Attachment 624 Details for Bug 1067 – patch

[patch] patch

xorg-x11-6.8.0-fbmmx-improvements.patch (text/plain), 37.80 KB, created by Nicholas Miell on 2004-08-13 00:08:12 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Nicholas Miell

Created: 2004-08-13 00:08:12 UTC

Size: 37.80 KB

patch

obsolete

>Index: xc/programs/Xserver/fb/Imakefile
>===================================================================
>RCS file: /cvs/xorg/xc/programs/Xserver/fb/Imakefile,v
>retrieving revision 1.7
>diff -u -p -r1.7 Imakefile
>--- xc/programs/Xserver/fb/Imakefile	6 Aug 2004 23:42:10 -0000	1.7
>+++ xc/programs/Xserver/fb/Imakefile	13 Aug 2004 04:13:47 -0000
>@@ -3,13 +3,21 @@ XCOMM 
> XCOMM 
> XCOMM Id: Imakefile,v 1.1 1999/11/02 03:54:44 keithp Exp $
> 
>-#if defined(i386Architecture) && defined(HasGcc34) && HasGcc34
>+#if defined(HasGcc34) && HasGcc34
> MMXOPTIONS= -mmmx -Winline --param inline-unit-growth=10000 \
>-	--param large-function-growth=10000 -DUSE_GCC34_MMX
>+	--param large-function-growth=10000 -DUSE_MMX
>+SSEOPTIONS= $(MMXOPTIONS) -msse -DUSE_SSE
> 
>+#if defined(i386Architecture)
> SpecialCObjectRule(fbmmx,fbmmx.c,$(MMXOPTIONS))
>+#elif defined(AMD64Architecture)
>+SpecialCObjectRule(fbmmx,fbmmx.c,$(SSEOPTIONS))
>+#endif
>+
>+#if defined(i386Architecture) || defined(AMD64Architecture)
> SpecialCObjectRule(fbpict,fbpict.c,$(MMXOPTIONS))
> SpecialCObjectRule(fbfill,fbfill.c,$(MMXOPTIONS))
>+#endif
> 
> #endif
> 
>Index: xc/programs/Xserver/fb/fbfill.c
>===================================================================
>RCS file: /cvs/xorg/xc/programs/Xserver/fb/fbfill.c,v
>retrieving revision 1.3
>diff -u -p -r1.3 fbfill.c
>--- xc/programs/Xserver/fb/fbfill.c	22 Jul 2004 19:24:50 -0000	1.3
>+++ xc/programs/Xserver/fb/fbfill.c	13 Aug 2004 04:13:47 -0000
>@@ -44,7 +44,7 @@ fbFill (DrawablePtr pDrawable,
> 
>     switch (pGC->fillStyle) {
>     case FillSolid:
>-#ifdef USE_GCC34_MMX
>+#ifdef USE_MMX
> 	if (!pPriv->and && fbHaveMMX())
> 	    if (fbSolidFillmmx (pDrawable, x, y, width, height, pPriv->xor))
> 		return;
>Index: xc/programs/Xserver/fb/fbmmx.c
>===================================================================
>RCS file: /cvs/xorg/xc/programs/Xserver/fb/fbmmx.c,v
>retrieving revision 1.1
>diff -u -p -r1.1 fbmmx.c
>--- xc/programs/Xserver/fb/fbmmx.c	22 Jul 2004 19:24:50 -0000	1.1
>+++ xc/programs/Xserver/fb/fbmmx.c	13 Aug 2004 04:13:50 -0000
>@@ -1,21 +1,25 @@
> /*
>  * Copyright © 2004 Red Hat, Inc.
>+ * Copyright © 2004 Nicholas Miell
>  *
>- * Permission to use, copy, modify, distribute, and sell this software and its
>- * documentation for any purpose is hereby granted without fee, provided that
>- * the above copyright notice appear in all copies and that both that
>- * copyright notice and this permission notice appear in supporting
>- * documentation, and that the name of Red Hat not be used in advertising or
>- * publicity pertaining to distribution of the software without specific,
>- * written prior permission.  Red Hat makes no representations about the
>- * suitability of this software for any purpose.  It is provided "as is"
>- * without express or implied warranty.
>+ * Permission to use, copy, modify, distribute, and sell this software
>+ * and its documentation for any purpose is hereby granted without
>+ * fee, provided that the above copyright notice appear in all copies
>+ * and that both that copyright notice and this permission notice
>+ * appear in supporting documentation, and that the name of Red Hat
>+ * not be used in advertising or publicity pertaining to distribution
>+ * of the software without specific, written prior permission.  Red
>+ * Hat makes no representations about the suitability of this software
>+ * for any purpose.  It is provided "as is" without express or implied
>+ * warranty.
>  *
>- * RED HAT DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE, INCLUDING ALL
>- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS, IN NO EVENT SHALL RED HAT
>- * BE LIABLE FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
>- * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION
>- * OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN 
>+ * RED HAT DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE,
>+ * INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS, IN
>+ * NO EVENT SHALL RED HAT BE LIABLE FOR ANY SPECIAL, INDIRECT OR
>+ * CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS
>+ * OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT,
>+ * NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN
>  * CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
>  *
>  * Author:  Søren Sandmann (sandmann@redhat.com)
>+ * Improvements: Nicholas Miell (nmiell@gmail.com)
>@@ -23,9 +25,15 @@
>  * Based on work by Owen Taylor
>  */
> 
>+
>+#ifdef USE_MMX
>+
> #include "fb.h"
>+#include "fbmmx.h"
>+
>+#include <mmintrin.h>
>+#include <xmmintrin.h> /* for _mm_shuffle_pi16 and _MM_SHUFFLE */
> 
>-#ifdef USE_GCC34_MMX
> 
> #ifdef RENDER
> 
>@@ -33,11 +41,6 @@
> #include "mipict.h"
> #include "fbpict.h"
> 
>-typedef int Vector1x64 __attribute__ ((mode(DI)));
>-typedef int Vector2x32 __attribute__ ((mode(V2SI)));
>-typedef int Vector4x16 __attribute__ ((mode(V4HI)));
>-typedef int Vector8x8  __attribute__ ((mode(V8QI)));
>-
> typedef unsigned long long ullong;
> 
> #define noVERBOSE
>@@ -50,7 +53,6 @@ typedef unsigned long long ullong;
> 
> typedef struct
> {
>-    ullong mmx_zero;
>     ullong mmx_4x00ff;
>     ullong mmx_4x0080;
>     ullong mmx_565_rgb;
>@@ -70,7 +72,6 @@ typedef struct
