nasm: match instruction length (movd/movq) to parameters

nasm requires the instruction length (movd/movq) to match to its
parameters. I find it more clear to really use 64bit instructions when
we use 64bit registers in the assembly.

Provide nasm compatibility. No binary change by this patch with yasm on
{x86_64,i686}-fedora13-linux-gnu. Few longer opcodes with nasm on
{x86_64,i686}-fedora13-linux-gnu have been checked as safe.

Change-Id: Id9b1a5cdfb1bc05697e523c317a296df43d42a91

9 files changed, 29 insertions, 29 deletions

diff --git a/vp8/common/x86/iwalsh_mmx.asm b/vp8/common/x86/iwalsh_mmx.asm
index 3f0671c58..10b5274dc 100644
--- a/vp8/common/x86/iwalsh_mmx.asm
+++ b/vp8/common/x86/iwalsh_mmx.asm
@@ -69,7 +69,7 @@ sym(vp8_short_inv_walsh4x4_mmx):
     movq    mm2, [rsi + 16]       ;ip[8]
     movq    mm3, [rsi + 24]       ;ip[12]
 
-    movd    mm7, rax
+    movq    mm7, rax
     movq    mm4, mm0
 
     punpcklwd mm7, mm7          ;0003000300030003h
diff --git a/vp8/decoder/x86/dequantize_mmx.asm b/vp8/decoder/x86/dequantize_mmx.asm
index 150d090b6..eb9d1f8aa 100644
--- a/vp8/decoder/x86/dequantize_mmx.asm
+++ b/vp8/decoder/x86/dequantize_mmx.asm
@@ -288,7 +288,7 @@ sym(vp8_dequant_dc_idct_add_mmx):
         psrlq       mm0,    16
         movzx       rcx,    word ptr arg(6) ;Dc
         psllq       mm0,    16
-        movd        mm7,    rcx
+        movq        mm7,    rcx
         por         mm0,    mm7
 
         movsxd      rax,            dword ptr arg(4) ;pitch
diff --git a/vp8/encoder/x86/encodeopt.asm b/vp8/encoder/x86/encodeopt.asm
index 413d74d61..c0f06bbbb 100644
--- a/vp8/encoder/x86/encodeopt.asm
+++ b/vp8/encoder/x86/encodeopt.asm
@@ -50,7 +50,7 @@ sym(vp8_block_error_xmm):
         psrldq      xmm0,       8
         paddd       xmm0,       xmm3
 
-        movd        rax,        xmm0
+        movq        rax,        xmm0
 
     pop rdi
     pop rsi
@@ -115,7 +115,7 @@ sym(vp8_block_error_mmx):
         psrlq       mm1,        32
         paddd       mm0,        mm1
 
-        movd        rax,        mm0
+        movq        rax,        mm0
 
     pop rdi
     pop rsi
@@ -192,7 +192,7 @@ mberror_loop_mmx:
         psrlq       mm2,        32
 
         paddd       mm0,        mm2
-        movd        rax,        mm0
+        movq        rax,        mm0
 
     pop rdi
     pop rsi
@@ -260,7 +260,7 @@ mberror_loop:
         psrldq      xmm0,       8
 
         paddd       xmm0,       xmm1
-        movd        rax,        xmm0
+        movq        rax,        xmm0
 
     pop rdi
     pop rsi
@@ -317,7 +317,7 @@ mbuverror_loop_mmx:
         psrlq           mm7,        32
 
         paddd           mm0,        mm7
-        movd            rax,        mm0
+        movq            rax,        mm0
 
     pop rdi
     pop rsi
@@ -374,7 +374,7 @@ mbuverror_loop:
         psrldq      xmm1,           8
         paddd       xmm1,           xmm2
 
