 
+#if ARCH_X86_64
+#    define REG_rb "rbx"
+#    define REG_rS "rsi"
+#    define REG_rBP "rbp"
+typedef int64_t x86_native_reg;
+#elif ARCH_X86_32
+#    define REG_rb "ebx"
+#    define REG_rS "esi"
+#    define REG_rBP "ebp"
+typedef int32_t x86_native_reg;
+#endif
+
 #define HAVE_7REGS (ARCH_X86_64 || (HAVE_EBX_AVAILABLE && HAVE_EBP_AVAILABLE))
 #define HAVE_6REGS (ARCH_X86_64 || (HAVE_EBX_AVAILABLE || HAVE_EBP_AVAILABLE))
 
diff --git a/libswscale/x86/output.asm b/libswscale/x86/output.asm
index ae2929c..7e18d1b 100644
--- a/libswscale/x86/output.asm
+++ b/libswscale/x86/output.asm
@@ -149,14 +149,14 @@ cglobal yuv2planeX_%1, %3, 7, %2
     movsx     cntr_reg,  r1m
 .filterloop_ %+ %%i:
     ; input pixels
-    mov             r6, [r2+gprsize*cntr_reg-2*gprsize]
+    mov            r6p, [r2+ptrsize*cntr_reg-2*ptrsize]
 %if %1 == 16
     mova            m3, [r6+r5*4]
     mova            m5, [r6+r5*4+mmsize]
 %else ; %1 == 8/9/10
     mova            m3, [r6+r5*2]
 %endif ; %1 == 8/9/10/16
-    mov             r6, [r2+gprsize*cntr_reg-gprsize]
+    mov            r6p, [r2+ptrsize*cntr_reg-ptrsize]
 %if %1 == 16
     mova            m4, [r6+r5*4]
     mova            m6, [r6+r5*4+mmsize]
diff --git a/libswscale/x86/swscale_template.c b/libswscale/x86/swscale_template.c
index 40188d8..d037738 100644
--- a/libswscale/x86/swscale_template.c
+++ b/libswscale/x86/swscale_template.c
@@ -774,12 +774,12 @@ static void RENAME(yuv2rgb32_2)(SwsContext *c, const int16_t *buf[2],
         const int16_t *abuf0 = abuf[0], *abuf1 = abuf[1];
 #if ARCH_X86_64
         __asm__ volatile(
-            YSCALEYUV2RGB(%%r8, %5)
-            YSCALEYUV2RGB_YA(%%r8, %5, %6, %7)
+            YSCALEYUV2RGB(%%REG8, %5)
+            YSCALEYUV2RGB_YA(%%REG8, %5, %6, %7)
             "psraw                  $3, %%mm1       \n\t" /* abuf0[eax] - abuf1[eax] >>7*/
             "psraw                  $3, %%mm7       \n\t" /* abuf0[eax] - abuf1[eax] >>7*/
             "packuswb            %%mm7, %%mm1       \n\t"
-            WRITEBGR32(%4, 8280(%5), %%r8, %%mm2, %%mm4, %%mm5, %%mm1, %%mm0, %%mm7, %%mm3, %%mm6)
+            WRITEBGR32(%4, 8280(%5), %%REG8, %%mm2, %%mm4, %%mm5, %%mm1, %%mm0, %%mm7, %%mm3, %%mm6)
             :: "c" (buf0), "d" (buf1), "S" (ubuf0), "D" (ubuf1), "r" (dest),
                "a" (&c->redDither),
                "r" (abuf0), "r" (abuf1)
@@ -791,7 +791,7 @@ static void RENAME(yuv2rgb32_2)(SwsContext *c, const int16_t *buf[2],
         __asm__ volatile(
             "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
             "mov        %4, %%"REG_b"               \n\t"
-            "push %%"REG_BP"                        \n\t"
+            "push %%"REG_rBP"                       \n\t"
             YSCALEYUV2RGB(%%REGBP, %5)
             "push                   %0              \n\t"
             "push                   %1              \n\t"
@@ -804,7 +804,7 @@ static void RENAME(yuv2rgb32_2)(SwsContext *c, const int16_t *buf[2],
             "pop                    %1              \n\t"
             "pop                    %0              \n\t"
             WRITEBGR32(%%REGb, 8280(%5), %%REGBP, %%mm2, %%mm4, %%mm5, %%mm1, %%mm0, %%mm7, %%mm3, %%mm6)
-            "pop %%"REG_BP"                         \n\t"
+            "pop %%"REG_rBP"                        \n\t"
             "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
             :: "c" (buf0), "d" (buf1), "S" (ubuf0), "D" (ubuf1), "m" (dest),
                "a" (&c->redDither)
@@ -814,11 +814,11 @@ static void RENAME(yuv2rgb32_2)(SwsContext *c, const int16_t *buf[2],
         __asm__ volatile(
             "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
             "mov        %4, %%"REG_b"               \n\t"
-            "push %%"REG_BP"                        \n\t"