> 
> static const MMXData c =
> {
>-    .mmx_zero =				0x0000000000000000ULL,
>     .mmx_4x00ff =			0x00ff00ff00ff00ffULL,
>     .mmx_4x0080 =			0x0080008000800080ULL,
>     .mmx_565_rgb =			0x000001f0003f001fULL,
>@@ -88,121 +89,101 @@ static const MMXData c =
>     .mmx_000000000000ffff =		0x000000000000ffffULL,
> };
> 
>-static __inline__ Vector1x64
>-shift (Vector1x64 v, int s)
>-{
>-    if (s > 0)
>-	return __builtin_ia32_psllq (v, s);
>-    else if (s < 0)
>-	return __builtin_ia32_psrlq (v, -s);
>-    else
>-	return v;
>-}
>+#define MC(x) ((__m64) c.mmx_##x)
> 
>-static __inline__ Vector4x16
>-negate (Vector4x16 mask)
>+static __inline__ __m64
>+negate (__m64 mask)
> {
>-    return (Vector4x16)__builtin_ia32_pxor (
>-	(Vector1x64)mask,
>-	(Vector1x64)c.mmx_4x00ff);
>+    return _mm_xor_si64 (mask, MC(4x00ff));
> }
> 
>-static __inline__ Vector4x16
>-pix_multiply (Vector4x16 a, Vector4x16 b)
>+static __inline__ __m64
>+pix_multiply (__m64 a, __m64 b)
> {
>-    Vector4x16 res;
>+    __m64 res;
>     
>-    res = __builtin_ia32_pmullw (a, b);
>-    res = __builtin_ia32_paddw (res, (Vector4x16)c.mmx_4x0080);
>-    res = __builtin_ia32_psrlw (res, 8);
>+    res = _mm_mullo_pi16 (a, b);
>+    res = _mm_add_pi16 (res, MC(4x0080));
>+    res = _mm_srli_pi16 (res, 8);
>     
>     return res;
> }
> 
>-#if 0
>+#ifdef USE_SSE
> #define HAVE_PSHUFW
> #endif
> 
> #ifdef HAVE_PSHUFW
> 
>-static __inline__ Vector4x16
>-expand_alpha (Vector4x16 pixel)
>+static __inline__ __m64
>+expand_alpha (__m64 pixel)
> {
>-    Vector4x16 result;
>-    __asm__ ("pshufw $0xFF, %1, %0\n\t" : "=y" (result) : "y" (pixel));
>-    return result;
>+    return _mm_shuffle_pi16 (pixel, _MM_SHUFFLE(3, 3, 3, 3));
> }
> 
>-static __inline__ Vector4x16
>-expand_alpha_rev (Vector4x16 pixel)
>+static __inline__ __m64
>+expand_alpha_rev (__m64 pixel)
> {
>-    Vector4x16 result;
>-    __asm__ ("pshufw $0x00, %1, %0\n\t" : "=y" (result) : "y" (pixel));
>-    return result;
>+    return _mm_shuffle_pi16 (pixel, _MM_SHUFFLE(0, 0, 0, 0));
> }    
> 
>-static __inline__ Vector4x16
>-invert_colors (Vector4x16 pixel)
>+static __inline__ __m64
>+invert_colors (__m64 pixel)
> {
>-    Vector4x16 result;
>-
>-    /* 0xC6 = 11000110 */
>-    /*         3 0 1 2 */
>-    
>-    __asm__ ("pshufw $0xC6, %1, %0\n\t" : "=y" (result) : "y" (pixel));
>-
>-    return result;
>+    return _mm_shuffle_pi16 (pixel, _MM_SHUFFLE(3, 0, 1, 2));
> }
> 
> #else
> 
>-static __inline__ Vector4x16
>-expand_alpha (Vector4x16 pixel)
>+static __inline__ __m64
>+expand_alpha (__m64 pixel)
> {
>-    Vector1x64 t1, t2;
>+    __m64 t1, t2;
> 
>-    t1 = shift ((Vector1x64)pixel, -48);
>-    t2 = shift (t1, 16);
>-    t1 = __builtin_ia32_por (t1, t2);
>-    t2 = shift (t1, 32);
>-    t1 = __builtin_ia32_por (t1, t2);
>+    t1 = _mm_srli_si64 (pixel, 48);
>+    t2 = _mm_slli_si64 (pixel, 16);
>+    t1 = _mm_or_si64 (t1, t2);
>+    t2 = _mm_slli_si64 (pixel, 32);
>+    t1 = _mm_or_si64 (t1, t2);
> 
>-    return (Vector4x16)t1;
>+    return t1;
> }
> 
>-static __inline__ Vector4x16
>-expand_alpha_rev (Vector4x16 pixel)
>+static __inline__ __m64
>+expand_alpha_rev (__m64 pixel)
> {
>-    Vector1x64 t1, t2;
>+    __m64 t1, t2;
>+
>+    /* move alpha to low 16 bits and zero the rest */
>+    t1 = _mm_slli_si64 (pixel,  48);
>+    t1 = _mm_srli_si64 (t1, 48);
> 
>-    t1 = shift ((Vector1x64)pixel,  48);
>-    t1 = shift (t1, -48);
>-    t2 = shift (t1, 16);
>-    t1 = __builtin_ia32_por (t1, t2);
>-    t2 = shift (t1, 32);
>-    t1 = __builtin_ia32_por (t1, t2);
>+    t2 = _mm_slli_si64 (t1, 16);
>+    t1 = _mm_or_si64 (t1, t2);
>+    t2 = _mm_slli_si64 (t1, 32);
>+    t1 = _mm_or_si64 (t1, t2);
> 
>-    return (Vector4x16)t1;
>+    return t1;
> }
> 
>-static __inline__ Vector4x16
>-invert_colors (Vector4x16 pixel)
>+static __inline__ __m64
>+invert_colors (__m64 pixel)
> {
>-    Vector1x64 x, y, z;
>+    __m64 x, y, z;
> 
>-    x = y = z = (Vector1x64)pixel;
>+    x = y = z = pixel;
> 
>-    x = __builtin_ia32_pand (x, (Vector1x64)c.mmx_ffff0000ffff0000);
>-    y = __builtin_ia32_pand (y, (Vector1x64)c.mmx_000000000000ffff);
>-    z = __builtin_ia32_pand (z, (Vector1x64)c.mmx_0000ffff00000000);
>+    x = _mm_and_si64 (x, MC(ffff0000ffff0000));
>+    y = _mm_and_si64 (y, MC(000000000000ffff));
>+    z = _mm_and_si64 (z, MC(0000ffff00000000));
> 
>-    y = shift (y, 32);
>-    z = shift (z, -32);
>+    y = _mm_slli_si64 (y, 32);
>+    z = _mm_srli_si64 (z, 32);
> 
>-    x = __builtin_ia32_por (x, y);
>-    x = __builtin_ia32_por (x, z);
>+    x = _mm_or_si64 (x, y);
>+    x = _mm_or_si64 (x, z);
> 
>-    return (Vector4x16)x;
>+    return x;
> }
> 
> #endif
>@@ -210,74 +191,69 @@ invert_colors (Vector4x16 pixel)
> /* Notes about writing mmx code
>  *
>  * give memory operands as the second operand. If you give it as the
>- * first, gcc will first load it into a register, then use that register
>+ * first, gcc will first load it into a register, then use that
>+ * register
>  *
>  *   ie. use
>  *
>- *         __builtin_pmullw (x, mmx_constant[8]);
>+ *         _mm_mullo_pi16 (x, mmx_constant);
>  *
>  *   not
>  *
>- *         __builtin_pmullw (mmx_constant[8], x);
>+ *         _mm_mullo_pi16 (mmx_constant, x);
>  *
>- * Also try to minimize dependencies. Ie. when you need a value, try to calculate
>- * it from a value that was calculated as early as possible.
>+ * Also try to minimize dependencies. i.e. when you need a value, try
>+ * to calculate it from a value that was calculated as early as
>+ * possible.
