summaryrefslogtreecommitdiffstats
path: root/mpeglib/lib/util/mmx
diff options
context:
space:
mode:
Diffstat (limited to 'mpeglib/lib/util/mmx')
-rw-r--r--mpeglib/lib/util/mmx/Makefile.am46
-rw-r--r--mpeglib/lib/util/mmx/attribute.h33
-rw-r--r--mpeglib/lib/util/mmx/cpu_accel.c162
-rw-r--r--mpeglib/lib/util/mmx/mm_accel.h58
-rw-r--r--mpeglib/lib/util/mmx/mmx.c73
-rw-r--r--mpeglib/lib/util/mmx/mmx.h26
-rw-r--r--mpeglib/lib/util/mmx/mmx_asm.h258
7 files changed, 656 insertions, 0 deletions
diff --git a/mpeglib/lib/util/mmx/Makefile.am b/mpeglib/lib/util/mmx/Makefile.am
new file mode 100644
index 00000000..ce8875e0
--- /dev/null
+++ b/mpeglib/lib/util/mmx/Makefile.am
@@ -0,0 +1,46 @@
+# ---- @OS_TYPE@/@ARCH_TYPE@ ----
+
+# For cpu_accel compile we cannot have ansi
+# (I dont have debugged why)
+
+INCLUDES = $(all_includes)
+
+noinst_LTLIBRARIES = libmmx.la
+noinst_HEADERS =
+
+kmpgincludedir = $(includedir)/$(THIS_LIB_NAME)/util/mmx
+
+kmpginclude_HEADERS = mmx.h mmx_asm.h mm_accel.h
+
+libmmx_la_SOURCES = cpu_accel.c mmx.c
+
+AM_ASFLAGS = $(DEFS) $(DEFAULT_INCLUDES) $(all_includes)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
diff --git a/mpeglib/lib/util/mmx/attribute.h b/mpeglib/lib/util/mmx/attribute.h
new file mode 100644
index 00000000..4dfd3e1e
--- /dev/null
+++ b/mpeglib/lib/util/mmx/attribute.h
@@ -0,0 +1,33 @@
+/*
+ align attribut definition (g++)
+ Copyright (C) 2001 Martin Vogt
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU Library General Public License as published by
+ the Free Software Foundation.
+
+ For more information look at the file COPYRIGHT in this package
+
+ */
+
+
+#ifndef __ATTRIBUTE_H
+#define __ATTRIBUTE_H
+
+
+
+#include "vidi/vidi_config.h"
+
+
+/* use gcc attribs to align critical data structures */
+
+#ifdef ATTRIBUTE_ALIGNED_MAX
+#define ATTR_ALIGN(align) __attribute__ \
+ ((__aligned__ ((ATTRIBUTE_ALIGNED_MAX <align) ? \
+ ATTRIBUTE_ALIGNED_MAX : align)))
+#else
+#define ATTR_ALIGN(align)
+#endif
+
+
+#endif
diff --git a/mpeglib/lib/util/mmx/cpu_accel.c b/mpeglib/lib/util/mmx/cpu_accel.c
new file mode 100644
index 00000000..8f138308
--- /dev/null
+++ b/mpeglib/lib/util/mmx/cpu_accel.c
@@ -0,0 +1,162 @@
+/*
+ * cpu_accel.c
+ * Copyright (C) 2000-2001 Michel Lespinasse <walken@zoy.org>
+ * Copyright (C) 1999-2000 Aaron Holtzman <aholtzma@ess.engr.uvic.ca>
+ *
+ * This file is part of mpeg2dec, a free MPEG-2 video stream decoder.
+ * See http://libmpeg2.sourceforge.net/ for updates.