+            "push %%"REG_rBP"                       \n\t"
             YSCALEYUV2RGB(%%REGBP, %5)
             "pcmpeqd %%mm7, %%mm7                   \n\t"
             WRITEBGR32(%%REGb, 8280(%5), %%REGBP, %%mm2, %%mm4, %%mm5, %%mm7, %%mm0, %%mm1, %%mm3, %%mm6)
-            "pop %%"REG_BP"                         \n\t"
+            "pop %%"REG_rBP"                        \n\t"
             "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
             :: "c" (buf0), "d" (buf1), "S" (ubuf0), "D" (ubuf1), "m" (dest),
                "a" (&c->redDither)
@@ -838,11 +838,11 @@ static void RENAME(yuv2bgr24_2)(SwsContext *c, const int16_t *buf[2],
     __asm__ volatile(
         "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
         "mov        %4, %%"REG_b"               \n\t"
-        "push %%"REG_BP"                        \n\t"
+        "push %%"REG_rBP"                       \n\t"
         YSCALEYUV2RGB(%%REGBP, %5)
         "pxor    %%mm7, %%mm7                   \n\t"
         WRITEBGR24(%%REGb, 8280(%5), %%REGBP)
-        "pop %%"REG_BP"                         \n\t"
+        "pop %%"REG_rBP"                        \n\t"
         "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
         :: "c" (buf0), "d" (buf1), "S" (ubuf0), "D" (ubuf1), "m" (dest),
            "a" (&c->redDither)
@@ -861,7 +861,7 @@ static void RENAME(yuv2rgb555_2)(SwsContext *c, const int16_t *buf[2],
     __asm__ volatile(
         "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
         "mov        %4, %%"REG_b"               \n\t"
-        "push %%"REG_BP"                        \n\t"
+        "push %%"REG_rBP"                       \n\t"
         YSCALEYUV2RGB(%%REGBP, %5)
         "pxor    %%mm7, %%mm7                   \n\t"
         /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */
@@ -871,7 +871,7 @@ static void RENAME(yuv2rgb555_2)(SwsContext *c, const int16_t *buf[2],
         "paddusb "RED_DITHER"(%5), %%mm5      \n\t"
 #endif
         WRITERGB15(%%REGb, 8280(%5), %%REGBP)
-        "pop %%"REG_BP"                         \n\t"
+        "pop %%"REG_rBP"                        \n\t"
         "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
         :: "c" (buf0), "d" (buf1), "S" (ubuf0), "D" (ubuf1), "m" (dest),
            "a" (&c->redDither)
@@ -890,7 +890,7 @@ static void RENAME(yuv2rgb565_2)(SwsContext *c, const int16_t *buf[2],
     __asm__ volatile(
         "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
         "mov        %4, %%"REG_b"               \n\t"
-        "push %%"REG_BP"                        \n\t"
+        "push %%"REG_rBP"                       \n\t"
         YSCALEYUV2RGB(%%REGBP, %5)
         "pxor    %%mm7, %%mm7                   \n\t"
         /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */
@@ -900,7 +900,7 @@ static void RENAME(yuv2rgb565_2)(SwsContext *c, const int16_t *buf[2],
         "paddusb "RED_DITHER"(%5), %%mm5      \n\t"
 #endif
         WRITERGB16(%%REGb, 8280(%5), %%REGBP)
-        "pop %%"REG_BP"                         \n\t"
+        "pop %%"REG_rBP"                        \n\t"
         "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
         :: "c" (buf0), "d" (buf1), "S" (ubuf0), "D" (ubuf1), "m" (dest),
            "a" (&c->redDither)
@@ -959,10 +959,10 @@ static void RENAME(yuv2yuyv422_2)(SwsContext *c, const int16_t *buf[2],
     __asm__ volatile(
         "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
         "mov %4, %%"REG_b"                        \n\t"
-        "push %%"REG_BP"                        \n\t"
+        "push %%"REG_rBP"                       \n\t"
         YSCALEYUV2PACKED(%%REGBP, %5)
         WRITEYUY2(%%REGb, 8280(%5), %%REGBP)
-        "pop %%"REG_BP"                         \n\t"
+        "pop %%"REG_rBP"                        \n\t"