>  */
> 
>-static __inline__ Vector4x16
>-over (Vector4x16 src, Vector4x16 srca, Vector4x16 dest)
>+static __inline__ __m64
>+over (__m64 src, __m64 srca, __m64 dest)
> {
>-    return (Vector4x16)__builtin_ia32_paddusb ((Vector8x8)src, (Vector8x8)pix_multiply(dest, negate(srca)));
>+    return  _mm_adds_pu8 (src, pix_multiply(dest, negate(srca)));
> }
> 
>-static __inline__ Vector4x16
>-over_rev_non_pre (Vector4x16 src, Vector4x16 dest)
>+static __inline__ __m64
>+over_rev_non_pre (__m64 src, __m64 dest)
> {
>-    Vector4x16 srca = expand_alpha (src);
>-    Vector4x16 srcfaaa = (Vector4x16)__builtin_ia32_por((Vector1x64)srca, (Vector1x64)c.mmx_full_alpha);
>+    __m64 srca = expand_alpha (src);
>+    __m64 srcfaaa = _mm_or_si64 (srca, MC(full_alpha));
> 
>     return over(pix_multiply(invert_colors(src), srcfaaa), srca, dest);
> }
> 
>-static __inline__ Vector4x16
>-in (Vector4x16 src,
>-    Vector4x16 mask)
>+static __inline__ __m64
>+in (__m64 src,
>+    __m64 mask)
> {
>     return pix_multiply (src, mask);
> }
> 
>-static __inline__ Vector4x16
>-in_over (Vector4x16 src,
>-	 Vector4x16 srca,
>-	 Vector4x16 mask,
>-	 Vector4x16 dest)
>+static __inline__ __m64
>+in_over (__m64 src,
>+	 __m64 srca,
>+	 __m64 mask,
>+	 __m64 dest)
> {
>     return over(in(src, mask), pix_multiply(srca, mask), dest);
> }
> 
>-static __inline__ Vector8x8
>-cvt32to64 (CARD32 v)
>-{
>-    ullong r = v;
>-    return (Vector8x8)r;
>-}
>-
>-static __inline__ Vector4x16
>+static __inline__ __m64
> load8888 (CARD32 v)
> {
>-    return (Vector4x16)__builtin_ia32_punpcklbw (cvt32to64 (v),
>-						 (Vector8x8)c.mmx_zero);
>+    return _mm_unpacklo_pi8 (_mm_cvtsi32_si64 (v), _mm_setzero_si64());
> }
> 
>-static __inline__ Vector8x8
>-pack8888 (Vector4x16 lo, Vector4x16 hi)
>+static __inline__ __m64
>+pack8888 (__m64 lo, __m64 hi)
> {
>-    Vector8x8 r;
>-    r = __builtin_ia32_packuswb ((Vector4x16)lo, (Vector4x16)hi);
>+    __m64 r;
>+    r = _mm_packs_pu16 (lo, hi);
>     return r;
> }
> 
>-/* Expand 16 bits positioned at @pos (0-3) of a mmx register into 00RR00GG00BB
>+/* Expand 16 bits positioned at @pos (0-3) of a mmx register into
>+   00RR00GG00BB
>    
> --- Expanding 565 in the low word ---
> 
>@@ -286,71 +262,65 @@ m = m & (01f0003f001f);
> m = m * (008404100840);
> m = m >> 8;
> 
>-Note the trick here - the top word is shifted by another nibble to avoid
>-it bumping into the middle word
>+Note the trick here - the top word is shifted by another nibble to
>+avoid it bumping into the middle word
> */
>-static __inline__ Vector4x16
>-expand565 (Vector4x16 pixel, int pos)
>+static __inline__ __m64
>+expand565 (__m64 pixel, int pos)
> {
>-    Vector1x64 p = (Vector1x64)pixel;
>+    __m64 p = pixel;
>     
>     /* move pixel to low 16 bit and zero the rest */
>-    p = shift (shift (p, (3 - pos) * 16), -48); 
>+    p = _mm_srli_si64 (_mm_slli_si64 (p, (3 - pos) * 16), 48); 
>     
>-    Vector1x64 t1 = shift (p, 36 - 11);
>-    Vector1x64 t2 = shift (p, 16 - 5);
>+    __m64 t1 = _mm_slli_si64 (p, 36 - 11);
>+    __m64 t2 = _mm_slli_si64 (p, 16 - 5);
>     
>-    p = __builtin_ia32_por (t1, p);
>-    p = __builtin_ia32_por (t2, p);
>-    p = __builtin_ia32_pand (p, (Vector1x64)c.mmx_565_rgb);
>+    p = _mm_or_si64 (t1, p);
>+    p = _mm_or_si64 (t2, p);
>+    p = _mm_and_si64 (p, MC(565_rgb));
>     
>-    pixel = __builtin_ia32_pmullw ((Vector4x16)p, (Vector4x16)c.mmx_565_unpack_multiplier);
>-    return __builtin_ia32_psrlw (pixel, 8);
>+    pixel = _mm_mullo_pi16 (p, MC(565_unpack_multiplier));
>+    return _mm_srli_pi16 (pixel, 8);
> }
> 
>-static __inline__ Vector4x16
>-expand8888 (Vector4x16 in, int pos)
>+static __inline__ __m64
>+expand8888 (__m64 in, int pos)
> {
>     if (pos == 0)
>-	return (Vector4x16)__builtin_ia32_punpcklbw ((Vector8x8)in, (Vector8x8)c.mmx_zero);
>+	 return _mm_unpacklo_pi8 (in, _mm_setzero_si64());
>     else
>-	return (Vector4x16)__builtin_ia32_punpckhbw ((Vector8x8)in, (Vector8x8)c.mmx_zero);
>+	 return _mm_unpackhi_pi8 (in, _mm_setzero_si64());
> }
> 
>-static __inline__ Vector4x16
>-pack565 (Vector4x16 pixel, Vector4x16 target, int pos)
>+static __inline__ __m64
>+pack565 (__m64 pixel, __m64 target, int pos)
> {
>-    Vector1x64 p = (Vector1x64)pixel;
>-    Vector1x64 t = (Vector1x64)target;
>-    Vector1x64 r, g, b;
>+    __m64 p = pixel;
>+    __m64 t = target;
>+    __m64 r, g, b;
>     
>-    r = __builtin_ia32_pand (p, (Vector1x64)c.mmx_565_r);
>-    g = __builtin_ia32_pand (p, (Vector1x64)c.mmx_565_g);
>-    b = __builtin_ia32_pand (p, (Vector1x64)c.mmx_565_b);
>+    r = _mm_and_si64 (p, MC(565_r));
>+    g = _mm_and_si64 (p, MC(565_g));
>+    b = _mm_and_si64 (p, MC(565_b));
>     
>-    r = shift (r, - (32 - 8) + pos * 16);
>-    g = shift (g, - (16 - 3) + pos * 16);
>-    b = shift (b, - (0  + 3) + pos * 16);
>+    r = _mm_slli_si64 (r, - (32 - 8) + pos * 16);