+ *
+ * mpeg2dec is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * mpeg2dec is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#include "config.h"
+
+#if defined(MMX_SUPPORT) || defined(INTEL)
+#define ARCH_X86 1
+#endif
+
+#include <inttypes.h>
+
+#include "mm_accel.h"
+
+#ifdef ARCH_X86
+static uint32_t arch_accel (void)
+{
+ uint32_t eax, ebx, ecx, edx;
+ int AMD;
+ uint32_t caps;
+
+#ifndef PIC
+#define cpuid(op,eax,ebx,ecx,edx) \
+ __asm__ ("cpuid" \
+ : "=a" (eax), \
+ "=b" (ebx), \
+ "=c" (ecx), \
+ "=d" (edx) \
+ : "a" (op) \
+ : "cc")
+#else /* PIC version : save ebx */
+#define cpuid(op,eax,ebx,ecx,edx) \
+ __asm__ ("pushl %%ebx\n\t" \
+ "cpuid\n\t" \
+ "movl %%ebx,%1\n\t" \
+ "popl %%ebx" \
+ : "=a" (eax), \
+ "=r" (ebx), \
+ "=c" (ecx), \
+ "=d" (edx) \
+ : "a" (op) \
+ : "cc")
+#endif
+
+ __asm__ ("pushfl\n\t"
+ "pushfl\n\t"
+ "popl %0\n\t"
+ "movl %0,%1\n\t"
+ "xorl $0x200000,%0\n\t"
+ "pushl %0\n\t"
+ "popfl\n\t"
+ "pushfl\n\t"
+ "popl %0\n\t"
+ "popfl"
+ : "=r" (eax),
+ "=r" (ebx)
+ :
+ : "cc");
+
+ if (eax == ebx) /* no cpuid */
+ return 0;
+
+ cpuid (0x00000000, eax, ebx, ecx, edx);
+ if (!eax) /* vendor string only */
+ return 0;
+
+ AMD = (ebx == 0x68747541) && (ecx == 0x444d4163) && (edx == 0x69746e65);
+
+ cpuid (0x00000001, eax, ebx, ecx, edx);
+ if (! (edx & 0x00800000)) /* no MMX */
+ return 0;
+
+ caps = MM_ACCEL_X86_MMX;
+ if (edx & 0x02000000) /* SSE - identical to AMD MMX extensions */
+ caps = MM_ACCEL_X86_MMX | MM_ACCEL_X86_MMXEXT;
+
+ cpuid (0x80000000, eax, ebx, ecx, edx);
+ if (eax < 0x80000001) /* no extended capabilities */
+ return caps;
+
+ cpuid (0x80000001, eax, ebx, ecx, edx);
+
+ if (edx & 0x80000000)
+ caps |= MM_ACCEL_X86_3DNOW;
+
+ if (AMD && (edx & 0x00400000)) /* AMD MMX extensions */
+ caps |= MM_ACCEL_X86_MMXEXT;
+
+ return caps;
+}
+#endif /* ARCH_X86 */
+
+#ifdef ARCH_PPC
+#include <signal.h>
+#include <setjmp.h>
+
+static sigjmp_buf jmpbuf;
+static volatile sig_atomic_t canjump = 0;
+
+static RETSIGTYPE sigill_handler (int sig)
+{
+ if (!canjump) {
+ signal (sig, SIG_DFL);
+ raise (sig);
+ }
+
+ canjump = 0;
+ siglongjmp (jmpbuf, 1);
+}
+
+static uint32_t arch_accel (void)
+{
+ signal (SIGILL, sigill_handler);
+ if (sigsetjmp (jmpbuf, 1)) {
+ signal (SIGILL, SIG_DFL);
+ return 0;
+ }
+
+ canjump = 1;
+
+ __asm__ volatile ("mtspr 256,%0" :: "r" (-1));
+
+ signal (SIGILL, SIG_DFL);
+ return MM_ACCEL_PPC_ALTIVEC;
+}
+#endif /* ARCH_PPC */
+
+uint32_t mm_accel (void)
+{
+#if defined (ARCH_X86) || defined (ARCH_PPC)
+ static int got_accel = 0;
+ static uint32_t accel;
+
+ if (!got_accel) {
+ got_accel = 1;
+ accel = arch_accel ();
+ }
+
+ return accel;
+#else
+ return 0;
+#endif
+}
diff --git a/mpeglib/lib/util/mmx/mm_accel.h b/mpeglib/lib/util/mmx/mm_accel.h
new file mode 100644
index 00000000..e0690804
--- /dev/null
+++ b/mpeglib/lib/util/mmx/mm_accel.h
@@ -0,0 +1,58 @@
+/*
+ * mm_accel.h
+ * Copyright (C) 2000-2001 Michel Lespinasse <walken@zoy.org>
+ * Copyright (C) 1999-2000 Aaron Holtzman <aholtzma@ess.engr.uvic.ca>
+ *
+ * This file is part of mpeg2dec, a free MPEG-2 video stream decoder.
