search for: opus_cpu_arm_neon

This adds run-time CPU detection for iOS devices. I found this page useful for CPU support and models: http://iossupportmatrix.com/ The patch could be simplified to just set the OPUS_CPU_ARM_NEON flag if you think it's unlikely that anyone would be running on the really old devices that don't support NEON. Tom -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.xiph.org/pipermail/opus/attachments/20150108/65653d3c/attachment.htm -----------...

On Jan 8, 2015, at 2:33 PM, Tom Hughes <tom at airtime.com<mailto:tom at airtime.com>> wrote: This adds run-time CPU detection for iOS devices. I found this page useful for CPU support and models: http://iossupportmatrix.com/ The patch could be simplified to just set the OPUS_CPU_ARM_NEON flag if you think it's unlikely that anyone would be running on the really old devices that don't support NEON. Tom Every armv7 (and armv8) iOS device has supported Neon, and Xcode support for armv6 was dropped with Xcode 4.5. Even if you?re compiling with an old Xcode version to support...

..._ASM) + __try{ + /*SHADD8 r3,r3,r3*/ + __emit(0xE6333F93); + flags|=OPUS_CPU_ARM_MEDIA; + } + __except(GetExceptionCode()==EXCEPTION_ILLEGAL_INSTRUCTION){ + /*Ignore exception.*/ + } +# if defined(ARM_HAVE_NEON) + __try{ + /*VORR q0,q0,q0*/ + __emit(0xF2200150); + flags|=OPUS_CPU_ARM_NEON; + } + __except(GetExceptionCode()==EXCEPTION_ILLEGAL_INSTRUCTION){ + /*Ignore exception.*/ + } +# endif +# endif +# endif + return flags; +} + +#elif defined(__linux__) +/* Linux based */ +opus_uint32 opus_cpu_capa(void) +{ + opus_uint32 flags = 0; + FILE *cpuinfo; + + /* Reading /pro...

Jonathan Lennox wrote: > +opus_int32 silk_noise_shape_quantizer_short_prediction_neon(const opus_int32 *buf32, const opus_int32 *coef32) > +{ > + int32x4_t coef0 = vld1q_s32(coef32); > + int32x4_t coef1 = vld1q_s32(coef32 + 4); > + int32x4_t coef2 = vld1q_s32(coef32 + 8); > + int32x4_t coef3 = vld1q_s32(coef32 + 12); > + > + int32x4_t a0 = vld1q_s32(buf32 -

...int opus_select_arch_real(void) { opus_uint32 flags = opus_cpu_capabilities(); int arch = 0; if(!(flags & OPUS_CPU_ARM_EDSP)) - return arch; + goto final; arch++; if(!(flags & OPUS_CPU_ARM_MEDIA)) - return arch; + goto final; arch++; if(!(flags & OPUS_CPU_ARM_NEON)) - return arch; + goto final; arch++; +final: + detected = 1; + arch_arm = arch; return arch; } +int opus_select_arch(void) +{ + return (detected?arch_arm:opus_select_arch_real()); +} + #endif diff --git a/celt/x86/x86cpu.c b/celt/x86/x86cpu.c index c82a4b7..ddf3cf3 100644 --...

Following Tim's comments, here are my reworked patches for the Neon intrinsic function patches of of my Aarch64 patchset, i.e. replacing patches 5-8 of the v2 series. Patches 1-4 and 9-18 of the old series still apply unmodified. The one new (as opposed to changed) patch is the first one in this series, to add named constants for the ARM architecture variants. There are also some minor code

...t ? what do you suggest instead? As the comment mentions, this can?t use the usual IMPL table implementation, because the parameters are different (due to the transformed coefficients.) Would something like an OPUS_ARCH_ARM_NEON #define in celt/arm/armcpu.h be okay? It?d be a bit confusing with OPUS_CPU_ARM_NEON in armcpu.c, but I could probably organize it to be sensible. > (also, I realize libopus doesn't have a line-length restriction, but a > few newlines in here might be a mercy to those of us who work in > 80-column terminals) Okay.

