Tue Jul 26 20:55:45 EDT 2011 athena@fftw.org * Honor WITH_OUR_MALLOC in libbench2 M ./libbench2/util.c -6 +5 Tue Jul 26 20:27:28 EDT 2011 athena@fftw.org * fixed typo: incorrect name of combined threads library on Windows M ./Makefile.am -1 +1 Mon Jul 25 14:38:20 EDT 2011 stevenj@fftw.org * 3.3 version bump & NEWS M ./NEWS -5 +16 M ./configure.ac -1 +1 Mon Jul 25 14:37:48 EDT 2011 stevenj@fftw.org * use int(..., C_SIZE_T) rather than declaring another variable in the Fortran examples M ./doc/modern-fortran.texi -12 +9 Wed Jul 13 05:02:32 EDT 2011 stevenj@fftw.org * typo, thanks to Rhys Ulerich for the comment M ./doc/mpi.texi -1 +1 Mon Jul 11 14:39:52 EDT 2011 athena@fftw.org * Fix bug in bubblesort Bubblesort was not sorting. This was a bug in the benchmark library (not in FFTW per se), and it impacted the benchmark program with --report-time and --report-mflops causing it to output an incorrect value for the median. (The minimum, maximum, and average value were correct.) Thanks Dima Baksheev of Intel for reporting this bug. M ./libbench2/report.c -2 +2 Fri Jul 8 13:35:59 EDT 2011 stevenj@fftw.org * small manual typos M ./doc/modern-fortran.texi -2 +2 Wed Jul 6 10:49:40 EDT 2011 athena@fftw.org * Detection of altivec.h requires $ALTIVEC_CFLAGS M ./configure.ac -5 +4 Tue Jul 5 19:58:47 EDT 2011 athena@fftw.org * Introduce fake dependency so that my-getopt.c is recompiled my-getopt.c does not depend on anything, and so it is not rebuilt when reconfiguring for a different ISA (e.g., CC="gcc -m32" vs CC="gcc -m64"). Add a fake dependency on so that the file is recompiled. M ./libbench2/my-getopt.c +1 Tue Jul 5 18:53:36 EDT 2011 stevenj@fftw.org * support compiling/installing --enable-threads --enable-openmp at the same time, although in this case the test program only uses the threads variety. Update documentation accordingly, and in general expand the documentation of the OpenMP support M ./Makefile.am -1 +1 M ./configure.ac -21 +13 M ./doc/install.texi -14 +13 M ./doc/threads.texi -19 +43 M ./mpi/Makefile.am -2 +6 M ./tests/Makefile.am -2 +6 M ./tests/fftw-bench.c -1 +3 M ./threads/Makefile.am -6 +20 M ./threads/conf.c -4 M ./threads/openmp.c -4 M ./threads/threads.c -3 M ./tools/Makefile.am -4 +8 Tue Jul 5 16:04:03 EDT 2011 stevenj@fftw.org * call omp_set_num_threads in fftw-bench so that the number of OpenMP threads corresponds with the number of FFTW threads M ./tests/fftw-bench.c +5 Tue Jul 5 16:03:06 EDT 2011 stevenj@fftw.org * when --enable-openmp, install as fftw3_omp rather than fftw3_threads, so that both the POSIX threads and OpenMP variants of FFTW can be installed at once M ./Makefile.am -1 +1 M ./configure.ac +4 M ./threads/Makefile.am -4 +4 Sat Jul 2 02:21:22 EDT 2011 stevenj@fftw.org * don't even declare an fftw_execute interface in Fortran, since it is unsafe and we recommend against it anyway; thanks to Arjen Markus for the suggestion M ./api/f03api.sh -1 +1 M ./doc/modern-fortran.texi -2 +4 Fri Jul 1 14:35:44 EDT 2011 athena@fftw.org * consistently use the order single, double, long double M ./support/BUILD-MINGW32.sh -1 +1 M ./support/BUILD-MINGW64.sh -1 +1 Wed Jun 29 17:27:06 EDT 2011 athena@fftw.org * MSVC AVX 64-bit detection does not work, punt for now. M ./simd-support/amd64-cpuid.h -16 +2 Wed Jun 29 15:52:27 EDT 2011 stevenj@fftw.org * fixed typo, added note on transposed flags for r2c/c2r; thanks to Rhys Ulerich for the suggestions M ./doc/mpi.texi -1 +14 Wed Jun 29 09:41:39 EDT 2011 athena@fftw.org * fixes for compiling with MSVC (untested) M ./simd-support/amd64-cpuid.h -2 +16 M ./simd-support/simd-avx.h -11 +16 M ./simd-support/x86-cpuid.h -1 +7 Tue Jun 28 16:48:36 EDT 2011 athena@fftw.org * comment M ./simd-support/sse2-nonportable.c +3 Mon Jun 27 21:01:56 EDT 2011 stevenj@fftw.org * rm extraneous line break in HTML output ... I hate texinfo M ./doc/reference.texi -6 +3 Mon Jun 27 06:43:57 EDT 2011 athena@fftw.org tagged fftw-3.3-beta1 Mon Jun 27 00:47:33 EDT 2011 stevenj@fftw.org * maintainer-clean should delete html directory (otherwwise we keep obsolete HTML files in the dist tarball, sigh) M ./doc/Makefile.am -2 +2 Mon Jun 27 00:45:28 EDT 2011 stevenj@fftw.org tagged fftw-3.3-beta1 Sun Jun 26 23:36:32 EDT 2011 stevenj@fftw.org * update copyright year in manual M ./doc/license.texi -2 +2 Sun Jun 26 22:52:54 EDT 2011 stevenj@fftw.org * whoops, don't dist .f03 headers, since those are built by the user's Makefile M ./api/Makefile.am -1 +2 M ./mpi/Makefile.am -1 +2 Sun Jun 26 22:43:49 EDT 2011 stevenj@fftw.org * fix embarrassing deadlock/crashing bug in my previous nowisdom_hook fix -- I forgot to handle the case where one process has wisdom and another one doesn't, requiring a nowisdom_hook in the latter case; this should only affect MPI transforms since otherwise these hook functions are NULL M ./kernel/ifftw.h +1 M ./kernel/planner.c -44 +48 M ./mpi/api.c +14 Sun Jun 26 21:02:15 EDT 2011 stevenj@fftw.org * subsubheadings, MPI transpose reference M ./doc/mpi.texi -7 +53 Sun Jun 26 20:48:53 EDT 2011 stevenj@fftw.org * add MPI plan reference M ./doc/mpi.texi -5 +197 Sun Jun 26 17:07:21 EDT 2011 stevenj@fftw.org * portions of MPI reference docs; tweaks to NEWS M ./NEWS -2 +3 M ./doc/mpi.texi -12 +264 Sun Jun 26 12:40:43 EDT 2011 stevenj@fftw.org * use $(CHECK_PL_OPTS) more consistently M ./tests/Makefile.am -1 +1 Sun Jun 26 10:04:54 EDT 2011 athena@fftw.org * accept \r\n as well as \n. Grrr... M ./tests/check.pl -1 +2 Sun Jun 26 09:52:11 EDT 2011 athena@fftw.org * new configure option --with-incoming-stack-boundary=N This option selects CFLAGS to align the stack at all externally-callable functions. This currently comprises api/* and threads/* M ./api/Makefile.am +1 M ./configure.ac +14 M ./support/BUILD-MINGW32.sh -1 +1 M ./threads/Makefile.am +1 Sun Jun 26 09:51:37 EDT 2011 athena@fftw.org * add -fomit-frame-pointer back Somehow -O3 does not imply -fomit-frame-pointer on ia32 M ./m4/ax_cc_maxopt.m4 -2 +3 Sun Jun 26 07:20:27 EDT 2011 athena@fftw.org * Note that removal of mips-ps is temporary. M ./NEWS -3 +5 Sat Jun 25 23:15:03 EDT 2011 stevenj@fftw.org * update copyright year M ./COPYRIGHT -2 +2 M ./api/api.h -2 +2 M ./api/apiplan.c -2 +2 M ./api/configure.c -2 +2 M ./api/execute-dft-c2r.c -2 +2 M ./api/execute-dft-r2c.c -2 +2 M ./api/execute-dft.c -2 +2 M ./api/execute-r2r.c -2 +2 M ./api/execute-split-dft-c2r.c -2 +2 M ./api/execute-split-dft-r2c.c -2 +2 M ./api/execute-split-dft.c -2 +2 M ./api/execute.c -2 +2 M ./api/export-wisdom-to-file.c -2 +2 M ./api/export-wisdom-to-string.c -2 +2 M ./api/export-wisdom.c -2 +2 M ./api/f77api.c -2 +2 M ./api/f77funcs.h -2 +2 M ./api/fftw3.h -2 +2 M ./api/flops.c -2 +2 M ./api/forget-wisdom.c -2 +2 M ./api/import-system-wisdom.c -2 +2 M ./api/import-wisdom-from-file.c -2 +2 M ./api/import-wisdom-from-string.c -2 +2 M ./api/import-wisdom.c -2 +2 M ./api/malloc.c -2 +2 M ./api/map-r2r-kind.c -2 +2 M ./api/mapflags.c -2 +2 M ./api/mkprinter-file.c -2 +2 M ./api/mktensor-iodims.h -2 +2 M ./api/mktensor-rowmajor.c -2 +2 M ./api/plan-dft-1d.c -2 +2 M ./api/plan-dft-2d.c -2 +2 M ./api/plan-dft-3d.c -2 +2 M ./api/plan-dft-c2r-1d.c -2 +2 M ./api/plan-dft-c2r-2d.c -2 +2 M ./api/plan-dft-c2r-3d.c -2 +2 M ./api/plan-dft-c2r.c -2 +2 M ./api/plan-dft-r2c-1d.c -2 +2 M ./api/plan-dft-r2c-2d.c -2 +2 M ./api/plan-dft-r2c-3d.c -2 +2 M ./api/plan-dft-r2c.c -2 +2 M ./api/plan-dft.c -2 +2 M ./api/plan-guru-dft-c2r.h -2 +2 M ./api/plan-guru-dft-r2c.h -2 +2 M ./api/plan-guru-dft.h -2 +2 M ./api/plan-guru-r2r.h -2 +2 M ./api/plan-guru-split-dft-c2r.h -2 +2 M ./api/plan-guru-split-dft-r2c.h -2 +2 M ./api/plan-guru-split-dft.h -2 +2 M ./api/plan-many-dft-c2r.c -2 +2 M ./api/plan-many-dft-r2c.c -2 +2 M ./api/plan-many-dft.c -2 +2 M ./api/plan-many-r2r.c -2 +2 M ./api/plan-r2r-1d.c -2 +2 M ./api/plan-r2r-2d.c -2 +2 M ./api/plan-r2r-3d.c -2 +2 M ./api/plan-r2r.c -2 +2 M ./api/print-plan.c -2 +2 M ./api/rdft2-pad.c -2 +2 M ./api/the-planner.c -2 +2 M ./api/version.c -2 +2 M ./api/x77.h -2 +2 M ./dft/bluestein.c -2 +2 M ./dft/buffered.c -2 +2 M ./dft/codelet-dft.h -2 +2 M ./dft/conf.c -2 +2 M ./dft/ct.c -2 +2 M ./dft/ct.h -2 +2 M ./dft/dft.h -2 +2 M ./dft/dftw-direct.c -2 +2 M ./dft/dftw-directsq.c -2 +2 M ./dft/dftw-generic.c -2 +2 M ./dft/dftw-genericbuf.c -2 +2 M ./dft/direct.c -2 +2 M ./dft/generic.c -2 +2 M ./dft/indirect-transpose.c -2 +2 M ./dft/indirect.c -2 +2 M ./dft/kdft-dif.c -2 +2 M ./dft/kdft-difsq.c -2 +2 M ./dft/kdft-dit.c -2 +2 M ./dft/kdft.c -2 +2 M ./dft/nop.c -2 +2 M ./dft/plan.c -2 +2 M ./dft/problem.c -2 +2 M ./dft/rader.c -2 +2 M ./dft/rank-geq2.c -2 +2 M ./dft/scalar/n.c -2 +2 M ./dft/scalar/n.h -2 +2 M ./dft/scalar/t.c -2 +2 M ./dft/scalar/t.h -2 +2 M ./dft/simd/common/genus.c -2 +2 M ./dft/simd/n1b.h -2 +2 M ./dft/simd/n1f.h -2 +2 M ./dft/simd/n2b.h -2 +2 M ./dft/simd/n2f.h -2 +2 M ./dft/simd/n2s.h -2 +2 M ./dft/simd/q1b.h -2 +2 M ./dft/simd/q1f.h -2 +2 M ./dft/simd/t1b.h -2 +2 M ./dft/simd/t1bu.h -2 +2 M ./dft/simd/t1f.h -2 +2 M ./dft/simd/t1fu.h -2 +2 M ./dft/simd/t2b.h -2 +2 M ./dft/simd/t2f.h -2 +2 M ./dft/simd/t3b.h -2 +2 M ./dft/simd/t3f.h -2 +2 M ./dft/simd/ts.h -2 +2 M ./dft/solve.c -2 +2 M ./dft/vrank-geq1.c -2 +2 M ./dft/zero.c -2 +2 M ./doc/f77_wisdom.f -2 +2 M ./genfft/algsimp.ml -2 +2 M ./genfft/algsimp.mli -2 +2 M ./genfft/annotate.ml -2 +2 M ./genfft/annotate.mli -2 +2 M ./genfft/assoctable.ml -2 +2 M ./genfft/assoctable.mli -2 +2 M ./genfft/c.ml -2 +2 M ./genfft/c.mli -2 +2 M ./genfft/complex.ml -2 +2 M ./genfft/complex.mli -2 +2 M ./genfft/conv.ml -2 +2 M ./genfft/conv.mli -2 +2 M ./genfft/dag.ml -2 +2 M ./genfft/dag.mli -2 +2 M ./genfft/expr.ml -2 +2 M ./genfft/expr.mli -2 +2 M ./genfft/fft.ml -2 +2 M ./genfft/fft.mli -2 +2 M ./genfft/gen_hc2c.ml -2 +2 M ./genfft/gen_hc2cdft.ml -2 +2 M ./genfft/gen_hc2cdft_c.ml -2 +2 M ./genfft/gen_hc2hc.ml -2 +2 M ./genfft/gen_mdct.ml -2 +2 M ./genfft/gen_notw.ml -2 +2 M ./genfft/gen_notw_c.ml -2 +2 M ./genfft/gen_r2cb.ml -2 +2 M ./genfft/gen_r2cf.ml -2 +2 M ./genfft/gen_r2r.ml -2 +2 M ./genfft/gen_twiddle.ml -2 +2 M ./genfft/gen_twiddle_c.ml -2 +2 M ./genfft/gen_twidsq.ml -2 +2 M ./genfft/gen_twidsq_c.ml -2 +2 M ./genfft/genutil.ml -2 +2 M ./genfft/littlesimp.ml -2 +2 M ./genfft/littlesimp.mli -2 +2 M ./genfft/magic.ml -2 +2 M ./genfft/monads.ml -2 +2 M ./genfft/number.ml -2 +2 M ./genfft/number.mli -2 +2 M ./genfft/oracle.ml -2 +2 M ./genfft/oracle.mli -2 +2 M ./genfft/schedule.ml -2 +2 M ./genfft/schedule.mli -2 +2 M ./genfft/simd.ml -2 +2 M ./genfft/simd.mli -2 +2 M ./genfft/simdmagic.ml -2 +2 M ./genfft/to_alist.ml -2 +2 M ./genfft/to_alist.mli -2 +2 M ./genfft/trig.ml -2 +2 M ./genfft/trig.mli -2 +2 M ./genfft/twiddle.ml -2 +2 M ./genfft/twiddle.mli -2 +2 M ./genfft/unique.ml -2 +2 M ./genfft/unique.mli -2 +2 M ./genfft/util.ml -2 +2 M ./genfft/util.mli -2 +2 M ./genfft/variable.ml -2 +2 M ./genfft/variable.mli -2 +2 M ./kernel/align.c -2 +2 M ./kernel/alloc.c -2 +2 M ./kernel/assert.c -2 +2 M ./kernel/awake.c -2 +2 M ./kernel/buffered.c -2 +2 M ./kernel/cpy1d.c -2 +2 M ./kernel/cpy2d-pair.c -2 +2 M ./kernel/cpy2d.c -2 +2 M ./kernel/ct.c -2 +2 M ./kernel/cycle.h -2 +2 M ./kernel/debug.c -2 +2 M ./kernel/extract-reim.c -2 +2 M ./kernel/hash.c -2 +2 M ./kernel/iabs.c -2 +2 M ./kernel/ifftw.h -2 +2 M ./kernel/kalloc.c -2 +2 M ./kernel/md5-1.c -2 +2 M ./kernel/md5.c -2 +2 M ./kernel/minmax.c -2 +2 M ./kernel/ops.c -2 +2 M ./kernel/pickdim.c -2 +2 M ./kernel/plan.c -2 +2 M ./kernel/primes.c -2 +2 M ./kernel/print.c -2 +2 M ./kernel/problem.c -2 +2 M ./kernel/rader.c -2 +2 M ./kernel/scan.c -2 +2 M ./kernel/solver.c -2 +2 M ./kernel/solvtab.c -2 +2 M ./kernel/stride.c -2 +2 M ./kernel/tensor.c -2 +2 M ./kernel/tensor1.c -2 +2 M ./kernel/tensor2.c -2 +2 M ./kernel/tensor3.c -2 +2 M ./kernel/tensor4.c -2 +2 M ./kernel/tensor5.c -2 +2 M ./kernel/tensor7.c -2 +2 M ./kernel/tensor8.c -2 +2 M ./kernel/tensor9.c -2 +2 M ./kernel/tile2d.c -2 +2 M ./kernel/timer.c -2 +2 M ./kernel/transpose.c -2 +2 M ./kernel/trig.c -2 +2 M ./kernel/twiddle.c -2 +2 M ./libbench2/dotens2.c -2 +2 M ./libbench2/my-getopt.c -2 +2 M ./libbench2/my-getopt.h -2 +2 M ./libbench2/verify-dft.c -2 +2 M ./libbench2/verify-lib.c -2 +2 M ./libbench2/verify-r2r.c -2 +2 M ./libbench2/verify-rdft2.c -2 +2 M ./libbench2/verify.h -2 +2 M ./mpi/any-true.c -2 +2 M ./mpi/api.c -2 +2 M ./mpi/block.c -2 +2 M ./mpi/choose-radix.c -2 +2 M ./mpi/conf.c -2 +2 M ./mpi/dft-problem.c -2 +2 M ./mpi/dft-rank-geq2-transposed.c -2 +2 M ./mpi/dft-rank-geq2.c -2 +2 M ./mpi/dft-rank1-bigvec.c -2 +2 M ./mpi/dft-rank1.c -2 +2 M ./mpi/dft-serial.c -2 +2 M ./mpi/dft-solve.c -2 +2 M ./mpi/dtensor.c -2 +2 M ./mpi/fftw3-mpi.h -2 +2 M ./mpi/ifftw-mpi.h -2 +2 M ./mpi/mpi-dft.h -2 +2 M ./mpi/mpi-rdft.h -2 +2 M ./mpi/mpi-rdft2.h -2 +2 M ./mpi/mpi-transpose.h -2 +2 M ./mpi/rdft-problem.c -2 +2 M ./mpi/rdft-rank-geq2-transposed.c -2 +2 M ./mpi/rdft-rank-geq2.c -2 +2 M ./mpi/rdft-rank1-bigvec.c -2 +2 M ./mpi/rdft-serial.c -2 +2 M ./mpi/rdft-solve.c -2 +2 M ./mpi/rdft2-problem.c -2 +2 M ./mpi/rdft2-rank-geq2-transposed.c -2 +2 M ./mpi/rdft2-rank-geq2.c -2 +2 M ./mpi/rdft2-serial.c -2 +2 M ./mpi/rdft2-solve.c -2 +2 M ./mpi/rearrange.c -2 +2 M ./mpi/testsched.c -1 +1 M ./mpi/transpose-alltoall.c -2 +2 M ./mpi/transpose-pairwise.c -2 +2 M ./mpi/transpose-problem.c -2 +2 M ./mpi/transpose-recurse.c -2 +2 M ./mpi/transpose-solve.c -2 +2 M ./mpi/wisdom-api.c -2 +2 M ./rdft/buffered.c -2 +2 M ./rdft/buffered2.c -2 +2 M ./rdft/codelet-rdft.h -2 +2 M ./rdft/conf.c -2 +2 M ./rdft/ct-hc2c-direct.c -2 +2 M ./rdft/ct-hc2c.c -2 +2 M ./rdft/ct-hc2c.h -2 +2 M ./rdft/dft-r2hc.c -2 +2 M ./rdft/dht-r2hc.c -2 +2 M ./rdft/dht-rader.c -2 +2 M ./rdft/direct-r2c.c -2 +2 M ./rdft/direct-r2r.c -2 +2 M ./rdft/direct2.c -2 +2 M ./rdft/generic.c -2 +2 M ./rdft/hc2hc-direct.c -2 +2 M ./rdft/hc2hc-generic.c -2 +2 M ./rdft/hc2hc.c -2 +2 M ./rdft/hc2hc.h -2 +2 M ./rdft/indirect.c -2 +2 M ./rdft/khc2c.c -2 +2 M ./rdft/khc2hc.c -2 +2 M ./rdft/kr2c.c -2 +2 M ./rdft/kr2r.c -2 +2 M ./rdft/nop.c -2 +2 M ./rdft/nop2.c -2 +2 M ./rdft/plan.c -2 +2 M ./rdft/plan2.c -2 +2 M ./rdft/problem.c -2 +2 M ./rdft/problem2.c -2 +2 M ./rdft/rank-geq2-rdft2.c -2 +2 M ./rdft/rank-geq2.c -2 +2 M ./rdft/rank0-rdft2.c -2 +2 M ./rdft/rank0.c -2 +2 M ./rdft/rdft-dht.c -2 +2 M ./rdft/rdft.h -2 +2 M ./rdft/rdft2-inplace-strides.c -2 +2 M ./rdft/rdft2-rdft.c -2 +2 M ./rdft/rdft2-strides.c -2 +2 M ./rdft/rdft2-tensor-max-index.c -2 +2 M ./rdft/scalar/hb.h -2 +2 M ./rdft/scalar/hc2c.c -2 +2 M ./rdft/scalar/hc2cb.h -2 +2 M ./rdft/scalar/hc2cf.h -2 +2 M ./rdft/scalar/hf.h -2 +2 M ./rdft/scalar/hfb.c -2 +2 M ./rdft/scalar/r2c.c -2 +2 M ./rdft/scalar/r2cb.h -2 +2 M ./rdft/scalar/r2cbIII.h -2 +2 M ./rdft/scalar/r2cf.h -2 +2 M ./rdft/scalar/r2cfII.h -2 +2 M ./rdft/scalar/r2r.c -2 +2 M ./rdft/scalar/r2r.h -2 +2 M ./rdft/simd/common/genus.c -2 +2 M ./rdft/simd/hc2cbv.h -2 +2 M ./rdft/simd/hc2cfv.h -2 +2 M ./rdft/solve.c -2 +2 M ./rdft/solve2.c -2 +2 M ./rdft/vrank-geq1-rdft2.c -2 +2 M ./rdft/vrank-geq1.c -2 +2 M ./rdft/vrank3-transpose.c -2 +2 M ./reodft/conf.c -2 +2 M ./reodft/redft00e-r2hc-pad.c -2 +2 M ./reodft/redft00e-r2hc.c -2 +2 M ./reodft/reodft.h -2 +2 M ./reodft/reodft010e-r2hc.c -2 +2 M ./reodft/reodft11e-r2hc-odd.c -2 +2 M ./reodft/reodft11e-r2hc.c -2 +2 M ./reodft/reodft11e-radix2.c -2 +2 M ./reodft/rodft00e-r2hc-pad.c -2 +2 M ./reodft/rodft00e-r2hc.c -2 +2 M ./simd/altivec.c -2 +2 M ./simd/nonportable/sse.c -2 +2 M ./simd/nonportable/sse2.c -2 +2 M ./simd/simd-altivec.h -2 +2 M ./simd/simd-sse.h -2 +2 M ./simd/simd-sse2.h -2 +2 M ./simd/simd.h -2 +2 M ./simd/sse.c -2 +2 M ./simd/sse2.c -2 +2 M ./simd/taint.c -2 +2 M ./simd/x86-cpuid.h -2 +2 M ./simd-support/altivec.c -2 +2 M ./simd-support/amd64-cpuid.h -2 +2 M ./simd-support/avx.c -2 +2 M ./simd-support/simd-altivec.h -2 +2 M ./simd-support/simd-avx.h -2 +2 M ./simd-support/simd-common.h -2 +2 M ./simd-support/simd-sse2.h -2 +2 M ./simd-support/sse2-nonportable.c -2 +2 M ./simd-support/sse2.c -2 +2 M ./simd-support/taint.c -2 +2 M ./simd-support/x86-cpuid.h -2 +2 M ./threads/api.c -2 +2 M ./threads/conf.c -2 +2 M ./threads/ct.c -2 +2 M ./threads/dft-vrank-geq1.c -2 +2 M ./threads/f77api.c -2 +2 M ./threads/f77funcs.h -2 +2 M ./threads/hc2hc.c -2 +2 M ./threads/openmp.c -2 +2 M ./threads/rdft-vrank-geq1.c -2 +2 M ./threads/threads.c -2 +2 M ./threads/threads.h -2 +2 M ./threads/vrank-geq1-rdft2.c -2 +2 M ./tools/fftw-wisdom-to-conf.1 -4 +4 M ./tools/fftw-wisdom-to-conf.in -2 +2 M ./tools/fftw-wisdom.c -2 +2 M ./tools/fftw_wisdom.1.in -4 +4 Sat Jun 25 21:33:13 EDT 2011 stevenj@fftw.org * updated NEWS M ./NEWS -4 +53 Sat Jun 25 20:29:55 EDT 2011 stevenj@fftw.org * fixes to Fortran interface and docs M ./api/genf03.pl -12 +22 M ./doc/modern-fortran.texi -5 +14 M ./doc/mpi.texi -13 +16 Sat Jun 25 17:43:31 EDT 2011 stevenj@fftw.org * initial stab at MPI Fortran docs M ./doc/mpi.texi +138 Sat Jun 25 16:43:31 EDT 2011 stevenj@fftw.org * correct description of what MPI standard says about I/O (I can't believe this crap) M ./doc/mpi.texi -11 +21 Sat Jun 25 15:14:07 EDT 2011 stevenj@fftw.org * more MPI documentation; mention `fftw_alloc' functions earlier in the manual M ./doc/fftw3.texi +12 M ./doc/mpi.texi -30 +158 M ./doc/other.texi +9 M ./doc/tutorial.texi -11 +14 Sat Jun 25 13:40:19 EDT 2011 stevenj@fftw.org * clarification about --enable-sse2 M ./doc/install.texi -6 +8 Sat Jun 25 13:31:25 EDT 2011 athena@fftw.org * Update mingw build scripts for fftw-3.3 M ./kernel/kalloc.c -2 +2 M ./support/BUILD-MINGW32.sh -3 +3 M ./support/BUILD-MINGW64.sh -4 +4 Sat Jun 25 08:52:13 EDT 2011 athena@fftw.org * Fix typo: EXTRADIST => EXTRA_DIST M ./mpi/Makefile.am -1 +1 Fri Jun 24 23:52:19 EDT 2011 stevenj@fftw.org * finished draft "modern fortran" chapter M ./doc/modern-fortran.texi -9 +91 Fri Jun 24 20:47:49 EDT 2011 stevenj@fftw.org * include FFTW_EXTERN prototypes for wrappers, so that they are properly exported to DLLs on Windows (sigh) M ./mpi/f03-wrap.sh +4 Fri Jun 24 16:52:30 EDT 2011 athena@fftw.org * use malloc() instead of alloca() for large buffers The proximate cause for this patch is that OpenBSD/i386 reserves 256KB stack size per thread. We were allocating a buffer of size 128*130*sizeof(fftw_complex) that exceeds the stack. While 128*130*sizeof(fftw_complex) = 260KiB is the worst case for normal configurations, it is a good idea to limit stack allocation just in case. Also, the generic solver might in principle generate unbounded buffers, even though it is normally disabled for n > 137. So, as an added precaution, we now never stack-allocate buffers larger than 64KiB, which ought to be enough for anybody. M ./dft/dftw-direct.c -2 +3 M ./dft/direct.c -2 +3 M ./dft/generic.c -2 +3 M ./kernel/ifftw.h -7 +31 M ./rdft/ct-hc2c-direct.c -2 +3 M ./rdft/direct-r2c.c -2 +3 M ./rdft/generic.c -4 +6 M ./rdft/hc2hc-direct.c -2 +3 Fri Jun 24 16:32:30 EDT 2011 stevenj@fftw.org * don't imply that AVX is available on Pentium III; note that MIPS Paired Single is currently only in FFTW 3.2.x M ./doc/install.texi -3 +2 M ./doc/other.texi -3 +4 Fri Jun 24 16:05:27 EDT 2011 stevenj@fftw.org * silence annoying gfortran warnings M ./api/genf03.pl -1 +6 M ./configure.ac +14 M ./mpi/Makefile.am -3 +6 Fri Jun 24 14:59:30 EDT 2011 stevenj@fftw.org * a couple MPI Fortran 2003 fixes; changed MPI flags to not use 1<<31 since Fortran (not having unsigned integers) does not allow us to declare that constant in a portable way M ./mpi/api.c -1 +1 M ./mpi/f03api.sh -2 +2 M ./mpi/fftw3-mpi.h -4 +4 Fri Jun 24 15:05:05 EDT 2011 athena@fftw.org * Fix libtool shared version info. FFTW-3.3.x should be a direct drop-in replacement for all FFTW-3.x.y versions. M ./configure.ac -1 +4 Fri Jun 24 14:38:47 EDT 2011 stevenj@fftw.org * add MPI Fortran API and wrappers M ./api/Makefile.am -2 +3 M ./api/f03api.sh -3 +5 M ./api/genf03.pl -2 +14 M ./mpi/Makefile.am -3 +14 R ./mpi/README A ./mpi/f03-wrap.sh A ./mpi/f03api.sh M ./mpi/fftw3-mpi.h -4 +4 A ./mpi/genf03-wrap.pl Fri Jun 24 14:51:12 EDT 2011 athena@fftw.org * Do not require fig2dev on the user's machine Distribute the manual's figures in PDF/PS/PNG form instead. M ./doc/Makefile.am -10 +16 Fri Jun 24 11:52:44 EDT 2011 athena@fftw.org * Remove --enable-portable-binary, --with-gcc-arch from documentation. M ./NEWS +3 M ./doc/install.texi -23 +2 Fri Jun 24 11:48:48 EDT 2011 athena@fftw.org * Forget about specifying nonportable CFLAGS. Let the user do it if he wants. M ./m4/Makefile.am -4 +3 M ./m4/ax_cc_maxopt.m4 -24 +10 R ./m4/ax_gcc_archflag.m4 R ./m4/ax_gcc_x86_cpuid.m4 Fri Jun 24 11:48:25 EDT 2011 athena@fftw.org * Add "-avx" to version string when appropriate. M ./api/version.c +4 Fri Jun 24 10:26:38 EDT 2011 athena@fftw.org * change 3.3-alpha => 3.3-beta1 M ./configure.ac -1 +1 Fri Jun 24 09:25:49 EDT 2011 athena@fftw.org * Extend OUR_MALLOC16 to larger alignments Make it work for 32-byte alignment and beyond, as needed by AVX. Rename --with-our-malloc16 to --with-our-malloc. Keep old --with-our-malloc16 flag for compatibility. M ./configure.ac -3 +4 M ./kernel/kalloc.c -9 +10 Fri Jun 24 09:19:38 EDT 2011 athena@fftw.org * Fix typo M ./doc/install.texi -1 +1 Fri Jun 24 09:10:26 EDT 2011 athena@fftw.org * One pass over the manual. M ./NEWS -1 +6 M ./doc/Makefile.am -5 +2 M ./doc/install.texi -16 +10 M ./doc/intro.texi -1 +1 M ./doc/other.texi -2 +2 M ./doc/reference.texi -5 +5 M ./doc/tutorial.texi -32 +32 Fri Jun 24 08:19:03 EDT 2011 athena@fftw.org * eliminate the WITH_ALIGNED_STACK hack This is 2011 and I have no system with incorrect stack alignment. M ./TODO -2 M ./api/apiplan.c -9 +4 M ./api/execute-dft-c2r.c -2 +2 M ./api/execute-dft-r2c.c -2 +2 M ./api/execute-dft.c -2 +2 M ./api/execute-r2r.c -2 +2 M ./api/execute-split-dft-c2r.c -2 +2 M ./api/execute-split-dft-r2c.c -2 +2 M ./api/execute-split-dft.c -2 +2 M ./api/execute.c -2 +2 M ./api/f77funcs.h -16 +16 M ./doc/Makefile.am -2 +2 M ./doc/fftw3.texi -1 +1 M ./doc/intro.texi -3 +4 M ./doc/legacy-fortran.texi -1 +1 M ./doc/other.texi -51 +4 M ./doc/reference.texi -1 +1 M ./doc/tutorial.texi -9 +5 M ./kernel/ifftw.h -51 M ./threads/ct.c -2 +2 M ./threads/dft-vrank-geq1.c -2 +2 M ./threads/hc2hc.c -2 +2 M ./threads/rdft-vrank-geq1.c -2 +2 M ./threads/vrank-geq1-rdft2.c -2 +2 Fri Jun 24 07:49:47 EDT 2011 athena@fftw.org * enable both threaded and unthreaded wisdom in tools/fftw-wisdom M ./TODO -7 +2 M ./tools/fftw-wisdom.c -18 +14 Fri Jun 24 02:40:04 EDT 2011 stevenj@fftw.org * clarification M ./doc/modern-fortran.texi -2 +2 Fri Jun 24 02:24:01 EDT 2011 stevenj@fftw.org * check for error code in example M ./doc/modern-fortran.texi +2 Fri Jun 24 02:22:18 EDT 2011 stevenj@fftw.org * cleanup - since NATIVE_MALLOC is always malloc, delete this #define M ./api/export-wisdom-to-string.c -1 +1 M ./kernel/ifftw.h -2 Fri Jun 24 02:19:44 EDT 2011 stevenj@fftw.org * document wisdom string import/export in Fortran M ./doc/modern-fortran.texi +49 Fri Jun 24 02:11:40 EDT 2011 stevenj@fftw.org * bug fix - NATIVE_MALLOC should always be plain malloc, even in debug_malloc mode, because it is used in the API to return things that should be deallocated with free(); correspondingly, be sure to use free() ansd not X(free) with this M ./kernel/ifftw.h -1 +1 M ./mpi/wisdom-api.c -1 +1 Fri Jun 24 01:35:27 EDT 2011 stevenj@fftw.org * declaration style M ./doc/modern-fortran.texi -2 +4 Fri Jun 24 01:25:36 EDT 2011 stevenj@fftw.org * document wisdom file export/import from Fortran; add export/import_to/from_filename functions for convenience M ./api/export-wisdom-to-file.c +11 M ./api/fftw3.h +2 M ./api/import-wisdom-from-file.c +10 M ./doc/modern-fortran.texi -9 +79 M ./doc/other.texi -5 +6 M ./doc/reference.texi -11 +22 Thu Jun 23 19:19:43 EDT 2011 stevenj@fftw.org * more fortran docs M ./doc/legacy-fortran.texi -5 +6 M ./doc/modern-fortran.texi -10 +167 M ./doc/reference.texi +2 Thu Jun 23 17:50:30 EDT 2011 stevenj@fftw.org * enforce 132-character line-length limit that is the default in Fortran M ./api/genf03.pl -3 +26 Wed Jun 22 23:27:31 EDT 2011 stevenj@fftw.org * the F03 standard is ambiguous about whether types can be assigned to wider types as formal parameters with VALUE attributes, and e.g. gfortran interprets it to disallow this code M ./doc/modern-fortran.texi -2 +4 Thu Jun 23 18:12:10 EDT 2011 athena@fftw.org * Add md5 hash of fftw's configuration to wisdom file People were already confused by threaded vs unthreaded wisdom, and now things will be even worse because we enable/disable AVX codelets at runtime. Accept incoming wisdom only if it was produced by the same configuration (modulo MD5). M ./kernel/planner.c -2 +33 Thu Jun 23 09:01:27 EDT 2011 athena@fftw.org * distribute fftw3.f03.in M ./api/Makefile.am -1 +1 Wed Jun 22 22:02:18 EDT 2011 stevenj@fftw.org * more Fortran documentation M ./doc/modern-fortran.texi -1 +95 Wed Jun 22 20:10:39 EDT 2011 stevenj@fftw.org * correct comment M ./api/genf03.pl -1 +2 Wed Jun 22 20:26:18 EDT 2011 athena@fftw.org * Use "sh FOO.sh" instead of "./FOO.sh" to avoid chmod +x. M ./api/Makefile.am -1 +1 Wed Jun 22 19:19:05 EDT 2011 stevenj@fftw.org * document fftw_alloc_real/complex ... should we switch to using these in the tutorial examples? M ./doc/reference.texi +19 Wed Jun 22 19:07:49 EDT 2011 stevenj@fftw.org * whoops, added missing file A ./api/genf03.pl Wed Jun 22 18:46:01 EDT 2011 athena@fftw.org * Note addition of AVX. M ./NEWS +2 Wed Jun 22 18:43:43 EDT 2011 athena@fftw.org * In SSE2, AVX: use FMA macros when applicable. Makes it easier to play with fma4 and fma3 when it comes out. M ./simd-support/simd-avx.h -21 +21 M ./simd-support/simd-sse2.h -19 +19 Wed Jun 22 18:16:45 EDT 2011 stevenj@fftw.org * all modern Fortran compilers can call FFTW's C interfface directly -- support this, and in particular generate a Fortran 2003 interface file from fftw3.h so that Fortran code calling FFTW can be typechecked ((addressing the source of a lot of Fortran-user problems) ./doc/fortran.texi -> ./doc/legacy-fortran.texi M ./api/Makefile.am -3 +12 A ./api/f03api.sh M ./api/fftw3.h -3 +11 M ./api/malloc.c +18 M ./configure.ac +12 M ./doc/Makefile.am -1 +1 M ./doc/fftw3.texi -2 +4 M ./doc/install.texi -2 +2 M ./doc/intro.texi -5 +6 M ./doc/legacy-fortran.texi -23 +38 A ./doc/modern-fortran.texi M ./doc/mpi.texi -1 +1 M ./doc/other.texi -2 +4 M ./doc/upgrading.texi -1 +1 Wed Jun 22 13:10:02 EDT 2011 athena@fftw.org * some cleanup of SSE2 macros M ./simd-support/simd-sse2.h -27 +16 Wed Jun 22 07:38:18 EDT 2011 athena@fftw.org * don't use -xHost on ICC -xHost with ICC is problematic. On icc-12.0.0, "-mavx -xHost" overrides -mavx with -xHost, generating SSE2 code instead of AVX code. ICC does not seem to support -mtune=host or equivalent non-ABI changing flag. M ./m4/ax_cc_maxopt.m4 -31 +5 Tue Jun 21 20:35:36 EDT 2011 athena@fftw.org * Complete AVX implementation for split codelets M ./configure.ac -36 +38 M ./simd-support/simd-avx.h -41 +26 Tue Jun 21 19:37:14 EDT 2011 stevenj@fftw.org * whoops, missing altivec conf patches M ./dft/codelet-dft.h +1 M ./dft/conf.c +4 M ./rdft/codelet-rdft.h +1 M ./rdft/conf.c +4 Tue Jun 21 19:12:45 EDT 2011 stevenj@fftw.org * some BSD ar versions (e.g. on MacOS X) give an error if there are no object files, so we cannot build empty libraries M ./dft/simd/altivec/Makefile.am -4 +3 M ./dft/simd/avx/Makefile.am -3 +4 M ./dft/simd/sse2/Makefile.am -4 +3 M ./rdft/simd/altivec/Makefile.am -4 +3 M ./rdft/simd/avx/Makefile.am -3 +4 M ./rdft/simd/sse2/Makefile.am -3 +4 Tue Jun 21 19:12:12 EDT 2011 stevenj@fftw.org * re-insertion of Altivec code M ./Makefile.am -1 +6 M ./configure.ac -8 +13 M ./dft/simd/Makefile.am -1 +1 A ./dft/simd/altivec/ A ./dft/simd/altivec/Makefile.am M ./kernel/ifftw.h -2 +9 M ./rdft/simd/Makefile.am -1 +1 A ./rdft/simd/altivec/ A ./rdft/simd/altivec/Makefile.am M ./simd-support/Makefile.am -1 +1 A ./simd-support/altivec.c A ./simd-support/simd-altivec.h M ./simd-support/simd-common.h +3 Tue Jun 21 16:26:09 EDT 2011 athena@fftw.org * Implement faster AVX loads/stores. M ./simd-support/simd-avx.h -21 +28 Tue Jun 21 16:03:24 EDT 2011 athena@fftw.org * Initial AVX256/single implementation This should be correct but slow. I need to figure out how to implement noncontiguous loads/stores efficiently. M ./simd-support/simd-avx.h -53 +133 Tue Jun 21 14:13:57 EDT 2011 athena@fftw.org * fix AVX alignment M ./simd-support/simd-common.h -6 +8 Tue Jun 21 14:07:28 EDT 2011 athena@fftw.org * rename avx256d -> avx AVX will work in both double and single precision, like SSE2. ./dft/simd/avx256d -> ./dft/simd/avx ./rdft/simd/avx256d -> ./rdft/simd/avx ./simd-support/avx256d.c -> ./simd-support/avx.c ./simd-support/simd-avx256d.h -> ./simd-support/simd-avx.h M ./Makefile.am -4 +4 M ./configure.ac -11 +6 M ./dft/codelet-dft.h -1 +1 M ./dft/conf.c -3 +3 M ./dft/simd/Makefile.am -1 +1 M ./dft/simd/avx/Makefile.am -4 +4 M ./kernel/ifftw.h -2 +2 M ./rdft/codelet-rdft.h -1 +1 M ./rdft/conf.c -3 +3 M ./rdft/simd/Makefile.am -1 +1 M ./rdft/simd/avx/Makefile.am -4 +4 M ./simd-support/Makefile.am -1 +1 M ./simd-support/avx.c -3 +3 M ./simd-support/simd-avx.h -2 +2 Tue Jun 21 13:52:20 EDT 2011 athena@fftw.org * remove CODELET_OPTIM In the old 32-bit gcc-3.x days we used to play games with gcc to force it to produce decent code. Now gcc has gotten smarter and it produces indecent code no matter what we do, so it is safe to remove these hacks. M ./api/version.c -5 +5 M ./configure.ac -40 M ./support/Makefile.codelets -6 Tue Jun 21 09:57:31 EDT 2011 athena@fftw.org * work around gcc/icc quirks M ./simd-support/simd-avx256d.h -14 +45 Tue Jun 21 09:56:07 EDT 2011 athena@fftw.org * Add remarks in places where we work around gcc quirks M ./simd-support/simd-sse2.h -7 +16 Mon Jun 20 21:17:59 EDT 2011 stevenj@fftw.org * remove the libbench directory (which we have kept lingering in the repository for years due to CVS's inability to remove directories) R ./libbench/Makefile.am R ./libbench/accopy-from.c R ./libbench/accopy-to.c R ./libbench/acopy.c R ./libbench/allocate.c R ./libbench/ascale.c R ./libbench/aset.c R ./libbench/bench-main.c R ./libbench/bench-user.h R ./libbench/bench.h R ./libbench/caadd.c R ./libbench/cacopy.c R ./libbench/can-do.c R ./libbench/cascale.c R ./libbench/caset.c R ./libbench/casub.c R ./libbench/ccopy-from.c R ./libbench/ccopy-to.c R ./libbench/copy-c2c-from.c R ./libbench/copy-c2c-to.c R ./libbench/copy-c2h-1d-fftpack.c R ./libbench/copy-c2h-1d-halfcomplex.c R ./libbench/copy-c2h-1d-packed.c R ./libbench/copy-c2h-1d-unpacked-ri.c R ./libbench/copy-c2h-unpacked.c R ./libbench/copy-c2h.c R ./libbench/copy-c2r-packed.c R ./libbench/copy-c2r-unpacked.c R ./libbench/copy-c2r.c R ./libbench/copy-c2ri.c R ./libbench/copy-h2c-1d-fftpack.c R ./libbench/copy-h2c-1d-halfcomplex.c R ./libbench/copy-h2c-1d-packed.c R ./libbench/copy-h2c-1d-unpacked-ri.c R ./libbench/copy-h2c-unpacked.c R ./libbench/copy-h2c.c R ./libbench/copy-r2c-packed.c R ./libbench/copy-r2c-unpacked.c R ./libbench/copy-r2c.c R ./libbench/copy-ri2c.c R ./libbench/deallocate.c R ./libbench/getopt-utils.c R ./libbench/getopt.c R ./libbench/getopt.h R ./libbench/getopt1.c R ./libbench/info.c R ./libbench/log2.c R ./libbench/main.c R ./libbench/mflops.c R ./libbench/mp.c R ./libbench/ovtpvt.c R ./libbench/pow2.c R ./libbench/prime.c R ./libbench/problem.c R ./libbench/report.c R ./libbench/speed.c R ./libbench/timer.c R ./libbench/unnormalize.c R ./libbench/util.c R ./libbench/verify.c R ./libbench/zero.c R ./libbench/ Mon Jun 20 21:17:14 EDT 2011 stevenj@fftw.org * update URLs M ./doc/install.texi -2 +2 M ./doc/intro.texi -1 +1 M ./doc/license.texi -1 +1 M ./doc/mpi.texi -1 +1 M ./doc/other.texi -1 +1 Mon Jun 20 20:53:31 EDT 2011 stevenj@fftw.org * whoops, forgot to check in alignment change M ./simd-support/simd-common.h -3 +6 Mon Jun 20 20:22:23 EDT 2011 athena@fftw.org * "test X = Y" requires spaces around "=" M ./configure.ac -1 +1 Mon Jun 20 19:18:52 EDT 2011 stevenj@fftw.org * indenting M ./simd-support/simd-sse2.h -3 +3 Mon Jun 20 18:57:10 EDT 2011 stevenj@fftw.org * merge back in SSE support, now combined with SSE2; --enable-sse2 now works in both single and double precision, and simd-sse2.h contains both the double- and single-precision code (which overlap a lot); in single precision it is still compiled for SSE-only (SSE2 is only required for double) M ./configure.ac -16 +9 M ./simd-support/simd-sse2.h -54 +181 M ./simd-support/sse2-nonportable.c -2 +6 M ./simd-support/sse2.c -4 +11 Mon Jun 20 16:02:07 EDT 2011 athena@fftw.org * Implement AVX autodetection (gcc-only so far) M ./simd-support/Makefile.am -1 +1 A ./simd-support/amd64-cpuid.h M ./simd-support/avx256d.c -2 +32 M ./simd-support/x86-cpuid.h +33 Mon Jun 20 14:25:54 EDT 2011 athena@fftw.org * Add VZEROUPPER at the end of AVX codelets If the Intel Optimization Manual is to be believed, we need to wave a dead chicken before transitioning from AVX code to SSE code. I am supposed to believe that there is a transition penalty for doing so, unless one uses a magic VZEROUPPER instruction that apparently has zero cost. Whatever. M ./genfft/c.ml +2 M ./genfft/c.mli +1 M ./genfft/gen_hc2c.ml -1 +1 M ./genfft/gen_hc2cdft.ml -1 +1 M ./genfft/gen_hc2cdft_c.ml -1 +1 M ./genfft/gen_hc2hc.ml -1 +1 M ./genfft/gen_mdct.ml -1 +1 M ./genfft/gen_notw.ml -1 +1 M ./genfft/gen_notw_c.ml -1 +1 M ./genfft/gen_r2cb.ml -1 +1 M ./genfft/gen_r2cf.ml -1 +1 M ./genfft/gen_r2r.ml -1 +1 M ./genfft/gen_twiddle.ml -1 +1 M ./genfft/gen_twiddle_c.ml -1 +1 M ./genfft/gen_twidsq.ml -1 +1 M ./genfft/gen_twidsq_c.ml -1 +1 M ./genfft/genutil.ml -2 +2 M ./genfft/simd.ml +1 M ./simd-support/simd-avx256d.h +5 M ./simd-support/simd-sse2.h +2 Mon Jun 20 10:21:25 EDT 2011 athena@fftw.org * Move RDFT to new simd scheme A ./rdft/simd/avx256d/ R ./rdft/simd/codelets/Makefile.am R ./rdft/simd/codelets/ A ./rdft/simd/common/ A ./rdft/simd/sse2/ M ./Makefile.am -2 +4 M ./configure.ac +4 M ./dft/simd/Makefile.am -1 +1 M ./genfft/gen_hc2cdft_c.ml -2 +2 M ./kernel/ifftw.h +1 M ./rdft/Makefile.am -1 +1 M ./rdft/codelet-rdft.h -1 +2 M ./rdft/conf.c -2 +7 M ./rdft/simd/Makefile.am -6 +2 A ./rdft/simd/avx256d/Makefile.am A ./rdft/simd/codlist.mk A ./rdft/simd/common/Makefile.am A ./rdft/simd/common/genus.c R ./rdft/simd/hc2cbv.c M ./rdft/simd/hc2cbv.h -2 +2 R ./rdft/simd/hc2cfv.c M ./rdft/simd/hc2cfv.h -2 +2 A ./rdft/simd/simd.mk A ./rdft/simd/sse2/Makefile.am Mon Jun 20 09:23:38 EDT 2011 athena@fftw.org * New SIMD build system We now support multiple SIMD extensions in the same binary, e.g. --enable-sse2 --enable-avx. This patch adds the necessary infrastructure for SSE2/AVX and complex DFT. Later patches will add RDFT and SSE/ALTIVEC/etc. A ./dft/simd/avx256d/ A ./simd-support/ R ./dft/simd/codelets/Makefile.am R ./dft/simd/codelets/ A ./dft/simd/common/ A ./dft/simd/sse2/ M ./Makefile.am -11 +13 M ./configure.ac -47 +74 M ./dft/codelet-dft.h -4 +2 M ./dft/conf.c -2 +7 M ./dft/scalar/codelets/Makefile.am -1 +1 M ./dft/simd/Makefile.am -6 +3 A ./dft/simd/avx256d/Makefile.am A ./dft/simd/codlist.mk A ./dft/simd/common/Makefile.am A ./dft/simd/common/genus.c R ./dft/simd/n1b.c M ./dft/simd/n1b.h -2 +2 R ./dft/simd/n1f.c M ./dft/simd/n1f.h -2 +2 R ./dft/simd/n2b.c M ./dft/simd/n2b.h -2 +2 R ./dft/simd/n2f.c M ./dft/simd/n2f.h -2 +2 R ./dft/simd/n2s.c M ./dft/simd/n2s.h -2 +2 R ./dft/simd/q1b.c M ./dft/simd/q1b.h -2 +2 R ./dft/simd/q1f.c M ./dft/simd/q1f.h -2 +2 A ./dft/simd/simd.mk A ./dft/simd/sse2/Makefile.am R ./dft/simd/t.c M ./dft/simd/t1b.h -2 +2 M ./dft/simd/t1bu.h -2 +2 M ./dft/simd/t1f.h -2 +2 M ./dft/simd/t1fu.h -2 +2 M ./dft/simd/t2b.h -2 +2 M ./dft/simd/t2f.h -2 +2 M ./dft/simd/t3b.h -2 +2 M ./dft/simd/t3f.h -2 +2 R ./dft/simd/ts.c M ./dft/simd/ts.h -2 +2 M ./genfft/gen_twiddle.ml -2 +2 M ./genfft/genutil.ml -2 +5 M ./kernel/ifftw.h -2 +6 M ./rdft/Makefile.am -1 +1 M ./rdft/conf.c -1 +1 M ./rdft/scalar/r2cb/Makefile.am +1 M ./rdft/scalar/r2cf/Makefile.am +1 M ./rdft/scalar/r2r/Makefile.am +1 A ./simd-support/Makefile.am A ./simd-support/avx256d.c A ./simd-support/simd-avx256d.h A ./simd-support/simd-common.h A ./simd-support/simd-sse2.h A ./simd-support/sse2-nonportable.c A ./simd-support/sse2.c A ./simd-support/taint.c A ./simd-support/x86-cpuid.h M ./support/Makefile.codelets -21 +22 Sun Jun 19 12:29:27 EDT 2011 stevenj@fftw.org * italicize Latin quote M ./doc/reference.texi -2 +2 Sun Jun 19 12:26:34 EDT 2011 stevenj@fftw.org * work around incredibly annoying makeinfo bug -- for HTML output, in any paragraph ending with an @index command, two blank lines are needed to create a paragraph break ... our HTML output has apparently been screwed up for years M ./doc/acknowledgements.texi +2 M ./doc/fortran.texi +1 M ./doc/install.texi +2 M ./doc/intro.texi +1 M ./doc/mpi.texi +20 M ./doc/other.texi +2 M ./doc/reference.texi +21 M ./doc/threads.texi +2 M ./doc/tutorial.texi +13 Sun Jun 19 12:01:39 EDT 2011 stevenj@fftw.org * conjugate-pair algorithm turns out not to be due to djb, but it was pointed out to us by djb M ./doc/intro.texi -3 +4 Sun Jun 19 11:59:21 EDT 2011 stevenj@fftw.org * new-array execute functions are *not* the same for MPI, since the problems are different M ./doc/mpi.texi -2 +3 Sun Jun 19 11:57:52 EDT 2011 stevenj@fftw.org * tweaks to MPI manual M ./doc/mpi.texi -19 +44 M ./doc/reference.texi -1 +1 Sun Jun 19 00:47:23 EDT 2011 stevenj@fftw.org * only call MPI_Init_thread for MPI version >= 2 M ./mpi/mpi-bench.c +5 Sun Jun 19 00:35:44 EDT 2011 stevenj@fftw.org * document quad precision in Fortran M ./doc/fortran.texi -9 +12 Sun Jun 19 00:16:10 EDT 2011 stevenj@fftw.org * use -lquadmath for quad-precision library in pkgconfig file M ./configure.ac +2 M ./fftw.pc.in -1 +1 Sun Jun 19 00:10:33 EDT 2011 stevenj@fftw.org * document quad precision M ./doc/install.texi -1 +11 M ./doc/reference.texi +8 Sat Jun 18 21:26:24 EDT 2011 stevenj@fftw.org * there is currently no standard quad-precision type in MPI, so don't allow MPI support to be compiled with --enable-quad-precision M ./configure.ac +3 M ./mpi/ifftw-mpi.h +2 M ./mpi/mpi-bench.c +2 Sat Jun 18 21:25:51 EDT 2011 stevenj@fftw.org * make quad-precision library libfftwq M ./configure.ac +1 Sat Jun 18 21:19:50 EDT 2011 stevenj@fftw.org * use --estimate in check script with --enable-random-estimator M ./configure.ac +2 M ./mpi/Makefile.am -1 +1 M ./tests/Makefile.am -5 +5 Sat Jun 18 20:36:51 EDT 2011 stevenj@fftw.org * bug fix - correct crashing interaction between threads and debug-malloc M ./threads/threads.c +4 Sat Jun 18 18:41:48 EDT 2011 stevenj@fftw.org * whoops, typo M ./doc/reference.texi -1 +1 Sat Jun 18 18:35:14 EDT 2011 stevenj@fftw.org * clarification of when fftw_cost may return 0 M ./doc/reference.texi -2 +5 Sat Jun 18 18:31:26 EDT 2011 stevenj@fftw.org * corrected manual and test program for proper interaction of MPI and threads M ./doc/mpi.texi -20 +59 M ./mpi/mpi-bench.c +13 M ./tests/fftw-bench.c -4 +10 M ./tests/fftw-bench.h +5 Sat Jun 18 17:13:52 EDT 2011 stevenj@fftw.org * need --mpi restriction in mpi+threads check M ./mpi/Makefile.am -1 +1 Sat Jun 18 12:09:04 EDT 2011 stevenj@fftw.org * split fftw3.texi into multiple files for ease of editing M ./doc/Makefile.am -2 +1 A ./doc/acknowledgements.texi A ./doc/cindex.texi M ./doc/fftw3.texi -6158 +14 A ./doc/findex.texi A ./doc/fortran.texi A ./doc/install.texi A ./doc/intro.texi A ./doc/license.texi A ./doc/mpi.texi A ./doc/other.texi A ./doc/reference.texi A ./doc/threads.texi A ./doc/tutorial.texi A ./doc/upgrading.texi Fri Jun 17 17:26:50 EDT 2011 stevenj@fftw.org * merge recent Cell deletion with MPI branch M ./Makefile.am -3 +2 M ./NEWS +40 M ./configure.ac -1 +1 M ./doc/fftw3.texi -95 +909 Fri Jun 17 01:52:59 EDT 2011 stevenj@fftw.org * whoops, incorrect assertion M ./mpi/transpose-problem.c -2 +2 Fri Jun 17 01:52:51 EDT 2011 stevenj@fftw.org * comment fix M ./mpi/block.c -2 +2 Thu Jun 16 23:30:27 EDT 2011 stevenj@fftw.org * check if pln creation failed (e.g. for split input) bbefore calling setup_gather_scatter, to prevent crashes M ./mpi/mpi-bench.c -1 +1 Thu Jun 16 23:26:48 EDT 2011 stevenj@fftw.org * bug fix -- transpose-recurse is only applicable if subtransposes fit in the same space (unless I change the allocation routine, but this would seem to require looking at all possible recursive invocations of transpose-recurse) M ./mpi/transpose-recurse.c +28 Mon Apr 11 17:58:03 EDT 2011 stevenj@fftw.org * yikes, any_true check on subplan creation should be in comm, not comm2, so that all processes know if failure occurred M ./mpi/transpose-recurse.c -6 +4 Mon Apr 11 17:00:46 EDT 2011 stevenj@fftw.org * add wisdom_ok_hook to enforce wisdom synchronization on MPI problems, apparently fixing a longstanding deadlock/crash bug M ./kernel/ifftw.h +1 M ./kernel/planner.c +7 M ./mpi/api.c -13 +61 Mon Apr 11 15:10:22 EDT 2011 stevenj@fftw.org * add a check (in DEBUG mode only) that all processes produce the same hash of MPI problems; don't include alignment in MPI problem hash because it may differ between processes for unaligned malloc M ./mpi/any-true.c +21 M ./mpi/dft-problem.c -2 +6 M ./mpi/ifftw-mpi.h +1 M ./mpi/rdft-problem.c -2 +6 M ./mpi/rdft2-problem.c -2 +6 M ./mpi/transpose-problem.c -2 +6 Fri Apr 8 18:46:54 EDT 2011 stevenj@fftw.org * use cost_hook in random_estimate M ./kernel/planner.c -1 +1 Sun Mar 6 23:33:53 EST 2011 stevenj@fftw.org * added mpi new-array execute functions; thanks to Guo Luo for the bug report M ./mpi/api.c +23 M ./mpi/fftw3-mpi.h -1 +6 Wed Feb 9 21:29:17 EST 2011 stevenj@fftw.org * MPI may not support tags > 2^15-1 (e.g. Cray MPI requires tags < 2^24); thanks to Jonathan Bentz for the bug report. M ./mpi/transpose-pairwise.c -4 +4 Wed Feb 2 12:21:30 EST 2011 stevenj@fftw.org * fix merge conflicts M! ./configure.ac -1 +1 Sat Nov 15 22:33:20 EST 2008 stevenj@fftw.org tagged fftw-3.3alpha1 Sat Nov 15 20:33:33 EST 2008 stevenj@fftw.org * version bump for 3.3alpha1 M! ./NEWS -40 M! ./configure.ac -2 +2 Sun Oct 26 22:47:07 EDT 2008 stevenj@fftw.org * re-added mpi/Makefile M ./configure.ac +2 Sat Oct 25 17:14:42 EDT 2008 stevenj@fftw.org * re-add MPI to dist M! ./Makefile.am -2 +3 M! ./doc/fftw3.texi -9 +134 Sat Jun 18 08:50:13 EDT 2011 athena@fftw.org * remove obsolete Cell code M ./kernel/align.c -1 +1 M ./kernel/ifftw.h -9 +2 M ./rdft/rank0.c -83 M ./simd/simd.h -6 M ./simd/taint.c -1 +1 M ./tests/fftw-bench.c -7 Fri Jun 17 23:31:33 EDT 2011 stevenj@fftw.org * bug fix in accuracy test, which prevented us from consiistently determining accuracy in > double precision M ./libbench2/mp.c -2 +2 Fri Jun 17 20:05:13 EDT 2011 athena@fftw.org * do not check for gcc version before checking for gcc M ./configure.ac -1 +1 Fri Jun 17 18:56:37 EDT 2011 stevenj@fftw.org * require gcc 4.6.0 or later for --enable-quad-precision, to match fftw3.h header file; no need to mark this as EXPERIMENTAL (make check passes, and support in gcc 4.6 seems reasonably complete) M ./configure.ac -1 +2 Fri Jun 17 18:51:41 EDT 2011 stevenj@fftw.org * need ugly __attribute__ to use __float128 with _Complex, ugh M ./api/fftw3.h +9 Fri Jun 17 18:23:05 EDT 2011 stevenj@fftw.org * --verify tolerance in quad precision changed to 1e-29 M ./libbench2/bench-main.c -1 +1 Fri Jun 17 18:22:38 EDT 2011 stevenj@fftw.org * quad-precision F77 api should use "qfftw" prefix M ./api/x77.h +3 Fri Jun 17 18:22:27 EDT 2011 stevenj@fftw.org * rm extraneous space from fftw3.h M ./api/fftw3.h -1 +1 Fri Jun 17 18:05:10 EDT 2011 stevenj@fftw.org * use cosq etcetera with libquadmath in libbench2, so that --verify correctly gives ~33 decimal places in shift test M ./libbench2/verify.h -1 +4 Fri Jun 17 17:52:51 EDT 2011 stevenj@fftw.org * libquadmath ships with gcc 4.6.0, so we should require this library for sinq/cosq with --enable-quad-precision; also, include the __float128 FFTW functions in the header file for gcc >= 4.6 on i86/x86[_\e2_][_\80_]_64/ia64 M ./Makefile.am +5 M ./api/fftw3.h -1 +6 M ./configure.ac -1 +5 M ./kernel/trig.c -13 +4 Fri Jun 17 16:54:01 EDT 2011 stevenj@fftw.org * typo in manual for fftw_cost M ./doc/fftw3.texi -1 +1 Fri Jun 17 16:48:24 EDT 2011 stevenj@fftw.org * fix fftw_cost function: pcost needs to be saved in mkapiplan, since the plan is re-created from wisdom M ./api/apiplan.c +5 M ./tests/fftw-bench.c -2 +3 Fri Jun 17 16:42:25 EDT 2011 athena@fftw.org * removed support for the Cell Broadband Engine R ./cell/spu/Makefile.am R ./cell/spu/alloc.spuc R ./cell/spu/copy.spuc R ./cell/spu/dft.spuc R ./cell/spu/dma.spuc R ./cell/spu/execute.spuc R ./cell/spu/fftw-spu.h R ./cell/spu/main.spuc R ./cell/spu/planner.spuc R ./cell/spu/spu-double.h R ./cell/spu/spu-single.h R ./cell/spu/transpose.spuc R ./cell/spu/ R ./cell/Makefile.am R ./cell/cell.c R ./cell/conf.c R ./cell/copy.c R ./cell/dft-direct-cell.c R ./cell/fftw-cell.h R ./cell/plans-double.c R ./cell/plans-single.c R ./cell/spufftw-embed.S R ./cell/transpose.c R ./cell/ M ./AUTHORS -2 +3 M ./Makefile.am -11 +5 M ./NEWS +2 R ./README.Cell M ./api/configure.c -3 M ./configure.ac -18 +1 M ./dft/dft.h -4 M ./dft/rank-geq2.c -6 M ./doc/fftw3.texi -128 +4 Tue May 24 06:51:07 EDT 2011 athena@fftw.org * Undo previous change; the typo was not a typo after all. M ./doc/fftw3.texi -1 +1 Mon May 23 05:08:05 EDT 2011 athena@fftw.org * Fix typo in manual M ./doc/fftw3.texi -1 +1 Sat May 21 17:37:50 EDT 2011 athena@fftw.org * clarify intent about canonicalization of tensor in tensor_compress_contiguous() M ./kernel/tensor7.c -1 +9 Sat May 21 17:30:31 EDT 2011 athena@fftw.org * avoid useless canonicalization in tensor_compress_contiguous() M ./kernel/tensor7.c -1 +1 Sat May 21 17:24:57 EDT 2011 athena@fftw.org * Fix tensor_compress_contiguous tensor_compress_contiguous() was supposed to sort dimensions by descending istride, and then compress adjacent dimensions. This property was lost once we changed the canonical order of strides to be sorted by descending min{istride,ostride}. Change tensor_compress_contiguous() to sort by descending istride again, which is necessary for its correctness, and then canonicalize at the end. M ./kernel/tensor7.c -20 +41 Sun May 8 18:47:26 EDT 2011 athena@fftw.org * Don't distribute obsolete .depend M ./genfft/Makefile.am -2 +2 Sun May 8 18:05:36 EDT 2011 athena@fftw.org * Use ocamlbuild for building genfft Remove the old Makefile cruft to support ocaml, and use ocamlbuild instead. M ./bootstrap.sh -2 M ./configure.ac -10 +1 M ./genfft/Makefile.am -151 +8 M ./m4/Makefile.am -1 +1 R ./m4/ocaml.m4 M ./support/Makefile.codelets -13 +13 Sun May 8 18:03:07 EDT 2011 athena@fftw.org * Do not use __float128 unless BENCHFFT_QUAD is defined Otherwise, compilation fails on compilers that do not support __float128. M ./libbench2/bench-user.h +5 Fri Apr 8 13:15:54 EDT 2011 stevenj@fftw.org * fix configure --help string for --disable-alloca (since default is enabled) M ./configure.ac -1 +1 Fri Apr 8 13:09:56 EDT 2011 stevenj@fftw.org * add "random estimator" for debugging purposes; note that this is best used with ESTIMATE_PATIENT mode M ./api/apiplan.c +4 M ./configure.ac +5 M ./kernel/ifftw.h +4 M ./kernel/planner.c +27 M ./tests/fftw-bench.c +7 Tue Apr 5 14:47:56 EDT 2011 stevenj@fftw.org * add AC_CHECK_DECLS for srand48; thanks to Ralf Wildenhues for the bug report M ./configure.ac -1 +1 M ./libbench2/util.c +3 Sat Feb 5 17:00:40 EST 2011 stevenj@fftw.org * experimental support for gcc's __float128 quad-precision type M ./api/fftw3.h +3 M ./configure.ac -2 +14 M ./kernel/ifftw.h -2 +10 M ./kernel/trig.c -1 +16 M ./libbench2/bench-main.c +2 M ./libbench2/bench-user.h +4 M ./libbench2/info.c -1 +2 M ./libbench2/verify.h +6 M ./simd/simd-sse2.h -1 +1 M ./tests/fftw-bench.h +2 M ./tools/fftw-wisdom.c +2 Sun Oct 24 14:33:59 EDT 2010 athena@fftw.org * guarantee that "timelimit < 0" means "no timeout" "timelimit < 0" was always meant to be equivalent to "timelimit = HUGENUM", but this was not true in all cases, causing some obscure wisdom behavior. Thanks William Andrew Burnson for the bug report. M ./api/mapflags.c -1 +1 Sun Oct 24 14:32:20 EDT 2010 athena@fftw.org * compile with --enable-fma and SSE, SSE2 Allow compilation with --enable-fma and --enable-sse, --enable-sse2. This is a bad idea performance-wise, but people will try anyway. M ./simd/simd-sse.h +5 M ./simd/simd-sse2.h +5 Sun Jul 11 13:34:06 EDT 2010 athena@fftw.org * Make threads.c compiler with c++ M ./threads/threads.c -1 +1 Sun Jul 11 10:05:05 EDT 2010 athena@fftw.org * Attempt at clarifying the advanced interface doc. M ./doc/fftw3.texi -72 +151 Sun Jul 11 07:37:27 EDT 2010 athena@fftw.org * rename rfftwnd html picture It turns out that texinfo with pdf output reads .png files in preference to .pdf files (when did this change?). I renamed the .png figure to avoid producing an ugly pdf file. M ./doc/Makefile.am -4 +4 M ./doc/fftw3.texi +5 Tue Mar 30 19:43:22 EDT 2010 stevenj@fftw.org * added fftw_cost function; this is the second time people have asked for this, and there is a reasonable use for it in comparing e.g. oout-of-place vs. in-place plans M ./NEWS +4 M ./api/f77funcs.h +10 M ./api/fftw3.h +1 M ./api/flops.c +5 M ./doc/fftw3.texi +17 Tue Mar 2 18:55:49 EST 2010 stevenj@fftw.org * documented that --enable-debug-malloc causes fftw_execute to be thread-unsafe (thanks to Alexis Rohou for the problem report) M ./doc/fftw3.texi +8 Fri Jan 22 19:42:08 EST 2010 athena@fftw.org * Added FAQ about how to transpose matrices using FFTW. M ./doc/FAQ/fftw-faq.bfnn +33 Thu Jan 7 20:16:57 EST 2010 stevenj@fftw.org * catch FMS (instead of generating FMA(_,_,NEG(_)) with h -generic-arith option M ./genfft/c.ml +2 Fri Dec 11 07:01:26 EST 2009 athena@fftw.org * note future wisdom enhancements. M ./TODO +5 Mon Oct 19 20:21:05 EDT 2009 athena@fftw.org * Use SIMD flags when checking for xmmintrin.h This prevents an obnoxious warning from configure. M ./configure.ac -3 +3 Sat Aug 29 20:47:56 EDT 2009 athena@fftw.org * new bug M ./TODO +4 Sun Jul 26 00:40:11 EDT 2009 stevenj@fftw.org * typo (s/man1/many) M ./doc/fftw3.texi -1 +1 Tue Jul 14 14:19:08 EDT 2009 stevenj@fftw.org * BUILD-MINGW32 script, updated Windows README A ./support/BUILD-MINGW32.sh M ./support/BUILD-MINGW64.sh -4 +4 Mon Jul 13 09:40:38 EDT 2009 fftw@fftw.org * cleanup BUILD-MINGW64.sh M ./support/BUILD-MINGW64.sh -19 +27 Sun Jul 12 06:34:46 EDT 2009 athena@fftw.org * Update NEWS, version number for 3.2.2 release. M ./NEWS +3 M ./configure.ac -2 +2 Sat Jul 11 22:28:38 EDT 2009 athena@fftw.org * Reintroduce the pruning heuristic in ESTIMATE mode for r2r problems. Somehow, we lost this feature between fftw-3.1.3 and fftw-3.2. M ./rdft/hc2hc-direct.c +3 Thu Jun 25 07:39:04 EDT 2009 athena@fftw.org * don't use pshared=1 in sem_init pshared is really not necessary, and it is not supported on GNU/kFreeBSD. Thanks Petr Salinger for the bug report. M ./threads/threads.c -2 +2 Thu Jun 11 19:35:40 EDT 2009 fftw@fftw.org * Add mingw64 build file so that we can track it. A ./support/BUILD-MINGW64.sh Wed Jun 10 12:10:58 EDT 2009 fftw@fftw.org * note 3.2.2 NEWS M ./NEWS +7 Wed Jun 10 12:04:54 EDT 2009 fftw@fftw.org * add --disable-alloca to configure It looks like alloca() is broken on mingw64, and thus we need to disable it explicitly. M ./configure.ac -1 +6 M ./kernel/ifftw.h -1 +1 Sun Apr 26 16:33:10 EDT 2009 athena@fftw.org * Note in FAQ that --enable-k7 has been discontinued. M ./doc/FAQ/fftw-faq.bfnn -4 +7 Tue Mar 24 09:16:18 EDT 2009 athena@fftw.org * clarified small confusion in fftw_cleanup documentation M ./doc/fftw3.texi -2 +3 Thu Mar 19 13:18:06 EDT 2009 stevenj@fftw.org * fix documentation of dfftw_init_threads to indicate thaat it takes an argument (since the C version returns a value); thanks t Hans Johnnston for the bug report M ./doc/fftw3.texi -1 +5 Thu Mar 12 13:12:13 EDT 2009 fftw@fftw.org * if possible, use a 128-bit type for copy M ./configure.ac -1 +1 M ./kernel/cpy2d.c -9 +38 Tue Mar 10 12:49:51 EDT 2009 fftw@fftw.org * add size-128 simd codelets It's about time M ./dft/simd/codelets/Makefile.am -2 +2 Mon Mar 9 20:29:16 EDT 2009 athena@fftw.org * copy two floats as a double when possible Resurrect the old hack of copying two floats as a double, which makes some difference in these days of 64 bit boxes. M ./kernel/cpy2d.c -16 +34 Sun Mar 8 18:08:04 EDT 2009 athena@fftw.org * fixed (harmless) confusion of strides RS and VS were swapped in dftw-direct.c. This is a bug, but it is harmless unless one uses fixed-stride codelets, which we do not. M ./dft/dftw-direct.c -7 +7 Sun Mar 8 10:29:49 EDT 2009 athena@fftw.org * oops, I checked in debug code accidentally. M ./kernel/ct.c -1 Sat Feb 14 19:01:00 EST 2009 athena@fftw.org * Change TLO email address since Magdalen is no longer there. M ./commercialize.sh -1 +1 Sat Feb 14 18:18:45 EST 2009 stevenj@fftw.org * quote arguments to bench in test script on the off-chance that '*' would be expanded by the shell into a valid filename, and also to avoid shell confusion on Cygwin that "//" begins the name of a Windows network mountpoint M ./tests/check.pl -1 +1 Sat Feb 14 18:17:23 EST 2009 stevenj@fftw.org tagged fftw-3.2.1 Sun Mar 8 10:02:59 EDT 2009 athena@fftw.org * stricter conditions for Cooley-Tukey being ugly It turns out that m=2 in the leaf of Cooley-Tukey may be advantageous in certain cases, eg. i512v512 on AMD Shanghai: (dft-buffered-512-x128/512-6 (dft-ct-dit/4 (dftw-direct-4/24-x128 "t2fv_4") (dft-vrank>=1-x4/1 (dft-ct-dit/64 (dftw-direct-64/504-x128 "t2fv_64") (dft-vrank>=1-x64/1 (dft-direct-2-x128 "n2fv_2"))))) (dft-r2hc-1 (rdft-rank0-tiled/2-x128-x512)) (dft-nop)) Presumably this works around the 2 way associativity of the L1 cache. M ./dft/dftw-direct.c -1 +1 M ./kernel/ct.c -2 +3 M ./kernel/ifftw.h -1 +1 M ./rdft/ct-hc2c-direct.c -1 +1 M ./rdft/hc2hc-direct.c -3 +3 Mon Feb 9 19:46:00 EST 2009 stevenj@fftw.org * disable Windows QueryPerformanceCounter code, since it requires us to pull in windows.h in ifftw.h and causes namespace conflicts; gettimeofday seems to work well enough and has had few complaints M ./kernel/ifftw.h -4 +1 M ./kernel/timer.c -1 +8 Wed Feb 4 22:55:54 EST 2009 stevenj@fftw.org * version bump for 3.2.1, updated NEWS M ./NEWS +10 M ./configure.ac -1 +1 Wed Feb 4 22:27:28 EST 2009 stevenj@fftw.org * recommend that users avoid fftw_execute in Fortran, instead using dfftw_execute_dft and friends so that the compiler knows that the input/output arrays are used M ./doc/fftw3.texi -10 +80 Wed Jan 21 16:02:08 EST 2009 stevenj@fftw.org * prefer windows queryperformancecounter to gettimeofday on Windows, thanks to David Price for the suggestion M ./kernel/ifftw.h -1 +4 M ./kernel/timer.c -1 +20 Sun Feb 1 14:34:49 EST 