search for: opus_fft_float_neon

...- 13 files changed, 214 insertions(+), 20 deletions(-) diff --git a/celt/arm/arm_celt_map.c b/celt/arm/arm_celt_map.c index 3b49f90..f132fe1 100644 --- a/celt/arm/arm_celt_map.c +++ b/celt/arm/arm_celt_map.c @@ -79,6 +79,15 @@ void (*const OPUS_FFT[OPUS_ARCHMASK+1])(const kiss_fft_state *cfg, opus_fft_float_neon /* Neon with NE10 */ }; +void (*const OPUS_IFFT[OPUS_ARCHMASK+1])(const kiss_fft_state *cfg, + const kiss_fft_cpx *fin, + kiss_fft_cpx *fout) = { + opus_ifft_c, /* ARMv4 */ + opus_ifft...

...- 13 files changed, 214 insertions(+), 20 deletions(-) diff --git a/celt/arm/arm_celt_map.c b/celt/arm/arm_celt_map.c index 3b49f90..f132fe1 100644 --- a/celt/arm/arm_celt_map.c +++ b/celt/arm/arm_celt_map.c @@ -79,6 +79,15 @@ void (*const OPUS_FFT[OPUS_ARCHMASK+1])(const kiss_fft_state *cfg, opus_fft_float_neon /* Neon with NE10 */ }; +void (*const OPUS_IFFT[OPUS_ARCHMASK+1])(const kiss_fft_state *cfg, + const kiss_fft_cpx *fin, + kiss_fft_cpx *fout) = { + opus_ifft_c, /* ARMv4 */ + opus_ifft...

Viswanath Puttagunta wrote: > What should we do for power-of-2? I really want to avoid putting > runtime checks if nfft is power of 2 in opus_fft_float_neon. Given the tests that had to be disabled for NE10, I suspect we will not really be able to use it for CUSTOM_MODES, which should be the only time nfft is a power of 2. So I'd suggest just disabling the support when CUSTOM_MODES is enabled and putting in an assert. >> It would not be...

Hello All, I extended the libNE10 optimizations for float towards mdct_backwards/opus_ifft. I am able to get about 14.26% improvement for Decode use case now on my Beaglebone Black. Please see [1] for measurements. Questions 1. Since this patch needs to go in after Encode [2] patch) should I submit this as patch series? 2. Since Jonathan Lennox posted intrinsics cleanup [3] patch, should

Hi all, I have already added support for scaled forward non-power-of-2 floating-point FFT: https://github.com/projectNe10/Ne10/commit/79c3d787302f8d74b9bcfe6545d487cdf1b101d9 Two flags are added to cfg structure: is_forward_scaled and is_backward_scaled. By setting is_forward_scaled to anything but zero, ne10_fft_c2c_1d_float32_neon will scale the output. So we can remove need for one buffer on

...ree_arm_neon /* Neon with NE10 */ }; #endif /* CUSTOM_MODES */ @@ -76,7 +58,7 @@ void (*const OPUS_FFT[OPUS_ARCHMASK+1])(const kiss_fft_state *cfg, opus_fft_c, /* ARMv4 */ opus_fft_c, /* EDSP */ opus_fft_c, /* Media */ - opus_fft_float_neon /* Neon with NE10 */ + opus_fft_neon /* Neon with NE10 */ }; void (*const OPUS_IFFT[OPUS_ARCHMASK+1])(const kiss_fft_state *cfg, @@ -85,9 +67,29 @@ void (*const OPUS_IFFT[OPUS_ARCHMASK+1])(const kiss_fft_state *cfg, opus_ifft_c, /* ARMv4 */...

...one module MUST be freed by the same module, because different modules do not share libc state (indeed, they can even be linked against different implementations of libc). If this is how the API of NE10 is designed to work, then that API needs to be fixed. > +} > +#endif > + > +void opus_fft_float_neon(const kiss_fft_state *st, > + const kiss_fft_cpx *fin, > + kiss_fft_cpx *fout) Again, these are mis-aligned. > +{ > + ne10_fft_cfg_float32_t cfg = (ne10_fft_cfg_float32_t)st->priv; > + VARDECL(ne10_fft_cpx_float32_t, temp); >...

...omething like typedef ne10_fft_cpx_float32 ne10_fft_cpx_float32_t typedef ne10_fft_cpx_float32 ne10_fft_cpx_float32_t Then from opus side, I can instead use ne10_fft_cfg_float32 cfg = (ne10_fft_cfg_float32)st->priv; VARDECL(ne10_fft_cpx_float32, temp); VARDECL(ne10_fft_cpx_float32, tempin); 2. opus_fft_float_neon: Remove at least one additional buffer that is currently needed (on stack) - Currently, we have to use 2 additional buffers on stack. - One for scaling (ALLOC(tempin, st->nfft, ne10_fft_cpx_float32_t);) - Another as temp buffer (cfg->buffer) - Is there any reason the scaling cannot be inside...

