w4kek/WSJT-X
1
0
Fork 0
mirror of https://github.com/saitohirga/WSJT-X.git synced 2024-11-01 08:07:10 -04:00
Code Issues Packages Projects Releases Wiki Activity
d918b1ab59
WSJT-X/lib/downsam9.f90

88 lines
2.2 KiB
Fortran
Raw Normal View History

Reintegrate Joe's experimental VHF & up features As at ^/branches/wsjtx_exp@5271 git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@5272 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-04-22 13:48:03 -04:00
subroutine downsam9(id2,npts8,nsps8,newdat,nspsd,fpk,c2)
Remove redundant copy of large array in JT9 decoder Also moved the same large array from stack to heap which along with other prior changes now allows the Windows jt9 OpenMP executable to run with a default stack size again. This also removes a crash on the Mac version which was probably due to excessive stack usage. Net result is an even faster JT9 decoder. git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@4942 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-02-08 04:53:20 -05:00
!Downsample from id2() into c2() so as to yield nspsd samples per symbol,
!mixing from fpk down to zero frequency. The downsample factor is 432.
use, intrinsic :: iso_c_binding
use FFTW3
Make Fortran profiling timer function a callback with a default null implementation Groundwork for calling the decoders directly from C/C++ threads. To access the timer module timer_module must now be used. Instrumented code need only use the module function 'timer' which is now a procedure pointer that is guaranteed to be associated (unless null() is assigned to it, which should not be done). The default behaviour of 'timer' is to do nothing. If a Fortran program wishes to profile code it should now use the timer_impl module which contains a default timer implementation. The main program should call 'init_timer([filename])' before using 'timer' or calling routines that are instrumented. If 'init_timer([filename])'. If it is called then an optional file name may be provided with 'timer.out' being used as a default. The procedure 'fini_timer()' may be called to close the file. The default timer implementation is thread safe if used with OpenMP multi-threaded code so long as the OpenMP thread team is given the copyin(/timer_private/) attribute for correct operation. The common block /timer_private/ should be included for OpenMP use by including the file 'timer_common.inc'. The module 'lib/timer_C_wrapper.f90' provides a Fortran wrapper along with 'init' and 'fini' subroutines which allow a C/C++ application to call timer instrumented Fortran code and for it to receive callbacks of 'timer()' subroutine invocations. No C/C++ timer implementation is provided at this stage. git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@6320 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-12-27 10:40:57 -05:00
use timer_module, only: timer
Remove redundant copy of large array in JT9 decoder Also moved the same large array from stack to heap which along with other prior changes now allows the Windows jt9 OpenMP executable to run with a default stack size again. This also removes a crash on the Mac version which was probably due to excessive stack usage. Net result is an even faster JT9 decoder. git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@4942 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-02-08 04:53:20 -05:00
include 'constants.f90'
Use correct type for FFTW3 argument git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@4944 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-02-08 06:10:28 -05:00
integer(C_SIZE_T) NMAX1
Reintegrate Joe's experimental VHF & up features As at ^/branches/wsjtx_exp@5271 git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@5272 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-04-22 13:48:03 -04:00
parameter (NMAX1=653184)
parameter (NFFT1=653184,NFFT2=1512)
Remove redundant copy of large array in JT9 decoder Also moved the same large array from stack to heap which along with other prior changes now allows the Windows jt9 OpenMP executable to run with a default stack size again. This also removes a crash on the Mac version which was probably due to excessive stack usage. Net result is an even faster JT9 decoder. git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@4942 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-02-08 04:53:20 -05:00
type(C_PTR) :: plan !Pointers plan for big FFT
integer*2 id2(0:8*npts8-1)
real*4, pointer :: x1(:)
Reintegrate Joe's experimental VHF & up features As at ^/branches/wsjtx_exp@5271 git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@5272 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-04-22 13:48:03 -04:00
complex c1(0:NFFT1/2)
complex c2(0:NFFT2-1)