2009 athena@fftw.org * compilation fixes in case snprintf() is defined as a macro. M ./libbench2/report.c -8 +7 Wed Jan 28 20:19:04 EST 2009 athena@fftw.org * Automake does not like continuation lines beginning with a comment. M ./rdft/scalar/r2cb/Makefile.am -1 +1 Wed Jan 28 18:24:39 EST 2009 athena@fftw.org * Add r2cb_2.c r2cb_2.c is needed for problem rb2, which is not equivalent to rf2 (unlike kb2, which is equivalent to kf2). This change would not matter much except that rb2 is generated when reducing backward rdft2 to dft, and the absence of the codelet was preventing radix 2 from being employed at all in this case. M ./rdft/scalar/r2cb/Makefile.am -6 +4 Sat Jan 10 06:47:22 EST 2009 athena@fftw.org * handle the case vecsz->rnk == 0 correctly. M ./dft/buffered.c -1 +3 M ./rdft/buffered.c -1 +3 M ./rdft/buffered2.c -1 +3 Fri Dec 19 15:20:36 EST 2008 stevenj@fftw.org * Macs are no longer ppc-based; thanks to Charles Collicutt for the FAQ update M ./doc/FAQ/fftw-faq.bfnn -3 +3 Mon Dec 8 18:08:33 EST 2008 stevenj@fftw.org * use new multiple-nbuf code in rdft/buffered, like for dft/buffered M ./rdft/buffered.c -11 +25 M ./rdft/buffered2.c -12 +26 Sat Dec 6 16:34:36 EST 2008 stevenj@fftw.org * make x86_cpuid macro work on x86_64 M ./m4/ax_gcc_x86_cpuid.m4 -1 +15 Sat Dec 6 09:20:37 EST 2008 athena@fftw.org * Allow automatic choice of buffer size in dft/buffered.c Try a couple of different buffer sizes in buffered transforms, since this seems to make a difference on some Core2 models. M ./api/mapflags.c -2 M ./dft/buffered.c -10 +25 M ./kernel/buffered.c -3 +19 M ./kernel/ifftw.h -1 +5 M ./rdft/buffered.c -2 +3 M ./rdft/buffered2.c -2 +3 M ./rdft/rdft2-rdft.c -1 +1 Tue Dec 2 19:18:30 EST 2008 athena@fftw.org * libbench2: do not assume that split-complex arrays are stride-1 M ./libbench2/verify-dft.c -4 +3 M ./libbench2/verify.c -1 +3 M ./tests/bench.c -3 +2 Tue Dec 2 18:39:43 EST 2008 stevenj@fftw.org * updated NEWS M ./NEWS +14 Tue Dec 2 18:30:00 EST 2008 stevenj@fftw.org * date fix M ./m4/ax_gcc_archflag.m4 -1 +1 Tue Dec 2 18:29:06 EST 2008 stevenj@fftw.org * updated icc flags -- now prefer -xHost (-xN etc. seem t be obsolete), check for new spelling -ansi-alias, and use -malign-double like we do for gcc M ./m4/ax_cc_maxopt.m4 -2 +14 Tue Dec 2 18:28:03 EST 2008 stevenj@fftw.org * use $ax_cv_c_compiler_vendor rather than $GCC, as the former is more reliable (icc incorrectly self-identifies as gcc on MacOS where we don't use -no-gcc) M ./configure.ac -1 +1 Tue Dec 2 17:55:36 EST 2008 stevenj@fftw.org * don't use -no-gcc for icc on MacOS M ./configure.ac -2 +7 Tue Dec 2 17:34:04 EST 2008 stevenj@fftw.org * document some more bench options M ./tests/README -1 +17 Wed Nov 19 16:55:13 EST 2008 stevenj@fftw.org * make it clearer that --enable-openmp and --enable-threads are mutually exclusive; thanks to Long To for his comments M ./doc/fftw3.texi -6 +10 Mon Nov 17 20:16:28 EST 2008 stevenj@fftw.org * version bump to 3.2.1, use explicit Makefile.am for m4 subdirectory so that tarball does not include random files in there when you do 'make dist' M ./Makefile.am -3 +2 M ./configure.ac -2 +4 A ./m4/Makefile.am Sat Nov 15 21:12:58 EST 2008 stevenj@fftw.org * document behavior of FFTW guru arrays, and in particular the odd behavior of the plan_guru_r2r routine in Fortran (thanks to Alexander Pozdneev for the bug report) M ./doc/fftw3.texi +10 Sat Nov 15 21:03:14 EST 2008 stevenj@fftw.org tagged fftw-3.2 Mon Nov 10 20:21:32 EST 2008 stevenj@fftw.org * version bump to 3.2, updated copyright year M ./COPYRIGHT -2 +2 M ./NEWS -1 +6 M ./api/api.h -2 +2 M ./api/apiplan.c -2 +2 M ./api/configure.c -2 +2 M ./api/execute-dft-c2r.c -2 +2 M ./api/execute-dft-r2c.c -2 +2 M ./api/execute-dft.c -2 +2 M ./api/execute-r2r.c -2 +2 M ./api/execute-split-dft-c2r.c -2 +2 M ./api/execute-split-dft-r2c.c -2 +2 M ./api/execute-split-dft.c -2 +2 M ./api/execute.c -2 +2 M ./api/export-wisdom-to-file.c -2 +2 M ./api/export-wisdom-to-string.c -2 +2 M ./api/export-wisdom.c -2 +2 M ./api/f77api.c -2 +2 M ./api/f77funcs.h -2 +2 M ./api/fftw3.h -2 +2 M ./api/flops.c -2 +2 M ./api/forget-wisdom.c -2 +2 M ./api/import-system-wisdom.c -2 +2 M ./api/import-wisdom-from-file.c -2 +2 M ./api/import-wisdom-from-string.c -2 +2 M ./api/import-wisdom.c -2 +2 M ./api/malloc.c -2 +2 M ./api/map-r2r-kind.c -2 +2 M ./api/mapflags.c -2 +2 M ./api/mkprinter-file.c -2 +2 M ./api/mktensor-iodims.h -2 +2 M ./api/mktensor-rowmajor.c -2 +2 M ./api/plan-dft-1d.c -2 +2 M ./api/plan-dft-2d.c -2 +2 M ./api/plan-dft-3d.c -2 +2 M ./api/plan-dft-c2r-1d.c -2 +2 M ./api/plan-dft-c2r-2d.c -2 +2 M ./api/plan-dft-c2r-3d.c -2 +2 M ./api/plan-dft-c2r.c -2 +2 M ./api/plan-dft-r2c-1d.c -2 +2 M ./api/plan-dft-r2c-2d.c -2 +2 M ./api/plan-dft-r2c-3d.c -2 +2 M ./api/plan-dft-r2c.c -2 +2 M ./api/plan-dft.c -2 +2 M ./api/plan-guru-dft-c2r.h -2 +2 M ./api/plan-guru-dft-r2c.h -2 +2 M ./api/plan-guru-dft.h -2 +2 M ./api/plan-guru-r2r.h -2 +2 M ./api/plan-guru-split-dft-c2r.h -2 +2 M ./api/plan-guru-split-dft-r2c.h -2 +2 M ./api/plan-guru-split-dft.h -2 +2 M ./api/plan-many-dft-c2r.c -2 +2 M ./api/plan-many-dft-r2c.c -2 +2 M ./api/plan-many-dft.c -2 +2 M ./api/plan-many-r2r.c -2 +2 M ./api/plan-r2r-1d.c -2 +2 M ./api/plan-r2r-2d.c -2 +2 M ./api/plan-r2r-3d.c -2 +2 M ./api/plan-r2r.c -2 +2 M ./api/print-plan.c -2 +2 M ./api/rdft2-pad.c -2 +2 M ./api/the-planner.c -2 +2 M ./api/version.c -2 +2 M ./api/x77.h -2 +2 M ./configure.ac -2 +2 M ./dft/bluestein.c -2 +2 M ./dft/buffered.c -2 +2 M ./dft/codelet-dft.h -2 +2 M ./dft/conf.c -2 +2 M ./dft/ct.c -2 +2 M ./dft/ct.h -2 +2 M ./dft/dft.h -2 +2 M ./dft/dftw-direct.c -2 +2 M ./dft/dftw-directsq.c -2 +2 M ./dft/dftw-generic.c -2 +2 M ./dft/dftw-genericbuf.c -2 +2 M ./dft/direct.c -2 +2 M ./dft/generic.c -2 +2 M ./dft/indirect-transpose.c -2 +2 M ./dft/indirect.c -2 +2 M ./dft/kdft-dif.c -2 +2 M ./dft/kdft-difsq.c -2 +2 M ./dft/kdft-dit.c -2 +2 M ./dft/kdft.c -2 +2 M ./dft/nop.c -2 +2 M ./dft/plan.c -2 +2 M ./dft/problem.c -2 +2 M ./dft/rader.c -2 +2 M ./dft/rank-geq2.c -2 +2 M ./dft/scalar/n.c -2 +2 M ./dft/scalar/n.h -2 +2 M ./dft/scalar/t.c -2 +2 M ./dft/scalar/t.h -2 +2 M ./dft/simd/n1b.c -2 +2 M ./dft/simd/n1b.h -2 +2 M ./dft/simd/n1f.c -2 +2 M ./dft/simd/n1f.h -2 +2 M ./dft/simd/n2b.c -2 +2 M ./dft/simd/n2b.h -2 +2 M ./dft/simd/n2f.c -2 +2 M ./dft/simd/n2f.h -2 +2 M ./dft/simd/n2s.c -2 +2 M ./dft/simd/n2s.h -2 +2 M ./dft/simd/q1b.c -2 +2 M ./dft/simd/q1b.h -2 +2 M ./dft/simd/q1f.c -2 +2 M ./dft/simd/q1f.h -2 +2 M ./dft/simd/t.c -2 +2 M ./dft/simd/t1b.h -2 +2 M ./dft/simd/t1bu.h -2 +2 M ./dft/simd/t1f.h -2 +2 M ./dft/simd/t1fu.h -2 +2 M ./dft/simd/t2b.h -2 +2 M ./dft/simd/t2f.h -2 +2 M ./dft/simd/t3b.h -2 +2 M ./dft/simd/t3f.h -2 +2 M ./dft/simd/ts.c -2 +2 M ./dft/simd/ts.h -2 +2 M ./dft/solve.c -2 +2 M ./dft/vrank-geq1.c -2 +2 M ./dft/zero.c -2 +2 M ./doc/f77_wisdom.f -2 +2 M ./genfft/algsimp.ml -2 +2 M ./genfft/algsimp.mli -2 +2 M ./genfft/annotate.ml -2 +2 M ./genfft/annotate.mli -2 +2 M ./genfft/assoctable.ml -2 +2 M ./genfft/assoctable.mli -2 +2 M ./genfft/c.ml -2 +2 M ./genfft/c.mli -2 +2 M ./genfft/complex.ml -2 +2 M ./genfft/complex.mli -2 +2 M ./genfft/conv.ml -2 +2 M ./genfft/conv.mli -2 +2 M ./genfft/dag.ml -2 +2 M ./genfft/dag.mli -2 +2 M ./genfft/expr.ml -2 +2 M ./genfft/expr.mli -2 +2 M ./genfft/fft.ml -2 +2 M ./genfft/fft.mli -2 +2 M ./genfft/gen_hc2c.ml -2 +2 M ./genfft/gen_hc2cdft.ml -2 +2 M ./genfft/gen_hc2cdft_c.ml -2 +2 M ./genfft/gen_hc2hc.ml -2 +2 M ./genfft/gen_mdct.ml -2 +2 M ./genfft/gen_notw.ml -2 +2 M ./genfft/gen_notw_c.ml -2 +2 M ./genfft/gen_r2cb.ml -2 +2 M ./genfft/gen_r2cf.ml -2 +2 M ./genfft/gen_r2r.ml -2 +2 M ./genfft/gen_twiddle.ml -2 +2 M ./genfft/gen_twiddle_c.ml -2 +2 M ./genfft/gen_twidsq.ml -2 +2 M ./genfft/gen_twidsq_c.ml -2 +2 M ./genfft/genutil.ml -2 +2 M ./genfft/littlesimp.ml -2 +2 M ./genfft/littlesimp.mli -2 +2 M ./genfft/magic.ml -2 +2 M ./genfft/monads.ml -2 +2 M ./genfft/number.ml -2 +2 M ./genfft/number.mli -2 +2 M ./genfft/oracle.ml -2 +2 M ./genfft/oracle.mli -2 +2 M ./genfft/schedule.ml -2 +2 M ./genfft/schedule.mli -2 +2 M ./genfft/simd.ml -2 +2 M ./genfft/simd.mli -2 +2 M ./genfft/simdmagic.ml -2 +2 M ./genfft/to_alist.ml -2 +2 M ./genfft/to_alist.mli -2 +2 M ./genfft/trig.ml -2 +2 M ./genfft/trig.mli -2 +2 M ./genfft/twiddle.ml -2 +2 M ./genfft/twiddle.mli -2 +2 M ./genfft/unique.ml -2 +2 M ./genfft/unique.mli -2 +2 M ./genfft/util.ml -2 +2 M ./genfft/util.mli -2 +2 M ./genfft/variable.ml -2 +2 M ./genfft/variable.mli -2 +2 M ./kernel/align.c -2 +2 M ./kernel/alloc.c -2 +2 M ./kernel/assert.c -2 +2 M ./kernel/awake.c -2 +2 M ./kernel/buffered.c -2 +2 M ./kernel/cpy1d.c -2 +2 M ./kernel/cpy2d-pair.c -2 +2 M ./kernel/cpy2d.c -2 +2 M ./kernel/ct.c -2 +2 M ./kernel/cycle.h -2 +2 M ./kernel/debug.c -2 +2 M ./kernel/extract-reim.c -2 +2 M ./kernel/hash.c -2 +2 M ./kernel/iabs.c -2 +2 M ./kernel/ifftw.h -2 +2 M ./kernel/kalloc.c -2 +2 M ./kernel/md5-1.c -2 +2 M ./kernel/md5.c -2 +2 M ./kernel/minmax.c -2 +2 M ./kernel/ops.c -2 +2 M ./kernel/pickdim.c -2 +2 M ./kernel/plan.c -2 +2 M ./kernel/primes.c -2 +2 M ./kernel/print.c -2 +2 M ./kernel/problem.c -2 +2 M ./kernel/rader.c -2 +2 M ./kernel/scan.c -2 +2 M ./kernel/solver.c -2 +2 M ./kernel/solvtab.c -2 +2 M ./kernel/stride.c -2 +2 M ./kernel/tensor.c -2 +2 M ./kernel/tensor1.c -2 +2 M ./kernel/tensor2.c -2 +2 M ./kernel/tensor3.c -2 +2 M ./kernel/tensor4.c -2 +2 M ./kernel/tensor5.c -2 +2 M ./kernel/tensor7.c -2 +2 M ./kernel/tensor8.c -2 +2 M ./kernel/tensor9.c -2 +2 M ./kernel/tile2d.c -2 +2 M ./kernel/timer.c -2 +2 M ./kernel/transpose.c -2 +2 M ./kernel/trig.c -2 +2 M ./kernel/twiddle.c -2 +2 M ./libbench2/dotens2.c -2 +2 M ./libbench2/my-getopt.c -2 +2 M ./libbench2/my-getopt.h -2 +2 M ./libbench2/verify-dft.c -2 +2 M ./libbench2/verify-lib.c -2 +2 M ./libbench2/verify-r2r.c -2 +2 M ./libbench2/verify-rdft2.c -2 +2 M ./libbench2/verify.h -2 +2 M ./mpi/any-true.c -2 +2 M ./mpi/api.c -2 +2 M ./mpi/block.c -2 +2 M ./mpi/choose-radix.c -2 +2 M ./mpi/conf.c -2 +2 M ./mpi/dft-problem.c -2 +2 M ./mpi/dft-rank-geq2-transposed.c -2 +2 M ./mpi/dft-rank-geq2.c -2 +2 M ./mpi/dft-rank1-bigvec.c -2 +2 M ./mpi/dft-rank1.c -2 +2 M ./mpi/dft-serial.c -2 +2 M ./mpi/dft-solve.c -2 +2 M ./mpi/dtensor.c -2 +2 M ./mpi/fftw3-mpi.h -2 +2 M ./mpi/ifftw-mpi.h -2 +2 M ./mpi/mpi-dft.h -2 +2 M ./mpi/mpi-rdft.h -2 +2 M ./mpi/mpi-rdft2.h -2 +2 M ./mpi/mpi-transpose.h -2 +2 M ./mpi/rdft-problem.c -2 +2 M ./mpi/rdft-rank-geq2-transposed.c -2 +2 M ./mpi/rdft-rank-geq2.c -2 +2 M ./mpi/rdft-rank1-bigvec.c -2 +2 M ./mpi/rdft-serial.c -2 +2 M ./mpi/rdft-solve.c -2 +2 M ./mpi/rdft2-problem.c -2 +2 M ./mpi/rdft2-rank-geq2-transposed.c -2 +2 M ./mpi/rdft2-rank-geq2.c -2 +2 M ./mpi/rdft2-serial.c -2 +2 M ./mpi/rdft2-solve.c -2 +2 M ./mpi/rearrange.c -2 +2 M ./mpi/testsched.c -2 +2 M ./mpi/transpose-alltoall.c -2 +2 M ./mpi/transpose-pairwise.c -2 +2 M ./mpi/transpose-problem.c -2 +2 M ./mpi/transpose-recurse.c -2 +2 M ./mpi/transpose-solve.c -2 +2 M ./mpi/wisdom-api.c -2 +2 M ./rdft/buffered.c -2 +2 M ./rdft/buffered2.c -2 +2 M ./rdft/codelet-rdft.h -2 +2 M ./rdft/conf.c -2 +2 M ./rdft/ct-hc2c-direct.c -2 +2 M ./rdft/ct-hc2c.c -2 +2 M ./rdft/ct-hc2c.h -2 +2 M ./rdft/dft-r2hc.c -2 +2 M ./rdft/dht-r2hc.c -2 +2 M ./rdft/dht-rader.c -2 +2 M ./rdft/direct-r2c.c -2 +2 M ./rdft/direct-r2r.c -2 +2 M ./rdft/direct2.c -2 +2 M ./rdft/generic.c -2 +2 M ./rdft/hc2hc-direct.c -2 +2 M ./rdft/hc2hc-generic.c -2 +2 M ./rdft/hc2hc.c -2 +2 M ./rdft/hc2hc.h -2 +2 M ./rdft/indirect.c -2 +2 M ./rdft/khc2c.c -2 +2 M ./rdft/khc2hc.c -2 +2 M ./rdft/kr2c.c -2 +2 M ./rdft/kr2r.c -2 +2 M ./rdft/nop.c -2 +2 M ./rdft/nop2.c -2 +2 M ./rdft/plan.c -2 +2 M ./rdft/plan2.c -2 +2 M ./rdft/problem.c -2 +2 M ./rdft/problem2.c -2 +2 M ./rdft/rank-geq2-rdft2.c -2 +2 M ./rdft/rank-geq2.c -2 +2 M ./rdft/rank0-rdft2.c -2 +2 M ./rdft/rank0.c -2 +2 M ./rdft/rdft-dht.c -2 +2 M ./rdft/rdft.h -2 +2 M ./rdft/rdft2-inplace-strides.c -2 +2 M ./rdft/rdft2-rdft.c -2 +2 M ./rdft/rdft2-strides.c -2 +2 M ./rdft/rdft2-tensor-max-index.c -2 +2 M ./rdft/scalar/hb.h -2 +2 M ./rdft/scalar/hc2c.c -2 +2 M ./rdft/scalar/hc2cb.h -2 +2 M ./rdft/scalar/hc2cf.h -2 +2 M ./rdft/scalar/hf.h -2 +2 M ./rdft/scalar/hfb.c -2 +2 M ./rdft/scalar/r2c.c -2 +2 M ./rdft/scalar/r2cb.h -2 +2 M ./rdft/scalar/r2cbIII.h -2 +2 M ./rdft/scalar/r2cf.h -2 +2 M ./rdft/scalar/r2cfII.h -2 +2 M ./rdft/scalar/r2r.c -2 +2 M ./rdft/scalar/r2r.h -2 +2 M ./rdft/simd/hc2cbv.c -2 +2 M ./rdft/simd/hc2cbv.h -2 +2 M ./rdft/simd/hc2cfv.c -2 +2 M ./rdft/simd/hc2cfv.h -2 +2 M ./rdft/solve.c -2 +2 M ./rdft/solve2.c -2 +2 M ./rdft/vrank-geq1-rdft2.c -2 +2 M ./rdft/vrank-geq1.c -2 +2 M ./rdft/vrank3-transpose.c -2 +2 M ./reodft/conf.c -2 +2 M ./reodft/redft00e-r2hc-pad.c -2 +2 M ./reodft/redft00e-r2hc.c -2 +2 M ./reodft/reodft.h -2 +2 M ./reodft/reodft010e-r2hc.c -2 +2 M ./reodft/reodft11e-r2hc-odd.c -2 +2 M ./reodft/reodft11e-r2hc.c -2 +2 M ./reodft/reodft11e-radix2.c -2 +2 M ./reodft/rodft00e-r2hc-pad.c -2 +2 M ./reodft/rodft00e-r2hc.c -2 +2 M ./simd/altivec.c -2 +2 M ./simd/nonportable/sse.c -2 +2 M ./simd/nonportable/sse2.c -2 +2 M ./simd/simd-altivec.h -2 +2 M ./simd/simd-sse.h -2 +2 M ./simd/simd-sse2.h -2 +2 M ./simd/simd.h -2 +2 M ./simd/sse.c -2 +2 M ./simd/sse2.c -2 +2 M ./simd/taint.c -2 +2 M ./simd/x86-cpuid.h -2 +2 M ./threads/api.c -2 +2 M ./threads/conf.c -2 +2 M ./threads/ct.c -2 +2 M ./threads/dft-vrank-geq1.c -2 +2 M ./threads/f77api.c -2 +2 M ./threads/f77funcs.h -2 +2 M ./threads/hc2hc.c -2 +2 M ./threads/openmp.c -2 +2 M ./threads/rdft-vrank-geq1.c -2 +2 M ./threads/threads.c -2 +2 M ./threads/threads.h -2 +2 M ./threads/vrank-geq1-rdft2.c -2 +2 M ./tools/fftw-wisdom-to-conf.1 -4 +4 M ./tools/fftw-wisdom-to-conf.in -2 +2 M ./tools/fftw-wisdom.c -2 +2 M ./tools/fftw_wisdom.1.in -4 +4 Wed Nov 5 16:40:31 EST 2008 athena@fftw.org * Store GPLv2 in darcs because automake installs GPLv3 these days. A ./COPYING Thu Oct 30 15:03:41 EDT 2008 athena@fftw.org * stylistic changes, comments M ./threads/threads.c -13 +17 Thu Oct 30 14:40:14 EDT 2008 athena@fftw.org * oops M ./threads/threads.c -27 +28 Thu Oct 30 14:30:08 EDT 2008 athena@fftw.org * simplification of the threading machinery M ./threads/threads.c -12 +7 Thu Oct 30 14:22:40 EDT 2008 athena@fftw.org * typo M ./threads/threads.c -1 +1 Thu Oct 30 13:42:07 EDT 2008 athena@fftw.org * [SECOND ATTEMPT] do not assume that a semaphore can be freed just because nobody is using it Let S be a semaphore, initially 0. Let thread A execute UP(S); let thread B execute DOWN(S); free(&S); It is unclear whether this code is correct with posix semaphores. The problem is whether UP() uses S after allowing DOWN() to continue; this seems to be the case in the glibc-2.7 implementation, and thus the pattern above seems to be incorrect. Avoid using such a pattern, and introduce a global semaphore for the unavoidable case when nothing else can be depended upon. M ./threads/threads.c -61 +80 Wed Oct 29 20:09:39 EDT 2008 stevenj@fftw.org * updated cpu codes from x86-1.21 M ./m4/ax_gcc_archflag.m4 -5 +7 Wed Oct 29 16:24:16 EDT 2008 athena@fftw.org * Previous change was bogus, need to find another way. M ./threads/threads.c -56 +42 Wed Oct 29 12:22:20 EDT 2008 athena@fftw.org * do not assume that a semaphore can be freed just because nobody is using it Let S be a semaphore, initially 0. Let thread A execute UP(S); let thread B execute DOWN(S); free(&S); It is unclear whether this code is correct with posix semaphores. The problem is whether UP() uses S after allowing DOWN() to continue; this seems to be the case in the glibc-2.7 implementation, and thus the pattern above seems to be incorrect. Avoid using such a pattern, and introduce a global semaphore for the unavoidable case when nothing else can be depended upon. M ./threads/threads.c -42 +56 Mon Oct 27 23:38:02 EDT 2008 stevenj@fftw.org * don't need PROG_AS any more M ./configure.ac -1 R ./m4/amx_prog_as.m4 Sun Oct 26 23:41:11 EDT 2008 stevenj@fftw.org * use AC_CONFIG_MACRO_DIR macro M ./Makefile.am -1 +1 M ./configure.ac +1 Sun Oct 26 10:08:44 EDT 2008 athena@fftw.org * Remove mpi/Makefile from configure.ac Otherwise, the tarball breaks because mpi/ is not in the distribution. M ./configure.ac -2 Sat Oct 25 17:13:50 EDT 2008 stevenj@fftw.org * remove MPI from dist until FFTW 3.3 M ./Makefile.am -1 +1 M ./NEWS +3 M ./doc/fftw3.texi -948 +9 Sat Oct 25 17:12:35 EDT 2008 stevenj@fftw.org * use MPIRUN even for -np 1 M ./mpi/Makefile.am -1 +1 Fri Jul 18 17:17:08 EDT 2008 stevenj@fftw.org * use new gcc arch=native flag as fallback M ./m4/ax_gcc_archflag.m4 -5 +5 Sat Oct 25 13:36:40 EDT 2008 athena@fftw.org * Use sem_t to implement mutexes Use sem_t instead of pthread_mutex_t to implement mutexes. It seems like pthread mutexes hang on linux-2.6.22 after several days of tests; the hang does not occur on linux >= 2.6.24 or when we use sem_t instead of pthread_mutex_t. The situation is still quite mysterious but this code seems to work. M ./threads/threads.c -22 +34 Thu Oct 23 13:32:45 EDT 2008 athena@fftw.org * print informative message when pstring is NULL. M ./libbench2/verify.c -1 +2 Sun Oct 19 16:00:07 EDT 2008 athena@fftw.org * Fix incorrect alignment in dftw-generic. Multithreaded dftw-generic is supposed to process only a slice of the array, but we were planning with the alignment of the original array rather than the slice. This led to unaligned accesses in certain obscure situations. M ./NEWS +3 M ./configure.ac -1 +1 M ./dft/dftw-generic.c -1 +2 Mon Aug 18 17:27:26 EDT 2008 Matteo Frigo * Paranoia: do not create OS threads while holding locks. Glibc at least plays silly games such as keeping a global variable that records whether there is more than one thread in the process, and it does not perform atomic operations if the variable says that there is only one thread. Who knows how this interacts with creating threads while holding a lock. Some day some genius will come up with some ``optimization'' that breaks everything. M ./threads/threads.c -7 +13 Wed Aug 6 07:41:46 EDT 2008 athena@fftw.org * Welcome to the quadcore era M ./mkdist.sh -1 +1 Mon Jun 16 16:46:39 EDT 2008 stevenj@fftw.org * backslash is technically not allowed in "echo" arguments; thanks to Debian Bug#486046 for pointing out problem and solution (and Raphael Geissert and Vincent Zweije, in particular) M ./tools/fftw-wisdom-to-conf.in -1 +1 Sun May 4 12:15:24 EDT 2008 stevenj@fftw.org * note problem with test program in gcc 4.1.2-4.2; thanks to Raymond Rogers for reporting it M ./doc/FAQ/fftw-faq.bfnn +6 Fri May 2 19:21:30 EDT 2008 stevenj@fftw.org * output count of constants along with other statistics M ./genfft/c.ml -3 +8 Sat Apr 19 14:15:03 EDT 2008 athena@fftw.org * Lower priority of unaligned SIMD codelets. List t1[fb]uv_* codelets before the corresponding aligned codelets, since the estimator picks the latter ones in case of a tie and aligned codelets are preferable. In other words, this is a hack. M ./dft/simd/codelets/Makefile.am -3 +2 Sat Apr 19 08:55:46 EDT 2008 Matteo Frigo * There is no point in using higher radices for unaligned codelets. M ./dft/simd/codelets/Makefile.am -4 +2 Fri Apr 18 19:01:27 EDT 2008 stevenj@fftw.org * support generating loopless, strideless r2r codelets M ./genfft/gen_r2r.ml -7 +17 Fri Apr 18 19:00:25 EDT 2008 stevenj@fftw.org * added Magic.threemult to use 3+3 complex-multiply variant when possible M ./genfft/complex.ml -2 +13 M ./genfft/magic.ml +4 Thu Apr 10 19:53:31 EDT 2008 stevenj@fftw.org * fix documentation bug - export_wisdom_to_string returns a string that should be deallocated with free, not fftw_free (thanks to Stein Vidar Hagfors Haugan for the bug report) M ./doc/fftw3.texi -1 +1 Mon Jan 21 01:11:44 EST 2008 stevenj@fftw.org * bsd calls x86_64 "amd64"; thanks to Fernando Herrero Carron for the bug report M ./m4/ax_gcc_archflag.m4 -3 +3 Tue Jan 1 12:29:56 EST 2008 stevenj@fftw.org * fix typo in manual, thanks to Yinon Ehrlich M ./doc/fftw3.texi -1 +1 Mon Dec 3 13:57:13 EST 2007 stevenj@fftw.org * note problem with gcc 3.4.4 on x86_64, thanks to Uwe Hollerbach for the report M ./doc/FAQ/fftw-faq.bfnn -2 +4 Tue Nov 13 16:19:22 EST 2007 stevenj@fftw.org * bump shared-lib revision M ./configure.ac -1 +1 Tue Nov 13 16:16:49 EST 2007 stevenj@fftw.org * update NEWS for alpha3 M ./NEWS +23 Tue Oct 2 13:53:04 EDT 2007 stevenj@fftw.org * fixed URL M ./m4/ax_openmp.m4 -1 +1 Mon Sep 17 19:38:29 EDT 2007 stevenj@fftw.org * added missing prototype M ./tests/fftw-bench.c +4 Tue Aug 14 22:35:06 EDT 2007 stevenj@fftw.org * terminology tweak M ./doc/fftw3.texi -1 +1 Wed Aug 1 18:44:21 EDT 2007 stevenj@fftw.org * check for pathscale compilers (thanks to Julian Cummings) M ./m4/ax_compiler_vendor.m4 -2 +2 Sat Sep 15 18:02:32 EDT 2007 athena@fftw.org * Avoid possible conflict with Windows include files. M ./threads/threads.c +5 Tue Aug 7 21:26:05 EDT 2007 athena@fftw.org * Distribute codlist.c for SIMD codelets in the commercial tarball. M ./commercialize.sh -1 +1 Wed Aug 1 10:33:41 EDT 2007 stevenj@fftw.org * some documentation clarifications, and documented FFTW_WISDOM_ONLY, at the suggestion of Mario Emmenlauer and Phil Dumont M ./api/apiplan.c -3 +3 M ./doc/fftw3.texi -3 +24 Tue Jul 31 16:52:56 EDT 2007 stevenj@fftw.org * bug fix in test program for vrank-3 transpose plans with vl=1 M ./mpi/mpi-bench.c -1 +1 Sun Jul 29 17:02:46 EDT 2007 stevenj@fftw.org * only run mpi checks for --enable-mpi M ./mpi/Makefile.am +4 Sun Jul 29 16:45:30 EDT 2007 stevenj@fftw.org * check for NULL return from spe_context_create in case SPE_MAP_PS not supported M ./cell/cell.c -4 +9 Sun Jul 29 15:56:57 EDT 2007 stevenj@fftw.org * use problem-state pointer to write SPE mailbox with lower latency (makes a significant performance difference for N < 32k), thanks to Jan Wagner for suggestion M ./cell/cell.c +16 Sun Jul 29 14:22:08 EDT 2007 stevenj@fftw.org * port cell code to SDK2.1 (libspe2), since libspe1 API is deprecated and can't be used in code that also uses libspe2 API M ./cell/cell.c -3 +49 M ./cell/copy.c -1 +1 M ./cell/dft-direct-cell.c -1 +1 M ./cell/fftw-cell.h -4 +5 M ./cell/spu/main.spuc -4 +4 M ./cell/transpose.c -1 +1 M ./configure.ac -2 +2 Sun Jul 29 11:46:24 EDT 2007 stevenj@fftw.org * bug fix: ego->W allocated with cell_aligned_malloc, so deallocate with free, not X(ifree0) M ./cell/dft-direct-cell.c -1 +1 Mon Jul 2 15:57:12 EDT 2007 stevenj@fftw.org * removed obsolete reference to CVS id M ./api/fftw3.h -4 Mon May 21 14:25:39 EDT 2007 athena@fftw.org * cycle counter for sun compiler M ./kernel/cycle.h -1 +1 Wed May 9 19:49:11 EDT 2007 stevenj@fftw.org * use __inline instead of inline for AIX routines (__inline is supported by gcc and xlc, whereas apparently "inline" is only supported by xlc if you specify -qlanglvl=stdc99 or similar); thanks to Jeff Haferman for the bug report M ./kernel/cycle.h -2 +2 Mon Apr 30 15:37:56 EDT 2007 stevenj@fftw.org * fixed incorrect type prefix (fftw_ vs. X(...)) in mpi/wisdom-api.c; thanks to Eric A. Borisch for the bug report M ./mpi/wisdom-api.c -6 +6 Wed Apr 25 21:21:39 EDT 2007 stevenj@fftw.org * some cleanups in MPI make check M ./mpi/Makefile.am -7 +11 Wed Apr 25 21:19:27 EDT 2007 stevenj@fftw.org * re-enable heuristic in the common case where we are not compiling for Cell M ./dft/rank-geq2.c -1 +3 Tue Apr 24 17:42:43 EDT 2007 athena@fftw.org * Removed duplicate codelet names, was breaking linker. M ./dft/simd/codelets/Makefile.am -4 +4 Tue Apr 24 11:38:16 EDT 2007 stevenj@fftw.org * added more codelets of sizes 5/10/20/25 to improve speed for round decimal sizes (speed improvements of 10-20%, at cost of 10-30% in library size) M ./dft/scalar/codelets/Makefile.am -3 +6 M ./dft/simd/codelets/Makefile.am -14 +26 M ./rdft/scalar/r2cb/Makefile.am -8 +16 M ./rdft/scalar/r2cf/Makefile.am -8 +16 M ./rdft/simd/codelets/Makefile.am -2 +4 Sat Mar 24 18:40:47 EDT 2007 stevenj@fftw.org * for 1d prime sizes, punt and return serial plan M ./mpi/api.c +4 Sat Mar 24 18:24:55 EDT 2007 stevenj@fftw.org * output reminders of the problem during bench --verify M ./libbench2/verify.c -2 +5 Sat Mar 24 18:10:24 EDT 2007 stevenj@fftw.org * bug fix - missing solver->destroy initializer in rdft2-rdft M ./rdft/rdft2-rdft.c -1 +1 Fri Mar 23 11:12:19 EDT 2007 stevenj@fftw.org * -static, in --enable-debug, doesn't work on MacOS X (according to Daniel Oberhoff) M ./configure.ac -1 Wed Mar 21 22:23:06 EDT 2007 stevenj@fftw.org * fix MPI r2c/c2r to work with howmany > 1 M ./mpi/rdft2-rank-geq2-transposed.c -1 +2 M ./mpi/rdft2-rank-geq2.c -1 +2 M ./mpi/rdft2-serial.c -2 +3 M ./tests/check.pl -1 Wed Mar 21 18:44:41 EDT 2007 stevenj@fftw.org * rm MPI version from TODO M ./TODO -2 Wed Mar 21 18:34:40 EDT 2007 stevenj@fftw.org * added 'make bigcheck' for MPI (no paranoid-check, unfortunately), and properly get MPIRUN from configure M ./configure.ac +2 M ./mpi/Makefile.am -7 +10 Wed Mar 21 18:23:18 EDT 2007 stevenj@fftw.org * bug fix - incorrect local_size returned for 1d bigvec case M ./mpi/api.c -1 +1 Wed Mar 21 03:13:54 EDT 2007 stevenj@fftw.org * hack to specify MPI_TRANSPOSED_IN/OUT via "[" and "]" in libbench2 problem M ./libbench2/bench-user.h +1 M ./libbench2/problem.c +6 M ./mpi/mpi-bench.c +12 M ./tests/check.pl -2 +7 Wed Mar 21 02:58:11 EDT 2007 stevenj@fftw.org * added MPI 'make check', still needs a bit of work M ./mpi/Makefile.am +28 M ./tests/check.pl +31 Wed Mar 21 02:47:10 EDT 2007 stevenj@fftw.org * bug fix in r2r transposed-input case M ./mpi/rdft-problem.c +3 Wed Mar 21 02:46:25 EDT 2007 stevenj@fftw.org * don't output more than 300 erroneous outputs (unless verbose > 2) M ./libbench2/verify-lib.c -2 +3 M ./libbench2/verify-r2r.c -2 +3 Wed Mar 21 01:48:54 EDT 2007 stevenj@fftw.org * fixed bug in transposed-in c2r MPI transforms ... seems to be working, finally M ./mpi/rdft2-problem.c -5 +6 Wed Mar 21 00:41:32 EDT 2007 stevenj@fftw.org * some fixes to MPI r2c/c2r transforms with transposed output/input M ./mpi/rdft2-rank-geq2-transposed.c -11 +17 Wed Mar 21 00:40:25 EDT 2007 stevenj@fftw.org * typos M ./NEWS -4 +4 Tue Mar 20 19:53:02 EDT 2007 stevenj@fftw.org * bug fix for mpi-bench with r2c/c2r: allocate a little bit extra to make sure that padding is allocated M ./libbench2/allocate.c -2 +4 Tue Mar 20 19:19:13 EDT 2007 stevenj@fftw.org * fix typo, thanks to Ernest Turro for the bug report M ./m4/ax_gcc_archflag.m4 -2 +2 Tue Mar 20 01:39:06 EDT 2007 stevenj@fftw.org * spacing tweaks M ./doc/fftw3.texi -2 +2 Tue Mar 20 00:53:11 EDT 2007 stevenj@fftw.org * Ralf Wildenhues is the one who pointed out that the self-communication could fill in the stalls in the pairwise schedule M ./mpi/testsched.c -3 +2 Tue Mar 20 00:22:25 EDT 2007 stevenj@fftw.org * add TRANSPOSED_OUT/IN support for r2c/c2r, respectively M ./doc/fftw3.texi -4 +1 M ./mpi/Makefile.am -1 +1 M ./mpi/README -3 M ./mpi/api.c -6 +15 M ./mpi/conf.c -1 +1 M ./mpi/rdft2-problem.c -18 +9 A ./mpi/rdft2-rank-geq2-transposed.c Mon Mar 19 21:45:34 EDT 2007 stevenj@fftw.org * yikes! fixed likely deadlock bug in MPI M ./mpi/transpose-recurse.c -2 +2 Mon Mar 19 21:38:52 EDT 2007 stevenj@fftw.org * comment M ./mpi/transpose-pairwise.c -2 +8 Mon Mar 19 21:30:44 EDT 2007 stevenj@fftw.org * s/alpha1/alpha2/ M ./NEWS -1 +1 Mon Mar 19 00:39:47 EDT 2007 stevenj@fftw.org * include README in dist tarball M ./mpi/Makefile.am -1 +1 Mon Mar 19 00:35:43 EDT 2007 stevenj@fftw.org * added MPI r2c/c2r transforms, some more documentation M ./NEWS -4 +9 M ./doc/fftw3.texi -6 +73 M ./libbench2/bench-user.h +2 M ./libbench2/problem.c -2 +4 M ./mpi/Makefile.am -1 +2 A ./mpi/README M ./mpi/api.c +140 M ./mpi/conf.c +4 M ./mpi/fftw3-mpi.h -88 +116 M ./mpi/mpi-bench.c -2 +102 A ./mpi/mpi-rdft2.h A ./mpi/rdft2-problem.c A ./mpi/rdft2-rank-geq2.c A ./mpi/rdft2-serial.c A ./mpi/rdft2-solve.c Sun Mar 18 23:14:29 EDT 2007 stevenj@fftw.org * set version to 3.2alpha2 M ./configure.ac -1 +1 Sun Mar 18 19:12:18 EDT 2007 stevenj@fftw.org * changed --enable-mips_ps to --enable-mips-ps; added Cell section to manual (from README.Cell); many minor updates to manual M ./configure.ac -1 +1 M ./doc/fftw3.texi -15 +243 Sun Mar 18 15:27:06 EDT 2007 stevenj@fftw.org * whoops, need to sync costs in problem_mpi_rdft M ./mpi/api.c +3 Sun Mar 18 12:44:49 EDT 2007 stevenj@fftw.org * documented guru64 interface M ./NEWS -1 +1 M ./doc/fftw3.texi -11 +81 Sun Mar 18 02:57:46 EDT 2007 stevenj@fftw.org * typo M ./doc/fftw3.texi -1 +1 Sun Mar 18 02:45:09 EDT 2007 stevenj@fftw.org * bumped copyright year to 2007 M ./COPYRIGHT -2 +2 M ./api/api.h -2 +2 M ./api/apiplan.c -2 +2 M ./api/configure.c -2 +2 M ./api/execute-dft-c2r.c -2 +2 M ./api/execute-dft-r2c.c -2 +2 M ./api/execute-dft.c -2 +2 M ./api/execute-r2r.c -2 +2 M ./api/execute-split-dft-c2r.c -2 +2 M ./api/execute-split-dft-r2c.c -2 +2 M ./api/execute-split-dft.c -2 +2 M ./api/execute.c -2 +2 M ./api/export-wisdom-to-file.c -2 +2 M ./api/export-wisdom-to-string.c -2 +2 M ./api/export-wisdom.c -2 +2 M ./api/f77api.c -2 +2 M ./api/f77funcs.h -2 +2 M ./api/fftw3.h -2 +2 M ./api/flops.c -2 +2 M ./api/forget-wisdom.c -2 +2 M ./api/import-system-wisdom.c -2 +2 M ./api/import-wisdom-from-file.c -2 +2 M ./api/import-wisdom-from-string.c -2 +2 M ./api/import-wisdom.c -2 +2 M ./api/malloc.c -2 +2 M ./api/map-r2r-kind.c -2 +2 M ./api/mapflags.c -2 +2 M ./api/mkprinter-file.c -2 +2 M ./api/mktensor-iodims.h -2 +2 M ./api/mktensor-rowmajor.c -2 +2 M ./api/plan-dft-1d.c -2 +2 M ./api/plan-dft-2d.c -2 +2 M ./api/plan-dft-3d.c -2 +2 M ./api/plan-dft-c2r-1d.c -2 +2 M ./api/plan-dft-c2r-2d.c -2 +2 M ./api/plan-dft-c2r-3d.c -2 +2 M ./api/plan-dft-c2r.c -2 +2 M ./api/plan-dft-r2c-1d.c -2 +2 M ./api/plan-dft-r2c-2d.c -2 +2 M ./api/plan-dft-r2c-3d.c -2 +2 M ./api/plan-dft-r2c.c -2 +2 M ./api/plan-dft.c -2 +2 M ./api/plan-guru-dft-c2r.h -2 +2 M ./api/plan-guru-dft-r2c.h -2 +2 M ./api/plan-guru-dft.h -2 +2 M ./api/plan-guru-r2r.h -2 +2 M ./api/plan-guru-split-dft-c2r.h -2 +2 M ./api/plan-guru-split-dft-r2c.h -2 +2 M ./api/plan-guru-split-dft.h -2 +2 M ./api/plan-many-dft-c2r.c -2 +2 M ./api/plan-many-dft-r2c.c -2 +2 M ./api/plan-many-dft.c -2 +2 M ./api/plan-many-r2r.c -2 +2 M ./api/plan-r2r-1d.c -2 +2 M ./api/plan-r2r-2d.c -2 +2 M ./api/plan-r2r-3d.c -2 +2 M ./api/plan-r2r.c -2 +2 M ./api/print-plan.c -2 +2 M ./api/rdft2-pad.c -2 +2 M ./api/the-planner.c -2 +2 M ./api/version.c -2 +2 M ./api/x77.h -2 +2 M ./dft/bluestein.c -2 +2 M ./dft/buffered.c -2 +2 M ./dft/codelet-dft.h -2 +2 M ./dft/conf.c -2 +2 M ./dft/ct.c -2 +2 M ./dft/ct.h -2 +2 M ./dft/dft.h -2 +2 M ./dft/dftw-direct.c -2 +2 M ./dft/dftw-directsq.c -2 +2 M ./dft/dftw-generic.c -2 +2 M ./dft/dftw-genericbuf.c -2 +2 M ./dft/direct.c -2 +2 M ./dft/generic.c -2 +2 M ./dft/indirect-transpose.c -2 +2 M ./dft/indirect.c -2 +2 M ./dft/kdft-dif.c -2 +2 M ./dft/kdft-difsq.c -2 +2 M ./dft/kdft-dit.c -2 +2 M ./dft/kdft.c -2 +2 M ./dft/nop.c -2 +2 M ./dft/plan.c -2 +2 M ./dft/problem.c -2 +2 M ./dft/rader.c -2 +2 M ./dft/rank-geq2.c -2 +2 M ./dft/scalar/n.c -2 +2 M ./dft/scalar/n.h -2 +2 M ./dft/scalar/t.c -2 +2 M ./dft/scalar/t.h -2 +2 M ./dft/simd/n1b.c -2 +2 M ./dft/simd/n1b.h -2 +2 M ./dft/simd/n1f.c -2 +2 M ./dft/simd/n1f.h -2 +2 M ./dft/simd/n2b.c -2 +2 M ./dft/simd/n2b.h -2 +2 M ./dft/simd/n2f.c -2 +2 M ./dft/simd/n2f.h -2 +2 M ./dft/simd/n2s.c -2 +2 M ./dft/simd/n2s.h -2 +2 M ./dft/simd/q1b.c -2 +2 M ./dft/simd/q1b.h -2 +2 M ./dft/simd/q1f.c -2 +2 M ./dft/simd/q1f.h -2 +2 M ./dft/simd/t.c -2 +2 M ./dft/simd/t1b.h -2 +2 M ./dft/simd/t1bu.h -2 +2 M ./dft/simd/t1f.h -2 +2 M ./dft/simd/t1fu.h -2 +2 M ./dft/simd/t2b.h -2 +2 M ./dft/simd/t2f.h -2 +2 M ./dft/simd/t3b.h -2 +2 M ./dft/simd/t3f.h -2 +2 M ./dft/simd/ts.c -2 +2 M ./dft/simd/ts.h -2 +2 M ./dft/solve.c -2 +2 M ./dft/vrank-geq1.c -2 +2 M ./dft/zero.c -2 +2 M ./doc/f77_wisdom.f -2 +2 M ./genfft/algsimp.ml -2 +2 M ./genfft/algsimp.mli -2 +2 M ./genfft/annotate.ml -2 +2 M ./genfft/annotate.mli -2 +2 M ./genfft/assoctable.ml -2 +2 M ./genfft/assoctable.mli -2 +2 M ./genfft/c.ml -2 +2 M ./genfft/c.mli -2 +2 M ./genfft/complex.ml -2 +2 M ./genfft/complex.mli -2 +2 M ./genfft/conv.ml -2 +2 M ./genfft/conv.mli -2 +2 M ./genfft/dag.ml -2 +2 M ./genfft/dag.mli -2 +2 M ./genfft/expr.ml -2 +2 M ./genfft/expr.mli -2 +2 M ./genfft/fft.ml -2 +2 M ./genfft/fft.mli -2 +2 M ./genfft/gen_hc2c.ml -2 +2 M ./genfft/gen_hc2cdft.ml -2 +2 M ./genfft/gen_hc2cdft_c.ml -2 +2 M ./genfft/gen_hc2hc.ml -2 +2 M ./genfft/gen_mdct.ml -2 +2 M ./genfft/gen_notw.ml -2 +2 M ./genfft/gen_notw_c.ml -2 +2 M ./genfft/gen_r2cb.ml -2 +2 M ./genfft/gen_r2cf.ml -2 +2 M ./genfft/gen_r2r.ml -2 +2 M ./genfft/gen_twiddle.ml -2 +2 M ./genfft/gen_twiddle_c.ml -2 +2 M ./genfft/gen_twidsq.ml -2 +2 M ./genfft/gen_twidsq_c.ml -2 +2 M ./genfft/genutil.ml -2 +2 M ./genfft/littlesimp.ml -2 +2 M ./genfft/littlesimp.mli -2 +2 M ./genfft/magic.ml -2 +2 M ./genfft/monads.ml -2 +2 M ./genfft/number.ml -2 +2 M ./genfft/number.mli -2 +2 M ./genfft/oracle.ml -2 +2 M ./genfft/oracle.mli -2 +2 M ./genfft/schedule.ml -2 +2 M ./genfft/schedule.mli -2 +2 M ./genfft/simd.ml -2 +2 M ./genfft/simd.mli -2 +2 M ./genfft/simdmagic.ml -2 +2 M ./genfft/to_alist.ml -2 +2 M ./genfft/to_alist.mli -2 +2 M ./genfft/trig.ml -2 +2 M ./genfft/trig.mli -2 +2 M ./genfft/twiddle.ml -2 +2 M ./genfft/twiddle.mli -2 +2 M ./genfft/unique.ml -2 +2 M ./genfft/unique.mli -2 +2 M ./genfft/util.ml -2 +2 M ./genfft/util.mli -2 +2 M ./genfft/variable.ml -2 +2 M ./genfft/variable.mli -2 +2 M ./kernel/align.c -2 +2 M ./kernel/alloc.c -2 +2 M ./kernel/assert.c -2 +2 M ./kernel/awake.c -2 +2 M ./kernel/buffered.c -2 +2 M ./kernel/cpy1d.c -2 +2 M ./kernel/cpy2d-pair.c -2 +2 M ./kernel/cpy2d.c -2 +2 M ./kernel/ct.c -2 +2 M ./kernel/cycle.h -2 +2 M ./kernel/debug.c -2 +2 M ./kernel/extract-reim.c -2 +2 M ./kernel/hash.c -2 +2 M ./kernel/iabs.c -2 +2 M ./kernel/ifftw.h -2 +2 M ./kernel/kalloc.c -2 +2 M ./kernel/md5-1.c -2 +2 M ./kernel/md5.c -2 +2 M ./kernel/minmax.c -2 +2 M ./kernel/ops.c -2 +2 M ./kernel/pickdim.c -2 +2 M ./kernel/plan.c -2 +2 M ./kernel/primes.c -2 +2 M ./kernel/print.c -2 +2 M ./kernel/problem.c -2 +2 M ./kernel/rader.c -2 +2 M ./kernel/scan.c -2 +2 M ./kernel/solver.c -2 +2 M ./kernel/solvtab.c -2 +2 M ./kernel/stride.c -2 +2 M ./kernel/tensor.c -2 +2 M ./kernel/tensor1.c -2 +2 M ./kernel/tensor2.c -2 +2 M ./kernel/tensor3.c -2 +2 M ./kernel/tensor4.c -2 +2 M ./kernel/tensor5.c -2 +2 M ./kernel/tensor7.c -2 +2 M ./kernel/tensor8.c -2 +2 M ./kernel/tensor9.c -2 +2 M ./kernel/tile2d.c -2 +2 M ./kernel/timer.c -2 +2 M ./kernel/transpose.c -2 +2 M ./kernel/trig.c -2 +2 M ./kernel/twiddle.c -2 +2 M ./libbench2/dotens2.c -2 +2 M ./libbench2/my-getopt.c -2 +2 M ./libbench2/my-getopt.h -2 +2 M ./libbench2/verify-dft.c -2 +2 M ./libbench2/verify-lib.c -2 +2 M ./libbench2/verify-r2r.c -2 +2 M ./libbench2/verify-rdft2.c -2 +2 M ./libbench2/verify.h -2 +2 M ./mpi/any-true.c -2 +2 M ./mpi/api.c -2 +2 M ./mpi/block.c -2 +2 M ./mpi/choose-radix.c -2 +2 M ./mpi/conf.c -2 +2 M ./mpi/dft-problem.c -2 +2 M ./mpi/dft-rank-geq2-transposed.c -2 +2 M ./mpi/dft-rank-geq2.c -2 +2 M ./mpi/dft-rank1-bigvec.c -2 +2 M ./mpi/dft-rank1.c -2 +2 M ./mpi/dft-serial.c -2 +2 M ./mpi/dft-solve.c -2 +2 M ./mpi/dtensor.c -2 +2 M ./mpi/fftw3-mpi.h -2 +2 M ./mpi/ifftw-mpi.h -2 +2 M ./mpi/mpi-dft.h -2 +2 M ./mpi/mpi-rdft.h -2 +2 M ./mpi/mpi-transpose.h -2 +2 M ./mpi/rdft-problem.c -2 +2 M ./mpi/rdft-rank-geq2-transposed.c -2 +2 M ./mpi/rdft-rank-geq2.c -2 +2 M ./mpi/rdft-rank1-bigvec.c -2 +2 M ./mpi/rdft-serial.c -2 +2 M ./mpi/rdft-solve.c -2 +2 M ./mpi/rearrange.c -2 +2 M ./mpi/testsched.c -2 +2 M ./mpi/transpose-alltoall.c -2 +2 M ./mpi/transpose-pairwise.c -2 +2 M ./mpi/transpose-problem.c -2 +2 M ./mpi/transpose-recurse.c -2 +2 M ./mpi/transpose-solve.c -2 +2 M ./mpi/wisdom-api.c -2 +2 M ./rdft/buffered.c -2 +2 M ./rdft/buffered2.c -2 +2 M ./rdft/codelet-rdft.h -2 +2 M ./rdft/conf.c -2 +2 M ./rdft/ct-hc2c-direct.c -2 +2 M ./rdft/ct-hc2c.c -2 +2 M ./rdft/ct-hc2c.h -2 +2 M ./rdft/dft-r2hc.c -2 +2 M ./rdft/dht-r2hc.c -2 +2 M ./rdft/dht-rader.c -2 +2 M ./rdft/direct-r2c.c -2 +2 M ./rdft/direct-r2r.c -2 +2 M ./rdft/direct2.c -2 +2 M ./rdft/generic.c -2 +2 M ./rdft/hc2hc-direct.c -2 +2 M ./rdft/hc2hc-generic.c -2 +2 M ./rdft/hc2hc.c -2 +2 M ./rdft/hc2hc.h -2 +2 M ./rdft/indirect.c -2 +2 M ./rdft/khc2c.c -2 +2 M ./rdft/khc2hc.c -2 +2 M ./rdft/kr2c.c -2 +2 M ./rdft/kr2r.c -2 +2 M ./rdft/nop.c -2 +2 M ./rdft/nop2.c -2 +2 M ./rdft/plan.c -2 +2 M ./rdft/plan2.c -2 +2 M ./rdft/problem.c -2 +2 M ./rdft/problem2.c -2 +2 M ./rdft/rank-geq2-rdft2.c -2 +2 M ./rdft/rank-geq2.c -2 +2 M ./rdft/rank0-rdft2.c -2 +2 M ./rdft/rank0.c -2 +2 M ./rdft/rdft-dht.c -2 +2 M ./rdft/rdft.h -2 +2 M ./rdft/rdft2-inplace-strides.c -2 +2 M ./rdft/rdft2-rdft.c -2 +2 M ./rdft/rdft2-strides.c -2 +2 M ./rdft/rdft2-tensor-max-index.c -2 +2 M ./rdft/scalar/hb.h -2 +2 M ./rdft/scalar/hc2c.c -2 +2 M ./rdft/scalar/hc2cb.h -2 +2 M ./rdft/scalar/hc2cf.h -2 +2 M ./rdft/scalar/hf.h -2 +2 M ./rdft/scalar/hfb.c -2 +2 M ./rdft/scalar/r2c.c -2 +2 M ./rdft/scalar/r2cb.h -2 +2 M ./rdft/scalar/r2cbIII.h -2 +2 M ./rdft/scalar/r2cf.h -2 +2 M ./rdft/scalar/r2cfII.h -2 +2 M ./rdft/scalar/r2r.c -2 +2 M ./rdft/scalar/r2r.h -2 +2 M ./rdft/simd/hc2cbv.c -2 +2 M ./rdft/simd/hc2cbv.h -2 +2 M ./rdft/simd/hc2cfv.c -2 +2 M ./rdft/simd/hc2cfv.h -2 +2 M ./rdft/solve.c -2 +2 M ./rdft/solve2.c -2 +2 M ./rdft/vrank-geq1-rdft2.c -2 +2 M ./rdft/vrank-geq1.c -2 +2 M ./rdft/vrank3-transpose.c -2 +2 M ./reodft/conf.c -2 +2 M ./reodft/redft00e-r2hc-pad.c -2 +2 M ./reodft/redft00e-r2hc.c -2 +2 M ./reodft/reodft.h -2 +2 M ./reodft/reodft010e-r2hc.c -2 +2 M ./reodft/reodft11e-r2hc-odd.c -2 +2 M ./reodft/reodft11e-r2hc.c -2 +2 M ./reodft/reodft11e-radix2.c -2 +2 M ./reodft/rodft00e-r2hc-pad.c -2 +2 M ./reodft/rodft00e-r2hc.c -2 +2 M ./simd/altivec.c -2 +2 M ./simd/nonportable/sse.c -2 +2 M ./simd/nonportable/sse2.c -2 +2 M ./simd/simd-altivec.h -2 +2 M ./simd/simd-sse.h -2 +2 M ./simd/simd-sse2.h -2 +2 M ./simd/simd.h -2 +2 M ./simd/sse.c -2 +2 M ./simd/sse2.c -2 +2 M ./simd/taint.c -2 +2 M ./simd/x86-cpuid.h -2 +2 M ./threads/api.c -2 +2 M ./threads/conf.c -2 +2 M ./threads/ct.c -2 +2 M ./threads/dft-vrank-geq1.c -2 +2 M ./threads/f77api.c -2 +2 M ./threads/f77funcs.h -2 +2 M ./threads/hc2hc.c -2 +2 M ./threads/openmp.c -2 +2 M ./threads/rdft-vrank-geq1.c -2 +2 M ./threads/threads.c -2 +2 M ./threads/threads.h -2 +2 M ./threads/vrank-geq1-rdft2.c -2 +2 M ./tools/fftw-wisdom-to-conf.1 -4 +4 M ./tools/fftw-wisdom-to-conf.in -2 +2 M ./tools/fftw-wisdom.c -2 +2 M ./tools/fftw_wisdom.1.in -4 +4 Sun Mar 18 01:41:40 EDT 2007 stevenj@fftw.org * noted CodeSourcery in AUTHORS M ./AUTHORS -1 +2 Sun Mar 18 01:25:00 EDT 2007 stevenj@fftw.org * more MPI documentation M ./doc/fftw3.texi -2 +61 Sat Mar 17 23:15:04 EDT 2007 stevenj@fftw.org * added MPI multi-dimensional rdft solvers & tests M ./mpi/Makefile.am -2 +4 M ./mpi/api.c +104 M ./mpi/conf.c +5 M ./mpi/fftw3-mpi.h -73 +90 M ./mpi/mpi-bench.c -1 +94 A ./mpi/mpi-rdft.h A ./mpi/rdft-problem.c A ./mpi/rdft-rank-geq2-transposed.c A ./mpi/rdft-rank-geq2.c A ./mpi/rdft-rank1-bigvec.c A ./mpi/rdft-serial.c A ./mpi/rdft-solve.c Sat Mar 17 22:52:00 EDT 2007 stevenj@fftw.org * whoops M ./mpi/dft-rank1-bigvec.c -1 +1 Sat Mar 17 22:43:54 EDT 2007 stevenj@fftw.org * clarification - fftw_mpi_init should be called before importing wisdom M ./doc/fftw3.texi -2 +4 Sat Mar 17 19:49:37 EDT 2007 stevenj@fftw.org * kindx/y/z -> kind0/1/2 for consistency M ./api/fftw3.h -3 +3 M ./doc/fftw3.texi -9 +9 Sat Mar 17 19:34:02 EDT 2007 stevenj@fftw.org * typo M ./mpi/dft-rank-geq2-transposed.c -1 +1 Sat Mar 17 19:14:16 EDT 2007 stevenj@fftw.org * some refactoring in preparation for mpi-rdft M ./mpi/Makefile.am -1 +1 M ./mpi/api.c -1 +1 M ./mpi/dft-problem.c -2 +2 M ./mpi/dft-rank1-bigvec.c -53 +5 M ./mpi/dtensor.c -3 +3 M ./mpi/ifftw-mpi.h -1 +29 A ./mpi/rearrange.c Sat Mar 17 18:12:45 EDT 2007 stevenj@fftw.org * documented more stuff for MPI M ./doc/fftw3.texi -2 +112 Sat Mar 17 15:41:23 EDT 2007 stevenj@fftw.org * added NEWS for 3.2alpha M ./NEWS -1 +37 Sat Mar 17 14:50:22 EDT 2007 stevenj@fftw.org * documented MPI transpose routines M ./doc/fftw3.texi -1 +130 Sat Mar 17 08:57:30 EDT 2007 athena@fftw.org * Removed unused variables M ./rdft/direct-r2c.c -3 Fri Mar 16 14:47:10 EDT 2007 athena@fftw.org * Preparing for interim release of Cell code. M ./AUTHORS +2 M ./configure.ac -1 +1 M ./doc/fftw3.texi +3 Thu Feb 8 12:23:43 EST 2007 athena@fftw.org * Added README.Cell A ./README.Cell M ./Makefile.am -1 +2 M ./README.Cell +103 Sat Mar 10 19:17:40 EST 2007 athena@fftw.org * Synchronized with main branch M ./simd/simd.h +10 Mon Jan 22 17:43:56 EST 2007 athena@fftw.org * Adapted vrecur heuristic to Cell. M ./cell/dft-direct-cell.c -1 +28 M ./dft/dft.h +4 Sun Jan 21 19:09:33 EST 2007 athena@fftw.org * synchronize with main branch. Sun Jan 21 14:42:00 EST 2007 athena@fftw.org * synchronized with main branch Thu Jan 18 20:29:22 EST 2007 athena@fftw.org * Increased MAX_N to 32K/sizeof(R). M ./cell/dft-direct-cell.c -23 +35 M ./cell/fftw-cell.h -3 +3 M ./cell/plans-double.c -9 +1033 M ./cell/plans-single.c -13 +525 M ./cell/spu/dft.spuc -2 +8 Thu Jan 18 13:43:51 EST 2007 Matteo Frigo * Added pointer to solver->destroy which is used in the Cell branch. M ./mpi/dft-rank-geq2-transposed.c -1 +1 M ./mpi/dft-rank1-bigvec.c -1 +1 M ./mpi/dft-rank1.c -1 +1 M ./mpi/transpose-recurse.c -1 +1 Thu Jan 18 12:09:26 EST 2007 athena@fftw.org * Updated copyright notices R ./mpi/transpose-radix2.c M ./cell/cell.c -2 +1 M ./cell/conf.c -2 +1 M ./cell/copy.c -2 +1 M ./cell/dft-direct-cell.c -2 +1 M ./cell/fftw-cell.h +19 M ./cell/spu/alloc.spuc +19 M ./cell/spu/copy.spuc +19 M ./cell/spu/dft.spuc +18 M ./cell/spu/dma.spuc +19 M ./cell/spu/execute.spuc +19 M ./cell/spu/fftw-spu.h +19 M ./cell/spu/main.spuc +19 M ./cell/spu/planner.spuc +19 M ./cell/spu/spu-double.h +19 M ./cell/spu/spu-single.h +19 M ./cell/spu/transpose.spuc +19 M ./cell/spufftw-embed.S +19 M ./cell/transpose.c -2 +1 Fri Jan 12 12:54:43 EST 2007 athena@fftw.org * Use mfc_read_tag_status_all() instead of spu_mfcstat(2), since the former seems to be standardized. M ./cell/spu/dma.spuc -1 +1 Thu Jan 11 14:55:08 EST 2007 athena@fftw.org * Silence some int/INT warnings. M ./cell/dft-direct-cell.c -21 +24 Wed Jan 10 18:19:53 EST 2007 athena@fftw.org * Note incompatibility of --enable-cell with --enable-threads M ./configure.ac +3 Wed Jan 10 17:57:10 EST 2007 athena@fftw.org * forgot to add file A ./cell/spufftw-embed.S Wed Jan 10 17:45:16 EST 2007 athena@fftw.org * 64-bit cleanup M ./cell/Makefile.am -4 +3 M ./cell/cell.c -7 +21 M ./cell/copy.c -2 +2 M ./cell/dft-direct-cell.c -11 +28 M ./cell/fftw-cell.h -8 +10 M ./cell/spu/dft.spuc -3 +3 M ./cell/spu/dma.spuc -7 +9 M ./cell/spu/fftw-spu.h -2 +2 M ./cell/transpose.c -1 +1 M ./configure.ac +3 M ./m4/ax_gcc_archflag.m4 -1 +1 Wed Jan 10 13:47:20 EST 2007 athena@fftw.org * Use -mcpu=cell where appropriate. M ./m4/ax_gcc_archflag.m4 +1 Tue Dec 26 21:35:59 EST 2006 athena@fftw.org * synchronized with main M ./dft/buffered.c -14 +1 Sun Dec 24 20:58:25 EST 2006 athena@fftw.org * synchronized with main branch M ./Makefile.am -12 +14 Sun Dec 24 13:47:37 EST 2006 athena@fftw.org * synchronized with main branch, updated to new sdk. M! ./cell/spu/Makefile.am -4 +3 M! ./cell/spu/dma.spuc -1 M! ./cell/spu/execute.spuc -2 +2 M! ./cell/spu/fftw-spu.h -29 +29 Thu Dec 21 17:17:41 EST 2006 athena@fftw.org * removed obsolete file R ./rdft/rdft2-radix2.c Tue Dec 19 15:17:20 EST 2006 athena@fftw.org * synchronized with main branch R ./rdft/direct.c M ./rdft/direct-r2c.c -1 +1 M ./rdft/direct-r2r.c -1 +1 Tue Dec 19 11:27:38 EST 2006 athena@fftw.org * Synchronized with main branch M! ./rdft/ct-hc2c.c -1 +1 M! ./rdft/direct2.c -13 +1 Fri Dec 15 16:04:31 EST 2006 athena@fftw.org * resolved conflict with main branch M ./tests/fftw-bench.c -3 +2 Fri Dec 8 14:43:50 EST 2006 athena@fftw.org * Fixes for compilation in subdirectories M ./cell/spu/Makefile.am -5 +3 Fri Dec 8 12:46:00 EST 2006 athena@fftw.org * Silence warning M ./rdft/rank0.c -2 Fri Dec 8 12:24:19 EST 2006 athena@fftw.org * silence warning M ./rdft/rank0.c -1 +1 Thu Dec 7 15:18:17 EST 2006 athena@fftw.org * Commented a particularly obscure piece of code. M ./cell/spu/dma.spuc -1 +14 Thu Dec 7 11:53:29 EST 2006 athena@fftw.org * Reorganized, clarified conditions for applicability of the DFT solver. M ./cell/dft-direct-cell.c -193 +178 Mon Dec 4 21:33:49 EST 2006 athena@fftw.org * Minor changes M ./cell/dft-direct-cell.c -1 +6 M ./cell/spu/dft.spuc -1 +7 Mon Dec 4 17:43:28 EST 2006 athena@fftw.org * Clarified comment M ./cell/dft-direct-cell.c -2 +2 Mon Dec 4 16:49:06 EST 2006 athena@fftw.org * Less incorrect conditions for fitting into local store. M ./cell/dft-direct-cell.c -5 +12 Mon Dec 4 16:08:24 EST 2006 athena@fftw.org * Implemented DECDIF+TRANSPOSE on Cell R ./dft/ctsq.c M ./cell/dft-direct-cell.c -93 +142 M ./cell/fftw-cell.h -8 +3 M ./cell/spu/dft.spuc -12 +21 M ./dft/indirect-transpose.c -1 +2 Fri Dec 1 17:42:55 EST 2006 athena@fftw.org * relaxed conditions of applicability of SPE M ./cell/dft-direct-cell.c -31 +42 Fri Dec 1 16:28:10 EST 2006 athena@fftw.org * tweaks M ./cell/dft-direct-cell.c -3 +2 Fri Dec 1 14:35:17 EST 2006 athena@fftw.org * Implemented Cell opcounts M ./cell/dft-direct-cell.c -4 +55 M ./cell/spu/execute.spuc +1 Fri Dec 1 13:38:44 EST 2006 athena@fftw.org * minor cleanup M ./cell/dft-direct-cell.c -9 +4 Fri Dec 1 11:16:52 EST 2006 athena@fftw.org * use [c0 s0 c1 s1] format for Cell twiddle factors, rather than [c0 c1 s0 s1]. This makes life easier and there is no speed penalty on Cell (unlike Altivec). M ./cell/copy.c -6 +6 M ./cell/dft-direct-cell.c -46 +15 M ./cell/spu/planner.spuc -2 +3 M ./cell/spu/spu-single.h -4 +4 Wed Nov 29 18:02:54 EST 2006 athena@fftw.org * Implemented SPE-accelerated copies A ./cell/copy.c M ./cell/Makefile.am -2 +2 M ./cell/copy.c +81 M ./cell/fftw-cell.h +14 M ./cell/spu/copy.spuc -69 +31 M ./cell/spu/dma.spuc -2 +78 M ./cell/spu/fftw-spu.h -4 +6 M ./cell/spu/main.spuc +4 M ./cell/spu/transpose.spuc -14 +6 M ./cell/transpose.c -4 +4 M ./kernel/ifftw.h -1 +3 M ./rdft/rank0.c +52 Wed Nov 29 12:11:08 EST 2006 athena@fftw.org * allow SPEs to compute vrank-0 problems. M ./cell/dft-direct-cell.c -10 +8 M ./cell/spu/dma.spuc -4 +4 Tue Nov 28 18:03:07 EST 2006 athena@fftw.org * eliminated DMA lists M ./cell/spu/dft.spuc -4 +4 M ./cell/spu/dma.spuc -119 +62 M ./cell/spu/fftw-spu.h -3 M ./cell/spu/main.spuc -2 +2 M ./cell/spu/transpose.spuc -6 +6 Tue Nov 28 14:22:05 EST 2006 athena@fftw.org * Conservatively force all dimensions to be 0 (mod VL) in cell, since otherwise it is too hard to get all cases right. M ./cell/dft-direct-cell.c -11 +10 M ./cell/spu/dft.spuc -1 +2 Tue Nov 28 12:39:01 EST 2006 athena@fftw.org * Check alignment of strides when transposing on Cell. M ./rdft/rank0.c -1 +10 Tue Nov 28 12:19:09 EST 2006 athena@fftw.org * consistent usage of FFT_SIGN M ./cell/dft-direct-cell.c -3 +3 M ./cell/spu/dft.spuc -2 +2 Tue Nov 28 11:35:38 EST 2006 athena@fftw.org * clever transposition algorithm without buffering M ./cell/spu/copy.spuc -178 +59 M ./cell/spu/dft.spuc -52 +8 M ./cell/spu/dma.spuc -54 +97 M ./cell/spu/fftw-spu.h -13 +3 M ./cell/spu/transpose.spuc -13 +11 Mon Nov 27 14:08:28 EST 2006 athena@fftw.org * Fixed tracking of dependencies M ./cell/spu/Makefile.am -1 +3 Mon Nov 27 14:03:53 EST 2006 athena@fftw.org * implemented 1D transforms, various tweaks M ./cell/cell.c -4 +14 M ./cell/conf.c +1 M ./cell/dft-direct-cell.c -15 +196 M ./cell/fftw-cell.h -2 +13 M ./cell/spu/Makefile.am +1 M ./cell/spu/alloc.spuc +1 M ./cell/spu/copy.spuc -34 +151 M ./cell/spu/dft.spuc -2 +37 M ./cell/spu/dma.spuc -13 +9 M ./cell/spu/fftw-spu.h -5 +9 M ./cell/spu/main.spuc -6 +3 Wed Nov 22 15:43:36 EST 2006 athena@fftw.org * no need to poll mailbox on spu side M ./cell/spu/main.spuc -1 Wed Nov 22 14:08:24 EST 2006 athena@fftw.org * increased maximum size handled by spe M ./cell/fftw-cell.h -2 +2 M ./cell/plans-double.c -2 +1027 M ./cell/plans-single.c -2 +514 M ./cell/spu/alloc.spuc -1 +1 M ./cell/spu/planner.spuc -1 +1 Tue Nov 21 16:23:17 EST 2006 athena@fftw.org * allow vrank<=2 problems in SPEs to avoid the vecloop overhead (grrr...) M ./cell/dft-direct-cell.c -40 +114 M ./cell/fftw-cell.h -1 +15 M ./cell/spu/dft.spuc -27 +37 M ./dft/rank-geq2.c +4 Mon Nov 20 14:41:45 EST 2006 athena@fftw.org * added emacs mode M ./cell/spu/copy.spuc +1 Mon Nov 20 09:34:12 EST 2006 athena@fftw.org * revised transpose, cleanup A ./cell/spu/transpose.spuc A ./cell/spu/dft.spuc M ./cell/spu/Makefile.am -1 +1 M ./cell/spu/dft.spuc +165 M ./cell/spu/fftw-spu.h +5 M ./cell/spu/main.spuc -224 +2 M ./cell/spu/transpose.spuc +70 Sun Nov 19 20:20:23 EST 2006 athena@fftw.org * added file A ./cell/spu/Makefile.am Sun Nov 19 20:18:35 EST 2006 athena@fftw.org * removed file R ./cell/spu/generate.sh Sun Nov 19 20:15:38 EST 2006 athena@fftw.org * better automake integration M! ./cell/Makefile.am -9 +4 R! ./cell/spu/alloc.c A! ./cell/spu/alloc.spuc R! ./cell/spu/copy.c A! ./cell/spu/copy.spuc R! ./cell/spu/dma.c A! ./cell/spu/dma.spuc R! ./cell/spu/execute.c A! ./cell/spu/execute.spuc R! ./cell/spu/main.c A! ./cell/spu/main.spuc R! ./cell/spu/planner.c A! ./cell/spu/planner.spuc R! ./cell/spu/spu_n2fv_10.c R! ./cell/spu/spu_n2fv_11.c R! ./cell/spu/spu_n2fv_12.c R! ./cell/spu/spu_n2fv_13.c R! ./cell/spu/spu_n2fv_14.c R! ./cell/spu/spu_n2fv_15.c R! ./cell/spu/spu_n2fv_16.c R! ./cell/spu/spu_n2fv_2.c R! ./cell/spu/spu_n2fv_3.c R! ./cell/spu/spu_n2fv_32.c R! ./cell/spu/spu_n2fv_4.c R! ./cell/spu/spu_n2fv_5.c R! ./cell/spu/spu_n2fv_6.c R! ./cell/spu/spu_n2fv_7.c R! ./cell/spu/spu_n2fv_8.c R! ./cell/spu/spu_n2fv_9.c R! ./cell/spu/spu_t1fv_10.c R! ./cell/spu/spu_t1fv_12.c R! ./cell/spu/spu_t1fv_15.c R! ./cell/spu/spu_t1fv_16.c R! ./cell/spu/spu_t1fv_2.c R! ./cell/spu/spu_t1fv_3.c R! ./cell/spu/spu_t1fv_32.c R! ./cell/spu/spu_t1fv_4.c R! ./cell/spu/spu_t1fv_5.c R! ./cell/spu/spu_t1fv_6.c R! ./cell/spu/spu_t1fv_7.c R! ./cell/spu/spu_t1fv_8.c R! ./cell/spu/spu_t1fv_9.c M! ./dft/indirect-transpose.c -1 +1 M! ./rdft/direct2.c -1 +13 Sat Nov 18 20:14:29 EST 2006 athena@fftw.org * changed algorithm for computing chunk size M ./cell/spu/main.c -32 +36 Sat Nov 18 19:18:11 EST 2006 athena@fftw.org * implemented transpose, various fixes. M! ./cell/Makefile.am -2 +2 M! ./cell/cell.c -2 +10 M! ./cell/conf.c -1 +1 M! ./cell/dft-direct-cell.c -13 +29 M! ./cell/fftw-cell.h -3 +13 A! ./cell/spu/copy.c A! ./cell/spu/dma.c M! ./cell/spu/fftw-spu.h +27 M! ./cell/spu/main.c -169 +83 A! ./cell/transpose.c M! ./configure.ac -2 +5 M! ./kernel/ifftw.h +5 M! ./rdft/rank0.c +24 M! ./tests/fftw-bench.c +4 Thu Nov 16 16:33:50 EST 2006 athena@fftw.org * Added explicit destructor to all solvers to help with the cell port. M! ./Makefile.am -2 +6 M! ./cell/cell.c -4 +22 M! ./cell/dft-direct-cell.c -1 +8 M! ./cell/fftw-cell.h -1 +3 M! ./configure.ac +1 M! ./dft/bluestein.c -1 +1 M! ./dft/buffered.c -1 +1 M! ./dft/ct.c -1 +1 A! ./dft/ctsq.c M! ./dft/direct.c -1 +1 M! ./dft/generic.c -1 +1 M! ./dft/indirect-transpose.c -1 +1 M! ./dft/indirect.c -1 +1 M! ./dft/nop.c -1 +1 M! ./dft/rader.c -1 +1 M! ./dft/rank-geq2.c -1 +1 M! ./dft/vrank-geq1.c -1 +1 M! ./kernel/ifftw.h +1 M! ./kernel/solver.c -1 +4 M! ./mpi/dft-rank-geq2.c -1 +1 M! ./mpi/dft-serial.c -1 +1 M! ./mpi/transpose-alltoall.c -1 +1 M! ./mpi/transpose-pairwise.c -1 +1 A! ./mpi/transpose-radix2.c M! ./rdft/buffered.c -1 +1 M! ./rdft/buffered2.c -1 +1 M! ./rdft/dft-r2hc.c -1 +1 M! ./rdft/dht-r2hc.c -1 +1 M! ./rdft/dht-rader.c -1 +1 A! ./rdft/direct.c M! ./rdft/direct2.c -1 +1 M! ./rdft/generic.c -1 +1 M! ./rdft/hc2hc.c -1 +1 M! ./rdft/indirect.c -1 +1 M! ./rdft/nop.c -1 +1 M! ./rdft/nop2.c -1 +1 M! ./rdft/rank-geq2-rdft2.c -1 +1 M! ./rdft/rank-geq2.c -1 +1 M! ./rdft/rank0-rdft2.c -1 +1 M! ./rdft/rank0.c -1 +1 M! ./rdft/rdft-dht.c -1 +1 A! ./rdft/rdft2-radix2.c M! ./rdft/vrank-geq1-rdft2.c -1 +1 M! ./rdft/vrank-geq1.c -1 +1 M! ./rdft/vrank3-transpose.c -1 +1 M! ./reodft/redft00e-r2hc-pad.c -1 +1 M! ./reodft/redft00e-r2hc.c -1 +1 M! ./reodft/reodft00e-splitradix.c -1 +1 M! ./reodft/reodft010e-r2hc.c -1 +1 M! ./reodft/reodft11e-r2hc-odd.c -1 +1 M! ./reodft/reodft11e-r2hc.c -1 +1 M! ./reodft/reodft11e-radix2.c -1 +1 M! ./reodft/rodft00e-r2hc-pad.c -1 +1 M! ./reodft/rodft00e-r2hc.c -1 +1 M! ./tests/fftw-bench.c -1 M! ./threads/ct.c -1 +1 M! ./threads/dft-vrank-geq1.c -1 +1 M! ./threads/hc2hc.c -1 +1 M! ./threads/rdft-vrank-geq1.c -1 +1 M! ./threads/vrank-geq1-rdft2.c -1 +1 Thu Nov 16 15:22:15 EST 2006 athena@fftw.org * consistent use of #if vs. #ifdef M ./api/configure.c +2 M ./cell/cell.c -1 +1 M ./cell/conf.c -1 +1 M ./cell/dft-direct-cell.c -1 +1 M ./cell/plans-double.c -1 +1 M ./cell/plans-single.c -1 +1 Thu Nov 16 15:15:34 EST 2006 athena@fftw.org * Additional Cell double codelets, better automake integration A! ./cell/spu/spu_n2fv_11.c A! ./cell/spu/spu_n2fv_13.c A! ./cell/spu/spu_n2fv_15.c A! ./cell/spu/spu_n2fv_3.c A! ./cell/spu/spu_n2fv_5.c A! ./cell/spu/spu_n2fv_7.c A! ./cell/spu/spu_n2fv_9.c M! ./Makefile.am -7 +5 M! ./cell/Makefile.am -2 +7 M! ./cell/cell.c -11 +7 M! ./cell/conf.c +26 M! ./cell/dft-direct-cell.c -9 +14 M! ./cell/fftw-cell.h -2 M! ./cell/plans-double.c -196 +201 M! ./cell/plans-single.c -3 +9 M! ./cell/spu/execute.c +10 M! ./cell/spu/fftw-spu.h -8 +22 M! ./cell/spu/generate.sh -1 +6 M! ./cell/spu/main.c -2 +1 M! ./cell/spu/planner.c +3 M! ./cell/spu/spu_n2fv_11.c +118 M! ./cell/spu/spu_n2fv_13.c +155 M! ./cell/spu/spu_n2fv_15.c +132 M! ./cell/spu/spu_n2fv_3.c +37 M! ./cell/spu/spu_n2fv_5.c +52 M! ./cell/spu/spu_n2fv_7.c +68 M! ./cell/spu/spu_n2fv_9.c +103 M! ./configure.ac +6 M! ./dft/buffered.c -1 +14 M! ./kernel/align.c -1 +1 M! ./kernel/ifftw.h -2 +2 M! ./simd/simd.h -4 M! ./simd/taint.c -1 +1 Thu Nov 16 12:43:34 EST 2006 athena@fftw.org * Use dma lists. M ./cell/spu/main.c -21 +86 Thu Nov 16 11:03:46 EST 2006 athena@fftw.org * converted to automake R ./cell/spu/Makefile.in M ./cell/Makefile.am +1 Wed Nov 15 18:00:12 EST 2006 athena@fftw.org * Initial port to Cell Broadband Engine. A! ./cell/ A! ./cell/spu/ A! ./cell/Makefile.am A! ./cell/fftw-cell.h A! ./cell/plans-single.c A! ./cell/spu/Makefile.in A! ./cell/spu/execute.c A! ./cell/spu/fftw-spu.h A! ./cell/spu/generate.sh A! ./cell/spu/main.c A! ./cell/spu/planner.c A! ./cell/spu/spu-single.h A! ./cell/spu/spu_n2fv_10.c A! ./cell/spu/spu_n2fv_12.c A! ./cell/spu/spu_n2fv_14.c A! ./cell/spu/spu_n2fv_16.c A! ./cell/spu/spu_n2fv_2.c A! ./cell/spu/spu_n2fv_32.c A! ./cell/spu/spu_n2fv_4.c A! ./cell/spu/spu_n2fv_6.c