+ * See http://libmpeg2.sourceforge.net/ for updates.
+ *
+ * mpeg2dec is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * mpeg2dec is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#ifndef MM_ACCEL_H
+#define MM_ACCEL_H
+
+#include <inttypes.h>
+
+#ifdef __cplusplus
+extern "C"
+{
+#endif /* __cplusplus */
+
+
+/* generic accelerations */
+#define MM_ACCEL_MLIB 0x00000001
+
+/* x86 accelerations */
+#define MM_ACCEL_X86_MMX 0x80000000
+#define MM_ACCEL_X86_3DNOW 0x40000000
+#define MM_ACCEL_X86_MMXEXT 0x20000000
+
+/* powerpc accelerations */
+#define MM_ACCEL_PPC_ALTIVEC 0x80000000
+
+/* detailed SIMD info */
+uint32_t mm_accel (void);
+
+/* true if MMX support */
+int mm_support();
+
+#ifdef __cplusplus
+}
+#endif /* __cplusplus */
+
+
+#endif
+/* MM_ACCEL_H */
diff --git a/mpeglib/lib/util/mmx/mmx.c b/mpeglib/lib/util/mmx/mmx.c
new file mode 100644
index 00000000..4ac3d6b8
--- /dev/null
+++ b/mpeglib/lib/util/mmx/mmx.c
@@ -0,0 +1,73 @@
+/*
+ wrapper for MMX calls
+ Copyright (C) 2000 Martin Vogt
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU Library General Public License as published by
+ the Free Software Foundation.
+
+ For more information look at the file COPYRIGHT in this package
+
+ */
+
+
+
+#include "mmx.h"
+
+static int mmSupport=-1;
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#if defined(MMX_SUPPORT) || defined(INTEL)
+#define HAVE_ACCEL
+#endif
+
+
+#ifndef HAVE_ACCEL
+int emms() {
+ printf("urgs! emms() never should happen\n");
+ exit(0);
+}
+
+int mm_support() {
+ return 0;
+}
+
+#endif
+
+
+#ifdef HAVE_ACCEL
+#include <inttypes.h>
+#include "mm_accel.h"
+
+
+
+
+int emms() {
+ __asm__ ("emms");
+ return 1;
+}
+
+
+int mm_support() {
+
+ int val;
+
+ if (mmSupport == -1) {
+
+ val=mm_accel();
+ if (val & MM_ACCEL_X86_MMX) {
+ mmSupport=1;
+ } else {
+ mmSupport=0;
+ }
+
+ }
+ /* Return */
+ return(mmSupport);
+}
+
+
+#endif
diff --git a/mpeglib/lib/util/mmx/mmx.h b/mpeglib/lib/util/mmx/mmx.h
new file mode 100644
index 00000000..c4b8340d
--- /dev/null
+++ b/mpeglib/lib/util/mmx/mmx.h
@@ -0,0 +1,26 @@
+
+
+
+#ifndef __MMX_H
+#define __MMX_H
+
+
+#include <stdio.h>
+#include <stdlib.h>
+
+#ifdef __cplusplus
+extern "C"
+{
+#endif /* __cplusplus */
+
+
+extern int emms();
+extern int mm_support();
+
+
+#ifdef __cplusplus
+}
+#endif /* __cplusplus */
+
+
+#endif
diff --git a/mpeglib/lib/util/mmx/mmx_asm.h b/mpeglib/lib/util/mmx/mmx_asm.h
new file mode 100644
index 00000000..8717eff5
--- /dev/null
+++ b/mpeglib/lib/util/mmx/mmx_asm.h
@@ -0,0 +1,258 @@
+/*
+ * mmx.h
+ * Copyright (C) 1997-2001 H. Dietz and R. Fisher
+ *
+ * This file is part of mpeg2dec, a free MPEG-2 video stream decoder.
