search for: use_asm

I think I've gotten FLAC__lpc_restore_signal() about as good as I'm going to get it. Here's what I have: -a new file, lpc_asm.s, which has the assembly routines -changes to cpu.h, cpu.c, and stream_decoder.c to enable them -changes to configure.in to support the new cpu stuff -a preliminary Makefile.am -maybe something else I'm forgetting Now automake complains that configure.in

.../* FLAC__NO_ASM */ + #endif /* FLAC__CPU_PPC */ + const unsigned FLAC__CPUINFO_IA32_CPUID_CMOV = 0x00008000; const unsigned FLAC__CPUINFO_IA32_CPUID_MMX = 0x00800000; const unsigned FLAC__CPUINFO_IA32_CPUID_FXSR = 0x01000000; *************** *** 78,83 **** --- 86,115 ---- #else info->use_asm = false; #endif + #elif defined FLAC__CPU_PPC + info->type = FLAC__CPUINFO_TYPE_PPC; + #if !defined FLAC__NO_ASM + info->use_asm = true; + #ifdef FLAC__USE_ALTIVEC + #if defined __APPLE__ && defined __MACH__ + { + int selectors[2] = { CTL_HW, HW_VECTORUNIT }; + int result = 0...

...lcome to redistribute it under certain conditions. Type `flac' for details. > > options: -P 0 -b 4608 -m -l 8 -q 0 -r 3,3 -R 0 -V > /home/mdz/english.wav: > @@@ FLAC__CPU_UNKNOWN > lt-flac: stream_decoder.c:173: FLAC__stream_decoder_init: Assertion `decoder->guts->cpuinfo.use_asm' failed. > Aborted It looks like this is because FLAC__NO_ASM isn't defined, but should be. configure.in should default to disabling asm optimizations unless the host architecture is one of those that are recognized. -- - mdz

There was a commit "Fall back to intrinsics if NASM is not available" ( http://git.xiph.org/?p=flac.git;a=commit;h=d163ef456797ef4807138b0ecb5da9ff05e147e2 ) But CPU detection code for IA-32 uses ASM code, and cpuinfo.use_asm == false when NASM isn't available. The attached patch fixes it. Now CPU features are also detected when only FLAC__HAS_X86INTRIN is defined. -------------- next part -------------- A non-text attachment was scrubbed... Name: cpuid_no_nasm.patch Type: application/octet-stream Size: 3994 bytes D...

...n 320X240, the environment is as following, - CPU Interl Celeron 2.66GHz,RAM1GB, - Windows XP SP2 According to profile about Iaxclient and our analysis it is found what consumed most of the CPU resources are Theora encode/decode functions,we recompiled Theora lib with the argument "USE_ASM" and make it MMX instruction set,but the effect is not obvious. Is there anyone else met this issue too? Is there any suggestion or solution about the issue?

...receive video -> Theora decode -> (video data with YUV format) -> Jiaxclient( show video picture). According to profile about Iaxclient and our analysis it is found what consumed most of the CPU resources are Theora encode/decode functions,we recompiled Theora lib with the argument "USE_ASM" and make it MMX instruction set,but the effect is not obvious. Is there anyone else met this issue too? Is there any suggestion or solution about the issue?

> > > Unfortunately, there is a bigger problem that affects both SDL > > > and FLAC, > > > which is that the assembly routines are not PIC. > > > > It's not? I think all the IA32 code only references data on the > > stack, and > > it doesn't call outside the library or export any functions outside > > the > > library. The

I'm pleased to announce that the port of the GCC (AT&T) inline assembly optimized code to Visual C (Intel) inline assembly code is now available in SVN. Thusnelda SVN branch is located here: http://svn.xiph.org/branches/theora-thusnelda For more information regarding Visual Studio 2005/2008 compilation read: http://svn.xiph.org/branches/theora-thusnelda/win32/VS2005/README

...together as it wasn't actually being used but that was difficult to tell because of all the #ifdef nonsense." But FLAC__cpu_have_cpuid_x86() actually WAS used in the code #if !defined FLAC__NO_ASM && (defined FLAC__HAS_NASM || FLAC__HAS_X86INTRIN) info->use_asm = true; /* we assume a minimum of 80386 with FLAC__CPU_IA32 */ #if FLAC__HAS_X86INTRIN if(!FLAC__cpu_have_cpuid_x86()) return; #else if(!FLAC__cpu_have_cpuid_asm_ia32()) return; #endif I'll try to explain it:...

...software, and you are welcome to redistribute it under certain conditions. Type `flac' for details. options: -P 0 -b 4608 -m -l 8 -q 0 -r 3,3 -R 0 -V /home/mdz/english.wav: @@@ FLAC__CPU_UNKNOWN lt-flac: stream_decoder.c:173: FLAC__stream_decoder_init: Assertion `decoder->guts->cpuinfo.use_asm' failed. Aborted -- - mdz

...amp; FLAC__CPUINFO_IA32_CPUID_EXTENDED_AMD_EXTMMX)? true : false; - -#ifndef FLAC__SSE_OS - if(!FLAC__cpu_info_sse_os_asm_ia32()) /* this function currently always returns false */ - info->data.ia32.fxsr = info->data.ia32.sse = info->data.ia32.sse2 = false; -#endif } #else info->use_asm = false; Index: src/libFLAC/ia32/cpu_asm.nasm =================================================================== RCS file: /cvsroot/flac/flac/src/libFLAC/ia32/cpu_asm.nasm,v retrieving revision 1.4 diff -u -r1.4 cpu_asm.nasm --- src/libFLAC/ia32/cpu_asm.nasm 2001/07/18 00:24:46 1.4 +++ src/libFLAC...

Hi, Attached is a patch against 1.0alpha7 to speed up Theora decoding. It is about 15~20% faster in my test. It consists of following things: * MMX loop filter based on Rudolf Marek's patch in http://lists.xiph.org/pipermail/theora-dev/2005-August/002838.html * MMX IDCT based on Rudolf Marek's patch in http://lists.xiph.org/pipermail/theora-dev/2005-July/002816.html and the code in

The following series adds initial vector support for PowerPC64. On POWER9, flac --best is about 3.3x faster. Amitay Isaacs (2): Add m4 macro to check for C __attribute__ features Check if compiler supports target attribute on ppc64 Anton Blanchard (5): configure.ac: Remove SPE detection code configure.ac: Add VSX enable/disable configure.ac: Fix FLAC__CPU_PPC on little endian, and add