A! ./cell/spu/spu_n2fv_8.c A! ./cell/spu/spu_t1fv_10.c A! ./cell/spu/spu_t1fv_12.c A! ./cell/spu/spu_t1fv_15.c A! ./cell/spu/spu_t1fv_16.c A! ./cell/spu/spu_t1fv_2.c A! ./cell/spu/spu_t1fv_3.c A! ./cell/spu/spu_t1fv_32.c A! ./cell/spu/spu_t1fv_4.c A! ./cell/spu/spu_t1fv_5.c A! ./cell/spu/spu_t1fv_6.c A! ./cell/spu/spu_t1fv_7.c A! ./cell/spu/spu_t1fv_8.c A! ./cell/spu/spu_t1fv_9.c A! ./cell/cell.c A! ./cell/conf.c A! ./cell/dft-direct-cell.c A! ./cell/spu/spu-double.h A! ./cell/plans-double.c A! ./cell/spu/alloc.c M! ./Makefile.am -14 +14 M! ./api/configure.c +1 M! ./cell/Makefile.am +22 M! ./cell/cell.c +90 M! ./cell/conf.c +12 M! ./cell/dft-direct-cell.c +302 M! ./cell/fftw-cell.h +64 M! ./cell/plans-double.c +2056 M! ./cell/plans-single.c +1032 M! ./cell/spu/Makefile.in +35 M! ./cell/spu/alloc.c +24 M! ./cell/spu/execute.c +68 M! ./cell/spu/fftw-spu.h +86 M! ./cell/spu/generate.sh +16 M! ./cell/spu/main.c +281 M! ./cell/spu/planner.c +187 M! ./cell/spu/spu-double.h +105 M! ./cell/spu/spu-single.h +120 M! ./cell/spu/spu_n2fv_10.c +100 M! ./cell/spu/spu_n2fv_12.c +109 M! ./cell/spu/spu_n2fv_14.c +142 M! ./cell/spu/spu_n2fv_16.c +146 M! ./cell/spu/spu_n2fv_2.c +33 M! ./cell/spu/spu_n2fv_32.c +313 M! ./cell/spu/spu_n2fv_4.c +44 M! ./cell/spu/spu_n2fv_6.c +62 M! ./cell/spu/spu_n2fv_8.c +75 M! ./cell/spu/spu_t1fv_10.c +104 M! ./cell/spu/spu_t1fv_12.c +114 M! ./cell/spu/spu_t1fv_15.c +162 M! ./cell/spu/spu_t1fv_16.c +153 M! ./cell/spu/spu_t1fv_2.c +32 M! ./cell/spu/spu_t1fv_3.c +41 M! ./cell/spu/spu_t1fv_32.c +329 M! ./cell/spu/spu_t1fv_4.c +44 M! ./cell/spu/spu_t1fv_5.c +60 M! ./cell/spu/spu_t1fv_6.c +63 M! ./cell/spu/spu_t1fv_7.c +81 M! ./cell/spu/spu_t1fv_8.c +77 M! ./cell/spu/spu_t1fv_9.c +120 M! ./configure.ac +2 M! ./kernel/ifftw.h -17 +21 M! ./libbench2/util.c -1 +1 M! ./tests/fftw-bench.c +1 Wed Mar 14 10:19:53 EDT 2007 athena@fftw.org * Remove Codesourcery contributions from commercial tarball. M ./commercialize.sh +7 Wed Mar 14 08:59:18 EDT 2007 athena@fftw.org * Added FFTW_WISDOM_ONLY, at the request of Phil Dumont. M ./api/apiplan.c -19 +29 M ./api/fftw3.h +1 M ./tests/fftw-bench.c +1 Tue Mar 13 00:32:05 EDT 2007 stevenj@fftw.org * fixed potential MPI deadlock if timer misbehaves M ./kernel/timer.c -3 +3 Mon Mar 12 23:31:52 EDT 2007 stevenj@fftw.org * more work on MPI documentation M ./doc/fftw3.texi -31 +309 M ./mpi/api.c -3 +4 M ./mpi/fftw3-mpi.h -19 +19 Tue Feb 27 13:48:43 EST 2007 stevenj@fftw.org * index M ./doc/fftw3.texi +1 Tue Feb 27 13:46:45 EST 2007 stevenj@fftw.org * rename "new-data execute" to "new-array execute", since of course you do not need a new array to have new data M ./doc/FAQ/fftw-faq.bfnn -1 +2 M ./doc/fftw3.texi -12 +12 Tue Feb 27 13:43:55 EST 2007 stevenj@fftw.org * consistency with manual (guru execute -> new-data execute) M ./doc/FAQ/fftw-faq.bfnn -1 +1 Tue Feb 27 13:42:24 EST 2007 stevenj@fftw.org * texinfo fixes; renamed "guru execute" section to "new-data execute", since previously it seemed to lead to endless confusion with the guru planner API M ./doc/fftw3.texi -60 +83 Mon Feb 26 18:57:11 EST 2007 stevenj@fftw.org * consistently use n0/n1/.. everywhere instead of nx/ny/... (for consistency with d-dimensional case n[0], n[1], ...) ... first start at MPI documentation M ./api/fftw3.h -10 +10 M ./doc/fftw3.texi -138 +374 M ./mpi/fftw3-mpi.h -27 +27 Sat Mar 10 18:48:05 EST 2007 athena@fftw.org * Changed C++-style comment into K&R M ./kernel/cycle.h -1 +1 Sat Mar 10 18:47:12 EST 2007 athena@fftw.org * Forgot to add file A ./simd/simd-mips_ps.h Sat Mar 10 18:44:39 EST 2007 athena@fftw.org * Note removal of K7 support. M ./AUTHORS -1 +4 Sat Mar 10 18:41:52 EST 2007 athena@fftw.org * Updated manual for MIPS PS M ./doc/fftw3.texi -9 +12 Sat Mar 10 18:37:07 EST 2007 athena@fftw.org * Adopted MIPS_PS patches from Codesourcery. A ./simd/mips_ps.c A ./simd/mips_ps.h M ./configure.ac +23 M ./kernel/cycle.h +43 M ./kernel/ifftw.h -1 +2 M ./kernel/timer.c -1 +3 M ./simd/Makefile.am -1 +1 M ./simd/mips_ps.c +45 M ./simd/mips_ps.h +273 M ./simd/simd.h +4 Sun Feb 25 11:34:51 EST 2007 athena@fftw.org * Incorrect initialization of win32 semaphores M ./threads/threads.c -1 +1 Tue Jan 30 11:43:09 EST 2007 stevenj@fftw.org * win32 fixes (I think, still untested) M ./threads/threads.c -9 +10 Fri Jan 19 17:31:47 EST 2007 stevenj@fftw.org * message-size heuristic in tranpose-recurse M ./mpi/transpose-recurse.c -3 +15 Tue Jan 30 08:53:55 EST 2007 athena@fftw.org * Threading layer for Win32, completely untested. M ./threads/threads.c +69 Mon Jan 29 14:26:30 EST 2007 athena@fftw.org * Check for EINTR after sem_wait(), as suggested by Chip Salzenberg. M ./threads/threads.c -1 +11 Mon Jan 22 13:58:23 EST 2007 athena@fftw.org * Force vector recursion by means of a separate function pointer. I need this for Cell. M ./dft/ct.c -3 +7 M ./dft/ct.h -2 +8 M ./dft/dft.h -15 M ./dft/dftw-direct.c -2 +2 M ./dft/dftw-directsq.c -2 +3 M ./dft/dftw-generic.c -2 +2 M ./dft/dftw-genericbuf.c -2 +2 M ./dft/vrank-geq1.c -4 M ./threads/ct.c -1 +3 M ./threads/threads.h -1 +3 Mon Jan 22 09:28:35 EST 2007 athena@fftw.org * Merge multiplications by twiddle with multiplications by i for faster r2c transforms. M ./genfft/algsimp.ml +4 M ./genfft/simd.ml +4 M ./simd/simd-altivec.h +24 M ./simd/simd-sse.h +18 M ./simd/simd-sse2.h +18 Sun Jan 21 19:02:44 EST 2007 athena@fftw.org * Disabled vector recursion, too messy. M ./dft/dft.h -2 +13 Sun Jan 21 14:23:35 EST 2007 athena@fftw.org * Changed heuristics for vector recursion. As in fftw-3.1, NO_VRECURSE disables vector recursion. As an exception, however, vector recursion is allowed when the predicate VRECURSE_ANYWAYP is true. We need some form of vector recursion to obtain decent plans on Cell, and this solution captures the common cases without increasing planning time too much. M ./api/fftw3.h -1 +1 M ./api/mapflags.c -2 +2 M ./dft/ct.c -10 +6 M ./dft/dft.h +4 M ./dft/vrank-geq1.c -6 +3 M ./kernel/ifftw.h -2 +2 M ./kernel/planner.c -1 +1 M ./rdft/ct-hc2c.c -9 +4 M ./rdft/hc2hc.c -9 +4 M ./rdft/vrank-geq1.c -7 Sun Dec 17 22:31:17 EST 2006 athena@fftw.org * fixed hc2c for vector-recursion branch M ./rdft/ct-hc2c.c -2 +7 Tue Dec 5 12:52:36 EST 2006 athena@fftw.org * switch to default vector recursion M ./api/fftw3.h -1 +1 M ./api/mapflags.c -2 +2 M ./dft/ct.c -4 +6 M ./dft/vrank-geq1.c +7 M ./kernel/ifftw.h -2 +2 M ./kernel/planner.c -1 +1 M ./rdft/hc2hc.c -2 +7 M ./rdft/vrank-geq1.c -2 +9 Sat Jan 20 23:37:33 EST 2007 athena@fftw.org * Smarter algorithm for selection of nbuf. M ./kernel/buffered.c -8 +6 Sat Jan 20 22:15:33 EST 2007 athena@fftw.org * Increased buffer sizes according to Moore's law. M ./kernel/buffered.c -3 +5 Fri Jan 19 16:02:00 EST 2007 stevenj@fftw.org * fix another MPI synchronization bug -- several more places where cost_hook must be called to synchronized process timings (sigh) M ./api/flops.c -3 +3 M ./kernel/ifftw.h -4 +5 M ./kernel/planner.c -10 +6 M ./kernel/timer.c -4 +15 M ./mpi/api.c -3 +1 Fri Jan 19 12:08:07 EST 2007 athena@fftw.org * Set havewisdom=0 when calling forget_wisdom() in the test program. M ./tests/fftw-bench.c -1 +3 Fri Jan 19 10:29:56 EST 2007 stevenj@fftw.org * remove redundant check M ./kernel/planner.c -4 +1 Thu Jan 18 22:37:59 EST 2007 stevenj@fftw.org * fixed potential (unlikely) bug in wisdom import (triggered when importing impatient wisdom after creating more patient plans, but apparently only for nonstandard configure.c configurations) M ./kernel/planner.c -1 +5 Thu Jan 18 21:50:14 EST 2007 stevenj@fftw.org * added functions to gather/broadcast wisdom for MPI M ./configure.ac -3 +8 M ./mpi/Makefile.am -1 +1 M ./mpi/api.c -3 +3 M ./mpi/fftw3-mpi.h -1 +5 M ./mpi/mpi-bench.c -19 +40 A ./mpi/wisdom-api.c M ./tests/bench.c +10 M ./tests/fftw-bench.c -3 +3 M ./tests/fftw-bench.h +2 Thu Jan 11 18:33:17 EST 2007 stevenj@fftw.org * whoops, another int/INT bug M ./mpi/transpose-pairwise.c -1 +1 Thu Jan 11 17:42:24 EST 2007 stevenj@fftw.org * whoops, fixed bug in transpose-recurse for r != m M ./mpi/transpose-recurse.c -10 +10 Thu Jan 11 17:25:36 EST 2007 stevenj@fftw.org * canonicalize mpi-transposed flags by setting TRANSPOSED_IN/OUT where possible M ./mpi/transpose-problem.c -1 +8 Thu Jan 11 17:16:24 EST 2007 stevenj@fftw.org * replace transpose-radix2 with much more general transpose-recurse solver M ./kernel/Makefile.am -3 +3 M ./kernel/ifftw.h +5 M ./kernel/tensor2.c -21 A ./kernel/tensor3.c M ./mpi/Makefile.am -1 +1 M ./mpi/conf.c -1 +1 M ./mpi/dft-rank-geq2-transposed.c -1 M ./mpi/mpi-bench.c +1 M ./mpi/mpi-transpose.h -1 +1 R ./mpi/transpose-radix2.c A ./mpi/transpose-recurse.c Wed Jan 10 20:23:48 EST 2007 stevenj@fftw.org * rename transpose-inplace to transpose-pairwise, as the algorithm is not restricted to inplace operation ./mpi/transpose-inplace.c -> ./mpi/transpose-pairwise.c M ./mpi/Makefile.am -1 +1 M ./mpi/conf.c -1 +1 M ./mpi/mpi-transpose.h -1 +1 M ./mpi/testsched.c -1 +1 M ./mpi/transpose-pairwise.c -9 +9 Wed Jan 10 14:39:08 EST 2007 stevenj@fftw.org * whoops, some int/INT bugs M ./mpi/api.c -2 +2 M ./mpi/block.c -3 +3 M ./mpi/ifftw-mpi.h -5 +5 Tue Jan 9 18:50:07 EST 2007 stevenj@fftw.org * fix FAQ Makefile for vpath builds M ./doc/FAQ/Makefile.am -4 +7 M ./doc/FAQ/fftw-faq.bfnn -1 +1 Tue Jan 9 20:22:11 EST 2007 athena@fftw.org * Missing ``static'' keyword. M ./rdft/direct-r2c.c -2 +2 Tue Jan 9 20:13:18 EST 2007 athena@fftw.org * Minor cleanup. M ./dft/direct.c -27 +16 Tue Jan 9 00:04:03 EST 2007 stevenj@fftw.org * interleave twiddle mults with DFTs (should we use dftw?) M ./mpi/dft-rank1.c -41 +44 Mon Jan 8 18:35:41 EST 2007 stevenj@fftw.org * simplified (and somewhat sped up) dft-rank1 by exploiting dft-rank1-bigvec M ./mpi/dft-rank-geq2-transposed.c -1 +1 M ./mpi/dft-rank-geq2.c -1 +1 M ./mpi/dft-rank1-bigvec.c -2 +4 M ./mpi/dft-rank1.c -175 +181 M ./mpi/transpose-alltoall.c -6 +3 M ./mpi/transpose-inplace.c -6 +3 M ./mpi/transpose-radix2.c -7 +2 Sun Jan 7 00:31:31 EST 2007 stevenj@fftw.org * rearranged TRANSPOSED format, numerous speedups Split the TRANSPOSED and non-TRANSPOSED rank-geq2 solvers, and changed the DFT TRANSPOSED format to be more like fftw2 (both globally and locally transposed). In general, more emphasis on arranging the data contiguously for the DFTs, and more flexibility in intermediate transposed formats. Also disable NO_SLOW when planning transposes, since otherwise non-square in-place transposes gratuitously put the planner in SLOW mode. Currently, dft-rank1-bigvec has 5 variants (or 10, if DESTROY_INPUT). It looks like only 2 of these are commonly used, so I should probably add some UGLY tags once I do more benchmarking. M ./mpi/Makefile.am -1 +1 M ./mpi/api.c -14 +34 M ./mpi/conf.c +1 M ./mpi/dft-problem.c -2 +12 A ./mpi/dft-rank-geq2-transposed.c M ./mpi/dft-rank-geq2.c -131 +60 M ./mpi/dft-rank1-bigvec.c -77 +123 M ./mpi/dft-rank1.c -2 +2 M ./mpi/ifftw-mpi.h -2 +3 M ./mpi/mpi-bench.c +13 M ./mpi/mpi-dft.h -1 +2 M ./mpi/transpose-alltoall.c -32 +39 M ./mpi/transpose-inplace.c -53 +69 M ./mpi/transpose-radix2.c -37 +43 Thu Jan 4 19:13:17 EST 2007 stevenj@fftw.org * add bench_cost_postprocess to prevent deadlocks in mpi-bench M ./libbench2/Makefile.am -5 +6 A ./libbench2/bench-cost-postprocess.c M ./libbench2/bench-user.h +1 M ./libbench2/speed.c -2 +2 M ./mpi/mpi-bench.c +7 Thu Jan 4 16:46:29 EST 2007 stevenj@fftw.org * whoops M ./m4/acx_pthread.m4 -1 +1 Wed Jan 3 14:23:42 EST 2007 stevenj@fftw.org * pass proper pointer types as arguments, so that ACX_PTHREAD still works with C++ and -Werror (thanks to Ewald Arnold for the suggestion) M ./m4/acx_pthread.m4 -3 +6 Mon Jan 1 19:30:43 EST 2007 athena@fftw.org * Renamed [io]vs => [io]vs_by_nbuf, which is more appropriate and would have saved me 30mins debugging. M ./dft/buffered.c -6 +6 Mon Jan 1 18:52:38 EST 2007 stevenj@fftw.org * add --with-g77-wrappers option & always include g77 wrappers on GNU systems and/or with gfortran Upcoming GNU/Linux distros will most likely switch to configuring FFTW with gfortran by default, since g77 isn't even included with recent gcc versions. However, we still want to include g77-compatible wrappers in this case (two underscores) in addition to gfortran wrappers (one underscore) lest we silently break binary compatibility and provoke lots of annoying emails. M ./api/x77.h +6 M ./configure.ac +11 Mon Jan 1 16:48:36 EST 2007 stevenj@fftw.org * use AC_HELP_STRING for --disable-fortran M ./configure.ac -1 +1 Mon Jan 1 15:56:12 EST 2007 stevenj@fftw.org * terminology M ./doc/FAQ/fftw-faq.bfnn -1 +1 Sat Dec 30 16:18:35 EST 2006 athena@fftw.org * Free buffers before calling cldrest. M ./dft/buffered.c -2 +2 M ./rdft/buffered.c -4 +4 Fri Dec 29 10:52:15 EST 2006 athena@fftw.org * Removed obsolete code. M ./simd/simd-sse.h -22 +9 Thu Dec 28 21:37:48 EST 2006 athena@fftw.org * Attempt to work-around old gcc bugs in a more efficient fashion that does not lose performance on newer gcc's. M ./simd/simd-sse.h -13 +24 Thu Dec 28 16:10:33 EST 2006 athena@fftw.org * Make sure that the speed() input is zero even in paranoid mode. M ./libbench2/speed.c +6 Thu Dec 28 11:41:46 EST 2006 athena@fftw.org * cld0 and cldm problems must be tainted because they are used in a v-loop. M ./rdft/ct-hc2c-direct.c -2 +5 M ./rdft/hc2hc-direct.c -2 +3 Wed Dec 27 17:17:45 EST 2006 athena@fftw.org * Run paranoid-check in patient mode. M ./tests/Makefile.am -5 +5 Wed Dec 27 10:51:42 EST 2006 athena@fftw.org * Fixed incorrect initialization to zero. M ./rdft/problem2.c -1 +1 Wed Dec 27 09:33:02 EST 2006 athena@fftw.org * Fixed wrong TAINT() M ./rdft/buffered2.c -1 +1 Tue Dec 26 22:50:28 EST 2006 athena@fftw.org * Grrrrr... M ./libbench2/verify-rdft2.c -1 +1 Tue Dec 26 22:48:44 EST 2006 athena@fftw.org * Give up trying to verify rdft2 when vrank=-infinity. M ./libbench2/verify-rdft2.c +3 Tue Dec 26 22:31:38 EST 2006 athena@fftw.org * typo M ./dft/direct.c -1 +2 Tue Dec 26 21:54:53 EST 2006 athena@fftw.org * Correctly verify rdft2 when vrank = -infinity. M ./libbench2/verify-rdft2.c -1 +1 Tue Dec 26 21:25:02 EST 2006 athena@fftw.org * rdft/buffered2.c now generates rdft2 subproblems, not rdft. The old rdft2->rdft reduction is now in rdft/rdft2-rdft.c and still does way too much. M ./TODO -3 M ./rdft/Makefile.am -1 +1 M ./rdft/buffered.c -1 +1 M ./rdft/buffered2.c -190 +200 M ./rdft/conf.c +1 M ./rdft/problem2.c -5 +6 M ./rdft/rdft.h -1 +2 M ./rdft/rdft2-inplace-strides.c -5 +5 A ./rdft/rdft2-rdft.c M ./rdft/rdft2-strides.c -5 +5 M ./rdft/vrank-geq1-rdft2.c -11 +10 Tue Dec 26 14:03:27 EST 2006 athena@fftw.org * Buffer the input in hc2r problems, as opposed to the output. M ./rdft/buffered.c -34 +104 M ./rdft/buffered2.c -1 Tue Dec 26 10:02:59 EST 2006 athena@fftw.org * streamlined buffered solvers M ./dft/buffered.c -66 +14 M ./kernel/buffered.c -4 +25 M ./kernel/ifftw.h -1 +4 M ./kernel/primes.c +11 M ./rdft/buffered.c -63 +13 M ./rdft/buffered2.c -54 +7 Mon Dec 25 16:08:22 EST 2006 athena@fftw.org * c++ compatibility M ./libbench2/verify-r2r.c -2 +2 Sun Dec 24 20:27:23 EST 2006 athena@fftw.org * Gratuitous renaming of directories and files since the old naming was becoming too inconsistent for my taste. ./dft/codelets -> ./dft/scalar ./dft/scalar/standard -> ./dft/scalar/codelets ./rdft/codelets -> ./rdft/scalar M ./Makefile.am -14 +15 M ./configure.ac -6 +6 M ./dft/Makefile.am -1 +1 M ./dft/scalar/Makefile.am -3 +3 M ./dft/scalar/codelets/Makefile.am -3 +3 M ./dft/simd/codelets/Makefile.am -2 +2 M ./doc/fftw3.texi -4 +4 M ./rdft/Makefile.am -1 +1 M ./rdft/scalar/Makefile.am -2 +2 M ./rdft/scalar/r2cb/Makefile.am -3 +3 M ./rdft/scalar/r2cf/Makefile.am -3 +3 M ./rdft/scalar/r2r/Makefile.am -3 +3 Sun Dec 24 20:11:50 EST 2006 athena@fftw.org * Fixed another dftw bug (sigh) M ./threads/ct.c -7 +3 Sun Dec 24 11:48:01 EST 2006 athena@fftw.org * Removed debugging leftovers. M ./dft/dftw-genericbuf.c -2 +2 Sun Dec 24 11:34:17 EST 2006 athena@fftw.org * Moved dftw-generic* to new dftw protocol. M ./dft/dftw-generic.c -30 +24 M ./dft/dftw-genericbuf.c -13 +13 Sun Dec 24 09:37:19 EST 2006 athena@fftw.org * Oops M ./tests/hook.c -1 +1 Sun Dec 24 09:31:46 EST 2006 athena@fftw.org * Fixed wrong verification of rank-1 rdft2 M ./tests/hook.c -4 +6 Sat Dec 23 20:11:29 EST 2006 athena@fftw.org * minor tweaks M ./dft/dftw-direct.c -2 +4 Sat Dec 23 19:50:03 EST 2006 athena@fftw.org * Removed obsolete items. M ./TODO -13 Sat Dec 23 17:56:37 EST 2006 athena@fftw.org * Modified the problem_dftw invocation protocol. apply() now requires pointers to the beginning of the full array. Each thread processes a slice mb <= m < me. This protocol is consistent with the one used in hc2hc, where there is no other choice. M ./dft/dftw-direct.c -27 +31 M ./dft/dftw-directsq.c -1 +3 M ./threads/ct.c -7 +2 Sat Dec 23 16:18:25 EST 2006 athena@fftw.org * typo M ./simd/simd-altivec.h -1 +1 Sat Dec 23 16:06:56 EST 2006 athena@fftw.org * changed hc2hc twiddle storage to be the same as hc2c M ./genfft/gen_hc2hc.ml -2 +3 M ./rdft/hc2hc-direct.c -14 +11 Sat Dec 23 15:16:36 EST 2006 athena@fftw.org * Allowed extra_iter in dftw-direct. Rationalized twiddle factors in hc2c. M ./dft/dftw-direct.c -10 +38 M ./dft/simd/Makefile.am -1 +1 M ./dft/simd/codelets/Makefile.am -1 +12 M ./dft/simd/t.c -2 +45 A ./dft/simd/t1bu.h A ./dft/simd/t1fu.h M ./genfft/gen_hc2c.ml -2 +3 M ./genfft/gen_hc2cdft.ml -2 +3 M ./genfft/gen_hc2cdft_c.ml -1 +1 M ./genfft/gen_hc2hc.ml -1 +1 M ./genfft/gen_twiddle.ml -1 +1 M ./genfft/gen_twiddle_c.ml -1 +1 M ./genfft/gen_twidsq.ml -1 +1 M ./genfft/gen_twidsq_c.ml -1 +1 M ./genfft/twiddle.ml -12 +10 M ./genfft/twiddle.mli -1 +1 M ./kernel/twiddle.c -4 +8 M ./rdft/ct-hc2c-direct.c -4 +7 M ./rdft/simd/hc2cbv.h -2 +2 M ./rdft/simd/hc2cfv.h -2 +2 M ./simd/simd-altivec.h -12 +9 M ./simd/simd-sse.h -12 +9 M ./simd/simd-sse2.h -11 +6 Sat Dec 23 10:37:11 EST 2006 athena@fftw.org * Implemented unmentionable hack to use 4-way SIMD with an odd number of iterations. M ./dft/direct.c -9 +34 M ./rdft/ct-hc2c-direct.c -43 +88 M ./simd/simd-altivec.h -1 +3 M ./simd/simd-sse.h -1 +3 Fri Dec 22 22:13:30 EST 2006 athena@fftw.org * altivec support for new codelets M ./simd/simd-altivec.h -7 +28 Fri Dec 22 19:09:15 EST 2006 athena@fftw.org * fixed incorrect computation of W M ./genfft/gen_hc2cdft_c.ml -1 +1 Fri Dec 22 18:51:22 EST 2006 athena@fftw.org * Implemented 4-way simd hc2cdftv Also eliminated the twiddle_shift hack. A zillion changes dictated by this choice, which was in turn necessary for the hc2cdftv thing to work. M ./Makefile.am -13 +8 M ./configure.ac -3 M ./dft/codelet-dft.h -9 +9 M ./dft/codelets/t.c -4 +4 M ./dft/dftw-direct.c -35 +30 M ./dft/dftw-directsq.c -7 +4 M ./dft/simd/q1b.c -6 +6 M ./dft/simd/q1f.c -6 +6 M ./dft/simd/t.c -14 +14 M ./dft/simd/t3b.h +1 M ./dft/simd/t3f.h +1 M ./dft/simd/ts.c -6 +6 M ./dft/simd/ts.h +1 M ./genfft/Makefile.am -17 +10 R ./genfft/gen_conv.ml M ./genfft/gen_hc2c.ml -25 +19 M ./genfft/gen_hc2cdft.ml -19 +19 M ./genfft/gen_hc2cdft_c.ml -20 +24 M ./genfft/gen_hc2hc.ml -23 +17 M ./genfft/gen_mdct.ml -9 +9 M ./genfft/gen_notw.ml -8 +8 M ./genfft/gen_notw_c.ml -6 +6 M ./genfft/gen_r2cb.ml -9 +9 M ./genfft/gen_r2cf.ml -9 +9 M ./genfft/gen_r2r.ml -6 +6 M ./genfft/gen_twiddle.ml -37 +39 M ./genfft/gen_twiddle_c.ml -38 +42 M ./genfft/gen_twidsq.ml -42 +44 M ./genfft/gen_twidsq_c.ml -44 +47 M ./genfft/genutil.ml -6 +6 M ./genfft/simd.ml -6 +7 M ./genfft/simd.mli -2 +1 M ./genfft/variable.ml -7 +12 M ./genfft/variable.mli -1 +3 M ./kernel/ifftw.h -7 M ./kernel/twiddle.c -14 M ./rdft/codelet-rdft.h -5 +5 M ./rdft/codelets/hc2c.c -2 +2 M ./rdft/conf.c -1 +1 M ./rdft/ct-hc2c-direct.c -47 +36 M ./rdft/ct-hc2c.c -1 +3 M ./rdft/hc2hc-direct.c -39 +31 M ./rdft/simd/codelets/Makefile.am -2 +2 M ./rdft/simd/hc2cbv.c -7 +8 M ./rdft/simd/hc2cbv.h -4 +3 M ./rdft/simd/hc2cfv.c -7 +8 M ./rdft/simd/hc2cfv.h -4 +3 M ./simd/simd-sse.h +4 M ./simd/simd-sse2.h -3 +5 M ./simd/simd.h -2 M ./tests/Makefile.am +9 Fri Dec 22 08:45:46 EST 2006 athena@fftw.org * Fixed verification of rdft2 problems with new format. M ./tests/hook.c -1 +5 Fri Dec 22 00:05:59 EST 2006 athena@fftw.org * Added file A ./rdft/simd/Makefile.am Fri Dec 22 00:02:50 EST 2006 athena@fftw.org * Hmm, previous commit did not work A ./rdft/simd/ A ./rdft/simd/codelets/ A ./rdft/simd/codelets/Makefile.am A ./rdft/simd/hc2hcv.h ./rdft/simd/hc2hcv.h -> ./rdft/simd/hc2cfv.h A ./rdft/simd/hc2cbv.h A ./rdft/simd/hc2cv.c ./rdft/simd/hc2cv.c -> ./rdft/simd/hc2cfv.c M ./rdft/simd/codelets/Makefile.am +51 A ./rdft/simd/hc2cbv.c M ./rdft/simd/hc2cbv.h +29 M ./rdft/simd/hc2cfv.c +41 M ./rdft/simd/hc2cfv.h +29 Thu Dec 21 23:58:33 EST 2006 athena@fftw.org * Added SIMD r2cdft codelets. A ./genfft/gen_hc2cdft_c.ml M ./Makefile.am -14 +22 M ./configure.ac +5 M ./dft/simd/n1b.h -1 M ./dft/simd/n1f.h -1 M ./dft/simd/n2b.h -1 M ./dft/simd/n2f.h -1 M ./dft/simd/n2s.h -1 M ./genfft/Makefile.am -8 +14 M ./genfft/algsimp.ml -2 +12 M ./genfft/c.ml +1 M ./genfft/expr.ml -1 +3 M ./genfft/expr.mli -1 +1 M ./genfft/gen_hc2cdft_c.ml +217 M ./genfft/simd.ml -2 +9 M ./genfft/to_alist.ml +1 M ./kernel/ifftw.h +6 M ./rdft/Makefile.am -1 +1 M ./rdft/codelet-rdft.h +5 M ./rdft/codelets/hc2c.c -2 +12 M ./rdft/conf.c +4 M ./rdft/ct-hc2c-direct.c -5 +60 M ./rdft/hc2hc-direct.c -5 M ./rdft/hc2hc.h -2 M ./rdft/khc2hc.c -1 M ./simd/nonportable/sse.c -9 +4 M ./simd/nonportable/sse2.c -9 +4 M ./simd/simd-sse.h -4 +4 M ./simd/simd-sse2.h -4 +7 M ./simd/sse.c -2 +2 M ./simd/sse2.c -2 +2 M ./support/Makefile.codelets +1 Thu Dec 21 21:19:21 EST 2006 athena@fftw.org * Bug in buffering, grrr... M ./rdft/ct-hc2c-direct.c -1 +1 Thu Dec 21 20:58:14 EST 2006 athena@fftw.org * Oops, memory leak. M ./rdft/hc2hc-direct.c +1 Thu Dec 21 17:12:31 EST 2006 athena@fftw.org * minor changes, cleanup. M ./rdft/ct-hc2c.c -2 M ./rdft/hc2hc.c -16 +10 M ./threads/hc2hc.c -17 +12 Wed Dec 20 22:09:28 EST 2006 athena@fftw.org * Unified hc2hc-direct, hc2hc-directbuf. Cleanup. M ./rdft/Makefile.am -4 +3 M ./rdft/ct-hc2c-direct.c -8 +8 R ./rdft/hc2hc-common.c M ./rdft/hc2hc-direct.c -44 +153 R ./rdft/hc2hc-directbuf.c M ./rdft/hc2hc.h -4 Wed Dec 20 17:55:56 EST 2006 athena@fftw.org * removed obsolete rdft2-radix2 M ./rdft/Makefile.am -1 +1 M ./rdft/conf.c -1 M ./rdft/rdft.h -1 R ./rdft/rdft2-radix2.c Wed Dec 20 17:51:20 EST 2006 athena@fftw.org * implemented reduction rdft2->dft A ./genfft/gen_hc2cdft.ml M ./genfft/Makefile.am -9 +16 M ./genfft/gen_hc2c.ml -1 +1 M ./genfft/gen_hc2cdft.ml +207 M ./rdft/codelet-rdft.h -2 +7 M ./rdft/codelets/r2cb/Makefile.am -1 +12 M ./rdft/codelets/r2cf/Makefile.am -1 +12 M ./rdft/ct-hc2c-direct.c -5 +8 M ./rdft/ct-hc2c.c -24 +95 M ./rdft/ct-hc2c.h -2 +6 M ./rdft/khc2c.c -2 +3 M ./support/Makefile.codelets +1 Wed Dec 20 09:29:39 EST 2006 athena@fftw.org * Implemented buffered direct-r2c, direct-hc2c. Also, removed some old cruft: * okp() functions were never used and a pain to maintain---now they are gone. * ``m'' in hc2hc and hc2c codelets is now the number of iterations, not the ``logical'' m. M ./genfft/gen_hc2c.ml -5 +3 M ./genfft/gen_hc2hc.ml -9 +9 M ./genfft/gen_r2cb.ml -4 +1 M ./genfft/gen_r2cf.ml -4 +1 M ./genfft/gen_r2r.ml -5 +1 M ./rdft/codelet-rdft.h -30 +4 M ./rdft/codelets/hc2c.c -14 +2 M ./rdft/codelets/hfb.c -14 +2 M ./rdft/codelets/r2c.c -18 +4 M ./rdft/codelets/r2r.c -15 +1 M ./rdft/ct-hc2c-direct.c -37 +127 M ./rdft/direct-r2c.c -41 +189 M ./rdft/direct-r2r.c -15 +6 M ./rdft/direct2.c -20 +11 M ./rdft/hc2hc-direct.c -23 +9 M ./rdft/hc2hc-directbuf.c -79 +42 M ./rdft/kr2c.c +1 M ./rdft/rdft.h +1 Tue Dec 19 17:07:04 EST 2006 stevenj@fftw.org * added memcpy-loop rank0 solver (it makes a 5-20% difference for transposes of large tuples) M ./rdft/rank0.c +33 Tue Dec 19 16:15:54 EST 2006 stevenj@fftw.org * new variable to disable libbench2's problem allocation during speed benchmarking (to benchmark MPI transforms where the array does not fit into the memory of a single process) M ./libbench2/bench-user.h +2 M ./libbench2/speed.c -3 +8 M ./mpi/mpi-bench.c -5 +15 Tue Dec 19 14:55:08 EST 2006 stevenj@fftw.org * allow transpose-inplace to use input as scratch for DESTROY_INPUT plans (to avoid non-square in-place transpositions) ... on supersgj, the planner often prefers transpose-inplace to transpose-alltoall in this case (apparently MPI_Alltoall in LAM MPI isn't that great) M ./mpi/transpose-inplace.c -52 +99 Tue Dec 19 17:07:14 EST 2006 athena@fftw.org * For some reason HB2 codelets were not generated. M ./rdft/codelets/r2cb/Makefile.am -2 +2 M ./rdft/codelets/r2cf/Makefile.am -1 +1 Tue Dec 19 15:12:39 EST 2006 athena@fftw.org * split rdft/direct.c into direct-r2r and direct-r2c, since the file was getting out of control. A ./rdft/direct-r2c.c A ./rdft/direct-r2r.c R ./rdft/direct.c M ./rdft/Makefile.am -4 +5 M ./rdft/direct-r2c.c +195 M ./rdft/direct-r2r.c +154 Tue Dec 19 02:59:35 EST 2006 stevenj@fftw.org * added dft-rank1 solver - MPI now supports 1d complex DFTs! M ./mpi/Makefile.am -2 +2 M ./mpi/api.c -9 +54 A ./mpi/choose-radix.c M ./mpi/conf.c +1 M ./mpi/dft-problem.c -2 +2 A ./mpi/dft-rank1.c M ./mpi/fftw3-mpi.h -2 +8 M ./mpi/ifftw-mpi.h +4 M ./mpi/mpi-bench.c -25 +24 M ./mpi/mpi-dft.h -4 +5 M ./mpi/mpi-transpose.h -3 +3 M ./mpi/transpose-problem.c -1 +1 M ./tests/bench.c -8 +8 M ./tests/fftw-bench.h -1 +1 Tue Dec 19 01:27:20 EST 2006 stevenj@fftw.org * fftw_flops must call cost_hook directly; iestimate_cost always uses COST_MAX M ./api/flops.c -1 +7 M ./kernel/ifftw.h -2 +2 M ./kernel/planner.c -4 +3 Tue Dec 19 01:16:54 EST 2006 stevenj@fftw.org * fftw_flops and fftw_estimate_cost must now be called from every process, to prevent deadlocks in the MPI code (since they sum/max the cost over all processes) M ./tests/fftw-bench.c -7 +11 Tue Dec 19 00:55:34 EST 2006 stevenj@fftw.org * whoops, typo in assert M ./mpi/block.c -1 +1 Tue Dec 19 00:51:07 EST 2006 stevenj@fftw.org * remove multiplication by FFT_SIGN M ./kernel/trig.c -2 +7 Mon Dec 18 22:51:45 EST 2006 stevenj@fftw.org * need to synchronize ESTIMATE costs in MPI planner, and sum ESTIMATE costs for flop reporting: generalize measure_hook to cost_hook(..., {COST_SUM, COST_MAX}) M ./api/flops.c -1 +1 M ./kernel/ifftw.h -3 +6 M ./kernel/planner.c -8 +12 M ./mpi/api.c -6 +7 Mon Dec 18 15:36:15 EST 2006 stevenj@fftw.org * previous patch slowed down transpose-alltoall when TRANSPOSED_IN and DESTROY_INPUT; now allow planner to choose old behavior in this case M ./mpi/transpose-alltoall.c -11 +31 Mon Dec 18 15:15:39 EST 2006 stevenj@fftw.org * transpose-alltoall doesn't require input to be destroyed if TRANSPOSED_IN is set M ./mpi/transpose-alltoall.c -26 +41 Mon Dec 18 17:41:25 EST 2006 athena@fftw.org * Added t2-style hc2c codelets, fixed typos. M ./api/plan-guru-dft-c2r.h -1 +1 M ./api/plan-guru-dft-r2c.h -1 +1 M ./api/plan-guru-dft.h -2 +2 M ./rdft/codelets/r2cb/Makefile.am -1 +9 M ./rdft/codelets/r2cf/Makefile.am -1 +9 Mon Dec 18 16:02:42 EST 2006 athena@fftw.org * Renamed certain variables to avoid calling an output stride `is'. M ./rdft/codelet-rdft.h -1 +1 M ./rdft/direct.c -17 +17 M ./rdft/direct2.c -7 +7 Mon Dec 18 15:54:43 EST 2006 athena@fftw.org * Oops---wrong stride M ./rdft/ct-hc2c.c -1 +1 Mon Dec 18 14:59:16 EST 2006 athena@fftw.org * Oops, forgot to add file A ./rdft/kr2c.c Mon Dec 18 14:17:02 EST 2006 athena@fftw.org * Renamed r2hc/hc2r codelets to r2c After the recent changes, r2hc/hc2r codelets became rdft2 problems, so I renamed them accordingly to r2cf/r2cb. Codelet parameters are now a real array and a complex array, instead of an input array and an output array, and forward and backward codelets have the same type, which removes some clutter from the rdft code. ./genfft/gen_hc2r.ml -> ./genfft/gen_r2cb.ml ./genfft/gen_r2hc.ml -> ./genfft/gen_r2cf.ml ./rdft/codelets/hc2r -> ./rdft/codelets/r2cb ./rdft/codelets/r2hc -> ./rdft/codelets/r2cf A ./rdft/codelets/r2c.c A ./rdft/codelets/r2cb.h A ./rdft/codelets/r2cbIII.h A ./rdft/codelets/r2cf.h A ./rdft/codelets/r2cfII.h M ./Makefile.am -2 +2 M ./configure.ac -2 +2 M ./doc/fftw3.texi -1 +1 M ./genfft/Makefile.am -9 +9 M ./genfft/gen_r2cb.ml -58 +51 M ./genfft/gen_r2cf.ml -58 +51 M ./rdft/Makefile.am -3 +3 M ./rdft/codelet-rdft.h -59 +16 M ./rdft/codelets/Makefile.am -3 +3 R ./rdft/codelets/hc2r.c R ./rdft/codelets/hc2r.h R ./rdft/codelets/hc2rIII.h M ./rdft/codelets/r2c.c +51 M ./rdft/codelets/r2cb/Makefile.am -21 +21 M ./rdft/codelets/r2cb.h +23 M ./rdft/codelets/r2cbIII.h +23 M ./rdft/codelets/r2cf/Makefile.am -20 +20 M ./rdft/codelets/r2cf.h +23 M ./rdft/codelets/r2cfII.h +23 R ./rdft/codelets/r2hc.c R ./rdft/codelets/r2hc.h R ./rdft/codelets/r2hcII.h M ./rdft/conf.c -2 +2 M ./rdft/ct-hc2c.c -2 +2 M ./rdft/direct.c -46 +28 M ./rdft/direct2.c -72 +23 R ./rdft/khc2r.c R ./rdft/kr2hc.c M ./rdft/rdft.h -4 +2 M ./support/Makefile.codelets -2 +2 Mon Dec 18 10:48:07 EST 2006 athena@fftw.org * Implemented backward radix-2k rdft2. A ./rdft/codelets/hc2cf.h A ./rdft/codelets/hc2cb.h M ./genfft/gen_hc2c.ml -26 +30 M ./rdft/codelets/Makefile.am -1 +1 M ./rdft/codelets/hc2c.c -1 +5 R ./rdft/codelets/hc2c.h M ./rdft/codelets/hc2cb.h +23 M ./rdft/codelets/hc2cf.h +23 M ./rdft/codelets/hc2r/Makefile.am -1 +11 M ./rdft/codelets/r2hc/Makefile.am -7 +7 M ./rdft/ct-hc2c.c -3 +25 M ./rdft/rdft2-radix2.c -72 +1 Mon Dec 18 13:56:09 EST 2006 stevenj@fftw.org * move extract_reim into kernel, since it is used by internal MPI stuff and not just in the API code any more ./api/extract-reim.c -> ./kernel/extract-reim.c M ./api/Makefile.am -1 +1 M ./api/api.h -1 +1 M ./api/plan-many-dft-c2r.c -1 +1 M ./api/plan-many-dft-r2c.c -1 +1 M ./api/plan-many-dft.c -2 +2 M ./kernel/Makefile.am -5 +6 M ./kernel/extract-reim.c -6 +6 M ./kernel/ifftw.h +1 M ./mpi/dft-rank-geq2.c -3 M ./mpi/dft-rank1-bigvec.c -3 M ./mpi/dft-serial.c -3 Mon Dec 18 08:40:14 EST 2006 athena@fftw.org * Do not check r1==cr unless rnk>0 M ./rdft/problem2.c -1 +5 Sun Dec 17 21:03:50 EST 2006 athena@fftw.org * Implemented radix-2k RDFT2, forward only for now M ./api/execute-dft-c2r.c -1 +2 M ./api/execute-dft-r2c.c -1 +2 M ./api/execute-split-dft-c2r.c -1 +2 M ./api/execute-split-dft-r2c.c -1 +2 M ./api/f77funcs.h -4 +8 M ./api/plan-guru-dft-c2r.h -8 +8 M ./api/plan-guru-dft-r2c.h -9 +9 M ./api/plan-guru-split-dft-c2r.h -6 +6 M ./api/plan-guru-split-dft-r2c.h -9 +9 M ./api/plan-many-dft-c2r.c -3 +3 M ./api/plan-many-dft-r2c.c -1 +1 M ./dft/problem.c -3 +5 M ./genfft/Makefile.am -5 +11 A ./genfft/gen_hc2c.ml M ./genfft/gen_hc2hc.ml -30 +30 M ./genfft/gen_hc2r.ml -7 +16 M ./genfft/gen_r2hc.ml -8 +18 M ./rdft/Makefile.am -1 +2 M ./rdft/buffered2.c -31 +45 M ./rdft/codelet-rdft.h -11 +41 M ./rdft/codelets/Makefile.am -1 +1 A ./rdft/codelets/hc2c.c A ./rdft/codelets/hc2c.h M ./rdft/codelets/hc2r.c -4 +5 M ./rdft/codelets/hfb.c -4 +4 M ./rdft/codelets/r2hc.c -4 +5 M ./rdft/codelets/r2hc/Makefile.am -1 +11 A ./rdft/ct-hc2c-direct.c A ./rdft/ct-hc2c.c A ./rdft/ct-hc2c.h M ./rdft/direct.c -16 +21 M ./rdft/direct2.c -15 +32 M ./rdft/hc2hc-common.c -2 +2 M ./rdft/hc2hc-direct.c -4 +4 M ./rdft/hc2hc-directbuf.c -8 +8 M ./rdft/hc2hc-generic.c -4 +3 M ./rdft/hc2hc.c -4 +4 A ./rdft/khc2c.c M ./rdft/nop2.c -7 +10 M ./rdft/problem.c -2 +1 M ./rdft/problem2.c -38 +117 M ./rdft/rank-geq2-rdft2.c -13 +17 M ./rdft/rank0-rdft2.c -35 +44 M ./rdft/rdft.h -8 +27 M ./rdft/rdft2-inplace-strides.c -2 +6 M ./rdft/rdft2-radix2.c -131 +37 M ./rdft/solve2.c -1 +3 M ./rdft/vrank-geq1-rdft2.c -7 +7 M ./support/Makefile.codelets +1 M ./tests/hook.c -8 +23 M ./threads/vrank-geq1-rdft2.c -10 +10 Mon Dec 18 01:23:45 EST 2006 stevenj@fftw.org * separate TRANSPOSED/SCRAMBLED flags internally (this is required so that dft-rank1-bigvec and the future dft-rank1 won't have incompatible SCRAMBLED formats) M ./mpi/api.c -16 +2 M ./mpi/dft-rank-geq2.c -7 +7 M ./mpi/dft-rank1-bigvec.c -8 +9 M ./mpi/dft-serial.c -1 +1 M ./mpi/ifftw-mpi.h -1 +8 M ./mpi/mpi-dft.h -1 +2 M ./mpi/mpi-transpose.h -2 +2 M ./mpi/transpose-alltoall.c -3 +4 M ./mpi/transpose-inplace.c -6 +7 M ./mpi/transpose-radix2.c -5 +6 Mon Dec 18 01:02:27 EST 2006 stevenj@fftw.org * ops_add -> ops_add2 where possible, to shrink code M ./mpi/dft-rank-geq2.c -8 +4 M ./mpi/dft-rank1-bigvec.c -4 +2 M ./mpi/transpose-alltoall.c -6 +3 M ./mpi/transpose-inplace.c -8 +4 M ./mpi/transpose-radix2.c -10 +5 M ./rdft/vrank3-transpose.c -9 +5 Mon Dec 18 00:43:02 EST 2006 stevenj@fftw.org * added dft-rank1-bigvec solver (easy case for 1d parallel transforms) M ./mpi/Makefile.am -1 +1 M ./mpi/api.c -2 +14 M ./mpi/conf.c +1 A ./mpi/dft-rank1-bigvec.c M ./mpi/mpi-dft.h +1 Sun Dec 17 20:42:21 EST 2006 stevenj@fftw.org * rewrote MPI stuff to use dtensor data structure A dtensor is an ordered tuple of triplets (n, ib, ob) giving the size of a dimension (n) and its input and output block sizes of a distributed row-major multi-dimensional array. An MPI DFT (etc.) is now specified in terms of dtensors, which provide a much more flexible data layout. For example, we can now describe multidimensional block distributions, which are important if the number of processors is greater than the size of any given dimension. Currently, we only have solvers for 1d slab distributions, and this is all that is supported in the basic and advanced APIs. The guru API allows one to specify more general distributions, however, which will be useful when/if we have solvers for this case. We now also don't need a TRANSPOSED flag, at least internally, since TRANSPOSED multi-dimensional DFT plans just correspond to dtensors where the input and output block distributions are different. Other changes include the use of the XM(foo) macro for X(mpi_foo). M ./mpi/Makefile.am -1 +1 M ./mpi/any-true.c -1 +1 M ./mpi/api.c -129 +335 M ./mpi/block.c -32 +95 M ./mpi/conf.c -6 +6 M ./mpi/dft-problem.c -46 +33 M ./mpi/dft-rank-geq2.c -60 +72 M ./mpi/dft-serial.c -26 +20 M ./mpi/dft-solve.c -1 +1 A ./mpi/dtensor.c M ./mpi/fftw3-mpi.h -24 +38 M ./mpi/ifftw-mpi.h -22 +60 M ./mpi/mpi-bench.c -115 +289 M ./mpi/mpi-dft.h -12 +13 M ./mpi/mpi-transpose.h -10 +11 M ./mpi/transpose-alltoall.c -17 +16 M ./mpi/transpose-inplace.c -24 +21 M ./mpi/transpose-problem.c -19 +18 M ./mpi/transpose-radix2.c -13 +11 M ./mpi/transpose-solve.c -1 +1 M ./tests/bench.c +4 M ./tests/fftw-bench.c +2 M ./tests/fftw-bench.h +1 Fri Dec 15 16:01:23 EST 2006 athena@fftw.org * Distinguished mutexes from semaphores. The distinction is useful because the linux implementation of sem_post() in unnecessarily slow when semaphores are used for mutual exclusion. This change made spinlocks messier to implement, so I excised them. M ./tests/fftw-bench.c -8 M ./threads/threads.c -105 +40 Fri Dec 15 12:46:11 EST 2006 athena@fftw.org * Use posix semaphores where available. Paranoid declaration of all shared variables as ``volatile''. Paranoid initialization of all shared variables within locks. M ./threads/threads.c -36 +68 Wed Dec 13 20:12:13 EST 2006 Matteo Frigo * paranoia M ./threads/threads.c -2 +2 Tue Dec 12 17:28:13 EST 2006 stevenj@fftw.org * punt on detecting unsolvable rdft2 problems; make r==iio rdft2 problems unsolvable, since it doesn't look like we've consistently checked for this case and it's not clear why we would want to support it (it was also not documented in the manual) M ./doc/fftw3.texi -1 +2 M ./rdft/problem2.c -30 +2 Sat Dec 9 12:14:13 EST 2006 athena@fftw.org * Obey stupid const rules M ./api/api.h -2 +2 M ./api/apiplan.c -1 +1 M ./dft/dft.h -3 +3 M ./dft/indirect.c -4 +4 M ./dft/problem.c -9 +8 M ./kernel/ifftw.h -5 +5 M ./kernel/planner.c -2 +2 M ./kernel/problem.c -4 +4 M ./rdft/buffered2.c -1 +1 M ./rdft/dht-rader.c -1 +1 M ./rdft/indirect.c -4 +4 M ./rdft/problem.c -14 +13 M ./rdft/problem2.c -9 +8 M ./rdft/rank-geq2-rdft2.c -1 +1 M ./rdft/rdft-dht.c -1 +1 M ./rdft/rdft.h -13 +13 M ./rdft/rdft2-radix2.c -5 +5 Fri Dec 8 18:21:50 EST 2006 stevenj@fftw.org * added unsolvable check for rdft2 problem An in-place rdft2 problem is ill-formed if the real data, including the extra "padding" elements, do not coincide with the complex data. CHANGE: the new code considers all in-place split r2c and c2r problems to be ill-formed. Previously, these could be done, but only if the entire multi-dimensional array fit into the buffer, which is kind of stupid. I'm not sure it's worth it to even try to support the split in-place r2c case. M ./rdft/problem2.c -1 +33 Fri Dec 8 13:47:53 EST 2006 stevenj@fftw.org * check in-placeness after joining taints M ./dft/problem.c -9 +9 Fri Dec 8 13:43:44 EST 2006 athena@fftw.org * Grrr... paranoid-check was not testing in exhaustive mode M ./tests/Makefile.am -5 +5 Fri Dec 8 10:00:30 EST 2006 Matteo Frigo * Implemented PROBLEM_UNSOLVABLE. In-place DFT and RDFT problems with inconsistent I/O strides are now unsolvable, and we don't check for them any longer in solvers. While I was at it, declared all problem pointers to be ``const'' for extra safety. M ./api/api.h -2 +2 M ./api/apiplan.c -3 +5 M ./dft/buffered.c -4 +1 M ./dft/dft.h -4 +4 M ./dft/direct.c -8 +6 M ./dft/indirect.c -4 +4 M ./dft/problem.c -11 +15 M ./dft/vrank-geq1.c -3 M ./kernel/ifftw.h -6 +10 M ./kernel/planner.c -6 +6 M ./kernel/problem.c -1 +40 M ./rdft/buffered.c -4 +1 M ./rdft/buffered2.c -1 +1 M ./rdft/dht-rader.c -1 +1 M ./rdft/direct.c -4 +3 M ./rdft/indirect.c -4 +4 M ./rdft/problem.c -14 +17 M ./rdft/problem2.c -10 +10 M ./rdft/rank-geq2-rdft2.c -1 +1 M ./rdft/rdft-dht.c -1 +1 M ./rdft/rdft.h -13 +13 M ./rdft/rdft2-radix2.c -5 +5 M ./rdft/vrank-geq1.c -2 Thu Dec 7 20:13:46 EST 2006 Matteo Frigo * Avoid qsort'ing one element. M ./kernel/tensor7.c -6 +2 Thu Dec 7 18:25:47 EST 2006 Matteo Frigo * In-place vrank>=1 is now applicable only if the problem is really in-place. M ./dft/buffered.c -7 +12 M ./dft/direct.c -11 +11 M ./dft/vrank-geq1.c +3 M ./kernel/ifftw.h +2 M ./kernel/tensor7.c +41 M ./rdft/buffered.c -7 +12 M ./rdft/direct.c -7 +5 M ./rdft/vrank-geq1.c +2 Tue Dec 5 12:21:38 EST 2006 athena@fftw.org * unused variable M ./kernel/align.c +1 Sun Dec 3 19:16:33 EST 2006 Matteo Frigo * Removed CVS $Id$ everywhere, since darcs does not update them. M ./api/fftw3.h -1 M ./api/version.c -1 M ./commercialize.sh +1 M ./dft/buffered.c -1 M ./dft/codelet-dft.h -1 M ./dft/codelets/standard/Makefile.am +1 M ./dft/conf.c -1 M ./dft/ct.c -1 M ./dft/dft.h -1 M ./dft/dftw-direct.c -1 M ./dft/dftw-directsq.c -1 M ./dft/direct.c -1 M ./dft/indirect.c -1 M ./dft/kdft-dif.c -1 M ./dft/kdft-difsq.c -1 M ./dft/kdft-dit.c -1 M ./dft/kdft.c -1 M ./dft/nop.c -1 M ./dft/plan.c -1 M ./dft/problem.c -1 M ./dft/rank-geq2.c -1 M ./dft/solve.c -1 M ./dft/vrank-geq1.c -1 M ./dft/zero.c -1 M ./doc/fftw3.texi -1 M ./genfft/algsimp.ml -2 M ./genfft/algsimp.mli -2 M ./genfft/annotate.ml -2 M ./genfft/annotate.mli -1 M ./genfft/assoctable.ml -1 M ./genfft/assoctable.mli -1 M ./genfft/c.ml -1 M ./genfft/c.mli -1 M ./genfft/complex.ml -1 M ./genfft/complex.mli -1 M ./genfft/conv.mli -1 M ./genfft/dag.ml -1 M ./genfft/dag.mli -1 M ./genfft/expr.ml -1 M ./genfft/expr.mli -1 M ./genfft/fft.ml -2 M ./genfft/fft.mli -2 M ./genfft/gen_conv.ml -3 +1 M ./genfft/gen_hc2hc.ml -3 +1 M ./genfft/gen_hc2r.ml -3 +1 M ./genfft/gen_mdct.ml -3 +1 M ./genfft/gen_notw.ml -3 +1 M ./genfft/gen_notw_c.ml -3 +1 M ./genfft/gen_r2hc.ml -3 +1 M ./genfft/gen_r2r.ml -3 +1 M ./genfft/gen_twiddle.ml -3 +1 M ./genfft/gen_twiddle_c.ml -3 +1 M ./genfft/gen_twidsq.ml -3 +1 M ./genfft/gen_twidsq_c.ml -3 +1 M ./genfft/genutil.ml -8 +1 M ./genfft/littlesimp.ml -1 M ./genfft/littlesimp.mli -1 M ./genfft/magic.ml -1 M ./genfft/monads.ml -1 M ./genfft/number.ml -1 M ./genfft/number.mli -1 M ./genfft/oracle.ml -1 M ./genfft/oracle.mli -1 M ./genfft/schedule.ml -1 M ./genfft/schedule.mli -1 M ./genfft/simd.ml -1 M ./genfft/simd.mli -1 M ./genfft/simdmagic.ml -1 M ./genfft/to_alist.ml -1 M ./genfft/to_alist.mli -1 M ./genfft/trig.ml -1 M ./genfft/trig.mli -1 M ./genfft/twiddle.ml -1 M ./genfft/twiddle.mli -1 M ./genfft/unique.ml -1 M ./genfft/unique.mli -1 M ./genfft/util.ml -1 M ./genfft/util.mli -1 M ./genfft/variable.ml -1 M ./genfft/variable.mli -1 M ./kernel/align.c -1 M ./kernel/alloc.c -1 M ./kernel/assert.c -1 M ./kernel/awake.c -1 M ./kernel/cycle.h -1 M ./kernel/debug.c -1 M ./kernel/iabs.c -1 M ./kernel/ifftw.h -1 M ./kernel/kalloc.c -1 M ./kernel/minmax.c -1 M ./kernel/ops.c -1 M ./kernel/pickdim.c -1 M ./kernel/plan.c -1 M ./kernel/planner.c -1 M ./kernel/primes.c -1 M ./kernel/print.c -1 M ./kernel/problem.c -1 M ./kernel/scan.c -1 M ./kernel/solver.c -1 M ./kernel/solvtab.c -1 M ./kernel/stride.c -1 M ./kernel/tensor.c -1 M ./kernel/tensor1.c -1 M ./kernel/tensor2.c -1 M ./kernel/tensor4.c -1 M ./kernel/tensor5.c -1 M ./kernel/tensor7.c -1 M ./kernel/tensor8.c -1 M ./kernel/tensor9.c -1 M ./kernel/timer.c -1 M ./kernel/trig.c -1 M ./kernel/twiddle.c -1 M ./libbench/accopy-from.c -1 M ./libbench/accopy-to.c -1 M ./libbench/allocate.c -1 M ./libbench/bench-main.c -1 M ./libbench/bench-user.h -1 M ./libbench/bench.h -1 M ./libbench/can-do.c -1 M ./libbench/ccopy-from.c -1 M ./libbench/ccopy-to.c -1 M ./libbench/deallocate.c -1 M ./libbench/getopt-utils.c -1 M ./libbench/info.c -1 M ./libbench/main.c -1 M ./libbench/prime.c -1 M ./libbench/problem.c -1 M ./libbench/report.c -1 M ./libbench/speed.c -1 M ./libbench/timer.c -1 M ./libbench/verify.c -1 M ./libbench/zero.c -1 M ./libbench2/aligned-main.c -1 M ./libbench2/allocate.c -1 M ./libbench2/bench-main.c -1 M ./libbench2/bench-user.h -1 M ./libbench2/bench.h -1 M ./libbench2/can-do.c -1 M ./libbench2/dotens2.c -1 M ./libbench2/info.c -1 M ./libbench2/main.c -1 M ./libbench2/problem.c -1 M ./libbench2/report.c -1 M ./libbench2/speed.c -1 M ./libbench2/tensor.c -1 M ./libbench2/timer.c -1 M ./libbench2/useropt.c -1 M ./libbench2/verify-dft.c -1 M ./libbench2/verify-lib.c -1 M ./libbench2/verify-rdft2.c -1 M ./libbench2/verify.c -1 M ./libbench2/zero.c -1 M ./mpi/conf.c -1 M ./rdft/buffered.c -1 M ./rdft/buffered2.c -1 M ./rdft/codelets/hc2r/Makefile.am +1 M ./rdft/codelets/r2hc/Makefile.am +1 M ./rdft/conf.c -1 M ./rdft/dft-r2hc.c -1 M ./rdft/dht-r2hc.c -1 M ./rdft/direct.c -1 M ./rdft/direct2.c -1 M ./rdft/hc2hc-direct.c -1 M ./rdft/hc2hc-directbuf.c -1 M ./rdft/indirect.c -1 M ./rdft/khc2hc.c -1 M ./rdft/khc2r.c -1 M ./rdft/kr2hc.c -1 M ./rdft/kr2r.c -1 M ./rdft/nop.c -1 M ./rdft/nop2.c -1 M ./rdft/plan.c -1 M ./rdft/plan2.c -1 M ./rdft/problem.c -1 M ./rdft/problem2.c -1 M ./rdft/rank-geq2-rdft2.c -1 M ./rdft/rank-geq2.c -1 M ./rdft/rank0-rdft2.c -1 M ./rdft/rank0.c -1 M ./rdft/rdft-dht.c -1 M ./rdft/rdft2-inplace-strides.c -1 M ./rdft/rdft2-radix2.c -1 M ./rdft/rdft2-tensor-max-index.c -1 M ./rdft/solve.c -1 M ./rdft/solve2.c -1 M ./rdft/vrank-geq1-rdft2.c -1 M ./rdft/vrank-geq1.c -1 M ./rdft/vrank3-transpose.c -1 M ./reodft/conf.c -1 M ./reodft/redft00e-r2hc-pad.c -1 M ./reodft/redft00e-r2hc.c -1 M ./reodft/reodft00e-splitradix.c -1 M ./reodft/reodft010e-r2hc.c -1 M ./reodft/reodft11e-r2hc-odd.c -1 M ./reodft/reodft11e-r2hc.c -1 M ./reodft/reodft11e-radix2.c -1 M ./reodft/rodft00e-r2hc-pad.c -1 M ./reodft/rodft00e-r2hc.c -1 M ./simd/altivec.c -1 M ./simd/nonportable/sse.c -1 M ./simd/nonportable/sse2.c -1 M ./simd/sse.c -1 M ./simd/sse2.c -1 M ./simd/taint.c -1 M ./simd/x86-cpuid.h -1 M ./threads/conf.c -1 M ./threads/ct.c -1 M ./threads/dft-vrank-geq1.c -1 M ./threads/rdft-vrank-geq1.c -1 M ./threads/vrank-geq1-rdft2.c -1 Sun Dec 3 16:11:17 EST 2006 Matteo Frigo * generalized dftw to encompass q codelets. As a side effect, q codelets are now threaded. R ./dft/ctsq.c A ./dft/dftw-directsq.c M ./dft/Makefile.am -2 +2 M ./dft/ct.c -25 +55 M ./dft/ct.h -6 +9 M ./dft/dftw-direct.c -59 +72 M ./dft/dftw-directsq.c +163 M ./dft/dftw-generic.c -24 +32 M ./dft/dftw-genericbuf.c -30 +37 M ./dft/indirect-transpose.c -1 +1 M ./dft/kdft-difsq.c -1 +1 M ./libbench2/my-getopt.c -3 +3 M ./threads/ct.c -28 +53 Sat Nov 25 16:34:38 EST 2006 stevenj@fftw.org * add missing __declspec attribute to threads API functions when compiling for Windows (thanks to Robert O. Morris for the bug report) M ./api/f77api.c -29 +1 M ./api/x77.h +28 M ./threads/f77api.c +13 M ./threads/f77funcs.h -3 +3 Mon Nov 20 17:39:20 EST 2006 stevenj@fftw.org * add AC_SUBST to AX_OPENMP, thanks to Sebastien Maret for the suggestion M ./m4/ax_openmp.m4 -1 +2 Mon Oct 16 23:02:29 EDT 2006 stevenj@fftw.org * not gcc bug for MIPS (thanks to Jonathan Day) M ./doc/FAQ/fftw-faq.bfnn +4 Sat Sep 23 17:52:36 EDT 2006 stevenj@fftw.org * in maintainer/debug mode, don't modify CFLAGS if they were explicitly set (-pedantic seems to cause problems with LAM's mpicc, so I need a way to override) M ./configure.ac +2 Wed Nov 22 18:14:47 EST 2006 athena@fftw.org * Removed obsolete comment. M ./dft/dftw-generic.c -1 Sun Nov 19 11:21:44 EST 2006 athena@fftw.org * Use p->v when comparing TW_FULL fields. M ./kernel/twiddle.c -2 +1 Thu Nov 16 14:49:05 EST 2006 athena@fftw.org * removed useless definition M ./simd/simd.h -1 Mon Nov 13 09:18:32 EST 2006 athena@fftw.org * paranoid avoidance of integer overflows M ./libbench2/timer.c -3 +3 Mon Nov 13 09:00:11 EST 2006 athena@fftw.org * avoid potential overflows in cycle counters At the suggestion of Alex Cichowski, convert all ticks to double before operating on them, to avoid potential signed/unsigned confusion and integer overflow. M ./kernel/cycle.h -5 +8 Sun Nov 5 09:00:52 EST 2006 Matteo Frigo * Removed unused struct field M ./threads/threads.c -1 Sat Nov 4 09:43:13 EST 2006 Matteo Frigo * use pthread condition variables instead of semaphores Condition variables are more likely to be portable everywhere, and somehow they appear to introduce less overhead at least on my linux box. M ./threads/threads.c -6 +19 Tue Oct 31 20:45:24 EST 2006 Matteo Frigo * Imprecise help message. M ./libbench2/my-getopt.c -1 +1 Mon Oct 30 20:13:35 EST 2006 athena@fftw.org * Experimental implementation of spinlocks. This patch implements spinlocks via a semi-portable hack, and adds the -ospinlocks option to the bench program so that we can play with them. M ./tests/fftw-bench.c +8 M ./threads/threads.c -28 +104 Thu Oct 26 22:29:18 EDT 2006 athena@fftw.org * Updated manual for new openmp configure options. M ./doc/fftw3.texi -20 +17 Thu Oct 26 21:52:39 EDT 2006 athena@fftw.org * Added back openmp. Rationalized threads naming conventions: * threads explicitly managed by us are enabled by --enable-threads, predicated on HAVE_THREADS, etc. * openmp is enabled by --enable-openmp, predicated on HAVE_OPENMP, etc. * SMP denotes either THREADS or OPENMP. M ./Makefile.am -1 +1 M ./configure.ac -10 +27 M ./kernel/alloc.c -1 +1 M ./kernel/ifftw.h -1 +1 M ./mpi/Makefile.am -1 +1 M ./tests/Makefile.am -5 +5 M ./tests/fftw-bench.c -3 +7 M ./threads/Makefile.am -2 +2 M ./threads/api.c -1 +16 M ./threads/conf.c -2 +2 A ./threads/openmp.c M ./threads/threads.c -13 +19 M ./threads/threads.h -2 +2 M ./tools/Makefile.am -1 +1 M ./tools/fftw-wisdom.c -5 +9 Mon Oct 23 20:14:31 EDT 2006 athena@fftw.org * different thread protocols M ./threads/threads.c -80 +71 Sun Oct 22 14:49:32 EDT 2006 athena@fftw.org * fix memory leak Added pthread_attr_destroy to avoid memory leak. M ./threads/threads.c +1 Sun Oct 22 14:23:30 EDT 2006 athena@fftw.org * Experimental new pthread implementation that recycles threads. M ./threads/threads.c -528 +155 Tue Oct 24 23:28:10 EDT 2006 athena@fftw.org * switched buddies Switched order of buddies in rdft2 rank-geq2 for consistency with analogous dft and rdft solvers. Furthermore, this change reduces the MEASURE planning time for rank == 3. M ./rdft/rank-geq2-rdft2.c -4 +1 Sat Oct 21 10:10:00 EDT 2006 Matteo Frigo * Typo. M ./tools/fftw_wisdom.1.in -1 +1 Sat Oct 21 09:56:50 EDT 2006 Matteo Frigo * Out of place is the default. Thanks to Kirk Kern for pointing this out. M ./tests/README -2 +2 Fri Sep 29 01:36:11 EDT 2006 stevenj@fftw.org * rename "test" to "tst", since a user (Igor Levicki) reports that "test" is a reserved words in some x86 assemblers M ./simd/x86-cpuid.h -10 +10 Tue Sep 26 09:01:08 EDT 2006 Matteo Frigo * Stylistic change. M ./kernel/timer.c -2 +1 Tue Sep 26 08:45:37 EDT 2006 athena@fftw.org * Do not set tmin=1e10, since a large FFT may take longer than that. M ./kernel/timer.c -5 +4 Sat Sep 23 22:07:10 EDT 2006 athena@fftw.org * Disable certain gcc optimizations. When PRECOMPUTE_ARRAY_INDICES is #define'd, array indices have the form array[stride[k]] for compile-time constant k. Apparently new gcc's copy stride[k] onto the stack before the codelet loop, which is an idiotic optimization if ever there was one. This patch confuses gcc enough to prevent this optimization. M ./kernel/ifftw.h -3 +6 M ./kernel/stride.c -1 +1 Sat Sep 23 13:02:58 EDT 2006 stevenj@fftw.org * re-enable TOMS algorithm - it is the best for large vector lengths, since for such sizes the cache line is not an issue and the bookkeeping overhead is negligible M ./rdft/vrank3-transpose.c -18 +15 Thu Sep 21 15:40:15 EDT 2006 stevenj@fftw.org * add measure_hook so that MPI can synchronize timing measurements (otherwise different processors might end up with different MPI plans, yikes!) M ./kernel/ifftw.h +2 M ./kernel/planner.c +5 M ./mpi/api.c -2 +26 Tue Sep 19 21:26:19 EDT 2006 stevenj@fftw.org * added O(p log p) transpose algorithm (radix 2) M ./mpi/Makefile.am -1 +1 M ./mpi/conf.c +1 M ./mpi/mpi-transpose.h +1 A ./mpi/transpose-radix2.c Tue Sep 19 21:05:09 EDT 2006 stevenj@fftw.org * comments M ./mpi/testsched.c -8 +10 Tue Sep 19 19:54:58 EDT 2006 stevenj@fftw.org * whoops ./mpi/any_true.c -> ./mpi/any-true.c M ./mpi/Makefile.am -1 +1 Tue Sep 19 19:49:01 EDT 2006 stevenj@fftw.org * synchronize planning so that if one process fails to create a plan then all of them do. M ./mpi/Makefile.am -1 +1 A ./mpi/any_true.c M ./mpi/dft-rank-geq2.c -4 +4 M ./mpi/dft-serial.c -1 +1 M ./mpi/ifftw-mpi.h +3 M ./mpi/transpose-alltoall.c -5 +5 M ./mpi/transpose-inplace.c -5 +5 Tue Sep 19 18:17:38 EDT 2006 stevenj@fftw.org * call MPI_Alltoall instead of MPI_Alltoallv for equal-blocks case, in case MPI implementation has special optimizations for the common case of equal sizes M ./mpi/transpose-alltoall.c -6 +19 Tue Sep 19 12:07:35 EDT 2006 stevenj@fftw.org * whoops M ./tools/Makefile.am -1 +1 Tue Sep 19 02:20:06 EDT 2006 stevenj@fftw.org * typo in comment M ./mpi/testsched.c -3 +3 Tue Sep 19 01:58:55 EDT 2006 stevenj@fftw.org * more filename simplifications ./mpi/mpi-dft-rank-geq2.c -> ./mpi/dft-rank-geq2.c ./mpi/mpi-dft-serial.c -> ./mpi/dft-serial.c ./mpi/mpi-transpose-alltoall.c -> ./mpi/transpose-alltoall.c ./mpi/mpi-transpose-inplace.c -> ./mpi/transpose-inplace.c ./mpi/problem.c -> ./mpi/dft-problem.c ./mpi/solve.c -> ./mpi/dft-solve.c ./mpi/tproblem.c -> ./mpi/transpose-problem.c ./mpi/tsolve.c -> ./mpi/transpose-solve.c M ./mpi/Makefile.am -3 +3 Tue Sep 19 01:49:52 EDT 2006 stevenj@fftw.org * canonicalize file names (hyphens, not underscores) ./mpi/fftw3_mpi.h -> ./mpi/fftw3-mpi.h ./mpi/ifftw_mpi.h -> ./mpi/ifftw-mpi.h ./mpi/mpi_bench.c -> ./mpi/mpi-bench.c ./mpi/mpi_dft.h -> ./mpi/mpi-dft.h ./mpi/mpi_transpose.h -> ./mpi/mpi-transpose.h ./mpi/mpi_transpose_alltoall.c -> ./mpi/mpi-transpose-alltoall.c ./mpi/mpi_transpose_inplace.c -> ./mpi/mpi-transpose-inplace.c ./tests/fftw_bench.h -> ./tests/fftw-bench.h ./tests/fftw_bench_common.c -> ./tests/fftw-bench.c M ./mpi/Makefile.am -5 +5 M ./mpi/api.c -4 +4 M ./mpi/block.c -1 +1 M ./mpi/conf.c -2 +2 M ./mpi/mpi-bench.c -2 +2 M ./mpi/mpi-dft-rank-geq2.c -2 +2 M ./mpi/mpi-dft-serial.c -1 +1 M ./mpi/mpi-dft.h -1 +1 M ./mpi/mpi-transpose-alltoall.c -1 +1 M ./mpi/mpi-transpose-inplace.c -1 +1 M ./mpi/mpi-transpose.h -1 +1 M ./mpi/problem.c -1 +1 M ./mpi/solve.c -1 +1 M ./mpi/tproblem.c -1 +1 M ./mpi/tsolve.c -1 +1 M ./tests/Makefile.am -1 +1 M ./tests/bench.c -2 +2 M ./tests/fftw-bench.c -1 +1 Tue Sep 19 01:34:07 EDT 2006 stevenj@fftw.org * add mpi-dft-serial M ./mpi/Makefile.am -1 +1 M ./mpi/conf.c +1 M ./mpi/mpi-dft-rank-geq2.c -3 +9 A ./mpi/mpi-dft-serial.c M ./mpi/mpi_dft.h +1 M ./mpi/mpi_transpose_alltoall.c -1 +1 M ./mpi/mpi_transpose_inplace.c -1 +1 Tue Sep 19 01:21:47 EDT 2006 stevenj@fftw.org * silence warnings M ./dft/indirect-transpose.c -1 M ./libbench2/problem.c -2 Tue Sep 19 00:31:59 EDT 2006 stevenj@fftw.org * make "t" problem semantics match FFTW_MPI_TRANSPOSED M ./libbench2/problem.c -4 +3 M ./mpi/mpi_bench.c -8 +18 Mon Sep 18 23:50:43 EDT 2006 stevenj@fftw.org * whoops, fixed backwards mpi_dft M ./mpi/mpi-dft-rank-geq2.c -7 +7 Mon Sep 18 22:26:31 EDT 2006 stevenj@fftw.org * initial stab at rank-geq2 mpi-dft; seems to be mostly working M ./libbench2/bench-user.h -1 +1 M ./mpi/Makefile.am -1 +4 M ./mpi/api.c -26 +180 M ./mpi/conf.c +2 M ./mpi/fftw3_mpi.h -11 +51 M ./mpi/ifftw_mpi.h -4 +1 A ./mpi/mpi-dft-rank-geq2.c M ./mpi/mpi_bench.c -55 +166 M ./mpi/mpi_dft.h -2 +2 M ./mpi/mpi_transpose.h -6 +1 M ./mpi/mpi_transpose_alltoall.c -7 +10 M ./mpi/mpi_transpose_inplace.c -7 +8 M ./mpi/problem.c -2 +7 A ./mpi/solve.c M ./mpi/tproblem.c -2 +2 Sun Sep 17 13:41:32 EDT 2006 stevenj@fftw.org * support SCRAMBLED_OUT in alltoall transpose M ./kernel/ifftw.h +4 M ./kernel/tensor2.c +20 M ./mpi/ifftw_mpi.h -1 +16 M ./mpi/mpi_dft.h -1 +1 M ./mpi/mpi_transpose_alltoall.c -23 +48 M ./mpi/mpi_transpose_inplace.c -23 +2 M ./mpi/problem.c -2 +3 Sun Sep 17 12:34:30 EDT 2006 stevenj@fftw.org * skeleton of future support for block-cyclic M ./mpi/block.c -7 +27 M ./mpi/ifftw_mpi.h +1 M ./mpi/mpi_transpose_alltoall.c +2 M ./mpi/mpi_transpose_inplace.c -2 +5 Sun Sep 17 12:11:19 EDT 2006 stevenj@fftw.org * test program now checks scrambled in/out via -obflag=28/29 M ./mpi/api.c -1 +1 M ./mpi/fftw3_mpi.h -1 +1 M ./mpi/mpi_bench.c -13 +59 M ./mpi/problem.c -1 Sun Sep 17 11:58:36 EDT 2006 stevenj@fftw.org * added -obflag to make it easier to set high-order bits M ./tests/fftw_bench_common.c +1 Sun Sep 17 01:30:51 EDT 2006 stevenj@fftw.org * use proper child plans for 2nd transpose in transpose_alltoall; implement opcount in transpose_inplace M ./mpi/mpi_transpose_alltoall.c -47 +47 M ./mpi/mpi_transpose_inplace.c -1 +10 Sun Sep 17 01:08:01 EDT 2006 stevenj@fftw.org * fix in test program for transposes of vectors -- transpose routines seem to completely work now (except for scrambled in/out, which is untested) M ./mpi/mpi_bench.c -1 +1 Sun Sep 17 01:01:16 EDT 2006 stevenj@fftw.org * fixed bug in transpose_alltoall for unequal blocks M ./mpi/mpi_transpose_alltoall.c -2 +2 M ./mpi/mpi_transpose_inplace.c -2 +2 Sat Sep 16 15:29:46 EDT 2006 stevenj@fftw.org * correctly handle cld2rest M ./mpi/mpi_transpose_inplace.c -24 +28 Sat Sep 16 15:29:31 EDT 2006 stevenj@fftw.org * some debugging code and other fixes M ./mpi/mpi_bench.c -13 +34 Sat Sep 16 14:54:30 EDT 2006 stevenj@fftw.org * whoops, forgot to check in mpi_bench.c file A ./mpi/mpi_bench.c Sat Sep 16 14:54:02 EDT 2006 stevenj@fftw.org * added bench_exit routine so that it can be overridden (by MPI_Abort) if needed M ./libbench2/Makefile.am -5 +5 A ./libbench2/bench-exit.c M ./libbench2/bench-user.h +1 M ./libbench2/util.c -1 +1 M ./libbench2/verify-lib.c -1 +1 M ./libbench2/verify-r2r.c -1 +1 Sat Sep 16 14:52:56 EDT 2006 stevenj@fftw.org * bug fix in mpi_transpose_inplace for case where some processors are idle M ./mpi/mpi_transpose_inplace.c -3 +4 Fri Sep 15 18:47:13 EDT 2006 stevenj@fftw.org * allow vecloop for sz->rnk==0 in exceptional (SLOW) cases, e.g. it is necessary for loops of non-square transposes (otherwise e.g. ik1v5:200:200x10:20:1x20:1:10 planning fails) M ./rdft/vrank-geq1.c -2 +6 Thu Sep 14 23:36:48 EDT 2006 stevenj@fftw.org * first pass at working mpi_bench test program; transpose seems to work iff dimensions are divisible by #processors M ./libbench2/Makefile.am -1 +3 A ./libbench2/after-ccopy-from.c A ./libbench2/after-ccopy-to.c A ./libbench2/after-hccopy-from.c A ./libbench2/after-hccopy-to.c A ./libbench2/after-rcopy-from.c A ./libbench2/after-rcopy-to.c M ./libbench2/bench-main.c -4 +6 M ./libbench2/bench-user.h +10 M ./libbench2/verify-dft.c +2 M ./libbench2/verify-r2r.c +4 M ./libbench2/verify-rdft2.c +4 M ./mpi/api.c -2 +6 Wed Sep 13 17:28:07 EDT 2006 stevenj@fftw.org * whoops M ./mpi/mpi_transpose_inplace.c -2 +2 Tue Sep 12 22:27:03 EDT 2006 stevenj@fftw.org * initial stub for mpi_bench M ./Makefile.am -2 +2 M ./libbench2/bench-main.c +3 M ./libbench2/bench-user.h +1 M ./libbench2/verify-lib.c -1 +1 M ./mpi/Makefile.am -1 +14 M ./tests/bench.c +9 M ./tests/fftw_bench.h +1 M ./tests/fftw_bench_common.c +2 Tue Sep 12 21:54:31 EDT 2006 stevenj@fftw.org * do no output at all if verbose < 0 (for use with MPI, where we only want output from process 0) M ./libbench2/bench-main.c -1 +1 M ./libbench2/bench.h +1 M ./libbench2/ovtpvt.c -1 +15 M ./libbench2/useropt.c -1 +1 M ./libbench2/util.c -2 +1 M ./libbench2/verify-lib.c -5 +4 M ./libbench2/verify-r2r.c -4 +4 Tue Sep 12 21:39:15 EDT 2006 stevenj@fftw.org * whoops M ./tests/fftw_bench_common.c +3 Tue Sep 12 21:31:40 EDT 2006 stevenj@fftw.org * split bench.c into bench.c and fftw_bench_common.c so that we can re-use some of the code in the MPI test program M ./tests/Makefile.am -1 +1 M ./tests/bench.c -210 +6 A ./tests/fftw_bench.h A ./tests/fftw_bench_common.c M ./tools/Makefile.am -1 +2 Tue Sep 12 21:00:36 EDT 2006 stevenj@fftw.org * MPI stuff at least compiles now M ./configure.ac -9 +11 M ./kernel/ifftw.h +7 M ./mpi/api.c -3 +4 M ./mpi/fftw3_mpi.h +1 M ./mpi/mpi_transpose_alltoall.c -10 +12 M ./mpi/mpi_transpose_inplace.c -11 +14 M ./mpi/problem.c -5 +4 M ./mpi/tproblem.c -4 +3 Mon Sep 11 22:26:36 EDT 2006 stevenj@fftw.org * initial (nonfunctional) start at MPI support (similar to FFTW 2.x in spirit, but mostly rewritten) A ./mpi/ M ./Makefile.am -1 +1 M ./configure.ac +11 A ./m4/acx_mpi.m4 A ./mpi/Makefile.am A ./mpi/api.c A ./mpi/block.c A ./mpi/conf.c A ./mpi/fftw3_mpi.h A ./mpi/ifftw_mpi.h A ./mpi/mpi_dft.h A ./mpi/mpi_transpose.h A ./mpi/mpi_transpose_alltoall.c A ./mpi/mpi_transpose_inplace.c A ./mpi/problem.c A ./mpi/testsched.c A ./mpi/tproblem.c A ./mpi/tsolve.c Mon Sep 11 22:25:38 EDT 2006 stevenj@fftw.org * make X(plan_awake) work for NULL argument to reduce code size M ./kernel/plan.c -3 +5 M ./rdft/vrank3-transpose.c -3 +3 Fri Sep 15 23:47:08 EDT 2006 stevenj@fftw.org * -mt should go before -mthreads to avoid spurious warnings on HPUX (thanks to Peter O'Gorman for the bug report) M ./m4/acx_pthread.m4 -2 +3 Mon Sep 11 13:53:44 EDT 2006 stevenj@fftw.org * Fortran init_threads wrapper didn't return result; thanks to Markus Wetzstein for the bug report M ./threads/f77funcs.h -2 +2 Thu Sep 7 18:43:55 EDT 2006 stevenj@fftw.org * make sure wrappers are included even if Fortran compiler was not detected (unless --disable-fortran was specified explicitly) ... this was supposed to be done before, but the definition was in the wrong place, grr M ./api/f77api.c -21 +21 Thu Aug 31 19:33:29 EDT 2006 stevenj@fftw.org * I'm sick of answering this question about non-deterministic results M ./doc/FAQ/fftw-faq.bfnn +13 Tue Aug 22 21:27:29 EDT 2006 Matteo Frigo * Add --tag=CC flag to libtool. This change is consistent with the libtool invocation in the latest automake, and is required to compile with (some version of) xlc. M ./support/Makefile.codelets -2 +3 Mon Aug 21 21:40:36 EDT 2006 athena@fftw.org * avoid ``fma'' because it is defined in c99. M ./tests/bench.c -3 +3 Sun Aug 20 11:40:53 EDT 2006 Matteo Frigo * Obey -standalone flag. M ./genfft/gen_twiddle.ml -2 +4 Sat Aug 19 13:34:27 EDT 2006 Matteo Frigo * obey -standalone when generating simd codelets M ./genfft/gen_notw_c.ml -2 +1 M ./genfft/gen_twiddle_c.ml -2 +3 Sat Aug 19 13:33:43 EDT 2006 Matteo Frigo * removed obsolete athfft M ./genfft/Makefile.am -10 +9 R ./genfft/gen_athnotw.ml R ./genfft/gen_athtw.ml Thu Aug 17 21:50:50 EDT 2006 stevenj@fftw.org * updated citation to Proc. IEEE paper M ./doc/fftw3.texi -8 +5 Thu Aug 17 21:47:05 EDT 2006 stevenj@fftw.org * use darcs changes --summary to make nice changelog; emacs fill-region hack is obsolete M ./mkdist.sh -3 +1 Mon Aug 14 17:53:19 EDT 2006 athena@fftw.org * removed timer calibration Timer calibration seems not to work any longer on recent processors--- too much noise. I have remove it completely. M ./libbench2/Makefile.am -1 +1 M ./libbench2/bench.h -2 M ./libbench2/timer.c -70 +3 R ./libbench2/timer2.c Mon Aug 14 10:47:15 EDT 2006 Matteo Frigo * removed k7 Removed obsolete k7 support. M ./Makefile.am -10 +2 M ./NEWS +3 M ./TODO -5 M ./api/version.c -3 M ./bootstrap.sh -2 M ./commercialize.sh -8 +1 M ./configure.ac -13 M ./dft/Makefile.am -1 +1 M ./dft/codelet-dft.h -4 M ./dft/conf.c -3 R ./dft/k7/Makefile.am R ./dft/k7/codelets/Makefile.am R ./dft/k7/codelets/ R ./dft/k7/k7.c R ./dft/k7/ M ./doc/fftw3.texi -31 +22 R ./genfft-k7/Makefile.am R ./genfft-k7/algsimp.ml R ./genfft-k7/algsimp.mli R ./genfft-k7/assignmentsToVfpinstrs.ml R ./genfft-k7/assignmentsToVfpinstrs.mli R ./genfft-k7/assoctable.ml R ./genfft-k7/assoctable.mli R ./genfft-k7/balanceVfpinstrs.ml R ./genfft-k7/balanceVfpinstrs.mli R ./genfft-k7/complex.ml R ./genfft-k7/complex.mli R ./genfft-k7/expr.ml R ./genfft-k7/expr.mli R ./genfft-k7/fft.ml R ./genfft-k7/fft.mli R ./genfft-k7/genUtil.ml R ./genfft-k7/gen_notw.ml R ./genfft-k7/gen_twiddle.ml R ./genfft-k7/id.ml R ./genfft-k7/id.mli R ./genfft-k7/k7Basics.ml R ./genfft-k7/k7Basics.mli R ./genfft-k7/k7ExecutionModel.ml R ./genfft-k7/k7ExecutionModel.mli R ./genfft-k7/k7FlatInstructionScheduling.ml R ./genfft-k7/k7FlatInstructionScheduling.mli R ./genfft-k7/k7InstructionSchedulingBasics.ml R ./genfft-k7/k7InstructionSchedulingBasics.mli R ./genfft-k7/k7RegisterAllocationBasics.ml R ./genfft-k7/k7RegisterAllocationBasics.mli R ./genfft-k7/k7RegisterAllocator.ml R ./genfft-k7/k7RegisterAllocator.mli R ./genfft-k7/k7RegisterAllocatorEATranslation.ml R ./genfft-k7/k7RegisterAllocatorEATranslation.mli R ./genfft-k7/k7RegisterAllocatorInit.ml R ./genfft-k7/k7RegisterAllocatorInit.mli R ./genfft-k7/k7RegisterReallocation.ml R ./genfft-k7/k7RegisterReallocation.mli R ./genfft-k7/k7Translate.ml R ./genfft-k7/k7Translate.mli R ./genfft-k7/k7Unparsing.ml R ./genfft-k7/k7Unparsing.mli R ./genfft-k7/k7Vectorization.ml R ./genfft-k7/k7Vectorization.mli R ./genfft-k7/littlesimp.ml R ./genfft-k7/littlesimp.mli R ./genfft-k7/magic.ml R ./genfft-k7/memoMonad.ml R ./genfft-k7/memoMonad.mli R ./genfft-k7/monads.ml R ./genfft-k7/nonDetMonad.ml R ./genfft-k7/nonDetMonad.mli R ./genfft-k7/nullVectorization.ml R ./genfft-k7/nullVectorization.mli R ./genfft-k7/number.ml R ./genfft-k7/number.mli R ./genfft-k7/oracle.ml R ./genfft-k7/oracle.mli R ./genfft-k7/stateMonad.ml R ./genfft-k7/stateMonad.mli R ./genfft-k7/to_alist.ml R ./genfft-k7/to_alist.mli R ./genfft-k7/twiddle.ml R ./genfft-k7/twiddle.mli R ./genfft-k7/util.ml R ./genfft-k7/util.mli R ./genfft-k7/vAnnotatedScheduler.ml R ./genfft-k7/vAnnotatedScheduler.mli R ./genfft-k7/vDag.ml R ./genfft-k7/vDag.mli R ./genfft-k7/vFpBasics.ml R ./genfft-k7/vFpBasics.mli R ./genfft-k7/vFpUnparsing.ml R ./genfft-k7/vFpUnparsing.mli R ./genfft-k7/vImproveSchedule.ml R ./genfft-k7/vImproveSchedule.mli R ./genfft-k7/vK7Optimization.ml R ./genfft-k7/vK7Optimization.mli R ./genfft-k7/vScheduler.ml R ./genfft-k7/vScheduler.mli R ./genfft-k7/vSimdBasics.ml R ./genfft-k7/vSimdBasics.mli R ./genfft-k7/vSimdIndexing.ml R ./genfft-k7/vSimdIndexing.mli R ./genfft-k7/vSimdUnparsing.ml R ./genfft-k7/vSimdUnparsing.mli R ./genfft-k7/variable.ml R ./genfft-k7/variable.mli R ./genfft-k7/ M ./kernel/align.c -2 M ./kernel/ifftw.h -5 +1 M ./mkdist.sh -1 +1 M ./support/Makefile.am -1 +1 M ./support/Makefile.codelets -6 +1 R ./support/codelet_asmprelude Sun Aug 13 11:02:11 EDT 2006 athena@fftw.org * Use darcs instead of cvs. M ./mkdist.sh -5 +2 Wed Jul 19 08:52:15 EDT 2006 athena * [project @ 2006-07-19 12:52:15 by athena] Treat a the string "-" as a nonoption. M ./libbench2/my-getopt.c -1 +4 Tue Jul 4 17:10:47 EDT 2006 stevenj * [project @ 2006-07-04 21:10:47 by stevenj] comment out pkginclude dir for now M ./Makefile.am -2 +2 M ./api/Makefile.am -2 +2 M ./dft/Makefile.am -2 +2 M ./kernel/Makefile.am -2 +2 M ./rdft/Makefile.am -2 +2 M ./reodft/Makefile.am -2 +2 M ./threads/Makefile.am -2 +2 Mon Jul 3 20:51:08 EDT 2006 stevenj * [project @ 2006-07-04 00:51:08 by stevenj] make sure CCAS = CC to avoid libtool confusion M ./configure.ac -1 +1 Fri Jun 23 04:07:31 EDT 2006 stevenj * [project @ 2006-06-23 08:07:31 by stevenj] install x77.h guru.h guru64.h in pkgincludedir M ./api/Makefile.am -1 +1 Fri Jun 23 04:03:42 EDT 2006 stevenj * [project @ 2006-06-23 08:03:42 by stevenj] whitespace M ./configure.ac -3 Fri Jun 23 02:33:45 EDT 2006 stevenj * [project @ 2006-06-23 06:33:45 by stevenj] support cycle counter with xlc on Linux/ppc M ./kernel/cycle.h -2 +2 Tue Jun 20 08:16:08 EDT 2006 athena * [project @ 2006-06-20 12:16:08 by athena] Stylistic change. M ./tools/fftw-wisdom.c -1 +1 Tue Jun 20 02:20:34 EDT 2006 stevenj * [project @ 2006-06-20 06:20:34 by stevenj] bump date M ./m4/ax_cc_maxopt.m4 -1 +1 Tue Jun 20 02:20:06 EDT 2006 stevenj * [project @ 2006-06-20 06:20:06 by stevenj] correct bug reported by Andrew Salamon ... --enable-portable-binary was ignored (or rather, treated unpredictably) due to typo, grrr M ./m4/ax_cc_maxopt.m4 -1 +1 Thu Jun 1 20:30:06 EDT 2006 stevenj * [project @ 2006-06-02 00:30:06 by stevenj] install 'internal' header files into includedir/fftw3/, includedir/fftw3f/, etcetera....this will make it easier to write external libraries that plug into FFTW internals, e.g. to add new solvers M ./Makefile.am +3 M ./api/Makefile.am +3 M ./dft/Makefile.am +3 M ./kernel/Makefile.am +3 M ./rdft/Makefile.am +3 M ./reodft/Makefile.am +3 M ./threads/Makefile.am +3 Mon May 29 23:59:19 EDT 2006 stevenj * [project @ 2006-05-30 03:59:19 by stevenj] bug fix, thanks to James Donald for the bug report (only affects experimental semaphore stuff) M ./threads/threads.c +1 Mon May 29 23:58:16 EDT 2006 stevenj * [project @ 2006-05-30 03:58:16 by stevenj] comment M ./NEWS -3 +4 Mon May 29 21:02:50 EDT 2006 stevenj * [project @ 2006-05-30 01:02:50 by stevenj] whoops M ./m4/acx_pthread.m4 -1 +3 Sat May 27 19:36:15 EDT 2006 stevenj * [project @ 2006-05-27 23:36:15 by stevenj] version bump M ./m4/acx_pthread.m4 -1 +1 Sat May 27 14:54:47 EDT 2006 stevenj * [project @ 2006-05-27 18:54:47 by stevenj] only check for xlc_r/cc_r if we are not using gcc M ./m4/acx_pthread.m4 -2 +3 Fri May 26 15:00:38 EDT 2006 stevenj * [project @ 2006-05-26 19:00:38 by stevenj] use ptrdiff_t (it's C89 and standard C++, hooray) M ./api/fftw3.h -10 +5 Fri May 26 12:59:33 EDT 2006 stevenj * [project @ 2006-05-26 16:59:33 by stevenj] version bump M ./configure.ac -2 +2 Fri May 26 12:57:32 EDT 2006 stevenj * [project @ 2006-05-26 16:57:32 by stevenj] noted 64-bit guru API M ./NEWS +7 Fri May 26 12:53:09 EDT 2006 stevenj * [project @ 2006-05-26 16:53:09 by stevenj] note that newer versions of VC++ support long long M ./api/fftw3.h -2 +3 Fri May 26 12:46:09 EDT 2006 stevenj * [project @ 2006-05-26 16:46:09 by stevenj] try harder to get a portable 64-bit type M ./api/fftw3.h -2 +6 Thu May 25 22:04:18 EDT 2006 stevenj * [project @ 2006-05-26 02:04:18 by stevenj] added draft guru64 API A ./api/guru.h A ./api/guru64.h A ./api/mktensor-iodims.h A ./api/mktensor-iodims64.c A ./api/plan-guru-dft-c2r.h A ./api/plan-guru-dft-r2c.h A ./api/plan-guru-dft.h A ./api/plan-guru-r2r.h A ./api/plan-guru-split-dft-c2r.h A ./api/plan-guru-split-dft-r2c.h A ./api/plan-guru-split-dft.h A ./api/plan-guru64-dft-c2r.c A ./api/plan-guru64-dft-r2c.c A ./api/plan-guru64-dft.c A ./api/plan-guru64-r2r.c A ./api/plan-guru64-split-dft-c2r.c A ./api/plan-guru64-split-dft-r2c.c A ./api/plan-guru64-split-dft.c M ./api/Makefile.am -1 +7 M ./api/api.h +3 M ./api/fftw3.h -2 +56 M ./api/guru.h +4 M ./api/guru64.h +4 M ./api/mktensor-iodims.c -62 +2 M ./api/mktensor-iodims.h +62 M ./api/mktensor-iodims64.c +2 M ./api/plan-guru-dft-c2r.c -44 +2 M ./api/plan-guru-dft-c2r.h +44 M ./api/plan-guru-dft-r2c.c -43 +2 M ./api/plan-guru-dft-r2c.h +43 M ./api/plan-guru-dft.c -44 +2 M ./api/plan-guru-dft.h +44 M ./api/plan-guru-r2r.c -47 +2 M ./api/plan-guru-r2r.h +47 M ./api/plan-guru-split-dft-c2r.c -40 +2 M ./api/plan-guru-split-dft-c2r.h +40 M ./api/plan-guru-split-dft-r2c.c -39 +2 M ./api/plan-guru-split-dft-r2c.h +39 M ./api/plan-guru-split-dft.c -39 +2 M ./api/plan-guru-split-dft.h +39 M ./api/plan-guru64-dft-c2r.c +2 M ./api/plan-guru64-dft-r2c.c +2 M ./api/plan-guru64-dft.c +2 M ./api/plan-guru64-r2r.c +2 M ./api/plan-guru64-split-dft-c2r.c +2 M ./api/plan-guru64-split-dft-r2c.c +2 M ./api/plan-guru64-split-dft.c +2 Mon May 22 16:41:44 EDT 2006 stevenj * [project @ 2006-05-22 20:41:44 by stevenj] added FIXME note M ./m4/acx_pthread.m4 +1 Mon May 22 16:40:30 EDT 2006 stevenj * [project @ 2006-05-22 20:40:30 by stevenj] check for xlc_r in addition to cc_r; thanks to Guy Moebs for the bug report M ./m4/acx_pthread.m4 -3 +3 Fri Apr 21 12:35:25 EDT 2006 stevenj * [project @ 2006-04-21 16:35:25 by stevenj] added note about gcc 4.0.1 on MacOS/Intel M ./doc/FAQ/fftw-faq.bfnn -1 +4 Thu Apr 20 23:08:42 EDT 2006 stevenj * [project @ 2006-04-21 03:08:42 by stevenj] added code for Core Duo; thanks to Eric Branlund M ./m4/ax_gcc_archflag.m4 -2 +3 Thu Apr 20 20:21:03 EDT 2006 stevenj * [project @ 2006-04-21 00:21:03 by stevenj] fixed failure for -fPIC or for gcc-4 on Apple Intel machines; thanks to Eric Branlund for the bug report M ./m4/ax_gcc_x86_cpuid.m4 -4 +12 Tue Apr 11 20:00:31 EDT 2006 athena * [project @ 2006-04-12 00:00:31 by athena] Use -maltivec when checking for altivec.h. M ./configure.ac -1 +15 Mon Apr 3 15:52:44 EDT 2006 stevenj * [project @ 2006-04-03 19:52:44 by stevenj] note planner overwriting input in planner-flags reference M ./doc/fftw3.texi -1 +6 Tue Mar 28 09:05:26 EST 2006 athena * [project @ 2006-03-28 14:05:26 by athena] FAQ entry about --enable-k7 in 64-bit mode. M ./doc/FAQ/fftw-faq.bfnn +11 Mon Mar 27 23:41:05 EST 2006 stevenj * [project @ 2006-03-28 04:41:05 by stevenj] sprintf -> snprintf, to avoid (harmless) complaints by users/compilers M ./configure.ac -1 +1 M ./libbench2/report.c -12 +19 M ./tools/fftw-wisdom.c +4 Mon Mar 27 23:30:22 EST 2006 stevenj * [project @ 2006-03-28 04:30:22 by stevenj] silence compiler warning M ./kernel/align.c -5 +6 Fri Mar 17 09:20:10 EST 2006 athena * [project @ 2006-03-17 14:20:10 by athena] Remove dft/codelets/inplace, add simd/nonportable to list of directories to be compiled on non-unix systems. M ./doc/fftw3.texi -13 +14 Sat Aug 12 23:52:49 EDT 2006 Unknown tagger tagged fftw-3-1-1 Sat Mar 4 16:17:56 EST 2006 stevenj * [project @ 2006-03-04 21:17:56 by stevenj] whoops M ./doc/fftw3.texi -2 +2 Sat Mar 4 16:13:08 EST 2006 stevenj * [project @ 2006-03-04 21:13:08 by stevenj] note that we align the stack ourselves if necessary, with gcc and icc M ./doc/fftw3.texi -8 +10 Sat Mar 4 16:08:16 EST 2006 stevenj * [project @ 2006-03-04 21:08:16 by stevenj] clearer distinction between static and automatic storage in C M ./doc/fftw3.texi -29 +30 Sat Feb 25 20:27:01 EST 2006 stevenj * [project @ 2006-02-26 01:27:01 by stevenj] rm unused var M ./libbench2/verify-lib.c -2 +1 Sat Feb 25 17:30:28 EST 2006 athena * [project @ 2006-02-25 22:30:28 by athena] Improved usage of goto (Dijkstra miserere nostri) M ./libbench2/my-getopt.c -18 +17 Sat Feb 25 14:19:15 EST 2006 stevenj * [project @ 2006-02-25 19:19:15 by stevenj] boilerplate M ./libbench2/my-getopt.h +13 Sat Feb 25 14:14:40 EST 2006 stevenj * [project @ 2006-02-25 19:14:40 by stevenj] update for upcoming 3.1.1 M ./NEWS +17 Sat Feb 25 13:57:34 EST 2006 stevenj * [project @ 2006-02-25 18:57:34 by stevenj] replace obsolete IMPATIENT with MEASURE M ./tools/fftw-wisdom.c -1 +3 M ./tools/fftw_wisdom.1.in -2 +2 Sat Feb 25 13:52:25 EST 2006 stevenj * [project @ 2006-02-25 18:52:25 by stevenj] corrected comment M ./tools/fftw-wisdom.c -1 +1 Sat Feb 25 10:19:26 EST 2006 athena * [project @ 2006-02-25 15:19:26 by athena] -v does not take an argument. M ./tools/fftw-wisdom.c -6 +3 Sat Feb 25 10:17:18 EST 2006 athena * [project @ 2006-02-25 15:17:18 by athena] Obey the unix convention that -ab = -a -b M ./libbench2/my-getopt.c -15 +24 Fri Feb 24 23:13:49 EST 2006 stevenj * [project @ 2006-02-25 04:13:49 by stevenj] minor fixes (return error on unrecognized option) M ./libbench2/bench-main.c -2 +3 M ./libbench2/my-getopt.c +5 M ./tools/fftw-wisdom.c -3 +4 Fri Feb 24 22:46:12 EST 2006 stevenj * [project @ 2006-02-25 03:46:12 by stevenj] ugh M ./tools/fftw-wisdom.c -34 +30 Fri Feb 24 21:42:56 EST 2006 athena * [project @ 2006-02-25 02:42:56 by athena] require exact match for long options. M ./libbench2/my-getopt.c -1 +2 Fri Feb 24 21:38:02 EST 2006 athena * [project @ 2006-02-25 02:38:02 by athena] better fix M ./libbench2/my-getopt.c -1 +2 Fri Feb 24 21:37:06 EST 2006 athena * [project @ 2006-02-25 02:37:06 by athena] Fix M ./libbench2/my-getopt.c -1 +1 Fri Feb 24 21:25:48 EST 2006 athena * [project @ 2006-02-25 02:25:48 by athena] nothing A ./libbench2/my-getopt.c A ./libbench2/my-getopt.h M ./libbench2/Makefile.am -5 +4 M ./libbench2/bench-main.c -53 +48 M ./libbench2/bench.h -5 +1 R ./libbench2/getopt-utils.c R ./libbench2/getopt.c R ./libbench2/getopt.h R ./libbench2/getopt1.c M ./libbench2/my-getopt.c +153 M ./libbench2/my-getopt.h +33 Mon Feb 20 17:37:21 EST 2006 stevenj * [project @ 2006-02-20 22:37:21 by stevenj] rm transpose-indirect-inplace solver, which was buggy M ./dft/indirect-transpose.c -54 +11 Wed Feb 15 08:43:05 EST 2006 athena * [project @ 2006-02-15 13:43:05 by athena] Comment fix. M ./kernel/cycle.h -2 +2 Wed Feb 15 08:18:41 EST 2006 athena * [project @ 2006-02-15 13:18:41 by athena] Cycle counter for Visual C++ x86-64, courtesy of Dirk Michaelis M ./kernel/cycle.h -4 +5 Tue Feb 14 19:17:30 EST 2006 stevenj * [project @ 2006-02-15 00:17:30 by stevenj] rfftwnd.png is in builddir M ./doc/Makefile.am -1 +1 Tue Feb 14 19:03:27 EST 2006 stevenj * [project @ 2006-02-15 00:03:27 by stevenj] fixed typo: --enable-portable-binary, not --with M ./doc/fftw3.texi -3 +3 Mon Feb 13 07:59:06 EST 2006 athena * [project @ 2006-02-13 12:59:06 by athena] estimator tweaks. M ./dft/dftw-direct.c -2 +3 M ./rdft/hc2hc-direct.c -2 +2 Sun Feb 12 20:43:39 EST 2006 athena * [project @ 2006-02-13 01:43:39 by athena] sse/sse2 support for t3?v codelets M ./simd/simd-sse.h +22 M ./simd/simd-sse2.h -8 +20 Sun Feb 12 20:39:22 EST 2006 athena * [project @ 2006-02-13 01:39:22 by athena] Use CEXP instead of SIN/COS. M ./simd/simd-altivec.h -1 +1 Sun Feb 12 20:12:10 EST 2006 athena * [project @ 2006-02-13 01:12:10 by athena] bug in randomized cse eliminator. M ./genfft/oracle.ml -3 +5 Sun Feb 12 18:34:12 EST 2006 athena * [project @ 2006-02-12 23:34:12 by athena] Added support for t2-style simd codelets. This is altivec only for now; sse/sse2 don't even compile yet. A ./dft/simd/t3b.h A ./dft/simd/t3f.h M ./dft/simd/Makefile.am -2 +2 M ./dft/simd/codelets/Makefile.am -2 +11 M ./dft/simd/t3b.h +34 M ./dft/simd/t3f.h +34 M ./genfft/algsimp.ml -4 +26 M ./genfft/annotate.ml -5 +3 M ./genfft/c.ml -10 +9 M ./genfft/c.mli -2 +2 M ./genfft/complex.ml -93 +6 M ./genfft/complex.mli -5 +3 M ./genfft/expr.ml -6 +15 M ./genfft/expr.mli -2 +4 M ./genfft/gen_athtw.ml -3 +3 M ./genfft/gen_conv.ml -5 +3 M ./genfft/gen_hc2hc.ml -4 +3 M ./genfft/gen_hc2r.ml -5 +3 M ./genfft/gen_mdct.ml -7 +3 M ./genfft/gen_notw.ml -5 +2 M ./genfft/gen_notw_c.ml -5 +2 M ./genfft/gen_r2hc.ml -4 +3 M ./genfft/gen_r2r.ml -4 +3 M ./genfft/gen_twiddle.ml -3 +3 M ./genfft/gen_twiddle_c.ml -10 +4 M ./genfft/gen_twidsq.ml -7 +3 M ./genfft/gen_twidsq_c.ml -15 +5 M ./genfft/magic.ml -4 +4 M ./genfft/oracle.ml -1 +5 M ./genfft/schedule.ml -3 +3 M ./genfft/simd.ml -5 +11 M ./genfft/to_alist.ml -4 +16 M ./genfft/trig.ml -3 +2 M ./genfft/twiddle.ml -265 +35 M ./genfft/twiddle.mli -4 +4 M ./simd/simd-altivec.h +28 Sun Feb 12 15:30:27 EST 2006 athena * [project @ 2006-02-12 20:30:27 by athena] Added support for t2-style simd split-complex codelets. A ./dft/simd/ts.c A ./dft/simd/ts.h M ./dft/simd/Makefile.am -2 +2 M ./dft/simd/codelets/Makefile.am -2 +7 R ./dft/simd/t1s.c R ./dft/simd/t1s.h M ./dft/simd/ts.c +50 M ./dft/simd/ts.h +33 M ./genfft/twiddle.ml -5 +3 Fri Feb 10 18:21:28 EST 2006 stevenj * [project @ 2006-02-10 23:21:28 by stevenj] M ./m4/ax_openmp.m4 -1 +1 Fri Feb 10 18:19:46 EST 2006 stevenj * [project @ 2006-02-10 23:19:46 by stevenj] punctuation M ./m4/ax_openmp.m4 -2 +2 Fri Feb 10 18:00:35 EST 2006 stevenj * [project @ 2006-02-10 23:00:35 by stevenj] windows DLL stuff for Fortran interface M ./api/f77api.c +7 M ./api/f77funcs.h -43 +43 Fri Feb 10 09:48:52 EST 2006 athena * [project @ 2006-02-10 14:48:52 by athena] Bumped version to 3.1.1 M ./configure.ac -2 +2 Fri Feb 10 09:18:39 EST 2006 athena * [project @ 2006-02-10 14:18:39 by athena] Precompute array indices on x86-64. Speeds up Pentium IV and makes no appreciable difference on AMD. M ./kernel/ifftw.h -2 +2 Tue Feb 7 22:01:36 EST 2006 athena * [project @ 2006-02-08 03:01:36 by athena] Check whether the processor supports CPUID before issuing the instruction. (Grrr...) Code contributed by Eric J. Korpela. A ./simd/x86-cpuid.h M ./simd/Makefile.am -1 +1 M ./simd/sse.c -24 +6 M ./simd/sse2.c -26 +8 M ./simd/x86-cpuid.h +127 Tue Feb 7 21:36:47 EST 2006 athena * [project @ 2006-02-08 02:36:47 by athena] icc supports x86_64 these days. M ./kernel/cycle.h -2 +2 Sun Feb 5 18:19:55 EST 2006 athena * [project @ 2006-02-05 23:19:55 by athena] Paranoia. M ./kernel/primes.c -2 +2 Mon Jan 30 15:27:53 EST 2006 stevenj * [project @ 2006-01-30 20:27:53 by stevenj] whoops, fixed assert (y <= x) M ./kernel/primes.c -2 +2 Mon Jan 30 15:26:22 EST 2006 stevenj * [project @ 2006-01-30 20:26:22 by stevenj] note that safe_mulmod requires {x,y} < p (or at least < 2p), and added assert M ./kernel/primes.c -2 +4 Mon Jan 30 11:09:32 EST 2006 athena * [project @ 2006-01-30 16:09:32 by athena] fixed aix/xlc lossage M ./libbench2/bench-user.h -2 +2 M ./libbench2/timer.c -4 +4 Sun Jan 29 20:42:51 EST 2006 athena * [project @ 2006-01-30 01:42:51 by athena] In the impuse test, normalize the impulse so that the impulse and the random vectors have roughly the same L2 norm. This change reduces the number of bits that we lose because of floating-point cancellation, so that we can focus on the bits that we lose because of bugs. M ./libbench2/verify-lib.c -3 +3 Sun Jan 29 20:37:47 EST 2006 athena * [project @ 2006-01-30 01:37:47 by athena] Compute omega in trigreal precision, as opposed to R. M ./rdft/dht-rader.c -2 +2 Sat Aug 12 23:51:14 EDT 2006 Unknown tagger tagged fftw-3-1 Fri Jan 27 19:16:22 EST 2006 stevenj * [project @ 2006-01-28 00:16:22 by stevenj] add --with-combined-threads option as workaround to Windows inability to build shared libs with dependencies M ./Makefile.am -1 +7 M ./configure.ac -1 +3 M ./tests/Makefile.am +2 M ./threads/Makefile.am +4 M ./tools/Makefile.am +2 Fri Jan 27 17:20:45 EST 2006 stevenj * [project @ 2006-01-27 22:20:45 by stevenj] libfftw3_threads should *not* used -no-undefined because, in fact, it is not true -- this library depends on -lfftw3, and is not self-contained M ./threads/Makefile.am -1 +1 Thu Jan 26 22:04:34 EST 2006 stevenj * [project @ 2006-01-27 03:04:34 by stevenj] updated M ./NEWS -3 +10 Thu Jan 26 21:10:50 EST 2006 athena * [project @ 2006-01-27 02:10:50 by athena] Added paranoid stack alignment when awaking plans. While I was at it, removed obsolete, redundant AWAKE macro. M ./api/apiplan.c -3 +8 M ./dft/bluestein.c -1 +1 M ./dft/buffered.c -4 +4 M ./dft/ct.c -3 +3 M ./dft/ctsq.c -2 +2 M ./dft/dftw-generic.c -1 +1 M ./dft/dftw-genericbuf.c -1 +1 M ./dft/indirect-transpose.c -3 +3 M ./dft/indirect.c -3 +3 M ./dft/rader.c -3 +3 M ./dft/rank-geq2.c -3 +3 M ./dft/vrank-geq1.c -2 +2 M ./kernel/ifftw.h -2 +1 M ./kernel/timer.c -3 +3 M ./rdft/buffered.c -4 +4 M ./rdft/buffered2.c -3 +3 M ./rdft/dft-r2hc.c -2 +2 M ./rdft/dht-r2hc.c -2 +2 M ./rdft/dht-rader.c -3 +3 M ./rdft/hc2hc-direct.c -3 +3 M ./rdft/hc2hc-directbuf.c -3 +3 M ./rdft/hc2hc-generic.c -2 +2 M ./rdft/hc2hc.c -2 +2 M ./rdft/indirect.c -3 +3 M ./rdft/rank-geq2-rdft2.c -3 +3 M ./rdft/rank-geq2.c -3 +3 M ./rdft/rank0-rdft2.c -2 +2 M ./rdft/rdft-dht.c -2 +2 M ./rdft/rdft2-radix2.c -2 +2 M ./rdft/vrank-geq1-rdft2.c -2 +2 M ./rdft/vrank-geq1.c -2 +2 M ./rdft/vrank3-transpose.c -4 +4 M ./reodft/redft00e-r2hc-pad.c -3 +3 M ./reodft/redft00e-r2hc.c -2 +2 M ./reodft/reodft00e-splitradix.c -3 +3 M ./reodft/reodft010e-r2hc.c -2 +2 M ./reodft/reodft11e-r2hc-odd.c -2 +2 M ./reodft/reodft11e-r2hc.c -2 +2 M ./reodft/reodft11e-radix2.c -2 +2 M ./reodft/rodft00e-r2hc-pad.c -3 +3 M ./reodft/rodft00e-r2hc.c -2 +2 M ./tests/hook.c -2 +2 M ./threads/ct.c -3 +3 M ./threads/dft-vrank-geq1.c -2 +2 M ./threads/hc2hc.c -2 +2 M ./threads/rdft-vrank-geq1.c -2 +2 M ./threads/vrank-geq1-rdft2.c -2 +2 Thu Jan 26 20:54:39 EST 2006 athena * [project @ 2006-01-27 01:54:39 by athena] Updated for 3.1. M ./NEWS -1 +1 Thu Jan 26 19:15:12 EST 2006 athena * [project @ 2006-01-27 00:15:12 by athena] ditched one alignment check and noted that we should eliminate the rest as well M ./TODO +3 M ./libbench2/bench-main.c -13 +1 Wed Jan 25 23:02:19 EST 2006 athena * [project @ 2006-01-26 04:02:19 by athena] alignment hack M ./libbench2/bench-main.c -2 +4 Wed Jan 25 22:05:11 EST 2006 athena * [project @ 2006-01-26 03:05:11 by athena] detect pentium M M ./m4/ax_gcc_archflag.m4 -1 +2 Wed Jan 25 18:42:58 EST 2006 stevenj * [project @ 2006-01-25 23:42:58 by stevenj] don't trust host_cpu if it claims we are on i386/i486, and call cpuid anyway (if it fails we use no arch flag). This is needed on FreeBSD M ./m4/ax_gcc_archflag.m4 -3 +1 Wed Jan 25 18:00:04 EST 2006 stevenj * [project @ 2006-01-25 23:00:04 by stevenj] suggest --with-our-malloc16 in error message M ./kernel/kalloc.c -2 +2 Tue Jan 24 19:53:34 EST 2006 stevenj * [project @ 2006-01-25 00:53:34 by stevenj] ditto for -no-gcc M ./configure.ac -2 +4 Tue Jan 24 19:51:08 EST 2006 stevenj * [project @ 2006-01-25 00:51:08 by stevenj] flags required for successfull compilation should be added even if the user overrides CFLAGS M ./configure.ac -9 +12 Tue Jan 24 18:43:59 EST 2006 stevenj * [project @ 2006-01-24 23:43:59 by stevenj] upcoming gcc OpenMP support uses -fopenmp M ./m4/ax_openmp.m4 -4 +4 Tue Jan 24 18:26:59 EST 2006 stevenj * [project @ 2006-01-24 23:26:59 by stevenj] note that PGI uses -mp as well M ./m4/ax_openmp.m4 -1 +1 Mon Jan 23 15:31:24 EST 2006 athena * [project @ 2006-01-23 20:31:24 by athena] my best guess at how to fix the microsoft crap du jour M ./kernel/cycle.h -2 +2 M ./simd/sse.c -65 +70 M ./simd/sse2.c -62 +67 Mon Jan 23 14:05:14 EST 2006 stevenj * [project @ 2006-01-23 19:05:14 by stevenj] use -Masmkeyword for PGI cycle counter, grr M ./configure.ac +5 M ./kernel/cycle.h -2 +4 Sun Jan 22 18:09:06 EST 2006 athena * [project @ 2006-01-22 23:09:06 by athena] Bumped version number to 3.1. M ./configure.ac -1 +1 Sat Jan 21 10:03:59 EST 2006 athena * [project @ 2006-01-21 15:03:59 by athena] Report that --enable-k7 is incompatible with --enable-shared. M ./configure.ac +3 Sat Jan 21 09:17:54 EST 2006 athena * [project @ 2006-01-21 14:17:54 by athena] Do not use empty libraries in LIBADD, since otherwise the linker fails on Solaris. M ./Makefile.am -7 +20 Wed Jan 18 10:47:59 EST 2006 stevenj * [project @ 2006-01-18 15:47:59 by stevenj] warn end-users away from this file M ./bootstrap.sh -1 +10 Tue Jan 17 16:16:42 EST 2006 athena * [project @ 2006-01-17 21:16:42 by athena] Gcc sucks. M ./simd/simd-sse.h -1 +2 Tue Jan 17 11:48:55 EST 2006 athena * [project @ 2006-01-17 16:48:55 by athena] Disabled checks that may turn out to be too paranoid. M ./tests/hook.c -1 +2 Tue Jan 17 10:35:03 EST 2006 athena * [project @ 2006-01-17 15:35:03 by athena] Some paranoid checks. M ./tests/hook.c +16 Tue Jan 17 09:31:08 EST 2006 athena * [project @ 2006-01-17 14:31:08 by athena] Flush stdout after printing. M ./libbench2/ovtpvt.c +1 Tue Jan 17 08:28:18 EST 2006 athena * [project @ 2006-01-17 13:28:18 by athena] Run the leak detector in all cases, not just when verbose > 2. M ./kernel/alloc.c -2 +7 M ./tests/bench.c -2 +1 Tue Jan 17 08:11:41 EST 2006 athena * [project @ 2006-01-17 13:11:41 by athena] Eliminate calls to pow(), rint(). M ./api/mapflags.c -8 +3 Tue Jan 17 00:45:06 EST 2006 stevenj * [project @ 2006-01-17 05:45:06 by stevenj] put # in first column, for stylistic consistency M ./kernel/ifftw.h -3 +3 Tue Jan 17 00:17:27 EST 2006 athena * [project @ 2006-01-17 05:17:27 by athena] Made timeout part of impatience flags, in order to improve the usability of wisdom. Also, fixed bogus error recovery logic in planner.c:imprt(). M ./api/mapflags.c -1 +36 M ./kernel/ifftw.h -4 +7 M ./kernel/planner.c -23 +70 Mon Jan 16 23:03:34 EST 2006 stevenj * [project @ 2006-01-17 04:03:33 by stevenj] make timelimit < 0 .eq. FFTW_NO_TIMELIMIT M ./api/apiplan.c -1 +1 M ./api/fftw3.h -1 +3 M ./doc/fftw3.texi -5 +9 M ./kernel/planner.c -3 +3 Mon Jan 16 21:52:01 EST 2006 athena * [project @ 2006-01-17 02:52:01 by athena] Eliminated the FFTW_TIMELIMIT flag in favor of this simpler logic: fftw_set_timelimit(0) disables time limit. fftw_set_timelimit(X), X>0 sets the time limit to X. M ./api/apiplan.c -6 +1 M ./api/fftw3.h -2 +1 M ./api/the-planner.c +7 M ./doc/fftw3.texi -30 +27 M ./kernel/planner.c -2 +4 M ./tests/bench.c -1 Mon Jan 16 08:38:04 EST 2006 athena * [project @ 2006-01-16 13:38:04 by athena] Force the use of the estimator when wisdom fails because of md5 collisions, otherwise the planner takes forever. M ./api/apiplan.c -2 +10 Sun Jan 15 21:30:31 EST 2006 athena * [project @ 2006-01-16 02:30:31 by athena] Ranted about how broken gcc-4 is. M ./kernel/ifftw.h -5 +40 Sun Jan 15 19:59:38 EST 2006 stevenj * [project @ 2006-01-16 00:59:38 by stevenj] change fftw_timelimit global var to fftw_set_timelimit(double) function, for simpler usage with shared libraries and for consistency with e.g. set_numthreads M ./api/apiplan.c -2 +4 M ./api/fftw3.h -2 +3 M ./doc/fftw3.texi -11 +11 M ./tests/bench.c -1 +1 Sun Jan 15 19:32:27 EST 2006 athena * [project @ 2006-01-16 00:32:27 by athena] Minor tweaks. M ./doc/fftw3.texi -8 +11 Sun Jan 15 16:32:54 EST 2006 athena * [project @ 2006-01-15 21:32:54 by athena] tweaks to make sure that time_n() is always called from the same stack position. M ./libbench2/timer.c -12 +12 Sun Jan 15 16:09:53 EST 2006 athena * [project @ 2006-01-15 21:09:53 by athena] Major simplification of the timer calibration logic. Also, use an FFT as a unit of work instead of the old pointer chasing, because God knows how pointer chasing interacts with the idiotic cache-hit speculation on the Pentium IV. M ./libbench2/bench.h -2 +3 M ./libbench2/speed.c -5 +8 M ./libbench2/timer.c -96 +35 M ./libbench2/timer2.c -7 +99 Sun Jan 15 15:12:08 EST 2006 athena * [project @ 2006-01-15 20:12:08 by athena] Fixed broken aligment checks when sizeof(R)==12. M ./kernel/align.c -3 +8 Sun Jan 15 10:36:40 EST 2006 athena * [project @ 2006-01-15 15:36:40 by athena] Manual unrolling of loop. M ./libbench2/timer2.c -5 +2 Sun Jan 15 10:12:55 EST 2006 athena * [project @ 2006-01-15 15:12:55 by athena] Various improvements to timer calibration routines. A ./libbench2/timer2.c M ./libbench2/Makefile.am -1 +1 M ./libbench2/bench.h -1 +2 M ./libbench2/timer.c -18 +32 M ./libbench2/timer2.c +36 Sat Jan 14 22:16:09 EST 2006 athena * [project @ 2006-01-15 03:16:09 by athena] cygwin defines __CYGWIN__, not __WIN32__ etc. M ./libbench2/timer.c -16 +17 Sat Jan 14 20:40:12 EST 2006 athena * [project @ 2006-01-15 01:40:12 by athena] fixed confusion between libbench and user timers M ./libbench2/bench-user.h -3 +7 M ./libbench2/speed.c -5 +5 M ./libbench2/timer.c -6 +8 M ./tests/bench.c -8 +8 Sat Jan 14 12:32:44 EST 2006 stevenj * [project @ 2006-01-14 17:32:44 by stevenj] update M ./NEWS -2 +4 Sat Jan 14 10:24:11 EST 2006 athena * [project @ 2006-01-14 15:24:11 by athena] Comment. M ./simd/simd-sse.h +5 Sat Jan 14 10:19:28 EST 2006 athena * [project @ 2006-01-14 15:19:28 by athena] Workaround gcc bug. M ./simd/simd-sse.h -2 +3 Fri Jan 13 19:13:18 EST 2006 athena * [project @ 2006-01-14 00:13:18 by athena] Switched to -beta2. M ./configure.ac -1 +1 Thu Jan 12 22:21:57 EST 2006 athena * [project @ 2006-01-13 03:21:57 by athena] Fixed technically correct but highly obfuscated use of the enum tag R2HC as a null pointer. M ./rdft/buffered.c -5 +3 M ./rdft/indirect.c -4 +3 M ./rdft/problem.c -1 +8 M ./rdft/rank0-rdft2.c -4 +3 M ./rdft/rdft.h +1 M ./rdft/vrank3-transpose.c -25 +13 Thu Jan 12 19:25:20 EST 2006 stevenj * [project @ 2006-01-13 00:25:20 by stevenj] --enable-unsafe-mulmod is obsolete M ./configure.ac -4 Thu Jan 12 19:23:18 EST 2006 athena * [project @ 2006-01-13 00:23:18 by athena] More thoughts. M ./TODO +2 Thu Jan 12 19:17:57 EST 2006 athena * [project @ 2006-01-13 00:17:57 by athena] Removed loop unrolling because it slows things down on at least one powerpc and it generates clumsy x86 code. M ./rdft/buffered2.c -50 +12 Thu Jan 12 19:17:35 EST 2006 stevenj * [project @ 2006-01-13 00:17:35 by stevenj] tweaks M ./kernel/kalloc.c -6 +6 Thu Jan 12 15:55:52 EST 2006 stevenj * [project @ 2006-01-12 20:55:52 by stevenj] MacOSX x86 ABI specifies that the stack is kept 16-byte aligned M ./kernel/ifftw.h -2 +3 Thu Jan 12 12:46:49 EST 2006 athena * [project @ 2006-01-12 17:46:49 by athena] ``ret'' is a reserved word in the evil empire. M ./kernel/cycle.h -5 +5 Thu Jan 12 08:31:43 EST 2006 athena * [project @ 2006-01-12 13:31:09 by athena] Changed ret => result because ret ``is a reserved word'' in the evil empire. M ./simd/sse.c -4 +4 M ./simd/sse2.c -4 +4 Wed Jan 11 19:30:42 EST 2006 athena * [project @ 2006-01-12 00:30:42 by athena] Workaround Visual c++ lossage. M ./simd/simd-sse2.h -5 +4 Wed Jan 11 19:26:16 EST 2006 athena * [project @ 2006-01-12 00:26:16 by athena] Workaround visual c++ lossage. M ./simd/simd-sse.h +38 Wed Jan 11 19:10:52 EST 2006 athena * [project @ 2006-01-12 00:10:52 by athena] isprint() is guaranteed to work for unsigned char + EOF only. M ./libbench2/getopt-utils.c -2 +3 Wed Jan 11 13:47:49 EST 2006 stevenj * [project @ 2006-01-11 18:47:49 by stevenj] rm obsolete fixme M ./rdft/vrank3-transpose.c -2 +2 Wed Jan 11 13:38:46 EST 2006 stevenj * [project @ 2006-01-11 18:38:46 by stevenj] M ./rdft/vrank3-transpose.c -2 +2 Wed Jan 11 13:32:26 EST 2006 stevenj * [project @ 2006-01-11 18:32:26 by stevenj] fix comment M ./rdft/vrank3-transpose.c -2 +3 Wed Jan 11 12:27:05 EST 2006 athena * [project @ 2006-01-11 17:27:05 by athena] Paranoid use of K(x) for all constants x, to avoid runtime double->float conversions on sufficiently stupid compilers. M ./dft/bluestein.c -2 +2 M ./rdft/buffered2.c -2 +2 M ./rdft/dht-rader.c -1 +1 M ./rdft/rank0-rdft2.c -11 +11 M ./reodft/rodft00e-r2hc-pad.c -3 +3 Tue Jan 10 20:10:38 EST 2006 athena * [project @ 2006-01-11 01:10:38 by athena] Workaround to gcc nonsense. M ./simd/simd-sse.h -1 +15 Tue Jan 10 18:44:28 EST 2006 stevenj * [project @ 2006-01-10 23:44:28 by stevenj] bug fix: infinite loop in transpose-cut planning M ./rdft/vrank3-transpose.c -6 +14 Tue Jan 10 18:12:14 EST 2006 stevenj * [project @ 2006-01-10 23:12:14 by stevenj] clarified comment M ./api/fftw3.h -6 +6 Tue Jan 10 18:10:32 EST 2006 stevenj * [project @ 2006-01-10 23:10:32 by stevenj] more Windows decorations M ./tests/bench.c -1 +1 Tue Jan 10 17:57:45 EST 2006 stevenj * [project @ 2006-01-10 22:57:45 by stevenj] added FIXME comment M ./support/Makefile.codelets +1 Tue Jan 10 17:52:07 EST 2006 stevenj * [project @ 2006-01-10 22:52:07 by stevenj] 'make clean' should not delete codlist.c since it is included in the dist tarball M ./support/Makefile.codelets -2 +3 Tue Jan 10 17:50:12 EST 2006 athena * [project @ 2006-01-10 22:50:12 by athena] Change threshold for ``large'' Cooley-Tukey to 256K from 64K, since it seems to benefit the Pentium IV with sse and the planning cost is not too horrible. M ./dft/dftw-direct.c -2 +2 Tue Jan 10 17:45:11 EST 2006 stevenj * [project @ 2006-01-10 22:45:11 by stevenj] more missing Windows DLL decorations M ./kernel/ifftw.h -3 +3 Tue Jan 10 17:41:28 EST 2006 stevenj * [project @ 2006-01-10 22:41:28 by stevenj] remove unused var M ./rdft/dht-rader.c -2 Tue Jan 10 14:00:50 EST 2006 stevenj * [project @ 2006-01-10 19:00:50 by stevenj] allow compiler threads, if enabled, to take precedence over explicit threads M ./threads/threads.c -17 +17 Tue Jan 10 12:30:09 EST 2006 stevenj * [project @ 2006-01-10 17:30:09 by stevenj] M ./api/api.h -2 +2 Tue Jan 10 12:21:56 EST 2006 stevenj * [project @ 2006-01-10 17:21:56 by stevenj] M ./kernel/planner.c -2 +2 Tue Jan 10 09:13:20 EST 2006 athena * [project @ 2006-01-10 14:13:20 by athena] Fixed comment typo. M ./kernel/planner.c -2 +2 Tue Jan 10 08:59:22 EST 2006 athena * [project @ 2006-01-10 13:59:22 by athena] Rearranged timeout checks so as to eliminate one of them. M ./kernel/planner.c -9 +4 Tue Jan 10 08:56:55 EST 2006 athena * [project @ 2006-01-10 13:56:55 by athena] Converted residual CK() -> A(). M ./kernel/plan.c -2 +2 Tue Jan 10 08:36:13 EST 2006 athena * [project @ 2006-01-10 13:36:13 by athena] Maintain the invariant TIMED_OUT ==> NEED_TIMEOUT_CHECK. M ./kernel/planner.c -5 +10 Tue Jan 10 08:24:41 EST 2006 athena * [project @ 2006-01-10 13:24:41 by athena] silence some 64-bit warnings M ./api/mapflags.c -1 +1 M ./dft/rank-geq2.c -2 +2 M ./dft/vrank-geq1.c -2 +2 M ./kernel/buffered.c -1 +1 M ./kernel/md5.c -1 +1 M ./kernel/scan.c -2 +3 M ./rdft/rank-geq2-rdft2.c -2 +2 M ./rdft/rank-geq2.c -2 +2 M ./rdft/vrank-geq1-rdft2.c -2 +2 M ./rdft/vrank-geq1.c -2 +2 M ./threads/dft-vrank-geq1.c -2 +2 M ./threads/rdft-vrank-geq1.c -2 +2 M ./threads/vrank-geq1-rdft2.c -2 +2 Tue Jan 10 07:58:48 EST 2006 athena * [project @ 2006-01-10 12:58:48 by athena] Assertions. M ./tests/hook.c +3 Tue Jan 10 00:14:00 EST 2006 stevenj * [project @ 2006-01-10 05:14:00 by stevenj] some condensing M ./kernel/timer.c -10 +5 Tue Jan 10 00:03:32 EST 2006 stevenj * [project @ 2006-01-10 05:03:32 by stevenj] eliminate X(seconds) in favor of X(elapsed_since), in paranoia of clock wrap M ./api/apiplan.c -2 +2 M ./kernel/ifftw.h -3 +33 M ./kernel/planner.c -4 +4 M ./kernel/timer.c -58 +25 Mon Jan 9 23:21:21 EST 2006 stevenj * [project @ 2006-01-10 04:21:21 by stevenj] M ./kernel/timer.c -2 +2 Mon Jan 9 23:21:06 EST 2006 stevenj * [project @ 2006-01-10 04:21:06 by stevenj] hmm, a bit more pessimistic about clock wrapping M ./kernel/timer.c -4 +4 Mon Jan 9 23:20:26 EST 2006 athena * [project @ 2006-01-10 04:20:26 by athena] Revert to md5uint = unsigned int whenever possible, so as to avoid wasting space for unsigned long on 64-bit machines. M ./configure.ac +2 M ./kernel/ifftw.h -1 +5 Mon Jan 9 23:12:27 EST 2006 stevenj * [project @ 2006-01-10 04:12:27 by stevenj] note why clock() wrap should not be a concern M ./kernel/timer.c -1 +6 Mon Jan 9 22:57:16 EST 2006 stevenj * [project @ 2006-01-10 03:57:16 by stevenj] bugfix in recent timeout changes - check for case where last solver times out M ./kernel/planner.c -2 +8 Mon Jan 9 22:40:26 EST 2006 stevenj * [project @ 2006-01-10 03:40:26 by stevenj] started changes list from beta M ./NEWS -1 +11 Mon Jan 9 22:34:13 EST 2006 athena * [project @ 2006-01-10 03:34:13 by athena] Paranoia. M ./api/mapflags.c +4 Mon Jan 9 22:27:37 EST 2006 athena * [project @ 2006-01-10 03:27:37 by athena] Paranoid assertions. M ./kernel/planner.c -3 +7 Mon Jan 9 22:13:32 EST 2006 athena * [project @ 2006-01-10 03:13:32 by athena] Added FIXME comment stating the 64-bit uncleaniness of fftw_tensor_to_bench_tensor(). M ./tests/hook.c +1 Mon Jan 9 22:06:05 EST 2006 athena * [project @ 2006-01-10 03:06:05 by athena] Another 64-bit bug. M ./dft/simd/t.c -1 +1 Mon Jan 9 21:54:07 EST 2006 stevenj * [project @ 2006-01-10 02:54:07 by stevenj] more Windows DLL nonsense M ./api/api.h -5 +15 M ./kernel/ifftw.h -3 +5 M ./tests/hook.c +2 Mon Jan 9 21:18:25 EST 2006 stevenj * [project @ 2006-01-10 02:18:25 by stevenj] some additional dllexport tags required to build the test program, due to internal stuff called by hook.c M ./api/api.h -2 +3 M ./kernel/ifftw.h -6 +17 Mon Jan 9 20:31:15 EST 2006 stevenj * [project @ 2006-01-10 01:31:15 by stevenj] M ./api/fftw3.h -2 +2 Mon Jan 9 20:30:19 EST 2006 stevenj * [project @ 2006-01-10 01:30:19 by stevenj] comment M ./api/fftw3.h -4 +6 Mon Jan 9 20:20:28 EST 2006 stevenj * [project @ 2006-01-10 01:17:11 by stevenj] M ./api/api.h -7 +7 M ./api/fftw3.h -4 +3 Mon Jan 9 20:16:50 EST 2006 stevenj * [project @ 2006-01-10 01:16:50 by stevenj] clarification M ./api/fftw3.h -3 +3 Mon Jan 9 20:12:23 EST 2006 stevenj * [project @ 2006-01-10 01:12:23 by stevenj] define FFTW_DLL if DLL_EXPORT (defined by libtool) is supplied M ./api/api.h +10 Mon Jan 9 20:05:11 EST 2006 stevenj * [project @ 2006-01-10 01:05:11 by stevenj] whoops M ./api/fftw3.h -2 +2 Mon Jan 9 20:00:47 EST 2006 stevenj * [project @ 2006-01-10 01:00:47 by stevenj] another stab at Windows DLL mess M ./api/fftw3.h -172 +185 Mon Jan 9 19:23:42 EST 2006 athena * [project @ 2006-01-10 00:23:42 by athena] 64-bit clean SIMD header file. I missed those because sparse does not know vector types. Grrr... M ./simd/simd-altivec.h -11 +11 M ./simd/simd-sse.h -6 +6 M ./simd/simd-sse2.h -4 +4 Mon Jan 9 19:08:36 EST 2006 stevenj * [project @ 2006-01-10 00:08:36 by stevenj] this option is called AC_DISABLE_SHARED in the documentation M ./configure.ac -1 +1 Mon Jan 9 17:34:13 EST 2006 stevenj * [project @ 2006-01-09 22:34:13 by stevenj] fixed --with-gcc-arch to work when cross-compiling M ./m4/ax_gcc_archflag.m4 -3 +4 Mon Jan 9 12:04:04 EST 2006 athena * [project @ 2006-01-09 17:04:04 by athena] Moved the timeout check back into the search loop, sicut erat in principio. This gives us a precise control over the timeout. To avoid the overhead of X(seconds)(), only call X(seconds)() if some time measurement was taken since the last call to X(seconds)(). M ./api/apiplan.c -11 +1 M ./kernel/ifftw.h -1 +2 M ./kernel/planner.c -5 +36 Mon Jan 9 00:07:40 EST 2006 stevenj * [project @ 2006-01-09 05:07:40 by stevenj] comments M ./rdft/vrank3-transpose.c -5 +5 Sun Jan 8 23:58:23 EST 2006 stevenj * [project @ 2006-01-09 04:58:23 by stevenj] generalized transpose-cut routine to be able to call transpose-gcd recursivly; TOMS follow-the-cycles algorithm now seems to be completely superseded M ./rdft/vrank3-transpose.c -111 +142 Sun Jan 8 20:53:18 EST 2006 stevenj * [project @ 2006-01-09 01:53:18 by stevenj] M ./threads/threads.c -3 +2 Sun Jan 8 20:52:16 EST 2006 stevenj * [project @ 2006-01-09 01:52:16 by stevenj] ignore errors from setscope -- POSIX standard does not require PTHREAD_SCOPE_SYSTEM to be supported, and PTHREAD_SCOPE_PROCESS is usually okay in that case M ./threads/threads.c -7 +9 Sun Jan 8 15:58:40 EST 2006 stevenj * [project @ 2006-01-08 20:58:40 by stevenj] added TODO comment M ./rdft/vrank3-transpose.c -1 +7 Sun Jan 8 15:39:28 EST 2006 stevenj * [project @ 2006-01-08 20:39:28 by stevenj] whoops M ./rdft/vrank3-transpose.c -8 +18 Sun Jan 8 14:44:23 EST 2006 athena * [project @ 2006-01-08 19:44:23 by athena] Boasted ``much faster altivec performance''. M ./NEWS -1 +1 Sun Jan 8 11:44:52 EST 2006 athena * [project @ 2006-01-08 16:44:52 by athena] Added a new pass to the generator to schedule for the pipeline latency. (This schedule modifies the ``optimal'' cache-oblivious schedule and hence it uses more registers.) This pass is currently: * disabled for non-fma code, under the assumption that this will run on a register-starved fma. * enabled for non-simd fma code, under the assumption that this will run on a processor with 32 or more FP registers. The latency of 4 is conservative and does not introduce too much register pressure. * enabled for simd fma code, under the assumption that this will run on altivec. The latency of 8 seems to produce the best results. M ./configure.ac -1 +1 M ./dft/simd/codelets/Makefile.am -1 +1 M ./genfft/annotate.ml -3 +58 M ./genfft/magic.ml -3 +5 M ./genfft/schedule.ml -33 +5 M ./support/Makefile.codelets -2 +2 M ./support/twovers.sh -1 +1 Sun Jan 8 03:13:53 EST 2006 stevenj * [project @ 2006-01-08 08:13:53 by stevenj] fixed estimator for vrank3-transpose M ./rdft/vrank3-transpose.c -6 +22 Sun Jan 8 02:02:11 EST 2006 stevenj * [project @ 2006-01-08 07:02:11 by stevenj] more detail on VC++ workaround M ./NEWS -1 +2 Sun Jan 8 00:19:19 EST 2006 stevenj * [project @ 2006-01-08 05:19:19 by stevenj] typo M ./rdft/vrank3-transpose.c -2 +2 Sun Jan 8 00:16:20 EST 2006 stevenj * [project @ 2006-01-08 05:16:20 by stevenj] screw it, just use planner for all sub-transposes in vrank3-transpose (still just use memcpy for contiguous copies, though) M ./rdft/vrank3-transpose.c -95 +185 Sat Jan 7 23:13:45 EST 2006 stevenj * [project @ 2006-01-08 04:13:45 by stevenj] add an assert M ./kernel/tile2d.c +2 Sat Jan 7 21:57:34 EST 2006 stevenj * [project @ 2006-01-08 02:57:34 by stevenj] vrank3-transpose now uses planner to decide whether to use cpy2d, cpy2d_tiled, etc. M ./kernel/ifftw.h -1 +7 M ./rdft/rank0.c -7 +5 M ./rdft/vrank3-transpose.c -217 +231 Sat Jan 7 20:57:16 EST 2006 stevenj * [project @ 2006-01-08 01:57:16 by stevenj] too annoying to have isqrt unexpectedly fail for n==0 M ./kernel/primes.c -2 +4 Sat Jan 7 17:49:37 EST 2006 stevenj * [project @ 2006-01-07 22:49:37 by stevenj] clarifications M ./NEWS -1 +1 M ./doc/fftw3.texi -2 +5 Sat Jan 7 16:39:20 EST 2006 stevenj * [project @ 2006-01-07 21:39:20 by stevenj] comment fix M ./rdft/vrank3-transpose.c -4 +4 Sat Jan 7 15:16:22 EST 2006 stevenj * [project @ 2006-01-07 20:16:22 by stevenj] more faq updates M ./doc/FAQ/fftw-faq.bfnn -8 +9 Sat Jan 7 15:12:16 EST 2006 stevenj * [project @ 2006-01-07 20:12:16 by stevenj] enable fma on hppa, update FAQ entry M ./configure.ac +1 M ./doc/FAQ/fftw-faq.bfnn -3 +11 Sat Aug 12 23:47:06 EDT 2006 Unknown tagger tagged fftw-3-1-beta1 Sat Jan 7 14:06:31 EST 2006 athena * [project @ 2006-01-07 19:06:31 by athena] Accomodate different semantics of 'const' in C and C++ M ./dft/simd/t.c +4 Fri Jan 6 23:40:53 EST 2006 athena * [project @ 2006-01-07 04:40:53 by athena] Altivec is called VMX in IBM land. M ./NEWS -1 +1 Fri Jan 6 23:40:16 EST 2006 athena * [project @ 2006-01-07 04:40:16 by athena] Noted faster altivec support. M ./NEWS +2 Fri Jan 6 21:49:10 EST 2006 stevenj * [project @ 2006-01-07 02:49:10 by stevenj] updated icc flag detection M ./m4/ax_cc_maxopt.m4 -1 +1 Fri Jan 6 10:01:50 EST 2006 athena * [project @ 2006-01-06 15:01:50 by athena] Note ``memoize triggen''. M ./TODO -2 +1 Fri Jan 6 09:36:51 EST 2006 athena * [project @ 2006-01-06 14:36:51 by athena] Use --enable-threads to generate dependencies in the threads/ directory. M ./mkdist.sh -1 +1 Fri Jan 6 09:26:29 EST 2006 athena * [project @ 2006-01-06 14:26:29 by athena] Workaround to icc #defining __GNUC__. M ./kernel/ifftw.h -14 +14 Fri Jan 6 09:21:19 EST 2006 athena * [project @ 2006-01-06 14:21:19 by athena] Switched name to 3.1-beta1. M ./configure.ac -1 +1 Thu Jan 5 23:08:44 EST 2006 athena * [project @ 2006-01-06 04:08:44 by athena] More thoughts. M ./TODO +12 Thu Jan 5 22:30:51 EST 2006 athena * [project @ 2006-01-06 03:30:51 by athena] Note wish that (block_size % 4) == 0. M ./TODO +3 Thu Jan 5 22:19:09 EST 2006 athena * [project @ 2006-01-06 03:19:09 by athena] Check alignment of mstart, mcount in SIMD codelets. M ./dft/codelet-dft.h -2 +3 M ./dft/codelets/t.c -2 +3 M ./dft/ctsq.c -2 +3 M ./dft/dftw-direct.c -11 +17 M ./dft/k7/k7.c -3 +3 M ./dft/simd/q1b.c -1 +3 M ./dft/simd/q1f.c -1 +3 M ./dft/simd/t.c -9 +11 M ./dft/simd/t1s.c -3 +5 M ./threads/ct.c -3 +1 M ./threads/hc2hc.c -2 Thu Jan 5 21:56:19 EST 2006 athena * [project @ 2006-01-06 02:56:19 by athena] Enable threads at bootstrap time, so I get the compiler warnings that I would otherwise ignore. M ./bootstrap.sh -1 +1 Thu Jan 5 18:23:15 EST 2006 athena * [project @ 2006-01-05 23:23:15 by athena] made compilable by c++ M ./threads/dft-vrank-geq1.c -2 +2 M ./threads/rdft-vrank-geq1.c -2 +2 M ./threads/vrank-geq1-rdft2.c -2 +2 Thu Jan 5 17:39:02 EST 2006 athena * [project @ 2006-01-05 22:39:02 by athena] FIXED: incorrect twiddle_shift() M ./kernel/twiddle.c -13 +19 Thu Jan 5 16:01:51 EST 2006 athena * [project @ 2006-01-05 21:01:51 by athena] Replaced remnants of awake flag with the new enum wakefulness type. M ./reodft/redft00e-r2hc.c -4 +5 M ./reodft/reodft11e-r2hc.c -5 +7 M ./reodft/rodft00e-r2hc.c -4 +5 M ./threads/ct.c -4 +4 M ./threads/dft-vrank-geq1.c -3 +3 M ./threads/hc2hc.c -3 +3 M ./threads/rdft-vrank-geq1.c -3 +3 M ./threads/vrank-geq1-rdft2.c -3 +3 Thu Jan 5 11:20:59 EST 2006 athena * [project @ 2006-01-05 16:20:59 by athena] Oops---there is no need to find a free slot. M ./kernel/planner.c -10 +1 Thu Jan 5 09:41:58 EST 2006 athena * [project @ 2006-01-05 14:41:58 by athena] Assertions. M ./kernel/planner.c -1 +3 Thu Jan 5 09:29:55 EST 2006 athena * [project @ 2006-01-05 14:29:55 by athena] Commented the hash table lookup algorithm. M ./kernel/planner.c -10 +14 Thu Jan 5 09:12:00 EST 2006 athena * [project @ 2006-01-05 14:12:00 by athena] Fixed infinite loop in hashtable lookup/insert. Grrr... M ./kernel/planner.c -12 +31 Wed Jan 4 22:04:28 EST 2006 stevenj * [project @ 2006-01-05 03:04:26 by stevenj] updated copyright years to 2006 M ./COPYRIGHT -2 +2 M ./api/api.h -2 +2 M ./api/apiplan.c -2 +2 M ./api/configure.c -2 +2 M ./api/execute-dft-c2r.c -2 +2 M ./api/execute-dft-r2c.c -2 +2 M ./api/execute-dft.c -2 +2 M ./api/execute-r2r.c -2 +2 M ./api/execute-split-dft-c2r.c -2 +2 M ./api/execute-split-dft-r2c.c -2 +2 M ./api/execute-split-dft.c -2 +2 M ./api/execute.c -2 +2 M ./api/export-wisdom-to-file.c -2 +2 M ./api/export-wisdom-to-string.c -2 +2 M ./api/export-wisdom.c -2 +2 M ./api/extract-reim.c -2 +2 M ./api/f77api.c -2 +2 M ./api/f77funcs.h -2 +2 M ./api/fftw3.h -3 +3 M ./api/flops.c -2 +2 M ./api/forget-wisdom.c -2 +2 M ./api/import-system-wisdom.c -2 +2 M ./api/import-wisdom-from-file.c -2 +2 M ./api/import-wisdom-from-string.c -2 +2 M ./api/import-wisdom.c -2 +2 M ./api/malloc.c -2 +2 M ./api/map-r2r-kind.c -2 +2 M ./api/mapflags.c -2 +2 M ./api/mkprinter-file.c -2 +2 M ./api/mktensor-iodims.c -2 +2 M ./api/mktensor-rowmajor.c -2 +2 M ./api/plan-dft-1d.c -2 +2 M ./api/plan-dft-2d.c -2 +2 M ./api/plan-dft-3d.c -2 +2 M ./api/plan-dft-c2r-1d.c -2 +2 M ./api/plan-dft-c2r-2d.c -2 +2 M ./api/plan-dft-c2r-3d.c -2 +2 M ./api/plan-dft-c2r.c -2 +2 M ./api/plan-dft-r2c-1d.c -2 +2 M ./api/plan-dft-r2c-2d.c -2 +2 M ./api/plan-dft-r2c-3d.c -2 +2 M ./api/plan-dft-r2c.c -2 +2 M ./api/plan-dft.c -2 +2 M ./api/plan-guru-dft-c2r.c -2 +2 M ./api/plan-guru-dft-r2c.c -2 +2 M ./api/plan-guru-dft.c -2 +2 M ./api/plan-guru-r2r.c -2 +2 M ./api/plan-guru-split-dft-c2r.c -2 +2 M ./api/plan-guru-split-dft-r2c.c -2 +2 M ./api/plan-guru-split-dft.c -2 +2 M ./api/plan-many-dft-c2r.c -2 +2 M ./api/plan-many-dft-r2c.c -2 +2 M ./api/plan-many-dft.c -2 +2 M ./api/plan-many-r2r.c -2 +2 M ./api/plan-r2r-1d.c -2 +2 M ./api/plan-r2r-2d.c -2 +2 M ./api/plan-r2r-3d.c -2 +2 M ./api/plan-r2r.c -2 +2 M ./api/print-plan.c -2 +2 M ./api/rdft2-pad.c -2 +2 M ./api/the-planner.c -2 +2 M ./api/version.c -3 +3 M ./api/x77.h -2 +2 M ./dft/bluestein.c -2 +2 M ./dft/buffered.c -3 +3 M ./dft/codelet-dft.h -3 +3 M ./dft/codelets/n.c -2 +2 M ./dft/codelets/n.h -2 +2 M ./dft/codelets/t.c -2 +2 M ./dft/codelets/t.h -2 +2 M ./dft/conf.c -3 +3 M ./dft/ct.c -3 +3 M ./dft/ct.h -2 +2 M ./dft/ctsq.c -3 +3 M ./dft/dft.h -3 +3 M ./dft/dftw-direct.c -3 +3 M ./dft/dftw-generic.c -2 +2 M ./dft/dftw-genericbuf.c -2 +2 M ./dft/direct.c -3 +3 M ./dft/generic.c -2 +2 M ./dft/indirect-transpose.c -2 +2 M ./dft/indirect.c -3 +3 M ./dft/k7/k7.c -3 +3 M ./dft/kdft-dif.c -3 +3 M ./dft/kdft-difsq.c -3 +3 M ./dft/kdft-dit.c -3 +3 M ./dft/kdft.c -3 +3 M ./dft/nop.c -3 +3 M ./dft/plan.c -3 +3 M ./dft/problem.c -3 +3 M ./dft/rader.c -2 +2 M ./dft/rank-geq2.c -3 +3 M ./dft/simd/n1b.c -2 +2 M ./dft/simd/n1b.h -2 +2 M ./dft/simd/n1f.c -2 +2 M ./dft/simd/n1f.h -2 +2 M ./dft/simd/n2b.c -2 +2 M ./dft/simd/n2b.h -2 +2 M ./dft/simd/n2f.c -2 +2 M ./dft/simd/n2f.h -2 +2 M ./dft/simd/n2s.c -2 +2 M ./dft/simd/n2s.h -2 +2 M ./dft/simd/q1b.c -2 +2 M ./dft/simd/q1b.h -2 +2 M ./dft/simd/q1f.c -2 +2 M ./dft/simd/q1f.h -2 +2 M ./dft/simd/t.c -2 +2 M ./dft/simd/t1b.h -2 +2 M ./dft/simd/t1f.h -2 +2 M ./dft/simd/t1s.c -2 +2 M ./dft/simd/t1s.h -2 +2 M ./dft/simd/t2b.h -2 +2 M ./dft/simd/t2f.h -2 +2 M ./dft/solve.c -3 +3 M ./dft/vrank-geq1.c -3 +3 M ./dft/zero.c -3 +3 M ./doc/f77_wisdom.f -2 +2 M ./doc/fftw3.texi -1 +1 M ./genfft/algsimp.ml -4 +4 M ./genfft/algsimp.mli -3 +3 M ./genfft/annotate.ml -4 +4 M ./genfft/annotate.mli -3 +3 M ./genfft/assoctable.ml -3 +3 M ./genfft/assoctable.mli -3 +3 M ./genfft/c.ml -3 +3 M ./genfft/c.mli -3 +3 M ./genfft/complex.ml -3 +3 M ./genfft/complex.mli -3 +3 M ./genfft/conv.ml -2 +2 M ./genfft/conv.mli -3 +3 M ./genfft/dag.ml -3 +3 M ./genfft/dag.mli -3 +3 M ./genfft/expr.ml -3 +3 M ./genfft/expr.mli -3 +3 M ./genfft/fft.ml -4 +4 M ./genfft/fft.mli -3 +3 M ./genfft/gen_athnotw.ml -4 +4 M ./genfft/gen_athtw.ml -4 +4 M ./genfft/gen_conv.ml -4 +4 M ./genfft/gen_hc2hc.ml -4 +4 M ./genfft/gen_hc2r.ml -4 +4 M ./genfft/gen_mdct.ml -4 +4 M ./genfft/gen_notw.ml -4 +4 M ./genfft/gen_notw_c.ml -4 +4 M ./genfft/gen_r2hc.ml -4 +4 M ./genfft/gen_r2r.ml -4 +4 M ./genfft/gen_twiddle.ml -4 +4 M ./genfft/gen_twiddle_c.ml -4 +4 M ./genfft/gen_twidsq.ml -4 +4 M ./genfft/gen_twidsq_c.ml -4 +4 M ./genfft/genutil.ml -3 +3 M ./genfft/littlesimp.ml -3 +3 M ./genfft/littlesimp.mli -3 +3 M ./genfft/magic.ml -3 +3 M ./genfft/monads.ml -3 +3 M ./genfft/number.ml -3 +3 M ./genfft/number.mli -3 +3 M ./genfft/oracle.ml -3 +3 M ./genfft/oracle.mli -3 +3 M ./genfft/schedule.ml -3 +3 M ./genfft/schedule.mli -3 +3 M ./genfft/simd.ml -3 +3 M ./genfft/simd.mli -3 +3 M ./genfft/simdmagic.ml -3 +3 M ./genfft/to_alist.ml -3 +3 M ./genfft/to_alist.mli -3 +3 M ./genfft/trig.ml -3 +3 M ./genfft/trig.mli -3 +3 M ./genfft/twiddle.ml -3 +3 M ./genfft/twiddle.mli -3 +3 M ./genfft/unique.ml -3 +3 M ./genfft/unique.mli -3 +3 M ./genfft/util.ml -3 +3 M ./genfft/util.mli -3 +3 M ./genfft/variable.ml -3 +3 M ./genfft/variable.mli -3 +3 M ./genfft-k7/algsimp.ml -4 +4 M ./genfft-k7/algsimp.mli -3 +3 M ./genfft-k7/assoctable.ml -3 +3 M ./genfft-k7/assoctable.mli -3 +3 M ./genfft-k7/complex.ml -1 +1 M ./genfft-k7/complex.mli -1 +1 M ./genfft-k7/expr.ml -3 +3 M ./genfft-k7/expr.mli -3 +3 M ./genfft-k7/fft.ml -4 +4 M ./genfft-k7/gen_notw.ml -1 +1 M ./genfft-k7/littlesimp.ml -3 +3 M ./genfft-k7/littlesimp.mli -3 +3 M ./genfft-k7/monads.ml -3 +3 M ./genfft-k7/number.ml -3 +3 M ./genfft-k7/number.mli -3 +3 M ./genfft-k7/oracle.ml -3 +3 M ./genfft-k7/oracle.mli -3 +3 M ./genfft-k7/to_alist.ml -3 +3 M ./genfft-k7/to_alist.mli -3 +3 M ./genfft-k7/twiddle.ml -3 +3 M ./genfft-k7/twiddle.mli -3 +3 M ./genfft-k7/vScheduler.mli -1 +1 M ./kernel/align.c -3 +3 M ./kernel/alloc.c -3 +3 M ./kernel/assert.c -3 +3 M ./kernel/awake.c -3 +3 M ./kernel/buffered.c -2 +2 M ./kernel/cpy1d.c -2 +2 M ./kernel/cpy2d-pair.c -2 +2 M ./kernel/cpy2d.c -2 +2 M ./kernel/ct.c -2 +2 M ./kernel/cycle.h -3 +3 M ./kernel/debug.c -3 +3 M ./kernel/hash.c -2 +2 M ./kernel/iabs.c -3 +3 M ./kernel/ifftw.h -3 +3 M ./kernel/kalloc.c -3 +3 M ./kernel/md5-1.c -2 +2 M ./kernel/md5.c -2 +2 M ./kernel/minmax.c -3 +3 M ./kernel/ops.c -3 +3 M ./kernel/pickdim.c -3 +3 M ./kernel/plan.c -3 +3 M ./kernel/primes.c -3 +3 M ./kernel/print.c -3 +3 M ./kernel/problem.c -3 +3 M ./kernel/rader.c -2 +2 M ./kernel/scan.c -3 +3 M ./kernel/solver.c -3 +3 M ./kernel/solvtab.c -3 +3 M ./kernel/stride.c -3 +3 M ./kernel/tensor.c -3 +3 M ./kernel/tensor1.c -3 +3 M ./kernel/tensor2.c -3 +3 M ./kernel/tensor4.c -3 +3 M ./kernel/tensor5.c -3 +3 M ./kernel/tensor7.c -3 +3 M ./kernel/tensor8.c -3 +3 M ./kernel/tensor9.c -3 +3 M ./kernel/tile2d.c -2 +2 M ./kernel/timer.c -3 +3 M ./kernel/transpose.c -2 +2 M ./kernel/trig.c -3 +3 M ./kernel/twiddle.c -3 +3 M ./libbench/accopy-from.c -1 +1 M ./libbench/accopy-to.c -1 +1 M ./libbench/allocate.c -1 +1 M ./libbench/bench-main.c -1 +1 M ./libbench/bench-user.h -1 +1 M ./libbench/bench.h -1 +1 M ./libbench/can-do.c -1 +1 M ./libbench/ccopy-from.c -1 +1 M ./libbench/ccopy-to.c -1 +1 M ./libbench/deallocate.c -1 +1 M ./libbench/getopt-utils.c -1 +1 M ./libbench/info.c -1 +1 M ./libbench/main.c -1 +1 M ./libbench/prime.c -1 +1 M ./libbench/problem.c -1 +1 M ./libbench/report.c -1 +1 M ./libbench/speed.c -1 +1 M ./libbench/timer.c -1 +1 M ./libbench/verify.c -1 +1 M ./libbench/zero.c -1 +1 M ./libbench2/aligned-main.c -1 +1 M ./libbench2/allocate.c -1 +1 M ./libbench2/can-do.c -1 +1 M ./libbench2/dotens2.c -3 +3 M ./libbench2/getopt-utils.c -1 +1 M ./libbench2/info.c -1 +1 M ./libbench2/main.c -1 +1 M ./libbench2/report.c -1 +1 M ./libbench2/tensor.c -1 +1 M ./libbench2/useropt.c -1 +1 M ./libbench2/verify-dft.c -3 +3 M ./libbench2/verify-lib.c -3 +3 M ./libbench2/verify-r2r.c -2 +2 M ./libbench2/verify-rdft2.c -3 +3 M ./libbench2/verify.c -1 +1 M ./libbench2/verify.h -2 +2 M ./libbench2/zero.c -1 +1 M ./m4/ax_gcc_archflag.m4 -1 +1 M ./rdft/buffered.c -3 +3 M ./rdft/buffered2.c -3 +3 M ./rdft/codelet-rdft.h -2 +2 M ./rdft/codelets/hb.h -2 +2 M ./rdft/codelets/hc2r.c -2 +2 M ./rdft/codelets/hc2r.h -2 +2 M ./rdft/codelets/hc2rIII.h -2 +2 M ./rdft/codelets/hf.h -2 +2 M ./rdft/codelets/hfb.c -2 +2 M ./rdft/codelets/r2hc.c -2 +2 M ./rdft/codelets/r2hc.h -2 +2 M ./rdft/codelets/r2hcII.h -2 +2 M ./rdft/codelets/r2r.c -2 +2 M ./rdft/codelets/r2r.h -2 +2 M ./rdft/conf.c -3 +3 M ./rdft/dft-r2hc.c -3 +3 M ./rdft/dht-r2hc.c -3 +3 M ./rdft/dht-rader.c -2 +2 M ./rdft/direct.c -3 +3 M ./rdft/direct2.c -3 +3 M ./rdft/generic.c -2 +2 M ./rdft/hc2hc-common.c -2 +2 M ./rdft/hc2hc-direct.c -3 +3 M ./rdft/hc2hc-directbuf.c -3 +3 M ./rdft/hc2hc-generic.c -2 +2 M ./rdft/hc2hc.c -2 +2 M ./rdft/hc2hc.h -2 +2 M ./rdft/indirect.c -3 +3 M ./rdft/khc2hc.c -3 +3 M ./rdft/khc2r.c -3 +3 M ./rdft/kr2hc.c -3 +3 M ./rdft/kr2r.c -3 +3 M ./rdft/nop.c -3 +3 M ./rdft/nop2.c -3 +3 M ./rdft/plan.c -3 +3 M ./rdft/plan2.c -3 +3 M ./rdft/problem.c -3 +3 M ./rdft/problem2.c -3 +3 M ./rdft/rank-geq2-rdft2.c -3 +3 M ./rdft/rank-geq2.c -3 +3 M ./rdft/rank0-rdft2.c -3 +3 M ./rdft/rank0.c -3 +3 M ./rdft/rdft-dht.c -3 +3 M ./rdft/rdft.h -2 +2 M ./rdft/rdft2-inplace-strides.c -3 +3 M ./rdft/rdft2-radix2.c -3 +3 M ./rdft/rdft2-strides.c -2 +2 M ./rdft/rdft2-tensor-max-index.c -3 +3 M ./rdft/solve.c -3 +3 M ./rdft/solve2.c -3 +3 M ./rdft/vrank-geq1-rdft2.c -3 +3 M ./rdft/vrank-geq1.c -3 +3 M ./rdft/vrank3-transpose.c -3 +3 M ./reodft/conf.c -3 +3 M ./reodft/redft00e-r2hc-pad.c -3 +3 M ./reodft/redft00e-r2hc.c -3 +3 M ./reodft/reodft.h -2 +2 M ./reodft/reodft010e-r2hc.c -3 +3 M ./reodft/reodft11e-r2hc-odd.c -3 +3 M ./reodft/reodft11e-r2hc.c -3 +3 M ./reodft/reodft11e-radix2.c -3 +3 M ./reodft/rodft00e-r2hc-pad.c -3 +3 M ./reodft/rodft00e-r2hc.c -3 +3 M ./simd/altivec.c -3 +3 M ./simd/nonportable/sse.c -3 +3 M ./simd/nonportable/sse2.c -3 +3 M ./simd/simd-altivec.h -2 +2 M ./simd/simd-sse.h -2 +2 M ./simd/simd-sse2.h -2 +2 M ./simd/simd.h -2 +2 M ./simd/sse.c -3 +3 M ./simd/sse2.c -3 +3 M ./simd/taint.c -3 +3 M ./threads/api.c -2 +2 M ./threads/conf.c -3 +3 M ./threads/ct.c -3 +3 M ./threads/dft-vrank-geq1.c -3 +3 M ./threads/f77api.c -2 +2 M ./threads/f77funcs.h -2 +2 M ./threads/hc2hc.c -2 +2 M ./threads/rdft-vrank-geq1.c -3 +3 M ./threads/threads.c -2 +2 M ./threads/threads.h -2 +2 M ./threads/vrank-geq1-rdft2.c -3 +3 M ./tools/fftw-wisdom-to-conf.1 -4 +4 M ./tools/fftw-wisdom-to-conf.in -2 +2 M ./tools/fftw-wisdom.c -2 +2 M ./tools/fftw_wisdom.1.in -4 +4 Wed Jan 4 21:57:23 EST 2006 stevenj * [project @ 2006-01-05 02:57:23 by stevenj] whoops M ./m4/ax_gcc_archflag.m4 -7 +4 Wed Jan 4 21:52:18 EST 2006 stevenj * [project @ 2006-01-05 02:52:18 by stevenj] whoops M ./m4/ax_gcc_archflag.m4 -1 +1 Wed Jan 4 21:51:40 EST 2006 stevenj * [project @ 2006-01-05 02:51:40 by stevenj] more updates for recent pentia/amd M ./m4/ax_gcc_archflag.m4 -5 +12 Wed Jan 4 20:57:47 EST 2006 athena * [project @ 2006-01-05 01:57:47 by athena] Pruned TODO. M ./TODO -6 +2 Wed Jan 4 20:43:41 EST 2006 athena * [project @ 2006-01-05 01:43:41 by athena] Prototype of problem_destroy() M ./libbench2/bench-user.h -1 +2 M ./libbench2/bench.h -2 +1 Wed Jan 4 20:43:13 EST 2006 stevenj * [project @ 2006-01-05 01:43:13 by stevenj] rm obsoleted TODOs M ./TODO -5 Wed Jan 4 20:37:24 EST 2006 athena * [project @ 2006-01-05 01:37:24 by athena] Fallback to 970 if neither -mcpu=power5 nor -mcpu=power4 are supported. M ./m4/ax_gcc_archflag.m4 -3 +3 Wed Jan 4 20:29:07 EST 2006 stevenj * [project @ 2006-01-05 01:29:07 by stevenj] NEWS updates, clarifications, and reorganization M ./NEWS -6 +10 Wed Jan 4 19:54:41 EST 2006 stevenj * [project @ 2006-01-05 00:54:40 by stevenj] remove some compiler warnings, add an assert check, make estimator work properly for nop plans M ./dft/dftw-genericbuf.c +3 M ./kernel/planner.c -1 +2 M ./kernel/trig.c -1 +3 M ./m4/ax_gcc_x86_cpuid.m4 +2 M ./rdft/dft-r2hc.c -1 +2 Tue Jan 3 19:34:04 EST 2006 athena * [project @ 2006-01-04 00:34:03 by athena] Two big changes: 1) revised the twiddle generation machinery, to avoid generating twiddles when measuring, and to use a faster O(sqrt(N)) table when this entails no loss of precision. 