...gs |= OPUS_CPU_ARM_EDSP; # endif -# if defined(OPUS_ARM_MAY_HAVE_NEON) +# if defined(OPUS_ARM_MAY_HAVE_NEON) || defined(OPUS_ARM_MAY_HAVE_NEON_INTR) p = strstr(buf, " neon"); if(p != NULL && (p[5] == ' ' || p[5] == '\n')) flags |= OPUS_CPU_ARM_NEON; diff --git a/celt/arm/pitch_arm.h b/celt/arm/pitch_arm.h index 125d1bc..8626ed7 100644 --- a/celt/arm/pitch_arm.h +++ b/celt/arm/pitch_arm.h @@ -54,10 +54,10 @@ opus_val32 celt_pitch_xcorr_edsp(const opus_val16 *_x, const opus_val16 *_y, #else /* Start !FIXED_POINT */ /* Float case */ -#if def...

...gs |= OPUS_CPU_ARM_EDSP; # endif -# if defined(OPUS_ARM_MAY_HAVE_NEON) +# if defined(OPUS_ARM_MAY_HAVE_NEON) || defined(OPUS_ARM_MAY_HAVE_NEON_INTR) p = strstr(buf, " neon"); if(p != NULL && (p[5] == ' ' || p[5] == '\n')) flags |= OPUS_CPU_ARM_NEON; diff --git a/celt/arm/pitch_arm.h b/celt/arm/pitch_arm.h index 125d1bc..8626ed7 100644 --- a/celt/arm/pitch_arm.h +++ b/celt/arm/pitch_arm.h @@ -54,10 +54,10 @@ opus_val32 celt_pitch_xcorr_edsp(const opus_val16 *_x, const opus_val16 *_y, #else /* Start !FIXED_POINT */ /* Float case */ -#if def...

The attached patch cleans up Opus's x86 intrinsics configury. It: * Makes ?enable-intrinsics work with clang and other non-GCC compilers * Enables RTCD for the floating-point-mode SSE code in Celt. * Disables use of RTCD in cases where the compiler targets an instruction set by default. * Enables the SSE4.1 Silk optimizations that apply to the common parts of Silk when Opus is built in

Hello opus-dev, I've been cooking up this patchset to integrate NE10 library into opus. Current patchset focuses on encode use case mainly effecting performance of clt_mdct_forward() and opus_fft() (for float only) Glad to report the following on Encode use case: (Measured on my Beaglebone Black Cortex-A8 board) - Performance improvement for encode use case ~= 12.34% (Based on time -p

Hi All, Since I continue to base my work on top of Jonathan's patch, and my previous Ne10 fft/ifft/mdct_forward/backward patches, I thought it would be better to just post all new patches as a patch series. Please let me know if anyone disagrees with this approach. You can see wip branch of all latest patches at https://git.linaro.org/people/viswanath.puttagunta/opus.git Branch:

Hi Timothy, As I mentioned earlier [1], I now fixed compile issues with fixed point and resubmitting the patch. I also have new patch that does intrinsics optimizations for celt_pitch_xcorr targetting aarch64. You can find my latest work-in-progress branch at [2] For reference, you can use the Ne10 pre-built libraries at [3] Note that I am working with Phil at ARM to get my patch at [4]

Hi All, As per Timothy's suggestion, disabling mdct_forward for fixed point. Only effects armv7,armv8: Extend fixed fft NE10 optimizations to mdct Rest of patches are same as in [1] For reference, latest wip code for opus is at [2] Still working with NE10 team at ARM to get corner cases of mdct_forward. Will update with another patch when issue in NE10 gets fixed. Regards, Vish [1]:

Hi All, Changes from RFC v2 [1] armv7,armv8: Extend fixed fft NE10 optimizations to mdct - Overflow issue fixed by Phil at ARM. Ne10 wip at [2]. Should be upstream soon. - So, re-enabled using fixed fft for mdct_forward which was disabled in RFCv2 armv7,armv8: Optimize fixed point fft using NE10 library - Thanks to Jonathan Lennox, fixed some build fixes on iOS and some copy-paste errors Rest

Hello Timothy / Jean-Marc / opus-dev, This patch series is follow up on work I posted on [1]. In addition to what was posted on [1], this patch series mainly integrates Fixed point FFT implementations in NE10 library into opus. You can view my opus wip code at [2]. Note that while I found some issues both with the NE10 library(fixed fft) and with Linaro toolchain (armv8 intrinsics), the work