+ *
+ * mpeg2dec is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ */
+
+/*
+ * The type of an value that fits in an MMX register (note that long
+ * long constant values MUST be suffixed by LL and unsigned long long
+ * values by ULL, lest they be truncated by the compiler)
+ */
+
+#ifndef __MMX_ASM_H
+#define __MMX_ASM_H
+
+#include "attribute.h"
+
+
+typedef union {
+ long long q; /* Quadword (64-bit) value */
+ unsigned long long uq; /* Unsigned Quadword */
+ int d[2]; /* 2 Doubleword (32-bit) values */
+ unsigned int ud[2]; /* 2 Unsigned Doubleword */
+ short w[4]; /* 4 Word (16-bit) values */
+ unsigned short uw[4]; /* 4 Unsigned Word */
+ char b[8]; /* 8 Byte (8-bit) values */
+ unsigned char ub[8]; /* 8 Unsigned Byte */
+ float s[2]; /* Single-precision (32-bit) value */
+} ATTR_ALIGN(8) mmx_t; /* On an 8-byte (64-bit) boundary */
+
+
+#define mmx_i2r(op,imm,reg) \
+ __asm__ __volatile__ (#op " %0, %%" #reg \
+ : /* nothing */ \
+ : "X" (imm) )
+
+#define mmx_m2r(op,mem,reg) \
+ __asm__ __volatile__ (#op " %0, %%" #reg \
+ : /* nothing */ \
+ : "X" (mem))
+
+#define mmx_r2m(op,reg,mem) \
+ __asm__ __volatile__ (#op " %%" #reg ", %0" \
+ : "=X" (mem) \
+ : /* nothing */ )
+
+#define mmx_r2r(op,regs,regd) \
+ __asm__ __volatile__ (#op " %" #regs ", %" #regd)
+
+
+#define emms() __asm__ __volatile__ ("emms")
+
+#define movd_m2r(var,reg) mmx_m2r (movd, var, reg)
+#define movd_r2m(reg,var) mmx_r2m (movd, reg, var)
+#define movd_r2r(regs,regd) mmx_r2r (movd, regs, regd)
+
+#define movq_m2r(var,reg) mmx_m2r (movq, var, reg)
+#define movq_r2m(reg,var) mmx_r2m (movq, reg, var)
+#define movq_r2r(regs,regd) mmx_r2r (movq, regs, regd)
+
+#define packssdw_m2r(var,reg) mmx_m2r (packssdw, var, reg)
+#define packssdw_r2r(regs,regd) mmx_r2r (packssdw, regs, regd)
+#define packsswb_m2r(var,reg) mmx_m2r (packsswb, var, reg)
+#define packsswb_r2r(regs,regd) mmx_r2r (packsswb, regs, regd)
+
+#define packuswb_m2r(var,reg) mmx_m2r (packuswb, var, reg)
+#define packuswb_r2r(regs,regd) mmx_r2r (packuswb, regs, regd)
+
+#define paddb_m2r(var,reg) mmx_m2r (paddb, var, reg)
+#define paddb_r2r(regs,regd) mmx_r2r (paddb, regs, regd)
+#define paddd_m2r(var,reg) mmx_m2r (paddd, var, reg)
+#define paddd_r2r(regs,regd) mmx_r2r (paddd, regs, regd)
+#define paddw_m2r(var,reg) mmx_m2r (paddw, var, reg)
+#define paddw_r2r(regs,regd) mmx_r2r (paddw, regs, regd)
+
