Re: [libav-devel] [PATCH 06/14] Move away x86 specific code from rgb2rgb part 2
On Apr 11, 2011, at 5:26 AM, Kostya wrote:
On Mon, Apr 11, 2011 at 11:07:39AM +0200, Luca Barbato wrote:
With this commit we should have all the arch specific code moved away.
Init pattern now:
- generic C init first
- arch specific init later overwriting
In future the arch specific init will overwrite just the generics for
which we have an optimization.
[..]
+void rgb2rgb_init_x86(int flags);
is that requirement by modern GCC to declare function before its
implementation in the same file?
Requirement of the compiler flags (-Werror=missing-prototypes).
If it's only used in this file it should be static.
+void rgb2rgb_init_x86(int flags)
+{
+
+#if HAVE_MMX2 || HAVE_AMD3DNOW || HAVE_MMX
+if (flags SWS_CPU_CAPS_SSE2)
+rgb2rgb_init_SSE2();
+else if (flags SWS_CPU_CAPS_MMX2)
+rgb2rgb_init_MMX2();
+else if (flags SWS_CPU_CAPS_3DNOW)
+rgb2rgb_init_3DNOW();
+else if (flags SWS_CPU_CAPS_MMX)
+rgb2rgb_init_MMX();
+#endif /* HAVE_MMX2 || HAVE_AMD3DNOW || HAVE_MMX */
+}
--
1.7.4.1
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
[libav-devel] [PATCH 06/14] Move away x86 specific code from rgb2rgb part 2
With this commit we should have all the arch specific code moved away. Init pattern now: - generic C init first - arch specific init later overwriting In future the arch specific init will overwrite just the generics for which we have an optimization. --- libswscale/Makefile |3 +- libswscale/rgb2rgb.c | 94 + libswscale/x86/rgb2rgb.c | 146 ++ 3 files changed, 152 insertions(+), 91 deletions(-) create mode 100644 libswscale/x86/rgb2rgb.c diff --git a/libswscale/Makefile b/libswscale/Makefile index 3ac60e5..b8f233f 100644 --- a/libswscale/Makefile +++ b/libswscale/Makefile @@ -12,7 +12,8 @@ OBJS-$(ARCH_BFIN) += bfin/internal_bfin.o \ bfin/yuv2rgb_bfin.o OBJS-$(CONFIG_MLIB)+= mlib/yuv2rgb_mlib.o OBJS-$(HAVE_ALTIVEC) += ppc/yuv2rgb_altivec.o -OBJS-$(HAVE_MMX) += x86/yuv2rgb_mmx.o +OBJS-$(HAVE_MMX) += x86/rgb2rgb.o \ + x86/yuv2rgb_mmx.o OBJS-$(HAVE_VIS) += sparc/yuv2rgb_vis.o TESTPROGS = colorspace swscale diff --git a/libswscale/rgb2rgb.c b/libswscale/rgb2rgb.c index e6d7971..9e27e82 100644 --- a/libswscale/rgb2rgb.c +++ b/libswscale/rgb2rgb.c @@ -24,50 +24,11 @@ */ #include inttypes.h #include config.h -#include libavutil/x86_cpu.h #include libavutil/bswap.h #include rgb2rgb.h #include swscale.h #include swscale_internal.h -#if ARCH_X86 -DECLARE_ASM_CONST(8, uint64_t, mmx_ff) = 0x00FFULL; -DECLARE_ASM_CONST(8, uint64_t, mmx_null) = 0xULL; -DECLARE_ASM_CONST(8, uint64_t, mmx_one) = 0xULL; -DECLARE_ASM_CONST(8, uint64_t, mask32b) = 0x00FF00FFULL; -DECLARE_ASM_CONST(8, uint64_t, mask32g) = 0xFF00FF00ULL; -DECLARE_ASM_CONST(8, uint64_t, mask32r) = 0x00FF00FFULL; -DECLARE_ASM_CONST(8, uint64_t, mask32a) = 0xFF00FF00ULL; -DECLARE_ASM_CONST(8, uint64_t, mask32) = 0x00FF00FFULL; -DECLARE_ASM_CONST(8, uint64_t, mask3216br) = 0x00F800F800F800F8ULL; -DECLARE_ASM_CONST(8, uint64_t, mask3216g)= 0xFC00FC00ULL; -DECLARE_ASM_CONST(8, uint64_t, mask3215g)= 0xF800F800ULL; -DECLARE_ASM_CONST(8, uint64_t, mul3216) = 0x20042004ULL; -DECLARE_ASM_CONST(8, uint64_t, mul3215) = 0x20082008ULL; -DECLARE_ASM_CONST(8, uint64_t, mask24b) = 0x00FFFFFFULL; -DECLARE_ASM_CONST(8, uint64_t, mask24g) = 0xFFFFFF00ULL; -DECLARE_ASM_CONST(8, uint64_t, mask24r) = 0xFFFFULL; -DECLARE_ASM_CONST(8, uint64_t, mask24l) = 0x00FFULL; -DECLARE_ASM_CONST(8, uint64_t, mask24h) = 0xFF00ULL; -DECLARE_ASM_CONST(8, uint64_t, mask24hh) = 0xULL; -DECLARE_ASM_CONST(8, uint64_t, mask24hhh)= 0xULL; -DECLARE_ASM_CONST(8, uint64_t, mask24) = 0xULL; -DECLARE_ASM_CONST(8, uint64_t, mask15b) = 0x001F001F001F001FULL; /* 0001 xxB */ -DECLARE_ASM_CONST(8, uint64_t, mask15rg) = 0x7FE07FE07FE07FE0ULL; /* 0111 1110 RGx */ -DECLARE_ASM_CONST(8, uint64_t, mask15s) = 0xFFE0FFE0FFE0FFE0ULL; -DECLARE_ASM_CONST(8, uint64_t, mask15g) = 0x03E003E003E003E0ULL; -DECLARE_ASM_CONST(8, uint64_t, mask15r) = 