Remove redundant copy of large array in JT9 decoder Also moved the same large array from stack to heap which along with other prior changes now allows the Windows jt9 OpenMP executable to run with a default stack size again. This also removes a crash on the Mac version which was probably due to excessive stack usage. Net result is an even faster JT9 decoder. git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@4942 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-02-08 04:53:20 -05:00
real s(5000)
logical first
common/patience/npatience,nthreads
data first/.true./
save plan,first,c1,s,x1
Reintegrate Joe's experimental VHF & up features As at ^/branches/wsjtx_exp@5271 git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@5272 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-04-22 13:48:03 -04:00
df1=12000.0/NFFT1
Remove redundant copy of large array in JT9 decoder Also moved the same large array from stack to heap which along with other prior changes now allows the Windows jt9 OpenMP executable to run with a default stack size again. This also removes a crash on the Mac version which was probably due to excessive stack usage. Net result is an even faster JT9 decoder. git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@4942 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-02-08 04:53:20 -05:00
npts=8*npts8
Reintegrate Joe's experimental VHF & up features As at ^/branches/wsjtx_exp@5271 git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@5272 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-04-22 13:48:03 -04:00
if(npts.gt.NFFT1) npts=NFFT1 !### Fix! ###
Remove redundant copy of large array in JT9 decoder Also moved the same large array from stack to heap which along with other prior changes now allows the Windows jt9 OpenMP executable to run with a default stack size again. This also removes a crash on the Mac version which was probably due to excessive stack usage. Net result is an even faster JT9 decoder. git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@4942 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-02-08 04:53:20 -05:00
if(first) then
nflags=FFTW_ESTIMATE
if(npatience.eq.1) nflags=FFTW_ESTIMATE_PATIENT
if(npatience.eq.2) nflags=FFTW_MEASURE
if(npatience.eq.3) nflags=FFTW_PATIENT
if(npatience.eq.4) nflags=FFTW_EXHAUSTIVE
! Plan the FFTs just once
!$omp critical(fftw) ! serialize non thread-safe FFTW3 calls
plan=fftwf_alloc_real(NMAX1)
call c_f_pointer(plan,x1,[NMAX1])
x1(0:NMAX1-1) => x1 !remap bounds
call fftwf_plan_with_nthreads(nthreads)
Reintegrate Joe's experimental VHF & up features As at ^/branches/wsjtx_exp@5271 git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@5272 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-04-22 13:48:03 -04:00
plan=fftwf_plan_dft_r2c_1d(NFFT1,x1,c1,nflags)
Remove redundant copy of large array in JT9 decoder Also moved the same large array from stack to heap which along with other prior changes now allows the Windows jt9 OpenMP executable to run with a default stack size again. This also removes a crash on the Mac version which was probably due to excessive stack usage. Net result is an even faster JT9 decoder. git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@4942 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-02-08 04:53:20 -05:00
call fftwf_plan_with_nthreads(1)
!$omp end critical(fftw)
first=.false.
endif
if(newdat.eq.1) then
The "big FFTs" should be computed only once for each new set of Rx data. This requires setting newdat=0 after the big FFT is computed. In the OMP code this must be done separately for each mode; so new variables newdat9 and newdat65 have been defined. Both are set to "newdat", the value forwarded from the GUI, each time jt9[_omp][.exe] goes into action. git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@4946 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-02-08 10:07:31 -05:00
x1(0:npts-1)=id2(0:npts-1)
Reintegrate Joe's experimental VHF & up features As at ^/branches/wsjtx_exp@5271 git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@5272 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-04-22 13:48:03 -04:00
x1(npts:NFFT1-1)=0. !Zero the rest of x1
Remove redundant copy of large array in JT9 decoder Also moved the same large array from stack to heap which along with other prior changes now allows the Windows jt9 OpenMP executable to run with a default stack size again. This also removes a crash on the Mac version which was probably due to excessive stack usage. Net result is an even faster JT9 decoder. git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@4942 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-02-08 04:53:20 -05:00
call timer('FFTbig9 ',0)
call fftwf_execute_dft_r2c(plan,x1,c1)
call timer('FFTbig9 ',1)
nadd=int(1.0/df1)
s=0.
