Skip to content

GitLab

Commit c2da68d4 authored by Martin Reinecke's avatar Martin Reinecke
Browse files
Options

consolidate trigonometric transforms to save cache space and precomputation time

parent d556ae21
Hide whitespace changes
Inline Side-by-side
Showing with 112 additions and 146 deletions
pocketfft_hdronly.h
View file @ c2da68d4
......@@ -2293,6 +2293,7 @@ template<typename T0> class pocketfft_r
//
// sine/cosine transforms
//
template<typename T0> class T_dct1
{
private:
......@@ -2302,7 +2303,8 @@ template<typename T0> class T_dct1
POCKETFFT_NOINLINE T_dct1(size_t length)
: fftplan(2*(length-1)) {}
template<typename T> POCKETFFT_NOINLINE void exec(T c[], T0 fct, bool ortho) const
template<typename T> POCKETFFT_NOINLINE void exec(T c[], T0 fct, bool ortho,
int /*type*/, bool /*cosine*/) const
{
constexpr T0 sqrt2=T0(1.414213562373095048801688724209698L);
size_t N=fftplan.length(), n=N/2+1;
......@@ -2323,22 +2325,42 @@ template<typename T0> class T_dct1
size_t length() const { return fftplan.length()/2+1; }
};
template<typename T0> class T_dct2
template<typename T0> class T_dst1
{
private:
pocketfft_r<T0> fftplan;
vector<T0> twiddle;
public:
POCKETFFT_NOINLINE T_dct2(size_t length)
: fftplan(length), twiddle(length)
POCKETFFT_NOINLINE T_dst1(size_t length)
: fftplan(2*(length+1)) {}
template<typename T> POCKETFFT_NOINLINE void exec(T c[], T0 fct,
bool /*ortho*/, int /*type*/, bool /*cosine*/) const
{
constexpr T0 pi = T0(3.141592653589793238462643383279502884197L);
for (size_t i=0; i<length; ++i)
twiddle[i] = T0(cos(0.5*pi*T0(i+1)/T0(length)));
size_t N=fftplan.length(), n=N/2-1;
arr<T> tmp(N);
tmp[0] = tmp[n+1] = c[0]*0;
for (size_t i=0; i<n; ++i)
{
tmp[i+1] = c[i];
tmp[N-1-i] = -c[i];
}
fftplan.forward(tmp.data(), fct);
for (size_t i=0; i<n; ++i)
c[i] = -tmp[2*i+2];
}
template<typename T> POCKETFFT_NOINLINE void exec(T c[], T0 fct, bool ortho) const
size_t length() const { return fftplan.length()/2-1; }
};
template<typename T0> class T_dcst23
{
private:
pocketfft_r<T0> fftplan;
vector<T0> twiddle;
template<typename T> POCKETFFT_NOINLINE void exec2c(T c[], T0 fct,
bool ortho) const
{
constexpr T0 sqrt2=T0(1.414213562373095048801688724209698L);
size_t N=length();
......@@ -2368,25 +2390,8 @@ template<typename T0> class T_dct2
if (ortho) c[0]/=sqrt2;
}
size_t length() const { return fftplan.length(); }
};
template<typename T0> class T_dct3
{
private:
pocketfft_r<T0> fftplan;
vector<T0> twiddle;
public:
POCKETFFT_NOINLINE T_dct3(size_t length)
: fftplan(length), twiddle(length)
{
constexpr T0 pi = T0(3.141592653589793238462643383279502884197L);
for (size_t i=0; i<length; ++i)
twiddle[i] = T0(cos(0.5*pi*T0(i+1)/T0(length)));
}
template<typename T> POCKETFFT_NOINLINE void exec(T c[], T0 fct, bool ortho) const