+#define paddsb_m2r(var,reg) mmx_m2r (paddsb, var, reg)
+#define paddsb_r2r(regs,regd) mmx_r2r (paddsb, regs, regd)
+#define paddsw_m2r(var,reg) mmx_m2r (paddsw, var, reg)
+#define paddsw_r2r(regs,regd) mmx_r2r (paddsw, regs, regd)
+
+#define paddusb_m2r(var,reg) mmx_m2r (paddusb, var, reg)
+#define paddusb_r2r(regs,regd) mmx_r2r (paddusb, regs, regd)
+#define paddusw_m2r(var,reg) mmx_m2r (paddusw, var, reg)
+#define paddusw_r2r(regs,regd) mmx_r2r (paddusw, regs, regd)
+
+#define pand_m2r(var,reg) mmx_m2r (pand, var, reg)
+#define pand_r2r(regs,regd) mmx_r2r (pand, regs, regd)
+
+#define pandn_m2r(var,reg) mmx_m2r (pandn, var, reg)
+#define pandn_r2r(regs,regd) mmx_r2r (pandn, regs, regd)
+
+#define pcmpeqb_m2r(var,reg) mmx_m2r (pcmpeqb, var, reg)
+#define pcmpeqb_r2r(regs,regd) mmx_r2r (pcmpeqb, regs, regd)
+#define pcmpeqd_m2r(var,reg) mmx_m2r (pcmpeqd, var, reg)
+#define pcmpeqd_r2r(regs,regd) mmx_r2r (pcmpeqd, regs, regd)
+#define pcmpeqw_m2r(var,reg) mmx_m2r (pcmpeqw, var, reg)
+#define pcmpeqw_r2r(regs,regd) mmx_r2r (pcmpeqw, regs, regd)
+
+#define pcmpgtb_m2r(var,reg) mmx_m2r (pcmpgtb, var, reg)
+#define pcmpgtb_r2r(regs,regd) mmx_r2r (pcmpgtb, regs, regd)
+#define pcmpgtd_m2r(var,reg) mmx_m2r (pcmpgtd, var, reg)
+#define pcmpgtd_r2r(regs,regd) mmx_r2r (pcmpgtd, regs, regd)
+#define pcmpgtw_m2r(var,reg) mmx_m2r (pcmpgtw, var, reg)
+#define pcmpgtw_r2r(regs,regd) mmx_r2r (pcmpgtw, regs, regd)
+
+#define pmaddwd_m2r(var,reg) mmx_m2r (pmaddwd, var, reg)
+#define pmaddwd_r2r(regs,regd) mmx_r2r (pmaddwd, regs, regd)
+
+#define pmulhw_m2r(var,reg) mmx_m2r (pmulhw, var, reg)
+#define pmulhw_r2r(regs,regd) mmx_r2r (pmulhw, regs, regd)
+
+#define pmullw_m2r(var,reg) mmx_m2r (pmullw, var, reg)
+#define pmullw_r2r(regs,regd) mmx_r2r (pmullw, regs, regd)
+
+#define por_m2r(var,reg) mmx_m2r (por, var, reg)
+#define por_r2r(regs,regd) mmx_r2r (por, regs, regd)
+
+#define pslld_i2r(imm,reg) mmx_i2r (pslld, imm, reg)
+#define pslld_m2r(var,reg) mmx_m2r (pslld, var, reg)
+#define pslld_r2r(regs,regd) mmx_r2r (pslld, regs, regd)
+#define psllq_i2r(imm,reg) mmx_i2r (psllq, imm, reg)
+#define psllq_m2r(var,reg) mmx_m2r (psllq, var, reg)
+#define psllq_r2r(regs,regd) mmx_r2r (psllq, regs, regd)
+#define psllw_i2r(imm,reg) mmx_i2r (psllw, imm, reg)
+#define psllw_m2r(var,reg) mmx_m2r (psllw, var, reg)
+#define psllw_r2r(regs,regd) mmx_r2r (psllw, regs, regd)
+