do i=1,5000
j=int((i-1)/df1)
do n=1,nadd
j=j+1
s(i)=s(i)+real(c1(j))**2 + aimag(c1(j))**2
enddo
enddo
The "big FFTs" should be computed only once for each new set of Rx data. This requires setting newdat=0 after the big FFT is computed. In the OMP code this must be done separately for each mode; so new variables newdat9 and newdat65 have been defined. Both are set to "newdat", the value forwarded from the GUI, each time jt9[_omp][.exe] goes into action. git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@4946 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-02-08 10:07:31 -05:00
newdat=0
Remove redundant copy of large array in JT9 decoder Also moved the same large array from stack to heap which along with other prior changes now allows the Windows jt9 OpenMP executable to run with a default stack size again. This also removes a crash on the Mac version which was probably due to excessive stack usage. Net result is an even faster JT9 decoder. git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@4942 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-02-08 04:53:20 -05:00
endif
Reintegrate Joe's experimental VHF & up features As at ^/branches/wsjtx_exp@5271 git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@5272 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-04-22 13:48:03 -04:00
ndown=8*nsps8/nspsd !Downsample factor = 432
nh2=NFFT2/2
Remove redundant copy of large array in JT9 decoder Also moved the same large array from stack to heap which along with other prior changes now allows the Windows jt9 OpenMP executable to run with a default stack size again. This also removes a crash on the Mac version which was probably due to excessive stack usage. Net result is an even faster JT9 decoder. git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@4942 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-02-08 04:53:20 -05:00
nf=nint(fpk)
i0=int(fpk/df1)
nw=100
ia=max(1,nf-nw)
ib=min(5000,nf+nw)
call pctile(s(ia),ib-ia+1,40,avenoise)
fac=sqrt(1.0/avenoise)
Reintegrate Joe's experimental VHF & up features As at ^/branches/wsjtx_exp@5271 git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@5272 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-04-22 13:48:03 -04:00
do i=0,NFFT2-1
Remove redundant copy of large array in JT9 decoder Also moved the same large array from stack to heap which along with other prior changes now allows the Windows jt9 OpenMP executable to run with a default stack size again. This also removes a crash on the Mac version which was probably due to excessive stack usage. Net result is an even faster JT9 decoder. git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@4942 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-02-08 04:53:20 -05:00
j=i0+i
Reintegrate Joe's experimental VHF & up features As at ^/branches/wsjtx_exp@5271 git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@5272 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-04-22 13:48:03 -04:00
if(i.gt.nh2) j=j-NFFT2
Remove redundant copy of large array in JT9 decoder Also moved the same large array from stack to heap which along with other prior changes now allows the Windows jt9 OpenMP executable to run with a default stack size again. This also removes a crash on the Mac version which was probably due to excessive stack usage. Net result is an even faster JT9 decoder. git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@4942 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-02-08 04:53:20 -05:00
c2(i)=fac*c1(j)
enddo
Reintegrate Joe's experimental VHF & up features As at ^/branches/wsjtx_exp@5271 git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@5272 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-04-22 13:48:03 -04:00
call four2a(c2,NFFT2,1,1,1) !FFT back to time domain
Remove redundant copy of large array in JT9 decoder Also moved the same large array from stack to heap which along with other prior changes now allows the Windows jt9 OpenMP executable to run with a default stack size again. This also removes a crash on the Mac version which was probably due to excessive stack usage. Net result is an even faster JT9 decoder. git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@4942 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
2015-02-08 04:53:20 -05:00
return
end subroutine downsam9
Reference in New Issue Copy Permalink