...ILD_SHARED=ON -DCMAKE_TOOLCHAIN_FILE=../GNUlinux_config.cmake .. make -j6 Timothy/All. Having scaling capability for non-power-of-2 will work for non-custom modes(nfft = 60, 120, 240 and 480). What should we do for power-of-2? I really want to avoid putting runtime checks if nfft is power of 2 in opus_fft_float_neon. > > As for the _t API style, it may confuse existing users if we provide two API for the same functionality or if we modify the existing API. How about we keep the typedefs in Opus? For example: > typedef ne10_int32_t ne10_int32; > typedef ne10_fft_cpx_float32_t ne10_fft_cpx_float32;...

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

...])(const kiss_fft_state *cfg, + const kiss_fft_cpx *fin, + kiss_fft_cpx *fout) = { + opus_fft_c, /* ARMv4 */ + opus_fft_c, /* EDSP */ + opus_fft_c, /* Media */ + opus_fft_float_neon /* Neon with NE10 */ +}; + +#endif diff --git a/celt/arm/arm_celt_ne10_mdct_map.c b/celt/arm/arm_celt_ne10_mdct_map.c new file mode 100644 index 0000000..6df7af3 --- /dev/null +++ b/celt/arm/arm_celt_ne10_mdct_map.c @@ -0,0 +1,53 @@ +/* Copyright (c) 2015-2016 Xiph.Org Foundation + Writ...

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

...(const kiss_fft_state *cfg, + const kiss_fft_cpx *fin, + kiss_fft_cpx *fout) = { + opus_fft_c, /* ARMv4 */ + opus_fft_c, /* EDSP */ + opus_fft_c, /* Media */ + opus_fft_float_neon /* Neon with NE10 */ +}; + +void (*const CLT_MDCT_FORWARD_IMPL[OPUS_ARCHMASK+1])(const mdct_lookup *l, + kiss_fft_scalar *in, + kiss_fft_scalar * OPUS_RESTRICT out, +...

...(const kiss_fft_state *cfg, + const kiss_fft_cpx *fin, + kiss_fft_cpx *fout) = { + opus_fft_c, /* ARMv4 */ + opus_fft_c, /* EDSP */ + opus_fft_c, /* Media */ + opus_fft_float_neon /* Neon with NE10 */ +}; + +void (*const CLT_MDCT_FORWARD_IMPL[OPUS_ARCHMASK+1])(const mdct_lookup *l, + kiss_fft_scalar *in, + kiss_fft_scalar * OPUS_RESTRICT out, +...

...(const kiss_fft_state *cfg, + const kiss_fft_cpx *fin, + kiss_fft_cpx *fout) = { + opus_fft_c, /* ARMv4 */ + opus_fft_c, /* EDSP */ + opus_fft_c, /* Media */ + opus_fft_float_neon /* Neon with NE10 */ +}; + +void (*const CLT_MDCT_FORWARD_IMPL[OPUS_ARCHMASK+1])(const mdct_lookup *l, + kiss_fft_scalar *in, + kiss_fft_scalar * OPUS_RESTRICT out, +...

Changes from RFC PATCH v3 - Just rebased on tip - For all else, please see notes from RFC PATCH v3 at http://lists.xiph.org/pipermail/opus/2015-March/002902.html - latest wip opus tree/branch https://git.linaro.org/people/viswanath.puttagunta/Ne10.git branch: rfcv4_final_fft_ne10 Viswanath Puttagunta (1): armv7(float): Optimize encode usecase using NE10 library Makefile.am

Changes from RFC PATCH v1: - passing arch parameter explicitly - reduced stack usage by ~3.5K by using scaled NE10 fft version - moved all optimization array functions to arm_celt_map.c - Other cleanups pointed out by Timothy Phil, As you mentioned earlier, could you please address all compile and linker errors/warnings coming out of Ne10 library? You can find my working Ne10 repo at [1] You

Changes from RFC PATCH v2 - fixed compile issue when just compiling for --enable-intrinsics for ARMv7 without NE10 - Notes for NE10: - All compile/link warnings are now in upstream NE10 - Only patch pending upstream in NE10 is the one that needs to add -funsafe-math-optimizations for ARMv7 targets. - Phil Wang @ ARM is working on getting this fixed. - Note that even without