2) implemented new ALLOW_PRUNING estimator hack. M ./api/apiplan.c -2 +10 M ./api/fftw3.h -1 +2 M ./api/mapflags.c -1 +4 M ./configure.ac -8 +9 M ./dft/bluestein.c -12 +18 M ./dft/buffered.c -5 +5 M ./dft/ct.c -4 +7 M ./dft/ctsq.c -4 +4 M ./dft/dftw-direct.c -3 +4 M ./dft/dftw-generic.c -131 +19 M ./dft/dftw-genericbuf.c -76 +36 M ./dft/direct.c -1 +2 M ./dft/generic.c -2 +2 M ./dft/indirect-transpose.c -4 +4 M ./dft/indirect.c -4 +4 M ./dft/rader.c -17 +22 M ./dft/rank-geq2.c -4 +4 M ./dft/vrank-geq1.c -3 +3 M ./genfft/twiddle.ml -22 +12 M ./kernel/awake.c -3 +3 M ./kernel/ifftw.h -23 +48 M ./kernel/plan.c -14 +9 M ./kernel/planner.c -1 +8 M ./kernel/timer.c -3 +3 M ./kernel/trig.c -41 +166 M ./kernel/twiddle.c -64 +103 M ./libbench2/bench-main.c -3 +8 M ./libbench2/bench.h -2 +2 M ./libbench2/problem.c -4 +3 M ./libbench2/speed.c -2 +8 M ./rdft/buffered.c -5 +5 M ./rdft/buffered2.c -4 +4 M ./rdft/dft-r2hc.c -3 +3 M ./rdft/dht-r2hc.c -3 +3 M ./rdft/dht-rader.c -16 +23 M ./rdft/direct.c -1 +3 M ./rdft/direct2.c -1 +2 M ./rdft/generic.c -2 +2 M ./rdft/hc2hc-direct.c -5 +7 M ./rdft/hc2hc-directbuf.c -5 +5 M ./rdft/hc2hc-generic.c -6 +7 M ./rdft/hc2hc.c -3 +7 M ./rdft/indirect.c -4 +4 M ./rdft/rank-geq2-rdft2.c -4 +4 M ./rdft/rank-geq2.c -4 +4 M ./rdft/rank0-rdft2.c -3 +3 M ./rdft/rdft-dht.c -3 +3 M ./rdft/rdft2-radix2.c -4 +5 M ./rdft/vrank-geq1-rdft2.c -3 +3 M ./rdft/vrank-geq1.c -3 +3 M ./reodft/redft00e-r2hc-pad.c -4 +4 M ./reodft/reodft00e-splitradix.c -5 +6 M ./reodft/reodft010e-r2hc.c -4 +5 M ./reodft/reodft11e-r2hc-odd.c -3 +3 M ./reodft/reodft11e-radix2.c -5 +7 M ./reodft/rodft00e-r2hc-pad.c -4 +4 M ./tests/hook.c -3 +6 Sat Dec 24 22:08:29 EST 2005 athena * [project @ 2005-12-25 03:08:29 by athena] Estimator tweaks, mostly to favor generic over rader for small n. M ./dft/generic.c +2 M ./rdft/generic.c +2 Sat Dec 24 17:55:47 EST 2005 athena * [project @ 2005-12-24 22:55:47 by athena] Grrr... missing break statement in switch. M ./tests/hook.c +1 Sat Dec 24 16:08:50 EST 2005 athena * [project @ 2005-12-24 21:08:49 by athena] Swapped fields TW and OPS in struct ct_desc_s, to make k7 asm code insensitive to -malign-double. For consistency, changed struct hc2hc_desc_s in the same way. M ./dft/codelet-dft.h -2 +2 M ./genfft/gen_hc2hc.ml -3 +3 M ./genfft/gen_twiddle.ml -3 +3 M ./genfft/gen_twiddle_c.ml -3 +3 M ./genfft/gen_twidsq.ml -3 +3 M ./genfft/gen_twidsq_c.ml -3 +3 M ./genfft-k7/gen_twiddle.ml -3 +2 M ./rdft/codelet-rdft.h -1 +1 Sat Dec 24 16:00:42 EST 2005 athena * [project @ 2005-12-24 21:00:42 by athena] Wrong check for infeasible slvndx in imprt(). M ./kernel/planner.c -4 +4 Sat Dec 24 15:56:59 EST 2005 athena * [project @ 2005-12-24 20:56:59 by athena] Removed obsolete function invoke_solver_if_correct_kind(). M ./kernel/planner.c -10 +1 Sat Dec 24 14:22:12 EST 2005 athena * [project @ 2005-12-24 19:22:12 by athena] Faster implementation of safe_mulmod(), avoiding divisions altogether. Works for 0 <= p <= INT_MAX. M ./kernel/primes.c -13 +22 Sat Dec 24 12:05:54 EST 2005 athena * [project @ 2005-12-24 17:05:54 by athena] FFTW_ALLOW_LARGE_GENERIC must belong to flags->l, it cannot be overridden by fftw. M ./api/mapflags.c -2 +2 Fri Dec 23 20:46:24 EST 2005 stevenj * [project @ 2005-12-24 01:46:24 by stevenj] no more need for limits.h, add some explanatory comments M ./kernel/primes.c -4 +9 Fri Dec 23 17:50:25 EST 2005 athena * [project @ 2005-12-23 22:50:25 by athena] Paranoia. M ./dft/k7/k7.c -1 +8 Fri Dec 23 17:40:41 EST 2005 athena * [project @ 2005-12-23 22:40:41 by athena] Fixed subtle bug involving overflow of the slvndx field in flags_t. M ./kernel/ifftw.h -5 +6 M ./kernel/planner.c -18 +26 Fri Dec 23 16:33:56 EST 2005 athena * [project @ 2005-12-23 21:33:56 by athena] Note 64-bit clean. M ./NEWS +2 Fri Dec 23 15:34:32 EST 2005 athena * [project @ 2005-12-23 20:34:32 by athena] Threads are now 64-bit clean M ./threads/ct.c -12 +13 M ./threads/dft-vrank-geq1.c -20 +17 M ./threads/hc2hc.c -7 +8 M ./threads/rdft-vrank-geq1.c -18 +15 M ./threads/threads.h -2 +2 M ./threads/vrank-geq1-rdft2.c -20 +20 Fri Dec 23 13:00:31 EST 2005 athena * [project @ 2005-12-23 18:00:31 by athena] Restored the old numbering TW_NEXT=3 etc, because the k7 code depends on it. M ./kernel/ifftw.h -3 +3 Fri Dec 23 11:58:00 EST 2005 athena * [project @ 2005-12-23 16:58:00 by athena] Portable implementation of MULMOD() and safe_mulmod(). Removed all unnecessary AC_CHECK_SIZEOF() from configure.ac. M ./configure.ac -5 M ./kernel/ifftw.h -27 +3 M ./kernel/primes.c -8 +3 Thu Dec 22 11:12:29 EST 2005 athena * [project @ 2005-12-22 16:12:29 by athena] Inline the loop body in r2r codelets like we do everywhere else. M ./genfft/gen_r2r.ml -44 +34 Thu Dec 22 10:48:53 EST 2005 athena * [project @ 2005-12-22 15:48:53 by athena] Oops. M ./dft/conf.c -3 +3 Thu Dec 22 10:25:15 EST 2005 athena * [project @ 2005-12-22 15:25:15 by athena] Renamed X(sin_and_cos)() to X(cexp)(). M ./dft/bluestein.c -1 +1 M ./dft/dftw-generic.c -2 +2 M ./dft/dftw-genericbuf.c -2 +2 M ./dft/rader.c -1 +1 M ./kernel/ifftw.h -2 +2 M ./kernel/trig.c -2 +2 M ./kernel/twiddle.c -4 +3 M ./rdft/dht-rader.c -1 +1 Wed Dec 21 22:49:58 EST 2005 athena * [project @ 2005-12-22 03:49:58 by athena] Somewhat faster generation of twiddle factors. M ./dft/bluestein.c -2 +1 M ./dft/conf.c -3 +3 M ./dft/dftw-generic.c -8 +5 M ./dft/dftw-genericbuf.c -8 +6 M ./dft/rader.c -3 +4 M ./kernel/Makefile.am -1 +1 M ./kernel/ifftw.h -4 +3 M ./kernel/trig.c -5 +67 R ./kernel/trig1.c M ./kernel/twiddle.c -11 +15 M ./rdft/dht-rader.c -2 +4 Tue Dec 20 23:50:01 EST 2005 athena * [project @ 2005-12-21 04:50:01 by athena] tweaks M ./kernel/md5.c -2 +1 Tue Dec 20 22:29:19 EST 2005 athena * [project @ 2005-12-21 03:29:19 by athena] Sped up planner, esp. in estimate mode. The planner now classifies all solvers into DFT, RDFT, and RDFT2, and it only invokes solvers appropriate for the problem being planned. Because we have several hundred solvers, the overhead of calling irrelevant solvers is significant, and this modification mitigates the issue somewhat. M ./dft/bluestein.c -14 +10 M ./dft/buffered.c -31 +30 M ./dft/ct.c -16 +13 M ./dft/ctsq.c -23 +20 M ./dft/dft.h -4 +1 M ./dft/direct.c -69 +62 M ./dft/generic.c -12 +8 M ./dft/indirect-transpose.c -23 +19 M ./dft/indirect.c -37 +33 M ./dft/nop.c -16 +15 M ./dft/problem.c -6 +2 M ./dft/rader.c -11 +7 M ./dft/rank-geq2.c -13 +9 M ./dft/vrank-geq1.c -16 +12 M ./kernel/ifftw.h -1 +25 M ./kernel/planner.c -5 +29 M ./rdft/buffered.c -31 +30 M ./rdft/buffered2.c -8 +5 M ./rdft/dft-r2hc.c -10 +6 M ./rdft/dht-r2hc.c -12 +9 M ./rdft/dht-rader.c -13 +9 M ./rdft/direct.c -49 +45 M ./rdft/direct2.c -45 +41 M ./rdft/generic.c -13 +9 M ./rdft/hc2hc.c -19 +15 M ./rdft/indirect.c -30 +26 M ./rdft/nop.c -16 +13 M ./rdft/nop2.c -17 +15 M ./rdft/problem.c -6 +2 M ./rdft/problem2.c -6 +2 M ./rdft/rank-geq2-rdft2.c -20 +16 M ./rdft/rank-geq2.c -13 +9 M ./rdft/rank0-rdft2.c -13 +10 M ./rdft/rank0.c -14 +10 M ./rdft/rdft-dht.c -15 +13 M ./rdft/rdft.h -6 M ./rdft/rdft2-radix2.c -25 +17 M ./rdft/vrank-geq1-rdft2.c -13 +12 M ./rdft/vrank-geq1.c -15 +11 M ./rdft/vrank3-transpose.c -27 +24 M ./reodft/redft00e-r2hc-pad.c -12 +9 M ./reodft/redft00e-r2hc.c -12 +9 M ./reodft/reodft00e-splitradix.c -17 +14 M ./reodft/reodft010e-r2hc.c -12 +9 M ./reodft/reodft11e-r2hc-odd.c -12 +9 M ./reodft/reodft11e-r2hc.c -11 +9 M ./reodft/reodft11e-radix2.c -12 +9 M ./reodft/rodft00e-r2hc-pad.c -12 +8 M ./reodft/rodft00e-r2hc.c -11 +8 M ./tests/hook.c -92 +98 Mon Dec 19 22:04:00 EST 2005 athena * [project @ 2005-12-20 03:04:00 by athena] Eliminated all calls to sprintf() in favor of own routines, so as not to force users to link stdio and the associated locale/pthreads crap. M ./kernel/print.c -35 +50 Mon Dec 19 21:27:25 EST 2005 athena * [project @ 2005-12-20 02:27:25 by athena] Implemented routine to print INT, removing the need for c99's %td format. M ./kernel/ifftw.h -2 +1 M ./kernel/print.c -7 +27 Mon Dec 19 12:06:33 EST 2005 athena * [project @ 2005-12-19 17:06:33 by athena] info->n is size_t M ./kernel/alloc.c -2 +2 Sun Dec 18 18:15:04 EST 2005 athena * [project @ 2005-12-18 23:15:04 by athena] Explicit casts in front of pointer difference in printf() context, just in case INT != ptrdiff_t. M ./configure.ac -1 +1 M ./dft/problem.c -3 +3 M ./rdft/problem.c -2 +2 M ./rdft/problem2.c -3 +3 Sun Dec 18 16:52:38 EST 2005 athena * [project @ 2005-12-18 21:52:38 by athena] Forgot to add %D to print.c M ./kernel/print.c -1 +6 Sun Dec 18 16:43:26 EST 2005 athena * [project @ 2005-12-18 21:43:26 by athena] Use %D as format character for type INT. M ./dft/bluestein.c -1 +1 M ./dft/buffered.c -2 +2 M ./dft/ct.c -2 +2 M ./dft/ctsq.c -2 +2 M ./dft/dftw-direct.c -3 +3 M ./dft/dftw-generic.c -1 +1 M ./dft/dftw-genericbuf.c -1 +1 M ./dft/direct.c -3 +3 M ./dft/generic.c -1 +1 M ./dft/problem.c -2 +2 M ./dft/rader.c -1 +1 M ./dft/vrank-geq1.c -2 +2 M ./kernel/print.c -9 +1 M ./kernel/tensor.c -2 +2 M ./rdft/buffered.c -2 +2 M ./rdft/buffered2.c -2 +2 M ./rdft/dft-r2hc.c -2 +2 M ./rdft/dht-r2hc.c -2 +2 M ./rdft/dht-rader.c -1 +1 M ./rdft/direct.c -2 +2 M ./rdft/direct2.c -2 +2 M ./rdft/generic.c -1 +1 M ./rdft/hc2hc-direct.c -2 +2 M ./rdft/hc2hc-directbuf.c -2 +2 M ./rdft/hc2hc-generic.c -1 +1 M ./rdft/hc2hc.c -1 +1 M ./rdft/problem.c -2 +2 M ./rdft/problem2.c -2 +2 M ./rdft/rank0.c -2 +2 M ./rdft/rdft-dht.c -2 +2 M ./rdft/rdft2-radix2.c -2 +2 M ./rdft/vrank-geq1-rdft2.c -2 +2 M ./rdft/vrank-geq1.c -2 +2 M ./rdft/vrank3-transpose.c -2 +2 M ./reodft/redft00e-r2hc-pad.c -2 +2 M ./reodft/redft00e-r2hc.c -2 +2 M ./reodft/reodft00e-splitradix.c -3 +3 M ./reodft/reodft010e-r2hc.c -2 +2 M ./reodft/reodft11e-r2hc-odd.c -2 +2 M ./reodft/reodft11e-r2hc.c -2 +2 M ./reodft/reodft11e-radix2.c -2 +2 M ./reodft/rodft00e-r2hc-pad.c -2 +2 M ./reodft/rodft00e-r2hc.c -2 +2 Sun Dec 18 15:14:03 EST 2005 athena * [project @ 2005-12-18 20:14:03 by athena] Changed type of an_int_guaranteed_to_be_zero. Changed name as well. M ./kernel/ifftw.h -3 +3 M ./kernel/stride.c -2 +2 Sun Dec 18 14:41:31 EST 2005 athena * [project @ 2005-12-18 19:41:31 by athena] converted %o -> INT M ./kernel/ifftw.h -1 +2 M ./kernel/planner.c -2 +2 M ./kernel/print.c -8 +6 Sat Dec 17 20:28:50 EST 2005 athena * [project @ 2005-12-18 01:28:50 by athena] Major 64-bit cleanup. M ./dft/bluestein.c -14 +15 M ./dft/buffered.c -20 +20 M ./dft/codelet-dft.h -17 +17 M ./dft/codelets/n.c -1 +1 M ./dft/codelets/t.c -1 +1 M ./dft/ct.c -6 +6 M ./dft/ct.h -5 +5 M ./dft/ctsq.c -6 +6 M ./dft/dftw-direct.c -17 +17 M ./dft/dftw-generic.c -28 +28 M ./dft/dftw-genericbuf.c -25 +25 M ./dft/direct.c -16 +14 M ./dft/generic.c -8 +8 M ./dft/indirect-transpose.c -5 +5 M ./dft/problem.c -3 +3 M ./dft/rader.c -10 +10 M ./dft/simd/n1b.c -1 +1 M ./dft/simd/n1f.c -1 +1 M ./dft/simd/n2b.c -1 +1 M ./dft/simd/n2f.c -1 +1 M ./dft/simd/n2s.c -1 +1 M ./dft/simd/q1b.c -1 +1 M ./dft/simd/q1f.c -1 +1 M ./dft/simd/t.c -5 +5 M ./dft/simd/t1s.c -1 +1 M ./dft/vrank-geq1.c -5 +5 M ./dft/zero.c -3 +3 M ./genfft/gen_hc2hc.ml -5 +5 M ./genfft/gen_hc2r.ml -6 +6 M ./genfft/gen_notw.ml -6 +6 M ./genfft/gen_notw_c.ml -6 +6 M ./genfft/gen_r2hc.ml -6 +6 M ./genfft/gen_r2r.ml -6 +6 M ./genfft/gen_twiddle.ml -5 +5 M ./genfft/gen_twiddle_c.ml -5 +5 M ./genfft/gen_twidsq.ml -5 +5 M ./genfft/gen_twidsq_c.ml -5 +5 M ./kernel/buffered.c -2 +2 M ./kernel/cpy1d.c -2 +2 M ./kernel/cpy2d-pair.c -9 +7 M ./kernel/cpy2d.c -20 +19 M ./kernel/ct.c -1 +1 M ./kernel/iabs.c -3 +3 M ./kernel/ifftw.h -81 +99 M ./kernel/md5-1.c -5 +5 M ./kernel/minmax.c -3 +3 M ./kernel/ops.c -4 +4 M ./kernel/planner.c -2 +2 M ./kernel/primes.c -27 +29 M ./kernel/rader.c -3 +3 M ./kernel/solvtab.c -1 +2 M ./kernel/stride.c -3 +3 M ./kernel/tensor.c -7 +8 M ./kernel/tensor1.c -2 +2 M ./kernel/tensor2.c -3 +21 M ./kernel/tensor4.c -8 +8 M ./kernel/tensor7.c -10 +16 M ./kernel/tile2d.c -31 +8 M ./kernel/transpose.c -12 +12 M ./kernel/trig.c -13 +3 M ./kernel/twiddle.c -23 +23 M ./rdft/buffered.c -15 +15 M ./rdft/buffered2.c -30 +30 M ./rdft/codelet-rdft.h -29 +29 M ./rdft/codelets/hc2r.c -1 +1 M ./rdft/codelets/hfb.c -1 +1 M ./rdft/codelets/r2hc.c -1 +1 M ./rdft/codelets/r2r.c -1 +1 M ./rdft/dft-r2hc.c -12 +12 M ./rdft/dht-r2hc.c -5 +5 M ./rdft/dht-rader.c -12 +12 M ./rdft/direct.c -11 +11 M ./rdft/direct2.c -8 +8 M ./rdft/generic.c -14 +14 M ./rdft/hc2hc-common.c -3 +3 M ./rdft/hc2hc-direct.c -13 +13 M ./rdft/hc2hc-directbuf.c -24 +24 M ./rdft/hc2hc-generic.c -48 +31 M ./rdft/hc2hc.c -5 +5 M ./rdft/hc2hc.h -7 +7 M ./rdft/problem.c -3 +3 M ./rdft/problem2.c -2 +2 M ./rdft/rank0-rdft2.c -7 +7 M ./rdft/rank0.c -11 +11 M ./rdft/rdft-dht.c -9 +9 M ./rdft/rdft.h -2 +2 M ./rdft/rdft2-inplace-strides.c -3 +3 M ./rdft/rdft2-radix2.c -23 +23 M ./rdft/rdft2-strides.c -1 +1 M ./rdft/rdft2-tensor-max-index.c -4 +4 M ./rdft/vrank-geq1-rdft2.c -6 +6 M ./rdft/vrank-geq1.c -5 +5 M ./rdft/vrank3-transpose.c -53 +54 M ./reodft/redft00e-r2hc-pad.c -11 +11 M ./reodft/redft00e-r2hc.c -11 +11 M ./reodft/reodft00e-splitradix.c -16 +17 M ./reodft/reodft010e-r2hc.c -26 +26 M ./reodft/reodft11e-r2hc-odd.c -19 +19 M ./reodft/reodft11e-r2hc.c -16 +16 M ./reodft/reodft11e-radix2.c -25 +25 M ./reodft/rodft00e-r2hc-pad.c -11 +11 M ./reodft/rodft00e-r2hc.c -11 +11 M ./simd/sse2.c -2 +3 M ./simd/taint.c -2 +2 Sat Aug 12 23:34:43 EDT 2006 Unknown tagger tagged before-64bit-rewrite Wed Dec 7 22:39:01 EST 2005 stevenj * [project @ 2005-12-08 03:39:01 by stevenj] PGI x86-64 cycle counter, courtesy Cristiano Calonaci M ./kernel/cycle.h -1 +12 Mon Dec 5 21:25:57 EST 2005 athena * [project @ 2005-12-06 02:25:57 by athena] Must insert into hash table when wisdom_state == WISDOM_ONLY, otherwise wisdom does not work. M ./kernel/planner.c -2 +3 Sat Oct 8 18:08:44 EDT 2005 stevenj * [project @ 2005-10-08 22:07:37 by stevenj] comment M ./m4/acx_pthread.m4 -1 +2 Sun Oct 2 11:49:13 EDT 2005 athena * [project @ 2005-10-02 15:49:13 by athena] Paranoia: made planner robust against MD5 collisions. M ./api/apiplan.c -6 +37 M ./kernel/ifftw.h -1 +19 M ./kernel/planner.c -25 +55 Tue Sep 27 22:33:18 EDT 2005 athena * [project @ 2005-09-28 02:33:18 by athena] Note that --enable-3dnow is unsupported. M ./doc/FAQ/fftw-faq.bfnn -2 +2 Tue Sep 27 22:31:04 EDT 2005 athena * [project @ 2005-09-28 02:31:04 by athena] * Removed --enable-3dnow support. * SIMD support for split complex arrays. M ./NEWS +4 Tue Sep 27 22:28:41 EDT 2005 athena * [project @ 2005-09-28 02:28:40 by athena] Removed --enabled-3dnow, since it is becoming useless as the world moves to x86-64, and it is a pain to maintain. (We should probably remove the k7 stuff as well.) M ./api/version.c -5 +1 M ./configure.ac -9 M ./genfft/gen_notw.ml -6 +4 M ./genfft/gen_notw_c.ml -4 +2 M ./genfft/gen_twiddle.ml -5 +3 M ./genfft/gen_twiddle_c.ml -4 +2 M ./genfft/gen_twidsq_c.ml -4 +2 M ./kernel/align.c -4 +2 M ./kernel/ifftw.h -2 +2 R ./simd/3dnow.c M ./simd/Makefile.am -2 +2 R ./simd/simd-3dnow.h M ./simd/simd-altivec.h -3 M ./simd/simd-sse.h -3 M ./simd/simd-sse2.h -3 M ./simd/simd.h -4 Tue Sep 27 21:59:16 EDT 2005 athena * [project @ 2005-09-28 01:59:16 by athena] Missing BEGIN_SIMD(), END_SIMD() statements. M ./genfft/gen_notw.ml -4 +6 M ./genfft/gen_twiddle.ml -3 +5 Tue Sep 27 12:16:08 EDT 2005 athena * [project @ 2005-09-27 16:16:08 by athena] Tweaks M ./simd/simd-sse.h -12 +9 Tue Sep 27 10:04:32 EDT 2005 athena * [project @ 2005-09-27 14:04:32 by athena] Fixed wrong opcount for simd codelets. M ./genfft/to_alist.ml -9 +9 Tue Sep 27 09:25:50 EDT 2005 athena * [project @ 2005-09-27 13:25:50 by athena] Fixed wrong opcount for simd codelets. M ./dft/dftw-direct.c -2 +2 Mon Sep 26 22:58:19 EDT 2005 athena * [project @ 2005-09-27 02:58:19 by athena] fixed flop counts M ./genfft/c.ml -23 +17 M ./simd/simd-altivec.h -2 +2 M ./simd/simd-sse2.h -9 +9 Mon Sep 26 22:34:40 EDT 2005 athena * [project @ 2005-09-27 02:34:40 by athena] Silence warnings M ./simd/simd-sse2.h +6 Mon Sep 26 20:52:36 EDT 2005 athena * [project @ 2005-09-27 00:52:36 by athena] Implemented split-complex SIMD codelets A ./dft/simd/n2s.c A ./dft/simd/n2s.h A ./dft/simd/t1s.c A ./dft/simd/t1s.h M ./dft/simd/Makefile.am -2 +3 M ./dft/simd/codelets/Makefile.am -6 +18 M ./dft/simd/n2s.c +50 M ./dft/simd/n2s.h +28 M ./dft/simd/t1s.c +48 M ./dft/simd/t1s.h +33 M ./genfft/annotate.ml -3 +3 M ./genfft/c.ml -1 +3 M ./genfft/gen_hc2hc.ml -3 +3 M ./genfft/gen_notw.ml -5 +26 M ./genfft/gen_twiddle.ml -4 +4 M ./genfft/gen_twiddle_c.ml -3 +3 M ./genfft/gen_twidsq_c.ml -3 +3 M ./genfft/genutil.ml -3 +3 M ./genfft/simd.ml -14 +8 M ./genfft/twiddle.ml -3 +3 M ./genfft/twiddle.mli -2 +2 M ./simd/simd-altivec.h -7 +33 M ./simd/simd-sse.h +33 M ./simd/simd-sse2.h -2 +38 M ./simd/simd.h +2 Sun Sep 25 22:25:35 EDT 2005 athena * [project @ 2005-09-26 02:25:35 by athena] Generalized the ``store pairs'' trick (now called ``store multiple''). M ./dft/simd/codelets/Makefile.am -2 +2 M ./genfft/annotate.ml -37 +25 M ./genfft/annotate.mli -4 +2 M ./genfft/expr.ml -6 +6 M ./genfft/expr.mli -2 +2 M ./genfft/gen_notw_c.ml -11 +13 M ./genfft/simd.ml -6 +10 M ./genfft/simdmagic.ml -3 +3 M ./simd/simd-3dnow.h -2 +2 M ./simd/simd-altivec.h -2 +4 M ./simd/simd-sse.h -3 +3 M ./simd/simd-sse2.h -2 +2 Sun Sep 25 18:58:20 EDT 2005 athena * [project @ 2005-09-25 22:58:20 by athena] Silence some warnings. M ./simd/simd-altivec.h -4 +4 Sat Sep 24 12:37:16 EDT 2005 athena * [project @ 2005-09-24 16:37:16 by athena] Removed obsolete cruft M ./simd/simd-altivec.h -6 +6 Mon Sep 19 22:55:19 EDT 2005 athena * [project @ 2005-09-20 02:55:19 by athena] Re-enabled check for because OSX requires it. M ./configure.ac -1 +1 M ./simd/simd-altivec.h +2 Sun Sep 11 11:03:03 EDT 2005 athena * [project @ 2005-09-11 15:03:03 by athena] Check for sizeof(unsigned int) unconditionally, because the result is used by ifftw.h. M ./configure.ac -2 +2 Sun Sep 11 10:59:40 EDT 2005 athena * [project @ 2005-09-11 14:59:40 by athena] Higher size limit for t2 codelets. M ./dft/simd/t.c -1 +1 Sun Sep 11 10:50:37 EDT 2005 athena * [project @ 2005-09-11 14:50:37 by athena] Heuristic: do not use t2 simd codelets for N>1024. A ./dft/simd/t.c M ./dft/simd/Makefile.am -2 +2 M ./dft/simd/t.c +98 R ./dft/simd/t1b.c R ./dft/simd/t1f.c M ./dft/simd/t2b.h -2 +1 M ./dft/simd/t2f.h -2 +1 Mon Sep 5 22:22:50 EDT 2005 athena * [project @ 2005-09-06 02:22:50 by athena] Larger tolerance in timer calibration routine. M ./libbench2/timer.c -2 +2 Mon Sep 5 16:03:33 EDT 2005 athena * [project @ 2005-09-05 20:03:33 by athena] #include unconditionally. (There is no point in checking.) M ./configure.ac -2 +1 M ./simd/simd-altivec.h -2 Mon Sep 5 15:23:27 EDT 2005 athena * [project @ 2005-09-05 19:23:27 by athena] Removed SSE and SSE2 asm because it was bitrotting. Use the Intel API instead, which seems to be supported by gcc >= 3.3. Moved files that require -msse, -msse2 to new directory. A ./simd/nonportable/ A ./simd/nonportable/Makefile.am A ./simd/nonportable/sse.c A ./simd/nonportable/sse2.c M ./Makefile.am +1 M ./configure.ac +1 M ./simd/Makefile.am -4 +3 M ./simd/nonportable/Makefile.am +8 M ./simd/nonportable/sse.c +43 M ./simd/nonportable/sse2.c +43 M ./simd/simd-sse.h -101 +19 M ./simd/simd-sse2.h -75 +7 R ./simd/sse-aux.c M ./simd/sse.c -5 +7 R ./simd/sse2-aux.c M ./simd/sse2.c -5 +7 Mon Sep 5 12:56:28 EDT 2005 athena * [project @ 2005-09-05 16:56:28 by athena] Parse cputypes of the form 7447A,altivecsupported M ./m4/ax_gcc_archflag.m4 -1 +1 Mon Sep 5 12:52:30 EDT 2005 athena * [project @ 2005-09-05 16:52:30 by athena] Distinguish powerpc 7400 from the 7450, which has a different pipeline. M ./m4/ax_gcc_archflag.m4 +2 Mon Sep 5 12:46:00 EDT 2005 athena * [project @ 2005-09-05 16:46:00 by athena] Paranoia: define RIGHT_CPU unconditionally. M ./simd/simd-altivec.h -3 +3 Thu Aug 11 20:56:41 EDT 2005 athena * [project @ 2005-08-12 00:56:41 by athena] Removed obsolete name fftw-wisdom2c. M ./tools/fftw-wisdom-to-conf.in -1 +1 Thu Aug 11 20:55:59 EDT 2005 athena * [project @ 2005-08-12 00:55:59 by athena] Avoid creation of temporary files---use cpp magic instead. This fix solves a security bug and avoids nonportable tempfile creation hacks. M ./tools/fftw-wisdom-to-conf.in -13 +20 Fri Aug 5 10:03:02 EDT 2005 athena * [project @ 2005-08-05 14:03:02 by athena] Workaround for with gcc-3.3 altivec bug. M ./configure.ac -1 +3 M ./simd/altivec.c -3 +2 M ./simd/simd-altivec.h -2 +2 Wed Jun 15 21:36:46 EDT 2005 stevenj * [project @ 2005-06-16 01:36:46 by stevenj] solaris fix: check -pthreads first since gcc does not like -pthread but chokes due to stubbed libc (grr) M ./m4/acx_pthread.m4 -2 +2 Fri Jun 3 17:19:56 EDT 2005 stevenj * [project @ 2005-06-03 21:19:56 by stevenj] note that VC++ bug was fixed in 2005 M ./doc/FAQ/fftw-faq.bfnn -4 +4 Mon May 30 16:30:45 EDT 2005 stevenj * [project @ 2005-05-30 20:30:32 by stevenj] generalized ax_cc_vendor to ax_compiler_vendor A ./m4/ax_compiler_vendor.m4 M ./configure.ac -2 +2 M ./m4/ax_cc_maxopt.m4 -5 +5 R ./m4/ax_cc_vendor.m4 M ./m4/ax_compiler_vendor.m4 +30 Mon May 30 15:55:07 EDT 2005 stevenj * [project @ 2005-05-30 19:55:07 by stevenj] updated message M ./m4/ax_cc_maxopt.m4 -2 +2 Mon May 30 15:45:14 EDT 2005 stevenj * [project @ 2005-05-30 19:45:14 by stevenj] update for new AC archive format M ./m4/acx_pthread.m4 -1 +4 M ./m4/ax_cc_maxopt.m4 -4 +9 M ./m4/ax_cc_vendor.m4 +6 M ./m4/ax_check_compiler_flags.m4 -1 +4 M ./m4/ax_gcc_aligns_stack.m4 -1 +4 M ./m4/ax_gcc_archflag.m4 -2 +5 M ./m4/ax_gcc_version.m4 -1 +4 M ./m4/ax_gcc_x86_cpuid.m4 -1 +4 M ./m4/ax_openmp.m4 -1 +4 Mon May 23 23:12:22 EDT 2005 stevenj * [project @ 2005-05-24 03:12:22 by stevenj] M ./api/fftw3.h -2 +2 Mon May 23 18:17:38 EDT 2005 stevenj * [project @ 2005-05-23 22:17:38 by stevenj] M ./NEWS -1 +1 Mon May 23 18:13:08 EDT 2005 stevenj * [project @ 2005-05-23 22:13:08 by stevenj] more notes M ./NEWS -1 +9 Sun May 22 23:37:08 EDT 2005 stevenj * [project @ 2005-05-23 03:37:08 by stevenj] whoops M ./m4/ax_cc_maxopt.m4 -8 +8 Sun May 22 22:37:50 EDT 2005 stevenj * [project @ 2005-05-23 02:37:50 by stevenj] note icc 8.x annoyance M ./doc/FAQ/fftw-faq.bfnn -1 +3 Sun May 22 22:36:04 EDT 2005 stevenj * [project @ 2005-05-23 02:36:04 by stevenj] M ./doc/FAQ/fftw-faq.bfnn -1 +1 Sun May 22 22:35:34 EDT 2005 stevenj * [project @ 2005-05-23 02:35:34 by stevenj] note gcc 3.4.[0123] bug, which is fixed in gcc 3.4.4 M ./doc/FAQ/fftw-faq.bfnn +5 Sun May 22 22:21:26 EDT 2005 stevenj * [project @ 2005-05-23 02:21:26 by stevenj] added automatic detection of icc architecture flag M ./m4/ax_cc_maxopt.m4 -1 +34 Sun May 22 21:47:19 EDT 2005 stevenj * [project @ 2005-05-23 01:47:19 by stevenj] add -no-gcc to icc flags...even if it is Intel's fault, I'm sick of dealing with bug reports about this M ./configure.ac +2 Sun May 22 21:40:59 EDT 2005 stevenj * [project @ 2005-05-23 01:40:59 by stevenj] added @cindex portability M ./doc/fftw3.texi -2 +11 Sun May 22 21:34:10 EDT 2005 stevenj * [project @ 2005-05-23 01:34:10 by stevenj] note --without-gcc-arch M ./doc/fftw3.texi -2 +4 Sun May 22 20:54:54 EDT 2005 stevenj * [project @ 2005-05-23 00:54:54 by stevenj] bsd ppc detection; some odd 603 types M ./m4/ax_gcc_archflag.m4 -2 +3 Sun May 22 11:53:20 EDT 2005 stevenj * [project @ 2005-05-22 15:53:20 by stevenj] M ./m4/ax_gcc_archflag.m4 -2 +2 Sat May 21 20:34:52 EDT 2005 stevenj * [project @ 2005-05-22 00:34:52 by stevenj] ensure no spaces in cputype M ./m4/ax_gcc_archflag.m4 -2 +2 Sat May 21 20:31:41 EDT 2005 stevenj * [project @ 2005-05-22 00:31:41 by stevenj] nevermind M ./m4/ax_gcc_archflag.m4 -2 +2 Sat May 21 20:30:08 EDT 2005 stevenj * [project @ 2005-05-22 00:30:08 by stevenj] more bsd stuff M ./m4/ax_gcc_archflag.m4 -2 +2 Sat May 21 20:28:40 EDT 2005 stevenj * [project @ 2005-05-22 00:28:40 by stevenj] added BSD cpu detection for SPARC and better super/hypersparc detection M ./m4/ax_gcc_archflag.m4 -5 +5 Sat May 21 20:22:11 EDT 2005 stevenj * [project @ 2005-05-22 00:22:11 by stevenj] comment M ./m4/ax_gcc_archflag.m4 -1 +4 Fri May 20 19:40:09 EDT 2005 stevenj * [project @ 2005-05-20 23:40:09 by stevenj] "alternate" == "alternative" is US-centric M ./doc/fftw3.texi -3 +3 Fri May 20 19:36:26 EDT 2005 stevenj * [project @ 2005-05-20 23:36:26 by stevenj] typo M ./doc/fftw3.texi -2 +2 Fri May 20 01:28:34 EDT 2005 stevenj * [project @ 2005-05-20 05:28:34 by stevenj] clarification M ./doc/FAQ/fftw-faq.bfnn -2 +2 Tue May 17 18:56:46 EDT 2005 stevenj * [project @ 2005-05-17 22:56:46 by stevenj] print out estimate-planner time from can_do in verbose>2 mode M ./tests/bench.c -1 +7 Mon May 9 00:47:19 EDT 2005 stevenj * [project @ 2005-05-09 04:47:19 by stevenj] comment M ./m4/ax_cc_vendor.m4 +1 Thu May 5 23:47:55 EDT 2005 stevenj * [project @ 2005-05-06 03:47:55 by stevenj] fixes for building Windows DLLs with Cygwin; thanks in part to Stephane Fillod M ./Makefile.am -1 +1 M ./api/api.h +2 M ./api/fftw3.h -2 +12 M ./configure.ac -2 +4 M ./threads/Makefile.am -1 +1 Fri Apr 22 19:47:43 EDT 2005 stevenj * [project @ 2005-04-22 23:47:43 by stevenj] -ffast-math seems to produce code that is either about the same speed or slightly faster (gcc 3.3 and 4.0, x86) M ./m4/ax_cc_maxopt.m4 -1 +4 Fri Apr 22 19:18:23 EDT 2005 stevenj * [project @ 2005-04-22 23:18:23 by stevenj] power5 fallback to power4 sched for older gcc's M ./m4/ax_gcc_archflag.m4 -2 +2 Fri Apr 22 19:14:53 EDT 2005 stevenj * [project @ 2005-04-22 23:14:53 by stevenj] check for power5 M ./m4/ax_gcc_archflag.m4 -1 +2 Tue Apr 19 21:55:13 EDT 2005 athena * [project @ 2005-04-20 01:55:13 by athena] Removed clause #3 M ./api/fftw3.h -5 +1 Tue Apr 19 21:44:57 EDT 2005 stevenj * [project @ 2005-04-20 01:44:57 by stevenj] license clarification M ./api/fftw3.h -1 +4 Tue Apr 19 21:42:51 EDT 2005 athena * [project @ 2005-04-20 01:42:51 by athena] Changed license of fftw3.h to X11. M ./api/fftw3.h -12 +24 Mon Apr 11 13:15:12 EDT 2005 stevenj * [project @ 2005-04-11 17:15:12 by stevenj] delete fixed-input code M ./genfft/gen_conv.ml -3 +2 Sun Apr 10 16:33:24 EDT 2005 athena * [project @ 2005-04-10 20:33:24 by athena] joned L-U-planner branch M ./api/apiplan.c -2 +2 M ./api/fftw3.h -4 +5 M ./api/mapflags.c -19 +35 M ./dft/bluestein.c -6 +7 M ./dft/buffered.c -18 +11 M ./dft/ct.c -7 +5 M ./dft/dftw-direct.c -1 +4 M ./dft/dftw-generic.c -2 +4 M ./dft/generic.c -1 +1 M ./dft/indirect-transpose.c -1 +1 M ./dft/indirect.c -8 +8 M ./dft/rader.c -14 +16 M ./dft/vrank-geq1.c -2 +2 M ./kernel/ifftw.h -49 +71 M ./kernel/planner.c -97 +146 M ./rdft/buffered.c -15 +9 M ./rdft/buffered2.c -9 +6 M ./rdft/dft-r2hc.c -7 +4 M ./rdft/dht-r2hc.c -7 +7 M ./rdft/dht-rader.c -13 +15 M ./rdft/generic.c -1 +1 M ./rdft/hc2hc-generic.c -1 +1 M ./rdft/hc2hc.c -3 +2 M ./rdft/indirect.c -6 +4 M ./rdft/rank-geq2-rdft2.c -2 +2 M ./rdft/rdft-dht.c -5 +5 M ./rdft/rdft2-radix2.c -2 +2 M ./rdft/vrank-geq1-rdft2.c -2 +2 M ./rdft/vrank-geq1.c -2 +2 M ./rdft/vrank3-transpose.c -6 +6 M ./reodft/redft00e-r2hc-pad.c -2 +2 M ./reodft/redft00e-r2hc.c -2 +2 M ./reodft/reodft00e-splitradix.c -34 +17 M ./reodft/reodft010e-r2hc.c -2 +2 M ./reodft/reodft11e-r2hc-odd.c -2 +2 M ./reodft/reodft11e-r2hc.c -2 +2 M ./reodft/reodft11e-radix2.c -2 +2 M ./reodft/rodft00e-r2hc-pad.c -2 +2 M ./reodft/rodft00e-r2hc.c -2 +2 M ./tests/bench.c -1 +1 M ./tests/hook.c -1 +1 Thu Apr 7 23:15:02 EDT 2005 stevenj * [project @ 2005-04-08 03:15:02 by stevenj] ref M ./reodft/reodft00e-splitradix.c -16 +33 Thu Apr 7 00:11:13 EDT 2005 stevenj * [project @ 2005-04-07 04:10:26 by stevenj] whoops M ./genfft/gen_r2r.ml -3 +3 Wed Apr 6 22:06:21 EDT 2005 stevenj * [project @ 2005-04-07 02:06:21 by stevenj] added (optional) new split-radix algorithm, enabled with -newsplit; also new -standalone option to omit desc; also -unitary, -normalization, and -normsqr options to generate r2r codelets with various normalization (to match lit. in DCT-II, use: -unitary -normsqr 2) M ./genfft/complex.ml -1 +22 M ./genfft/complex.mli -1 +7 M ./genfft/fft.ml -4 +86 M ./genfft/gen_athtw.ml -3 +3 M ./genfft/gen_hc2r.ml -4 +4 M ./genfft/gen_notw.ml -5 +4 M ./genfft/gen_notw_c.ml -3 +3 M ./genfft/gen_r2hc.ml -4 +4 M ./genfft/gen_r2r.ml -19 +43 M ./genfft/magic.ml -1 +6 M ./genfft/number.ml -4 +15 M ./genfft/number.mli -1 +2 Fri Mar 25 08:59:43 EST 2005 athena * [project @ 2005-03-25 13:59:43 by athena] Moved timeout check outside the search loop, because X(seconds) is expensive. M ./kernel/planner.c -9 +7 Sun Mar 20 18:35:53 EST 2005 athena * [project @ 2005-03-20 23:35:53 by athena] Enable vector recursion for in-place problems, otherwise dftw-genericbuf works only in PATIENT mode. M ./dft/ct.c -2 +4 Sun Mar 20 17:53:58 EST 2005 athena * [project @ 2005-03-20 22:53:58 by athena] oops M ./dft/dftw-genericbuf.c -1 +1 Sun Mar 20 17:49:13 EST 2005 athena * [project @ 2005-03-20 22:49:13 by athena] make solver UGLY for small N M ./dft/dftw-genericbuf.c -3 +13 Sun Mar 20 17:16:37 EST 2005 athena * [project @ 2005-03-20 22:16:37 by athena] new dftw-genericbuf solver A ./dft/dftw-genericbuf.c Sun Mar 20 16:12:44 EST 2005 athena * [project @ 2005-03-20 21:12:44 by athena] new dftw-genericbuf solver M ./dft/Makefile.am -4 +4 M ./dft/conf.c -1 +2 M ./dft/dft.h -1 +2 Thu Mar 17 21:48:19 EST 2005 athena * [project @ 2005-03-18 02:48:19 by athena] Hmm... what was I thinking? M ./simd/sse2-aux.c -2 +2 Thu Mar 17 19:20:54 EST 2005 athena * [project @ 2005-03-18 00:20:54 by athena] Workaround for a MSVC bug. M ./simd/simd-sse2.h -1 +6 M ./simd/sse2-aux.c -1 +8 Thu Mar 17 08:18:39 EST 2005 athena * [project @ 2005-03-17 13:18:39 by athena] Workaround for a MSVC bug that was reported by Eddie Yee. M ./simd/simd-sse.h -1 +6 M ./simd/sse-aux.c -1 +8 Tue Mar 15 13:25:53 EST 2005 athena * [project @ 2005-03-15 18:25:53 by athena] try both contiguous input and contiguous output when in doubt M ./rdft/rank0.c -3 +26 Tue Mar 15 08:44:41 EST 2005 athena * [project @ 2005-03-15 13:44:41 by athena] Added genfft flag -precompute-twiddles which moves the computation of the twiddle factors before the main schedule. This flag produces smaller code everywhere, and slightly faster code on powerpc. I observe no speed difference on x86. M ./dft/codelets/standard/Makefile.am -3 +3 M ./genfft/genutil.ml -3 +5 M ./genfft/magic.ml -1 +5 M ./genfft/schedule.ml -1 +53 M ./genfft/schedule.mli -1 +2 M ./rdft/codelets/hc2r/Makefile.am -1 +1 M ./rdft/codelets/r2hc/Makefile.am -1 +1 Mon Mar 14 21:43:53 EST 2005 stevenj * [project @ 2005-03-15 02:43:53 by stevenj] sp M ./kernel/kalloc.c -2 +2 Mon Mar 14 21:43:05 EST 2005 stevenj * [project @ 2005-03-15 02:43:05 by stevenj] whoops, spelling error (thanks to Steve Eddins for bug report) M ./kernel/alloc.c -2 +2 Sat Mar 12 15:03:47 EST 2005 athena * [project @ 2005-03-12 20:03:45 by athena] Do not approximate pcost = vl * child->pcost unless child is guaranteed not to be a simple codelet. M ./dft/vrank-geq1.c -2 +4 M ./rdft/vrank-geq1-rdft2.c -2 +4 M ./rdft/vrank-geq1.c -2 +4 Wed Mar 9 20:00:02 EST 2005 athena * [project @ 2005-03-10 01:00:02 by athena] Relaxed applicability conditions. M ./dft/direct.c -6 +3 Wed Mar 9 00:05:47 EST 2005 athena * [project @ 2005-03-09 05:05:47 by athena] Minor optimization M ./dft/dftw-generic.c -8 +8 Tue Mar 8 22:14:02 EST 2005 athena * [project @ 2005-03-09 03:14:02 by athena] Interpret K to mean *1024. Similarly for M. M ./libbench2/problem.c -1 +12 Tue Mar 8 20:44:25 EST 2005 athena * [project @ 2005-03-09 01:44:25 by athena] Hmm... somehow some previous commit got lost. M ./kernel/primes.c -2 +8 Tue Mar 8 20:30:42 EST 2005 athena * [project @ 2005-03-09 01:30:42 by athena] Paranoia M ./dft/ct.c -2 +2 Mon Mar 7 14:30:01 EST 2005 stevenj * [project @ 2005-03-07 19:30:01 by stevenj] whoops M ./configure.ac +1 Mon Mar 7 14:29:43 EST 2005 stevenj * [project @ 2005-03-07 19:29:43 by stevenj] move fftw-specific HP/UX tweak into configure.ac M ./configure.ac +3 M ./m4/ax_cc_maxopt.m4 -2 +2 Mon Mar 7 14:19:24 EST 2005 stevenj * [project @ 2005-03-07 19:19:24 by stevenj] ax_cc_family -> ax_cc_vendor (vendor names are easier to remember), add checks for many new compilers, use in ax_cc_maxopt A ./m4/ax_cc_vendor.m4 M ./configure.ac -3 +3 R ./m4/ax_cc_family.m4 M ./m4/ax_cc_maxopt.m4 -65 +42 M ./m4/ax_cc_vendor.m4 +23 Sun Mar 6 21:36:05 EST 2005 athena * [project @ 2005-03-07 02:36:05 by athena] Count FMA as one flop in estimator when HAVE_FMA M ./kernel/planner.c -1 +7 Sun Mar 6 19:16:06 EST 2005 athena * [project @ 2005-03-07 00:16:06 by athena] Do not try radix-2 generic. M ./dft/dftw-generic.c -1 +1 Sun Mar 6 13:04:23 EST 2005 athena * [project @ 2005-03-06 18:04:23 by athena] Use -O3 for xlc now that we use -O for CODELET_OPTIM M ./m4/ax_cc_maxopt.m4 -4 +4 Sun Mar 6 13:02:41 EST 2005 athena * [project @ 2005-03-06 18:02:41 by athena] New AX_CC_FAMILY macro, that detects the compiler based on symbols that it defines (as opposed to the name of the compiler). We need to start use this strategy everywhere else. A ./m4/ax_cc_family.m4 M ./configure.ac -25 +29 M ./m4/ax_cc_family.m4 +18 Sun Mar 6 11:33:15 EST 2005 athena * [project @ 2005-03-06 16:33:15 by athena] Runtime checks to guarantee small strides. M ./dft/direct.c -2 +18 Sat Mar 5 20:09:25 EST 2005 athena * [project @ 2005-03-06 01:09:25 by athena] Reduced the search space for rank-0 transforms M ./dft/vrank-geq1.c -6 +6 M ./kernel/tensor7.c -8 +20 M ./rdft/rank0.c -50 +48 M ./rdft/vrank-geq1.c -6 +5 Fri Mar 4 17:50:29 EST 2005 stevenj * [project @ 2005-03-04 22:50:29 by stevenj] little assert M ./kernel/primes.c -8 +3 Tue Mar 1 09:19:16 EST 2005 athena * [project @ 2005-03-01 14:19:16 by athena] Implemented directbuf, enabled for now. M ./dft/dft.h -1 +2 M ./dft/dftw-direct.c -10 +7 M ./dft/direct.c -11 +135 M ./dft/kdft.c -3 +3 Mon Feb 28 22:21:14 EST 2005 athena * [project @ 2005-03-01 03:21:14 by athena] Unified dftw-direct, dftw-directbuf in an attempt to tame code growth M ./dft/Makefile.am -4 +4 M ./dft/dftw-direct.c -15 +130 R ./dft/dftw-directbuf.c M ./dft/kdft-dif.c -2 +1 M ./dft/kdft-dit.c -2 +1 Sun Feb 27 13:51:24 EST 2005 stevenj * [project @ 2005-02-27 18:51:24 by stevenj] fixed copyright M ./doc/fftw3.texi -2 +2 Sat Feb 26 22:21:03 EST 2005 athena * [project @ 2005-02-27 03:21:03 by athena] silence warnings M ./rdft/rank0.c -4 +5 Sat Feb 26 22:19:16 EST 2005 athena * [project @ 2005-02-27 03:19:16 by athena] oops M ./rdft/rank0.c -2 +2 Sat Feb 26 21:28:39 EST 2005 athena * [project @ 2005-02-27 02:28:39 by athena] Tweaking while thinking about a higher-rank transposer (bitreverser) M ./rdft/rank0.c -44 +63 Sat Feb 26 20:06:49 EST 2005 athena * [project @ 2005-02-27 01:06:49 by athena] Transposed the buffer, and skewed it. This allows for contiguous copy operations, and the codelet should not incur associativity conflicts if the buffer is large. M ./dft/dftw-directbuf.c -20 +26 Sat Feb 26 18:14:11 EST 2005 stevenj * [project @ 2005-02-26 23:14:11 by stevenj] make tensor_max_index more reasonable (take maximum of input and output max indices, computed separately) M ./kernel/tensor4.c -4 +5 Sat Feb 26 10:04:30 EST 2005 athena * [project @ 2005-02-26 15:04:30 by athena] Use cpy2d instead of cpy2d_tiled, because vl may be too large. M ./rdft/vrank3-transpose.c -5 +5 Sat Feb 26 00:31:52 EST 2005 athena * [project @ 2005-02-26 05:31:52 by athena] Fixed old bug that was introduced with yesterday's changes. M ./genfft/annotate.ml -4 +6 Fri Feb 25 21:54:23 EST 2005 athena * [project @ 2005-02-26 02:54:23 by athena] ``Interesting'' switch statement. M ./kernel/cpy1d.c -3 +23 Fri Feb 25 12:29:54 EST 2005 athena * [project @ 2005-02-25 17:29:54 by athena] Disabled -reorder-loads -reorder-stores, since they seem to do nothing. M ./support/Makefile.codelets -1 +1 Fri Feb 25 12:19:10 EST 2005 stevenj * [project @ 2005-02-25 17:19:03 by stevenj] Because of the recent changes to kernel/pickdim.c, splitrnk=0 is no longer equivalent to splitrnk=1 for rnk < 4, where the latter is the FFTW2 behavior. For small rnk, however, I observe the planner to pretty consistently choose the FFTW2 behavior (splitrnk=1), despite its not being asymptotically optimal in the cache oblivious sense. So, make splitrnk=1 instead of splitrnk=0 the default in FFTW_MEASURE and FFTW_ESTIMATE modes (rnk > 3 is pretty rare in practice anyway). M ./dft/rank-geq2.c -5 +2 M ./rdft/rank-geq2.c -5 +2 Fri Feb 25 00:33:27 EST 2005 stevenj * [project @ 2005-02-25 05:33:27 by stevenj] tweak M ./dft/indirect-transpose.c -3 +4 Fri Feb 25 00:29:09 EST 2005 stevenj * [project @ 2005-02-25 05:29:09 by stevenj] slight relaxation M ./dft/indirect-transpose.c -2 +4 Fri Feb 25 00:21:00 EST 2005 stevenj * [project @ 2005-02-25 05:21:00 by stevenj] cruft M ./dft/indirect-transpose.c -2 Fri Feb 25 00:03:14 EST 2005 stevenj * [project @ 2005-02-25 05:03:13 by stevenj] added experimental indirect-transpose solver: when transforming the columns of the matrix, allow us to do a transpose to make the DFTs contiguous A ./dft/indirect-transpose.c M ./dft/Makefile.am -3 +3 M ./dft/conf.c -1 +2 M ./dft/dft.h -1 +2 M ./dft/indirect-transpose.c +280 M ./dft/indirect.c -3 +10 M ./kernel/ifftw.h -1 +3 M ./kernel/tensor4.c -1 +32 Thu Feb 24 23:04:58 EST 2005 stevenj * [project @ 2005-02-25 04:04:58 by stevenj] check for abort() M ./configure.ac -1 +1 Thu Feb 24 23:04:43 EST 2005 stevenj * [project @ 2005-02-25 04:04:43 by stevenj] call abort() on failed assertion M ./kernel/assert.c -1 +5 Thu Feb 24 21:17:23 EST 2005 athena * [project @ 2005-02-25 02:17:23 by athena] Forgot to change X(isqrt) -> isqrt_maybe M ./kernel/primes.c -2 +2 Thu Feb 24 20:18:59 EST 2005 stevenj * [project @ 2005-02-25 01:17:59 by stevenj] require finite_rnk M ./dft/rank-geq2.c -1 +2 M ./rdft/rank-geq2-rdft2.c -1 +2 M ./rdft/rank-geq2.c -1 +2 Thu Feb 24 20:07:38 EST 2005 stevenj * [project @ 2005-02-25 01:07:38 by stevenj] #ifdef HAVE_STRING_H must come after rdft.h so that we get config.h M ./rdft/vrank3-transpose.c -3 +3 Thu Feb 24 18:59:40 EST 2005 athena * [project @ 2005-02-24 23:59:38 by athena] Implemented reordering of loads and stores so that the real and imaginary part are loaded/stored together. This should improve out-of-cache performance in the presence of associativity conflicts, and maybe worsen in-cache performance because of worse scheduling. Enabled for now, for experimental purposes. M ./genfft/annotate.ml -30 +76 M ./genfft/magic.ml -4 +10 M ./support/Makefile.codelets -1 +1 M ./support/twovers.sh -1 +1 Thu Feb 24 18:10:49 EST 2005 stevenj * [project @ 2005-02-24 23:10:49 by stevenj] fix comment M ./m4/ax_gcc_aligns_stack.m4 -2 +2 Thu Feb 24 18:10:23 EST 2005 stevenj * [project @ 2005-02-24 23:10:23 by stevenj] better message M ./m4/ax_gcc_aligns_stack.m4 -2 +2 Thu Feb 24 18:08:36 EST 2005 stevenj * [project @ 2005-02-24 23:08:36 by stevenj] use gcc version > 3.0 as fallback in check for alignment bug M ./m4/ax_gcc_aligns_stack.m4 -2 +2 Thu Feb 24 18:02:31 EST 2005 stevenj * [project @ 2005-02-24 23:02:31 by stevenj] don't use -malign-double unconditionally (it is only available on x86) M ./m4/ax_gcc_aligns_stack.m4 -2 +3 Thu Feb 24 12:03:30 EST 2005 athena * [project @ 2005-02-24 17:03:30 by athena] Subtler selection of tilesz. M ./kernel/transpose.c -3 +8 Thu Feb 24 11:52:25 EST 2005 athena * [project @ 2005-02-24 16:52:25 by athena] Call cpy2d_tiledbuf, not cpy2d_tiled. M ./rdft/rank0.c -5 +5 Thu Feb 24 11:29:28 EST 2005 athena * [project @ 2005-02-24 16:29:28 by athena] buffer sizes were wrong :-( M ./kernel/cpy2d.c -1 +1 M ./kernel/transpose.c -2 +2 Thu Feb 24 11:19:01 EST 2005 athena * [project @ 2005-02-24 16:19:01 by athena] Single function for computing tile size. Eliminate spurious assertions. M ./kernel/cpy2d.c -3 +8 M ./kernel/ifftw.h -1 +3 M ./kernel/tile2d.c +5 M ./kernel/transpose.c -4 +6 M ./rdft/rank0.c -3 +1 Thu Feb 24 10:00:02 EST 2005 athena * [project @ 2005-02-24 15:00:02 by athena] Do tiling recursively. M ./kernel/tile2d.c +28 Thu Feb 24 09:40:30 EST 2005 athena * [project @ 2005-02-24 14:40:30 by athena] Reworked tiled transposes; provide tiling with and without buffering. I can't believe that one has to waste his life with this @#$%. A ./kernel/tile2d.c M ./kernel/Makefile.am -1 +1 M ./kernel/cpy2d.c -36 +60 M ./kernel/ifftw.h -2 +10 M ./kernel/tile2d.c +41 M ./kernel/transpose.c -64 +122 M ./rdft/rank0.c -6 +41 M ./rdft/vrank3-transpose.c -2 +2 Wed Feb 23 22:21:19 EST 2005 athena * [project @ 2005-02-24 03:21:19 by athena] Clarified logic. I am not sure why the code was so confusing to begin with. The computation of *dp in the which_dim == 0 case was also wrong, returning e.g. *dp == -1 if sz->rnk == 1. M ./kernel/pickdim.c -6 +6 Wed Feb 23 22:00:15 EST 2005 athena * [project @ 2005-02-24 03:00:15 by athena] Enable aggressive inlining in codelets only, to avoid code bloat. M ./configure.ac -9 +10 Wed Feb 23 21:51:50 EST 2005 athena * [project @ 2005-02-24 02:51:50 by athena] Removed cache-oblivious copy/transpose algorithms in favor of explicitly blocked algorithms. The cache-oblivious algorithms fail if there are associativity conflicts, in which case buffering is necessary, as per Carter and Gatlin. Once you set the buffer size, there is no point whatsoever to do the algorithm recursively, and you may as well use blocking. M ./kernel/Makefile.am -2 +1 M ./kernel/cpy2d.c -62 +40 M ./kernel/ifftw.h -11 +11 M ./kernel/primes.c -4 +10 R ./kernel/transpose-rec.c M ./kernel/transpose.c +54 M ./rdft/rank0.c -49 +28 M ./rdft/vrank3-transpose.c -6 +6 Wed Feb 23 18:46:12 EST 2005 stevenj * [project @ 2005-02-23 23:46:12 by stevenj] --disable-fortran now differs from --enable-fortran that fails M ./configure.ac -2 +4 Wed Feb 23 18:42:21 EST 2005 stevenj * [project @ 2005-02-23 23:42:21 by stevenj] comment tweak M ./api/f77api.c -1 +1 Wed Feb 23 18:41:14 EST 2005 stevenj * [project @ 2005-02-23 23:41:14 by stevenj] If a Fortran compiler was not detected, just make our best guess at what wrappers to use...I'm sick of dealing with user complaints from cases where wrapper detection fails for whatever reason. M ./api/f77api.c +21 Wed Feb 23 18:10:40 EST 2005 stevenj * [project @ 2005-02-23 23:10:40 by stevenj] fflush(stdout) after print_plan, in case F77 doesn't M ./api/f77funcs.h +1 Tue Feb 22 22:54:42 EST 2005 athena * [project @ 2005-02-23 03:54:42 by athena] --enable-sse is necessary after all, to generate all dependencies correctly. M ./mkdist.sh -1 +1 Tue Feb 22 22:32:06 EST 2005 athena * [project @ 2005-02-23 03:32:06 by athena] Put cpy2d_pair into its own file, so that I can experiment with buffering of nontwiddle codelets. A ./kernel/cpy2d-pair.c M ./dft/dftw-directbuf.c -28 +7 M ./kernel/Makefile.am -5 +6 M ./kernel/cpy2d-pair.c +61 M ./kernel/ifftw.h -1 +11 Tue Feb 22 20:07:11 EST 2005 athena * [project @ 2005-02-23 01:07:11 by athena] Copy rfftwnd.png from ${srcdir}, not $PWD M ./doc/Makefile.am -1 +1 Tue Feb 22 17:08:48 EST 2005 athena * [project @ 2005-02-22 22:08:48 by athena] Do not bother memcpy-ing complex numbers. M ./rdft/rank0.c -2 +6 Tue Feb 22 16:20:46 EST 2005 athena * [project @ 2005-02-22 21:20:46 by athena] Tighther layout of buffers. I am not sure it matters, but just in case... M ./kernel/cpy2d.c -2 +2 M ./kernel/transpose-rec.c -4 +4 Tue Feb 22 10:13:02 EST 2005 athena * [project @ 2005-02-22 15:13:02 by athena] Usec cpy1d for rank-0 copies M ./rdft/rank0.c -3 +2 Tue Feb 22 10:06:13 EST 2005 athena * [project @ 2005-02-22 15:06:13 by athena] Implemented in-place transposes with buffering. Moved copy/transposition routines into own files, so that we can reuse them from multiple places. TODO: merge vrank3-transpose.c with rank0.c, or rename vrank3-transpose.c to rank0-fancy.c or something like that; decide whether square in-place transposes should be in rank0.c or vrank3-transpose.c; apply FIXME's in vrank3-transpose.c. A ./kernel/cpy1d.c A ./kernel/cpy2d.c A ./kernel/transpose-rec.c A ./kernel/transpose.c M ./kernel/Makefile.am -5 +5 M ./kernel/cpy1d.c +50 M ./kernel/cpy2d.c +154 M ./kernel/ifftw.h -1 +28 M ./kernel/transpose-rec.c +143 M ./kernel/transpose.c +72 M ./rdft/rank0.c -160 +85 M ./rdft/vrank3-transpose.c -231 +24 Mon Feb 21 23:29:52 EST 2005 athena * [project @ 2005-02-22 04:29:52 by athena] Indentation should be printed after newline, not at the beginning of print() M ./kernel/print.c -6 +11 Mon Feb 21 10:07:24 EST 2005 athena * [project @ 2005-02-21 15:07:24 by athena] generalized in anticipation of more complicated solvers. M ./rdft/rank0.c -24 +45 Sun Feb 20 22:18:59 EST 2005 athena * [project @ 2005-02-21 03:18:59 by athena] Implemented buffered recursive transpose M ./rdft/rank0.c -7 +85 Sun Feb 20 18:27:29 EST 2005 athena * [project @ 2005-02-20 23:27:29 by athena] Fixed comment M ./rdft/rank0.c -2 +2 Sun Feb 20 18:22:15 EST 2005 athena * [project @ 2005-02-20 23:22:15 by athena] grand unification of rank0 solvers M ./rdft/Makefile.am -3 +2 M ./rdft/conf.c -2 +1 R ./rdft/rank0-vrank2.c M ./rdft/rank0.c -93 +173 M ./rdft/rdft.h -1 Sun Feb 20 15:35:24 EST 2005 athena * [project @ 2005-02-20 20:35:24 by athena] manual tail-recursion optimization M ./rdft/vrank3-transpose.c -7 +11 Sat Feb 19 17:57:44 EST 2005 athena * [project @ 2005-02-19 22:57:44 by athena] implemented check for transpositions M ./libbench2/verify-lib.c -11 +11 M ./libbench2/verify-r2r.c -1 +1 M ./tests/check.pl -1 +19 Sat Feb 19 17:28:43 EST 2005 athena * [project @ 2005-02-19 22:28:43 by athena] Previous fix was wrong for rdft2 problems. M ./libbench2/verify-lib.c -3 +3 Sat Feb 19 17:23:36 EST 2005 athena * [project @ 2005-02-19 22:23:36 by athena] vecsz->rnk must be finite for this solver to apply. M ./rdft/dft-r2hc.c -2 +2 Sat Feb 19 17:15:19 EST 2005 athena * [project @ 2005-02-19 22:15:19 by athena] unified the various simple'' transposers M ./rdft/vrank3-transpose.c -101 +51 Sat Feb 19 16:55:29 EST 2005 athena * [project @ 2005-02-19 21:55:29 by athena] Fixed stupid bug in rec_transpose_swap. Fixed stupid verifier that did not catch the bug. M ./libbench2/verify-lib.c -17 +26 M ./libbench2/verify-r2r.c -2 +3 M ./rdft/vrank3-transpose.c -5 +5 Sat Feb 19 15:24:03 EST 2005 athena * [project @ 2005-02-19 20:24:03 by athena] Minor cleanup of transposition routines. M ./rdft/vrank3-transpose.c -188 +83 Sat Feb 19 09:31:14 EST 2005 athena * [project @ 2005-02-19 14:31:14 by athena] Make the batch size B=Theta(r) instead of B=Theta(1) in buffered twiddle solvers. Theory: for cache line size L, we want B = Omega(L) to utilize the cache line fully. We also want B*r =O(Z), where Z is the size