>+    g = _mm_slli_si64 (g, - (16 - 3) + pos * 16);
>+    b = _mm_slli_si64 (b, - (0  + 3) + pos * 16);
> 
>     if (pos == 0)
>-	t = __builtin_ia32_pand (t, (Vector1x64)c.mmx_mask_0);
>+	 t = _mm_and_si64 (t, MC(mask_0));
>     else if (pos == 1)
>-	t = __builtin_ia32_pand (t, (Vector1x64)c.mmx_mask_1);
>+	 t = _mm_and_si64 (t, MC(mask_1));
>     else if (pos == 2)
>-	t = __builtin_ia32_pand (t, (Vector1x64)c.mmx_mask_2);
>+	 t = _mm_and_si64 (t, MC(mask_2));
>     else if (pos == 3)
>-	t = __builtin_ia32_pand (t, (Vector1x64)c.mmx_mask_3);
>+	 t = _mm_and_si64 (t, MC(mask_3));
>     
>-    p = __builtin_ia32_por (r, t);
>-    p = __builtin_ia32_por (g, p);
>+    p = _mm_or_si64 (r, t);
>+    p = _mm_or_si64 (g, p);
>     
>-    return (Vector4x16)__builtin_ia32_por (b, p);
>-}
>-
>-static __inline__ void
>-emms (void)
>-{
>-    __asm__ __volatile__ ("emms");
>+    return _mm_or_si64 (b, p);
> }
> 
> void
>@@ -371,7 +341,7 @@ fbCompositeSolid_nx8888mmx (CARD8	op,
>     CARD32	*dstLine, *dst;
>     CARD16	w;
>     FbStride	dstStride;
>-    Vector4x16	vsrc, vsrca;
>+    __m64	vsrc, vsrca;
> 
>     CHECKPOINT();
>     
>@@ -395,7 +365,7 @@ fbCompositeSolid_nx8888mmx (CARD8	op,
> 	
> 	while (w && (unsigned long)dst & 7)
> 	{
>-	    *dst = (ullong) pack8888(over(vsrc, vsrca, load8888(*dst)), (Vector4x16)c.mmx_zero);
>+	     *dst = (ullong) pack8888(over(vsrc, vsrca, load8888(*dst)), _mm_setzero_si64());
> 	    
> 	    w--;
> 	    dst++;
>@@ -403,15 +373,15 @@ fbCompositeSolid_nx8888mmx (CARD8	op,
> 
> 	while (w >= 2)
> 	{
>-	    Vector4x16 vdest;
>-	    Vector4x16 dest0, dest1;
>+	    __m64 vdest;
>+	    __m64 dest0, dest1;
> 
>-	    vdest = *(Vector4x16 *)dst;
>+	    vdest = *(__m64 *)dst;
> 	    
> 	    dest0 = over(vsrc, vsrca, expand8888(vdest, 0));
> 	    dest1 = over(vsrc, vsrca, expand8888(vdest, 1));
> 	    
>-	    *(Vector8x8 *)dst = (Vector8x8)pack8888(dest0, dest1);
>+	    *(__m64 *)dst = pack8888(dest0, dest1);
> 	    
> 	    dst += 2;
> 	    w -= 2;
>@@ -421,14 +391,14 @@ fbCompositeSolid_nx8888mmx (CARD8	op,
> 	
> 	while (w)
> 	{
>-	    *dst = (ullong) pack8888(over(vsrc, vsrca, load8888(*dst)), (Vector4x16)c.mmx_zero);
>+	     *dst = (ullong) pack8888(over(vsrc, vsrca, load8888(*dst)), _mm_setzero_si64());
> 	    
> 	    w--;
> 	    dst++;
> 	}
>     }
>     
>-    emms();
>+    _mm_empty();
> }
> 
> void
>@@ -449,7 +419,7 @@ fbCompositeSolid_nx0565mmx (CARD8	op,
>     CARD16	*dstLine, *dst;
>     CARD16	w;
>     FbStride	dstStride;
>-    Vector4x16	vsrc, vsrca;
>+    __m64	vsrc, vsrca;
> 
>     CHECKPOINT();
>     
>@@ -474,7 +444,7 @@ fbCompositeSolid_nx0565mmx (CARD8	op,
> 	while (w && (unsigned long)dst & 7)
> 	{
> 	    ullong d = *dst;
>-	    Vector4x16 vdest = expand565 ((Vector4x16)d, 0);
>+	    __m64 vdest = expand565 ((__m64)d, 0);
> 	    vdest = pack565(over(vsrc, vsrca, vdest), vdest, 0);
> 	    *dst = (ullong)vdest;
> 	    
>@@ -484,16 +454,16 @@ fbCompositeSolid_nx0565mmx (CARD8	op,
> 
> 	while (w >= 4)
> 	{
>-	    Vector4x16 vdest;
>+	    __m64 vdest;
> 
>-	    vdest = *(Vector4x16 *)dst;
>+	    vdest = *(__m64 *)dst;
> 	    
> 	    vdest = pack565 (over(vsrc, vsrca, expand565(vdest, 0)), vdest, 0);
> 	    vdest = pack565 (over(vsrc, vsrca, expand565(vdest, 1)), vdest, 1);
> 	    vdest = pack565 (over(vsrc, vsrca, expand565(vdest, 2)), vdest, 2);
> 	    vdest = pack565 (over(vsrc, vsrca, expand565(vdest, 3)), vdest, 3);
> 	    
>-	    *(Vector8x8 *)dst = (Vector8x8)vdest;
>+	    *(__m64 *)dst = vdest;
> 	    
> 	    dst += 4;
> 	    w -= 4;
>@@ -504,7 +474,7 @@ fbCompositeSolid_nx0565mmx (CARD8	op,
> 	while (w)
> 	{
> 	    ullong d = *dst;
>-	    Vector4x16 vdest = expand565 ((Vector4x16)d, 0);
>+	    __m64 vdest = expand565 ((__m64)d, 0);
> 	    vdest = pack565(over(vsrc, vsrca, vdest), vdest, 0);
> 	    *dst = (ullong)vdest;
> 	    
>@@ -513,7 +483,7 @@ fbCompositeSolid_nx0565mmx (CARD8	op,
> 	}
>     }
>     
>-    emms();
>+    _mm_empty();
> }
> 
> void
>@@ -534,7 +504,7 @@ fbCompositeSolidMask_nx8888x8888Cmmx (CA
>     CARD32	*dstLine;
>     CARD32	*maskLine;
>     FbStride	dstStride, maskStride;
>-    Vector4x16	vsrc, vsrca;
>+    __m64	vsrc, vsrca;
> 
>     CHECKPOINT();
>     
>@@ -562,9 +532,9 @@ fbCompositeSolidMask_nx8888x8888Cmmx (CA
> 	    
> 	    if (m)
> 	    {
>-		Vector4x16 vdest = load8888(*q);
>+		__m64 vdest = load8888(*q);
> 		vdest = in_over(vsrc, vsrca, load8888(m), vdest);
>-		*q = (ullong)pack8888(vdest, (Vector4x16)c.mmx_zero);
>+		*q = (ullong)pack8888(vdest, _mm_setzero_si64());
> 	    }
> 	    
> 	    twidth--;
>@@ -580,15 +550,15 @@ fbCompositeSolidMask_nx8888x8888Cmmx (CA
> 	    
> 	    if (m0 | m1)
> 	    {
>-		Vector4x16 dest0, dest1;
>-		Vector4x16 vdest = *(Vector4x16 *)q;
>+		__m64 dest0, dest1;
>+		__m64 vdest = *(__m64 *)q;
> 		
> 		dest0 = in_over(vsrc, vsrca, load8888(m0),
> 				expand8888 (vdest, 0));
> 		dest1 = in_over(vsrc, vsrca, load8888(m1),
