similar to: [PATCH] for FLAC__cpu_info() in cpu.c

I found that FLAC__cpu_have_cpuid_x86() was removed in the commit <http://git.xiph.org/?p=flac.git;a=commitdiff;h=fa24613ad94ba8fb8e23bcb9ca80b4548bb617e6> with the message: "Remove `FLAC__cpu_have_cpuid_x86` altogether 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

1) in config.h FLAC__HAS_X86INTRIN macro is always defined and empty, even if x86intrin.h is not available. 2) sse_os is defined as 'yes' or 'no', but AM_CONDITIONAL tests it for 'true': AM_CONDITIONAL(FLaC__SSE_OS, test "x$sse_os" = xtrue) It seems that it should be changed to AM_CONDITIONAL(FLaC__SSE_OS, test "x$sse_os" = xyes) 3) configure

Currently the structure of FLAC__cpu_info() is: void FLAC__cpu_info(FLAC__CPUInfo *info) { ... // init if(info->ia32.cpuid) { ... // very long code ... ... ... ... ... ... ... } } This patch changes it to: void FLAC__cpu_info(FLAC__CPUInfo *info) { ... // init if(info->ia32.cpuid == false)

It's not possible to use ia32/*.nasm code in 64-bit compiles. There's still no 64-bit asm code in FLAC. I'm not familiar with asm too, so I wrote SSE-accelerated code using intrinsics. This code uses two new preprocessor macros: FLAC__CPU_X86_64 (analogous to FLAC__CPU_IA32) and FLAC__HAS_X86INTRIN (analogous to FLAC__HAS_NASM) Patch for cpu.c/cpu.h adds CPU features (sse3, ssse3)

> lvqcl wrote: > >> 1) in config.h FLAC__HAS_X86INTRIN macro is always defined and empty, >> even if x86intrin.h is not available. >> >> 2) sse_os is defined as 'yes' or 'no', but AM_CONDITIONAL tests it for 'true': > > The patch is attached. Please check it. Ping. Is the fix of FLAC__HAS_X86INTRIN definition necessary? Anyway I think

Erik de Castro Lopo wrote: > lvqcl.mail wrote: > >> Currently libFLAC doesn't check the existence of CPUID instruction if >> FLAC__HAS_X86INTRIN is set to 1. >> It's not a real problem because x86 CPUs without CPUID are probably >> extinct, but if libFLAC performs this check then it should do it >> in all cases (when NASM is available or intrinsics are

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. --------------

configure.ac still has two places that aren't fully correct: 1) it creates incorrect definition of FLAC__HAS_X86INTRIN in config.h file. 2)AM_CONDITIONAL(FLaC__SSE_OS...) compares sse_os with 'true' instead of 'yes'. -------------- next part -------------- A non-text attachment was scrubbed... Name: conf_ac.patch Type: application/octet-stream Size: 1074 bytes Desc: not

Building 1.3.2 on OpenBSD/i386, I get this error: cpu.c: In function 'ia32_cpu_info': cpu.c:128: error: 'FLAC__CPUINFO_IA32_CPUID_SSE3' undeclared (first use in this function) cpu.c:128: error: (Each undeclared identifier is reported only once cpu.c:128: error: for each function it appears in.) cpu.c:129: error: 'FLAC__CPUINFO_IA32_CPUID_SSSE3' undeclared (first use in

On 14 March 2016 at 09:02, Erik de Castro Lopo <mle+la at mega-nerd.com> wrote: > commit a9f84425cfd5d1dbfb564730ca80d0b588cb4f24 > Author: Erik de Castro Lopo <erikd at mega-nerd.com> > Date: Mon Mar 14 18:14:31 2016 +1100 > > libFLAC/cpu.c: Use `sigemptyset` instead of `__sigemptyset` > > The former is POSIX while the later is a GNU

Hi Guys, The following code in CPU.c (line 155) won't link if you don't have NASM code built even if FLAC__HAS_X86INTRIN is true as FLAC__cpu_info_asm_ia32 don't exists and the else is compiled if there is no dead code stripping if (FLAC__HAS_X86INTRIN) { FLAC__cpu_info_x86(0, &flags_eax, &flags_ebx, &flags_ecx, &flags_edx); info->ia32.intel =

On 03/14/16 03:51 PM, lvqcl wrote: > With --disable-sse, FLAC__SSE_OS is undefined and FLAC__cpu_info() uses > sigemptyset/sigaction to determine OS SSE support. That's not quite right as I have to build binaries with --disable-sse (I build and distribute both with and without) so that some users on PII's don't get a sigill, even though the OS supports SSE. So it seems to

lvqcl wrote: > Found a bug in Android OS SSE test. Sorry, what is the bug and how does this fix it? Erik -- ---------------------------------------------------------------------- Erik de Castro Lopo http://www.mega-nerd.com/

Erik de Castro Lopo wrote: > When should FLAC__HAS_X86INTRIN be defined? What header file should I be > checking for? Ah, should be checking for <x86intrin.h>. The rest seems to be coming together. Testing this now. Erik -- ---------------------------------------------------------------------- Erik de Castro Lopo http://www.mega-nerd.com/

lvqcl wrote: > Erik de Castro Lopo wrote: > > >> Found a bug in Android OS SSE test. > > > > Sorry, what is the bug and how does this fix it? > > FLAC__cpu_info(FLAC__CPUInfo *info): detects CPU features > that can be used and sets corresponding flags in the info > struct. This code is in flux and in the current state the logic probably has

In the precompute_partition_info_sums_ function, instead of selecting 64-bit accumulator when the signal bps is larger than 16, revert to the original approach based on partition size, but make room for few extra bits to not overflow with unusual signals where the average residual magnitude may be larger than bps. It slightly improves the performance with standard encoding levels and 16-bit files

lvqcl wrote: > I have a couple of questions: > > 1) Do you plan to release 1.3.1 pre1, pre2 etc or just 1.3.1 w/o any pre-releases? I had not planned to do a pre-release. > 2) Do you plan to release any official binaries (flac, metaflac, maybe something else)? Nor had I planned to release binaries. The source code tarball ends up here: https://svn.xiph.org/releases/flac/ I

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

lvqcl wrote: > lvqcl wrote: > > > 1) in config.h FLAC__HAS_X86INTRIN macro is always defined and empty, > > even if x86intrin.h is not available. > > > > 2) sse_os is defined as 'yes' or 'no', but AM_CONDITIONAL tests it for 'true': > > The patch is attached. Please check it. Looks good. I need to do a little testing. > > 3)

Is Nasm always available when the target architecture is IA-32? For example, FLAC has FLAC__lpc_compute_autocorrelation_asm_ia32_sse_lag_NN() functions that require NASM presence, and FLAC__lpc_compute_autocorrelation_intrin_sse_lag_NN() functions that require x86 intrinsics support. FLAC makes use of the former (if NASM is available) but doesn't try to use the latter when NASM is not found.