Add macros for 64- and 128-bit write-combining optimization to intreadwrite.h.

Add x86 implementation using MMX/SSE.

Originally committed as revision 21281 to svn://svn.ffmpeg.org/ffmpeg/trunk

2 files changed, 141 insertions, 2 deletions

diff --git a/libavutil/intreadwrite.h b/libavutil/intreadwrite.h
index 78f83300bc..bdc3c4366a 100644
--- a/libavutil/intreadwrite.h
+++ b/libavutil/intreadwrite.h
@@ -25,8 +25,9 @@
 
 /*
  * Arch-specific headers can provide any combination of
- * AV_[RW][BLN](16|24|32|64) macros.  Preprocessor symbols must be
- * defined, even if these are implemented as inline functions.
+ * AV_[RW][BLN](16|24|32|64) and AV_(COPY|SWAP|ZERO)(64|128) macros.
+ * Preprocessor symbols must be defined, even if these are implemented
+ * as inline functions.
  */
 
 #if   ARCH_ARM
@@ -37,6 +38,8 @@
 #   include "mips/intreadwrite.h"
 #elif ARCH_PPC
 #   include "ppc/intreadwrite.h"
+#elif ARCH_X86
+#   include "x86/intreadwrite.h"
 #endif
 
 /*
@@ -397,4 +400,44 @@ struct unaligned_16 { uint16_t l; } __attribute__((packed));
     } while(0)
 #endif
 
+/* Parameters for AV_COPY*, AV_SWAP*, AV_ZERO* must be
+ * naturally aligned. They may be implemented using MMX,
+ * so emms_c() must be called before using any float code
+ * afterwards.
+ */
+
+#define AV_COPY(n, d, s) (*(uint##n##_t*)(d) = *(const uint##n##_t*)(s))
+
+#ifndef AV_COPY64
+#   define AV_COPY64(d, s) AV_COPY(64, d, s)
+#endif
+
+#ifndef AV_COPY128
+#   define AV_COPY128(d, s)                    \
+    do {                                       \
+        AV_COPY64(d, s);                       \
+        AV_COPY64((char*)(d)+8, (char*)(s)+8); \
+    } while(0)
+#endif
+
+#define AV_SWAP(n, a, b) FFSWAP(uint##n##_t, *(uint##n##_t*)(a), *(uint##n##_t*)(b))
+
+#ifndef AV_SWAP64
+#   define AV_SWAP64(a, b) AV_SWAP(64, a, b)
+#endif
+
+#define AV_ZERO(n, d) (*(uint##n##_t*)(d) = 0)
+
+#ifndef AV_ZERO64
+#   define AV_ZERO64(d) AV_ZERO(64, d)
+#endif
+
+#ifndef AV_ZERO128
+#   define AV_ZERO128(d)         \
+    do {                         \
+        AV_ZERO64(d);            \
+        AV_ZERO64((char*)(d)+8); \
+    } while(0)
+#endif
+
 #endif /* AVUTIL_INTREADWRITE_H */
diff --git a/libavutil/x86/intreadwrite.h b/libavutil/x86/intreadwrite.h
new file mode 100644
index 0000000000..4621f064d1
--- /dev/null
+++ b/libavutil/x86/intreadwrite.h
@@ -0,0 +1,96 @@
+/*
+ * Copyright (c) 2010 Alexander Strange <astrange@ithinksw.com>
+ *
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * FFmpeg 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
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with FFmpeg; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#ifndef AVUTIL_X86_INTREADWRITE_H
+#define AVUTIL_X86_INTREADWRITE_H
+
+#include <stdint.h>
+#include "config.h"
+
+#if HAVE_MMX
+
+#if !HAVE_FAST_64BIT && defined(__MMX__)
+
+#define AV_COPY64 AV_COPY64
+static av_always_inline void AV_COPY64(void *d, const void *s)
+{
+    __asm__("movq   %1, %%mm0  \n\t"
+            "movq   %%mm0, %0  \n\t"
+            : "=m"(*(uint64_t*)d)
+            : "m" (*(const uint64_t*)s)
+            : "mm0");
+}
+
+#define AV_SWAP64 AV_SWAP64
+static av_always_inline void AV_SWAP64(void *a, void *b)
+{
+    __asm__("movq   %1, %%mm0  \n\t"
+            "movq   %0, %%mm1  \n\t"
+            "movq   %%mm0, %0  \n\t"
+            "movq   %%mm1, %1  \n\t"
+            : "+m"(*(uint64_t*)a), "+m"(*(uint64_t*)b)
+            ::"mm0", "mm1");
+}
+
+#define AV_ZERO64 AV_ZERO64
+static av_always_inline void AV_ZERO64(void *d)
+{
+    __asm__("pxor %%mm0, %%mm0  \n\t"
+            "movq %%mm0, %0     \n\t"
+            : "=m"(*(uint64_t*)d)
+            :: "mm0");
+}
+
+#endif /* !HAVE_FAST_64BIT && defined(__MMX__) */
+
+#ifdef __SSE__
+
+#define AV_COPY128 AV_COPY128
+static av_always_inline void AV_COPY128(void *d, const void *s)
+{
+    struct v {uint64_t v[2];};
+
+    __asm__("movaps   %1, %%xmm0  \n\t"
+            "movaps   %%xmm0, %0  \n\t"
+            : "=m"(*(struct v*)d)
+            : "m" (*(const struct v*)s)
+            : "xmm0");
+}
+
+#endif /* __SSE__ */
+
+#ifdef __SSE2__
+
+#define AV_ZERO128 AV_ZERO128
+static av_always_inline void AV_ZERO128(void *d)
+{
+    struct v {uint64_t v[2];};
+
+    __asm__("pxor %%xmm0, %%xmm0  \n\t"
+            "movdqa   %%xmm0, %0  \n\t"
+            : "=m"(*(struct v*)d)
+            :: "xmm0");
+}
+
+#endif /* __SSE2__ */
+
+#endif /* HAVE_MMX */
+
+#endif /* AVUTIL_X86_INTREADWRITE_H */