-        movd            rax,            xmm1
+        movq            rax,            xmm1
 
     pop rdi
     pop rsi
diff --git a/vp8/encoder/x86/quantize_mmx.asm b/vp8/encoder/x86/quantize_mmx.asm
index a867409b5..51cd94078 100644
--- a/vp8/encoder/x86/quantize_mmx.asm
+++ b/vp8/encoder/x86/quantize_mmx.asm
@@ -249,7 +249,7 @@ sym(vp8_fast_quantize_b_impl_mmx):
         paddd           mm0,        mm5
 
         ; eob adjustment begins here
-        movd            rcx,        mm0
+        movq            rcx,        mm0
         and             rcx,        0xffff
 
         xor             rdx,        rdx
@@ -262,7 +262,7 @@ sym(vp8_fast_quantize_b_impl_mmx):
         and             rax,        rdx
         ; Substitute the sse assembly for the old mmx mixed assembly/C. The
         ; following is kept as reference
-        ;    movd            rcx,        mm0
+        ;    movq            rcx,        mm0
         ;    bsr             rax,        rcx
         ;
         ;    mov             eob,        rax
@@ -418,7 +418,7 @@ sym(vp8_fast_quantize_b_impl_sse):
         psrldq          xmm0,       4
         paddd           xmm1,       xmm0
 
-        movd            rcx,        xmm1
+        movq            rcx,        xmm1
         and             rcx,        0xffff
 
         xor             rdx,        rdx
diff --git a/vp8/encoder/x86/sad_mmx.asm b/vp8/encoder/x86/sad_mmx.asm
index ad9658bf6..19041d49f 100644
--- a/vp8/encoder/x86/sad_mmx.asm
+++ b/vp8/encoder/x86/sad_mmx.asm
@@ -100,7 +100,7 @@ x16x16sad_mmx_loop:
         psrlq           mm0,        32
         paddw           mm7,        mm0
 
-        movd            rax,        mm7
+        movq            rax,        mm7
 
     pop rdi
     pop rsi
@@ -172,7 +172,7 @@ x8x16sad_mmx_loop:
         psrlq           mm0,        32
 
         paddw           mm7,        mm0
-        movd            rax,        mm7
+        movq            rax,        mm7
 
     pop rdi
     pop rsi
@@ -242,7 +242,7 @@ x8x8sad_mmx_loop:
         psrlq           mm0,        32
 
         paddw           mm7,        mm0
-        movd            rax,        mm7
+        movq            rax,        mm7
 
     pop rdi
     pop rsi
@@ -331,7 +331,7 @@ sym(vp8_sad4x4_mmx):
         psrlq           mm0,        32
         paddw           mm0,        mm1
 
-        movd            rax,        mm0
+        movq            rax,        mm0
 
     pop rdi
     pop rsi
@@ -418,7 +418,7 @@ x16x8sad_mmx_loop:
         psrlq           mm0,        32
 
         paddw           mm7,        mm0
-        movd            rax,        mm7
+        movq            rax,        mm7
 
     pop rdi
     pop rsi
diff --git a/vp8/encoder/x86/sad_sse2.asm b/vp8/encoder/x86/sad_sse2.asm
index 9f34a7ac4..0f6c5d9c4 100644
--- a/vp8/encoder/x86/sad_sse2.asm
+++ b/vp8/encoder/x86/sad_sse2.asm
@@ -75,7 +75,7 @@ x16x16sad_wmt_loop:
         psrldq          xmm7,       8
 
         paddw           xmm0,       xmm7
-        movd            rax,        xmm0
+        movq            rax,        xmm0
 
     ; begin epilog
     pop rdi
@@ -113,7 +113,7 @@ sym(vp8_sad8x16_wmt):
 
 x8x16sad_wmt_loop:
 
-        movd            rax,        mm7
+        movq            rax,        mm7
         cmp             rax,        arg(4)
         jg              x8x16sad_wmt_early_exit
 
@@ -135,7 +135,7 @@ x8x16sad_wmt_loop:
         cmp             rsi,        rcx
         jne             x8x16sad_wmt_loop
 
-        movd            rax,        mm7
+        movq            rax,        mm7
 
 x8x16sad_wmt_early_exit:
 