         "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
         :: "c" (buf0), "d" (buf1), "S" (ubuf0), "D" (ubuf1), "m" (dest),
            "a" (&c->redDither)
@@ -1100,11 +1100,11 @@ static void RENAME(yuv2rgb32_1)(SwsContext *c, const int16_t *buf0,
             __asm__ volatile(
                 "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
                 "mov        %4, %%"REG_b"               \n\t"
-                "push %%"REG_BP"                        \n\t"
+                "push %%"REG_rBP"                       \n\t"
                 YSCALEYUV2RGB1(%%REGBP, %5)
                 YSCALEYUV2RGB1_ALPHA(%%REGBP)
                 WRITEBGR32(%%REGb, 8280(%5), %%REGBP, %%mm2, %%mm4, %%mm5, %%mm7, %%mm0, %%mm1, %%mm3, %%mm6)
-                "pop %%"REG_BP"                         \n\t"
+                "pop %%"REG_rBP"                        \n\t"
                 "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
                 :: "c" (buf0), "d" (abuf0), "S" (ubuf0), "D" (ubuf1), "m" (dest),
                    "a" (&c->redDither)
@@ -1113,11 +1113,11 @@ static void RENAME(yuv2rgb32_1)(SwsContext *c, const int16_t *buf0,
             __asm__ volatile(
                 "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
                 "mov        %4, %%"REG_b"               \n\t"
-                "push %%"REG_BP"                        \n\t"
+                "push %%"REG_rBP"                       \n\t"
                 YSCALEYUV2RGB1(%%REGBP, %5)
                 "pcmpeqd %%mm7, %%mm7                   \n\t"
                 WRITEBGR32(%%REGb, 8280(%5), %%REGBP, %%mm2, %%mm4, %%mm5, %%mm7, %%mm0, %%mm1, %%mm3, %%mm6)
-                "pop %%"REG_BP"                         \n\t"
+                "pop %%"REG_rBP"                        \n\t"
                 "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
                 :: "c" (buf0), "d" (buf1), "S" (ubuf0), "D" (ubuf1), "m" (dest),
                    "a" (&c->redDither)
@@ -1128,11 +1128,11 @@ static void RENAME(yuv2rgb32_1)(SwsContext *c, const int16_t *buf0,
             __asm__ volatile(
                 "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
                 "mov        %4, %%"REG_b"               \n\t"
-                "push %%"REG_BP"                        \n\t"
+                "push %%"REG_rBP"                       \n\t"
                 YSCALEYUV2RGB1b(%%REGBP, %5)
                 YSCALEYUV2RGB1_ALPHA(%%REGBP)
                 WRITEBGR32(%%REGb, 8280(%5), %%REGBP, %%mm2, %%mm4, %%mm5, %%mm7, %%mm0, %%mm1, %%mm3, %%mm6)
-                "pop %%"REG_BP"                         \n\t"
+                "pop %%"REG_rBP"                        \n\t"
                 "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
                 :: "c" (buf0), "d" (abuf0), "S" (ubuf0), "D" (ubuf1), "m" (dest),
                    "a" (&c->redDither)
@@ -1141,11 +1141,11 @@ static void RENAME(yuv2rgb32_1)(SwsContext *c, const int16_t *buf0,
             __asm__ volatile(
                 "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
                 "mov        %4, %%"REG_b"               \n\t"
-                "push %%"REG_BP"                        \n\t"
+                "push %%"REG_rBP"                       \n\t"
                 YSCALEYUV2RGB1b(%%REGBP, %5)
                 "pcmpeqd %%mm7, %%mm7                   \n\t"
                 WRITEBGR32(%%REGb, 8280(%5), %%REGBP, %%mm2, %%mm4, %%mm5, %%mm7, %%mm0, %%mm1, %%mm3, %%mm6)
-                "pop %%"REG_BP"                         \n\t"
+                "pop %%"REG_rBP"                        \n\t"
                 "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
                 :: "c" (buf0), "d" (buf1), "S" (ubuf0), "D" (ubuf1), "m" (dest),
                    "a" (&c->redDither)