of the cache. It is safe to assume that Z = Theta(L^2): cache designers will tend to make L as large as they can get away with, because they don't have to program the machines that they build, and Z < Theta(L^2) will screw up the little matrix transposition benchmarks that they use to design the cache. Hence, B=Theta(r) is the right number. M ./dft/dftw-directbuf.c -12 +17 M ./rdft/hc2hc-directbuf.c -16 +20 Fri Feb 18 23:47:22 EST 2005 stevenj * [project @ 2005-02-19 04:47:22 by stevenj] for --enable-portable-binary, only try -mcpu=$arch and -m$arch on x86, since these generate non-portable code on every other target (and some other targets, like Alpha, don't support -mtune=$arch). M ./m4/ax_gcc_archflag.m4 -2 +5 Thu Feb 17 21:15:42 EST 2005 athena * [project @ 2005-02-18 02:15:42 by athena] gcc/aix defines _POWER, not __powerpc__ like the rest of the world does. M ./kernel/ifftw.h -2 +2 Wed Feb 16 22:30:27 EST 2005 athena * [project @ 2005-02-17 03:30:27 by athena] enable fma for ia64, since it seems to help with the hpux compiler. M ./configure.ac +1 Wed Feb 16 21:47:48 EST 2005 athena * [project @ 2005-02-17 02:47:48 by athena] M ./TODO -1 +2 Wed Feb 16 15:27:18 EST 2005 athena * [project @ 2005-02-16 20:27:18 by athena] Fixes for darwin M ./simd/simd-altivec.h -2 +2 Wed Feb 16 14:27:42 EST 2005 athena * [project @ 2005-02-16 19:27:42 by athena] Made the correctness of the code more obvious. M ./api/apiplan.c -14 +16 Wed Feb 16 12:30:29 EST 2005 stevenj * [project @ 2005-02-16 17:30:29 by stevenj] s/with-portable-binary/enable-portable-binary/ to be GNUlly correct; I'm sticking with --with-gcc-arch=arch, however, as --enable-gcc-arch=arch has the wrong connotations for me M ./NEWS -2 +2 M ./m4/ax_cc_maxopt.m4 -3 +3 Wed Feb 16 11:44:48 EST 2005 stevenj * [project @ 2005-02-16 16:44:48 by stevenj] whoops M ./api/apiplan.c -2 +2 Wed Feb 16 11:23:38 EST 2005 stevenj * [project @ 2005-02-16 16:23:38 by stevenj] bless wisdom with patience used to create it M ./api/apiplan.c -1 +1 Wed Feb 16 11:18:56 EST 2005 stevenj * [project @ 2005-02-16 16:18:56 by stevenj] whoops M ./api/apiplan.c -7 +10 Wed Feb 16 10:50:28 EST 2005 stevenj * [project @ 2005-02-16 15:50:28 by stevenj] whoops M ./api/apiplan.c -1 +2 Tue Feb 15 23:53:53 EST 2005 stevenj * [project @ 2005-02-16 04:53:53 by stevenj] added 'timed' planner option M ./NEWS +3 M ./TODO -6 M ./api/apiplan.c -16 +56 M ./api/fftw3.h -1 +3 M ./doc/fftw3.texi -1 +27 M ./kernel/ifftw.h -1 +6 M ./kernel/planner.c -2 +14 M ./kernel/timer.c -1 +13 M ./tests/bench.c +5 Tue Feb 15 23:08:29 EST 2005 athena * [project @ 2005-02-16 04:08:27 by athena] Do not use SIMD_CFLAGS. The theory is that if taint.c is unsafe with SIMD_CFLAGS, then all files in this directory are as well. Conversely, if these files require SIMD_CFLAGS because they include "simd.h", then taint.c requires SIMD_CFLAGS as well, and thus we need some other hack. M ./dft/simd/Makefile.am -1 M ./simd/Makefile.am -8 +3 Tue Feb 15 22:49:05 EST 2005 athena * [project @ 2005-02-16 03:49:05 by athena] Do not override CFLAGS in Makefile.am. M ./dft/codelets/standard/Makefile.am -1 M ./dft/simd/Makefile.am -1 +1 M ./dft/simd/codelets/Makefile.am -1 +1 M ./rdft/codelets/hc2r/Makefile.am -1 M ./rdft/codelets/r2hc/Makefile.am -1 M ./rdft/codelets/r2r/Makefile.am -1 M ./support/Makefile.codelets -1 +6 Tue Feb 15 10:30:12 EST 2005 athena * [project @ 2005-02-15 15:30:12 by athena] Allow users to build long double version even if sizeof(long double) == sizeof(double) M ./configure.ac -3 Mon Feb 14 19:55:38 EST 2005 athena * [project @ 2005-02-15 00:55:38 by athena] Updated for 3.1 M ./commercialize.sh -12 +20 Mon Feb 14 19:07:14 EST 2005 athena * [project @ 2005-02-15 00:07:14 by athena] Oops, version.h is no longer used M ./api/version.c -2 +1 Mon Feb 14 18:51:05 EST 2005 athena * [project @ 2005-02-14 23:51:05 by athena] unified fma and non-fma versions A ./support/twovers.sh M ./api/Makefile.am -5 +2 M ./api/version.c -2 +6 M ./configure.ac -5 +10 M ./dft/codelets/standard/Makefile.am -7 +7 M ./dft/simd/codelets/Makefile.am -10 +10 M ./m4/ocaml.m4 -9 +3 M ./mkdist.sh -5 M ./rdft/codelets/hc2r/Makefile.am -4 +4 M ./rdft/codelets/r2hc/Makefile.am -4 +4 M ./rdft/codelets/r2r/Makefile.am -9 +9 M ./support/Makefile.am -2 +2 M ./support/Makefile.codelets -7 +2 M ./support/twovers.sh +17 Mon Feb 14 14:12:09 EST 2005 athena * [project @ 2005-02-14 19:12:09 by athena] forgot to remove inplace/Makefile from configure.ac M ./configure.ac -1 Mon Feb 14 12:08:52 EST 2005 athena * [project @ 2005-02-14 17:08:49 by athena] Merged dft/codelets/inplace with the main dft/codelets/standard directory. This step makes dft codelets consistent with the rest of the naming conventions, and will simplify the eventual merge of fma and non-fma codelets. M ./Makefile.am -1 M ./dft/codelet-dft.h -2 +1 M ./dft/codelets/Makefile.am -1 +1 R ./dft/codelets/inplace/Makefile.am R ./dft/codelets/inplace/ M ./dft/codelets/standard/Makefile.am -9 +31 M ./dft/conf.c -2 +1 Mon Feb 14 11:16:15 EST 2005 athena * [project @ 2005-02-14 16:16:15 by athena] inline altivec constants, since gcc seems to generate better code this way. M ./simd/altivec.c -21 +1 M ./simd/simd-altivec.h -16 +18 Sun Feb 13 18:17:32 EST 2005 athena * [project @ 2005-02-13 23:17:32 by athena] group altivec constants into a single array, for faster access M ./simd/altivec.c -11 +17 M ./simd/simd-altivec.h -20 +11 Sun Feb 13 18:15:37 EST 2005 athena * [project @ 2005-02-13 23:15:37 by athena] code cleanup M ./genfft/c.ml -23 +9 M ./genfft/c.mli -2 +1 M ./genfft/simd.ml -6 +6 Sun Feb 13 10:29:32 EST 2005 athena * [project @ 2005-02-13 15:29:32 by athena] removed some unused stuff M ./genfft/c.ml -3 +1 M ./genfft/c.mli -2 +1 Sat Feb 12 22:04:40 EST 2005 athena * [project @ 2005-02-13 03:04:40 by athena] New twiddle scheme for altivec, 3dnow M ./simd/simd-3dnow.h -4 +34 M ./simd/simd-altivec.h -4 +27 Sat Feb 12 20:17:35 EST 2005 athena * [project @ 2005-02-13 01:17:35 by athena] Implemented new twiddle scheme for sse2 M ./simd/simd-sse2.h -4 +26 Sat Feb 12 19:57:46 EST 2005 athena * [project @ 2005-02-13 00:57:40 by athena] Implemented experimental t2* codelets, which store twiddle factors in a more convenient format, at the expense of twice the storage. Currently only SSE works; I have to port SSE2, altivec, etc. to the new scheme. After this, we will decide whether these codelets are worth the price. A ./dft/simd/t2b.h A ./dft/simd/t2f.h M ./dft/simd/Makefile.am -1 +1 M ./dft/simd/codelets/Makefile.am -2 +14 M ./dft/simd/q1b.h +5 M ./dft/simd/q1f.h +5 M ./dft/simd/t1b.h +5 M ./dft/simd/t1f.h +5 M ./dft/simd/t2b.h +36 M ./dft/simd/t2f.h +36 M ./simd/simd-sse.h -36 +26 Fri Feb 11 08:07:12 EST 2005 athena * [project @ 2005-02-11 13:07:12 by athena] Forgot to define SIMD_STRIDE_OKPAIR M ./simd/simd-altivec.h +1 Thu Feb 10 22:20:00 EST 2005 athena * [project @ 2005-02-11 03:20:00 by athena] fixed sse2, 3dnow, and altivec, as promised M ./simd/simd-3dnow.h -5 +3 M ./simd/simd-altivec.h -1 +4 M ./simd/simd-sse.h +1 M ./simd/simd-sse2.h -5 +3 Thu Feb 10 21:47:40 EST 2005 athena * [project @ 2005-02-11 02:47:33 by athena] Generate n2?v_* codelets in such a way that we may or may not pair stores, depending on which mode happens to work best on a particular SIMD implementation. sse2, 3dnow, and altivec are currently broken---will fix soon. M ./dft/simd/n2b.c -1 +1 M ./dft/simd/n2f.c -1 +1 M ./genfft/annotate.ml -7 +9 M ./genfft/expr.ml -4 +6 M ./genfft/expr.mli -2 +2 M ./genfft/simd.ml -3 +7 M ./simd/simd-sse.h -1 +6 Thu Feb 10 08:53:22 EST 2005 athena * [project @ 2005-02-10 13:53:22 by athena] instantiate altivec constants only once M ./simd/altivec.c -1 +6 M ./simd/simd-altivec.h -4 +10 Thu Feb 10 06:37:56 EST 2005 athena * [project @ 2005-02-10 11:37:56 by athena] Fixed alignment checks for new SIMD scheme M ./dft/simd/n2b.c -3 +3 M ./dft/simd/n2f.c -3 +3 Wed Feb 9 21:35:01 EST 2005 athena * [project @ 2005-02-10 02:35:01 by athena] Change n2?v_* codelets to store pairs of vectors, with implicit 2x2 transposition. Works for 2-way SIMD as well. Tested with sse and sse2. I haven't tried altivec yet, but I observed a huge speedup when I transformed one codelet by hand. M ./dft/simd/codelets/Makefile.am -8 +6 M ./genfft/annotate.ml -3 +48 M ./genfft/annotate.mli -2 +5 M ./genfft/expr.ml -2 +5 M ./genfft/expr.mli -2 +2 M ./genfft/gen_notw_c.ml -3 +17 M ./genfft/genutil.ml -3 +2 M ./genfft/simd.ml -5 +9 M ./genfft/simdmagic.ml -1 +3 M ./simd/simd-3dnow.h +6 M ./simd/simd-altivec.h +10 M ./simd/simd-sse.h +6 M ./simd/simd-sse2.h +6 Tue Feb 8 21:28:38 EST 2005 athena * [project @ 2005-02-09 02:28:38 by athena] Resurrected old DIF codelets for experimental purposes. They are disabled for now, but I am keeping the setup around for future reference. M ./dft/codelets/standard/Makefile.am -1 +14 Tue Feb 8 20:10:19 EST 2005 stevenj * [project @ 2005-02-09 01:09:12 by stevenj] M ./doc/fftw3.texi -7 +7 Tue Feb 8 19:37:09 EST 2005 stevenj * [project @ 2005-02-09 00:37:09 by stevenj] clarifications, document --with-portable-binary and --with-gcc-arch M ./doc/fftw3.texi -17 +30 Tue Feb 8 19:23:41 EST 2005 stevenj * [project @ 2005-02-09 00:20:56 by stevenj] M ./NEWS -7 +7 Tue Feb 8 01:36:22 EST 2005 stevenj * [project @ 2005-02-08 06:36:22 by stevenj] more change comments M ./NEWS +61 Tue Feb 8 00:41:38 EST 2005 stevenj * [project @ 2005-02-08 05:41:38 by stevenj] fma is definitely beneficial on Itanium with the HP/UX compiler M ./doc/FAQ/fftw-faq.bfnn -1 +1 Mon Feb 7 22:58:47 EST 2005 athena * [project @ 2005-02-08 03:58:47 by athena] Silence warnings. M ./libbench2/bench-main.c -3 +4 Mon Feb 7 22:55:49 EST 2005 stevenj * [project @ 2005-02-08 03:55:49 by stevenj] when we compile our own getopt, change symbol names to avoid conflicts (e.g. avoid build failure on MacOS X with --enable-shared) M ./libbench2/getopt.h +26 Mon Feb 7 22:36:42 EST 2005 stevenj * [project @ 2005-02-08 03:36:42 by stevenj] grr, more bugfixes for in-place case M ./reodft/reodft00e-splitradix.c -9 +13 Mon Feb 7 22:29:35 EST 2005 athena * [project @ 2005-02-08 03:29:35 by athena] removed relics of FRANZ mode M ./dft/codelets/standard/Makefile.am -19 Mon Feb 7 18:48:36 EST 2005 athena * [project @ 2005-02-07 23:48:36 by athena] Somehow xlc does not like ``vector int dummy;'' M ./simd/altivec.c -2 +2 Mon Feb 7 13:59:47 EST 2005 athena * [project @ 2005-02-07 18:59:47 by athena] There is no need to enable sse to make the distribution. This might have been true in the past but not anymore. M ./mkdist.sh -6 +8 Mon Feb 7 13:55:17 EST 2005 athena * [project @ 2005-02-07 18:55:17 by athena] Oops---included fortran file in C sources M ./api/Makefile.am -1 +1 Mon Feb 7 13:42:45 EST 2005 athena * [project @ 2005-02-07 18:42:45 by athena] Set version string at ``make dist'' time, not at ``configure'' time, so we know whether a user is using the fma version or not. M ./api/Makefile.am -2 +9 M ./api/version.c -2 +3 Sun Feb 6 17:00:33 EST 2005 athena * [project @ 2005-02-06 22:00:33 by athena] Removed useless files R ./genfft/gen_hc2r_noinline.ml R ./genfft/gen_notw_noinline.ml R ./genfft/gen_notw_noinline_c.ml R ./genfft/gen_r2hc_noinline.ml Sun Feb 6 16:59:39 EST 2005 athena * [project @ 2005-02-06 21:59:39 by athena] Different (simpler?) way to prevent the compiler from optimizing loop inductive variables. We now explicitly corrupt stride variables by xor-ing them with another variable that happens to be zero (but the compiler does not know it). In this way, the compiler does not attempt to extract a zillion loop indices from codelets, which would overflow the register set. Set the -fno-loop-optimize flag to further help the process. Consequences: removed m* codelets. Smaller library size. Slightly faster code with gcc/powerpc (including altivec). Much faster code with xlc/powerpc. No changes for gcc/pentium. Maybe slightly faster with icc/pentium. M ./configure.ac -11 +9 M ./dft/codelets/standard/Makefile.am -52 +3 M ./dft/simd/codelets/Makefile.am -30 +6 M ./genfft/Makefile.am -36 +10 M ./genfft/c.ml -1 +4 M ./genfft/c.mli -1 +2 M ./genfft/gen_hc2hc.ml -3 +5 M ./genfft/gen_hc2r.ml -3 +7 M ./genfft/gen_notw.ml -3 +6 M ./genfft/gen_notw_c.ml -3 +6 M ./genfft/gen_r2hc.ml -3 +7 M ./genfft/gen_twiddle.ml -3 +5 M ./genfft/gen_twiddle_c.ml -3 +5 M ./genfft/gen_twidsq.ml -3 +6 M ./genfft/gen_twidsq_c.ml -3 +6 M ./genfft/genutil.ml -1 +2 M ./genfft/simd.ml -1 +2 M ./kernel/ifftw.h -3 +7 M ./kernel/stride.c -1 +3 M ./rdft/codelets/hc2r/Makefile.am -13 +3 M ./rdft/codelets/r2hc/Makefile.am -13 +3 M ./support/Makefile.codelets -4 Sat Feb 5 18:51:08 EST 2005 stevenj * [project @ 2005-02-05 23:51:08 by stevenj] paranoia about in-place rodft00 plans M ./reodft/reodft00e-splitradix.c -3 +17 Sat Feb 5 18:39:55 EST 2005 stevenj * [project @ 2005-02-05 23:39:55 by stevenj] don't believe pcost when using the estimator...there is no point, and it screws up estimator hacks to prefer in-codelet loops to vecloops M ./kernel/planner.c -2 +2 Sat Feb 5 18:34:25 EST 2005 athena * [project @ 2005-02-05 23:34:25 by athena] Reduced optimization level from -O3 to -O for xlc, since -O generates faster code. M ./m4/ax_cc_maxopt.m4 -4 +4 Sat Feb 5 16:26:58 EST 2005 stevenj * [project @ 2005-02-05 21:26:58 by stevenj] whoops, only applicable to redft00/rodft00 plans M ./reodft/reodft00e-splitradix.c -1 +2 Sat Feb 5 16:22:39 EST 2005 stevenj * [project @ 2005-02-05 21:22:39 by stevenj] fixed in-place operation, and don't create size-0 sub-plans M ./reodft/reodft00e-splitradix.c -17 +16 Fri Feb 4 11:30:30 EST 2005 athena * [project @ 2005-02-04 16:30:30 by athena] Autodetect altivec on linux. This code works with gcc-3.4 and -maltivec, with or without -mabi=altivec. The code *should* work with gcc-3.3 without -mabi=altivec. However, disabling -mabi=altivec on gcc-3.4 produces much worse code (I don't know why). M ./simd/altivec.c -5 +31 Fri Jan 28 00:04:58 EST 2005 stevenj * [project @ 2005-01-28 05:04:58 by stevenj] update reference M ./doc/fftw3.texi -3 +3 Thu Jan 27 15:48:28 EST 2005 stevenj * [project @ 2005-01-27 20:48:28 by stevenj] note that DCT-II/III are often called the'' DCT/DCT M ./doc/fftw3.texi -11 +17 Fri Jan 21 14:42:04 EST 2005 stevenj * [project @ 2005-01-21 19:42:04 by stevenj] added MSVC++ for ia64 (based on information at http://www.intel.com/cd/ids/developer/asmo-na/eng/19949.htm?prn=Y) M ./kernel/cycle.h -1 +21 Fri Jan 21 14:22:50 EST 2005 stevenj * [project @ 2005-01-21 19:22:50 by stevenj] vc++ defines _M_AMD64 on x86-64, apparently M ./kernel/cycle.h -3 +3 Tue Jan 18 22:30:27 EST 2005 stevenj * [project @ 2005-01-19 03:30:27 by stevenj] avoid gratuitous breakage with -Werror, requested by Simon Perreault M ./m4/acx_pthread.m4 -2 +2 Mon Jan 17 18:54:55 EST 2005 stevenj * [project @ 2005-01-17 23:54:55 by stevenj] comment typo M ./m4/ax_gcc_aligns_stack.m4 -2 +2 Sat Jan 15 16:56:23 EST 2005 stevenj * [project @ 2005-01-15 21:56:23 by stevenj] bumped shared-lib revision# M ./configure.ac -1 +1 Sat Jan 15 16:35:42 EST 2005 stevenj * [project @ 2005-01-15 21:35:42 by stevenj] add X(estimate_cost) to get estimator cost, and print from bench, to aid in tweaking estimator M ./api/fftw3.h -1 +2 M ./api/flops.c +5 M ./kernel/ifftw.h -1 +2 M ./kernel/planner.c -6 +11 M ./tests/bench.c -1 +1 Sat Jan 15 14:57:56 EST 2005 stevenj * [project @ 2005-01-15 19:57:56 by stevenj] M ./doc/fftw3.texi -3 +3 Sat Jan 15 12:57:07 EST 2005 stevenj * [project @ 2005-01-15 17:57:07 by stevenj] formatting fix M ./doc/fftw3.texi -3 +9 Sat Jan 15 12:31:28 EST 2005 stevenj * [project @ 2005-01-15 17:31:28 by stevenj] tweaks M ./doc/fftw3.texi -17 +17 M ./reodft/Makefile.am +1 M ./reodft/conf.c -4 +7 Sat Jan 15 12:03:24 EST 2005 stevenj * [project @ 2005-01-15 17:03:24 by stevenj] use less buffer space M ./reodft/reodft00e-splitradix.c -22 +23 Sat Jan 15 01:41:58 EST 2005 stevenj * [project @ 2005-01-15 06:41:58 by stevenj] added split-radix-based dct/dst I for odd n A ./reodft/reodft00e-splitradix.c M ./doc/fftw3.texi -16 +21 M ./reodft/Makefile.am -1 +1 M ./reodft/conf.c -1 +2 M ./reodft/redft00e-r2hc.c -2 +4 M ./reodft/reodft.h +1 M ./reodft/reodft00e-splitradix.c +337 M ./reodft/rodft00e-r2hc.c -2 +4 Fri Jan 14 21:50:08 EST 2005 stevenj * [project @ 2005-01-15 02:50:08 by stevenj] M ./api/fftw3.h -3 +3 Fri Jan 14 21:49:55 EST 2005 stevenj * [project @ 2005-01-15 02:49:55 by stevenj] warn silly users who confuse CVS id with FFTW version M ./api/fftw3.h -1 +3 Fri Jan 14 16:57:36 EST 2005 stevenj * [project @ 2005-01-14 21:57:36 by stevenj] get sparc cpu type on solaris as well as with linux M ./m4/ax_gcc_archflag.m4 -13 +11 Thu Jan 13 19:21:58 EST 2005 stevenj * [project @ 2005-01-14 00:21:58 by stevenj] detect prescott mobile (f37) M ./m4/ax_gcc_archflag.m4 -2 +2 Thu Jan 13 18:09:52 EST 2005 stevenj * [project @ 2005-01-13 23:09:52 by stevenj] use cpuid for x86_64 as well as i[56]86 M ./bootstrap.sh +1 M ./m4/ax_gcc_archflag.m4 -2 +2 Thu Jan 13 17:59:55 EST 2005 stevenj * [project @ 2005-01-13 22:59:55 by stevenj] update with x86info 1.7 and other sources (identify k8, nocona, etc), handle nonzero leading bytes in eax M ./m4/ax_gcc_archflag.m4 -23 +27 Thu Jan 13 16:30:33 EST 2005 stevenj * [project @ 2005-01-13 21:30:33 by stevenj] compactified check for JOINABLE; use AC_DEFINE_UNQUOTED instead of AC_DEFINE for PTHREAD_CREATE_JOINABLE (thanks to Oliver Niekrenz for the bug report) M ./m4/acx_pthread.m4 -23 +16 Wed Jan 12 12:22:13 EST 2005 athena * [project @ 2005-01-12 17:22:13 by athena] The scheduler hack was incorrect because it swapped instructions of the form A = *B and *B = C. Fixed. M ./genfft/annotate.ml -7 +13 Tue Jan 11 22:13:24 EST 2005 athena * [project @ 2005-01-12 03:13:24 by athena] Quote expressions such as ``if test $FOO = yes'' when $FOO may be empty. Also, $GCC is set to either ``yes'' or empty, never to ``no''. M ./m4/ax_cc_maxopt.m4 -2 +2 M ./m4/ax_gcc_archflag.m4 -2 +2 Tue Jan 11 19:30:47 EST 2005 athena * [project @ 2005-01-12 00:30:47 by athena] Hmm---somehow the previous commit did not work. M ./TODO -6 M ./configure.ac +9 M ./simd/altivec.c -3 +3 Tue Jan 11 16:54:45 EST 2005 athena * [project @ 2005-01-11 21:54:45 by athena] Fixed various gcc-related problems on powerpc: - gcc-3.4 becomes totally confused by expressions like vec_add(a, vec_add(b, vec_add(c, ...))) The compiler uses gigabytes of memory and then crashes, presumably because of the exponential-time search problem involved in typing the above expression (since vec_add can take either ints or floats). I changed VADD and similar macros to be inline functions, thus constraining the type system. - New flags --param inline-unit-growth=1000 --param large-function-growth=1000 to work around limitations of the gcc-3.4 inliner. M ./simd/simd-altivec.h -14 +12 Mon Jan 10 21:27:24 EST 2005 athena * [project @ 2005-01-11 02:27:24 by athena] Check for HAVE_ALTIVEC_H M ./simd/simd-altivec.h -1 +2 Mon Jan 10 21:09:30 EST 2005 athena * [project @ 2005-01-11 02:09:30 by athena] Remove support for altivec using gcc builtins, since these keep changing across gcc versions. These changes work on gcc-3.4/linux; I haven't tried MacOS X yet. (The altivec ``spec'' differs between Motorola/Apple and gcc, grrr...) M ./configure.ac -1 +2 M ./simd/altivec.c -8 +5 M ./simd/simd-altivec.h -69 +5 Mon Jan 10 18:57:30 EST 2005 athena * [project @ 2005-01-10 23:57:30 by athena] Stylistic changes M ./rdft/rank0-vrank2.c -21 +21 Mon Jan 10 17:34:41 EST 2005 athena * [project @ 2005-01-10 22:34:41 by athena] Changed incorrect ugliness condition. M ./rdft/dft-r2hc.c -4 +2 Mon Jan 10 16:09:43 EST 2005 stevenj * [project @ 2005-01-10 21:09:43 by stevenj] note x86info version number that was used, to make it easier to update the cpuid for changes in later versions M ./m4/ax_gcc_archflag.m4 -2 +2 Mon Jan 10 15:00:51 EST 2005 athena * [project @ 2005-01-10 20:00:51 by athena] Make dft-r2hc non-UGLY for rank-0 problems M ./rdft/dft-r2hc.c -1 +6 Mon Jan 10 14:50:23 EST 2005 athena * [project @ 2005-01-10 19:50:23 by athena] Do not use -mcpu=970 on power4 processors, because power4 does not have altivec. M ./m4/ax_gcc_archflag.m4 -2 +2 Mon Jan 10 14:48:47 EST 2005 athena * [project @ 2005-01-10 19:48:47 by athena] Note gcc-3.4 problem with inlining. M ./TODO +5 Mon Jan 10 13:51:08 EST 2005 athena * [project @ 2005-01-10 18:51:08 by athena] Oops, forgot to remove ``static'' from the declaration of noninlinable functions. M ./genfft/gen_hc2r_noinline.ml -3 +3 M ./genfft/gen_notw_noinline_c.ml -3 +3 Mon Jan 10 12:31:26 EST 2005 athena * [project @ 2005-01-10 17:31:26 by athena] Recognize power4. Use ``head -n COUNT'' instead of obsolete ``head -COUNT'' (which fails on gentoo). M ./m4/ax_gcc_archflag.m4 -3 +4 Sun Jan 9 22:12:16 EST 2005 athena * [project @ 2005-01-10 03:12:16 by athena] Remind to add FAQ entry concerning gcc-3.4.[1-3] crashes. M ./TODO +4 Sun Jan 9 21:53:08 EST 2005 stevenj * [project @ 2005-01-10 02:53:08 by stevenj] whoops M ./m4/ax_gcc_version.m4 -3 +3 Sun Jan 9 21:48:02 EST 2005 stevenj * [project @ 2005-01-10 02:48:02 by stevenj] support checking for major.minor.patchlevel M ./m4/ax_gcc_version.m4 -10 +11 Sun Jan 9 21:40:18 EST 2005 athena * [project @ 2005-01-10 02:40:18 by athena] Revert CODELET_OPTIM to -O on IA32, which is faster than -O2. M ./configure.ac -3 +4 Sun Jan 9 20:30:12 EST 2005 athena * [project @ 2005-01-10 01:30:12 by athena] /bin/sh allows no spaces in assignments. M ./configure.ac -1 +1 Sun Jan 9 20:05:55 EST 2005 athena * [project @ 2005-01-10 01:05:55 by athena] Make non-inlinable functions external, so that gcc becomes confused and does not try to inline them. M ./genfft/gen_hc2r_noinline.ml -5 +5 M ./genfft/gen_notw_noinline.ml -5 +5 M ./genfft/gen_notw_noinline_c.ml -5 +5 Sun Jan 9 13:44:25 EST 2005 athena * [project @ 2005-01-09 18:44:25 by athena] Add -fno-web to CFLAGS, because -fweb destroys FMAs. M ./configure.ac -13 +18 Sun Jan 9 10:31:47 EST 2005 athena * [project @ 2005-01-09 15:31:47 by athena] Allow -mcpu=970 besides -mcpu=G5 M ./m4/ax_gcc_archflag.m4 -2 +2 Sun Jan 9 10:26:20 EST 2005 athena * [project @ 2005-01-09 15:26:20 by athena] configure was not using -fno-schedule-insns :-( M ./configure.ac -2 +1 Sun Jan 9 08:52:40 EST 2005 athena * [project @ 2005-01-09 13:52:40 by athena] In mkplan() and elsewhere, use solver index instead of solver *pointer*, which looks marginally clearer. M ./kernel/planner.c -18 +22 Sun Jan 9 08:15:36 EST 2005 athena * [project @ 2005-01-09 13:15:36 by athena] Split planner hash table into two tables, for blessed and unblessed solutions respectively. Now an unblessed solution never overwrites a blessed solution, thus avoiding wisdom leakage by construction. Further, forget() is now a O(1) operation, which speeds up the estimator when the wisdom table is large. M ./TODO -6 M ./kernel/ifftw.h -14 +17 M ./kernel/planner.c -157 +126 Sat Jan 8 21:19:45 EST 2005 athena * [project @ 2005-01-09 02:19:45 by athena] New TODO idea. M ./TODO +7 Thu Jan 6 11:02:29 EST 2005 athena * [project @ 2005-01-06 16:02:29 by athena] Split search() into two routines to make the UGLY/NO_UGLY logic obvious. M ./kernel/planner.c -41 +48 Fri Dec 17 16:08:54 EST 2004 stevenj * [project @ 2004-12-17 21:08:54 by stevenj] push/pop 64-bit registers on ia64; thanks to Orion Poplawski for the fix M ./simd/3dnow.c -1 +13 M ./simd/sse.c -1 +7 M ./simd/sse2.c -1 +7 Thu Dec 9 21:41:09 EST 2004 stevenj * [project @ 2004-12-10 02:41:09 by stevenj] patch from FreeBSD ports - FreeBSD does not have memalign, but its malloc is 16-byte aligned M ./kernel/kalloc.c -1 +5 Tue Nov 23 17:06:47 EST 2004 stevenj * [project @ 2004-11-23 22:06:47 by stevenj] don't compile taint.c with SIMD_CFLAGS (fixed Debian bug #259612) M ./simd/Makefile.am -3 +10 Thu Nov 18 11:37:32 EST 2004 stevenj * [project @ 2004-11-18 16:37:32 by stevenj] revert incorrect change -- codlist.c should be rebuilt, but it is built in the build directory and not in the source directory M ./support/Makefile.codelets -1 +1 Wed Nov 17 22:53:53 EST 2004 stevenj * [project @ 2004-11-18 03:53:53 by stevenj] $(CODLIST) should be rebuilt only if Makefile.am changes, or alternatively only in maintainer mode, to prevent stomping in the source directory during user builds. (Thanks to Grant Cook for the bug report.) M ./support/Makefile.codelets -1 +1 Sat Nov 13 13:43:01 EST 2004 stevenj * [project @ 2004-11-13 18:43:01 by stevenj] corrected #ifdef for icc/ia64, thanks to Matt Boman M ./kernel/cycle.h -16 +18 Sat Nov 13 13:34:55 EST 2004 stevenj * [project @ 2004-11-13 18:34:55 by stevenj] spelling correction (Larsen, not Larson) M ./NEWS -1 +1 Mon Nov 8 22:12:39 EST 2004 stevenj * [project @ 2004-11-09 03:12:39 by stevenj] use standard withval M ./m4/ax_gcc_archflag.m4 -3 +3 Mon Nov 8 22:09:16 EST 2004 stevenj * [project @ 2004-11-09 03:09:16 by stevenj] match doc M ./m4/ax_gcc_x86_cpuid.m4 -2 +2 Mon Nov 8 22:00:34 EST 2004 stevenj * [project @ 2004-11-09 03:00:34 by stevenj] formatting M ./m4/ax_openmp.m4 -3 +3 Mon Nov 8 21:59:33 EST 2004 stevenj * [project @ 2004-11-09 02:59:33 by stevenj] make sure OPENMP_CFLAGS environment variable is used correctly M ./m4/ax_openmp.m4 -2 +6 Mon Nov 8 21:46:50 EST 2004 stevenj * [project @ 2004-11-09 02:46:50 by stevenj] replace ax_check_cc_flags with more generic ax_check_compiler_flags A ./m4/ax_check_compiler_flags.m4 M ./configure.ac -13 +8 M ./m4/ax_cc_maxopt.m4 -7 +6 R ./m4/ax_check_cc_flags.m4 M ./m4/ax_check_compiler_flags.m4 +37 M ./m4/ax_gcc_aligns_stack.m4 -5 +4 M ./m4/ax_gcc_archflag.m4 -3 +3 Mon Nov 8 17:49:42 EST 2004 stevenj * [project @ 2004-11-08 22:49:42 by stevenj] separate macro for OpenMP test A ./m4/ax_openmp.m4 M ./configure.ac -40 +6 M ./m4/ax_cc_maxopt.m4 -3 +5 M ./m4/ax_openmp.m4 +58 Fri Nov 5 16:24:22 EST 2004 stevenj * [project @ 2004-11-05 21:24:22 by stevenj] typo M ./doc/fftw3.texi -2 +2 Fri Oct 29 00:48:13 EDT 2004 stevenj * [project @ 2004-10-29 04:48:13 by stevenj] M ./configure.ac -12 +19 Thu Oct 28 00:09:38 EDT 2004 stevenj * [project @ 2004-10-28 04:09:38 by stevenj] better guessing of sparc type on Linux M ./m4/ax_gcc_archflag.m4 -2 +15 Wed Oct 27 13:44:08 EDT 2004 stevenj * [project @ 2004-10-27 17:44:08 by stevenj] note default M ./m4/ax_gcc_archflag.m4 -2 +2 Wed Oct 27 13:41:57 EDT 2004 stevenj * [project @ 2004-10-27 17:41:57 by stevenj] tweak M ./m4/ax_gcc_archflag.m4 -2 +2 Wed Oct 27 13:34:25 EDT 2004 stevenj * [project @ 2004-10-27 17:34:25 by stevenj] comment M ./m4/ax_gcc_x86_cpuid.m4 -2 +3 Wed Oct 27 13:31:10 EDT 2004 stevenj * [project @ 2004-10-27 17:31:10 by stevenj] whoops, m4 is EXTRA_DIST, not SUBDIR, since it doesn't have a Makefile M ./Makefile.am -2 +2 Wed Oct 27 13:16:57 EDT 2004 stevenj * [project @ 2004-10-27 17:16:57 by stevenj] silence warnings M ./m4/ocaml.m4 -2 +2 Wed Oct 27 13:14:22 EDT 2004 stevenj * [project @ 2004-10-27 17:14:22 by stevenj] clean up m4 macros; try to detect correct gcc -march flag on x86; new --with-portable-binary, --with-gcc-arch= flags; use -O2 for codelets with gcc 3.4 to work around bug A ./m4/ A ./m4/acx_pthread.m4 A ./m4/amx_prog_as.m4 A ./m4/ax_cc_maxopt.m4 A ./m4/ax_check_cc_flags.m4 A ./m4/ax_gcc_aligns_stack.m4 A ./m4/ax_gcc_archflag.m4 A ./m4/ax_gcc_version.m4 A ./m4/ax_gcc_x86_cpuid.m4 A ./m4/ocaml.m4 M ./Makefile.am -2 +4 R ./acinclude.m4 R ./acx_pthread.m4 M ./bootstrap.sh -12 +10 M ./configure.ac -11 +26 M ./m4/acx_pthread.m4 +240 M ./m4/amx_prog_as.m4 +11 M ./m4/ax_cc_maxopt.m4 +126 M ./m4/ax_check_cc_flags.m4 +37 M ./m4/ax_gcc_aligns_stack.m4 +47 M ./m4/ax_gcc_archflag.m4 +147 M ./m4/ax_gcc_version.m4 +34 M ./m4/ax_gcc_x86_cpuid.m4 +37 M ./m4/ocaml.m4 +84 Tue Oct 26 16:46:14 EDT 2004 stevenj * [project @ 2004-10-26 20:46:14 by stevenj] rename cexp -> mcexp to avoid conflict with C99 builtin M ./libbench2/mp.c -3 +3 Mon Oct 25 16:58:23 EDT 2004 stevenj * [project @ 2004-10-25 20:58:23 by stevenj] use basename , w/o args, for compiler-name comparisons; also detect Compaq ccc on alpha-linus M ./acinclude.m4 -4 +10 Sun Oct 24 22:05:10 EDT 2004 stevenj * [project @ 2004-10-25 02:05:10 by stevenj] note recent icc problems M ./doc/FAQ/fftw-faq.bfnn -6 +9 Sun Oct 24 02:10:12 EDT 2004 stevenj * [project @ 2004-10-24 06:10:12 by stevenj] whoops, disable semaphores again (for now) M ./threads/threads.c -1 +1 Sun Oct 24 02:04:58 EDT 2004 stevenj * [project @ 2004-10-24 06:04:58 by stevenj] POSIX semaphores are *not* the same as SYSV semaphores M ./threads/threads.c -2 +2 Sun Oct 24 01:18:14 EDT 2004 stevenj * [project @ 2004-10-24 05:18:14 by stevenj] re-implement threaded stuff; dftw now takes parameters to indicate a portion of m loop A ./rdft/hc2hc.c A ./rdft/hc2hc.h A ./threads/ct.c A ./threads/hc2hc.c M ./dft/conf.c -2 +2 M ./dft/ct.c -8 +11 M ./dft/ct.h -8 +11 M ./dft/ctsq.c -2 +2 M ./dft/dft.h -5 +2 M ./dft/dftw-direct.c -15 +29 M ./dft/dftw-directbuf.c -15 +30 M ./dft/dftw-generic.c -14 +31 M ./dft/kdft-dif.c -7 +3 M ./dft/kdft-difsq.c -2 +2 M ./dft/kdft-dit.c -7 +3 M ./kernel/ifftw.h -1 +2 M ./kernel/twiddle.c -1 +10 M ./rdft/Makefile.am -1 +1 M ./rdft/conf.c -2 +2 R ./rdft/ct.c R ./rdft/ct.h M ./rdft/hc2hc-common.c -4 +33 M ./rdft/hc2hc-direct.c -26 +42 M ./rdft/hc2hc-directbuf.c -25 +44 M ./rdft/hc2hc-generic.c -37 +104 M ./rdft/hc2hc.c +221 M ./rdft/hc2hc.h +60 M ./rdft/khc2hc.c -8 +4 M ./rdft/rdft.h -3 +1 M ./threads/Makefile.am -2 +2 R ./threads/ct-dit.c M ./threads/ct.c +253 M ./threads/dft-vrank-geq1.c -2 +2 R ./threads/hc2hc-dif.c R ./threads/hc2hc-dit.c M ./threads/hc2hc.c +238 M ./threads/threads.c -27 +8 M ./threads/threads.h -5 +4 Thu Oct 21 20:44:51 EDT 2004 stevenj * [project @ 2004-10-22 00:44:51 by stevenj] more C++ notes M ./doc/fftw3.texi -4 +11 Thu Oct 14 09:50:38 EDT 2004 stevenj * [project @ 2004-10-14 13:50:38 by stevenj] note bug report for VC++ 6.0 from Dale Dickerhoof M ./doc/FAQ/fftw-faq.bfnn -1 +4 Fri Oct 1 16:06:59 EDT 2004 stevenj * [project @ 2004-10-01 20:06:59 by stevenj] fmt M ./api/fftw3.h -3 +3 Fri Oct 1 15:59:17 EDT 2004 stevenj * [project @ 2004-10-01 19:58:50 by stevenj] comment typo M ./rdft/vrank3-transpose.c -3 +3 Fri Oct 1 15:48:09 EDT 2004 stevenj * [project @ 2004-10-01 19:48:09 by stevenj] bug fix -- ishift/oshift only apply to execution of child plan M ./rdft/dft-r2hc.c -9 +4 Thu Sep 30 21:12:47 EDT 2004 athena * [project @ 2004-10-01 01:12:47 by athena] New planner that tries never to lose wisdom. M ./api/fftw3.h -2 +1 M ./api/mapflags.c -1 M ./kernel/ifftw.h -6 +4 M ./kernel/planner.c -137 +183 Thu Sep 30 13:36:43 EDT 2004 athena * [project @ 2004-09-30 17:36:43 by athena] Nested comment was triggering a warning. M ./api/fftw3.h -2 +2 Fri Sep 10 15:20:07 EDT 2004 stevenj * [project @ 2004-09-10 19:20:07 by stevenj] system "root" under dgjpp is /dev/env/DJDIR, not /dev/env/DJGPP, according to djgpp's libc.info; patch confirmed with J. M. Guerrero M ./api/import-system-wisdom.c -1 +1 Wed Sep 8 18:50:03 EDT 2004 stevenj * [project @ 2004-09-08 22:50:03 by stevenj] some minor portability fixes for djgpp; thanks to Juan Manuel Guerrero for the patch M ./api/import-system-wisdom.c -1 +8 M ./tests/Makefile.am -4 +4 M ./tools/fftw-wisdom-to-conf.in -1 +1 Thu Aug 19 12:41:23 EDT 2004 stevenj * [project @ 2004-08-19 16:41:23 by stevenj] pointer to tutorial for quick start M ./README +3 Thu Aug 19 12:39:50 EDT 2004 stevenj * [project @ 2004-08-19 16:39:50 by stevenj] point users to manual M ./api/fftw3.h -1 +16 Sat Aug 7 13:42:22 EDT 2004 stevenj * [project @ 2004-08-07 17:42:22 by stevenj] minor typo M ./doc/fftw3.texi -2 +2 Sun Jul 18 18:54:18 EDT 2004 stevenj * [project @ 2004-07-18 22:54:18 by stevenj] use __DECCXX for Compaq cxx, not Linux-specific symbol M ./kernel/cycle.h -2 +2 Fri Jul 16 13:55:25 EDT 2004 stevenj * [project @ 2004-07-16 17:55:25 by stevenj] patch by John Bowman to make cycle counter work with DEC cxx under Linux M ./kernel/cycle.h -2 +2 Wed Jun 30 00:45:10 EDT 2004 stevenj * [project @ 2004-06-30 04:45:10 by stevenj] updated pruned FFT discussion, with link to further details on www.fftw.org/pruned.html M ./doc/FAQ/fftw-faq.bfnn -8 +16 M ./doc/FAQ/html.refs +1 Mon Jun 14 20:08:27 EDT 2004 stevenj * [project @ 2004-06-15 00:08:27 by stevenj] darwin is based on freebsd M ./acx_pthread.m4 -2 +2 Thu Jun 3 14:23:41 EDT 2004 stevenj * [project @ 2004-06-03 18:23:41 by stevenj] in --with-windows-f77-mangling, add lowercase + single underscore for Intel compilers, etc. (thanks to David Gomez for the bug report) M ./api/f77api.c +5 Wed Apr 7 00:46:07 EDT 2004 stevenj * [project @ 2004-04-07 04:46:07 by stevenj] whoops, extra alignment check M ./rdft/rank0-vrank2.c -1 +5 Wed Apr 7 00:16:49 EDT 2004 stevenj * [project @ 2004-04-07 04:16:49 by stevenj] disable most 2-float-as-double copying, add alignment check in one remaining place M ./kernel/ifftw.h -1 +11 M ./rdft/rank0-vrank2.c -22 +25 M ./rdft/vrank3-transpose.c -6 +6 Tue Apr 6 13:49:13 EDT 2004 stevenj * [project @ 2004-04-06 17:49:13 by stevenj] make sure it is clear that real-even/odd refers to symmetry, not size M ./doc/fftw3.texi -7 +7 Mon Apr 5 20:18:29 EDT 2004 stevenj * [project @ 2004-04-06 00:18:29 by stevenj] optimization M ./rdft/vrank3-transpose.c -11 +48 Fri Apr 2 21:31:00 EST 2004 stevenj * [project @ 2004-04-03 02:31:00 by stevenj] separate cutoff for ugliness...these cutoffs are still not ideal M ./rdft/vrank3-transpose.c -3 +4 Fri Apr 2 21:30:17 EST 2004 stevenj * [project @ 2004-04-03 02:30:17 by stevenj] transpose.c is gone M ./kernel/ifftw.h -15 +1 Fri Apr 2 21:18:27 EST 2004 stevenj * [project @ 2004-04-03 02:18:27 by stevenj] move all rank0 transforms to rdft A ./rdft/rank0-vrank2.c A ./rdft/vrank3-transpose.c M ./configure.ac +1 M ./dft/Makefile.am -2 +2 M ./dft/conf.c -4 +1 R ./dft/rank0.c R ./dft/vrank2-transpose.c R ./dft/vrank3-transpose.c M ./kernel/Makefile.am -2 +2 R ./kernel/transpose.c M ./rdft/Makefile.am -2 +3 M ./rdft/conf.c -1 +3 M ./rdft/dft-r2hc.c -20 +36 M ./rdft/rank0-vrank2.c +289 M ./rdft/rdft.h -2 +2 M ./rdft/vrank3-transpose.c +943 Fri Apr 2 20:35:35 EST 2004 stevenj * [project @ 2004-04-03 01:35:35 by stevenj] enable fp-moves/us comparison of rank-0 transforms M ./libbench2/mflops.c -2 +8 M ./libbench2/report.c -3 +7 Thu Apr 1 16:13:22 EST 2004 stevenj * [project @ 2004-04-01 21:13:22 by stevenj] whoops M ./kernel/transpose.c -1 +1 Thu Apr 1 15:25:30 EST 2004 stevenj * [project @ 2004-04-01 20:25:30 by stevenj] whoops M ./kernel/tensor7.c -3 +3 Wed Mar 31 18:11:02 EST 2004 stevenj * [project @ 2004-03-31 23:11:02 by stevenj] sort tensor dims by stride absolute values, not strides M ./kernel/tensor7.c -5 +7 Tue Mar 30 20:22:50 EST 2004 stevenj * [project @ 2004-03-31 01:22:50 by stevenj] M ./kernel/transpose.c -2 +1 Tue Mar 30 19:44:54 EST 2004 stevenj * [project @ 2004-03-31 00:44:54 by stevenj] added improved transpose algorithm for N x M where |N-M| is small M ./dft/dftw-generic.c +1 M ./dft/vrank2-transpose.c -7 +10 M ./dft/vrank3-transpose.c -6 +7 M ./kernel/ifftw.h -2 +2 M ./kernel/transpose.c -15 +73 M ./libbench2/problem.c -2 +4 Tue Mar 30 19:41:14 EST 2004 stevenj * [project @ 2004-03-31 00:41:14 by stevenj] check to make sure SIMD matches precision, and make sure user doesn't select both SSE and SSE2 M ./configure.ac -2 +7 Sun Mar 28 09:26:38 EST 2004 athena * [project @ 2004-03-28 14:26:38 by athena] Implemented hc2hc-generic hc2r. M ./rdft/hc2hc-generic.c -29 +79 Thu Mar 25 11:19:25 EST 2004 athena * [project @ 2004-03-25 16:19:25 by athena] Inverted loop for stride-1 access. M ./rdft/hc2hc-generic.c -12 +18 Thu Mar 25 11:18:49 EST 2004 athena * [project @ 2004-03-25 16:18:49 by athena] Swapped j <-> k for consistency M ./dft/dftw-generic.c -3 +3 Tue Mar 23 12:08:07 EST 2004 athena * [project @ 2004-03-23 17:08:07 by athena] Require that R be odd M ./rdft/hc2hc-generic.c -1 +1 Tue Mar 23 11:49:01 EST 2004 athena * [project @ 2004-03-23 16:49:01 by athena] Implemented hc2hc-generic (DIT only for now). A ./rdft/hc2hc-generic.c M ./rdft/Makefile.am -3 +3 M ./rdft/conf.c -5 +4 M ./rdft/dft-r2hc.c -2 +2 M ./rdft/hc2hc-generic.c +216 M ./rdft/rdft.h +1 Mon Mar 22 14:43:16 EST 2004 athena * [project @ 2004-03-22 19:43:16 by athena] Relax equality of twiddle description, since the `i' field is not used by TW_FULL or TW_HALF. M ./kernel/twiddle.c -5 +17 Mon Mar 22 13:22:44 EST 2004 athena * [project @ 2004-03-22 18:22:41 by athena] Do not allocate tw_instr's on the stack. Thus, the ``consistency check'' in twiddle.c becomes wrong. M ./dft/dftw-generic.c -8 +1 M ./kernel/twiddle.c -2 +1 Mon Mar 22 13:21:28 EST 2004 athena * [project @ 2004-03-22 18:21:28 by athena] Fixed incorrect malloc()/free() logic. M ./libbench2/mp.c -3 +6 Mon Mar 22 09:04:37 EST 2004 athena * [project @ 2004-03-22 14:04:37 by athena] Silence warnings M ./rdft/hc2hc-directbuf.c -2 +2 Mon Mar 22 09:02:55 EST 2004 athena * [project @ 2004-03-22 14:02:55 by athena] Separate file for hc2hc common routines A ./rdft/hc2hc-common.c M ./rdft/Makefile.am -4 +4 M ./rdft/ct.c -35 M ./rdft/hc2hc-common.c +58 Mon Mar 22 08:23:56 EST 2004 athena * [project @ 2004-03-22 13:23:56 by athena] (re)Implemented buffered hc2hc. Slight simplification of twiddle-factors management. A ./rdft/hc2hc-directbuf.c M ./dft/dftw-directbuf.c -2 +2 M ./rdft/Makefile.am -2 +3 M ./rdft/ct.h +2 M ./rdft/hc2hc-direct.c -16 +8 M ./rdft/hc2hc-directbuf.c +238 M ./rdft/khc2hc.c -1 +2 Sun Mar 21 19:53:05 EST 2004 athena * [project @ 2004-03-22 00:53:05 by athena] Incremented libtool revision number before we forget. M ./configure.ac -1 +1 Sun Mar 21 19:25:56 EST 2004 athena * [project @ 2004-03-22 00:25:56 by athena] Fixed opcnt M ./rdft/hc2hc-direct.c -2 +5 Sun Mar 21 17:56:15 EST 2004 athena * [project @ 2004-03-21 22:56:15 by athena] Renamed files. These solvers are not really cooley-tukey. A ./dft/dftw-direct.c A ./dft/dftw-directbuf.c A ./dft/dftw-generic.c M ./dft/Makefile.am -5 +5 R ./dft/ct-directw.c R ./dft/ct-directwbuf.c R ./dft/ct-generic.c M ./dft/dftw-direct.c +153 M ./dft/dftw-directbuf.c +213 M ./dft/dftw-generic.c +304 Sun Mar 21 12:38:45 EST 2004 athena * [project @ 2004-03-21 17:38:45 by athena] Started moving rdft/ to the new cooley-tukey ontology A ./rdft/ct.c A ./rdft/ct.h A ./rdft/hc2hc-direct.c A ./rdft/khc2hc.c M ./dft/ct.h -1 +1 M ./genfft/gen_hc2hc.ml -6 +3 M ./rdft/Makefile.am -5 +4 M ./rdft/codelet-rdft.h -3 +1 M ./rdft/ct.c +253 M ./rdft/ct.h +54 R ./rdft/hc2hc-buf.c R ./rdft/hc2hc-dif.c M ./rdft/hc2hc-direct.c +182 R ./rdft/hc2hc-dit.c R ./rdft/hc2hc.c R ./rdft/hc2hc.h R ./rdft/khc2hc-dif.c R ./rdft/khc2hc-dit.c M ./rdft/khc2hc.c +32 M ./rdft/rdft.h -6 +1 Sun Mar 21 10:59:42 EST 2004 athena * [project @ 2004-03-21 15:59:42 by athena] Plans in ct-*.c are subtypes of plan_dftw, not plan_dft M ./dft/ct-directw.c -2 +2 M ./dft/ct-directwbuf.c -2 +2 M ./dft/ct-generic.c -1 +1 Sun Mar 21 10:38:18 EST 2004 athena * [project @ 2004-03-21 15:38:18 by athena] Slight simplification M ./dft/ct-directw.c -2 +2 Sun Mar 21 10:20:06 EST 2004 athena * [project @ 2004-03-21 15:20:06 by athena] Minor simplification M ./dft/ct.c -4 +2 Sat Mar 20 08:43:57 EST 2004 athena * [project @ 2004-03-20 13:43:57 by athena] Workarounds for icc-8.0 nonsense. M ./simd/simd-sse.h +6 M ./simd/simd-sse2.h +6 Sun Mar 7 07:56:08 EST 2004 athena * [project @ 2004-03-07 12:56:08 by athena] FFTW_FORWARD is not technically an ``option''. M ./doc/fftw3.texi -2 +2 Tue Feb 24 12:17:06 EST 2004 stevenj * [project @ 2004-02-24 17:17:06 by stevenj] Alejandro requested that his name be removed from @author M ./acx_pthread.m4 -4 +5 Mon Feb 23 17:42:56 EST 2004 stevenj * [project @ 2004-02-23 22:42:56 by stevenj] GNU Pth emulation library check M ./acx_pthread.m4 -3 +12 Sat Feb 21 17:51:13 EST 2004 stevenj * [project @ 2004-02-21 22:51:13 by stevenj] calling can-do calls the estimating-planner, which creates wisdom that we don't want ...we should be able to do all of the documented problems, anyway M ./tools/fftw-wisdom.c -1 +1 Sat Feb 21 17:46:06 EST 2004 stevenj * [project @ 2004-02-21 22:46:06 by stevenj] don't forget_wisdom because of side effects M ./tests/bench.c -1 Sat Feb 21 17:42:47 EST 2004 stevenj * [project @ 2004-02-21 22:42:47 by stevenj] forget wisdom from can_do M ./tests/bench.c +1 Thu Feb 19 14:11:14 EST 2004 stevenj * [project @ 2004-02-19 19:11:14 by stevenj] parenthesization M ./api/malloc.c -1 +1 Fri Feb 13 07:20:31 EST 2004 athena * [project @ 2004-02-13 12:20:31 by athena] Split malloc into kernel_malloc and API malloc A ./api/malloc.c A ./kernel/kalloc.c M ./api/Makefile.am -14 +14 M ./api/malloc.c +32 M ./kernel/Makefile.am -5 +5 M ./kernel/alloc.c -127 +5 M ./kernel/ifftw.h -1 +6 M ./kernel/kalloc.c +140 M ./tests/bench.c +4 Thu Feb 12 15:42:20 EST 2004 stevenj * [project @ 2004-02-12 20:42:20 by stevenj] X(malloc) must be extern "C" M ./kernel/alloc.c -1 +4 Thu Feb 12 15:41:44 EST 2004 stevenj * [project @ 2004-02-12 20:41:44 by stevenj] satsify C++ compiler M ./dft/bluestein.c -2 +2 Thu Feb 5 20:39:14 EST 2004 stevenj * [project @ 2004-02-06 01:39:14 by stevenj] with the new flags, fma is definitely beneficial on PA-RISC with HP/UX cc M ./doc/FAQ/fftw-faq.bfnn -6 +7 Thu Feb 5 19:52:17 EST 2004 stevenj * [project @ 2004-02-06 00:52:17 by stevenj] grr, Ofaster etcetera are not supported under older versions of the compiler. Note that +Ofltacc *disables* fp-reordering optimizations (which are enabled by +Oall). +Optrs_ansi is the older version of the aliasing stuff M ./acinclude.m4 -1 +1 Thu Feb 5 19:26:01 EST 2004 stevenj * [project @ 2004-02-06 00:26:01 by stevenj] +Otype_safety=ansi on hpux M ./acinclude.m4 -1 +1 Thu Feb 5 19:22:34 EST 2004 stevenj * [project @ 2004-02-06 00:22:34 by stevenj] just use +Ofaster on hpux (+O3 +Onolimit +Olibcalls +Ofltacc=relaxed -Wl,+mergeseg) M ./acinclude.m4 -1 +1 Fri Jan 30 14:17:15 EST 2004 stevenj * [project @ 2004-01-30 19:17:15 by stevenj] check for win32 threads for mingw32; thanks to Alessio Massaro M ./configure.ac +8 Thu Jan 29 15:23:33 EST 2004 stevenj * [project @ 2004-01-29 20:23:33 by stevenj] added missing 'static', thanks to Alessio Massaro M ./threads/threads.c -3 +3 Fri Jan 9 16:36:48 EST 2004 stevenj * [project @ 2004-01-09 21:36:48 by stevenj] print more like bluestein M ./rdft/dht-rader.c -1 +1 Fri Jan 9 15:45:22 EST 2004 stevenj * [project @ 2004-01-09 20:45:22 by stevenj] fixed op count for R2HC_ONLY_CONV M ./rdft/dht-rader.c -3 +2 Fri Jan 9 15:41:50 EST 2004 stevenj * [project @ 2004-01-09 20:41:50 by stevenj] include DESTROY_INPUT in buffered flags for in-place...otherwise in-place hc2r uses rdft-dhtcvs diff M ./dft/buffered.c -1 +10 M ./rdft/buffered.c -1 +10 M ./rdft/buffered2.c -1 +9 Fri Jan 9 15:41:09 EST 2004 stevenj * [project @ 2004-01-09 20:41:09 by stevenj] resurrected R2HC_ONLY_CONV option to share plans and save on planning time M ./rdft/dht-rader.c -3 +46 Fri Jan 9 14:47:00 EST 2004 stevenj * [project @ 2004-01-09 19:47:00 by stevenj] precompute folding for cyclic convolution M ./rdft/dht-rader.c -13 +9 Wed Jan 7 16:48:39 EST 2004 stevenj * [project @ 2004-01-07 21:48:39 by stevenj] minor M ./doc/FAQ/fftw-faq.bfnn -1 +1 Wed Jan 7 16:48:25 EST 2004 stevenj * [project @ 2004-01-07 21:48:25 by stevenj] note reports of successful compilation on Windows M ./doc/FAQ/fftw-faq.bfnn -5 +5 Wed Jan 7 14:16:16 EST 2004 stevenj * [project @ 2004-01-07 19:16:16 by stevenj] citation year M ./reodft/reodft010e-r2hc.c -2 +2 Tue Jan 6 01:07:36 EST 2004 stevenj * [project @ 2004-01-06 06:07:36 by stevenj] comment M ./rdft/dht-rader.c -1 +1 Tue Jan 6 01:07:08 EST 2004 stevenj * [project @ 2004-01-06 06:07:08 by stevenj] comment fix M ./rdft/dht-rader.c -1 +1 Tue Jan 6 01:06:57 EST 2004 stevenj * [project @ 2004-01-06 06:06:57 by stevenj] fixed naming cruft M ./rdft/dht-rader.c -18 +18 Tue Jan 6 00:56:16 EST 2004 stevenj * [project @ 2004-01-06 05:56:16 by stevenj] space M ./rdft/dht-rader.c -1 Tue Jan 6 00:55:53 EST 2004 stevenj * [project @ 2004-01-06 05:55:45 by stevenj] comment M ./rdft/dht-rader.c -1 +1 Tue Jan 6 00:54:07 EST 2004 stevenj * [project @ 2004-01-06 05:54:07 by stevenj] moved assert M ./rdft/dht-rader.c -1 +1 Tue Jan 6 00:49:16 EST 2004 stevenj * [project @ 2004-01-06 05:49:16 by stevenj] comment M ./rdft/dht-rader.c +5 Tue Jan 6 00:41:06 EST 2004 stevenj * [project @ 2004-01-06 05:41:06 by stevenj] delete old R2HC_ONLY_CONV hack, now defunct M ./rdft/dht-rader.c -46 Tue Jan 6 00:32:58 EST 2004 stevenj * [project @ 2004-01-06 05:32:58 by stevenj] added padded real rader M ./rdft/dht-rader.c -56 +95 Mon Jan 5 22:56:58 EST 2004 stevenj * [project @ 2004-01-06 03:56:58 by stevenj] removed unused var M ./rdft/generic.c -1 +1 Mon Jan 5 21:20:29 EST 2004 stevenj * [project @ 2004-01-06 02:20:29 by stevenj] handle both FFT_SIGN values M ./rdft/generic.c -2 +14 Fri Jan 2 06:07:51 EST 2004 athena * [project @ 2004-01-02 11:07:51 by athena] Oops: d->ros ==> d->ios M ./rdft/codelets/r2hc.c -1 +1 Fri Jan 2 06:05:10 EST 2004 athena * [project @ 2004-01-02 11:05:10 by athena] Oops: d->ris should have been d->iis M ./rdft/codelets/hc2r.c -1 +1 Thu Jan 1 16:00:07 EST 2004 athena * [project @ 2004-01-01 21:00:07 by athena] Removed rdft rader cooley-tukey, to be superseded by a generic reduction of rdft twiddle problems to dft + pre/post processing M ./dft/Makefile.am -2 +2 M ./dft/dft.h -5 +1 R ./dft/rader-omega.c M ./dft/rader.c -2 +38 M ./rdft/Makefile.am -2 +2 M ./rdft/conf.c -2 +1 R ./rdft/rader-hc2hc.c Thu Jan 1 15:44:09 EST 2004 athena * [project @ 2004-01-01 20:44:09 by athena] In anticipation of the upcoming revision of rdft, removed rdft generic dit/dif cooley-tukey, in favor of generic rh2c and hc2r solvers. Cleaned up stuff that became unused after this change, such as TW_GENERIC. M ./dft/ct.c -33 +3 M ./dft/generic.c -1 +2 M ./kernel/ifftw.h -2 +3 M ./kernel/primes.c -1 +32 M ./kernel/twiddle.c -18 +1 M ./rdft/generic.c -258 +103 Thu Jan 1 12:59:30 EST 2004 athena * [project @ 2004-01-01 17:59:30 by athena] Removed useless file M ./kernel/Makefile.am -3 +3 M ./kernel/ifftw.h -2 +1 R ./kernel/square.c Fri Dec 26 13:54:00 EST 2003 stevenj * [project @ 2003-12-26 18:54:00 by stevenj] whoops, don't call AC_F77_DUMMY_MAIN if no Fortran compiler is found; thanks to Charles Radley for the bug report. M ./configure.ac -2 +3 Fri Dec 19 13:58:05 EST 2003 stevenj * [project @ 2003-12-19 18:58:05 by stevenj] guess good flags for Solaris/intel, suggested by J. Gregory Wright M ./acinclude.m4 -1 +1 Fri Dec 5 19:55:13 EST 2003 stevenj * [project @ 2003-12-06 00:55:13 by stevenj] blah M ./doc/FAQ/fftw-faq.bfnn -1 +1 M ./doc/FAQ/html.refs -1 +1 Sun Nov 30 06:59:41 EST 2003 athena * [project @ 2003-11-30 11:59:41 by athena] DIF generic solver was destroying the input. M ./rdft/generic.c -2 +16 Sat Nov 29 19:28:39 EST 2003 athena * [project @ 2003-11-30 00:28:39 by athena] Fixed bug that caused HC2R transforms to destroy the input in certain cases, even if the user specified FFTW_PRESERVE_INPUT. M ./NEWS +5 M ./rdft/rader-hc2hc.c -2 +14 Sat Nov 29 16:49:01 EST 2003 athena * [project @ 2003-11-29 21:49:01 by athena] Implemented swap_io hack for r2r verifier. M ./libbench2/verify-r2r.c -1 +8 Thu Nov 20 22:00:53 EST 2003 stevenj * [project @ 2003-11-21 03:00:53 by stevenj] citation M ./reodft/reodft010e-r2hc.c -2 +6 Fri Nov 14 20:57:55 EST 2003 athena * [project @ 2003-11-15 01:57:55 by athena] Trying to get ``make paranoid-check'' to work. (Still broken.) M ./kernel/ifftw.h -2 +3 M ./kernel/planner.c -2 +2 M ./tests/hook.c -9 +8 Fri Nov 14 20:05:54 EST 2003 stevenj * [project @ 2003-11-15 01:05:54 by stevenj] fixes for input-preservation tests M ./libbench2/bench-user.h -1 +2 M ./libbench2/tensor.c -1 +15 M ./libbench2/verify-dft.c -3 +12 M ./libbench2/verify-lib.c -1 +4 M ./libbench2/verify-r2r.c -13 +33 M ./libbench2/verify-rdft2.c -6 +22 M ./libbench2/verify.h -1 +1 Fri Nov 14 19:19:31 EST 2003 athena * [project @ 2003-11-15 00:19:31 by athena] Assume FFTW_PRESERVE_INPUT unless either the `d' flag is given in the problem, or the problem is multidimensional c2r (which fftw3 cannot without destroying the input). With this change, we can at least test that FFTW_PRESERVE_INPUT works in the c2r 1d case. M ./tests/bench.c -12 +21 Fri Nov 14 19:14:40 EST 2003 stevenj * [project @ 2003-11-15 00:14:40 by stevenj] apply should copy back input for input-preservation check M ./libbench2/verify-dft.c -1 +3 M ./libbench2/verify-r2r.c +13 M ./libbench2/verify-rdft2.c -1 +4 Fri Nov 14 19:01:36 EST 2003 athena * [project @ 2003-11-15 00:01:14 by athena] Undone previous bogus changes M ./rdft/rank-geq2-rdft2.c -1 +10 M ./tests/bench.c -14 +10 M ./tests/check.pl -1 Fri Nov 14 18:27:12 EST 2003 athena * [project @ 2003-11-14 23:27:12 by athena] Check dr[fb] in addition to r[fb] M ./tests/check.pl +1 Fri Nov 14 17:33:44 EST 2003 athena * [project @ 2003-11-14 22:33:42 by athena] Fixed conditions under which the rank-geq2-rdft2 solver is applicable. The old solver was not applicable for out-of-place problems unless DESTROY_INPUT. This is bogus. As long as the subsolvers honor !DESTROY_INPUT, the solver is always applicable. Changed semantics of test program, so that PRESERVE_INPUT is always true unless the problem specifies destroy_input explicitly. Without this change, there is no way to test the new solver. M ./rdft/rank-geq2-rdft2.c -10 +1 M ./tests/bench.c -10 +14 Thu Oct 30 15:10:42 EST 2003 stevenj * [project @ 2003-10-30 20:10:42 by stevenj] added AIX OpenMP (-qsmp=omp) support; thanks to Greg Bauer M ./configure.ac +5 Thu Oct 30 10:11:39 EST 2003 athena * [project @ 2003-10-30 15:11:39 by athena] G5 CFLAGS M ./acinclude.m4 +6 Fri Oct 24 04:17:39 EDT 2003 stevenj * [project @ 2003-10-24 08:17:39 by stevenj] western FAQ M ./doc/FAQ/fftw-faq.bfnn +5 Thu Oct 23 11:34:11 EDT 2003 athena * [project @ 2003-10-23 15:34:11 by athena] Oops. M ./simd/altivec.c -1 +2 Thu Oct 23 11:28:28 EDT 2003 athena * [project @ 2003-10-23 15:28:27 by athena] Autodetect altivec M ./configure.ac -2 +3 M ./simd/altivec.c -1 +28 M ./simd/simd-altivec.h -1 +2 Wed Oct 22 01:14:10 EDT 2003 stevenj * [project @ 2003-10-22 05:14:10 by stevenj] MinGW gets confused by a single / M ./tests/check.pl -4 +4 Fri Oct 17 10:46:41 EDT 2003 athena * [project @ 2003-10-17 14:46:41 by athena] Paranoid portability fix M ./libbench2/mp.c -2 +2 Thu Oct 16 11:07:46 EDT 2003 athena * [project @ 2003-10-16 15:07:46 by athena] size -> length, which should make clear that we are not talking about arbitrary precision. M ./doc/fftw3.texi -4 +4 Wed Oct 15 15:01:40 EDT 2003 stevenj * [project @ 2003-10-15 19:01:40 by stevenj] pruned transforms are a FAQ M ./doc/FAQ/fftw-faq.bfnn +11 Wed Oct 8 23:54:17 EDT 2003 stevenj * [project @ 2003-10-09 03:54:17 by stevenj] NO_SEARCH has already been mapped to FFTW_WISDOM_ONLY M ./TODO -1 +1 Wed Oct 8 23:53:19 EDT 2003 stevenj * [project @ 2003-10-09 03:53:19 by stevenj] newline M ./TODO +1 Sat Sep 27 20:27:32 EDT 2003 stevenj * [project @ 2003-09-28 00:27:32 by stevenj] fix M ./doc/fftw3.texi -2 +2 Sat Sep 27 20:24:39 EDT 2003 stevenj * [project @ 2003-09-28 00:24:39 by stevenj] clarification M ./doc/fftw3.texi -2 +4 Sat Sep 27 17:43:57 EDT 2003 stevenj * [project @ 2003-09-27 21:43:57 by stevenj] minor fix M ./doc/fftw3.texi -2 +2 Sat Sep 27 17:42:30 EDT 2003 stevenj * [project @ 2003-09-27 21:42:30 by stevenj] grammar M ./doc/fftw3.texi -2 +2 Sat Sep 27 17:29:04 EDT 2003 stevenj * [project @ 2003-09-27 21:29:04 by stevenj] html output fix M ./doc/fftw3.texi -5 +5 Sat Sep 27 17:22:48 EDT 2003 stevenj * [project @ 2003-09-27 21:22:48 by stevenj] mentioned sqrt(2) factors for DCT/DST M ./doc/fftw3.texi -5 +48 Sat Sep 27 17:07:18 EDT 2003 stevenj * [project @ 2003-09-27 21:07:18 by stevenj] FFTW_WISDOM_ONLY flag (undocumented for now), suggested by Phil Dumont M ./api/fftw3.h -1 +2 M ./api/mapflags.c +1 Tue Sep 23 23:36:19 EDT 2003 stevenj * [project @ 2003-09-24 03:36:19 by stevenj] removed UpTime code M ./kernel/cycle.h -20 +3 Tue Sep 23 23:27:29 EDT 2003 stevenj * [project @ 2003-09-24 03:27:29 by stevenj] updated documentation for mach_absolute_time M ./kernel/cycle.h -3 +3 Tue Sep 23 23:25:52 EDT 2003 stevenj * [project @ 2003-09-24 03:25:52 by stevenj] use mach_absolute_time on MacOS/Darwin, as a fallback; don't bother checking for UpTime since it requires extra libs M ./configure.ac -3 +2 M ./kernel/cycle.h -1 +11 Tue Sep 23 22:59:29 EDT 2003 stevenj * [project @ 2003-09-24 02:59:29 by stevenj] support Apple UpTime function for asm-less xlc, grrr... M ./configure.ac -2 +3 M ./kernel/cycle.h -1 +19 Tue Sep 23 15:42:29 EDT 2003 stevenj * [project @ 2003-09-23 19:42:29 by stevenj] additional paranoia for xlc etc. M ./api/api.h -1 +1 M ./api/fftw3.h -2 +2 Mon Sep 22 15:28:56 EDT 2003 stevenj * [project @ 2003-09-22 19:28:56 by stevenj] work around _Complex_I weirdness in xlc, reported by Greg Allen M ./api/api.h -2 +4 M ./api/fftw3.h -2 +2 Fri Sep 5 18:03:11 EDT 2003 stevenj * [project @ 2003-09-05 22:03:11 by stevenj] typo M ./doc/FAQ/fftw-faq.bfnn -2 +2 Fri Sep 5 13:11:40 EDT 2003 athena * [project @ 2003-09-05 17:11:40 by athena] New script that produces commercial version. A ./commercialize.sh Fri Sep 5 07:27:06 EDT 2003 athena * [project @ 2003-09-05 11:27:06 by athena] Noted that VC++ is buggy. Noted that we know nothing about Windows. Noted that the sky is blue as well. M ./doc/FAQ/fftw-faq.bfnn -10 +22 Tue Sep 2 09:04:19 EDT 2003 athena * [project @ 2003-09-02 13:04:19 by athena] Noted that certain arrays are no longer used after the planner has completed. M ./doc/fftw3.texi -2 +12 Tue Aug 26 08:22:38 EDT 2003 athena * [project @ 2003-08-26 12:22:38 by athena] Typo M ./doc/fftw3.texi -2 +2 Mon Aug 25 21:27:43 EDT 2003 athena * [project @ 2003-08-26 01:27:43 by athena] New item M ./TODO -2 +1 Thu Aug 21 17:36:08 EDT 2003 stevenj * [project @ 2003-08-21 21:36:08 by stevenj] try creating output file before planning (thanks to Phil Dumont for the suggestion) M ./tools/fftw-wisdom.c -10 +10 Tue Aug 19 10:08:07 EDT 2003 athena * [project @ 2003-08-19 14:08:07 by athena] Clarified fftw_cleanup() M ./doc/fftw3.texi -4 +7 Sat Aug 16 03:13:41 EDT 2003 stevenj * [project @ 2003-08-16 07:13:41 by stevenj] typo M ./doc/fftw3.texi -2 +2 Mon Jul 28 18:01:13 EDT 2003 stevenj * [project @ 2003-07-28 22:01:13 by stevenj] use time() instead of clock() (FIXME: what to do for non-POSIX systems?) ...thanks to JP Sugarbroad and James A. Treacy for the bug report M ./tools/fftw-wisdom.c -4 +4 Thu Jul 24 18:58:10 EDT 2003 athena * [project @ 2003-07-24 22:58:10 by athena] Need __volatile__ in sparc cycle counter. This is why the debian