@@ -174,7 +174,7 @@ sym(vp8_sad8x8_wmt):
 
 x8x8sad_wmt_loop:
 
-        movd            rax,        mm7
+        movq            rax,        mm7
         cmp             rax,        arg(4)
         jg              x8x8sad_wmt_early_exit
 
@@ -190,7 +190,7 @@ x8x8sad_wmt_loop:
         cmp             rsi,        rcx
         jne             x8x8sad_wmt_loop
 
-        movd            rax,        mm7
+        movq            rax,        mm7
 x8x8sad_wmt_early_exit:
 
     ; begin epilog
@@ -246,7 +246,7 @@ sym(vp8_sad4x4_wmt):
         psadbw          mm4,        mm5
 
         paddw           mm0,        mm4
-        movd            rax,        mm0
+        movq            rax,        mm0
 
     ; begin epilog
     pop rdi
@@ -283,7 +283,7 @@ sym(vp8_sad16x8_wmt):
 
 x16x8sad_wmt_loop:
 
-        movd            rax,        mm7
+        movq            rax,        mm7
         cmp             rax,        arg(4)
         jg              x16x8sad_wmt_early_exit
 
@@ -317,7 +317,7 @@ x16x8sad_wmt_loop:
         cmp             rsi,        rcx
         jne             x16x8sad_wmt_loop
 
-        movd            rax,        mm7
+        movq            rax,        mm7
 
 x16x8sad_wmt_early_exit:
 
diff --git a/vp8/encoder/x86/sad_sse3.asm b/vp8/encoder/x86/sad_sse3.asm
index c2a1ae70a..b12c81562 100644
--- a/vp8/encoder/x86/sad_sse3.asm
+++ b/vp8/encoder/x86/sad_sse3.asm
@@ -530,7 +530,7 @@ sym(vp8_sad16x16_sse3):
 
 vp8_sad16x16_sse3_loop:
 
-        movd            rax,        mm7
+        movq            rax,        mm7
         cmp             rax,        arg(4)
         jg              vp8_sad16x16_early_exit
 
@@ -564,7 +564,7 @@ vp8_sad16x16_sse3_loop:
         cmp             rsi,        rcx
         jne             vp8_sad16x16_sse3_loop
 
-        movd            rax,        mm7
+        movq            rax,        mm7
 
 vp8_sad16x16_early_exit:
 
diff --git a/vp8/encoder/x86/variance_impl_mmx.asm b/vp8/encoder/x86/variance_impl_mmx.asm
index 173238e24..d4ec63bd6 100644
--- a/vp8/encoder/x86/variance_impl_mmx.asm
+++ b/vp8/encoder/x86/variance_impl_mmx.asm
@@ -498,7 +498,7 @@ sym(vp8_get4x4sse_cs_mmx):
         psrlq       mm7,    32
 
         paddd       mm0,    mm7
-        movd        rax,    mm0
+        movq        rax,    mm0
 
 
     ; begin epilog
diff --git a/vp8/encoder/x86/variance_impl_sse2.asm b/vp8/encoder/x86/variance_impl_sse2.asm
index f47d9ccdd..38b3f33ee 100644
--- a/vp8/encoder/x86/variance_impl_sse2.asm
+++ b/vp8/encoder/x86/variance_impl_sse2.asm
@@ -58,7 +58,7 @@ NEXTROW:
         movdqa      xmm3,xmm4
         psrldq      xmm4,4
         paddd       xmm4,xmm3
-        movd        rax,xmm4
+        movq        rax,xmm4
 
 
     ; begin epilog
@@ -471,7 +471,7 @@ sym(vp8_get8x8var_sse2):
         mov         rax,            arg(5) ;[Sum]
         mov         rdi,            arg(4) ;[SSE]
 
-        movd        rdx,            xmm7
+        movq        rdx,            xmm7
         movsx       rcx,            dx
 
         mov  dword ptr [rax],       ecx