> 				expand8888 (vdest, 1));
> 		
>-		*(Vector8x8 *)q = (Vector8x8)pack8888(dest0, dest1);
>+		*(__m64 *)q = pack8888(dest0, dest1);
> 	    }
> 	    
> 	    p += 2;
>@@ -602,9 +572,9 @@ fbCompositeSolidMask_nx8888x8888Cmmx (CA
> 	    
> 	    if (m)
> 	    {
>-		Vector4x16 vdest = load8888(*q);
>+		__m64 vdest = load8888(*q);
> 		vdest = in_over(vsrc, vsrca, load8888(m), vdest);
>-		*q = (ullong)pack8888(vdest, (Vector4x16)c.mmx_zero);
>+		*q = (ullong)pack8888(vdest, _mm_setzero_si64());
> 	    }
> 	    
> 	    twidth--;
>@@ -616,7 +586,7 @@ fbCompositeSolidMask_nx8888x8888Cmmx (CA
> 	maskLine += maskStride;
>     }
>     
>-    emms();
>+    _mm_empty();
> }
> 
> void
>@@ -638,7 +608,7 @@ fbCompositeSolidMask_nx8x8888mmx (CARD8 
>     CARD8	*maskLine, *mask;
>     FbStride	dstStride, maskStride;
>     CARD16	w;
>-    Vector4x16	vsrc, vsrca;
>+    __m64	vsrc, vsrca;
>     ullong	srcsrc;
>     
>     CHECKPOINT();
>@@ -673,8 +643,8 @@ fbCompositeSolidMask_nx8x8888mmx (CARD8 
> 	    
> 	    if (m)
> 	    {
>-		Vector4x16 vdest = in_over(vsrc, vsrca, expand_alpha_rev ((Vector4x16)m), load8888(*dst));
>-		*dst = (ullong)pack8888(vdest, (Vector4x16)c.mmx_zero);
>+		__m64 vdest = in_over(vsrc, vsrca, expand_alpha_rev ((__m64)m), load8888(*dst));
>+		*dst = (ullong)pack8888(vdest, _mm_setzero_si64());
> 	    }
> 	    
> 	    w--;
>@@ -696,15 +666,15 @@ fbCompositeSolidMask_nx8x8888mmx (CARD8 
> 	    }
> 	    else if (m0 | m1)
> 	    {
>-		Vector4x16 vdest;
>-		Vector4x16 dest0, dest1;
>+		__m64 vdest;
>+		__m64 dest0, dest1;
> 
>-		vdest = *(Vector4x16 *)dst;
>+		vdest = *(__m64 *)dst;
> 		
>-		dest0 = in_over(vsrc, vsrca, expand_alpha_rev ((Vector4x16)m0), expand8888(vdest, 0));
>-		dest1 = in_over(vsrc, vsrca, expand_alpha_rev ((Vector4x16)m1), expand8888(vdest, 1));
>+		dest0 = in_over(vsrc, vsrca, expand_alpha_rev ((__m64)m0), expand8888(vdest, 0));
>+		dest1 = in_over(vsrc, vsrca, expand_alpha_rev ((__m64)m1), expand8888(vdest, 1));
> 		
>-		*(Vector8x8 *)dst = (Vector8x8)pack8888(dest0, dest1);
>+		*(__m64 *)dst = pack8888(dest0, dest1);
> 	    }
> 	    
> 	    mask += 2;
>@@ -720,9 +690,9 @@ fbCompositeSolidMask_nx8x8888mmx (CARD8 
> 	    
> 	    if (m)
> 	    {
>-		Vector4x16 vdest = load8888(*dst);
>-		vdest = in_over(vsrc, vsrca, expand_alpha_rev ((Vector4x16)m), vdest);
>-		*dst = (ullong)pack8888(vdest, (Vector4x16)c.mmx_zero);
>+		__m64 vdest = load8888(*dst);
>+		vdest = in_over(vsrc, vsrca, expand_alpha_rev ((__m64)m), vdest);
>+		*dst = (ullong)pack8888(vdest, _mm_setzero_si64());
> 	    }
> 	    
> 	    w--;
>@@ -731,7 +701,7 @@ fbCompositeSolidMask_nx8x8888mmx (CARD8 
> 	}
>     }
>     
>-    emms();
>+    _mm_empty();
> }
> 
> 
>@@ -754,7 +724,7 @@ fbCompositeSolidMask_nx8x0565mmx (CARD8 
>     CARD8	*maskLine, *mask;
>     FbStride	dstStride, maskStride;
>     CARD16	w;
>-    Vector4x16	vsrc, vsrca;
>+    __m64	vsrc, vsrca;
>     unsigned long long srcsrcsrcsrc, src16;
>     
>     CHECKPOINT();
>@@ -771,7 +741,7 @@ fbCompositeSolidMask_nx8x0565mmx (CARD8 
>     vsrc = load8888 (src);
>     vsrca = expand_alpha (vsrc);
> 
>-    src16 = (ullong)pack565(vsrc, (Vector4x16)c.mmx_zero, 0);
>+    src16 = (ullong)pack565(vsrc, _mm_setzero_si64(), 0);
> 
>     srcsrcsrcsrc = (ullong)src16 << 48 | (ullong)src16 << 32 |
> 	(ullong)src16 << 16 | (ullong)src16;
>@@ -793,9 +763,9 @@ fbCompositeSolidMask_nx8x0565mmx (CARD8 
> 	    if (m)
> 	    {
> 		ullong d = *dst;
>-		Vector4x16 vd = (Vector4x16)d;
>-		Vector4x16 vdest = in_over(vsrc, vsrca, expand_alpha_rev ((Vector4x16)m), expand565(vd, 0));
>-		*dst = (ullong)pack565(vdest, (Vector4x16)c.mmx_zero, 0);
>+		__m64 vd = (__m64)d;
>+		__m64 vdest = in_over(vsrc, vsrca, expand_alpha_rev ((__m64)m), expand565(vd, 0));
>+		*dst = (ullong)pack565(vdest, _mm_setzero_si64(), 0);
> 	    }
> 	    
> 	    w--;
>@@ -819,21 +789,21 @@ fbCompositeSolidMask_nx8x0565mmx (CARD8 
> 	    }
> 	    else if (m0 | m1 | m2 | m3)
> 	    {
>-		Vector4x16 vdest;
>-		Vector4x16 vm0, vm1, vm2, vm3;
>+		__m64 vdest;
>+		__m64 vm0, vm1, vm2, vm3;
> 
>-		vdest = *(Vector4x16 *)dst;
>+		vdest = *(__m64 *)dst;
> 
>-		vm0 = (Vector4x16)m0;
>+		vm0 = (__m64)m0;
> 		vdest = pack565(in_over(vsrc, vsrca, expand_alpha_rev(vm0), expand565(vdest, 0)), vdest, 0);
>-		vm1 = (Vector4x16)m1;
>+		vm1 = (__m64)m1;
> 		vdest = pack565(in_over(vsrc, vsrca, expand_alpha_rev(vm1), expand565(vdest, 1)), vdest, 1);
>-		vm2 = (Vector4x16)m2;
>+		vm2 = (__m64)m2;
> 		vdest = pack565(in_over(vsrc, vsrca, expand_alpha_rev(vm2), expand565(vdest, 2)), vdest, 2);
>-		vm3 = (Vector4x16)m3;
>+		vm3 = (__m64)m3;
> 		vdest = pack565(in_over(vsrc, vsrca, expand_alpha_rev(vm3), expand565(vdest, 3)), vdest, 3);
> 		
>-		*(Vector4x16 *)dst = vdest;
>+		*(__m64 *)dst = vdest;
> 	    }
> 	    
> 	    w -= 4;
>@@ -850,9 +820,9 @@ fbCompositeSolidMask_nx8x0565mmx (CARD8 
> 	    if (m)
> 	    {
> 		ullong d = *dst;