0x7C007C007C007C00ULL; -#define mask16b mask15b -DECLARE_ASM_CONST(8, uint64_t, mask16g) = 0x07E007E007E007E0ULL; -DECLARE_ASM_CONST(8, uint64_t, mask16r) = 0xF800F800F800F800ULL; -DECLARE_ASM_CONST(8, uint64_t, red_16mask) = 0xf800f800ULL; -DECLARE_ASM_CONST(8, uint64_t, green_16mask) = 0x07e007e0ULL; -DECLARE_ASM_CONST(8, uint64_t, blue_16mask) = 0x001f001fULL; -DECLARE_ASM_CONST(8, uint64_t, red_15mask) = 0x7c007c00ULL; -DECLARE_ASM_CONST(8, uint64_t, green_15mask) = 0x03e003e0ULL; -DECLARE_ASM_CONST(8, uint64_t, blue_15mask) = 0x001f001fULL; -#endif /* ARCH_X86 */ - #define RGB2YUV_SHIFT 8 #define BY ((int)( 0.098*(1RGB2YUV_SHIFT)+0.5)) #define BV ((int)(-0.071*(1RGB2YUV_SHIFT)+0.5)) @@ -79,49 +40,9 @@ DECLARE_ASM_CONST(8, uint64_t, blue_15mask) = 0x001f001fULL; #define RV ((int)( 0.439*(1RGB2YUV_SHIFT)+0.5)) #define RU ((int)(-0.148*(1RGB2YUV_SHIFT)+0.5)) -//Note: We have C, MMX, MMX2, 3DNOW versions, there is no 3DNOW + MMX2 one. //plain C versions #include rgb2rgb_template.c -#if ARCH_X86 -#define COMPILE_TEMPLATE_MMX 0 -#define COMPILE_TEMPLATE_MMX2 0 -#define COMPILE_TEMPLATE_AMD3DNOW 0 -#define COMPILE_TEMPLATE_SSE2 0 - -//MMX versions -#undef RENAME -#undef COMPILE_TEMPLATE_MMX -#define COMPILE_TEMPLATE_MMX 1 -#define RENAME(a) a ## _MMX -#include x86/rgb2rgb_template.c - -//MMX2 versions -#undef RENAME -#undef COMPILE_TEMPLATE_MMX2 -#define COMPILE_TEMPLATE_MMX2 1 -#define RENAME(a) a ## _MMX2 -#include x86/rgb2rgb_template.c - -//SSE2 versions -#undef RENAME -#undef COMPILE_TEMPLATE_SSE2 -#define
Re: [libav-devel] [PATCH 06/14] Move away x86 specific code from rgb2rgb part 2
On Mon, Apr 11, 2011 at 11:07:39AM +0200, Luca Barbato wrote: With this commit we should have all the arch specific code moved away. Init pattern now: - generic C init first - arch specific init later overwriting --- a/libswscale/rgb2rgb.c +++ b/libswscale/rgb2rgb.c @@ -130,21 +51,14 @@ DECLARE_ASM_CONST(8, uint64_t, blue_15mask) = 0x001f001fULL; 32-bit C version, and andadd trick by Michael Niedermayer */ +void rgb2rgb_init_x86(); Do we not have a suitable header for this forward declaration? --- /dev/null +++ b/libswscale/x86/rgb2rgb.c @@ -0,0 +1,146 @@ +/* + * software RGB to RGB converter + * pluralize by software PAL8 to RGB converter + * software YUV to YUV converter + * software YUV to RGB converter + * Written by Nick Kurshev. + * palette YUV runtime CPU stuff by Michael (michae...@gmx.at) + * + * This file is part of FFmpeg. U + * 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 + */ +#include inttypes.h +#include config.h +#include libavutil/x86_cpu.h +#include libavutil/bswap.h +#include libswscale/rgb2rgb.h +#include libswscale/swscale.h +#include libswscale/swscale_internal.h nits: I suspect just stdint.h is enough; add empty lines before the #includes and between system and local #includes. +#if ARCH_X86 This file is only compiled for x86, so this is unnecessary. +#if ARCH_X86 again +void rgb2rgb_init_x86(int flags); Yeah, we need a header file, now the forward declaration is duplicated. Diego ___ libav-devel mailing list libav-devel@libav.org https://lists.libav.org/mailman/listinfo/libav-devel
Re: [libav-devel] [PATCH 06/14] Move away x86 specific code from rgb2rgb part 2
On Mon, Apr 11, 2011 at 11:26:16AM +0200, Kostya wrote: On Mon, Apr 11, 2011 at 11:07:39AM +0200, Luca Barbato wrote: With this commit we should have all the arch specific code moved away. Init pattern now: - generic C init first - arch specific init later overwriting --- /dev/null +++ b/libswscale/x86/rgb2rgb.c +void rgb2rgb_init_x86(int flags); is that requirement by modern GCC to declare function before its implementation in the same file? It is a requirement of modern FFmpeg configure which adds -Werror=implicit-declaration to CFLAGS. The declaration should go in a header file instead. Diego ___ libav-devel mailing list libav-devel@libav.org https://lists.libav.org/mailman/listinfo/libav-devel
