search for: as_gcc_inline_assembly

On 25 November 2014 at 10:11, Viswanath Puttagunta <viswanath.puttagunta at linaro.org> wrote: > > On 25 November 2014 at 09:39, Jonathan Lennox <jonathan at vidyo.com> wrote: > > > > On Nov 25, 2014, at 10:07 AM, Viswanath Puttagunta <viswanath.puttagunta at linaro.org> wrote: > >> > >> > Also is there plans to make the NEON optimisations

...s" case $host_cpu in + arm64*|aarch64*) + dnl Currently we only have asm for fixed-point + AS_IF([test "$enable_float" != "yes"],[ + cpu_arm64=yes + AC_DEFINE([OPUS_ARM64_ASM], [], [Make use of ARM64 asm optimization]) + AS_GCC_INLINE_ASSEMBLY( + [inline_optimization="ARM64"], + [inline_optimization="disabled"] + ) + AS_IF([test x"$inline_optimization" = x"ARM64"],[ + AC_DEFINE([OPUS_ARM64_INLINE_ASM], 1, + [Use A...

...s" case $host_cpu in + arm64*|aarch64*) + dnl Currently we only have asm for fixed-point + AS_IF([test "$enable_float" != "yes"],[ + cpu_arm64=yes + AC_DEFINE([OPUS_ARM64_ASM], [], [Make use of ARM64 asm optimization]) + AS_GCC_INLINE_ASSEMBLY( + [inline_optimization="ARM64"], + [inline_optimization="disabled"] + ) + AS_IF([test x"$inline_optimization" = x"ARM64"],[ + AC_DEFINE([OPUS_ARM64_INLINE_ASM], 1, + [Use A...

...e should be much smaller as we can make valid assumption that NEON will be supported... so as Jonathan Lennox suggested earlier, most of these flags can be hardcoded..... even run time checks can be disabled as ARMv8 cpu in aarch64 mode is guaranteed to have NEON/ASIMD. > >> AS_GCC_INLINE_ASSEMBLY( >> [inline_optimization="ARM"], >> [inline_optimization="disabled"] >> @@ -212,6 +211,35 @@ AS_IF([test x"${enable_asm}" = x"yes"],[ >> AS_ASM_ARM_MEDIA([OPUS_ARM_INLINE_MEDIA=1], >...

...rm64*|aarch64*) 204 dnl Currently we only have asm for fixed-point 205 AS_IF([test "$enable_float" != "yes"],[ 206 cpu_arm64=yes 207 AC_DEFINE([OPUS_ARM64_ASM], [], [Make use of ARM64 asm optimization]) 208 AS_GCC_INLINE_ASSEMBLY( 209 [inline_optimization="ARM64"], 210 [inline_optimization="disabled"] 211 ) 212 AS_IF([test x"$inline_optimization" = x"ARM64"],[ 213 AM_CONDITIONAL([OPUS_ARM64_INL...

...dnl Currently we only have asm for fixed-point > 205 AS_IF([test "$enable_float" != "yes"],[ > 206 cpu_arm64=yes > 207 AC_DEFINE([OPUS_ARM64_ASM], [], [Make use of ARM64 asm > optimization]) > 208 AS_GCC_INLINE_ASSEMBLY( > 209 [inline_optimization="ARM64"], > 210 [inline_optimization="disabled"] > 211 ) > 212 AS_IF([test x"$inline_optimization" = x"ARM64"],[ > 213 AM_CO...

Hello, I received feedback from engineers working on NE10 [1] that it would be better to use NE10 [1] for FFT optimizations for opus use cases. However, these FFT patches are currently in review and haven't been integrated into NE10 yet. While the FFT functions in NE10 are getting baked, I wanted to optimize the celt_pitch_xcorr (floating point only) and use it to introduce ARM NEON

...6 * OPUS_RESTRICT x_lp, opus_val16 * OPUS_RESTRICT y, int len, int max_pitch, int *pitch); diff --git a/configure.ac b/configure.ac index ee6df9a..6b9612f 100644 --- a/configure.ac +++ b/configure.ac @@ -167,6 +167,7 @@ if test "x${ac_enable_asm}" = xyes ; then AS_GCC_INLINE_ASSEMBLY([asm_optimization="ARM"], [asm_optimization="disabled"]) if test "x${asm_optimization}" = "xARM" ; then + AC_DEFINE([ARM_ASM], 1, [Use generic ARM asm optimizations]) AC_DEFINE([ARMv4_ASM], 1, [Use generic ARMv4 a...

...ve asm for fixed-point - AS_IF([test "$enable_float" != "yes"],[ cpu_arm=yes - AC_DEFINE([OPUS_ARM_ASM], [], [Make use of ARM asm optimization]) + AC_DEFINE([OPUS_ARM_ASM], [], [Make use of ARM asm/intrinsic optimization]) AS_GCC_INLINE_ASSEMBLY( [inline_optimization="ARM"], [inline_optimization="disabled"] @@ -212,6 +211,35 @@ AS_IF([test x"${enable_asm}" = x"yes"],[ AS_ASM_ARM_MEDIA([OPUS_ARM_INLINE_MEDIA=1], [OPUS_ARM_INLINE_MEDIA=0])...

...]) > + AC_DEFINE([OPUS_ARM_ASM], [], [Make use of ARM asm/intrinsic optimization]) Not sure I'm in love with conflating intrinsics with inline assembly. For example, are these tests (especially the PRESUME_NEON stuff) going to do the right thing on aarch64? > AS_GCC_INLINE_ASSEMBLY( > [inline_optimization="ARM"], > [inline_optimization="disabled"] > @@ -212,6 +211,35 @@ AS_IF([test x"${enable_asm}" = x"yes"],[ > AS_ASM_ARM_MEDIA([OPUS_ARM_INLINE_MEDIA=1], >...

This sequence of patches provides arm64 support for Opus. Tested on iOS, Android, and Ubuntu 14.04. The patch sequence was written on top of Viswanath Puttagunta's Ne10 patches, but all but the second ("Reorganize pitch_arm.h") should, I think, apply independently of it. It does depends on my previous intrinsics configury reorganization, however. Comments welcome. With this and

Here are my aarch64 patches rebased to the current tip of Opus master. They're largely the same as my previous patch set, with the addition of the final one (the Neon fixed-point implementation of xcorr_kernel). This replaces Viswanath's Neon fixed-point celt_pitch_xcorr, since xcorr_kernel is used in celt_fir and celt_iir as well. These have been tested for correctness under qemu