>-		Vector4x16 vd = (Vector4x16)d;
>-		Vector4x16 vdest = in_over(vsrc, vsrca, expand_alpha_rev ((Vector4x16)m), expand565(vd, 0));
>-		*dst = (ullong)pack565(vdest, (Vector4x16)c.mmx_zero, 0);
>+		__m64 vd = (__m64)d;
>+		__m64 vdest = in_over(vsrc, vsrca, expand_alpha_rev ((__m64)m), expand565(vd, 0));
>+		*dst = (ullong)pack565(vdest, _mm_setzero_si64(), 0);
> 	    }
> 	    
> 	    w--;
>@@ -861,7 +831,7 @@ fbCompositeSolidMask_nx8x0565mmx (CARD8 
> 	}
>     }
>     
>-    emms();
>+    _mm_empty();
> }
> 
> void
>@@ -902,9 +872,9 @@ fbCompositeSrc_8888RevNPx0565mmx (CARD8 
> 	
> 	while (w && (unsigned long)dst & 7)
> 	{
>-	    Vector4x16 vsrc = load8888 (*src);
>+	    __m64 vsrc = load8888 (*src);
> 	    ullong d = *dst;
>-	    Vector4x16 vdest = expand565 ((Vector4x16)d, 0);
>+	    __m64 vdest = expand565 ((__m64)d, 0);
> 	    
> 	    vdest = pack565(over_rev_non_pre(vsrc, vdest), vdest, 0);
> 	    
>@@ -934,24 +904,24 @@ fbCompositeSrc_8888RevNPx0565mmx (CARD8 
> 	    
> 	    if ((a0 & a1 & a2 & a3) == 0xFF)
> 	    {
>-		Vector4x16 vdest;
>-		vdest = pack565(invert_colors(load8888(s0)), (Vector4x16)c.mmx_zero, 0);
>+		__m64 vdest;
>+		vdest = pack565(invert_colors(load8888(s0)), _mm_setzero_si64(), 0);
> 		vdest = pack565(invert_colors(load8888(s1)), vdest, 1);
> 		vdest = pack565(invert_colors(load8888(s2)), vdest, 2);
> 		vdest = pack565(invert_colors(load8888(s3)), vdest, 3);
> 
>-		*(Vector4x16 *)dst = vdest;
>+		*(__m64 *)dst = vdest;
> 	    }
> 	    else if (a0 | a1 | a2 | a3)
> 	    {
>-		Vector4x16 vdest = *(Vector4x16 *)dst;
>+		__m64 vdest = *(__m64 *)dst;
> 
> 		vdest = pack565(over_rev_non_pre(load8888(s0), expand565(vdest, 0)), vdest, 0);
> 	        vdest = pack565(over_rev_non_pre(load8888(s1), expand565(vdest, 1)), vdest, 1);
> 		vdest = pack565(over_rev_non_pre(load8888(s2), expand565(vdest, 2)), vdest, 2);
> 		vdest = pack565(over_rev_non_pre(load8888(s3), expand565(vdest, 3)), vdest, 3);
> 
>-		*(Vector4x16 *)dst = vdest;
>+		*(__m64 *)dst = vdest;
> 	    }
> 	    
> 	    w -= 4;
>@@ -963,9 +933,9 @@ fbCompositeSrc_8888RevNPx0565mmx (CARD8 
> 	
> 	while (w)
> 	{
>-	    Vector4x16 vsrc = load8888 (*src);
>+	    __m64 vsrc = load8888 (*src);
> 	    ullong d = *dst;
>-	    Vector4x16 vdest = expand565 ((Vector4x16)d, 0);
>+	    __m64 vdest = expand565 ((__m64)d, 0);
> 	    
> 	    vdest = pack565(over_rev_non_pre(vsrc, vdest), vdest, 0);
> 	    
>@@ -977,7 +947,7 @@ fbCompositeSrc_8888RevNPx0565mmx (CARD8 
> 	}
>     }
> 
>-    emms();
>+    _mm_empty();
> }
> 
> /* "888RevNP" is GdkPixbuf's format: ABGR, non premultiplied */
>@@ -1018,10 +988,10 @@ fbCompositeSrc_8888RevNPx8888mmx (CARD8 
> 
> 	while (w && (unsigned long)dst & 7)
> 	{
>-	    Vector4x16 s = load8888 (*src);
>-	    Vector4x16 d = load8888 (*dst);
>+	    __m64 s = load8888 (*src);
>+	    __m64 d = load8888 (*dst);
> 	    
>-	    *dst = (ullong)pack8888 (over_rev_non_pre (s, d), (Vector4x16)c.mmx_zero);
>+	    *dst = (ullong)pack8888 (over_rev_non_pre (s, d), _mm_setzero_si64());
> 	    
> 	    w--;
> 	    dst++;
>@@ -1032,7 +1002,7 @@ fbCompositeSrc_8888RevNPx8888mmx (CARD8 
> 	{
> 	    ullong s0, s1;
> 	    unsigned char a0, a1;
>-	    Vector4x16 d0, d1;
>+	    __m64 d0, d1;
> 
> 	    s0 = *src;
> 	    s1 = *(src + 1);
>@@ -1045,16 +1015,16 @@ fbCompositeSrc_8888RevNPx8888mmx (CARD8 
> 		d0 = invert_colors(load8888(s0));
> 		d1 = invert_colors(load8888(s1));
> 
>-		*(Vector8x8 *)dst = pack8888 (d0, d1);
>+		*(__m64 *)dst = pack8888 (d0, d1);
> 	    }
> 	    else if (a0 | a1)
> 	    {
>-		Vector4x16 vdest = *(Vector4x16 *)dst;
>+		__m64 vdest = *(__m64 *)dst;
> 
> 		d0 = over_rev_non_pre (load8888(s0), expand8888 (vdest, 0));
> 		d1 = over_rev_non_pre (load8888(s1), expand8888 (vdest, 1));
> 	    
>-		*(Vector8x8 *)dst = pack8888 (d0, d1);
>+		*(__m64 *)dst = pack8888 (d0, d1);
> 	    }
> 	    
> 	    w -= 2;
>@@ -1064,10 +1034,10 @@ fbCompositeSrc_8888RevNPx8888mmx (CARD8 
> 	
> 	while (w)
> 	{
>-	    Vector4x16 s = load8888 (*src);
>-	    Vector4x16 d = load8888 (*dst);
>+	    __m64 s = load8888 (*src);
>+	    __m64 d = load8888 (*dst);
> 	    
>-	    *dst = (ullong)pack8888 (over_rev_non_pre (s, d), (Vector4x16)c.mmx_zero);
>+	    *dst = (ullong)pack8888 (over_rev_non_pre (s, d), _mm_setzero_si64());
> 	    
> 	    w--;
> 	    dst++;
>@@ -1075,7 +1045,7 @@ fbCompositeSrc_8888RevNPx8888mmx (CARD8 
> 	}
>     }
> 
>-    emms();
>+    _mm_empty();
> }
> 
> void
>@@ -1096,7 +1066,7 @@ fbCompositeSolidMask_nx8888x0565Cmmx (CA
>     CARD16	*dstLine;
>     CARD32	*maskLine;
>     FbStride	dstStride, maskStride;
>-    Vector4x16  vsrc, vsrca;
>+    __m64  vsrc, vsrca;
>     
>     CHECKPOINT();
>     
>@@ -1125,7 +1095,7 @@ fbCompositeSolidMask_nx8888x0565Cmmx (CA
> 	    if (m)
> 	    {
> 		ullong d = *q;
>-		Vector4x16 vdest = expand565 ((Vector4x16)d, 0);
>+		__m64 vdest = expand565 ((__m64)d, 0);