@@ -1166,11 +1166,11 @@ static void RENAME(yuv2bgr24_1)(SwsContext *c, const int16_t *buf0,
         __asm__ volatile(
             "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
             "mov        %4, %%"REG_b"               \n\t"
-            "push %%"REG_BP"                        \n\t"
+            "push %%"REG_rBP"                       \n\t"
             YSCALEYUV2RGB1(%%REGBP, %5)
             "pxor    %%mm7, %%mm7                   \n\t"
             WRITEBGR24(%%REGb, 8280(%5), %%REGBP)
-            "pop %%"REG_BP"                         \n\t"
+            "pop %%"REG_rBP"                        \n\t"
             "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
             :: "c" (buf0), "d" (buf1), "S" (ubuf0), "D" (ubuf1), "m" (dest),
                "a" (&c->redDither)
@@ -1179,11 +1179,11 @@ static void RENAME(yuv2bgr24_1)(SwsContext *c, const int16_t *buf0,
         __asm__ volatile(
             "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
             "mov        %4, %%"REG_b"               \n\t"
-            "push %%"REG_BP"                        \n\t"
+            "push %%"REG_rBP"                       \n\t"
             YSCALEYUV2RGB1b(%%REGBP, %5)
             "pxor    %%mm7, %%mm7                   \n\t"
             WRITEBGR24(%%REGb, 8280(%5), %%REGBP)
-            "pop %%"REG_BP"                         \n\t"
+            "pop %%"REG_rBP"                        \n\t"
             "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
             :: "c" (buf0), "d" (buf1), "S" (ubuf0), "D" (ubuf1), "m" (dest),
                "a" (&c->redDither)
@@ -1203,7 +1203,7 @@ static void RENAME(yuv2rgb555_1)(SwsContext *c, const int16_t *buf0,
         __asm__ volatile(
             "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
             "mov        %4, %%"REG_b"               \n\t"
-            "push %%"REG_BP"                        \n\t"
+            "push %%"REG_rBP"                       \n\t"
             YSCALEYUV2RGB1(%%REGBP, %5)
             "pxor    %%mm7, %%mm7                   \n\t"
             /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */
@@ -1213,7 +1213,7 @@ static void RENAME(yuv2rgb555_1)(SwsContext *c, const int16_t *buf0,
             "paddusb "RED_DITHER"(%5), %%mm5      \n\t"
 #endif
             WRITERGB15(%%REGb, 8280(%5), %%REGBP)
-            "pop %%"REG_BP"                         \n\t"
+            "pop %%"REG_rBP"                        \n\t"
             "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
             :: "c" (buf0), "d" (buf1), "S" (ubuf0), "D" (ubuf1), "m" (dest),
                "a" (&c->redDither)
@@ -1222,7 +1222,7 @@ static void RENAME(yuv2rgb555_1)(SwsContext *c, const int16_t *buf0,
         __asm__ volatile(
             "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
             "mov        %4, %%"REG_b"               \n\t"
-            "push %%"REG_BP"                        \n\t"
+            "push %%"REG_rBP"                       \n\t"
             YSCALEYUV2RGB1b(%%REGBP, %5)
             "pxor    %%mm7, %%mm7                   \n\t"
             /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */
@@ -1232,7 +1232,7 @@ static void RENAME(yuv2rgb555_1)(SwsContext *c, const int16_t *buf0,
             "paddusb "RED_DITHER"(%5), %%mm5      \n\t"
 #endif
             WRITERGB15(%%REGb, 8280(%5), %%REGBP)
-            "pop %%"REG_BP"                         \n\t"
+            "pop %%"REG_rBP"                        \n\t"
             "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
             :: "c" (buf0), "d" (buf1), "S" (ubuf0), "D" (ubuf1), "m" (dest),
                "a" (&c->redDither)
@@ -1252,7 +1252,7 @@ static void RENAME(yuv2rgb565_1)(SwsContext *c, const int16_t *buf0,