template<typename T> POCKETFFT_NOINLINE void exec3c(T c[], T0 fct,
bool ortho) const
{
constexpr T0 sqrt2=T0(1.414213562373095048801688724209698L);
size_t N=length();
......@@ -2417,10 +2422,55 @@ template<typename T0> class T_dct3
}
}
template<typename T> POCKETFFT_NOINLINE void exec2s(T c[], T0 fct,
bool ortho) const
{
constexpr T0 sqrt2=T0(1.414213562373095048801688724209698L);
size_t N=length();
for (size_t k=1; k<N; k+=2)
c[k] = -c[k];
exec2c(c, fct, false);
for (size_t k=0, kc=N-1; k<kc; ++k, --kc)
swap(c[k], c[kc]);
if (ortho) c[0]/=sqrt2;
}
template<typename T> POCKETFFT_NOINLINE void exec3s(T c[], T0 fct,
bool ortho) const
{
constexpr T0 sqrt2=T0(1.414213562373095048801688724209698L);
size_t N=length();
if (ortho) c[0]*=sqrt2;
size_t NS2 = N/2;
for (size_t k=0, kc=N-1; k<NS2; ++k, --kc)
swap(c[k], c[kc]);
exec3c(c, fct, false);
for (size_t k=1; k<N; k+=2)
c[k] = -c[k];
}
public:
POCKETFFT_NOINLINE T_dcst23(size_t length)
: fftplan(length), twiddle(length)
{
constexpr T0 pi = T0(3.141592653589793238462643383279502884197L);
for (size_t i=0; i<length; ++i)
twiddle[i] = T0(cos(0.5*pi*T0(i+1)/T0(length)));
}
template<typename T> POCKETFFT_NOINLINE void exec(T c[], T0 fct, bool ortho,
int type, bool cosine) const
{
if (type==2)
cosine ? exec2c(c, fct, ortho) : exec2s(c, fct, ortho);
else
cosine ? exec3c(c, fct, ortho) : exec3s(c, fct, ortho);
}
size_t length() const { return fftplan.length(); }
};
template<typename T0> class T_dct4
template<typename T0> class T_dcst4
{
// even length algorithm from
// https://www.appletonaudio.com/blog/2013/derivation-of-fast-dct-4-algorithm-based-on-dft/
......@@ -2430,23 +2480,7 @@ template<typename T0> class T_dct4
unique_ptr<pocketfft_r<T0>> rfft;
arr<cmplx<T0>> C2;
public:
POCKETFFT_NOINLINE T_dct4(size_t length)
: N(length),
fft((N&1) ? nullptr : new pocketfft_c<T0>(N/2)),
rfft((N&1)? new pocketfft_r<T0>(N) : nullptr),
C2((N&1) ? 0 : N/2)
{
constexpr T0 pi = T0(3.141592653589793238462643383279502884197L);
if ((N&1)==0)
for (size_t i=0; i<N/2; ++i)
{
T0 ang = -pi/T0(N)*(T0(i)+T0(0.125));
C2[i].Set(cos(ang), sin(ang));
}
}
template<typename T> POCKETFFT_NOINLINE void exec(T c[], T0 fct, bool /*ortho*/) const
template<typename T> POCKETFFT_NOINLINE void exec4c(T c[], T0 fct) const
{
constexpr T0 sqrt2=T0(1.414213562373095048801688724209698L);
if (N&1)
......@@ -2512,106 +2546,38 @@ template<typename T0> class T_dct4
}
}
size_t length() const { return N; }
};
template<typename T0> class T_dst1
{
private:
pocketfft_r<T0> fftplan;
public:
POCKETFFT_NOINLINE T_dst1(size_t length)
: fftplan(2*(length+1)) {}
template<typename T> POCKETFFT_NOINLINE void exec(T c[], T0 fct, bool /*ortho*/) const
template<typename T> POCKETFFT_NOINLINE void exec4s(T c[], T0 fct) const
{
size_t N=fftplan.length(), n=N/2-1;