> 		vdest = pack565 (in_over (vsrc, vsrca, load8888 (m), vdest), vdest, 0);
> 		*q = (ullong)vdest;
> 	    }
>@@ -1146,14 +1116,14 @@ fbCompositeSolidMask_nx8888x0565Cmmx (CA
> 	    
> 	    if ((m0 | m1 | m2 | m3))
> 	    {
>-		Vector4x16 vdest = *(Vector4x16 *)q;
>+		__m64 vdest = *(__m64 *)q;
> 		
> 		vdest = pack565(in_over(vsrc, vsrca, load8888(m0), expand565(vdest, 0)), vdest, 0);
> 		vdest = pack565(in_over(vsrc, vsrca, load8888(m1), expand565(vdest, 1)), vdest, 1);
> 		vdest = pack565(in_over(vsrc, vsrca, load8888(m2), expand565(vdest, 2)), vdest, 2);
> 		vdest = pack565(in_over(vsrc, vsrca, load8888(m3), expand565(vdest, 3)), vdest, 3);
> 		
>-		*(Vector4x16 *)q = vdest;
>+		*(__m64 *)q = vdest;
> 	    }
> 	    twidth -= 4;
> 	    p += 4;
>@@ -1168,7 +1138,7 @@ fbCompositeSolidMask_nx8888x0565Cmmx (CA
> 	    if (m)
> 	    {
> 		ullong d = *q;
>-		Vector4x16 vdest = expand565((Vector4x16)d, 0);
>+		__m64 vdest = expand565((__m64)d, 0);
> 		vdest = pack565 (in_over(vsrc, vsrca, load8888(m), vdest), vdest, 0);
> 		*q = (ullong)vdest;
> 	    }
>@@ -1182,7 +1152,7 @@ fbCompositeSolidMask_nx8888x0565Cmmx (CA
> 	dstLine += dstStride;
>     }
>     
>-    emms ();
>+    _mm_empty ();
> }
> 
> void
>@@ -1234,13 +1204,7 @@ fbCompositeSrcAdd_8000x8000mmx (CARD8	op
> 	
> 	while (w >= 8)
> 	{
>-	    __asm__ __volatile__ (
>-		"movq (%0), %%mm2\n\t"
>-		"movq (%1), %%mm3\n\t"
>-		"paddusb %%mm2, %%mm3\n\t"
>-		"movq %%mm3, (%1)\n\t"
>-		: /* no output */ : "r" (src), "r" (dst));
>-	    
>+	    *(__m64*)dst = _mm_adds_pu8(*(__m64*)src, *(__m64*)dst);
> 	    dst += 8;
> 	    src += 8;
> 	    w -= 8;
>@@ -1260,7 +1224,7 @@ fbCompositeSrcAdd_8000x8000mmx (CARD8	op
> 	}
>     }
> 
>-    emms();
>+    _mm_empty();
> }
> 
> void
>@@ -1297,13 +1261,8 @@ fbCompositeSrcAdd_8888x8888mmx (CARD8		o
> 	
> 	while (w && (unsigned long)dst & 7)
> 	{
>-	    __asm__ __volatile__ (
>-		"movd %0, %%mm2\n\t"
>-		"movd %1, %%mm3\n\t"
>-		"paddusb %%mm2, %%mm3\n\t"
>-		"movd %%mm3, %1\n\t"
>-		: /* no output */ : "m" (*src), "m" (*dst));
>-	    
>+	    *dst = _mm_cvtsi64_si32(_mm_adds_pu8(_mm_cvtsi32_si64(*src),
>+						 _mm_cvtsi32_si64(*dst)));
> 	    dst++;
> 	    src++;
> 	    w--;
>@@ -1311,13 +1270,7 @@ fbCompositeSrcAdd_8888x8888mmx (CARD8		o
> 	
> 	while (w >= 2)
> 	{
>-	    __asm__ __volatile__ (
>-		"movq (%0), %%mm2\n\t"
>-		"movq (%1), %%mm3\n\t"
>-		"paddusb %%mm2, %%mm3\n\t"
>-		"movq %%mm3, (%1)\n\t"
>-		: /* no output */ : "r" (src), "r" (dst));
>-	    
>+	    *(ullong*)dst = (ullong) _mm_adds_pu8(*(__m64*)src, *(__m64*)dst);
> 	    dst += 2;
> 	    src += 2;
> 	    w -= 2;
>@@ -1325,16 +1278,13 @@ fbCompositeSrcAdd_8888x8888mmx (CARD8		o
> 	
> 	if (w)
> 	{
>-	    __asm__ __volatile__ (
>-		"movd %0, %%mm2\n\t"
>-		"movd %1, %%mm3\n\t"
>-		"paddusb %%mm2, %%mm3\n\t"
>-		"movd %%mm3, %1\n\t"
>-		: /* no output */ : "m" (*src), "m" (*dst));
>+	    *dst = _mm_cvtsi64_si32(_mm_adds_pu8(_mm_cvtsi32_si64(*src),
>+						 _mm_cvtsi32_si64(*dst)));
>+
> 	}
>     }
> 
>-    emms();
>+    _mm_empty();
> }
> 
> #define GetStart(drw,x,y,type,stride,line,bpp) {\
>@@ -1358,7 +1308,7 @@ fbSolidFillmmx (DrawablePtr	pDraw,
>     FbStride	stride;
>     int		bpp;
>     ullong	fill;
>-    Vector8x8	vfill;
>+    __m64	vfill;
>     CARD32	byte_width;
>     CARD8	*byte_line;
>     FbBits      *bits;
>@@ -1390,7 +1340,7 @@ fbSolidFillmmx (DrawablePtr	pDraw,
>     }
> 
>     fill = ((ullong)xor << 32) | xor;
>-    vfill = (Vector8x8)fill;
>+    vfill = (__m64)fill;
>     
>     while (height--)
>     {
>@@ -1406,7 +1356,7 @@ fbSolidFillmmx (DrawablePtr	pDraw,
> 	    d += 2;
> 	}
> 	
>-	while (w >= 4 && ((unsigned int)d & 7))
>+	while (w >= 4 && ((unsigned long)d & 7))
> 	{
> 	    *(CARD32 *)d = xor;
> 
>@@ -1416,18 +1366,15 @@ fbSolidFillmmx (DrawablePtr	pDraw,
> 	
> 	while (w >= 64)
> 	{
>-	    __asm__ __volatile  (
>-		"movq %0, (%1)\n\t"
>-		"movq %0, 8(%1)\n\t"
>-		"movq %0, 16(%1)\n\t"
>-		"movq %0, 24(%1)\n\t"
>-		"movq %0, 32(%1)\n\t"
>-		"movq %0, 40(%1)\n\t"
>-		"movq %0, 48(%1)\n\t"
>-		"movq %0, 56(%1)\n\t"
>-		: /* no output */
>-		: "y" (vfill), "r" (d)
>-		: "memory");
>+	    *(__m64*) (d +  0) = vfill;
>+	    *(__m64*) (d +  8) = vfill;
>+	    *(__m64*) (d + 16) = vfill;
>+	    *(__m64*) (d + 24) = vfill;
>+	    *(__m64*) (d + 32) = vfill;
>+	    *(__m64*) (d + 40) = vfill;
>+	    *(__m64*) (d + 48) = vfill;
>+	    *(__m64*) (d + 56) = vfill;
>+
> 	    w -= 64;
> 	    d += 64;
> 	}
>@@ -1446,10 +1393,11 @@ fbSolidFillmmx (DrawablePtr	pDraw,
> 	}
>     }
>     
>-    emms();
>+    _mm_empty();
>     return TRUE;
> }
> 
>+#ifndef __amd64__
> Bool
> fbHaveMMX (void)
> {
>@@ -1508,7 +1456,8 @@ fbHaveMMX (void)
>     
>     return mmx_present;
> }
>+#endif /* __amd64__ */
> 
> 
> #endif /* RENDER */
>-#endif /* USE_GCC34_MMX */
>+#endif /* USE_MMX */
>Index: xc/programs/Xserver/fb/fbmmx.h
>===================================================================