port hangs. M ./kernel/cycle.h -2 +2 Sun Jul 20 16:02:43 EDT 2003 stevenj * [project @ 2003-07-20 20:02:43 by stevenj] merged 3.0.1 notes M ./NEWS +30 Sun Jul 13 20:57:34 EDT 2003 stevenj * [project @ 2003-07-14 00:57:34 by stevenj] whoops M ./libbench2/bench-main.c -2 +2 Thu Jul 10 11:48:50 EDT 2003 athena * [project @ 2003-07-10 15:48:50 by athena] Dealing with constants in a way that seems to confuse gcc less. M ./simd/simd-sse.h -5 +2 M ./simd/simd-sse2.h -6 +3 Wed Jul 9 17:39:23 EDT 2003 athena * [project @ 2003-07-09 21:39:16 by athena] Enabled scheduler hack for FMA, where it seems to help. M ./genfft/annotate.ml -7 +3 M ./genfft/magic.ml -1 +5 M ./support/Makefile.codelets -1 +1 Wed Jul 9 12:57:38 EDT 2003 athena * [project @ 2003-07-09 16:57:38 by athena] Hmm---the new scheduler seems make things worse for gcc/x86, better for gcc/ppc, and about the same for icc/x86. Disabled for now. M ./genfft/annotate.ml -3 +6 Wed Jul 9 08:09:53 EDT 2003 athena * [project @ 2003-07-09 12:09:53 by athena] New scheduling pass that keeps ``x = a + b'' and ``y = a - b'' close together. This property was no longer automatic for the dags generated in SIMD mode. I cannot measure any speed difference due to this change. However, the change is justified by a minimal-screwup argument. Moreover, the sse2 fftw library is now 1% smaller than it was before. M ./genfft/annotate.ml -32 +70 Tue Jul 8 20:42:22 EDT 2003 athena * [project @ 2003-07-09 00:42:22 by athena] -(FNMS()) => FMS() M ./genfft/c.ml -1 +2 Sun Jul 6 13:53:23 EDT 2003 stevenj * [project @ 2003-07-06 17:53:23 by stevenj] added more convenient target name M ./doc/FAQ/Makefile.am +2 Sat Jul 5 13:30:10 EDT 2003 stevenj * [project @ 2003-07-05 17:30:10 by stevenj] typo M ./kernel/ifftw.h -2 +2 Sat Jul 5 13:19:36 EDT 2003 athena * [project @ 2003-07-05 17:19:36 by athena] Consistent naming M ./dft/ct-generic.c -2 +2 Sat Jul 5 13:05:51 EDT 2003 athena * [project @ 2003-07-05 17:05:51 by athena] Got rid of problemw. A ./dft/ct-directw.c A ./dft/ct-directwbuf.c A ./dft/ct-generic.c A ./dft/ct.h M ./dft/Makefile.am -5 +4 M ./dft/conf.c -3 +2 M ./dft/ct-directw.c +153 M ./dft/ct-directwbuf.c +213 M ./dft/ct-generic.c +304 M ./dft/ct.c -62 +24 M ./dft/ct.h +56 M ./dft/ctsq.c -3 +3 M ./dft/dft.h -39 +2 R ./dft/dftw-dft.c M ./dft/direct.c -2 +1 R ./dft/directw.c R ./dft/directwbuf.c M ./dft/generic.c -2 M ./dft/kdft-dif.c -4 +4 M ./dft/kdft-difsq.c -3 +3 M ./dft/kdft-dit.c -4 +4 M ./dft/plan.c -11 +1 R ./dft/problemw.c M ./dft/rader.c -2 M ./dft/solve.c -8 +1 Fri Jul 4 06:56:26 EDT 2003 athena * [project @ 2003-07-04 10:56:26 by athena] Increase TIME_MIN on intel only M ./kernel/cycle.h -1 +3 M ./kernel/timer.c -2 +2 Fri Jul 4 06:36:02 EDT 2003 athena * [project @ 2003-07-04 10:36:02 by athena] A little hack to get more consistent scheduling. M ./genfft/schedule.ml -7 +8 Thu Jul 3 16:47:42 EDT 2003 athena * [project @ 2003-07-03 20:47:42 by athena] New experimental scheduler (currently disabled). The old scheduler is ``optimal'' in the sense that it minimizes register pressure. The only way to reduce register pressure is to schedule dependent instructions as closely as possible, so as to minimize the life time of registers. This strategy maximizes the number of pipeline stalls, however. With enough registers and short enough pipelines, this tradeoff is fine. This is no longer the case for the devilish pipeline of the Pentium IV or (probably) the PowerPC 970. The new scheduler switches to a ``list scheduler'' for dags smaller than a specified size. The list scheduler executes a butterfly left to right one column at the time. This amounts to the best possible pipeline utilization, and the worst possible register pressure. The ``specified size'' defaults to 0, i.e., no change from fftw2 and fftw-3.0. It seems like a value of 7--10 produces the best results for Pentium IV (probably screwing the G3/G4 powerpcs and sparc, but I haven't tried.) As time goes by, we may want to increase this number to favor newer processors over older processors. M ./genfft/magic.ml -1 +3 M ./genfft/schedule.ml -9 +35 Wed Jun 25 17:43:59 EDT 2003 stevenj * [project @ 2003-06-25 21:43:59 by stevenj] remove non-portable use of tempfile; thanks to Nicolas Decoster for the patch M ./tools/fftw-wisdom-to-conf.in -1 +1 Wed Jun 25 17:14:03 EDT 2003 stevenj * [project @ 2003-06-25 21:14:03 by stevenj] increase stupid HP preprocessor limits M ./acinclude.m4 -1 +1 Thu Jun 19 15:21:52 EDT 2003 athena * [project @ 2003-06-19 19:21:52 by athena] Distribute gen_mdct.ml M ./genfft/Makefile.am -8 +8 Wed Jun 11 06:55:21 EDT 2003 athena * [project @ 2003-06-11 10:55:21 by athena] Cleared int/ptrdiff_t confusions M ./rdft/buffered2.c -2 +3 M ./rdft/rdft2-radix2.c -5 +5 Tue Jun 10 22:15:42 EDT 2003 athena * [project @ 2003-06-11 02:15:41 by athena] Cleared int/ptrdiff_t confusion M ./dft/dftw-dft.c -1 +1 M ./dft/directwbuf.c -2 +2 M ./dft/rank0.c -2 +2 M ./dft/vrank2-transpose.c -2 +2 M ./dft/vrank3-transpose.c -2 +2 M ./kernel/planner.c -2 +2 Sun Jun 8 09:52:57 EDT 2003 athena * [project @ 2003-06-08 13:52:57 by athena] Increased TIME_MIN. This seems to produce more reliable plans on Pentium IV. M ./kernel/timer.c -2 +2 Sat Jun 7 21:43:00 EDT 2003 athena * [project @ 2003-06-08 01:43:00 by athena] Removed relic -trivial-stores, which dates back to Franz's early experiments. Speed improved on SSE2, both with gcc and icc. M ./dft/simd/codelets/Makefile.am -1 +1 Thu Jun 5 22:29:52 EDT 2003 stevenj * [project @ 2003-06-06 02:29:52 by stevenj] fix direntry M ./doc/fftw3.texi -2 +2 Thu Jun 5 13:41:34 EDT 2003 stevenj * [project @ 2003-06-05 17:41:34 by stevenj] added imdct M ./genfft/gen_mdct.ml -13 +66 Wed Jun 4 19:54:38 EDT 2003 athena * [project @ 2003-06-04 23:54:38 by athena] Collect pattern (a * b) +- (c * d) in generic-arith, because this operation can usually be computed with one rounding in fixed-point (and it possibly exposes a FMA instruction) M ./genfft/c.ml -5 +15 Wed Jun 4 15:11:29 EDT 2003 athena * [project @ 2003-06-04 19:11:29 by athena] Generic-arithmetic unparser M ./genfft/c.ml -8 +31 M ./genfft/magic.ml -1 +5 Sun Jun 1 09:05:30 EDT 2003 athena * [project @ 2003-06-01 13:05:30 by athena] Oops---randomized CSE was using the same random numbers over and over M ./genfft/oracle.ml -8 +6 M ./genfft-k7/oracle.ml -11 +11 Sun Jun 1 07:01:17 EDT 2003 athena * [project @ 2003-06-01 11:01:17 by athena] Paranoia. M ./genfft/c.ml -3 +3 Sun Jun 1 07:00:54 EDT 2003 athena * [project @ 2003-06-01 11:00:54 by athena] Use relative error instead of absolute error, to avoid problems when normalization factors are used. M ./genfft/oracle.ml -3 +5 Sat May 31 22:11:28 EDT 2003 stevenj * [project @ 2003-06-01 02:11:28 by stevenj] slight opt M ./reodft/reodft11e-radix2.c -3 +3 Sat May 31 22:10:45 EDT 2003 stevenj * [project @ 2003-06-01 02:10:45 by stevenj] slight optimization M ./reodft/reodft11e-radix2.c -3 +3 Sat May 31 20:43:31 EDT 2003 stevenj * [project @ 2003-06-01 00:43:31 by stevenj] *W is const M ./genfft/gen_mdct.ml -3 +3 Sat May 31 20:41:15 EDT 2003 stevenj * [project @ 2003-06-01 00:41:15 by stevenj] comment M ./genfft/gen_mdct.ml -2 +4 Thu May 29 21:31:31 EDT 2003 stevenj * [project @ 2003-05-30 01:31:31 by stevenj] added experimental MDCT A ./genfft/gen_mdct.ml M ./genfft/Makefile.am -1 +7 M ./genfft/gen_mdct.ml +208 Wed May 28 22:01:37 EDT 2003 stevenj * [project @ 2003-05-29 02:01:37 by stevenj] altivec (fma) needs simd codlist.c too M ./mkdist.sh -1 +1 Wed May 28 22:00:49 EDT 2003 stevenj * [project @ 2003-05-29 02:00:49 by stevenj] make sure we include SIMD codlist.c for non-Unix folks M ./mkdist.sh -1 +1 Tue May 27 20:31:25 EDT 2003 stevenj * [project @ 2003-05-28 00:31:25 by stevenj] noted howmany_rank == 0 is a single transform M ./doc/fftw3.texi -1 +3 Tue May 27 20:02:31 EDT 2003 stevenj * [project @ 2003-05-28 00:02:31 by stevenj] further stride clarification M ./doc/fftw3.texi -4 +4 Mon May 26 10:21:22 EDT 2003 athena * [project @ 2003-05-26 14:21:22 by athena] Removed transposed dftw problems. I now consider transposed dftw a Bad Idea, since it does not apply to the case that it was originally meant for (speed up four-step) and it complicates the implementation of the other thing I want to try (dftw m-slices). M ./dft/Makefile.am -3 +3 M ./dft/conf.c -2 +1 M ./dft/ct.c -5 +3 M ./dft/ctsq.c -74 +57 M ./dft/dft.h -9 +5 M ./dft/dftw-dft.c -83 M ./dft/directw.c -5 +1 M ./dft/directwbuf.c -5 +1 R ./dft/directwsq.c M ./dft/kdft-difsq.c -2 +2 M ./dft/problemw.c -13 +9 Mon May 26 07:22:59 EDT 2003 athena * [project @ 2003-05-26 11:22:59 by athena] Obsolete comment M ./dft/buffered.c -6 +1 Sat May 24 15:00:53 EDT 2003 athena * [project @ 2003-05-24 19:00:53 by athena] comment M ./dft/ct.c -2 +2 Sat May 24 07:20:35 EDT 2003 athena * [project @ 2003-05-24 11:20:35 by athena] Oops---wrong test NO_UGLYP instead of !NO_UGLYP M ./dft/dftw-dft.c -3 +3 Sat May 24 07:05:34 EDT 2003 athena * [project @ 2003-05-24 11:05:34 by athena] Implemented radix r, where n=r^2 * p M ./dft/ct.c -19 +53 Wed May 21 01:54:32 EDT 2003 stevenj * [project @ 2003-05-21 05:54:32 by stevenj] xlc seems to properly use fma as well M ./doc/FAQ/fftw-faq.bfnn -2 +2 Tue May 20 23:07:43 EDT 2003 stevenj * [project @ 2003-05-21 03:07:43 by stevenj] print warning if there is no cycle counter M ./configure.ac +18 M ./doc/fftw3.texi -2 +3 Tue May 20 17:32:04 EDT 2003 stevenj * [project @ 2003-05-20 21:32:04 by stevenj] updated Funda reference M ./libbench2/verify-lib.c -1 +4 M ./libbench2/verify-r2r.c +3 Mon May 19 20:12:36 EDT 2003 athena * [project @ 2003-05-20 00:12:36 by athena] const M ./dft/ct.c -2 +2 Mon May 19 15:41:09 EDT 2003 athena * [project @ 2003-05-19 19:41:09 by athena] Implemented generic dif square transposed (q-style) solver. M ./dft/dftw-dft.c -8 +88 M ./dft/directwsq.c -2 +2 Mon May 19 07:00:36 EDT 2003 athena * [project @ 2003-05-19 11:00:36 by athena] applicable() is now a property of the solver (in anticipation of transposed solvers) M ./dft/dftw-dft.c -35 +35 Mon May 19 06:33:40 EDT 2003 athena * [project @ 2003-05-19 10:33:40 by athena] Slight cleanup M ./dft/dftw-dft.c -7 +7 Sun May 18 13:05:51 EDT 2003 athena * [project @ 2003-05-18 17:05:51 by athena] Nothing, really M ./dft/bluestein.c -9 +3 M ./kernel/ifftw.h -1 +2 M ./kernel/primes.c -1 +9 Sun May 18 09:05:20 EDT 2003 athena * [project @ 2003-05-18 13:05:20 by athena] Moved vector loop inside bytwiddle(), in anticipation of a q-style dftw-dit transposed solver. M ./dft/dftw-dft.c -38 +41 Sun May 18 08:52:02 EDT 2003 athena * [project @ 2003-05-18 12:52:02 by athena] Fixed flops count M ./dft/dftw-dft.c -1 +1 Sun May 18 08:47:20 EDT 2003 athena * [project @ 2003-05-18 12:47:20 by athena] style M ./dft/dftw-dft.c -5 +3 Sun May 18 07:16:34 EDT 2003 athena * [project @ 2003-05-18 11:16:34 by athena] Faster inner loop. M ./dft/dftw-dft.c -13 +25 Sat May 17 08:02:38 EDT 2003 athena * [project @ 2003-05-17 12:02:38 by athena] Print vector length M ./dft/dftw-dft.c -2 +2 Sat May 17 07:55:33 EDT 2003 athena * [project @ 2003-05-17 11:55:33 by athena] Oops M ./dft/dftw-dft.c -7 +7 Sat May 17 07:50:35 EDT 2003 athena * [project @ 2003-05-17 11:50:35 by athena] Allow vl > 1 M ./dft/dftw-dft.c -7 +16 Sat May 17 07:01:42 EDT 2003 athena * [project @ 2003-05-17 11:01:42 by athena] Radix can be derived from problem---no need to pre-specify it. M ./dft/ctsq.c -21 +12 Fri May 16 22:50:50 EDT 2003 stevenj * [project @ 2003-05-17 02:50:50 by stevenj] fixed comment M ./kernel/transpose.c -1 +1 Fri May 16 22:48:20 EDT 2003 stevenj * [project @ 2003-05-17 02:48:20 by stevenj] whoops, gcd should be static M ./kernel/transpose.c -1 +1 Fri May 16 22:40:32 EDT 2003 stevenj * [project @ 2003-05-17 02:40:32 by stevenj] more unrolling M ./kernel/transpose.c -7 +31 Fri May 16 20:03:48 EDT 2003 athena * [project @ 2003-05-17 00:03:48 by athena] Hack to avoid infinite recursion. M ./dft/bluestein.c +3 Fri May 16 19:52:43 EDT 2003 stevenj * [project @ 2003-05-16 23:52:43 by stevenj] consistency M ./dft/codelet-dft.h -4 +4 Fri May 16 19:45:15 EDT 2003 athena * [project @ 2003-05-16 23:45:15 by athena] Wrong comment. M ./dft/bluestein.c -1 +1 Fri May 16 19:45:03 EDT 2003 athena * [project @ 2003-05-16 23:45:03 by athena] Style. M ./dft/bluestein.c -3 +3 Fri May 16 18:35:27 EDT 2003 stevenj * [project @ 2003-05-16 22:35:27 by stevenj] punctuation M ./doc/FAQ/fftw-faq.bfnn -2 +2 Fri May 16 18:33:45 EDT 2003 stevenj * [project @ 2003-05-16 22:33:45 by stevenj] added allzero FAQ M ./doc/FAQ/fftw-faq.bfnn +7 Fri May 16 18:22:45 EDT 2003 stevenj * [project @ 2003-05-16 22:22:45 by stevenj] simplification: instead of cldb, just use cldf with inputs/output values swapped M ./dft/bluestein.c -22 +11 Fri May 16 15:47:17 EDT 2003 athena * [project @ 2003-05-16 19:47:17 by athena] Allow more general transform sizes. M ./dft/bluestein.c -15 +21 Fri May 16 14:22:37 EDT 2003 stevenj * [project @ 2003-05-16 18:22:37 by stevenj] slight change M ./kernel/ifftw.h -4 +4 Fri May 16 14:22:05 EDT 2003 stevenj * [project @ 2003-05-16 18:22:05 by stevenj] MS has __int64 type, not long long (grr) M ./kernel/ifftw.h -1 +4 Fri May 16 13:34:16 EDT 2003 athena * [project @ 2003-05-16 17:34:16 by athena] Fixed printout M ./dft/ct.c -2 +4 Fri May 16 13:23:00 EDT 2003 athena * [project @ 2003-05-16 17:23:00 by athena] Fixed flop count M ./dft/bluestein.c +5 Fri May 16 13:02:06 EDT 2003 athena * [project @ 2003-05-16 17:02:06 by athena] New bluestein solver A ./dft/bluestein.c M ./dft/Makefile.am -5 +6 M ./dft/bluestein.c +255 M ./dft/conf.c -1 +2 M ./dft/dft.h -1 +2 Fri May 16 09:51:05 EDT 2003 athena * [project @ 2003-05-16 13:51:05 by athena] Implemented generic radix. M ./dft/ct.c -14 +35 Fri May 16 08:19:38 EDT 2003 athena * [project @ 2003-05-16 12:19:38 by athena] Removed conditional branch from inner loop in generic.c M ./dft/generic.c -18 +20 M ./kernel/ifftw.h -2 +2 M ./kernel/twiddle.c -1 +18 Fri May 16 07:48:28 EDT 2003 athena * [project @ 2003-05-16 11:48:28 by athena] Simplified indexing M ./dft/generic.c -29 +20 Fri May 16 06:53:56 EDT 2003 athena * [project @ 2003-05-16 10:53:56 by athena] Better still. M ./dft/generic.c -26 +37 Fri May 16 06:24:31 EDT 2003 athena * [project @ 2003-05-16 10:24:31 by athena] Further improvement of generic solver M ./dft/generic.c -8 +8 Fri May 16 05:57:07 EDT 2003 athena * [project @ 2003-05-16 09:57:07 by athena] Cleanup M ./dft/rader.c -54 +51 Fri May 16 05:42:57 EDT 2003 athena * [project @ 2003-05-16 09:42:57 by athena] Cleanup M ./dft/dftw-dft.c -30 +25 Fri May 16 05:31:40 EDT 2003 athena * [project @ 2003-05-16 09:31:40 by athena] Generic now only works for odd sized. Added check. M ./dft/generic.c -1 +2 Thu May 15 21:53:25 EDT 2003 athena * [project @ 2003-05-16 01:53:25 by athena] Increased GENERIC_MIN_BAD because of new algorithm. M ./kernel/ifftw.h -2 +2 Thu May 15 21:40:27 EDT 2003 athena * [project @ 2003-05-16 01:40:27 by athena] Much, much better. M ./dft/generic.c -9 +30 Thu May 15 21:25:00 EDT 2003 athena * [project @ 2003-05-16 01:25:00 by athena] Still trying to understand why rdft-generic-dit is faster then dft-generic... M ./dft/generic.c -9 +8 Thu May 15 21:04:33 EDT 2003 athena * [project @ 2003-05-16 01:04:33 by athena] Nothing, really M ./dft/generic.c -2 +2 Thu May 15 20:59:45 EDT 2003 athena * [project @ 2003-05-16 00:59:45 by athena] Never be clever for the sake of being clever. M ./dft/generic.c -1 +1 Thu May 15 20:58:06 EDT 2003 athena * [project @ 2003-05-16 00:58:06 by athena] Simplified. generic-dit is gone. The solver is now out-of-place only---buffering is done by the buffered solver. M ./dft/generic.c -94 +45 Thu May 15 19:18:18 EDT 2003 athena * [project @ 2003-05-15 23:18:18 by athena] rader-dit is gone. M ./dft/rader.c -200 Thu May 15 19:13:03 EDT 2003 athena * [project @ 2003-05-15 23:13:03 by athena] Cast M ./dft/plan.c -2 +2 Thu May 15 19:09:07 EDT 2003 athena * [project @ 2003-05-15 23:09:07 by athena] Introduced twiddle problem ``dftw''. Changed most other things to deal with this change. A ./dft/ctsq.c A ./dft/dftw-dft.c A ./dft/directw.c A ./dft/directwbuf.c A ./dft/directwsq.c A ./dft/problemw.c M ./configure.ac -1 +1 M ./dft/Makefile.am -5 +5 M ./dft/buffered.c -2 +1 M ./dft/codelet-dft.h -13 +9 M ./dft/conf.c -1 +4 R ./dft/ct-dif.c R ./dft/ct-dit.c R ./dft/ct-ditbuf.c R ./dft/ct-ditf.c M ./dft/ct.c -91 +157 R ./dft/ct.h M ./dft/ctsq.c +213 M ./dft/dft.h -7 +45 M ./dft/dftw-dft.c +297 M ./dft/directw.c +172 M ./dft/directwbuf.c +230 M ./dft/directwsq.c +148 M ./dft/kdft-dif.c -4 +5 M ./dft/kdft-difsq.c -3 +3 M ./dft/kdft-dit.c -5 +5 M ./dft/plan.c -1 +11 M ./dft/problemw.c +103 M ./dft/rader.c -1 +1 M ./dft/solve.c -1 +8 Sat Aug 12 23:17:14 EDT 2006 Unknown tagger tagged fftw-3-0-1-branch Thu May 15 18:47:18 EDT 2003 stevenj * [project @ 2003-05-15 22:47:18 by stevenj] whoops, X(safe_mulmod) not fftw_safe_mulmod M ./kernel/primes.c -3 +3 Thu May 15 16:53:16 EDT 2003 stevenj * [project @ 2003-05-15 20:53:16 by stevenj] add VC++ versions of asm M ./simd/sse.c -1 +17 M ./simd/sse2.c -1 +17 Thu May 15 15:03:06 EDT 2003 stevenj * [project @ 2003-05-15 19:03:05 by stevenj] VC++ reportedly supports the intel intrinsics, but requires __inline instead of __inline__ M ./simd/simd-sse.h -1 +4 M ./simd/simd-sse2.h -1 +5 Thu May 15 14:32:06 EDT 2003 stevenj * [project @ 2003-05-15 18:32:06 by stevenj] precompute array indices with VC++ M ./kernel/ifftw.h -2 +2 Wed May 14 21:57:39 EDT 2003 stevenj * [project @ 2003-05-15 01:57:39 by stevenj] added doc note M ./acx_pthread.m4 -1 +5 Wed May 14 19:45:54 EDT 2003 stevenj * [project @ 2003-05-14 23:45:54 by stevenj] autodetect windows M ./threads/threads.c -3 +4 Wed May 14 15:08:49 EDT 2003 stevenj * [project @ 2003-05-14 19:08:49 by stevenj] don't bother with #ifdef HAVE_CONFIG_H, since non-Unix users always forget to define it M ./libbench2/getopt.c -2 Tue May 13 16:58:07 EDT 2003 stevenj * [project @ 2003-05-13 20:58:07 by stevenj] VC++ uses __inline M ./kernel/cycle.h -2 +2 Tue May 13 14:51:26 EDT 2003 stevenj * [project @ 2003-05-13 18:51:26 by stevenj] added leak question M ./doc/FAQ/fftw-faq.bfnn +8 Mon May 12 18:26:51 EDT 2003 stevenj * [project @ 2003-05-12 22:26:51 by stevenj] LARGE_INTEGER needs windows.h (supposedly, there is some problem converting _itnt64 to double...damn MS and their nonstandard types) M ./kernel/cycle.h -1 +2 Mon May 12 18:22:16 EDT 2003 stevenj * [project @ 2003-05-12 22:22:16 by stevenj] whoops M ./libbench2/timer.c -2 +2 Mon May 12 17:16:19 EDT 2003 stevenj * [project @ 2003-05-12 21:16:19 by stevenj] added 256x256 to canonical list M ./tools/fftw-wisdom.c -1 +1 Mon May 12 07:02:06 EDT 2003 athena * [project @ 2003-05-12 11:02:06 by athena] Oops... M ./kernel/transpose.c +1 Sun May 11 11:04:46 EDT 2003 athena * [project @ 2003-05-11 15:04:46 by athena] Unrolled loops, changed cutoff M ./kernel/transpose.c -23 +80 Sun May 11 10:20:04 EDT 2003 athena * [project @ 2003-05-11 14:20:04 by athena] Do not multiply strides by 2 twice. M ./tests/bench.c -16 +15 Wed May 7 21:09:43 EDT 2003 stevenj * [project @ 2003-05-08 01:09:43 by stevenj] added 'make smallcheck' M ./tests/Makefile.am +13 Wed May 7 20:46:10 EDT 2003 stevenj * [project @ 2003-05-08 00:46:10 by stevenj] --without-cycle-counter becomes --with-slow-timer, updated docs M ./configure.ac -3 +3 M ./doc/fftw3.texi -15 +15 M ./kernel/timer.c -3 +3 Wed May 7 18:05:29 EDT 2003 stevenj * [project @ 2003-05-07 22:05:29 by stevenj] remove duplicate -openmp check; Sun requires -xopenmp M ./configure.ac -5 +5 Wed May 7 17:59:23 EDT 2003 stevenj * [project @ 2003-05-07 21:59:23 by stevenj] fixed compilation under Sun C++ M ./dft/ct-ditbuf.c -3 +3 M ./rdft/hc2hc-buf.c -3 +3 Wed May 7 14:24:46 EDT 2003 athena * [project @ 2003-05-07 18:24:46 by athena] Use estimator if cycle counter is unavailable, regardless of the FFTW_MEASURE/ESTIMATE setting. M ./kernel/planner.c -3 +13 M ./kernel/timer.c -57 +67 Tue May 6 23:15:34 EDT 2003 stevenj * [project @ 2003-05-07 03:15:34 by stevenj] _WIN32 (not __WIN32__) is always defined M ./kernel/cycle.h -3 +2 Tue May 6 23:11:52 EDT 2003 stevenj * [project @ 2003-05-07 03:11:52 by stevenj] minor cleanup M ./kernel/cycle.h -2 +2 Tue May 6 22:50:07 EDT 2003 stevenj * [project @ 2003-05-07 02:50:07 by stevenj] tentative VC++ stuff, some consolidation M ./kernel/cycle.h -51 +56 Tue May 6 12:17:56 EDT 2003 stevenj * [project @ 2003-05-06 16:17:56 by stevenj] made cycle.h more self-contained M ./kernel/cycle.h -3 +13 M ./kernel/timer.c -15 +18 Tue May 6 08:30:39 EDT 2003 athena * [project @ 2003-05-06 12:30:39 by athena] Use ``%'' flag to denote commutative operations. M ./simd/simd-3dnow.h -3 +3 M ./simd/simd-sse.h -3 +3 M ./simd/simd-sse2.h -3 +3 Mon May 5 20:42:30 EDT 2003 stevenj * [project @ 2003-05-06 00:42:30 by stevenj] MIT license, brief documentation M ./kernel/cycle.h -12 +41 Mon May 5 20:31:16 EDT 2003 stevenj * [project @ 2003-05-06 00:31:16 by stevenj] whoops, forgot f77_wisdom.f M ./doc/Makefile.am -1 +1 Sun May 4 19:37:09 EDT 2003 athena * [project @ 2003-05-04 23:37:09 by athena] Improved speed of accuracy test. M ./dft/problem.c -6 +3 M ./libbench/mp.c -16 +26 M ./libbench2/bench.h -1 +2 M ./libbench2/mp.c -46 +92 M ./libbench2/verify-lib.c -1 +3 M ./rdft/problem2.c -3 +2 Tue Apr 29 11:45:34 EDT 2003 athena * [project @ 2003-04-29 15:45:34 by athena] s390 cycle counter M ./kernel/cycle.h -1 +21 Sat Apr 26 12:26:15 EDT 2003 stevenj * [project @ 2003-04-26 16:26:15 by stevenj] forgot r2r directory M ./doc/fftw3.texi -10 +10 Fri Apr 25 20:52:23 EDT 2003 stevenj * [project @ 2003-04-26 00:52:23 by stevenj] delete unused files, since they don't compile any more M ./rdft/Makefile.am -2 R ./rdft/vrank2-transpose.c R ./rdft/vrank3-transpose.c Thu Apr 24 06:37:41 EDT 2003 athena * [project @ 2003-04-24 10:37:41 by athena] Better gcc code generation M ./simd/simd-sse2.h -3 +5 Wed Apr 23 15:30:50 EDT 2003 stevenj * [project @ 2003-04-23 19:30:50 by stevenj] ccc is the Compaq C compiler on Linux/alpha M ./acinclude.m4 -1 +1 Wed Apr 23 00:06:03 EDT 2003 stevenj * [project @ 2003-04-23 04:06:03 by stevenj] whoops M ./doc/fftw3.texi -3 +3 Sat Aug 12 23:16:25 EDT 2006 Unknown tagger tagged fftw-3-0 Sat Apr 19 09:18:25 EDT 2003 athena * [project @ 2003-04-19 13:18:25 by athena] ia64 cycle counter with intel compiler. M ./kernel/cycle.h -1 +19 Fri Apr 18 18:27:30 EDT 2003 athena * [project @ 2003-04-18 22:27:30 by athena] More gcc bugs. Sigh. M ./doc/FAQ/fftw-faq.bfnn -4 +12 Fri Apr 18 18:01:49 EDT 2003 athena * [project @ 2003-04-18 22:01:49 by athena] touch ChangeLog to observe GNU standards M ./bootstrap.sh +1 Fri Apr 18 18:01:12 EDT 2003 athena * [project @ 2003-04-18 22:01:12 by athena] We now build ChangeLog automatically at distribution time R ./ChangeLog Fri Apr 18 18:00:17 EDT 2003 athena * [project @ 2003-04-18 22:00:17 by athena] Automatic ChangeLog hackery M ./mkdist.sh +7 Fri Apr 18 13:25:26 EDT 2003 stevenj * [project @ 2003-04-18 17:25:26 by stevenj] plural M ./doc/FAQ/fftw-faq.bfnn -1 +1 Fri Apr 18 13:25:01 EDT 2003 stevenj * [project @ 2003-04-18 17:25:01 by stevenj] updated M ./NEWS +12 Fri Apr 18 12:59:41 EDT 2003 athena * [project @ 2003-04-18 16:59:41 by athena] Updated M ./ChangeLog -6789 +3415 Fri Apr 18 11:48:39 EDT 2003 stevenj * [project @ 2003-04-18 15:48:39 by stevenj] a -> an M ./doc/FAQ/fftw-faq.bfnn -1 +1 Fri Apr 18 11:47:56 EDT 2003 stevenj * [project @ 2003-04-18 15:47:56 by stevenj] hyphen M ./doc/FAQ/fftw-faq.bfnn -1 +1 Fri Apr 18 11:47:28 EDT 2003 stevenj * [project @ 2003-04-18 15:47:28 by stevenj] comma M ./doc/FAQ/fftw-faq.bfnn -1 +1 Fri Apr 18 11:46:59 EDT 2003 stevenj * [project @ 2003-04-18 15:46:59 by stevenj] minor M ./doc/FAQ/fftw-faq.bfnn -4 +4 Fri Apr 18 10:37:31 EDT 2003 athena * [project @ 2003-04-18 14:37:31 by athena] Updated M ./doc/FAQ/fftw-faq.bfnn -9 +36 Fri Apr 18 10:14:59 EDT 2003 athena * [project @ 2003-04-18 14:14:59 by athena] New script that builds the distributions A ./mkdist.sh Fri Apr 18 08:51:07 EDT 2003 athena * [project @ 2003-04-18 12:51:07 by athena] Oops again M ./dft/simd/codelets/Makefile.am -4 +4 Fri Apr 18 08:39:05 EDT 2003 athena * [project @ 2003-04-18 12:39:05 by athena] Oops, forgot -sign 1 M ./dft/simd/codelets/Makefile.am -4 +4 Fri Apr 18 08:28:25 EDT 2003 athena * [project @ 2003-04-18 12:28:25 by athena] Reorganization of simd codelets M ./configure.ac -4 M ./dft/simd/codelets/Makefile.am -136 +22 M ./dft/simd/n1b.c -3 +3 M ./dft/simd/n1b.h -3 M ./dft/simd/n1f.c -3 +3 M ./dft/simd/n1f.h -3 M ./dft/simd/n2b.c -3 +3 M ./dft/simd/n2b.h +3 M ./dft/simd/n2f.c -3 +3 M ./dft/simd/n2f.h +3 Thu Apr 17 21:21:45 EDT 2003 athena * [project @ 2003-04-18 01:21:45 by athena] k7 assembly was not updated after conversion of opcnt from int to double M ./genfft-k7/gen_notw.ml -5 +5 M ./genfft-k7/gen_twiddle.ml -5 +6 Thu Apr 17 19:15:53 EDT 2003 athena * [project @ 2003-04-17 23:15:53 by athena] Capital `X' looks bad in all-lowercase plans M ./dft/vrank2-transpose.c -2 +2 M ./dft/vrank3-transpose.c -2 +2 Thu Apr 17 18:53:29 EDT 2003 athena * [project @ 2003-04-17 22:53:29 by athena] Removed redundant inline/noinline codelets M ./dft/codelets/standard/Makefile.am -3 +2 M ./dft/simd/codelets/Makefile.am -27 +12 M ./rdft/codelets/hc2r/Makefile.am -8 +4 M ./rdft/codelets/r2hc/Makefile.am -8 +4 Thu Apr 17 15:25:50 EDT 2003 athena * [project @ 2003-04-17 19:25:50 by athena] New noinline Noinline real codelets A ./genfft/gen_hc2r_noinline.ml A ./genfft/gen_r2hc_noinline.ml M ./genfft/Makefile.am -10 +24 M ./genfft/gen_hc2hc.ml -5 +5 M ./genfft/gen_hc2r.ml -48 +31 M ./genfft/gen_hc2r_noinline.ml +185 M ./genfft/gen_r2hc.ml -49 +31 M ./genfft/gen_r2hc_noinline.ml +181 M ./genfft/gen_r2r.ml -3 +3 M ./kernel/ifftw.h -4 +1 M ./rdft/codelets/hc2r/Makefile.am -1 +15 M ./rdft/codelets/r2hc/Makefile.am -1 +15 M ./support/Makefile.codelets -1 +3 Thu Apr 17 15:23:03 EDT 2003 stevenj * [project @ 2003-04-17 19:23:03 by stevenj] more ideas M ./TODO +6 Thu Apr 17 13:18:45 EDT 2003 athena * [project @ 2003-04-17 17:18:45 by athena] Removed duplicate rules. M ./dft/simd/codelets/Makefile.am -6 Thu Apr 17 10:51:09 EDT 2003 athena * [project @ 2003-04-17 14:51:09 by athena] acx_pthread.m4 was not distributed M ./Makefile.am -1 +1 Thu Apr 17 07:21:17 EDT 2003 athena * [project @ 2003-04-17 11:21:17 by athena] Oops M ./support/Makefile.codelets -1 +1 Thu Apr 17 07:07:19 EDT 2003 athena * [project @ 2003-04-17 11:07:19 by athena] Both inlined and non-inlined notw codelets. A ./genfft/gen_notw_noinline.ml M ./dft/codelets/standard/Makefile.am -1 +10 M ./genfft/Makefile.am -11 +17 M ./genfft/gen_notw.ml -51 +37 M ./genfft/gen_notw_noinline.ml +164 M ./support/Makefile.codelets +1 Thu Apr 17 06:44:21 EDT 2003 athena * [project @ 2003-04-17 10:44:21 by athena] Initial experiment with both inlined and non-inlined simd codelets. Both are included for now. A ./genfft/gen_notw_noinline_c.ml M ./dft/simd/codelets/Makefile.am -3 +68 M ./genfft/Makefile.am -9 +17 M ./genfft/gen_notw_noinline_c.ml +160 M ./support/Makefile.codelets -1 +2 Thu Apr 17 05:57:36 EDT 2003 athena * [project @ 2003-04-17 09:57:36 by athena] --enable-fma to build FMA distribution M ./configure.ac +7 M ./support/Makefile.codelets -1 +7 Wed Apr 16 17:21:53 EDT 2003 athena * [project @ 2003-04-16 21:21:53 by athena] Inline SIMD nontwiddle codelets M ./genfft/gen_notw_c.ml -47 +41 Wed Apr 16 16:18:29 EDT 2003 athena * [project @ 2003-04-16 20:18:29 by athena] Pathetic attempt at saving a couple of registers... M ./simd/simd-sse.h -6 +9 M ./simd/simd-sse2.h -6 +9 Wed Apr 16 15:51:27 EDT 2003 athena * [project @ 2003-04-16 19:51:27 by athena] for (i = 0; i < m; ++i) ==> for (i = m; i > 0; --i) No proof of evidence that this is any faster, but just in case... M ./genfft/gen_hc2r.ml -3 +3 M ./genfft/gen_notw.ml -3 +3 M ./genfft/gen_notw_c.ml -3 +3 M ./genfft/gen_r2hc.ml -3 +3 M ./genfft/gen_r2r.ml -3 +3 M ./genfft/gen_twiddle.ml -5 +5 M ./genfft/gen_twiddle_c.ml -5 +5 Tue Apr 15 15:03:20 EDT 2003 stevenj * [project @ 2003-04-15 19:03:20 by stevenj] added hack to make sure that codelet loops are preferred to vecloop solvers in the estimator M ./dft/vrank-geq1.c -1 +2 M ./rdft/vrank-geq1-rdft2.c -1 +2 M ./rdft/vrank-geq1.c -1 +2 Tue Apr 15 14:53:44 EDT 2003 stevenj * [project @ 2003-04-15 18:53:44 by stevenj] use double for flops M ./api/f77funcs.h -1 +1 M ./api/fftw3.h -2 +2 M ./api/flops.c -1 +1 M ./doc/fftw3.texi -3 +6 M ./kernel/ifftw.h -5 +5 M ./tests/bench.c -2 +2 Tue Apr 15 14:51:50 EDT 2003 stevenj * [project @ 2003-04-15 18:51:50 by stevenj] metrowerks reportedly supports gcc assembly extensions on ppc M ./kernel/cycle.h -2 +2 Mon Apr 14 15:00:50 EDT 2003 athena * [project @ 2003-04-14 19:00:48 by athena] foo_CFLAGS generates some automake junk that breaks the build on Redhat 7.3. Screw it. M ./dft/simd/Makefile.am -1 +1 M ./simd/Makefile.am -1 +1 Mon Apr 14 12:22:59 EDT 2003 athena * [project @ 2003-04-14 16:22:59 by athena] Carefully check return status M ./tests/check.pl -1 +21 Sun Apr 13 16:46:12 EDT 2003 athena * [project @ 2003-04-13 20:46:12 by athena] Removed annoying -FMA() expressions. M ./genfft/c.ml -2 +6 M ./genfft/simd.ml -2 +2 M ./kernel/ifftw.h -1 +9 M ./support/Makefile.codelets -1 +1 Sat Apr 12 14:32:22 EDT 2003 athena * [project @ 2003-04-12 18:32:22 by athena] Major fma hackery M ./kernel/ifftw.h -5 +17 Sat Apr 12 14:25:43 EDT 2003 athena * [project @ 2003-04-12 18:25:43 by athena] Slight cleanup M ./api/apiplan.c -8 +5 Sat Apr 12 10:04:51 EDT 2003 athena * [project @ 2003-04-12 14:04:51 by athena] Updated version number M ./configure.ac -1 +1 Sat Apr 12 08:03:07 EDT 2003 athena * [project @ 2003-04-12 12:03:07 by athena] Damn autoconf M ./acinclude.m4 -2 +2 Sat Apr 12 07:54:20 EDT 2003 athena * [project @ 2003-04-12 11:54:20 by athena] Recognize all 74xx processors M ./acinclude.m4 -4 +5 Sat Apr 12 07:35:17 EDT 2003 athena * [project @ 2003-04-12 11:35:17 by athena] Detect 7400 processor. M ./acinclude.m4 -3 +7 Fri Apr 11 20:42:11 EDT 2003 athena * [project @ 2003-04-12 00:42:11 by athena] No need to check for gcc-2.95 M ./acinclude.m4 -4 +2 Fri Apr 11 16:14:39 EDT 2003 stevenj * [project @ 2003-04-11 20:14:39 by stevenj] removed duplicate M ./NEWS -3 Fri Apr 11 08:45:37 EDT 2003 athena * [project @ 2003-04-11 12:45:37 by athena] mflops ==> ``mflops'' M ./libbench2/report.c -2 +2 Fri Apr 11 07:00:53 EDT 2003 athena * [project @ 2003-04-11 11:00:53 by athena] Print setup time as well M ./libbench2/report.c -3 +5 Sat Aug 12 23:14:42 EDT 2006 Unknown tagger tagged fftw-3-0-beta3 Thu Apr 10 15:36:18 EDT 2003 athena * [project @ 2003-04-10 19:36:15 by athena] Enforce pointer equality for in-place problems. M ./dft/problem.c -4 +8 M ./kernel/ifftw.h -1 +4 M ./rdft/problem.c -4 +3 M ./rdft/problem2.c -4 +7 M ./simd/taint.c -1 +8 Wed Apr 9 17:47:54 EDT 2003 stevenj * [project @ 2003-04-09 21:47:26 by stevenj] updated M ./ChangeLog +406 M ./NEWS +28 Wed Apr 9 14:53:38 EDT 2003 stevenj * [project @ 2003-04-09 18:53:38 by stevenj] cross-ref fftw-wisdom man page M ./tests/README -1 +1 Wed Apr 9 10:13:00 EDT 2003 athena * [project @ 2003-04-09 14:13:00 by athena] Undone previous change, committed by mistake. M ./kernel/planner.c -2 +1 Wed Apr 9 10:12:24 EDT 2003 athena * [project @ 2003-04-09 14:12:18 by athena] Quick and dirty README for bench A ./tests/README M ./kernel/planner.c -1 +2 M ./tests/Makefile.am -1 +1 M ./tests/README +57 Wed Apr 9 08:50:25 EDT 2003 athena * [project @ 2003-04-09 12:50:25 by athena] Consider additional command-line arguments as problems to be benchmarked. M ./libbench2/bench-main.c -7 +6 M ./libbench2/timer.c -1 +7 Wed Apr 9 08:44:13 EDT 2003 athena * [project @ 2003-04-09 12:44:13 by athena] Default report format is now human-readable. Removed unnecessary complexity in benchmark reporting. M ./libbench2/bench-main.c -18 +7 M ./libbench2/bench.h -5 +2 M ./libbench2/report.c -71 +39 Wed Apr 9 06:10:40 EDT 2003 athena * [project @ 2003-04-09 10:10:40 by athena] Updated for new interleaved/split api. M ./doc/fftw3.texi -100 +180 Wed Apr 9 03:01:03 EDT 2003 stevenj * [project @ 2003-04-09 07:01:03 by stevenj] updated citation M ./doc/fftw3.texi -3 +3 Tue Apr 8 19:35:59 EDT 2003 athena * [project @ 2003-04-08 23:35:59 by athena] Time for beta3 M ./configure.ac -1 +1 Tue Apr 8 17:40:59 EDT 2003 stevenj * [project @ 2003-04-08 21:40:59 by stevenj] whoops, added A ./reodft/redft00e-r2hc-pad.c Tue Apr 8 17:33:47 EDT 2003 stevenj * [project @ 2003-04-08 21:33:47 by stevenj] more comparison of different R*DFT types M ./doc/fftw3.texi -1 +32 Tue Apr 8 16:48:08 EDT 2003 stevenj * [project @ 2003-04-08 20:48:08 by stevenj] comments M ./reodft/redft00e-r2hc.c -5 +18 M ./reodft/rodft00e-r2hc.c -5 +18 Tue Apr 8 16:19:39 EDT 2003 stevenj * [project @ 2003-04-08 20:19:39 by stevenj] more accurate DCT-I and DST-I, at the expense of up to a factor of 2 in speed and memory A ./reodft/rodft00e-r2hc-pad.c M ./reodft/Makefile.am -4 +5 M ./reodft/conf.c -3 +5 M ./reodft/reodft.h +2 M ./reodft/rodft00e-r2hc-pad.c +200 Tue Apr 8 05:38:09 EDT 2003 athena * [project @ 2003-04-08 09:38:09 by athena] Workaround gcc/sparc bug M ./kernel/planner.c -2 +3 Tue Apr 8 01:34:12 EDT 2003 stevenj * [project @ 2003-04-08 05:34:12 by stevenj] rumors M ./doc/fftw3.texi -2 +2 Mon Apr 7 18:54:11 EDT 2003 stevenj * [project @ 2003-04-07 22:54:11 by stevenj] added rdft2 paranoid mode M ./tests/hook.c +33 Mon Apr 7 18:47:37 EDT 2003 stevenj * [project @ 2003-04-07 22:47:37 by stevenj] added paranoid mode for r2r M ./tests/hook.c +51 Mon Apr 7 15:10:08 EDT 2003 stevenj * [project @ 2003-04-07 19:10:08 by stevenj] whoops, sincos is predefined on some systems M ./libbench2/verify-r2r.c -3 +3 Sat Apr 5 16:50:57 EST 2003 athena * [project @ 2003-04-05 21:50:57 by athena] bp->destroy_input was not initialized M ./tests/hook.c +1 Sat Apr 5 09:29:11 EST 2003 athena * [project @ 2003-04-05 14:29:11 by athena] Asserted correctness conditions for tainted pointers. (For now, use CK() while we test. They should be changed into A() at some point.) M ./dft/problem.c -7 +16 M ./kernel/ifftw.h -1 +3 M ./rdft/problem.c -3 +7 M ./rdft/problem2.c -6 +12 Sat Apr 5 08:18:23 EST 2003 athena * [project @ 2003-04-05 13:18:23 by athena] Untaint pointers before zero'ing arrays and before hashing M ./dft/problem.c -8 +10 M ./rdft/problem.c -4 +4 M ./rdft/problem2.c -9 +9 Sat Apr 5 07:11:56 EST 2003 athena * [project @ 2003-04-05 12:11:56 by athena] Alignment check did not work with icc, which seems to be confused by the fact that the variable is not used. M ./libbench2/bench-main.c -5 +4 Sat Apr 5 06:41:20 EST 2003 athena * [project @ 2003-04-05 11:41:20 by athena] More paranoid paranoid-check M ./tests/Makefile.am -1 +7 Sat Apr 5 06:19:25 EST 2003 athena * [project @ 2003-04-05 11:19:25 by athena] 0 == x & 7 parses as (0 == x) & 7, which is wrong M ./kernel/ifftw.h -2 +2 Fri Apr 4 21:35:49 EST 2003 stevenj * [project @ 2003-04-05 02:35:49 by stevenj] alignment checks M ./dft/direct.c -1 +2 M ./kernel/ifftw.h -1 +12 M ./kernel/planner.c -1 +2 M ./libbench2/bench-main.c -1 +3 M ./rdft/direct.c -1 +4 M ./rdft/direct2.c -1 +3 Fri Apr 4 21:04:14 EST 2003 stevenj * [project @ 2003-04-05 02:04:14 by stevenj] prevent infinite loops in exhaustive planning M ./rdft/rdft-dht.c -1 +6 Fri Apr 4 20:58:20 EST 2003 stevenj * [project @ 2003-04-05 01:58:20 by stevenj] split/unsplit guru interface A ./api/execute-split-dft-c2r.c A ./api/execute-split-dft-r2c.c A ./api/execute-split-dft.c A ./api/plan-guru-split-dft-c2r.c A ./api/plan-guru-split-dft-r2c.c A ./api/plan-guru-split-dft.c M ./api/Makefile.am -1 +3 M ./api/api.h -2 +3 M ./api/apiplan.c -3 +4 M ./api/execute-dft-c2r.c -2 +2 M ./api/execute-dft-r2c.c -2 +2 M ./api/execute-dft.c -3 +6 M ./api/execute-split-dft-c2r.c +29 M ./api/execute-split-dft-r2c.c +29 M ./api/execute-split-dft.c +29 M ./api/f77funcs.h -8 +73 M ./api/fftw3.h -5 +25 M ./api/mktensor-iodims.c -3 +3 M ./api/plan-guru-dft-c2r.c -4 +9 M ./api/plan-guru-dft-r2c.c -4 +9 M ./api/plan-guru-dft.c -4 +10 M ./api/plan-guru-r2r.c -3 +4 M ./api/plan-guru-split-dft-c2r.c +40 M ./api/plan-guru-split-dft-r2c.c +39 M ./api/plan-guru-split-dft.c +39 M ./api/plan-many-dft-c2r.c -1 +1 M ./api/plan-many-dft-r2c.c -1 +1 M ./api/plan-many-dft.c -1 +1 M ./api/plan-many-r2r.c -1 +1 M ./tests/bench.c -29 +15 Fri Apr 4 20:39:55 EST 2003 athena * [project @ 2003-04-05 01:39:55 by athena] Need UNTAINT in verifier too. M ./tests/hook.c -4 +4 Fri Apr 4 19:36:46 EST 2003 athena * [project @ 2003-04-05 00:36:46 by athena] Forgot #if HAVE_SIMD M ./simd/taint.c -1 +5 Fri Apr 4 19:30:37 EST 2003 athena * [project @ 2003-04-05 00:30:37 by athena] Keep track of two separate taint bits A ./simd/taint.c M ./api/fftw3.h -2 +2 M ./kernel/align.c -15 +1 M ./kernel/ifftw.h -2 +2 M ./simd/Makefile.am -1 +1 M ./simd/simd.h -2 +16 M ./simd/taint.c +33 Fri Apr 4 19:16:32 EST 2003 stevenj * [project @ 2003-04-05 00:16:32 by stevenj] added NO_SIMD problem flag, made UNALIGNED an API issue (taints input pointers) M ./api/api.h +2 M ./api/fftw3.h -1 +3 M ./api/mapflags.c -1 +1 M ./api/plan-guru-dft-c2r.c -1 +3 M ./api/plan-guru-dft-r2c.c -1 +3 M ./api/plan-guru-dft.c -1 +4 M ./api/plan-guru-r2r.c -1 +2 M ./api/plan-many-dft-c2r.c -1 +3 M ./api/plan-many-dft-r2c.c -1 +3 M ./api/plan-many-dft.c -1 +4 M ./api/plan-many-r2r.c -1 +2 M ./dft/k7/k7.c -5 +13 M ./dft/simd/n1b.c -1 +1 M ./dft/simd/n1f.c -1 +1 M ./dft/simd/n2b.c +1 M ./dft/simd/n2f.c +1 M ./dft/simd/q1b.c -1 +1 M ./dft/simd/q1f.c -1 +1 M ./dft/simd/t1b.c -1 +1 M ./dft/simd/t1f.c -1 +1 M ./kernel/ifftw.h -3 +3 M ./tests/bench.c +1 Fri Apr 4 18:14:14 EST 2003 stevenj * [project @ 2003-04-04 23:14:14 by stevenj] bugfix in buffered: wrong pointers passed for cldrest; also use TAINT instead of UNALIGNED in buffered2 M ./dft/buffered.c -6 +10 M ./rdft/buffered.c -6 +10 M ./rdft/buffered2.c -7 +5 Fri Apr 4 17:19:51 EST 2003 athena * [project @ 2003-04-04 22:19:51 by athena] Reverted previous change, committed accidentally M ./dft/vrank-geq1.c -3 +3 Fri Apr 4 17:18:39 EST 2003 athena * [project @ 2003-04-04 22:18:39 by athena] What was I thinking? M ./kernel/align.c -1 +2 Fri Apr 4 17:18:21 EST 2003 athena * [project @ 2003-04-04 22:18:21 by athena] M ./dft/vrank-geq1.c -3 +3 Fri Apr 4 16:48:32 EST 2003 stevenj * [project @ 2003-04-04 21:48:32 by stevenj] added --enable-debug-alignment M ./configure.ac +5 M ./libbench2/aligned-main.c -1 +7 Fri Apr 4 16:29:43 EST 2003 stevenj * [project @ 2003-04-04 21:29:43 by stevenj] X(taint) prototype, define corresponding function only if HAVE_SIMD M ./kernel/align.c -1 +3 M ./kernel/ifftw.h -1 +2 Fri Apr 4 16:15:53 EST 2003 athena * [project @ 2003-04-04 21:15:53 by athena] Initial checkin of tained pointers M ./dft/buffered.c -4 +9 M ./dft/solve.c -2 +4 M ./dft/vrank-geq1.c -3 +3 M ./kernel/align.c -6 +4 M ./kernel/ifftw.h -2 +9 M ./rdft/buffered.c -4 +4 M ./rdft/buffered2.c -3 +2 M ./rdft/solve.c -2 +2 M ./rdft/solve2.c -2 +2 M ./rdft/vrank-geq1-rdft2.c -3 +4 M ./rdft/vrank-geq1.c -3 +3 Fri Apr 4 13:12:58 EST 2003 athena * [project @ 2003-04-04 18:12:52 by athena] More conservative preservation of alignment M ./dft/buffered.c -1 +2 M ./dft/rader.c -4 +3 M ./dft/simd/n2b.c -1 M ./dft/simd/n2f.c -1 M ./dft/vrank-geq1.c -6 +4 M ./kernel/align.c -35 +8 M ./kernel/ifftw.h -3 +2 M ./rdft/buffered.c -1 +2 M ./rdft/buffered2.c -1 +4 M ./rdft/dht-rader.c -2 +2 M ./rdft/vrank-geq1-rdft2.c -5 +3 M ./rdft/vrank-geq1.c -5 +5 Thu Apr 3 23:16:27 EST 2003 stevenj * [project @ 2003-04-04 04:16:27 by stevenj] plan/execute with aligned stack M ./api/apiplan.c -19 +24 M ./api/execute-dft-c2r.c -2 +2 M ./api/execute-dft-r2c.c -2 +2 M ./api/execute-dft.c -2 +2 M ./api/execute-r2r.c -2 +2 M ./api/execute.c -2 +2 M ./api/f77funcs.h -10 +10 Thu Apr 3 15:40:01 EST 2003 stevenj * [project @ 2003-04-03 20:40:01 by stevenj] whoops, missed FFTW_MEASURE in fftw3.f M ./api/Makefile.am -1 +1 Thu Apr 3 13:44:46 EST 2003 stevenj * [project @ 2003-04-03 18:44:46 by stevenj] use WITH_ALIGNED_STACK for experimental semaphore stuff, too M ./threads/threads.c -8 +3 Thu Apr 3 09:04:23 EST 2003 athena * [project @ 2003-04-03 14:04:23 by athena] Removed old file R ./kernel/stack.c Thu Apr 3 07:50:43 EST 2003 athena * [project @ 2003-04-03 12:50:43 by athena] Improved stack-alignment hack M ./kernel/Makefile.am -3 +3 M ./kernel/ifftw.h -4 +52 M ./threads/ct-dit.c -7 +6 M ./threads/dft-vrank-geq1.c -7 +6 M ./threads/hc2hc-dif.c -7 +6 M ./threads/hc2hc-dit.c -7 +6 M ./threads/rdft-vrank-geq1.c -6 +4 M ./threads/threads.h -5 M ./threads/vrank-geq1-rdft2.c -7 +6 Thu Apr 3 02:37:57 EST 2003 stevenj * [project @ 2003-04-03 07:37:57 by stevenj] use aligned stack for experimental semaphores, too M ./threads/threads.c -2 +8 Thu Apr 3 02:17:58 EST 2003 stevenj * [project @ 2003-04-03 07:17:58 by stevenj] whoops M ./kernel/ifftw.h -2 +4 M ./kernel/stack.c -2 +2 M ./threads/ct-dit.c -2 +4 M ./threads/dft-vrank-geq1.c -2 +4 M ./threads/hc2hc-dif.c -2 +4 M ./threads/hc2hc-dit.c -2 +4 M ./threads/rdft-vrank-geq1.c -2 +4 M ./threads/threads.c -24 +3 M ./threads/threads.h +5 M ./threads/vrank-geq1-rdft2.c -2 +4 Thu Apr 3 01:58:32 EST 2003 stevenj * [project @ 2003-04-03 06:58:32 by stevenj] fix(?) for SIMD thread problems M ./kernel/ifftw.h -2 +2 M ./kernel/stack.c -3 +3 M ./threads/ct-dit.c -7 +10 M ./threads/hc2hc-dif.c -2 +2 M ./threads/hc2hc-dit.c -2 +2 M ./threads/threads.c -6 +27 Wed Apr 2 20:33:12 EST 2003 stevenj * [project @ 2003-04-03 01:33:12 by stevenj] noted n=1 REDFT01 case M ./doc/fftw3.texi -1 +11 Wed Apr 2 20:32:07 EST 2003 stevenj * [project @ 2003-04-03 01:32:07 by stevenj] note about n=2 REDFT00 formula M ./doc/fftw3.texi -2 +3 Wed Apr 2 20:30:10 EST 2003 stevenj * [project @ 2003-04-03 01:30:10 by stevenj] note about undefined REDFT00 M ./doc/fftw3.texi -1 +2 Wed Apr 2 20:18:03 EST 2003 stevenj * [project @ 2003-04-03 01:18:03 by stevenj] noted n=1 RODFT01 case M ./doc/fftw3.texi -2 +12 Wed Apr 2 20:14:07 EST 2003 stevenj * [project @ 2003-04-03 01:14:07 by stevenj] corrected definitions M ./doc/equation-redft11.png M ./doc/equation-rodft01.png M ./doc/equation-rodft11.png M ./doc/fftw3.texi -6 +6 Wed Apr 2 19:43:59 EST 2003 stevenj * [project @ 2003-04-03 00:43:59 by stevenj] added REODFT_KINDP, fixed nontrivial test for R2HC11 and HC2R11 (not that we support these yet anyway) M ./rdft/codelet-rdft.h +1 M ./rdft/problem.c -3 +3 M ./rdft/vrank-geq1.c -3 +3 Wed Apr 2 19:16:54 EST 2003 stevenj * [project @ 2003-04-03 00:16:53 by stevenj] size 2 hc2r and dht are equivalent to r2hc M ./rdft/codelets/hc2r/Makefile.am -3 +5 M ./rdft/problem.c -3 +5 Wed Apr 2 15:09:08 EST 2003 stevenj * [project @ 2003-04-02 20:09:08 by stevenj] noted overwriting in upgrading section M ./doc/fftw3.texi -1 +4 Wed Apr 2 05:25:56 EST 2003 athena * [project @ 2003-04-02 10:25:56 by athena] Moved with_aligned_stack to its own file A ./kernel/stack.c M ./kernel/Makefile.am -3 +3 M ./kernel/align.c -46 +1 M ./kernel/stack.c +67 Tue Apr 1 21:11:31 EST 2003 athena * [project @ 2003-04-02 02:11:31 by athena] Fixed comments M ./kernel/align.c -6 +2 M ./libbench2/aligned-main.c -6 +2 Tue Apr 1 20:57:39 EST 2003 athena * [project @ 2003-04-02 01:57:39 by athena] Alignment hacks M ./kernel/align.c -1 +50 M ./kernel/ifftw.h -2 +3 M ./libbench2/aligned-main.c -26 +14 M ./libbench2/bench-main.c -1 +9 Tue Apr 1 14:26:48 EST 2003 stevenj * [project @ 2003-04-01 19:26:48 by stevenj] phew, no, previous version was okay M ./threads/threads.c -1 +1 Tue Apr 1 14:26:15 EST 2003 stevenj * [project @ 2003-04-01 19:26:15 by stevenj] whoops, crap M ./threads/threads.c -2 +2 Tue Apr 1 08:01:06 EST 2003 athena * [project @ 2003-04-01 13:01:06 by athena] support sse2 in forthcoming gcc-3.3 M ./simd/simd-sse2.h -3 +11 Tue Apr 1 01:17:15 EST 2003 stevenj * [project @ 2003-04-01 06:17:15 by stevenj] comment M ./kernel/cycle.h -2 +2 Tue Apr 1 01:16:46 EST 2003 stevenj * [project @ 2003-04-01 06:16:46 by stevenj] noted ac_check_headers M ./kernel/cycle.h -1 +4 Tue Apr 1 01:11:31 EST 2003 stevenj * [project @ 2003-04-01 06:11:31 by stevenj] comment M ./kernel/cycle.h -1 +2 Tue Apr 1 01:06:53 EST 2003 stevenj * [project @ 2003-04-01 06:06:53 by stevenj] documented autoconf tests, so that cycle.h can be distributed separately M ./kernel/cycle.h -1 +27 Sat Aug 12 23:11:17 EDT 2006 Unknown tagger tagged fftw-3-0-beta2 Mon Mar 31 22:12:02 EST 2003 stevenj * [project @ 2003-04-01 03:12:02 by stevenj] IRIX is all-caps M ./NEWS -1 +1 Mon Mar 31 22:11:42 EST 2003 stevenj * [project @ 2003-04-01 03:11:42 by stevenj] noted Irix fix M ./NEWS +3 Mon Mar 31 22:10:33 EST 2003 stevenj * [project @ 2003-04-01 03:10:33 by stevenj] whoops M ./threads/api.c -1 +1 M ./threads/threads.h -1 +1 Mon Mar 31 22:04:35 EST 2003 stevenj * [project @ 2003-04-01 03:04:35 by stevenj] use ithreads_init so as not to confuse fftw 2 users M ./threads/threads.c -6 +8 Mon Mar 31 22:00:42 EST 2003 stevenj * [project @ 2003-04-01 03:00:42 by stevenj] IRIX lossage M ./threads/threads.c -1 +5 Mon Mar 31 21:19:20 EST 2003 stevenj * [project @ 2003-04-01 02:19:20 by stevenj] check for -openmp (icc) among the OpenMP flags (TODO: make this a separate macro, with a loop instead of repeated checks) M ./configure.ac +5 Mon Mar 31 17:12:19 EST 2003 stevenj * [project @ 2003-03-31 22:12:19 by stevenj] clarification M ./doc/fftw3.texi -1 +4 Mon Mar 31 17:01:16 EST 2003 athena * [project @ 2003-03-31 22:01:16 by athena] More liberal test for solaris CC M ./acinclude.m4 -1 +1 Mon Mar 31 15:13:33 EST 2003 athena * [project @ 2003-03-31 20:13:33 by athena] Allow x86-64 simd M ./simd/simd-sse.h -1 +1 M ./simd/simd-sse2.h -1 +1 Mon Mar 31 15:13:21 EST 2003 athena * [project @ 2003-03-31 20:13:21 by athena] Added x86-64 timer code M ./kernel/cycle.h -1 +22 Mon Mar 31 13:10:54 EST 2003 stevenj * [project @ 2003-03-31 18:10:54 by stevenj] updated M ./NEWS +4 Mon Mar 31 13:07:19 EST 2003 stevenj * [project @ 2003-03-31 18:07:19 by stevenj] updated M ./ChangeLog -1378 +3497 Mon Mar 31 13:05:27 EST 2003 stevenj * [project @ 2003-03-31 18:05:27 by stevenj] colon M ./doc/FAQ/fftw-faq.bfnn -1 +1 Mon Mar 31 07:20:20 EST 2003 athena * [project @ 2003-03-31 12:20:20 by athena] Reorganized compiler bugs section (which is growing out of control) M ./doc/FAQ/fftw-faq.bfnn -11 +18 Mon Mar 31 07:15:20 EST 2003 athena * [project @ 2003-03-31 12:15:20 by athena] solaris gcc bug appears to be also in 2.95.2 M ./doc/FAQ/fftw-faq.bfnn -1 +1 Mon Mar 31 07:13:45 EST 2003 athena * [project @ 2003-03-31 12:13:45 by athena] Workaround works---there is another gcc/sparc bug elsehwere M ./kernel/planner.c -3 +3 Mon Mar 31 07:08:56 EST 2003 athena * [project @ 2003-03-31 12:08:56 by athena] Grrr, workaround does not work. M ./kernel/planner.c -3 +3 Mon Mar 31 07:02:23 EST 2003 athena * [project @ 2003-03-31 12:02:23 by athena] ADDMOD is now function, which seems to avoid gcc bugs. M ./kernel/planner.c -10 +16 Sun Mar 30 16:40:26 EST 2003 athena * [project @ 2003-03-30 21:40:26 by athena] Workaround sparc gcc bug M ./kernel/planner.c -3 +5 Sun Mar 30 15:51:59 EST 2003 stevenj * [project @ 2003-03-30 20:50:59 by stevenj] note M ./doc/fftw3.texi -4 +5 Sun Mar 30 15:34:57 EST 2003 stevenj * [project @ 2003-03-30 20:34:57 by stevenj] make non-square UGLY, for now M ./dft/vrank2-transpose.c -4 +4 M ./dft/vrank3-transpose.c -1 +4 Sun Mar 30 15:33:57 EST 2003 stevenj * [project @ 2003-03-30 20:33:57 by stevenj] added -o amnesia to forget_wisdom before each plan M ./tests/bench.c +5 Sun Mar 30 09:41:27 EST 2003 athena * [project @ 2003-03-30 14:41:27 by athena] Report setup time in benchmark M ./libbench2/bench-user.h -1 +4 M ./libbench2/report.c -2 +2 M ./libbench2/speed.c -1 +4 Sat Mar 29 20:21:15 EST 2003 stevenj * [project @ 2003-03-30 01:20:52 by stevenj] comment M ./kernel/transpose.c -1 +2 Sat Mar 29 19:11:10 EST 2003 stevenj * [project @ 2003-03-30 00:11:10 by stevenj] slight change M ./doc/fftw3.texi -4 +4 Sat Mar 29 18:46:16 EST 2003 athena * [project @ 2003-03-29 23:46:16 by athena] More relaxed definition of UGLYness M ./kernel/ct.c -1 +1 Sat Mar 29 15:28:01 EST 2003 stevenj * [project @ 2003-03-29 20:28:01 by stevenj] no more cvs id strings in header files...I'm tired of having to rebuild everything after a commit M ./rdft/codelet-rdft.h -2 M ./rdft/hc2hc.h -2 M ./rdft/rdft.h -2 M ./reodft/reodft.h -2 M ./threads/threads.h -2 Sat Mar 29 15:22:28 EST 2003 stevenj * [project @ 2003-03-29 20:22:28 by stevenj] rdft2 stride unification A ./rdft/rdft2-strides.c M ./rdft/Makefile.am -1 +1 M ./rdft/buffered2.c -20 +11 M ./rdft/direct2.c -5 +5 M ./rdft/rdft.h -1 +2 M ./rdft/rdft2-inplace-strides.c -9 +2 M ./rdft/rdft2-strides.c +38 M ./rdft/rdft2-tensor-max-index.c -7 +4 M ./rdft/vrank-geq1-rdft2.c -26 +9 M ./threads/vrank-geq1-rdft2.c -9 +3 Sat Mar 29 14:38:23 EST 2003 stevenj * [project @ 2003-03-29 19:38:23 by stevenj] preserve in-place-ness M ./rdft/vrank-geq1-rdft2.c -3 +3 Sat Mar 29 14:23:31 EST 2003 stevenj * [project @ 2003-03-29 19:23:31 by stevenj] make nowisdom the default M ./tests/Makefile.am -7 +7 M ./tests/bench.c -2 +2 M ./tests/check.pl -3 +3 Sat Mar 29 14:13:18 EST 2003 athena * [project @ 2003-03-29 19:13:18 by athena] --verbose in paranoid-check produces too much output. Make it quiet. M ./tests/Makefile.am -1 +1 Sat Mar 29 13:45:13 EST 2003 stevenj * [project @ 2003-03-29 18:45:13 by stevenj] fixed transpose bugs...need to check ri-ii before deciding whether Ntuple fits M ./dft/vrank2-transpose.c -5 +9 M ./dft/vrank3-transpose.c -7 +13 M ./kernel/ifftw.h -3 +5 M ./kernel/transpose.c -11 +21 Sat Mar 29 08:10:40 EST 2003 athena * [project @ 2003-03-29 13:10:40 by athena] try more 2^k M ./tests/check.pl +2 Sat Mar 29 08:05:41 EST 2003 athena * [project @ 2003-03-29 13:05:41 by athena] MIN_ALIGNMENT was defined after being used, causing crash in sse2. M ./kernel/ifftw.h -4 +4 Sat Mar 29 03:07:34 EST 2003 stevenj * [project @ 2003-03-29 08:07:34 by stevenj] real transposes are currently unused, and are not needed for MPI code either M ./kernel/Makefile.am -2 +2 M ./kernel/ifftw.h -2 +1 R ./kernel/tensor10.c M ./kernel/transpose.c -180 +18 M ./rdft/Makefile.am -2 +3 M ./rdft/conf.c -3 +5 Sat Mar 29 02:58:39 EST 2003 stevenj * [project @ 2003-03-29 07:58:39 by stevenj] added general transpose A ./kernel/transpose.c M ./dft/vrank2-transpose.c -9 +43 M ./dft/vrank3-transpose.c -25 +57 M ./kernel/Makefile.am -2 +2 M ./kernel/ifftw.h -1 +13 M ./kernel/transpose.c +523 Fri Mar 28 22:49:04 EST 2003 stevenj * [project @ 2003-03-29 03:49:04 by stevenj] added transposition option M ./libbench2/problem.c -1 +18 Fri Mar 28 22:09:22 EST 2003 stevenj * [project @ 2003-03-29 03:09:22 by stevenj] yikes, fixed incorrect applicability of transpose plans A ./kernel/tensor10.c M ./dft/vrank2-transpose.c -3 +3 M ./dft/vrank3-transpose.c -2 +2 M ./kernel/Makefile.am -2 +2 M ./kernel/ifftw.h -1 +2 M ./kernel/tensor10.c +31 M ./rdft/vrank2-transpose.c -2 +2 M ./rdft/vrank3-transpose.c -2 +2 Fri Mar 28 22:06:14 EST 2003 stevenj * [project @ 2003-03-29 03:06:14 by stevenj] in the future, we might want to allow sz->rnk == 0, vecsz->rnk arbitrary to be converted to r2hc (the apply function already should work for this case)...disabled for now, though M ./rdft/dft-r2hc.c -7 +21 Fri Mar 28 19:12:08 EST 2003 stevenj * [project @ 2003-03-29 00:12:08 by stevenj] use most_unaligned in rdft2 M ./kernel/align.c -6 +1 M ./kernel/ifftw.h -2 +1 M ./rdft/vrank-geq1-rdft2.c -6 +11 Fri Mar 28 19:11:47 EST 2003 stevenj * [project @ 2003-03-29 00:11:47 by stevenj] slight change M ./tests/Makefile.am -2 +2 Fri Mar 28 19:00:21 EST 2003 stevenj * [project @ 2003-03-29 00:00:21 by stevenj] output message when checks pass M ./tests/Makefile.am +12 Fri Mar 28 17:21:47 EST 2003 stevenj * [project @ 2003-03-28 22:21:47 by stevenj] added ifndef alloca around alloca stuff M ./kernel/ifftw.h -1 +3 Fri Mar 28 13:45:50 EST 2003 athena * [project @ 2003-03-28 18:45:43 by athena] Proper alignment in rader M ./dft/rader.c -2 +4 M ./dft/vrank-geq1.c -4 +3 M ./kernel/align.c -3 +2 M ./kernel/ifftw.h -2 +2 M ./rdft/dht-rader.c -2 +5 Fri Mar 28 12:43:23 EST 2003 stevenj * [project @ 2003-03-28 17:43:23 by stevenj] whitespace M ./kernel/ifftw.h -1 +2 Fri Mar 28 12:41:39 EST 2003 stevenj * [project @ 2003-03-28 17:41:39 by stevenj] whoops, alloca stuff inside HAVE_ALLOCA M ./kernel/ifftw.h -3 +4 Fri Mar 28 12:35:21 EST 2003 stevenj * [project @ 2003-03-28 17:35:21 by stevenj] make check can afford to be a little bigger M ./tests/Makefile.am -2 +2 Fri Mar 28 12:31:32 EST 2003 stevenj * [project @ 2003-03-28 17:31:32 by stevenj] use same alloca macrology as configure script M ./kernel/ifftw.h -28 +45 Fri Mar 28 03:05:15 EST 2003 stevenj * [project @ 2003-03-28 08:05:15 by stevenj] fallback is no longer needed for mingw M ./kernel/ifftw.h -6 +1 Fri Mar 28 02:58:45 EST 2003 stevenj * [project @ 2003-03-28 07:58:45 by stevenj] alloca fallback for gcc M ./kernel/ifftw.h -3 +4 Fri Mar 28 02:49:59 EST 2003 stevenj * [project @ 2003-03-28 07:49:59 by stevenj] _alloca was added for MinGW, but it causes problems there M ./configure.ac -1 +1 Thu Mar 27 22:06:07 EST 2003 stevenj * [project @ 2003-03-28 03:06:07 by stevenj] fixed most_unaligned for split format M ./kernel/align.c -5 +8 Thu Mar 27 19:01:58 EST 2003 stevenj * [project @ 2003-03-28 00:01:58 by stevenj] whoops M ./Makefile.am -1 +1 Thu Mar 27 19:00:20 EST 2003 stevenj * [project @ 2003-03-28 00:00:20 by stevenj] added pkg-config A ./fftw.pc.in M ./Makefile.am -1 +6 M ./configure.ac +1 M ./fftw.pc.in +10 Thu Mar 27 15:59:01 EST 2003 stevenj * [project @ 2003-03-27 20:59:01 by stevenj] fixed asserts M ./dft/vrank-geq1.c -2 +2 M ./rdft/vrank-geq1-rdft2.c -2 +3 M ./rdft/vrank-geq1.c -2 +2 Thu Mar 27 15:49:53 EST 2003 athena * [project @ 2003-03-27 20:49:53 by athena] Do not adjust r/i pointers separately. M ./kernel/align.c -4 +5 Thu Mar 27 15:17:40 EST 2003 athena * [project @ 2003-03-27 20:17:40 by athena] iForgot to add files A ./dft/simd/n2b.h A ./dft/simd/n2f.h M ./dft/simd/n2b.h +25 M ./dft/simd/n2f.h +25 Thu Mar 27 15:10:41 EST 2003 athena * [project @ 2003-03-27 20:10:36 by athena] Specialized n simd codelets for unit vector stride. A ./dft/simd/n2b.c A ./dft/simd/n2f.c M ./configure.ac +4 M ./dft/simd/Makefile.am -2 +2 M ./dft/simd/codelets/Makefile.am -8 +54 M ./dft/simd/n1b.c -3 +3 M ./dft/simd/n1b.h +4 M ./dft/simd/n1f.c -3 +3 M ./dft/simd/n1f.h +4 M ./dft/simd/n2b.c +49 M ./dft/simd/n2f.c +49 Thu Mar 27 08:22:03 EST 2003 athena * [project @ 2003-03-27 13:22:03 by athena] Changed