arr<T> tmp(N);
tmp[0] = tmp[n+1] = c[0]*0;
for (size_t i=0; i<n; ++i)
{
tmp[i+1] = c[i];
tmp[N-1-i] = -c[i];
}
fftplan.forward(tmp.data(), fct);
for (size_t i=0; i<n; ++i)
c[i] = -tmp[2*i+2];
}
size_t length() const { return fftplan.length()/2-1; }
};
template<typename T0> class T_dst2
{
private:
T_dct2<T0> dct;
public:
POCKETFFT_NOINLINE T_dst2(size_t length)
: dct(length) {}
template<typename T> POCKETFFT_NOINLINE void exec(T c[], T0 fct, bool ortho) const
{
constexpr T0 sqrt2=T0(1.414213562373095048801688724209698L);
size_t N=length();
for (size_t k=1; k<N; k+=2)
c[k] = -c[k];
dct.exec(c, fct, false);
for (size_t k=0, kc=N-1; k<kc; ++k, --kc)
swap(c[k], c[kc]);
if (ortho) c[0]/=sqrt2;
}
size_t length() const { return dct.length(); }
};
template<typename T0> class T_dst3
{
private:
T_dct3<T0> dct;
public:
POCKETFFT_NOINLINE T_dst3(size_t length)
: dct(length) {}
template<typename T> POCKETFFT_NOINLINE void exec(T c[], T0 fct, bool ortho)
{
constexpr T0 sqrt2=T0(1.414213562373095048801688724209698L);
size_t N=length();
if (ortho) c[0]*=sqrt2;
size_t NS2 = N/2;
for (size_t k=0, kc=N-1; k<NS2; ++k, --kc)
swap(c[k], c[kc]);
dct.exec(c, fct, false);
exec4c(c, fct);
for (size_t k=1; k<N; k+=2)
c[k] = -c[k];
}
size_t length() const { return dct.length(); }
};
template<typename T0> class T_dst4
{
private:
T_dct4<T0> dct;
public:
POCKETFFT_NOINLINE T_dst4(size_t length)
: dct(length) {}
template<typename T> POCKETFFT_NOINLINE void exec(T c[], T0 fct, bool /*ortho*/)
POCKETFFT_NOINLINE T_dcst4(size_t length)
: N(length),
fft((N&1) ? nullptr : new pocketfft_c<T0>(N/2)),
rfft((N&1)? new pocketfft_r<T0>(N) : nullptr),
C2((N&1) ? 0 : N/2)
{
size_t N=length();
size_t NS2 = N/2;
for (size_t k=0, kc=N-1; k<NS2; ++k, --kc)
swap(c[k], c[kc]);
dct.exec(c, fct, false);
for (size_t k=1; k<N; k+=2)
c[k] = -c[k];
constexpr T0 pi = T0(3.141592653589793238462643383279502884197L);
if ((N&1)==0)
for (size_t i=0; i<N/2; ++i)
{
T0 ang = -pi/T0(N)*(T0(i)+T0(0.125));
C2[i].Set(cos(ang), sin(ang));
}
}
size_t length() const { return dct.length(); }
template<typename T> POCKETFFT_NOINLINE void exec(T c[], T0 fct,
bool /*ortho*/, int /*type*/, bool cosine) const
{ cosine ? exec4c(c, fct) : exec4s(c, fct); }
size_t length() const { return N; }
};
......@@ -3060,7 +3026,7 @@ template<typename T> POCKETFFT_NOINLINE void general_hartley(
template<typename Trafo, typename T> POCKETFFT_NOINLINE void general_dcst(
const cndarr<T> &in, ndarr<T> &out, const shape_t &axes,
T fct, bool ortho, size_t POCKETFFT_NTHREADS)
T fct, bool ortho, int type, bool cosine, size_t POCKETFFT_NTHREADS)
{
shared_ptr<Trafo> plan;
......@@ -3088,7 +3054,7 @@ template<typename Trafo, typename T> POCKETFFT_NOINLINE void general_dcst(
for (size_t i=0; i<len; ++i)