         __asm__ volatile(
             "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
             "mov        %4, %%"REG_b"               \n\t"
-            "push %%"REG_BP"                        \n\t"
+            "push %%"REG_rBP"                       \n\t"
             YSCALEYUV2RGB1(%%REGBP, %5)
             "pxor    %%mm7, %%mm7                   \n\t"
             /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */
@@ -1262,7 +1262,7 @@ static void RENAME(yuv2rgb565_1)(SwsContext *c, const int16_t *buf0,
             "paddusb "RED_DITHER"(%5), %%mm5      \n\t"
 #endif
             WRITERGB16(%%REGb, 8280(%5), %%REGBP)
-            "pop %%"REG_BP"                         \n\t"
+            "pop %%"REG_rBP"                        \n\t"
             "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
             :: "c" (buf0), "d" (buf1), "S" (ubuf0), "D" (ubuf1), "m" (dest),
                "a" (&c->redDither)
@@ -1271,7 +1271,7 @@ static void RENAME(yuv2rgb565_1)(SwsContext *c, const int16_t *buf0,
         __asm__ volatile(
             "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
             "mov        %4, %%"REG_b"               \n\t"
-            "push %%"REG_BP"                        \n\t"
+            "push %%"REG_rBP"                       \n\t"
             YSCALEYUV2RGB1b(%%REGBP, %5)
             "pxor    %%mm7, %%mm7                   \n\t"
             /* mm2=B, %%mm4=G, %%mm5=R, %%mm7=0 */
@@ -1281,7 +1281,7 @@ static void RENAME(yuv2rgb565_1)(SwsContext *c, const int16_t *buf0,
             "paddusb "RED_DITHER"(%5), %%mm5      \n\t"
 #endif
             WRITERGB16(%%REGb, 8280(%5), %%REGBP)
-            "pop %%"REG_BP"                         \n\t"
+            "pop %%"REG_rBP"                        \n\t"
             "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
             :: "c" (buf0), "d" (buf1), "S" (ubuf0), "D" (ubuf1), "m" (dest),
                "a" (&c->redDither)
@@ -1338,10 +1338,10 @@ static void RENAME(yuv2yuyv422_1)(SwsContext *c, const int16_t *buf0,
         __asm__ volatile(
             "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
             "mov        %4, %%"REG_b"               \n\t"
-            "push %%"REG_BP"                        \n\t"
+            "push %%"REG_rBP"                       \n\t"
             YSCALEYUV2PACKED1(%%REGBP, %5)
             WRITEYUY2(%%REGb, 8280(%5), %%REGBP)
-            "pop %%"REG_BP"                         \n\t"
+            "pop %%"REG_rBP"                        \n\t"
             "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
             :: "c" (buf0), "d" (buf1), "S" (ubuf0), "D" (ubuf1), "m" (dest),
                "a" (&c->redDither)
@@ -1350,10 +1350,10 @@ static void RENAME(yuv2yuyv422_1)(SwsContext *c, const int16_t *buf0,
         __asm__ volatile(
             "mov %%"REG_b", "ESP_OFFSET"(%5)        \n\t"
             "mov        %4, %%"REG_b"               \n\t"
-            "push %%"REG_BP"                        \n\t"
+            "push %%"REG_rBP"                       \n\t"
             YSCALEYUV2PACKED1b(%%REGBP, %5)
             WRITEYUY2(%%REGb, 8280(%5), %%REGBP)
-            "pop %%"REG_BP"                         \n\t"
+            "pop %%"REG_rBP"                        \n\t"
             "mov "ESP_OFFSET"(%5), %%"REG_b"        \n\t"
             :: "c" (buf0), "d" (buf1), "S" (ubuf0), "D" (ubuf1), "m" (dest),
                "a" (&c->redDither)
@@ -1510,7 +1510,7 @@ static void RENAME(hyscale_fast)(SwsContext *c, int16_t *dst,
 {
     int32_t *filterPos = c->hLumFilterPos;
     int16_t *filter    = c->hLumFilter;
-    void    *mmx2FilterCode= c->lumMmx2FilterCode;
+    x86_native_reg mmx2FilterCode = (uintptr_t)c->lumMmx2FilterCode;
     int i;
 #if defined(PIC)
     uint64_t ebxsave;
-- 
1.8.1.2