Should call celt_inner_prod().
This requires the API change of celt_pitch_xcorr() by passing in
"arch".
We tested on x86 and arm, and got bit exact results as original.
Change-Id: I606915da6a196f327ce81f4a5ae32811f4c1fabb
Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
Should call celt_inner_prod().
This change is bit exact as original, except for x86 floating-point.
In x86 floating-point, it calls celt_inner_prod_sse() which may have
different output with the change of floating-point operations' orders.
Change-Id: Ia2381e2e198a84296ac28305183f15be842b3454
Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
Enable x86 intrinsics when building in floating-point mode.
Support SSE as an arch value.
Use RTCD to conditionally enable existing floating-point Celt SSE code.
Call functions directly (without RTCD) when their architecture can be presumed.
Use SSE4.1 intrinsics optimized code for Silk even in floating-point mode.
This should not take an arch parameter, so it can properly be used
as a fallback for accelerated versions which do not.
This patch instead provides a separate version which can call
accelerated helpers for platforms that have taken that approach.
1. Only for fixed point on x86 platform (32bit and 64bit, uses SIMD
intrinsics up to SSE4.2)
2. Use "configure --enable-fixed-point --enable-intrinsics" to enable
optimization, default is disabled.
3. Official test cases are verified and passed.
Signed-off-by: Timothy B. Terriberry <tterribe@xiph.org>
Rename y0 and y1 because of the name clash with Bessel functions.
Initialize y_3 to zero because gcc is too dumb to realize it can't
be used uninitialized.
Computes most of the auto-correlation by reusing pitch_xcorr(). We only
need lag*(lag-1)/2 MACs to complete the calculations.
To do this, pitch_xcorr() was modified so that it no longer truncates the
length to a multiple of 4. Also, the xcorr didn't need the floor at -1.
As a side benefit, this speeds up the PLC, which uses a higher order LPC
filter.
The Microsoft compiler warns about precision reduction from
default double literals to the floats we generally use outside
the fixed-point build. Avoid these by qualifying the literals
as floats.
Thanks to derf for review.
We had previously advised people to -Drestrict on
non-C99 compilers, but this creates problems for
some of the MSVC headers. Instead this just
uses a macro and defines it sanely.
Tuned the tonality estimator to trigger on signals where only part of the
spectrum is tonal. Also tuned the pitch detector not to be confused
by short-term correlation.
The overflows reported were:
- Syy in find_best_pitch() in pitch.c
- xy, xx and yy in remove_doubling() in pitch.c
The fixes are:
- Adaptive scaling in pitch_downsample() which also improves quality
- Adding a missing downshift in find_best_pitch()
- More conservative yshift when calling find_best_pitch()
C reserves identifiers of the from _[A-Z]+ and we have a number of
those in the code. This patch renames the various function arguments,
MACROS and preprocessor symbols to avoid the reserved form.
It also removes the CHANNELS() macro altogether. This was a
minor optimization for TI DSP to force a mono-only build,
as were the associated local 'const' versions. Since stereo
support is manditory, it wasn't worth keeping.
Thanks to John Ridges for raising the issue, and Jean-Marc Valin
and Greg Maxwell for reviewing the changes.