>RCS file: /cvs/xorg/xc/programs/Xserver/fb/fbmmx.h,v
>retrieving revision 1.1
>diff -u -p -r1.1 fbmmx.h
>--- xc/programs/Xserver/fb/fbmmx.h	22 Jul 2004 19:24:50 -0000	1.1
>+++ xc/programs/Xserver/fb/fbmmx.h	13 Aug 2004 04:13:50 -0000
>@@ -22,13 +22,19 @@
>  * 
>  * Based on work by Owen Taylor
>  */
>-#ifdef USE_GCC34_MMX
>+#ifdef USE_MMX
>+
>+#ifndef __amd64__
> Bool fbHaveMMX(void);
> #else
>-#define fbHaveMMX FALSE
>+#define fbHaveMMX() TRUE
>+#endif
>+
>+#else
>+#define fbHaveMMX() FALSE
> #endif
> 
>-#ifdef USE_GCC34_MMX
>+#ifdef USE_MMX
> 
> void fbCompositeSolidMask_nx8888x0565Cmmx (CARD8      op,
> 					   PicturePtr pSrc,
>@@ -157,4 +163,4 @@ Bool fbSolidFillmmx (DrawablePtr	pDraw,
> 		     int		height,
> 		     FbBits		xor);
> 
>-#endif /* USE_GCC34_MMX */
>+#endif /* USE_MMX */
>Index: xc/programs/Xserver/fb/fbpict.c
>===================================================================
>RCS file: /cvs/xorg/xc/programs/Xserver/fb/fbpict.c,v
>retrieving revision 1.5
>diff -u -p -r1.5 fbpict.c
>--- xc/programs/Xserver/fb/fbpict.c	6 Aug 2004 23:42:10 -0000	1.5
>+++ xc/programs/Xserver/fb/fbpict.c	13 Aug 2004 04:13:53 -0000
>@@ -877,7 +877,7 @@ fbComposite (CARD8      op,
> 			switch (pDst->format) {
> 			case PICT_r5g6b5:
> 			case PICT_b5g6r5:
>-#ifdef USE_GCC34_MMX
>+#ifdef USE_MMX
> 			    if (fbHaveMMX())
> 				func = fbCompositeSolidMask_nx8x0565mmx;
> 			    else
>@@ -892,7 +892,7 @@ fbComposite (CARD8      op,
> 			case PICT_x8r8g8b8:
> 			case PICT_a8b8g8r8:
> 			case PICT_x8b8g8r8:
>-#ifdef USE_GCC34_MMX
>+#ifdef USE_MMX
> 			    if (fbHaveMMX())
> 				func = fbCompositeSolidMask_nx8x8888mmx;
> 			    else
>@@ -906,7 +906,7 @@ fbComposite (CARD8      op,
> 			    switch (pDst->format) {
> 			    case PICT_a8r8g8b8:
> 			    case PICT_x8r8g8b8:
>-#ifdef USE_GCC34_MMX
>+#ifdef USE_MMX
> 				if (fbHaveMMX())
> 				    func = fbCompositeSolidMask_nx8888x8888Cmmx;
> 				else
>@@ -914,7 +914,7 @@ fbComposite (CARD8      op,
> 				    func = fbCompositeSolidMask_nx8888x8888C;
> 				break;
> 			    case PICT_r5g6b5:
>-#ifdef USE_GCC34_MMX
>+#ifdef USE_MMX
> 				if (fbHaveMMX())
> 				    func = fbCompositeSolidMask_nx8888x0565Cmmx;
> 				else
>@@ -929,7 +929,7 @@ fbComposite (CARD8      op,
> 			    switch (pDst->format) {
> 			    case PICT_a8b8g8r8:
> 			    case PICT_x8b8g8r8:
>-#ifdef USE_GCC34_MMX
>+#ifdef USE_MMX
> 				if (fbHaveMMX())
> 				    func = fbCompositeSolidMask_nx8888x8888Cmmx;
> 				else
>@@ -937,7 +937,7 @@ fbComposite (CARD8      op,
> 				    func = fbCompositeSolidMask_nx8888x8888C;
> 				break;
> 			    case PICT_b5g6r5:
>-#ifdef USE_GCC34_MMX
>+#ifdef USE_MMX
> 				if (fbHaveMMX())
> 				    func = fbCompositeSolidMask_nx8888x0565Cmmx;
> 				else
>@@ -978,13 +978,13 @@ fbComposite (CARD8      op,
> 			    switch (pDst->format) {
> 			    case PICT_a8r8g8b8:
> 			    case PICT_x8r8g8b8:
>-#ifdef USE_GCC34_MMX
>+#ifdef USE_MMX
> 				if (fbHaveMMX())
> 				    func = fbCompositeSrc_8888RevNPx8888mmx;
> #endif
> 				break;
> 			    case PICT_r5g6b5:
>-#ifdef USE_GCC34_MMX
>+#ifdef USE_MMX
> 				if (fbHaveMMX())
> 				    func = fbCompositeSrc_8888RevNPx0565mmx;
> #endif
>@@ -1000,13 +1000,13 @@ fbComposite (CARD8      op,
> 			    switch (pDst->format) {
> 			    case PICT_a8b8g8r8:
> 			    case PICT_x8b8g8r8:
>-#ifdef USE_GCC34_MMX
>+#ifdef USE_MMX
> 				if (fbHaveMMX())
> 				    func = fbCompositeSrc_8888RevNPx8888mmx;
> #endif
> 				break;
> 			    case PICT_r5g6b5:
>-#ifdef USE_GCC34_MMX
>+#ifdef USE_MMX
> 				if (fbHaveMMX())
> 				    func = fbCompositeSrc_8888RevNPx0565mmx;
> #endif
>@@ -1032,7 +1032,7 @@ fbComposite (CARD8      op,
> 		    switch (pDst->format) {
> 		    case PICT_a8r8g8b8:
> 		    case PICT_x8r8g8b8:
>-#ifdef USE_GCC34_MMX
>+#ifdef USE_MMX
> 			if (fbHaveMMX())
> 			{
> 			    srcRepeat = FALSE;
>@@ -1041,7 +1041,7 @@ fbComposite (CARD8      op,
> #endif
> 			break;
> 		    case PICT_r5g6b5:
>-#ifdef USE_GCC34_MMX
>+#ifdef USE_MMX
> 			if (fbHaveMMX())
> 			{
> 			    srcRepeat = FALSE;
>@@ -1109,7 +1109,7 @@ fbComposite (CARD8      op,
> 	    case PICT_a8r8g8b8:
> 		switch (pDst->format) {
> 		case PICT_a8r8g8b8:
>-#ifdef USE_GCC34_MMX
>+#ifdef USE_MMX
> 		    if (fbHaveMMX())
> 			func = fbCompositeSrcAdd_8888x8888mmx;
> 		    else
>@@ -1121,7 +1121,7 @@ fbComposite (CARD8      op,
> 	    case PICT_a8b8g8r8:
> 		switch (pDst->format) {
> 		case PICT_a8b8g8r8:
>-#ifdef USE_GCC34_MMX
>+#ifdef USE_MMX
> 		    if (fbHaveMMX())
> 			func = fbCompositeSrcAdd_8888x8888mmx;
> 		    else
>@@ -1133,7 +1133,7 @@ fbComposite (CARD8      op,
> 	    case PICT_a8:
> 		switch (pDst->format) {
> 		case PICT_a8:
>-#ifdef USE_GCC34_MMX
>+#ifdef USE_MMX
> 		    if (fbHaveMMX())
> 			func = fbCompositeSrcAdd_8000x8000mmx;
> 		    else

Actions: View | Splinter Review

Attachments on bug 1067: 624 | 1591

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.