+#define psrad_i2r(imm,reg) mmx_i2r (psrad, imm, reg)
+#define psrad_m2r(var,reg) mmx_m2r (psrad, var, reg)
+#define psrad_r2r(regs,regd) mmx_r2r (psrad, regs, regd)
+#define psraw_i2r(imm,reg) mmx_i2r (psraw, imm, reg)
+#define psraw_m2r(var,reg) mmx_m2r (psraw, var, reg)
+#define psraw_r2r(regs,regd) mmx_r2r (psraw, regs, regd)
+
+#define psrld_i2r(imm,reg) mmx_i2r (psrld, imm, reg)
+#define psrld_m2r(var,reg) mmx_m2r (psrld, var, reg)
+#define psrld_r2r(regs,regd) mmx_r2r (psrld, regs, regd)
+#define psrlq_i2r(imm,reg) mmx_i2r (psrlq, imm, reg)
+#define psrlq_m2r(var,reg) mmx_m2r (psrlq, var, reg)
+#define psrlq_r2r(regs,regd) mmx_r2r (psrlq, regs, regd)
+#define psrlw_i2r(imm,reg) mmx_i2r (psrlw, imm, reg)
+#define psrlw_m2r(var,reg) mmx_m2r (psrlw, var, reg)
+#define psrlw_r2r(regs,regd) mmx_r2r (psrlw, regs, regd)
+
+#define psubb_m2r(var,reg) mmx_m2r (psubb, var, reg)
+#define psubb_r2r(regs,regd) mmx_r2r (psubb, regs, regd)
+#define psubd_m2r(var,reg) mmx_m2r (psubd, var, reg)
+#define psubd_r2r(regs,regd) mmx_r2r (psubd, regs, regd)
+#define psubw_m2r(var,reg) mmx_m2r (psubw, var, reg)
+#define psubw_r2r(regs,regd) mmx_r2r (psubw, regs, regd)
+
+#define psubsb_m2r(var,reg) mmx_m2r (psubsb, var, reg)
+#define psubsb_r2r(regs,regd) mmx_r2r (psubsb, regs, regd)
+#define psubsw_m2r(var,reg) mmx_m2r (psubsw, var, reg)
+#define psubsw_r2r(regs,regd) mmx_r2r (psubsw, regs, regd)
+
+#define psubusb_m2r(var,reg) mmx_m2r (psubusb, var, reg)
+#define psubusb_r2r(regs,regd) mmx_r2r (psubusb, regs, regd)
+#define psubusw_m2r(var,reg) mmx_m2r (psubusw, var, reg)
+#define psubusw_r2r(regs,regd) mmx_r2r (psubusw, regs, regd)
+
+#define punpckhbw_m2r(var,reg) mmx_m2r (punpckhbw, var, reg)
+#define punpckhbw_r2r(regs,regd) mmx_r2r (punpckhbw, regs, regd)
+#define punpckhdq_m2r(var,reg) mmx_m2r (punpckhdq, var, reg)
+#define punpckhdq_r2r(regs,regd) mmx_r2r (punpckhdq, regs, regd)
+#define punpckhwd_m2r(var,reg) mmx_m2r (punpckhwd, var, reg)
+#define punpckhwd_r2r(regs,regd) mmx_r2r (punpckhwd, regs, regd)
+
+#define punpcklbw_m2r(var,reg) mmx_m2r (punpcklbw, var, reg)
+#define punpcklbw_r2r(regs,regd) mmx_r2r (punpcklbw, regs, regd)
+#define punpckldq_m2r(var,reg) mmx_m2r (punpckldq, var, reg)
+#define punpckldq_r2r(regs,regd) mmx_r2r (punpckldq, regs, regd)
+#define punpcklwd_m2r(var,reg) mmx_m2r (punpcklwd, var, reg)
+#define punpcklwd_r2r(regs,regd) mmx_r2r (punpcklwd, regs, regd)
+