for (size_t j=0; j<vlen; ++j)
tdatav[i][j] = tin[it.iofs(j,i)];
plan->exec(tdatav, fct, ortho);
plan->exec(tdatav, fct, ortho, type, cosine);
for (size_t i=0; i<len; ++i)
for (size_t j=0; j<vlen; ++j)
out[it.oofs(j,i)] = tdatav[i][j];
......@@ -3099,18 +3065,18 @@ template<typename Trafo, typename T> POCKETFFT_NOINLINE void general_dcst(
it.advance(1);
auto tdata = reinterpret_cast<T *>(storage.data());
if ((&tin[0]==&out[0]) && (it.stride_out()==sizeof(T))) // fully in-place
plan->exec(&out[it.oofs(0)], fct, ortho);
plan->exec(&out[it.oofs(0)], fct, ortho, type, cosine);
else if (it.stride_out()==sizeof(T)) // compute FFT in output location
{
for (size_t i=0; i<len; ++i)
out[it.oofs(i)] = tin[it.iofs(i)];
plan->exec(&out[it.oofs(0)], fct, ortho);
plan->exec(&out[it.oofs(0)], fct, ortho, type, cosine);
}
else
{
for (size_t i=0; i<len; ++i)
tdata[i] = tin[it.iofs(i)];
plan->exec(tdata, fct, ortho);
plan->exec(tdata, fct, ortho, type, cosine);
for (size_t i=0; i<len; ++i)
out[it.oofs(i)] = tdata[i];
}
......@@ -3374,13 +3340,13 @@ template<typename T> void dct(const shape_t &shape,
cndarr<T> ain(data_in, shape, stride_in);
ndarr<T> aout(data_out, shape, stride_out);
if (type==1)
general_dcst<T_dct1<T>>(ain, aout, axes, fct, ortho, nthreads);
general_dcst<T_dct1<T>>(ain, aout, axes, fct, ortho, type, true, nthreads);
else if (type==2)
general_dcst<T_dct2<T>>(ain, aout, axes, fct, ortho, nthreads);
general_dcst<T_dcst23<T>>(ain, aout, axes, fct, ortho, type, true, nthreads);
else if (type==3)
general_dcst<T_dct3<T>>(ain, aout, axes, fct, ortho, nthreads);
general_dcst<T_dcst23<T>>(ain, aout, axes, fct, ortho, type, true, nthreads);
else if (type==4)
general_dcst<T_dct4<T>>(ain, aout, axes, fct, ortho, nthreads);
general_dcst<T_dcst4<T>>(ain, aout, axes, fct, ortho, type, true, nthreads);
else
throw runtime_error("unsupported DCT type");
}
......@@ -3395,13 +3361,13 @@ template<typename T> void dst(const shape_t &shape,
cndarr<T> ain(data_in, shape, stride_in);
ndarr<T> aout(data_out, shape, stride_out);
if (type==1)
general_dcst<T_dst1<T>>(ain, aout, axes, fct, ortho, nthreads);
general_dcst<T_dst1<T>>(ain, aout, axes, fct, ortho, type, false, nthreads);
else if (type==2)
general_dcst<T_dst2<T>>(ain, aout, axes, fct, ortho, nthreads);
general_dcst<T_dcst23<T>>(ain, aout, axes, fct, ortho, type, false, nthreads);
else if (type==3)
general_dcst<T_dst3<T>>(ain, aout, axes, fct, ortho, nthreads);
general_dcst<T_dcst23<T>>(ain, aout, axes, fct, ortho, type, false, nthreads);
else if (type==4)
general_dcst<T_dst4<T>>(ain, aout, axes, fct, ortho, nthreads);
general_dcst<T_dcst4<T>>(ain, aout, axes, fct, ortho, type, false, nthreads);
else
throw runtime_error("unsupported DST type");
}
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment