Subversion Repositories shark

1999-05-17  fftw  <fftw@glauke.lcs.mit.edu>

	* NEWS: Noted codelet generator speed.

	* gensrc/symmetry.ml, gensrc/genfft.ml, gensrc/fft.ml: Improved

 	real{even|odd}2 transforms

	* doc/fftw.texi: Fixed typo.

	* gensrc/exprdag.ml: Improved simplifier for DCT-type transforms.

	* TODO: Noted that the transpose routines for MPI could use

 	improving.

	* gensrc/symmetry.ml: Fixed comment.

	* gensrc/symmetry.ml, gensrc/genfft.ml, gensrc/fft.ml: Fixed

 	modified DCT/DST generation (realeven2/realodd2) so that it works

 	now.  The simplifier really sucks for this, at the moment.  Why?

1999-05-16  fftw  <fftw@glauke.lcs.mit.edu>

	* gensrc/genfft.ml: Fixed real[even,odd]2 codelet node type

 	output.

	* gensrc/variable.ml, gensrc/variable.mli, gensrc/symmetry.ml,

 	gensrc/genfft.ml: Added generators realeven2 and realodd2 for the

 	modified DCT and DST (i.e. transforms for real data that are

 	even/odd about n=-1/2, not n=0).

	* gensrc/exprdag.ml: Memoized eval for speed reasons.

	* gensrc/util.ml, gensrc/exprdag.ml: Implemented better

 	statistics.

	* gensrc/exprdag.ml: Improved complexity of network transposition

 	from O(n^2)=O(slow) to O(n) [times O(polylog)]

	* TODO: Noted that the generator can now output efficient

 	hard-coded DCT/DST routines of small sizes.

	* gensrc/variable.mli, gensrc/variable.ml, gensrc/symmetry.ml,

 	gensrc/genfft.ml, gensrc/complex.mli, gensrc/complex.ml: Added

 	realeven and realodd codelet generation options.  (They work.)

1999-05-10  fftw  <fftw@glauke.lcs.mit.edu>

	* doc/fftw.texi: Noted how to force compiler choice via CC

 	env. variable.  Also added "compiler" index entry.

1999-05-07  fftw  <fftw@glauke.lcs.mit.edu>

	* TODO, NEWS: Updated TODO.

	* README: Fixed Cilk URL.

	* NEWS: Noted GNU-style long options in the test programs.

1999-05-06  fftw  <fftw@glauke.lcs.mit.edu>

	* tests/test_main.c: --help option no longer causes an exit with

 	an error code.

	* tests/test_main.c, configure.in, fftw/config.h.in: Test programs

 	now use GNU-ly correct long options if getopt_long is available.

	* FAQ/fftw-faq.bfnn, NEWS: MPICH bug fix is apparently not

 	specific to Linux (it also fixed the same problem on some

 	Ultrasparcs).

	* fftw.spec: Updated for 2.1.2.

	* mpi/fftw_mpi_test.c: --only-parallel argument hack should also

 	only be accessed by process 0.

1999-05-05  fftw  <fftw@glauke.lcs.mit.edu>

	* NEWS: Noted addition of omitted fftw_f77_threads_init function.

	* FAQ/fftw-faq.bfnn: Bug fix (dates need to have two digits for

 	the day).

	* threads/rfftw_threads_test.c, threads/fftw_threads_test.c,

 	tests/test_main.c, tests/test_main.h, tests/rfftw_test.c,

 	tests/fftw_test.c, mpi/rfftw_mpi_test.c, mpi/test_transpose_mpi.c,

 	FAQ/fftw-faq.bfnn, mpi/fftw_mpi_test.c, NEWS, configure.in: Fixed

 	bug when running test programs under MPICH; prepared for 2.1.2

 	release.

1999-04-18  stevenj  <stevenj@glauke.lcs.mit.edu>

	* fftw.spec: Made package relocatable.

	* fftw.spec: Created spec file for generating RPM packages of

 	FFTW.

	* FAQ/fftw-faq.bfnn: Revised linker FAQ entry.

1999-04-12  fftw  <fftw@glauke.lcs.mit.edu>

	* tests/test_main.c: Added joke.

1999-04-10  fftw  <fftw@glauke.lcs.mit.edu>

	* FAQ/fftw-faq.bfnn: Added linker FAQ entry.

1999-04-07  stevenj  <stevenj@glauke.lcs.mit.edu>

	* threads/fftw_f77_threads.c: Added missing fftw_f77_threads_init

 	function (thanks to V. Sundararajan for pointing out the

 	omission).

1999-04-06  stevenj  <stevenj@glauke.lcs.mit.edu>

	* configure.in: Check for hcc in addition to mpicc (hcc is used by

 	the LAM MPI implementation).

1999-03-31  stevenj  <stevenj@glauke.lcs.mit.edu>

	* NEWS, doc/fftw.texi, ChangeLog: Fixed credit for 2.1.1 bug fix

 	(real credit goes to Ming-Chang Liu, according to Jeff

 	Briedenbach, whose name was misspelled anyway).

1999-03-30  stevenj  <stevenj@glauke.lcs.mit.edu>

	* FAQ/fftw-faq.bfnn: Moderated language.

	* NEWS: This isn't LaTeX, Matteo; an en dash in a monospaced font

 	is rendered the same as a hyphen (i.e. "-" not "--").

1999-03-29  Matteo Frigo  <athena@supertech.lcs.mit.edu>

	* ChangeLog: Updated

	* NEWS: Minor change.

1999-03-28  Matteo Frigo  <athena@supertech.lcs.mit.edu>

	* tests/rfftw_test.c, tests/fftw_test.c:

	Implemented paranoid check for real->complex and complex->real

	transforms

1999-03-28  Matteo Frigo  <athena@stern.lcs.mit.edu>

	* NEWS:

	Moderated language.  I do not want to claim that the bug occurs in

	`rare' circumstances.  A bug is a bug, period.

	* tests/fftw_test.c, rfftw/rplanner.c:

	Implemented paranoid check for in-place complex planners.

	I still don't know how to check real->complex plans,

	though (apart from rewriting test_ergun for real->complex

	and complex->real)

1999-03-26  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* doc/fftw.texi, FAQ/fftw-faq.bfnn, NEWS:

	Updated documentation for FFTW 2.1.1.

	* configure.in:

	Updated version number for 2.1.1 (as well as shared lib. version).

	* threads/Makefile.am, mpi/Makefile.am: Fixed typo in comment.

	* threads/fftw_threads_test.c, threads/rfftw_threads_test.c,

	mpi/rfftw_mpi_test.c, mpi/fftw_mpi_test.c:

	Added definition of enter_paranoid_mode() to threads & MPI test

	programs.

	* fftw/planner.c:

	Slight modification for future safety (not necessary now).

1999-03-25  Matteo Frigo  <athena@stern.lcs.mit.edu>

	* tests/test_main.c, tests/test_main.h, tests/rfftw_test.c,

	tests/fftw_test.c, fftw/planner.c:

	Added hooks for paranoid verification of every single plan.

1999-03-23  Matteo Frigo  <athena@stern.lcs.mit.edu>

	* fftw/generic.c:

	Array read out of bounds---fixed.  Thanks to Ming-Chang Liu.

1999-03-09  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* FAQ/fftw-faq.bfnn:

	Noted free-ness in "What is FFTW?" description.

1999-03-08  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* threads/fftw_threads.c:

	By default, don't even try to specify PTHREAD_SCOPE_SYSTEM, since

	it causes problems on @!#%$ IRIX 6.5 (in which

	PTHREAD_SCOPE_SYSTEM is not supported, but pthread_attr_setscope

	doesn't return an error!!!).  Just use the default attributes

	(fftw_pthread_attributes_p == NULL).

	* threads/fftw_threads.c: Fixed typo in comment.

	* NEWS: Added 2.1 news.

1999-03-08  Matteo Frigo  <athena@stern.lcs.mit.edu>

	* tests/rfftw_test.c, tests/fftw_test.c:

	Fixed pow(8192, x) --> pow(8192.0, x) .  This failed on

	Digital unix, I don't know why.

1999-03-08  Matteo Frigo  <athena@stern.lcs.mit.edu>

	* ChangeLog: Updated

	* configure.in: Updated version number

1999-03-07  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* doc/fftw.texi: a_slow_array --> a_bad_array

	* doc/fftw.texi:

	Removed all reference to performance from "Dynamic Arrays--The

	Wrong Way" section.

	* doc/fftw.texi:

	Clarified output parameter of Fortran multi-dimensional in-place

	transforms.

	* NEWS, doc/fftw.texi:

	Thanks to Erik Scheirer (boom@sonyx.com) for testing the Mach C

	threads code.

1999-03-06  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* mpi/fftw_mpi_test.c: Added undocumented --only-parallel flag.

	* threads/rfftw_f77_threads.c, threads/fftw_f77_threads.c,

	threads/Makefile.am, doc/fftw.texi, NEWS:

	Added Fortran-callable wrappers for the multi-threaded routines.

1999-02-25  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* threads/fftw_threads.c, threads/fftw_threads-int.h:

	Made pthreads code more tolerant if user doesn't call

	fftw_threads_init.

	* tests/rfftw_test.c, tests/fftw_test.c, mpi/rfftw_mpi_test.c,

	mpi/fftw_mpi_test.c:

	Reduced array sizes somewhat for planner tests.

1999-02-24  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* doc/fftw.texi: Added user thank-you to acknowledgements.

	* doc/fftw.texi: Clarified output format of 1d MPI transforms.

	* FAQ/fftw-faq.bfnn: Small change.

1999-02-24  fftw  <fftw@stern.lcs.mit.edu>

	* FAQ/fftw-faq.bfnn: Changed wording regarding VC++5.0.

	* doc/texi2html: Fixed bug in index generation

1999-02-24  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* doc/fftw.texi: moved configure to fpindex, and added more entries.

	* doc/fftw.texi: Added more padding index entries.

	* doc/fftw.texi: Objective Caml -> Caml in the index.

	* doc/fftw.texi: Added @cindex floating-point precision

	* doc/fftw.texi: Updated index.

	* doc/fftw.texi:

	typo fix: rfftwnd_threads_*_one -> rfftwnd_threads_one_*

	* doc/fftw.texi:

	Print table of contents at the end, as directed by the texinfo

	manual, so that pagination is not screwed up when the TOC has an

	odd number of pages.

1999-02-23  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* doc/fftw.texi: Fixed typo.

	* doc/fftw.texi: Indentation fix.

1999-02-22  Matteo Frigo  <athena@stern.lcs.mit.edu>

	* ChangeLog: Updated

1999-02-22  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* FAQ/fftw-faq.bfnn: Noted workaround for buggy SGI compilers.

	* ChangeLog:

	Line-wrapped (I wish you would remember to do this, Matteo).

	* gensrc/Makefile.rfftw.am, mpi/Makefile.am, threads/Makefile.am,

	gensrc/Makefile.fftw.am:

	Got rid of hackery in header creation, since circular dependencies are

	gone.  Also added prefixed headers to CLEANFILES so they get deleted

	by 'make clean'.

	* configure.in:

	Used xyz for prefix1 instead of just x, to make future conflicts less

	likely.

1999-02-22  Matteo Frigo  <athena@stern.lcs.mit.edu>

	* configure.in: Fix for irix didn't work, let's try this

	* ChangeLog: Fixed typo

	* ChangeLog: Updated

	* configure.in: Accounted for irix thread lossage

	* tests/test_main.c:

	usage() must normally exit, otherwise the test programs tries to

	read argv[] out of bounds.

	* doc/Makefile.am: fftw.ps used compressed fonts

	* gensrc/Makefile.rfftw.am, gensrc/Makefile.fftw.am:

	rfftw.h didn't go into distribution.  Fixed

	* threads/Makefile.am, gensrc/Makefile.rfftw.am, mpi/Makefile.am,

	configure.in, gensrc/Makefile.fftw.am:

        Removed circular dependencies

	* configure.in: Fixed typo

1999-02-22  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* tests/test_main.c:

	Added some new jokes.  Also, print out floating-point precision

	for -v.

	* FAQ/fftw-faq.bfnn: Noted workaround for buggy SGI compilers.

	* ChangeLog:

	Line-wrapped (I wish you would remember to do this, Matteo).

	* gensrc/Makefile.rfftw.am, mpi/Makefile.am, threads/Makefile.am, gensrc/Makefile.fftw.am:

	Got rid of hackery in header creation, since circular dependencies are

	gone.  Also added prefixed headers to CLEANFILES so they get deleted

	by 'make clean'.

	* configure.in:

	Used xyz for prefix1 instead of just x, to make future conflicts less

	likely.

1999-02-22  Matteo Frigo  <athena@stern.lcs.mit.edu>

	* configure.in: Fix for irix didn't work, let's try this

	* ChangeLog: Fixed typo

	* ChangeLog: Updated

	* configure.in: Accounted for irix thread lossage

	* tests/test_main.c:

	usage() must normally exit, otherwise the test programs tries to

	read argv[] out of bounds.

	* doc/Makefile.am: fftw.ps used compressed fonts

	* gensrc/Makefile.rfftw.am, gensrc/Makefile.fftw.am:

	rfftw.h didn't go into distribution.  Fixed

	* threads/Makefile.am, gensrc/Makefile.rfftw.am, mpi/Makefile.am, configure.in, gensrc/Makefile.fftw.am:

	Removed circular dependencies

	* configure.in: Fixed typo

1999-02-21  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* NEWS: Feature set for 2.1 is now frozen (I think).

	* doc/fftw.texi: slight change

	* mpi/Makefile.am, threads/Makefile.am, gensrc/Makefile.fftw.am,

	gensrc/Makefile.rfftw.am: Whoops!  Don't include

	XXX_FFTW_PREFIX_XXX header files in CLEANFILES, because that will

	cause 'make clean' to delete irreplaceable files when no prefix is

	being used.

	* mpi/Makefile.am, threads/Makefile.am, gensrc/Makefile.fftw.am,

	gensrc/Makefile.rfftw.am: Fixed installed headers to use

	FFTW_PREFIX for their own header inclusions.

	* configure.in, doc/fftw.texi, NEWS: Documented

	--enable-type-prefix.

	* mpi/Makefile.am, tests/Makefile.am, threads/Makefile.am,

	gensrc/Makefile.fftw.am, gensrc/Makefile.rfftw.am, acinclude.m4,

	configure.in: Use FFTW_PREFIX for all installed files.

	* configure.in: Use g77 in preference to fort77.

1999-02-21  Matteo Frigo  <athena@stern.lcs.mit.edu>

	* tests/Makefile.am: Fixed forgotted XXX_FFTW_NAME_XXX

	* configure.in: Allowed user specification of [r]fftw-name

1999-02-21  fftw  <fftw@stern.lcs.mit.edu>

	* tests/Makefile.am, configure.in, gensrc/Makefile.fftw.am,

	gensrc/Makefile.rfftw.am, acinclude.m4: (Preliminary) hack: allow

	user to change name of fftw library.

	* configure.in: Incremented shared library version number.

	* configure.in: Added rfftw_mpi.h to list of files to install for

	--enable-mpi.

1999-02-20  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* FAQ/fftw-faq.bfnn: The ascii and info (but not html) versions of

	the FAQ were chopping off the answer to the last question.  Rather

	than fix the bug, I just added a trailing comment like in the

	Linux FAQ, which seems to correct the problem.

	* FAQ/fftw-faq.bfnn: Minor fixes and updates for 2.1.

	* mpi/rfftw_mpi_test.c, mpi/fftw_mpi.h, mpi/fftw_mpi_test.c,

	mpi/fftw_mpi.c: Share plans between forward and backward fftw_mpi

	plans (and between other fftw_mpi plans of the same size).

	* tests/rfftw_test.c, tests/fftw_test.c: In -p test with rank > 1,

	reduce the size of the arrays for the plan.

	* mpi/transpose_mpi.c: Got rid of compiler warning (code was okay,

	though).

	* doc/fftw.texi: small fix.

1999-02-20  fftw  <fftw@stern.lcs.mit.edu>

	* doc/fftw.texi: Minor tweaks

1999-02-20  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* doc/Makefile.am, doc/rfftwnd.gif: fig2dev gif output stinks;

	"manually" add better version of gif figure.

	* doc/fftw.texi: Added link.

	* doc/Makefile.am: Make sure to distribute rfftwnd.fig.

	* doc/fftw.texi: Some post-reorganization grooming.

	* configure.in: Whoops!  Use $enableval, not $withval, in

	AC_ARG_ENABLE.

1999-02-19  Matteo Frigo  <athena@stern.lcs.mit.edu>

	* doc/Makefile.am, doc/fftw.texi: Added picture to postscript

	manual

1999-02-19  fftw  <fftw@stern.lcs.mit.edu>

	* configure.in: Added fort77 to list of fortran compiler (linux

	f2c comes with a program with this name)

	* doc/rfftwnd.gif, doc/rfftwnd.fig, doc/Makefile.am:

	Reverse-engineered rfftwnd picture from postscirpt to fig, so I

	can edit it and produce smaller postscript. (Linux rules!)

	* doc/fftw.texi: Restructured manual

1999-02-19  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* fftw/config.h.in, threads/fftw_threads-int.h, configure.in,

	doc/fftw.texi, NEWS: Added untested support for Mach C threads.

	Also changed three --with options to --enable.

1999-02-19  Matteo Frigo  <athena@stern.lcs.mit.edu>

	* threads/rfftw_threads_test.c, threads/rfftwnd_threads.c,

	threads/rexec2_threads.c, threads/rexec_threads.c,

	threads/rfftw_threads.h, threads/fftw_threads_test.c,

	threads/fftwnd_threads.c, threads/fftw_threads.c,

	threads/fftw_threads.h, threads/executor_threads.c,

	threads/fftw_threads-int.h, tests/test_main.c, tests/fftw_test.c,

	tests/rfftw_test.c, rfftw/rgeneric.c, rfftw/rplanner.c,

	rfftw/rfftwf77.c, rfftw/rfftwnd.c, rfftw/rexec2.c, rfftw/rfftw.h,

	mpi/transpose_mpi.c, rfftw/rexec.c, mpi/sched.h, mpi/test_sched.c,

	mpi/test_transpose_mpi.c, mpi/rfftw_mpi_test.c, mpi/rfftwnd_mpi.c,

	mpi/sched.c, mpi/fftwnd_mpi.c, mpi/rfftw_mpi.h, mpi/fftw_mpi.c,

	mpi/fftw_mpi.h, mpi/fftw_mpi_test.c, mpi/TOMS_transpose.h,

	matlab/fftw.c, mpi/TOMS_transpose.c, gensrc/variable.ml,

	gensrc/variable.mli, gensrc/util.mli, gensrc/twiddle.ml,

	gensrc/util.ml, gensrc/to_c.mli, gensrc/symmetry.ml,

	gensrc/to_c.ml, gensrc/schedule.ml, gensrc/schedule.mli,

	gensrc/number.mli, gensrc/rconfig_prelude, gensrc/magic.ml,

	gensrc/number.ml, gensrc/fft.ml, gensrc/genfft.ml,

	gensrc/exprdag.mli, gensrc/expr.ml, gensrc/expr.mli,

	gensrc/exprdag.ml, gensrc/dag.ml, gensrc/dag.mli,

	gensrc/complex.ml, gensrc/complex.mli, gensrc/asched.mli,

	gensrc/ast.ml, gensrc/asched.ml, fftw/wisdomio.c,

	fortran/f77_test.F, fftw/twiddle.c, fftw/wisdom.c, fftw/rader.c,

	fftw/timer.c, fftw/planner.c, fftw/putils.c, fftw/generic.c,

	fftw/malloc.c, fftw/fftwf77.c, fftw/fftwnd.c, fftw/fftw-int.h,

	fftw/fftw.h.in, fftw/executor.c, fftw/f77_func.h,

	fftw/config.h.in, cilk/time_cilk.cilk, doc/fftw.texi,

	cilk/fftwnd_cilk.cilk, cilk/test_cilk.cilk,

	cilk/executor_cilk.cilk, cilk/fftw_cilk.cilkh, COPYRIGHT,

	ChangeLog: Fixed copyright year

1999-02-19  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* doc/fftw.texi: Minor changes.

	* doc/fftw.texi: Various updates.

	* doc/fftw.texi: Acknowledged Steven's long-suffering advisor.  =)

	* doc/fftw.texi: Noted that 1d complex MPI transforms are

	supported.

	* NEWS: Noted rfftwnd illustration.

	* README: Moved threads, mpi, and fortran descriptions to be under

	"official" header.

	* TODO: Removed parallel rfftw from TODO (it's done).

	* doc/rfftwnd.gif, doc/fftw.texi, doc/Makefile.am: Added rfftwnd

	illustration to the manual (HTML version only).

1999-02-18  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* doc/fftw.texi: rfftwnd_mpi documentation bug fix.

1999-02-16  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* doc/fftw.texi: Noted library requirements of 1D MPI.

	* mpi/fftw_mpi.h, mpi/fftw_mpi.c, doc/fftw.texi: Added scrambled

	input/output options (flags) for 1d mpi transforms.

	* cilk/Makefile: Don't complain if Makefile.common doesn't exist.

	* mpi/fftw_mpi_test.c: Fixed printf.

	* mpi/Makefile.am: Made sure headers are included in make dist.

	* doc/fftw.texi: Documented 1D MPI transforms in manual.

	* mpi/fftw_mpi.c, mpi/fftw_mpi.h, mpi/fftw_mpi_test.c,

	mpi/Makefile.am, NEWS: Added MPI 1D FFT for those crazy guys at

	Caltech.

	* NEWS, doc/fftw.texi: Documented real MPI transforms.

	* mpi/rfftw_mpi_test.c: Small fix.

1999-02-15  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* mpi/rfftw_mpi_test.c, mpi/rfftwnd_mpi.c, mpi/fftw_mpi_test.c,

	mpi/rfftw_mpi.h, configure.in, mpi/Makefile.am: Added rfftw MPI

	transforms.

	* mpi/transpose_mpi.h: this header file is now part of fftw_mpi.h

	* doc/fftw.texi: fixed quote marks.

	* doc/fftw.texi: Fixed typo.

	* doc/fftw.texi: Fixed parallel ref. in intro.

	* doc/fftw.texi: fixed typo

	* doc/fftw.texi: Noted stupid Fortran identifier length

	limitation.

	* doc/fftw.texi: hyphenation fix

	* doc/fftw.texi: Fixed cross-reference.

	* doc/fftw.texi: small fix

	* doc/fftw.texi: Clarifications in the MPI section.

	* doc/fftw.texi: bug fix.

	* doc/fftw.texi: Fixed @uref tags.

	* NEWS: Noted MPI updates.

	* doc/fftw.texi: Added documentation reference to MPI FFTW in

	parallel overview.

	* mpi/README, doc/fftw.texi: Folded MPI documentation into main

	manual.

	* mpi/transpose_mpi.c: bug fix: when a process didn't have any

	local data, we weren't able to distinguish between in-place and

	out-of-place transposes, and were also incorrectly thinking that

	all block sizes were equal.

1999-02-14  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* tests/test_main.h, tests/test_main.c, mpi/test_fftwnd_mpi.c,

	mpi/time_fftwnd_mpi.c, mpi/fftw_mpi_test.c, configure.in,

	mpi/Makefile.am: Use standard fftw test program template for MPI

	tests.

	* fftw/malloc.c: Whoops!

	* fftw/malloc.c: fixed omission from last checkin.

	* configure.in: Fixed typo and updated version number to 2.1.

	* mpi/transpose_mpi.c, mpi/time_fftwnd_mpi.c,

	mpi/test_transpose_mpi.c, mpi/test_fftwnd_mpi.c, mpi/fftwnd_mpi.c,

	mpi/fftw_mpi.h: Added option of doing out-of-place transpose, so

	that we can take advantage of MPI_Alltoall primitive if the user

	provides enough space.

	* threads/fftw_threads.c: Don't require system to support

	PTHREAD_SCOPE_SYSTEM (although we prefer this over

	PTHREAD_SCOPE_PROCESS).

	* gensrc/Makefile.fftw.am, gensrc/Makefile.rfftw.am: Include f77

	wrapper files in sources.

	* mpi/transpose_mpi.c, mpi/test_transpose_mpi.c,

	mpi/time_fftwnd_mpi.c, mpi/fftwnd_mpi.h, mpi/sched.c,

	mpi/test_fftwnd_mpi.c, mpi/fftw_mpi.h, mpi/fftwnd_mpi.c,

	mpi/Makefile.am, mpi/TOMS_transpose.c, mpi/TOMS_transpose.h,

	mpi/Makefile, Makefile.am, configure.in: Integrated mpi stuff into

	automake/autoconf.

	* threads/Makefile.am: Fixed typo in comment.

1999-02-13  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* mpi/transpose_mpi.c, mpi/transpose_mpi.h,

	mpi/test_transpose_mpi.c, mpi/time_fftwnd_mpi.c, mpi/sched.h,

	mpi/test_fftwnd_mpi.c, mpi/test_sched.c, mpi/fftwnd_mpi.h,

	mpi/sched.c, mpi/TOMS_transpose.c, mpi/TOMS_transpose.h,

	mpi/fftwnd_mpi.c, mpi/Makefile: Updated MPI routines.  The new

	implementation is almost completely rewritten from before, and

	should (hopefully) be easier to manage and (maybe) faster in some

	cases.

	* configure.in: Check for pthread_create outside of -lpthread,

	since some systems may have threads built into the standard C

	libraries.

	* fftw/malloc.c: Allow allocation/freeing of 0-size blocks/NULL

	pointers (since that is ANSI-okay, after all).

	* cilk/executor_cilk.cilk: Use ntwiddle instead of r-1 in case

	twiddle policy changes.

	* mpi/time_fftwnd_mpi.c: Whoops!  don't pass uninitialized out

	parameter (shouldn't be a problem in any case since transform is

	in-place, but better safe than sorry).

1999-02-11  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* doc/fftw.texi: Noted necessity of linking threads-using code

	with -lpthread or whatever.

	* rfftw/rfftwf77.c, fortran/rfftw_f77.c, fortran/fortranize.h,

	fortran/fftw_f77.c, fortran/fftw_f77.i, fortran/README,

	fortran/f77_test.F, fftw/f77_func.h, fftw/fftwf77.c,

	fftw/config.h.in, configure.in, doc/fftw.texi, NEWS, acinclude.m4:

	Fortran wrapper functions are now automatically included in the

	main FFTW libraries, unless the --without-fortran option is passed

	to configure.  They are also documented in the main manual.

	* doc/fftw.texi: Fixed broken cross-reference.

1999-02-10  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* doc/fftw.texi: Small fixes to multi-threaded FFTW section.

	* threads/README, doc/fftw.texi, NEWS: Documented threads routines

	in the main FFTW manual.

	* threads/fftw_threads.c: Make sure spawn_loop handles nthreads ==

	0.

	* threads/fftw_threads-int.h, threads/rfftwnd_threads.c,

	threads/executor_threads.c: Set things up so that we could pass

	workspace to fftw_many_inplace_threads.

	* threads/fftw_threads.c: Improved load-balancing of threads

	slightly, although in practice it doesn't seem to make much

	difference.

	* threads/rfftwnd_threads.c, threads/rexec_threads.c,

	threads/rfftw_threads.h, threads/rfftw_threads_test.c: For greater

	consistency, make sure "threads" suffix always directly follows

	rfftw or rfftwnd.

	* threads/rfftwnd_threads.c, threads/rexec2_threads.c,

	threads/rexec_threads.c, threads/rfftw_threads.h,

	threads/fftw_threads-int.h, threads/fftw_threads.c,

	threads/fftwnd_threads.c, threads/executor_threads.c: Cleaned up

	loop-parallelizing interface considerably.

	* threads/rfftwnd_threads.c, threads/rfftw_threads.h,

	threads/rfftw_threads_test.c, threads/rexec2_threads.c,

	threads/rexec_threads.c, threads/executor_threads.c,

	threads/fftw_threads-int.h, threads/fftw_threads.h, configure.in,

	threads/Makefile.am, NEWS: Added first stab at parallel (threads)

	rfftw.

	* rfftw/rfftwnd.c, rfftw/rexec.c: Slight stride/dist fix (to make

	behavior in in-place transforms more in line with what the manual

	says),

1999-02-05  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* tests/test_main.c: Noted -x and -f options in -h help.

	* threads/executor_threads.c: Yikes!  Use ntwiddle instead of r-1

	in twiddle_thread.  (This would have been a problem if we ever

	changed the twiddle policy.)

	* threads/fftwnd_threads.c, threads/executor_threads.c,

	threads/fftw_threads-int.h, fftw/config.h.in: Use autoconf to

	determine if alloca is available and use malloc/free if it is not.

	* threads/fftw_threads_test.c: Call fftw_threads_init when

	starting up.

	* threads/time_threads.c, threads/fftwnd_threads.c,

	threads/test_threads.c, threads/fftw_threads.h,

	threads/fftw_threads_test.c, threads/executor_threads.c,

	threads/fftw_threads-int.h, threads/fftw_threads.c,

	threads/Makefile.am, threads/README, tests/test_main.c,

	tests/test_main.h, tests/rfftw_test.c, configure.in,

	tests/fftw_test.c: Made the threads test program follow the model

	of the uniprocessor test programs.

	* doc/fftw.texi: Documented --with-threads configure option.

	* threads/time_threads.c, threads/Makefile.am, threads/README,

	threads/fftw_threads.h, threads/test_threads.c, threads/Makefile,

	configure.in, fftw/config.h.in, Makefile.am, NEWS: Autoconfiscated

	threads stuff.

1999-02-01  fftw  <fftw@stern.lcs.mit.edu>

	* FAQ/fftw-faq.bfnn: Added entry about VC++ 5.0

1999-01-29  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* NEWS: Noted improvements to <n> argument for test programs.

	* tests/fftw_test.c: Removed unnecessary testing of both real and

	imaginary impulses--this is unnecessary since we check linearity

	under multiplication by complex scalars.

	* tests/test_main.c: (Whoops!)  Turn interactive mode back off

	when getopt is available.

	* tests/test_main.c: Added support for new <n> format to

	interactive mode.

1999-01-29  Matteo Frigo  <athena@stern.lcs.mit.edu>

	* tests/test_main.c, tests/test_main.h, tests/rfftw_test.c,

	tests/fftw_test.c, tests/Makefile.am, tests/README, ChangeLog:

	Allow specification of dimensions in ND test program.

1998-12-11  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* fftw/malloc.c: Use typedefs consistently for fftw_*_hook.

1998-12-09  Matteo Frigo  <athena@stern.lcs.mit.edu>

	* gensrc/dag.ml: Changed to avoid warning from ocaml-2.01

1998-11-17  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* doc/fftw.texi: Updated Cilk URL.

1998-11-11  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* tests/test_main.c: Print FFTW version on -v.

	* tests/Makefile.am: Removed explicit -lm -- it is added

	automatically by configure when it is available.

1998-10-19  Matteo Frigo  <athena@stern.lcs.mit.edu>

	* tests/test_main.c: Replaced 2x2 linear system with fib() in

	timer iteration.  The linear system was converging to 0 in

	floating point, and it caused underflow problems.

1998-10-07  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* configure.in: Fixed usage of ` in echo.

1998-10-06  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* configure.in: Got rid of message about enabling the x86 cycle

	counter--why are we loudly suggesting a deprecated option?

	Interested users can read the manual.

1998-10-01  Matteo Frigo  <athena@stern.lcs.mit.edu>

	* gensrc/genfft.ml: Changed <athfft.h> -> "athfft.h"

1998-09-29  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* FAQ/fftw-faq.bfnn, README: Don't mention the current version

	number, so that we don't have to keep updating these files with

	each new version.

	* README.hacks: Postponed computation of the phase of the moon

	until FFTW 17.0.

	* doc/fftw.texi: Explained "out of place" in tutorial.

	* doc/fftw.texi: Added a couple of clarifications to the

	installation on non-Unix section.

	* NEWS: Fixed typo.

1998-09-28  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* tests/test_main.c: Deleted some unused variables.

1998-09-28  Matteo Frigo  <athena@stern.lcs.mit.edu>

	* fftw/wisdom.c, fftw/config.h.in, fftw/executor.c, Makefile.am,

	configure.in: Moved version number to configure.in and tweaked

	make dist to change config.h with the right version number.

1998-09-27  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* FAQ/fftw-faq.bfnn, NEWS: Noted bug fixes for large rfftwnd

	transforms in 2.0.1.

	* tests/rfftw_test.c: Fixed bug in -s for large multi-dimensional

	transforms.

	* tests/test_main.c: Added -b flag for testing really large

	transforms.

	* fftw/fftwnd.c: Initialize nwork to zero (not really necessary,

	but it doesn't hurt to make sure things are initialized).

	* rfftw/rfftwnd.c: Fixed another parenthesization problem which

	caused overflow problems with rank > 2 transforms.

1998-09-26  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* rfftw/rfftwnd.c: Yeow!  Integer precision overflow occurs for

	final dim. >= 2^16 due to inauspicious parenthesization.  Fixed.

1998-09-25  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* NEWS: Some small changes to 2.0.1 release notes.

1998-09-25  Matteo Frigo  <athena@stern.lcs.mit.edu>

	* gensrc/Makefile.rfftw.am, gensrc/exprdag.ml,

	gensrc/Makefile.fftw.am, NEWS, configure.in, ChangeLog: Changed

	version numbers for 2.0.1 release.

	* tests/fftw_test.c: Added FFTW_OUT_OF_PLACE to flags just in case

	we forget it again in the header file.

	* gensrc/magic.ml, gensrc/genfft.ml, gensrc/exprdag.ml: Added a

	couple of hacks useful for athenafft

1998-09-24  Matteo Frigo  <athena@theory.lcs.mit.edu>

	* gensrc/magic.ml, gensrc/genfft.ml: Added experimental `athena'

	mode

1998-09-23  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* doc/fftw.texi: Fixed another typo.

1998-09-23  Matteo Frigo  <athena@theory.lcs.mit.edu>

	* fftw/fftw.h.in, doc/fftw.texi: Added definition of

	FFTW_OUT_OF_PLACE and fixed typo.

1998-09-23  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* doc/fftw.texi: Noted how to compile in single precision in

	installation for non-unix section.

1998-09-18  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* fftw/fftwnd.c: Modified buffered transforms to use a skew

	(padding) between buffers, and also to do contigous writes instead

	of reads when copying result from buffers back to array.  Thanks

	to Geert van Kempen for the tips.

	* fftw/fftw.h.in: Added Andrew Sterian's patch to allow FFTW to be

	used as a shared library more easily on Win32.

1998-09-17  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* tests/test_main.c, tests/test_main.h, tests/rfftw_test.c,

	tests/fftw_test.c: Added -e flag to use FFTW_ESTIMATE in speed

	tests.

1998-09-16  Matteo Frigo  <athena@theory.lcs.mit.edu>

	* rfftw/rplanner.c, fftw/planner.c: Do not use twiddle(1)

	codelets, even if somebody (like me :-)) includes them

	accidentally in [r]config.c.

1998-09-15  Matteo Frigo  <athena@theory.lcs.mit.edu>

	* doc/fftw.texi: Fixed typo 0 <= i instead of 0 < i

1998-09-15  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* doc/fftw.texi: Added hyphen.

1998-09-15  Matteo Frigo  <athena@theory.lcs.mit.edu>

	* gensrc/variable.ml, gensrc/number.ml, gensrc/exprdag.ml: Better

	hashing scheme for variables.  Removed caching of float value of

	Numbers since it makes little difference with the new hashing

	scheme.

1998-09-15  Matteo Frigo  <athena@theory.lcs.mit.edu>

	* gensrc/number.ml, gensrc/exprdag.ml: Made oracle smarter (as it

	was in 2.0)

1998-09-14  Matteo Frigo  <athena@theory.lcs.mit.edu>

	* gensrc/variable.mli, gensrc/number.ml, gensrc/variable.ml,

	gensrc/Makefile.genfft, gensrc/exprdag.ml: Many, many speed

	improvements

1998-09-14  Matteo Frigo  <athena@theory.lcs.mit.edu>

	* gensrc/Makefile.genfft, gensrc/util.ml: Added timestamp to info

	messages

	* gensrc/util.ml, gensrc/util.mli, gensrc/genfft.ml,

	gensrc/magic.ml, gensrc/exprdag.ml, gensrc/Makefile.genfft: Added

	informative messages

	* gensrc/variable.mli, gensrc/to_c.ml, gensrc/variable.ml,

	gensrc/fft.ml, gensrc/exprdag.ml, gensrc/exprdag.mli,

	gensrc/expr.ml, gensrc/expr.mli, gensrc/complex.mli,

	gensrc/asched.ml, gensrc/complex.ml: Added LittleSimplifier to do

	a first quick simplification pass.  This speeds up the algebraic

	simplifier a lot.

1998-09-14  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* doc/fftw.texi: Fixed library ordering for -lrfftw.

1998-09-13  Matteo Frigo  <athena@theory.lcs.mit.edu>

	* gensrc/number.mli, gensrc/fft.ml, gensrc/number.ml,

	gensrc/exprdag.mli, gensrc/complex.mli, gensrc/exprdag.ml,

	gensrc/complex.ml: Improved speed of generator

1998-09-11  Matteo Frigo  <athena@theory.lcs.mit.edu>

	* acinclude.m4: Dec's compiler does not like -malign-double.

1998-09-11  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* acinclude.m4: Make sure ACX_PROG_CC_EGCS is called.

1998-09-11  Matteo Frigo  <athena@theory.lcs.mit.edu>

	* Makefile.am: Added COPYRIGHT to distribution

	* tests/Makefile.am: Added missing dependency

	* threads/Makefile, cilk/Makefile, configure.in, ChangeLog,

	acinclude.m4: Disabled egcs scheduler.

1998-09-10  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* acinclude.m4: Added ppc/linux flags.

1998-09-10  Matteo Frigo  <athena@theory.lcs.mit.edu>

	* Makefile.am: Added bootstrap.sh to distributed files

	* ChangeLog: Formatted The Right Way

	* gensrc/to_c.mli, gensrc/fft.ml, gensrc/to_c.ml,

	gensrc/exprdag.ml, gensrc/exprdag.mli, gensrc/Makefile.sources,

	gensrc/codelet_prelude, gensrc/Makefile.genfft: Tweaked generator

	to print fused mult/add count and more revision infos

1998-09-09  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* threads/README, threads/fftw_threads.h, threads/Makefile,

	doc/fftw.texi: Noted that Win32 threads code has been reported as

	working.

1998-09-09  Matteo Frigo  <athena@theory.lcs.mit.edu>

	* threads/executor_threads.c, cilk/executor_cilk.cilk,

	gensrc/magic.ml: Fixed parallel transform after X_codelet ->

	fftw_X_codelet conversion

1998-09-08  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* ChangeLog: Line-wrapped.

1998-09-08  Matteo Frigo  <athena@theory.lcs.mit.edu>

	* gensrc/magic.ml, gensrc/genfft.ml, gensrc/Makefile.sources,

	gensrc/exprdag.ml, ChangeLog: More experiments with generation of

	twiddle factors (disabled by default)

1998-09-08  fftw  <fftw@stern.lcs.mit.edu>

	* doc/fftw.texi: Added reference to ocaml 2.00

1998-09-07  Matteo Frigo  <athena@theory.lcs.mit.edu>

	* gensrc/Makefile.sources: Disabled fused madd flags for now

	* gensrc/magic.ml, gensrc/genfft.ml, gensrc/Makefile.sources,

	gensrc/exprdag.ml: Added option to expand all FMA's explicitly,

	even at the cost of increasing the operation count.

	* gensrc/number.ml, gensrc/number.mli, gensrc/magic.ml,

	gensrc/genfft.ml, gensrc/Makefile.sources, gensrc/exprdag.ml:

	Implemented fused-multiply-add rules

1998-09-07  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* fftw/malloc.c: Got rid of COMMA hack.

1998-09-07  Matteo Frigo  <athena@theory.lcs.mit.edu>

	* rfftw/rgeneric.c, rfftw/rplanner.c, rfftw/rexec2.c,

	rfftw/rfftwnd.c, rfftw/.indent.pro, rfftw/rexec.c,

	gensrc/genfft.ml, fftw/putils.c, gensrc/.indent.pro,

	fftw/planner.c, fftw/fftw-int.h, fftw/fftw.h.in, fftw/.indent.pro,

	fftw/executor.c: Alpha-converted X_codelet => fftw_X_codelet to

	avoid namespace pollution.  Replaced ``for (i = 0; i < m; ++i)''

	with ``for (i = m; i > 0; --i)'' in twiddle codelets.

1998-09-05  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* configure.in: Removed redundant inclusion of acinclude.m4.

	* acinclude.m4, configure.in: Moved max. optimization checking

	into acinclude.m4.

	* ChangeLog: Line-wrapped.

	* gensrc/codelet_prelude, gensrc/Makefile.fftw.am,

	gensrc/Makefile.rfftw.am: Changed from perfect to past tense in

	notices about being automatically generated.

	* FAQ/fftw-faq.bfnn: Slight change.

	* fftw/putils.c, fftw/fftw.h.in: added fftw_sizeof_fftw_real().

1998-09-03  Steven G. Johnson  <stevenj@stern.lcs.mit.edu>

	* doc/fftw.texi: Fixed bug in documentation of how factors of 11

	and 13 may be efficiently handled.

1998-09-02  fftw  <fftw@stern.lcs.mit.edu>

	* configure.in: Respect existing CFLAGS in the environment if any

1998-09-02  Matteo Frigo  <athena@theory.lcs.mit.edu>

	* cilk/Makefile, configure.in, acinclude.m4: Added -cpu=pentium et

	al. for latest egcs

1998-09-02  fftw  <fftw@stern.lcs.mit.edu>

	* FAQ/fftw-faq.bfnn: Added FAQ antry about the darned ucbcc

1998-09-01  fftw  <fftw@stern.lcs.mit.edu>

	* FAQ/fftw-faq.bfnn: Mentioned that matlab computes normalized

	FFTs

	* fortran/f77_test.F: Used 'implicit none' to be sure to catch

	Fortran compilers that don't call cpp.

1998-08-31  fftw  <fftw@stern.lcs.mit.edu>

	* ChangeLog: Updated changelog

	* gensrc/Makefile.rfftw.am, gensrc/Makefile.fftw.am: Added

	`automatically generated' notice

	* rfftw/rplanner.c, rfftw/rgeneric.c, rfftw/rfftwnd.c,

	rfftw/rexec2.c, rfftw/rexec.c, gensrc/rconfig_prelude,

	gensrc/config_prelude, configure.in: Tried to be more tolerant of

	systems with nonworking ``const''.

1998-08-30  fftw  <fftw@stern.lcs.mit.edu>

	* gensrc/Makefile.rfftw.am, gensrc/Makefile.fftw.am: Switched

	version number of shared library to 2.0.0.

1998-08-28  fftw  <fftw@stern.lcs.mit.edu>

	* fftw/fftwnd.c: Removed redundant include.

	* doc/fftw.texi: Recommend using GNU make.

	* cilk/Makefile: use rm -f for make clean.

	* doc/fftw.texi: Various small changes.

	* Makefile.am: Fixed bug: CVS dir was not removed from

	distribution

	* threads/Makefile, tests/Makefile.am, mpi/Makefile,

	gensrc/Makefile.rfftw.am, gensrc/Makefile.fftw.am,