+#define pxor_m2r(var,reg) mmx_m2r (pxor, var, reg)
+#define pxor_r2r(regs,regd) mmx_r2r (pxor, regs, regd)
+
+
+/* 3DNOW extensions */
+
+#define pavgusb_m2r(var,reg) mmx_m2r (pavgusb, var, reg)
+#define pavgusb_r2r(regs,regd) mmx_r2r (pavgusb, regs, regd)
+
+
+/* AMD MMX extensions - also available in intel SSE */
+
+
+#define mmx_m2ri(op,mem,reg,imm) \
+ __asm__ __volatile__ (#op " %1, %0, %%" #reg \
+ : /* nothing */ \
+ : "X" (mem), "X" (imm))
+#define mmx_r2ri(op,regs,regd,imm) \
+ __asm__ __volatile__ (#op " %0, %%" #regs ", %%" #regd \
+ : /* nothing */ \
+ : "X" (imm) )
+
+#define mmx_fetch(mem,hint) \
+ __asm__ __volatile__ ("prefetch" #hint " %0" \
+ : /* nothing */ \
+ : "X" (mem))
+
+
+#define maskmovq(regs,maskreg) mmx_r2ri (maskmovq, regs, maskreg)
+
+#define movntq_r2m(mmreg,var) mmx_r2m (movntq, mmreg, var)
+
+#define pavgb_m2r(var,reg) mmx_m2r (pavgb, var, reg)
+#define pavgb_r2r(regs,regd) mmx_r2r (pavgb, regs, regd)
+#define pavgw_m2r(var,reg) mmx_m2r (pavgw, var, reg)
+#define pavgw_r2r(regs,regd) mmx_r2r (pavgw, regs, regd)
+
+#define pextrw_r2r(mmreg,reg,imm) mmx_r2ri (pextrw, mmreg, reg, imm)
+
+#define pinsrw_r2r(reg,mmreg,imm) mmx_r2ri (pinsrw, reg, mmreg, imm)
+
+#define pmaxsw_m2r(var,reg) mmx_m2r (pmaxsw, var, reg)
+#define pmaxsw_r2r(regs,regd) mmx_r2r (pmaxsw, regs, regd)
+
+#define pmaxub_m2r(var,reg) mmx_m2r (pmaxub, var, reg)
+#define pmaxub_r2r(regs,regd) mmx_r2r (pmaxub, regs, regd)
+
+#define pminsw_m2r(var,reg) mmx_m2r (pminsw, var, reg)
+#define pminsw_r2r(regs,regd) mmx_r2r (pminsw, regs, regd)
+
+#define pminub_m2r(var,reg) mmx_m2r (pminub, var, reg)
+#define pminub_r2r(regs,regd) mmx_r2r (pminub, regs, regd)
+
+#define pmovmskb(mmreg,reg) \
+ __asm__ __volatile__ ("movmskps %" #mmreg ", %" #reg)
+
+#define pmulhuw_m2r(var,reg) mmx_m2r (pmulhuw, var, reg)
+#define pmulhuw_r2r(regs,regd) mmx_r2r (pmulhuw, regs, regd)
+
+#define prefetcht0(mem) mmx_fetch (mem, t0)
+#define prefetcht1(mem) mmx_fetch (mem, t1)
+#define prefetcht2(mem) mmx_fetch (mem, t2)
+#define prefetchnta(mem) mmx_fetch (mem, nta)
+
+#define psadbw_m2r(var,reg) mmx_m2r (psadbw, var, reg)
+#define psadbw_r2r(regs,regd) mmx_r2r (psadbw, regs, regd)
+
+#define pshufw_m2r(var,reg,imm) mmx_m2ri(pshufw, var, reg, imm)
+#define pshufw_r2r(regs,regd,imm) mmx_r2ri(pshufw, regs, regd, imm)
+
+#define sfence() __asm__ __volatile__ ("sfence\n\t")
+
+
+#endif
+
+