version number to beta2 M ./configure.ac -1 +1 Thu Mar 27 06:37:07 EST 2003 athena * [project @ 2003-03-27 11:37:07 by athena] Changed alignment requirements for n1 simd codelets. Changed mechanism for detecting lack of alignment. M ./api/mapflags.c -1 +1 M ./dft/simd/n1b.c -4 +4 M ./dft/simd/n1b.h -3 M ./dft/simd/n1f.c -4 +4 M ./dft/simd/n1f.h -3 M ./dft/simd/q1b.c -1 +1 M ./dft/simd/q1f.c -1 +1 M ./dft/simd/t1b.c -1 +1 M ./dft/simd/t1f.c -1 +1 M ./dft/vrank-geq1.c -7 +7 M ./kernel/align.c -5 +42 M ./kernel/ifftw.h -3 +5 M ./rdft/vrank-geq1-rdft2.c -4 +7 M ./rdft/vrank-geq1.c -5 +5 Thu Mar 27 04:25:06 EST 2003 athena * [project @ 2003-03-27 09:25:06 by athena] Oops, wrong place for hook M ./tests/bench.c -2 +1 Thu Mar 27 02:37:52 EST 2003 stevenj * [project @ 2003-03-27 07:37:52 by stevenj] added comments to codelet makefiles, to aid people wanting to generate their own code M ./dft/codelets/inplace/Makefile.am -1 +23 M ./dft/codelets/standard/Makefile.am +24 M ./dft/k7/codelets/Makefile.am +25 M ./dft/simd/codelets/Makefile.am -1 +32 M ./rdft/codelets/hc2r/Makefile.am +25 M ./rdft/codelets/r2hc/Makefile.am +25 M ./rdft/codelets/r2r/Makefile.am +18 Thu Mar 27 01:42:27 EST 2003 stevenj * [project @ 2003-03-27 06:42:27 by stevenj] Matteo is also a copyright holder M ./doc/FAQ/fftw-faq.bfnn -1 +1 Thu Mar 27 01:41:08 EST 2003 stevenj * [project @ 2003-03-27 06:41:08 by stevenj] FORTRAN is officially Fortran, these days M ./doc/FAQ/fftw-faq.bfnn -2 +2 Thu Mar 27 01:40:32 EST 2003 stevenj * [project @ 2003-03-27 06:40:32 by stevenj] punctuation M ./doc/FAQ/fftw-faq.bfnn -1 +1 Thu Mar 27 01:40:14 EST 2003 stevenj * [project @ 2003-03-27 06:40:14 by stevenj] don't use "wrapper" M ./doc/FAQ/fftw-faq.bfnn -1 +1 Thu Mar 27 01:37:53 EST 2003 stevenj * [project @ 2003-03-27 06:37:53 by stevenj] plural M ./doc/FAQ/fftw-faq.bfnn -1 +1 Thu Mar 27 01:35:32 EST 2003 stevenj * [project @ 2003-03-27 06:35:32 by stevenj] grammar M ./doc/FAQ/fftw-faq.bfnn -1 +1 Thu Mar 27 01:33:35 EST 2003 stevenj * [project @ 2003-03-27 06:33:35 by stevenj] better phrasing M ./doc/FAQ/fftw-faq.bfnn -1 +1 Wed Mar 26 22:47:58 EST 2003 stevenj * [project @ 2003-03-27 03:47:58 by stevenj] stddef.h should not be needed anymore for this file M ./kernel/align.c -2 +1 Wed Mar 26 22:13:48 EST 2003 stevenj * [project @ 2003-03-27 03:13:48 by stevenj] added comments for Franz mode M ./dft/codelets/standard/Makefile.am -5 +5 Wed Mar 26 22:11:58 EST 2003 stevenj * [project @ 2003-03-27 03:10:11 by stevenj] clarification M ./dft/simd/codelets/Makefile.am -3 +4 Wed Mar 26 22:08:51 EST 2003 stevenj * [project @ 2003-03-27 03:08:51 by stevenj] commented on FRANZ codelets M ./dft/simd/codelets/Makefile.am -5 +5 Wed Mar 26 22:06:45 EST 2003 stevenj * [project @ 2003-03-27 03:06:45 by stevenj] updated M ./NEWS +23 Wed Mar 26 21:51:15 EST 2003 stevenj * [project @ 2003-03-27 02:51:15 by stevenj] disable DIF codelets, since they are never used (apparently) except for some non-power-of-two sizes...improve support for the latter by adding size 3, 5, and 6 q^2 codelets. M ./dft/codelets/inplace/Makefile.am -4 +8 Wed Mar 26 20:07:11 EST 2003 stevenj * [project @ 2003-03-27 01:07:11 by stevenj] DHT has no forward/backward M ./doc/fftw3.texi -3 +3 Wed Mar 26 19:46:12 EST 2003 fftw * [project @ 2003-03-27 00:46:12 by fftw] added hacky way to use an arbitrary flag M ./tests/bench.c +1 Wed Mar 26 19:44:31 EST 2003 athena * [project @ 2003-03-27 00:44:31 by athena] Better place to install hook M ./tests/bench.c -2 +2 Wed Mar 26 19:40:28 EST 2003 stevenj * [project @ 2003-03-27 00:40:28 by stevenj] noted that the user should run make check if they think FFTW has a bug M ./doc/FAQ/fftw-faq.bfnn -6 +8 Wed Mar 26 17:31:16 EST 2003 athena * [project @ 2003-03-26 22:31:16 by athena] Oops, what am I thinking M ./kernel/planner.c -5 +8 Wed Mar 26 17:23:56 EST 2003 athena * [project @ 2003-03-26 22:23:56 by athena] Grrr.... fixed bug in estimator M ./kernel/planner.c -8 +5 Wed Mar 26 17:16:19 EST 2003 athena * [project @ 2003-03-26 22:16:19 by athena] Oops---the flop count was right. The estimator is broken elsewhere. M ./genfft/c.ml -5 +2 Wed Mar 26 14:28:41 EST 2003 athena * [project @ 2003-03-26 19:28:41 by athena] Fixed SIMD estimator M ./genfft/c.ml -2 +5 Wed Mar 26 07:45:03 EST 2003 athena * [project @ 2003-03-26 12:45:03 by athena] Added twidsq simd codelets A ./dft/simd/q1b.c A ./dft/simd/q1b.h A ./dft/simd/q1f.c A ./dft/simd/q1f.h A ./genfft/gen_twidsq_c.ml M ./dft/simd/Makefile.am -1 +2 M ./dft/simd/codelets/Makefile.am -1 +12 M ./dft/simd/q1b.c +44 M ./dft/simd/q1b.h +25 M ./dft/simd/q1f.c +44 M ./dft/simd/q1f.h +25 M ./dft/simd/t1b.c +1 M ./dft/simd/t1f.c +1 M ./genfft/Makefile.am -8 +14 M ./genfft/gen_twiddle_c.ml -3 +3 M ./genfft/gen_twidsq_c.ml +195 M ./support/Makefile.codelets -1 +2 Tue Mar 25 23:33:03 EST 2003 stevenj * [project @ 2003-03-26 04:33:03 by stevenj] gensrc -> genfft M ./doc/fftw3.texi -2 +2 Tue Mar 25 23:32:16 EST 2003 stevenj * [project @ 2003-03-26 04:32:16 by stevenj] newline M ./TODO +1 Tue Mar 25 19:17:08 EST 2003 athena * [project @ 2003-03-26 00:17:08 by athena] Noted need to add dif simd codelets M ./TODO +1 Tue Mar 25 13:03:47 EST 2003 stevenj * [project @ 2003-03-25 18:03:47 by stevenj] noted shift M ./doc/fftw3.texi -3 +2 Tue Mar 25 13:02:47 EST 2003 stevenj * [project @ 2003-03-25 18:02:47 by stevenj] clarification M ./doc/fftw3.texi -4 +14 Tue Mar 25 12:46:44 EST 2003 stevenj * [project @ 2003-03-25 17:46:44 by stevenj] need make after bootstrap M ./doc/fftw3.texi -2 +3 Tue Mar 25 12:31:49 EST 2003 stevenj * [project @ 2003-03-25 17:31:49 by stevenj] slight change M ./doc/fftw3.texi -3 +3 Tue Mar 25 12:30:56 EST 2003 stevenj * [project @ 2003-03-25 17:30:56 by stevenj] libtool is also needed M ./doc/fftw3.texi -4 +4 Tue Mar 25 12:29:52 EST 2003 stevenj * [project @ 2003-03-25 17:29:52 by stevenj] added code generator introduction M ./doc/fftw3.texi -13 +63 Tue Mar 25 11:51:49 EST 2003 stevenj * [project @ 2003-03-25 16:51:49 by stevenj] added support for REDFT/RODFT/DHT direct codelets A ./genfft/gen_r2r.ml A ./rdft/kr2r.c A ./rdft/codelets/r2r/ A ./rdft/codelets/r2r.c A ./rdft/codelets/r2r.h A ./rdft/codelets/r2r/Makefile.am M ./Makefile.am +1 M ./configure.ac +1 M ./genfft/Makefile.am -6 +6 M ./genfft/complex.ml -1 +8 M ./genfft/complex.mli -1 +2 M ./genfft/gen_r2r.ml +240 R ./genfft/gen_trig.ml M ./genfft/trig.ml -60 +47 M ./rdft/Makefile.am -3 +3 M ./rdft/codelet-rdft.h -2 +2 M ./rdft/codelets/Makefile.am -2 +2 M ./rdft/codelets/r2r/Makefile.am +72 M ./rdft/codelets/r2r.c +38 M ./rdft/codelets/r2r.h +23 M ./rdft/conf.c -1 +2 M ./rdft/direct.c -16 +58 M ./rdft/kr2r.c +28 M ./rdft/rdft.h -1 +2 M ./support/Makefile.codelets -1 +2 Tue Mar 25 11:29:29 EST 2003 stevenj * [project @ 2003-03-25 16:29:29 by stevenj] noted ARM bug; thanks to Jay Treacy M ./doc/FAQ/fftw-faq.bfnn -1 +2 Tue Mar 25 07:55:54 EST 2003 athena * [project @ 2003-03-25 12:55:54 by athena] bugfix from Stefan M ./genfft-k7/vK7Optimization.ml -1 +2 Mon Mar 24 15:59:08 EST 2003 stevenj * [project @ 2003-03-24 20:59:08 by stevenj] slight change M ./doc/fftw3.texi -2 +2 Mon Mar 24 15:58:44 EST 2003 stevenj * [project @ 2003-03-24 20:58:44 by stevenj] caveat M ./doc/fftw3.texi -2 +2 Mon Mar 24 15:58:04 EST 2003 stevenj * [project @ 2003-03-24 20:58:04 by stevenj] warning about DHT M ./doc/fftw3.texi -1 +8 Mon Mar 24 08:34:14 EST 2003 athena * [project @ 2003-03-24 13:34:14 by athena] Oops M ./dft/k7/codelets/Makefile.am -2 +2 Mon Mar 24 08:13:15 EST 2003 athena * [project @ 2003-03-24 13:13:15 by athena] Regression test for p4fftwgel M ./dft/k7/codelets/Makefile.am -2 +2 M ./tests/Makefile.am +3 M ./tests/check.pl +4 Mon Mar 24 03:09:06 EST 2003 stevenj * [project @ 2003-03-24 08:09:06 by stevenj] make check is faster, old tests are in make bigcheck M ./tests/Makefile.am -1 +6 Sat Mar 22 00:41:21 EST 2003 stevenj * [project @ 2003-03-22 05:41:21 by stevenj] note M ./doc/FAQ/fftw-faq.bfnn -1 +1 Sat Mar 22 00:40:05 EST 2003 stevenj * [project @ 2003-03-22 05:40:05 by stevenj] whoops, line wrapping M ./doc/FAQ/fftw-faq.bfnn -2 +1 Fri Mar 21 15:10:00 EST 2003 athena * [project @ 2003-03-21 20:10:00 by athena] Franz-mode codelets even without SIMD. (disabled) M ./dft/codelets/standard/Makefile.am -2 +59 M ./genfft/gen_notw.ml -5 +6 M ./genfft/gen_twiddle.ml -4 +5 Fri Mar 21 09:09:30 EST 2003 athena * [project @ 2003-03-21 14:09:30 by athena] Bug is in netbsd-1.6, not 1.5 M ./doc/FAQ/fftw-faq.bfnn -1 +1 Fri Mar 21 07:45:48 EST 2003 athena * [project @ 2003-03-21 12:45:48 by athena] const cast, should placate c++ compilers. M ./simd/simd-altivec.h -1 +1 Thu Mar 20 18:49:49 EST 2003 stevenj * [project @ 2003-03-20 23:49:49 by stevenj] added FAQ on why plans are array-specific M ./doc/FAQ/fftw-faq.bfnn +26 Thu Mar 20 16:12:56 EST 2003 stevenj * [project @ 2003-03-20 21:12:56 by stevenj] comment fix M ./reodft/reodft010e-r2hc.c -3 +3 Thu Mar 20 16:12:15 EST 2003 stevenj * [project @ 2003-03-20 21:12:15 by stevenj] noted comparison to NR M ./reodft/reodft010e-r2hc.c -1 +5 Wed Mar 19 20:13:16 EST 2003 stevenj * [project @ 2003-03-20 01:13:16 by stevenj] whoops, C99 complex didn't work if complex is a macro (as it is with glibc); thanks to Keh-Cheng Chu for the bug report M ./api/fftw3.h -5 +4 Wed Mar 19 16:52:54 EST 2003 stevenj * [project @ 2003-03-19 21:52:54 by stevenj] noted in help that --enable-k7 enables 3dnow, and that --enable-3dnow is only a fallback M ./configure.ac -2 +2 Wed Mar 19 15:09:52 EST 2003 athena * [project @ 2003-03-19 20:09:52 by athena] New gcc bug. html.refs was not in repository/distribution. A ./doc/FAQ/html.refs M ./doc/FAQ/Makefile.am -1 +1 M ./doc/FAQ/fftw-faq.bfnn -1 +13 M ./doc/FAQ/html.refs +6 Wed Mar 19 10:09:16 EST 2003 athena * [project @ 2003-03-19 15:09:16 by athena] Don't write wisdom if you don't have it. M ./tests/bench.c -1 +1 Tue Mar 18 15:44:41 EST 2003 athena * [project @ 2003-03-18 20:44:41 by athena] Added index entries for DHT. Similarly for DCT, DST M ./doc/fftw3.texi -1 +6 Tue Mar 18 14:50:04 EST 2003 stevenj * [project @ 2003-03-18 19:50:04 by stevenj] execute should not go through C api, for efficiency M ./api/f77api.c +2 M ./api/f77funcs.h -5 +10 Tue Mar 18 06:14:51 EST 2003 athena * [project @ 2003-03-18 11:14:51 by athena] Renamed FFTW_IODIM, FFTW_R2R_KIND M ./api/fftw3.h -6 +6 Tue Mar 18 00:30:17 EST 2003 stevenj * [project @ 2003-03-18 05:30:17 by stevenj] added rfftwnd.eps to dist, so that transfig is not required for people trying to build other formats (e.g. ps); thanks to Brian Gough for the bug report M ./doc/Makefile.am -1 +1 Mon Mar 17 15:17:59 EST 2003 stevenj * [project @ 2003-03-17 20:17:59 by stevenj] pointer to upgrading section from tutorial M ./doc/fftw3.texi -1 +4 Mon Mar 17 14:44:40 EST 2003 stevenj * [project @ 2003-03-17 19:44:40 by stevenj] make print_plan and fprint_plan, so that the former can be more easily called from other languages M ./api/f77funcs.h -2 +2 M ./api/fftw3.h -2 +3 M ./api/print-plan.c -1 +6 M ./doc/fftw3.texi -3 +5 M ./tests/bench.c -1 +1 Mon Mar 17 14:19:10 EST 2003 stevenj * [project @ 2003-03-17 19:19:10 by stevenj] whoops, forgot to change equation image links to .png M ./doc/fftw3.texi -14 +14 Mon Mar 17 04:15:50 EST 2003 athena * [project @ 2003-03-17 09:15:50 by athena] fixed c++ linkage problems M ./api/fftw3.h -4 +4 M ./api/version.c -8 +5 M ./support/Makefile.codelets -1 +2 Mon Mar 17 03:25:17 EST 2003 athena * [project @ 2003-03-17 08:25:17 by athena] Removed ``const'', otherwise c++ link fails M ./api/fftw3.h -4 +4 M ./api/version.c -4 +4 Sun Mar 16 20:24:31 EST 2003 stevenj * [project @ 2003-03-17 01:24:31 by stevenj] fixed C++ annoyances: void* casts, and global variables are static by default(?!?) M ./api/f77api.c -4 +5 M ./api/f77funcs.h -3 +5 M ./api/version.c -4 +4 M ./libbench2/allocate.c -9 +9 M ./libbench2/getopt-utils.c -2 +2 M ./libbench2/problem.c -3 +3 M ./libbench2/speed.c -2 +2 M ./libbench2/timer.c -2 +2 M ./libbench2/verify-r2r.c -3 +3 M ./libbench2/zero.c -9 +9 M ./support/Makefile.codelets -1 +1 M ./tests/bench.c -37 +66 M ./tests/hook.c -2 +2 M ./tools/fftw-wisdom.c -1 +1 Sat Aug 12 23:06:56 EDT 2006 Unknown tagger tagged fftw-3-0-beta1 Sun Mar 16 15:29:11 EST 2003 stevenj * [project @ 2003-03-16 20:29:11 by stevenj] ranlib bug is in binutils M ./doc/FAQ/fftw-faq.bfnn -4 +5 Sun Mar 16 15:26:42 EST 2003 stevenj * [project @ 2003-03-16 20:26:42 by stevenj] ranlib Irix bug M ./doc/FAQ/fftw-faq.bfnn +5 Sun Mar 16 15:13:35 EST 2003 stevenj * [project @ 2003-03-16 20:13:35 by stevenj] start with random tests M ./tests/check.pl -1 +1 Sun Mar 16 15:00:04 EST 2003 stevenj * [project @ 2003-03-16 20:00:03 by stevenj] silenced some compiler warnings, eliminated unused variables, and fixed Makefile.am for f77funcs.h M ./api/Makefile.am -5 +1 M ./dft/direct.c -3 +2 M ./kernel/ifftw.h -4 +4 M ./libbench2/verify-r2r.c -3 M ./rdft/direct.c -3 +2 M ./rdft/direct2.c -3 +2 M ./threads/Makefile.am -3 +1 Sun Mar 16 14:55:13 EST 2003 stevenj * [project @ 2003-03-16 19:55:13 by stevenj] whoops M ./doc/FAQ/fftw-faq.bfnn -1 +1 Sun Mar 16 14:28:22 EST 2003 stevenj * [project @ 2003-03-16 19:28:22 by stevenj] 3dnow is float M ./doc/fftw3.texi -2 +2 Sun Mar 16 14:27:45 EST 2003 stevenj * [project @ 2003-03-16 19:27:45 by stevenj] fixed k7 docs M ./doc/fftw3.texi -17 +10 Sun Mar 16 14:19:10 EST 2003 stevenj * [project @ 2003-03-16 19:19:10 by stevenj] SGI compilers now support inline M ./kernel/cycle.h -3 +3 Sun Mar 16 14:18:32 EST 2003 stevenj * [project @ 2003-03-16 19:18:32 by stevenj] cruft M ./kernel/cycle.h -2 +1 Sun Mar 16 14:15:47 EST 2003 stevenj * [project @ 2003-03-16 19:15:47 by stevenj] texinfo doesn't like commas in nodes M ./doc/fftw3.texi -16 +15 Sun Mar 16 13:52:04 EST 2003 stevenj * [project @ 2003-03-16 18:48:30 by stevenj] updated M ./ChangeLog -47 +1389 M ./README +28 Sun Mar 16 13:47:44 EST 2003 stevenj * [project @ 2003-03-16 18:47:44 by stevenj] f77funcs.c -> f77funcs.h so that people don't try to compile it A ./api/f77funcs.h A ./threads/f77funcs.h M ./api/f77api.c -4 +4 R ./api/f77funcs.c M ./api/f77funcs.h +366 M ./threads/f77api.c -4 +4 R ./threads/f77funcs.c M ./threads/f77funcs.h +39 Sun Mar 16 13:46:11 EST 2003 stevenj * [project @ 2003-03-16 18:46:11 by stevenj] minor changes M ./doc/FAQ/fftw-faq.bfnn -7 +4 Sun Mar 16 13:39:24 EST 2003 stevenj * [project @ 2003-03-16 18:39:24 by stevenj] updated compiler bug list M ./doc/FAQ/fftw-faq.bfnn -40 +20 Sun Mar 16 13:39:07 EST 2003 stevenj * [project @ 2003-03-16 18:39:07 by stevenj] noted how to set CC M ./doc/fftw3.texi -4 +10 Sun Mar 16 13:01:01 EST 2003 stevenj * [project @ 2003-03-16 18:01:01 by stevenj] TODONE M ./TODO -10 Sun Mar 16 13:00:42 EST 2003 stevenj * [project @ 2003-03-16 18:00:42 by stevenj] yikes, bugfix M ./threads/vrank-geq1-rdft2.c -19 +11 Sun Mar 16 10:26:28 EST 2003 stevenj * [project @ 2003-03-16 15:26:28 by stevenj] whoops M ./kernel/ifftw.h -2 +2 Sun Mar 16 09:24:19 EST 2003 athena * [project @ 2003-03-16 14:24:19 by athena] Report SIMD extensions in version string M ./api/version.c -2 +26 Sat Mar 15 18:56:11 EST 2003 stevenj * [project @ 2003-03-15 23:56:11 by stevenj] more verbose output M ./tests/bench.c +2 Sat Mar 15 17:41:25 EST 2003 stevenj * [project @ 2003-03-15 22:41:25 by stevenj] a couple of additional non-Unix instructions M ./doc/fftw3.texi -1 +9 Sat Mar 15 17:15:26 EST 2003 stevenj * [project @ 2003-03-15 22:15:26 by stevenj] hyphen M ./doc/FAQ/fftw-faq.bfnn -1 +1 Sat Mar 15 17:12:29 EST 2003 stevenj * [project @ 2003-03-15 22:12:29 by stevenj] softened M ./doc/FAQ/fftw-faq.bfnn -3 +3 Sat Mar 15 17:09:44 EST 2003 stevenj * [project @ 2003-03-15 22:09:44 by stevenj] added FAQ, used PNGs A ./doc/equation-dft.png A ./doc/equation-dht.png A ./doc/equation-idft.png A ./doc/equation-redft00.png A ./doc/equation-redft01.png A ./doc/equation-redft10.png A ./doc/equation-redft11.png A ./doc/equation-rodft00.png A ./doc/equation-rodft01.png A ./doc/equation-rodft10.png A ./doc/FAQ/ A ./doc/FAQ/Makefile.am A ./doc/FAQ/bfnnconv.pl A ./doc/FAQ/fftw-faq.bfnn A ./doc/FAQ/m-ascii.pl A ./doc/FAQ/m-html.pl A ./doc/FAQ/m-info.pl A ./doc/FAQ/m-lout.pl A ./doc/FAQ/m-post.pl A ./doc/equation-rodft11.png M ./configure.ac +1 M ./doc/FAQ/Makefile.am +14 M ./doc/FAQ/bfnnconv.pl +298 M ./doc/FAQ/fftw-faq.bfnn +492 M ./doc/FAQ/m-ascii.pl +189 M ./doc/FAQ/m-html.pl +337 M ./doc/FAQ/m-info.pl +226 M ./doc/FAQ/m-lout.pl +242 M ./doc/FAQ/m-post.pl +189 M ./doc/Makefile.am -4 +6 R ./doc/equation-dft.gif M ./doc/equation-dft.png R ./doc/equation-dht.gif M ./doc/equation-dht.png R ./doc/equation-idft.gif M ./doc/equation-idft.png R ./doc/equation-redft00.gif M ./doc/equation-redft00.png R ./doc/equation-redft01.gif M ./doc/equation-redft01.png R ./doc/equation-redft10.gif M ./doc/equation-redft10.png R ./doc/equation-redft11.gif M ./doc/equation-redft11.png R ./doc/equation-rodft00.gif M ./doc/equation-rodft00.png R ./doc/equation-rodft01.gif M ./doc/equation-rodft01.png R ./doc/equation-rodft10.gif M ./doc/equation-rodft10.png R ./doc/equation-rodft11.gif M ./doc/equation-rodft11.png Sat Mar 15 15:29:43 EST 2003 stevenj * [project @ 2003-03-15 20:29:42 by stevenj] great copyright update M ./COPYRIGHT -2 +2 M ./TODO -6 M ./api/api.h -2 +2 M ./api/apiplan.c -2 +2 M ./api/configure.c -2 +2 M ./api/execute-dft-c2r.c -2 +2 M ./api/execute-dft-r2c.c -2 +2 M ./api/execute-dft.c -2 +2 M ./api/execute-r2r.c -2 +2 M ./api/execute.c -2 +2 M ./api/export-wisdom-to-file.c -2 +2 M ./api/export-wisdom-to-string.c -2 +2 M ./api/export-wisdom.c -2 +2 M ./api/extract-reim.c -2 +2 M ./api/f77api.c -2 +2 M ./api/f77funcs.c -2 +2 M ./api/fftw3.h -3 +3 M ./api/flops.c -2 +2 M ./api/forget-wisdom.c -2 +2 M ./api/import-system-wisdom.c -2 +2 M ./api/import-wisdom-from-file.c -2 +2 M ./api/import-wisdom-from-string.c -2 +2 M ./api/import-wisdom.c -2 +2 M ./api/map-r2r-kind.c -2 +2 M ./api/mapflags.c -2 +2 M ./api/mkprinter-file.c -2 +2 M ./api/mktensor-iodims.c -2 +2 M ./api/mktensor-rowmajor.c -2 +2 M ./api/plan-dft-1d.c -2 +2 M ./api/plan-dft-2d.c -2 +2 M ./api/plan-dft-3d.c -2 +2 M ./api/plan-dft-c2r-1d.c -2 +2 M ./api/plan-dft-c2r-2d.c -2 +2 M ./api/plan-dft-c2r-3d.c -2 +2 M ./api/plan-dft-c2r.c -2 +2 M ./api/plan-dft-r2c-1d.c -2 +2 M ./api/plan-dft-r2c-2d.c -2 +2 M ./api/plan-dft-r2c-3d.c -2 +2 M ./api/plan-dft-r2c.c -2 +2 M ./api/plan-dft.c -2 +2 M ./api/plan-guru-dft-c2r.c -2 +2 M ./api/plan-guru-dft-r2c.c -2 +2 M ./api/plan-guru-dft.c -2 +2 M ./api/plan-guru-r2r.c -2 +2 M ./api/plan-many-dft-c2r.c -2 +2 M ./api/plan-many-dft-r2c.c -2 +2 M ./api/plan-many-dft.c -2 +2 M ./api/plan-many-r2r.c -2 +2 M ./api/plan-r2r-1d.c -2 +2 M ./api/plan-r2r-2d.c -2 +2 M ./api/plan-r2r-3d.c -2 +2 M ./api/plan-r2r.c -2 +2 M ./api/print-plan.c -2 +2 M ./api/rdft2-pad.c -2 +2 M ./api/the-planner.c -2 +2 M ./api/version.c -3 +3 M ./api/x77.h -2 +2 M ./dft/buffered.c -3 +3 M ./dft/codelet-dft.h -3 +3 M ./dft/codelets/n.c -2 +2 M ./dft/codelets/n.h -2 +2 M ./dft/codelets/t.c -2 +2 M ./dft/codelets/t.h -2 +2 M ./dft/conf.c -3 +3 M ./dft/ct-dif.c -3 +3 M ./dft/ct-dit.c -3 +3 M ./dft/ct-ditbuf.c -3 +3 M ./dft/ct-ditf.c -3 +3 M ./dft/ct.c -3 +3 M ./dft/ct.h -3 +3 M ./dft/dft.h -3 +3 M ./dft/direct.c -3 +3 M ./dft/generic.c -2 +2 M ./dft/indirect.c -3 +3 M ./dft/k7/k7.c -3 +3 M ./dft/kdft-dif.c -3 +3 M ./dft/kdft-difsq.c -3 +3 M ./dft/kdft-dit.c -3 +3 M ./dft/kdft.c -3 +3 M ./dft/nop.c -3 +3 M ./dft/plan.c -3 +3 M ./dft/problem.c -3 +3 M ./dft/rader-omega.c -2 +2 M ./dft/rader.c -2 +2 M ./dft/rank-geq2.c -3 +3 M ./dft/rank0.c -3 +3 M ./dft/simd/n1b.c -2 +2 M ./dft/simd/n1b.h -2 +2 M ./dft/simd/n1f.c -2 +2 M ./dft/simd/n1f.h -2 +2 M ./dft/simd/t1b.c -2 +2 M ./dft/simd/t1b.h -2 +2 M ./dft/simd/t1f.c -2 +2 M ./dft/simd/t1f.h -2 +2 M ./dft/solve.c -3 +3 M ./dft/vrank-geq1.c -3 +3 M ./dft/vrank2-transpose.c -3 +3 M ./dft/vrank3-transpose.c -3 +3 M ./dft/zero.c -3 +3 M ./doc/f77_wisdom.f -2 +2 M ./doc/fftw3.texi -4 +4 M ./genfft/algsimp.ml -4 +4 M ./genfft/algsimp.mli -3 +3 M ./genfft/annotate.ml -4 +4 M ./genfft/annotate.mli -3 +3 M ./genfft/assoctable.ml -3 +3 M ./genfft/assoctable.mli -3 +3 M ./genfft/c.ml -3 +3 M ./genfft/c.mli -3 +3 M ./genfft/complex.ml -3 +3 M ./genfft/complex.mli -3 +3 M ./genfft/conv.ml -2 +2 M ./genfft/conv.mli -3 +3 M ./genfft/dag.ml -3 +3 M ./genfft/dag.mli -3 +3 M ./genfft/expr.ml -3 +3 M ./genfft/expr.mli -3 +3 M ./genfft/fft.ml -4 +4 M ./genfft/fft.mli -3 +3 M ./genfft/gen_athnotw.ml -4 +4 M ./genfft/gen_athtw.ml -4 +4 M ./genfft/gen_conv.ml -4 +4 M ./genfft/gen_hc2hc.ml -4 +4 M ./genfft/gen_hc2r.ml -4 +4 M ./genfft/gen_notw.ml -4 +4 M ./genfft/gen_notw_c.ml -4 +4 M ./genfft/gen_r2hc.ml -4 +4 M ./genfft/gen_trig.ml -4 +4 M ./genfft/gen_twiddle.ml -4 +4 M ./genfft/gen_twiddle_c.ml -4 +4 M ./genfft/gen_twidsq.ml -4 +4 M ./genfft/genutil.ml -3 +3 M ./genfft/littlesimp.ml -3 +3 M ./genfft/littlesimp.mli -3 +3 M ./genfft/magic.ml -3 +3 M ./genfft/monads.ml -3 +3 M ./genfft/number.ml -3 +3 M ./genfft/number.mli -3 +3 M ./genfft/oracle.ml -3 +3 M ./genfft/oracle.mli -3 +3 M ./genfft/schedule.ml -3 +3 M ./genfft/schedule.mli -3 +3 M ./genfft/simd.ml -3 +3 M ./genfft/simd.mli -3 +3 M ./genfft/simdmagic.ml -3 +3 M ./genfft/to_alist.ml -3 +3 M ./genfft/to_alist.mli -3 +3 M ./genfft/trig.ml -3 +3 M ./genfft/trig.mli -3 +3 M ./genfft/twiddle.ml -3 +3 M ./genfft/twiddle.mli -3 +3 M ./genfft/unique.ml -3 +3 M ./genfft/unique.mli -3 +3 M ./genfft/util.ml -3 +3 M ./genfft/util.mli -3 +3 M ./genfft/variable.ml -3 +3 M ./genfft/variable.mli -3 +3 M ./genfft-k7/algsimp.ml -4 +4 M ./genfft-k7/algsimp.mli -3 +3 M ./genfft-k7/assoctable.ml -3 +3 M ./genfft-k7/assoctable.mli -3 +3 M ./genfft-k7/expr.ml -3 +3 M ./genfft-k7/expr.mli -3 +3 M ./genfft-k7/fft.ml -4 +4 M ./genfft-k7/littlesimp.ml -3 +3 M ./genfft-k7/littlesimp.mli -3 +3 M ./genfft-k7/monads.ml -3 +3 M ./genfft-k7/number.ml -3 +3 M ./genfft-k7/number.mli -3 +3 M ./genfft-k7/oracle.ml -3 +3 M ./genfft-k7/oracle.mli -3 +3 M ./genfft-k7/to_alist.ml -3 +3 M ./genfft-k7/to_alist.mli -3 +3 M ./genfft-k7/twiddle.ml -3 +3 M ./genfft-k7/twiddle.mli -3 +3 M ./kernel/align.c -3 +3 M ./kernel/alloc.c -3 +3 M ./kernel/assert.c -3 +3 M ./kernel/awake.c -3 +3 M ./kernel/buffered.c -2 +2 M ./kernel/ct.c -2 +2 M ./kernel/cycle.h -3 +3 M ./kernel/debug.c -3 +3 M ./kernel/hash.c -2 +2 M ./kernel/iabs.c -3 +3 M ./kernel/ifftw.h -3 +3 M ./kernel/md5-1.c -2 +2 M ./kernel/md5.c -2 +2 M ./kernel/minmax.c -3 +3 M ./kernel/ops.c -3 +3 M ./kernel/pickdim.c -3 +3 M ./kernel/plan.c -3 +3 M ./kernel/planner.c -2 +2 M ./kernel/primes.c -3 +3 M ./kernel/print.c -3 +3 M ./kernel/problem.c -3 +3 M ./kernel/rader.c -2 +2 M ./kernel/scan.c -3 +3 M ./kernel/solver.c -3 +3 M ./kernel/solvtab.c -3 +3 M ./kernel/square.c -3 +3 M ./kernel/stride.c -3 +3 M ./kernel/tensor.c -3 +3 M ./kernel/tensor1.c -3 +3 M ./kernel/tensor2.c -3 +3 M ./kernel/tensor4.c -3 +3 M ./kernel/tensor5.c -3 +3 M ./kernel/tensor7.c -3 +3 M ./kernel/tensor8.c -3 +3 M ./kernel/tensor9.c -3 +3 M ./kernel/timer.c -3 +3 M ./kernel/trig.c -3 +3 M ./kernel/trig1.c -3 +3 M ./kernel/twiddle.c -3 +3 M ./libbench/bench-main.c -2 +2 M ./libbench/bench-user.h -2 +2 M ./libbench/bench.h -2 +2 M ./libbench/can-do.c -2 +2 M ./libbench/getopt-utils.c -2 +2 M ./libbench/info.c -2 +2 M ./libbench/main.c -2 +2 M ./libbench/prime.c -2 +2 M ./libbench/problem.c -2 +2 M ./libbench/report.c -2 +2 M ./libbench/speed.c -2 +2 M ./libbench/timer.c -2 +2 M ./libbench/util.c -1 +1 M ./libbench/verify.c -2 +2 M ./libbench/zero.c -2 +2 M ./libbench2/aligned-main.c -2 +2 M ./libbench2/bench-main.c -2 +2 M ./libbench2/bench-user.h -2 +2 M ./libbench2/bench.h -2 +2 M ./libbench2/can-do.c -2 +2 M ./libbench2/dotens2.c -3 +3 M ./libbench2/getopt-utils.c -2 +2 M ./libbench2/info.c -2 +2 M ./libbench2/main.c -2 +2 M ./libbench2/problem.c -2 +2 M ./libbench2/report.c -2 +2 M ./libbench2/speed.c -2 +2 M ./libbench2/tensor.c -2 +2 M ./libbench2/timer.c -2 +2 M ./libbench2/useropt.c -2 +2 M ./libbench2/util.c -1 +1 M ./libbench2/verify-dft.c -3 +3 M ./libbench2/verify-lib.c -3 +3 M ./libbench2/verify-r2r.c -2 +2 M ./libbench2/verify-rdft2.c -3 +3 M ./libbench2/verify.c -2 +2 M ./libbench2/verify.h -2 +2 M ./libbench2/zero.c -2 +2 M ./rdft/buffered.c -3 +3 M ./rdft/buffered2.c -3 +3 M ./rdft/codelet-rdft.h -3 +3 M ./rdft/codelets/hb.h -2 +2 M ./rdft/codelets/hc2r.c -2 +2 M ./rdft/codelets/hc2r.h -2 +2 M ./rdft/codelets/hc2rIII.h -2 +2 M ./rdft/codelets/hf.h -2 +2 M ./rdft/codelets/hfb.c -2 +2 M ./rdft/codelets/r2hc.c -2 +2 M ./rdft/codelets/r2hc.h -2 +2 M ./rdft/codelets/r2hcII.h -2 +2 M ./rdft/conf.c -3 +3 M ./rdft/dft-r2hc.c -3 +3 M ./rdft/dht-r2hc.c -3 +3 M ./rdft/dht-rader.c -2 +2 M ./rdft/direct.c -3 +3 M ./rdft/direct2.c -3 +3 M ./rdft/generic.c -2 +2 M ./rdft/hc2hc-buf.c -3 +3 M ./rdft/hc2hc-dif.c -3 +3 M ./rdft/hc2hc-dit.c -3 +3 M ./rdft/hc2hc.c -3 +3 M ./rdft/hc2hc.h -3 +3 M ./rdft/indirect.c -3 +3 M ./rdft/khc2hc-dif.c -3 +3 M ./rdft/khc2hc-dit.c -3 +3 M ./rdft/khc2r.c -3 +3 M ./rdft/kr2hc.c -3 +3 M ./rdft/nop.c -3 +3 M ./rdft/nop2.c -3 +3 M ./rdft/plan.c -3 +3 M ./rdft/plan2.c -3 +3 M ./rdft/problem.c -3 +3 M ./rdft/problem2.c -3 +3 M ./rdft/rader-hc2hc.c -2 +2 M ./rdft/rank-geq2-rdft2.c -3 +3 M ./rdft/rank-geq2.c -3 +3 M ./rdft/rank0-rdft2.c -3 +3 M ./rdft/rank0.c -3 +3 M ./rdft/rdft-dht.c -3 +3 M ./rdft/rdft.h -3 +3 M ./rdft/rdft2-inplace-strides.c -3 +3 M ./rdft/rdft2-radix2.c -3 +3 M ./rdft/rdft2-tensor-max-index.c -3 +3 M ./rdft/solve.c -3 +3 M ./rdft/solve2.c -3 +3 M ./rdft/vrank-geq1-rdft2.c -3 +3 M ./rdft/vrank-geq1.c -3 +3 M ./rdft/vrank2-transpose.c -3 +3 M ./rdft/vrank3-transpose.c -3 +3 M ./reodft/conf.c -3 +3 M ./reodft/redft00e-r2hc.c -3 +3 M ./reodft/reodft.h -3 +3 M ./reodft/reodft010e-r2hc.c -3 +3 M ./reodft/reodft11e-r2hc-odd.c -3 +3 M ./reodft/reodft11e-r2hc.c -3 +3 M ./reodft/reodft11e-radix2.c -3 +3 M ./reodft/rodft00e-r2hc.c -3 +3 M ./simd/3dnow.c -3 +3 M ./simd/altivec.c -3 +3 M ./simd/simd-3dnow.h -2 +2 M ./simd/simd-altivec.h -2 +2 M ./simd/simd-sse.h -2 +2 M ./simd/simd-sse2.h -2 +2 M ./simd/simd.h -2 +2 M ./simd/sse-aux.c -3 +3 M ./simd/sse.c -3 +3 M ./simd/sse2-aux.c -3 +3 M ./simd/sse2.c -3 +3 M ./threads/api.c -2 +2 M ./threads/conf.c -3 +3 M ./threads/ct-dit.c -3 +3 M ./threads/dft-vrank-geq1.c -3 +3 M ./threads/f77api.c -2 +2 M ./threads/f77funcs.c -2 +2 M ./threads/hc2hc-dif.c -3 +3 M ./threads/hc2hc-dit.c -3 +3 M ./threads/rdft-vrank-geq1.c -3 +3 M ./threads/threads.c -2 +2 M ./threads/threads.h -3 +3 M ./threads/vrank-geq1-rdft2.c -3 +3 M ./tools/fftw-wisdom-to-conf.1 -4 +4 M ./tools/fftw-wisdom-to-conf.in -2 +2 M ./tools/fftw-wisdom.c -2 +2 M ./tools/fftw_wisdom.1.in -4 +4 Sat Mar 15 15:14:02 EST 2003 stevenj * [project @ 2003-03-15 20:14:02 by stevenj] threads in make check M ./TODO -4 M ./tests/Makefile.am -1 +5 M ./tests/check.pl +3 Sat Mar 15 15:11:24 EST 2003 stevenj * [project @ 2003-03-15 20:11:24 by stevenj] fixed const warnings M ./threads/ct-dit.c -2 +2 M ./threads/hc2hc-dif.c -2 +2 M ./threads/hc2hc-dit.c -2 +2 Sat Mar 15 15:08:25 EST 2003 stevenj * [project @ 2003-03-15 20:08:25 by stevenj] make sure spawn_loop size > 1 (it has to be at least > 0 lest we crash, but > 1 is an optimization) M ./threads/ct-dit.c -1 +2 M ./threads/hc2hc-dif.c -1 +2 M ./threads/hc2hc-dit.c -1 +2 Sat Mar 15 14:00:17 EST 2003 athena * [project @ 2003-03-15 19:00:17 by athena] hpux seems to want machine/sys/inline.h as opposed to machine/inline.h. M ./kernel/cycle.h -2 +2 Sat Mar 15 13:36:56 EST 2003 stevenj * [project @ 2003-03-15 18:36:56 by stevenj] Sourceforge is really SourceForge.net, and is run by VA M ./doc/fftw3.texi -3 +4 Sat Mar 15 13:34:05 EST 2003 stevenj * [project @ 2003-03-15 18:33:07 by stevenj] comma M ./doc/fftw3.texi -2 +2 Sat Mar 15 13:31:42 EST 2003 stevenj * [project @ 2003-03-15 18:31:42 by stevenj] fixed AMD company name M ./doc/fftw3.texi -2 +2 Sat Mar 15 13:29:41 EST 2003 stevenj * [project @ 2003-03-15 18:29:41 by stevenj] minor changes M ./doc/fftw3.texi -19 +19 Sat Mar 15 13:13:55 EST 2003 stevenj * [project @ 2003-03-15 18:13:55 by stevenj] more emitter->read_char renaming M ./api/f77api.c -5 +5 M ./api/f77funcs.c -4 +4 Sat Mar 15 13:08:45 EST 2003 stevenj * [project @ 2003-03-15 18:08:45 by stevenj] more wisdom docs, noted wisdom utilities M ./doc/fftw3.texi -35 +95 Sat Mar 15 11:41:32 EST 2003 stevenj * [project @ 2003-03-15 16:41:32 by stevenj] compound adjectives are hyphenated M ./doc/fftw3.texi -2 +2 Sat Mar 15 11:40:30 EST 2003 stevenj * [project @ 2003-03-15 16:40:30 by stevenj] fftw does support another type of packed array via r2r M ./doc/fftw3.texi -2 +2 Sat Mar 15 11:29:12 EST 2003 stevenj * [project @ 2003-03-15 16:29:12 by stevenj] write_char/read_char for export/import functions M ./api/export-wisdom.c -4 +4 M ./api/f77api.c -5 +5 M ./api/f77funcs.c -4 +4 M ./api/fftw3.h -3 +3 M ./api/import-wisdom.c -4 +4 M ./doc/f77_wisdom.f -6 +6 M ./doc/fftw3.texi -5 +5 Sat Mar 15 11:19:19 EST 2003 stevenj * [project @ 2003-03-15 16:19:19 by stevenj] comments M ./threads/threads.c -4 +8 Sat Mar 15 10:08:26 EST 2003 athena * [project @ 2003-03-15 15:08:26 by athena] Enabled randomized-cse M ./support/Makefile.codelets -1 +1 Sat Mar 15 09:47:49 EST 2003 athena * [project @ 2003-03-15 14:47:49 by athena] Changed to 3.0-beta1 M ./configure.ac -1 +1 Sat Mar 15 09:07:31 EST 2003 athena * [project @ 2003-03-15 14:07:31 by athena] First complete draft M ./doc/fftw3.texi -18 +131 Sat Mar 15 08:37:52 EST 2003 athena * [project @ 2003-03-15 13:37:52 by athena] EMITTER is a misnomer M ./api/fftw3.h -2 +2 M ./api/import-wisdom.c -8 +6 Sat Mar 15 05:50:50 EST 2003 athena * [project @ 2003-03-15 10:50:50 by athena] Revision, wisdom tutorial, acks. M ./doc/fftw3.texi -44 +219 Fri Mar 14 22:59:04 EST 2003 stevenj * [project @ 2003-03-15 03:59:04 by stevenj] noted OpenMP M ./NEWS +2 Fri Mar 14 22:38:49 EST 2003 stevenj * [project @ 2003-03-15 03:38:49 by stevenj] comment M ./threads/threads.c -1 +1 Fri Mar 14 22:38:30 EST 2003 stevenj * [project @ 2003-03-15 03:38:30 by stevenj] comments M ./threads/threads.c -2 +2 Fri Mar 14 22:38:05 EST 2003 stevenj * [project @ 2003-03-15 03:38:05 by stevenj] reformatting M ./threads/threads.c -3 +1 Fri Mar 14 22:26:28 EST 2003 stevenj * [project @ 2003-03-15 03:26:28 by stevenj] whoops M ./threads/threads.c -2 +2 Fri Mar 14 22:11:23 EST 2003 stevenj * [project @ 2003-03-15 03:11:23 by stevenj] some threads fixes, and added experimental semaphore (pre-thread-spawning) and Linux spinlock support M ./tests/bench.c -1 +10 M ./threads/api.c +1 M ./threads/threads.c -8 +159 M ./threads/threads.h -1 +2 Fri Mar 14 20:50:46 EST 2003 stevenj * [project @ 2003-03-15 01:50:46 by stevenj] whoops M ./threads/f77funcs.c -2 +2 Fri Mar 14 18:23:03 EST 2003 stevenj * [project @ 2003-03-14 23:23:03 by stevenj] added note that FFTW_PATIENT will disable threads if they are not beneficial M ./doc/fftw3.texi -2 +5 Fri Mar 14 18:20:44 EST 2003 stevenj * [project @ 2003-03-14 23:20:44 by stevenj] made fftw_cleanup* more restrictive, in that we don't want to guarantee that previously created plans will still work (they won't, in the case of threaded plans and fftw_cleanup_threads), and there is no reason to provide such a guarantee anyway. M ./doc/fftw3.texi -6 +14 Fri Mar 14 17:23:13 EST 2003 athena * [project @ 2003-03-14 22:23:13 by athena] Moved version.c from kernel/ into api/ A ./api/version.c M ./api/Makefile.am -1 +2 M ./api/version.c +28 M ./kernel/Makefile.am -1 +1 M ./kernel/ifftw.h -4 +1 R ./kernel/version.c Fri Mar 14 17:19:50 EST 2003 athena * [project @ 2003-03-14 22:19:50 by athena] icc-7.0 requires -openmp M ./configure.ac +5 Fri Mar 14 14:47:52 EST 2003 athena * [project @ 2003-03-14 19:47:52 by athena] Ensure that one can do make dist given the distribution M ./doc/Makefile.am -5 +5 Fri Mar 14 14:38:11 EST 2003 athena * [project @ 2003-03-14 19:38:11 by athena] Dist fftw3.pdf, not fftw.pdf M ./doc/Makefile.am -1 +2 Fri Mar 14 14:36:25 EST 2003 athena * [project @ 2003-03-14 19:36:25 by athena] Support -onthreads=%d M ./tests/bench.c -4 +8 Fri Mar 14 14:34:21 EST 2003 stevenj * [project @ 2003-03-14 19:34:21 by stevenj] comment M ./kernel/alloc.c -2 +2 Fri Mar 14 14:33:27 EST 2003 stevenj * [project @ 2003-03-14 19:33:27 by stevenj] whoops M ./threads/Makefile.am -1 +3 Fri Mar 14 12:32:18 EST 2003 stevenj * [project @ 2003-03-14 17:32:18 by stevenj] fftw_real is gone M ./doc/rfftwnd.fig -1 +1 Fri Mar 14 12:26:04 EST 2003 stevenj * [project @ 2003-03-14 17:26:04 by stevenj] typos M ./doc/fftw3.texi -3 +3 Fri Mar 14 06:21:43 EST 2003 athena * [project @ 2003-03-14 11:21:43 by athena] More BENCH_DOC strings M ./api/fftw3.h -3 +8 M ./tests/bench.c -4 +11 Fri Mar 14 05:58:53 EST 2003 athena * [project @ 2003-03-14 10:58:53 by athena] Fixed xref's M ./doc/fftw3.texi -15 +15 Fri Mar 14 05:38:26 EST 2003 athena * [project @ 2003-03-14 10:38:26 by athena] Revised manual (esp. intro and tutorial), fixed texinfo hackery for figures. M ./doc/Makefile.am -8 +13 M ./doc/fftw3.texi -165 +280 R ./doc/rfftwnd.gif Wed Mar 12 02:42:33 EST 2003 stevenj * [project @ 2003-03-12 07:42:33 by stevenj] redirect users from guru execute to advanced interface, if possible M ./doc/fftw3.texi -1 +5 Wed Mar 12 02:35:22 EST 2003 stevenj * [project @ 2003-03-12 07:35:22 by stevenj] punctuation M ./doc/fftw3.texi -2 +2 Wed Mar 12 02:28:51 EST 2003 stevenj * [project @ 2003-03-12 07:28:51 by stevenj] use correct heading level M ./doc/fftw3.texi -7 +7 Wed Mar 12 02:24:37 EST 2003 stevenj * [project @ 2003-03-12 07:24:37 by stevenj] html generation M ./doc/Makefile.am +15 M ./doc/fftw3.texi -91 +90 Wed Mar 12 01:44:00 EST 2003 stevenj * [project @ 2003-03-12 06:44:00 by stevenj] added equation GIFs A ./doc/equation-dft.gif A ./doc/equation-dht.gif A ./doc/equation-idft.gif A ./doc/equation-redft00.gif A ./doc/equation-redft01.gif A ./doc/equation-redft10.gif A ./doc/equation-redft11.gif A ./doc/equation-rodft00.gif A ./doc/equation-rodft01.gif A ./doc/equation-rodft10.gif A ./doc/equation-rodft11.gif M ./doc/equation-dft.gif M ./doc/equation-dht.gif M ./doc/equation-idft.gif M ./doc/equation-redft00.gif M ./doc/equation-redft01.gif M ./doc/equation-redft10.gif M ./doc/equation-redft11.gif M ./doc/equation-rodft00.gif M ./doc/equation-rodft01.gif M ./doc/equation-rodft10.gif M ./doc/equation-rodft11.gif Wed Mar 12 01:43:27 EST 2003 stevenj * [project @ 2003-03-12 06:43:27 by stevenj] punctuation M ./doc/fftw3.texi -18 +18 Wed Mar 12 01:26:46 EST 2003 stevenj * [project @ 2003-03-12 06:26:46 by stevenj] punctuation M ./doc/fftw3.texi -14 +14 Wed Mar 12 01:25:12 EST 2003 stevenj * [project @ 2003-03-12 06:25:12 by stevenj] added multi-dimensional transform definitions M ./doc/fftw3.texi -3 +124 Wed Mar 12 00:14:03 EST 2003 stevenj * [project @ 2003-03-12 05:14:03 by stevenj] slight changes M ./doc/fftw3.texi -3 +5 Wed Mar 12 00:06:34 EST 2003 stevenj * [project @ 2003-03-12 05:06:34 by stevenj] typo M ./doc/fftw3.texi -2 +2 Tue Mar 11 23:50:43 EST 2003 stevenj * [project @ 2003-03-12 04:50:43 by stevenj] added 1d version of What FFTW Really Computes M ./doc/fftw3.texi -1 +432 Tue Mar 11 21:17:54 EST 2003 stevenj * [project @ 2003-03-12 02:17:54 by stevenj] note in upgrading section about FFTW_PATIENT M ./doc/fftw3.texi -1 +7 Tue Mar 11 15:18:39 EST 2003 stevenj * [project @ 2003-03-11 20:18:39 by stevenj] added cycle-counter section M ./doc/fftw3.texi -1 +32 Tue Mar 11 14:53:44 EST 2003 stevenj * [project @ 2003-03-11 19:53:44 by stevenj] more ideas M ./TODO +7 Mon Mar 10 17:41:35 EST 2003 stevenj * [project @ 2003-03-10 22:41:35 by stevenj] noted that indirect should probably be merged with rank-geq2, to make a rank-split solver M ./dft/indirect.c -1 +4 M ./rdft/indirect.c -1 +4 Fri Mar 7 03:01:52 EST 2003 stevenj * [project @ 2003-03-07 08:01:52 by stevenj] added non-Unix installation instructions M ./doc/fftw3.texi -1 +38 Fri Mar 7 02:30:59 EST 2003 stevenj * [project @ 2003-03-07 07:30:59 by stevenj] also talk about stack alignment with SSE/SSE2 M ./doc/fftw3.texi -1 +7 Fri Mar 7 02:24:07 EST 2003 stevenj * [project @ 2003-03-07 07:24:07 by stevenj] made warning more dire M ./doc/fftw3.texi -3 +4 Fri Mar 7 02:13:25 EST 2003 stevenj * [project @ 2003-03-07 07:13:25 by stevenj] fix M ./doc/fftw3.texi -2 +2 Fri Mar 7 02:09:55 EST 2003 stevenj * [project @ 2003-03-07 07:09:55 by stevenj] number M ./doc/fftw3.texi -2 +2 Fri Mar 7 02:09:08 EST 2003 stevenj * [project @ 2003-03-07 07:09:08 by stevenj] fix M ./doc/fftw3.texi -3 +3 Fri Mar 7 02:08:01 EST 2003 stevenj * [project @ 2003-03-07 07:08:01 by stevenj] minor M ./doc/fftw3.texi -2 +2 Fri Mar 7 02:04:45 EST 2003 stevenj * [project @ 2003-03-07 07:04:45 by stevenj] minor fix M ./doc/fftw3.texi -3 +3 Fri Mar 7 01:58:15 EST 2003 stevenj * [project @ 2003-03-07 06:58:15 by stevenj] cross-ref M ./doc/fftw3.texi -2 +3 Fri Mar 7 01:57:31 EST 2003 stevenj * [project @ 2003-03-07 06:57:31 by stevenj] minor M ./doc/fftw3.texi -7 +8 Fri Mar 7 01:53:28 EST 2003 stevenj * [project @ 2003-03-07 06:53:28 by stevenj] more installation manual M ./doc/fftw3.texi -7 +151 Fri Mar 7 00:43:40 EST 2003 stevenj * [project @ 2003-03-07 05:43:40 by stevenj] GNU-lly correct M ./doc/fftw3.texi -2 +2 Fri Mar 7 00:38:48 EST 2003 stevenj * [project @ 2003-03-07 05:38:48 by stevenj] started installation section M ./doc/fftw3.texi -3 +33 Fri Mar 7 00:25:02 EST 2003 stevenj * [project @ 2003-03-07 05:25:02 by stevenj] added --without-cycle-counter option as a last resort M ./configure.ac +5 M ./kernel/timer.c -11 +24 Fri Mar 7 00:07:12 EST 2003 stevenj * [project @ 2003-03-07 05:07:12 by stevenj] macros with () arguments were only standardized in C99, and we don't need them anyway M ./kernel/cycle.h -3 +3 Thu Mar 6 23:10:41 EST 2003 stevenj * [project @ 2003-03-07 04:10:41 by stevenj] wording M ./doc/fftw3.texi -2 +2 Thu Mar 6 23:03:03 EST 2003 stevenj * [project @ 2003-03-07 04:03:03 by stevenj] parallelism M ./doc/fftw3.texi -2 +2 Thu Mar 6 23:01:47 EST 2003 stevenj * [project @ 2003-03-07 04:01:47 by stevenj] additions to upgrading chapter M ./doc/fftw3.texi -1 +31 Thu Mar 6 22:39:36 EST 2003 stevenj * [project @ 2003-03-07 03:39:36 by stevenj] noted additional humility of FFTW 3 wisdom M ./doc/fftw3.texi -1 +9 Thu Mar 6 22:32:44 EST 2003 stevenj * [project @ 2003-03-07 03:32:44 by stevenj] renaming M ./doc/fftw3.texi -6 +6 Thu Mar 6 22:31:00 EST 2003 stevenj * [project @ 2003-03-07 03:31:00 by stevenj] added placeholder for wisdom reference M ./doc/fftw3.texi -3 +10 Thu Mar 6 22:29:38 EST 2003 stevenj * [project @ 2003-03-07 03:29:38 by stevenj] wrote upgrading chapter M ./doc/fftw3.texi -1 +139 Thu Mar 6 18:01:10 EST 2003 stevenj * [project @ 2003-03-06 23:01:10 by stevenj] slight change M ./doc/fftw3.texi -6 +6 Thu Mar 6 18:00:43 EST 2003 stevenj * [project @ 2003-03-06 23:00:43 by stevenj] placeholder for upgrade chapter M ./doc/fftw3.texi -3 +18 Thu Mar 6 13:47:49 EST 2003 stevenj * [project @ 2003-03-06 18:47:49 by stevenj] whoops M ./tools/fftw-wisdom.c -2 +2 Thu Mar 6 13:36:38 EST 2003 stevenj * [project @ 2003-03-06 18:36:38 by stevenj] strengthed warning about time M ./tools/fftw_wisdom.1.in -1 +1 Thu Mar 6 13:35:42 EST 2003 stevenj * [project @ 2003-03-06 18:35:42 by stevenj] noted -t in example M ./tools/fftw_wisdom.1.in -1 +3 Thu Mar 6 13:21:03 EST 2003 stevenj * [project @ 2003-03-06 18:21:03 by stevenj] pay attention to WINDOWS_F77_MANGLING M ./threads/f77api.c -3 +26 Thu Mar 6 02:52:30 EST 2003 stevenj * [project @ 2003-03-06 07:52:30 by stevenj] punctuation M ./doc/fftw3.texi -2 +2 Thu Mar 6 02:51:02 EST 2003 stevenj * [project @ 2003-03-06 07:51:02 by stevenj] index M ./doc/fftw3.texi -1 +3 Thu Mar 6 02:50:38 EST 2003 stevenj * [project @ 2003-03-06 07:50:38 by stevenj] documented C++ usage M ./doc/fftw3.texi -3 +18 Thu Mar 6 02:25:32 EST 2003 stevenj * [project @ 2003-03-06 07:25:32 by stevenj] got rid of overfull hbox TeX warnings M ./doc/fftw3.texi -6 +6 Thu Mar 6 02:20:38 EST 2003 stevenj * [project @ 2003-03-06 07:20:38 by stevenj] whoops M ./doc/fftw3.texi -2 +2 Thu Mar 6 02:20:13 EST 2003 stevenj * [project @ 2003-03-06 07:20:13 by stevenj] noted fftw_iodim split for Fortran guru interface M ./doc/fftw3.texi -5 +16 Thu Mar 6 02:14:21 EST 2003 stevenj * [project @ 2003-03-06 07:14:21 by stevenj] added guru reference M ./doc/fftw3.texi -3 +187 Wed Mar 5 22:56:05 EST 2003 stevenj * [project @ 2003-03-06 03:56:05 by stevenj] minor M ./doc/fftw3.texi -3 +3 Wed Mar 5 22:45:31 EST 2003 stevenj * [project @ 2003-03-06 03:45:31 by stevenj] use @r{...} for comment text in code examples M ./doc/fftw3.texi -3 +3 Wed Mar 5 13:14:04 EST 2003 stevenj * [project @ 2003-03-05 18:14:04 by stevenj] eliminate warning M ./simd/sse.c -1 +2 Wed Mar 5 13:12:56 EST 2003 stevenj * [project @ 2003-03-05 18:12:56 by stevenj] SIMD_CFLAGS only for simd code M ./configure.ac -5 +8 M ./dft/simd/Makefile.am -1 +1 M ./dft/simd/codelets/Makefile.am -1 +1 M ./kernel/align.c -12 +9 M ./simd/Makefile.am +1 Wed Mar 5 11:06:41 EST 2003 athena * [project @ 2003-03-05 16:06:41 by athena] Minor changes. M ./doc/fftw3.texi -89 +96 Wed Mar 5 02:13:34 EST 2003 stevenj * [project @ 2003-03-05 07:13:34 by stevenj] cross-compiling with MinGW can't detect f77 mangling, so add an option to use what seems to be the most common styles M ./api/f77api.c -3 +34 M ./configure.ac +5 Tue Mar 4 20:00:31 EST 2003 stevenj * [project @ 2003-03-05 01:00:31 by stevenj] comment M ./libbench2/util.c -2 +3 Tue Mar 4 20:00:13 EST 2003 stevenj * [project @ 2003-03-05 01:00:13 by stevenj] we only use our-malloc-16 on machines where size_t == uintptr_t, so don't bother doing the right thing with the benchmark M ./libbench2/util.c -1 +1 Tue Mar 4 19:46:09 EST 2003 stevenj * [project @ 2003-03-05 00:46:09 by stevenj] support WITH_OUR_MALLOC16 M ./libbench2/util.c -2 +19 Tue Mar 4 18:50:53 EST 2003 fftw * [project @ 2003-03-04 23:50:53 by fftw] automatically add -msse etcetera for --enable-sse etcetera M ./configure.ac +22 Tue Mar 4 18:24:26 EST 2003 fftw * [project @ 2003-03-04 23:24:26 by fftw] got rid of const warning M ./tools/fftw-wisdom.c -2 +2 Tue Mar 4 18:22:48 EST 2003 fftw * [project @ 2003-03-04 23:22:48 by fftw] missing header M ./libbench2/problem.c -1 +2 Tue Mar 4 15:55:47 EST 2003 stevenj * [project @ 2003-03-04 20:55:47 by stevenj] fixes M ./doc/fftw3.texi -15 +49 Tue Mar 4 15:53:26 EST 2003 stevenj * [project @ 2003-03-04 20:53:26 by stevenj] whoops M ./api/import-system-wisdom.c -1 +1 Tue Mar 4 02:22:14 EST 2003 stevenj * [project @ 2003-03-04 07:22:14 by stevenj] started guru reference M ./doc/fftw3.texi -1 +110 Tue Mar 4 01:44:09 EST 2003 stevenj * [project @ 2003-03-04 06:44:09 by stevenj] use same FFTW_IODIM between precisions M ./api/fftw3.h -6 +8 Tue Mar 4 00:25:57 EST 2003 stevenj * [project @ 2003-03-04 05:25:57 by stevenj] renamed section M ./doc/fftw3.texi -5 +5 Tue Mar 4 00:21:49 EST 2003 stevenj * [project @ 2003-03-04 05:21:49 by stevenj] no need for "advanced" in subheadings M ./doc/fftw3.texi -4 +4 Tue Mar 4 00:20:05 EST 2003 stevenj * [project @ 2003-03-04 05:20:05 by stevenj] typo M ./doc/fftw3.texi -2 +2 Tue Mar 4 00:17:23 EST 2003 stevenj * [project @ 2003-03-04 05:17:23 by stevenj] finished advanced interface M ./doc/fftw3.texi -1 +34 Mon Mar 3 23:26:12 EST 2003 stevenj * [project @ 2003-03-04 04:26:12 by stevenj] more advance interface docs M ./doc/fftw3.texi -1 +46 Mon Mar 3 23:12:09 EST 2003 stevenj * [project @ 2003-03-04 04:12:09 by stevenj] fail for win32 M ./api/import-system-wisdom.c -2 +2 Mon Mar 3 17:18:48 EST 2003 fftw * [project @ 2003-03-03 22:18:48 by fftw] shortened help string M ./configure.ac -1 +1 Mon Mar 3 17:16:17 EST 2003 fftw * [project @ 2003-03-03 22:16:17 by fftw] fixed cross-refs M ./doc/fftw3.texi -3 +3 Mon Mar 3 17:07:27 EST 2003 fftw * [project @ 2003-03-03 22:07:27 by fftw] FFTW_POSSIBLY_UNALIGNED -> simpler FFTW_UNALIGNED in API, added bench option M ./api/fftw3.h -2 +2 M ./api/mapflags.c -1 +1 M ./doc/fftw3.texi -3 +3 M ./tests/bench.c +1 Mon Mar 3 16:58:07 EST 2003 fftw * [project @ 2003-03-03 21:58:07 by fftw] whoops M ./kernel/alloc.c -3 +3 Mon Mar 3 16:52:58 EST 2003 fftw * [project @ 2003-03-03 21:52:58 by fftw] noted assumption M ./kernel/alloc.c -2 +2 Mon Mar 3 16:50:33 EST 2003 fftw * [project @ 2003-03-03 21:50:33 by fftw] provide our own malloc16 routine because of Windows lossage M ./configure.ac +5 M ./kernel/alloc.c -3 +35 Mon Mar 3 13:28:12 EST 2003 stevenj * [project @ 2003-03-03 18:28:12 by stevenj] capitalization M ./doc/fftw3.texi -6 +6 Mon Mar 3 13:26:32 EST 2003 stevenj * [project @ 2003-03-03 18:26:32 by stevenj] whoops M ./doc/fftw3.texi -2 +2 Mon Mar 3 12:55:57 EST 2003 stevenj * [project @ 2003-03-03 17:55:57 by stevenj] vertical skip looks better than indenting for setting off short paragraphs M ./doc/fftw3.texi -1 +6 Mon Mar 3 06:34:09 EST 2003 athena * [project @ 2003-03-03 11:34:09 by athena] Removed franz-mode. Automake was distributing franz files whether franz mode was enabled or not. M ./configure.ac -3 M ./dft/simd/codelets/Makefile.am -7 +2 Mon Mar 3 01:44:00 EST 2003 stevenj * [project @ 2003-03-03 06:44:00 by stevenj] made output boundary conditions more prominent; they are important, because they make the different transform types inequivalent in parity M ./doc/fftw3.texi -4 +8 Mon Mar 3 01:17:28 EST 2003 stevenj * [project @ 2003-03-03 06:17:28 by stevenj] clarification M ./doc/fftw3.texi -2 +2 Mon Mar 3 01:17:07 EST 2003 stevenj * [project @ 2003-03-03 06:17:07 by stevenj] typo M ./doc/fftw3.texi -2 +2 Mon Mar 3 01:10:28 EST 2003 stevenj * [project @ 2003-03-03 06:10:28 by stevenj] started advanced reference M ./doc/fftw3.texi -6 +51 Mon Mar 3 00:52:02 EST 2003 stevenj * [project @ 2003-03-03 05:52:02 by stevenj] r2r reference M ./doc/fftw3.texi -2 +209 Sun Mar 2 23:51:21 EST 2003 stevenj * [project @ 2003-03-03 04:51:21 by stevenj] workaround for info formatting bug M ./doc/fftw3.texi -5 +4 Sun Mar 2 23:47:19 EST 2003 stevenj * [project @ 2003-03-03 04:47:19 by stevenj] noted lack of fftw_malloc in Fortran M ./doc/fftw3.texi -3 +9 Sun Mar 2 23:42:52 EST 2003 stevenj * [project @ 2003-03-03 04:42:32 by stevenj] parallelism M ./doc/fftw3.texi -4 +4 Sun Mar 2 23:39:54 EST 2003 stevenj * [project @ 2003-03-03 04:39:05 by stevenj] whoops M ./doc/fftw3.texi -7 +7 Sun Mar 2 23:33:02 EST 2003 stevenj * [project @ 2003-03-03 04:33:02 by stevenj] r2c/c2r reference M ./doc/fftw3.texi -40 +248 Sun Mar 2 22:44:10 EST 2003 stevenj * [project @ 2003-03-03 03:44:10 by stevenj] table of contents was being included twice M ./doc/fftw3.texi -2 +1 Sun Mar 2 22:42:29 EST 2003 stevenj * [project @ 2003-03-03 03:42:29 by stevenj] minor changes M ./doc/fftw3.texi -15 +16 Sun Mar 2 21:54:13 EST 2003 stevenj * [project @ 2003-03-03 02:54:13 by stevenj] started reference section M ./doc/fftw3.texi -2 +238 Sun Mar 2 19:10:02 EST 2003 stevenj * [project @ 2003-03-03 00:10:02 by stevenj] whoops M ./doc/Makefile.am -2 +2 Sun Mar 2 19:03:23 EST 2003 stevenj * [project @ 2003-03-03 00:03:23 by stevenj] started ref. section M ./doc/fftw3.texi -2 +146 Sun Mar 2 18:50:58 EST 2003 stevenj * [project @ 2003-03-02 23:50:58 by stevenj] fftw_flops takes const plan M ./api/fftw3.h -2 +2 M ./api/flops.c -1 +1 Sun Mar 2 15:54:14 EST 2003 stevenj * [project @ 2003-03-02 20:54:14 by stevenj] typo M ./doc/fftw3.texi -2 +2 Sun Mar 2 15:52:41 EST 2003 stevenj * [project @ 2003-03-02 20:52:41 by stevenj] added "Wisdom of Fortran?" section M ./doc/fftw3.texi -1 +39 Sun Mar 2 15:50:37 EST 2003 stevenj * [project @ 2003-03-02 20:50:37 by stevenj] typo M ./doc/f77_wisdom.f -2 +2 Sun Mar 2 15:49:57 EST 2003 stevenj * [project @ 2003-03-02 20:49:57 by stevenj] wording M ./doc/f77_wisdom.f -1 +1 Sun Mar 2 15:46:13 EST 2003 stevenj * [project @ 2003-03-02 20:46:13 by stevenj] added comments M ./doc/f77_wisdom.f +25 Sun Mar 2 15:44:01 EST 2003 stevenj * [project @ 2003-03-02 20:44:01 by stevenj] added example file A ./doc/f77_wisdom.f Sun Mar 2 15:37:32 EST 2003 stevenj * [project @ 2003-03-02 20:37:32 by stevenj] don't print out READ WISDOM unless we have M ./tests/bench.c -6 +11 Sun Mar 2 15:36:28 EST 2003 stevenj * [project @ 2003-03-02 20:36:28 by stevenj] EOF is not a space M ./kernel/scan.c -2 +2 Sun Mar 2 09:14:37 EST 2003 athena * [project @ 2003-03-02 14:14:37 by athena] Turn on inline by default M ./kernel/ifftw.h -2 +3 Sun Mar 2 07:11:56 EST 2003 athena * [project @ 2003-03-02 12:11:56 by athena] Optionally inline loop in notw codelets M ./genfft/gen_hc2r.ml -3 +3 M ./genfft/gen_notw.ml -3 +3 M ./genfft/gen_notw_c.ml -3 +3 M ./genfft/gen_r2hc.ml -3 +3 M ./kernel/ifftw.h -1 +3 Sun Mar 2 01:37:41 EST 2003 stevenj * [project @ 2003-03-02 06:37:41 by stevenj] updated nodes M ./doc/fftw3.texi -5 +9 Sun Mar 2 01:37:19 EST 2003 stevenj * [project @ 2003-03-02 06:37:19 by stevenj] wrote most of Fortran chapter M ./doc/fftw3.texi -1 +208 Sun Mar 2 00:58:37 EST 2003 stevenj * [project @ 2003-03-02 05:58:37 by stevenj] citation M ./doc/fftw3.texi -2 +2 Sun Mar 2 00:57:22 EST 2003 stevenj * [project @ 2003-03-02 05:57:22 by stevenj] added parallel FFTW chapter M ./doc/fftw3.texi -3 +209 Sat Mar 1 20:42:23 EST 2003 stevenj * [project @ 2003-03-02 01:42:23 by stevenj] typo M ./doc/fftw3.texi -2 +2 Sat Mar 1 20:34:38 EST 2003 stevenj * [project @ 2003-03-02 01:34:38 by stevenj] added inlining to TODO M ./TODO +2 Sat Mar 1 19:36:26 EST 2003 stevenj * [project @ 2003-03-02 00:36:26 by stevenj] added K M ./CONVENTIONS -1 +2 Sat Mar 1 19:15:18 EST 2003 stevenj * [project @ 2003-03-02 00:15:18 by stevenj] use K for constants M ./dft/zero.c -3 +3 M ./kernel/trig1.c -5 +5 M ./rdft/generic.c -8 +8 M ./rdft/problem.c -3 +3 M ./rdft/rdft-dht.c -3 +3 M ./rdft/rdft2-radix2.c -11 +11 M ./reodft/redft00e-r2hc.c -3 +3 M ./reodft/reodft010e-r2hc.c -11 +11 M ./reodft/reodft11e-r2hc.c -7 +7 M ./reodft/reodft11e-radix2.c -11 +11 M ./reodft/rodft00e-r2hc.c -3 +3 Sat Mar 1 19:14:54 EST 2003 stevenj * [project @ 2003-03-02 00:14:54 by stevenj] fixed cross-ref M ./doc/fftw3.texi -2 +2 Sat Mar 1 19:14:16 EST 2003 stevenj * [project @ 2003-03-02 00:14:16 by stevenj] whoops M ./doc/fftw3.texi -6 +6 Sat Mar 1 18:50:43 EST 2003 stevenj * [project @ 2003-03-01 23:50:43 by stevenj] cleanup M ./doc/fftw3.texi -8 +8 Sat Mar 1 18:46:38 EST 2003 stevenj * [project @ 2003-03-01 23:46:38 by stevenj] "words of wisdom" by itself is a little too obscure M ./doc/fftw3.texi -6 +6 Sat Mar 1 18:43:21 EST 2003 stevenj * [project @ 2003-03-01 23:43:21 by stevenj] re-added multi-dimensional array stuff M ./doc/fftw3.texi -1 +203 Sat Mar 1 18:15:22 EST 2003 stevenj * [project @ 2003-03-01 23:15:22 by stevenj] added alignment section M ./doc/fftw3.texi -4 +93 Sat Mar 1 16:34:21 EST 2003 stevenj * [project @ 2003-03-01 21:34:21 by stevenj] shrunk code M ./reodft/reodft11e-r2hc-odd.c -147 +57 Fri Feb 28 20:22:00 EST 2003 stevenj * [project @ 2003-03-01 01:22:00 by stevenj] slight compression M ./reodft/reodft11e-r2hc-odd.c -146 +122 Fri Feb 28 19:01:20 EST 2003 stevenj * [project @ 2003-03-01 00:01:20 by stevenj] style M ./doc/fftw3.texi -7 +7 M ./reodft/reodft11e-radix2.c -2 +2 Fri Feb 28 18:46:53 EST 2003 stevenj * [project @ 2003-02-28 23:46:53 by stevenj] noted not in API M ./CONVENTIONS -1 +1 Fri Feb 28 18:43:14 EST 2003 stevenj * [project @ 2003-02-28 23:43:14 by stevenj] more updates M ./CONVENTIONS -3 +4 Fri Feb 28 18:38:42 EST 2003 stevenj * [project @ 2003-02-28 23:38:42 by stevenj] slight updates M ./CONVENTIONS -3 +4 Fri Feb 28 18:28:58 EST 2003 stevenj * [project @ 2003-02-28 23:28:58 by stevenj] great const-ification of apply/solve and print M ./api/f77funcs.c -1 +1 M ./api/fftw3.h -2 +2 M ./api/print-plan.c -1 +1 M ./dft/buffered.c -5 +5 M ./dft/ct-dif.c -3 +3 M ./dft/ct-dit.c -3 +3 M ./dft/ct-ditbuf.c -3 +3 M ./dft/ct-ditf.c -3 +3 M ./dft/ct.c -3 +3 M ./dft/dft.h -3 +3 M ./dft/direct.c -5 +5 M ./dft/generic.c -4 +4 M ./dft/indirect.c -7 +7 M ./dft/nop.c -3 +3 M ./dft/rader.c -12 +12 M ./dft/rank-geq2.c -5 +5 M ./dft/rank0.c -12 +12 M ./dft/solve.c -3 +3 M ./dft/vrank-geq1.c -5 +5 M ./dft/vrank2-transpose.c -5 +5 M ./dft/vrank3-transpose.c -5 +5 M ./kernel/ifftw.h -3 +3 M ./rdft/buffered.c -5 +5 M ./rdft/buffered2.c -7 +7 M ./rdft/dft-r2hc.c -5 +5 M ./rdft/dht-r2hc.c -5 +5 M ./rdft/dht-rader.c -4 +4 M ./rdft/direct.c -7 +7 M ./rdft/direct2.c -7 +7 M ./rdft/generic.c -6 +6 M ./rdft/hc2hc-buf.c -5 +5 M ./rdft/hc2hc-dif.c -3 +3 M ./rdft/hc2hc-dit.c -3 +3 M ./rdft/hc2hc.c -3 +3 M ./rdft/indirect.c -7 +7 M ./rdft/nop.c -3 +3 M ./rdft/nop2.c -3 +3 M ./rdft/rader-hc2hc.c -6 +6 M ./rdft/rank-geq2-rdft2.c -7 +7 M ./rdft/rank-geq2.c -5 +5 M ./rdft/rank0-rdft2.c -9 +9 M ./rdft/rank0.c -8 +8 M ./rdft/rdft-dht.c -9 +9 M ./rdft/rdft.h -5 +5 M ./rdft/rdft2-radix2.c -12 +12 M ./rdft/solve.c -3 +3 M ./rdft/solve2.c -3 +3 M ./rdft/vrank-geq1-rdft2.c -7 +7 M ./rdft/vrank-geq1.c -5 +5 M ./rdft/vrank2-transpose.c -5 +5 M ./rdft/vrank3-transpose.c -5 +5 M ./reodft/redft00e-r2hc.c -5 +5 M ./reodft/reodft010e-r2hc.c -11 +11 M ./reodft/reodft11e-r2hc-odd.c -7 +7 M ./reodft/reodft11e-r2hc.c -7 +7 M ./reodft/reodft11e-radix2.c -9 +9 M ./reodft/rodft00e-r2hc.c -5 +5 M ./threads/ct-dit.c -3 +3 M ./threads/dft-vrank-geq1.c -5 +5 M ./threads/hc2hc-dif.c -3 +3 M ./threads/hc2hc-dit.c -3 +3 M ./threads/rdft-vrank-geq1.c -5 +5 M ./threads/vrank-geq1-rdft2.c -7 +7 Fri Feb 28 17:51:15 EST 2003 stevenj * [project @ 2003-02-28 22:51:15 by stevenj] make fftw_execute take a const plan, to remind the user that it is re-entrant (or should be)... M ./api/execute-dft-c2r.c -1 +1 M ./api/execute-dft-r2c.c -1 +1 M ./api/execute-dft.c -1 +1 M ./api/execute-r2r.c -1 +1 M ./api/execute.c -1 +1 M ./api/f77funcs.c -5 +6 M ./api/fftw3.h -6 +6 M ./doc/fftw3.texi -2 +2 Fri Feb 28 17:29:40 EST 2003 stevenj * [project @ 2003-02-28 22:29:40 by stevenj] weakening M ./doc/fftw3.texi -2 +2 Fri Feb 28 17:28:48 EST 2003 stevenj * [project @ 2003-02-28 22:28:48 by stevenj] note M ./doc/fftw3.texi -4 +4 Fri Feb 28 17:27:10 EST 2003 stevenj * [project @ 2003-02-28 22:27:10 by stevenj] footnote about why DHT is provided M ./doc/fftw3.texi -7 +12 Fri Feb 28 15:07:03 EST 2003 stevenj * [project @ 2003-02-28 20:07:03 by stevenj] index M ./doc/fftw3.texi -1 +2 Fri Feb 28 15:05:48 EST 2003 stevenj * [project @ 2003-02-28 20:05:48 by stevenj] added DHT tutorial M ./doc/fftw3.texi -2 +32 Fri Feb 28 14:36:45 EST 2003 stevenj * [project @ 2003-02-28 19:36:45 by stevenj] fixed O(n log n) M ./doc/fftw3.texi -4 +16 Fri Feb 28 14:12:15 EST 2003 stevenj * [project @ 2003-02-28 19:08:03 by stevenj] whoops M ./doc/fftw3.texi -4 +10 Fri Feb 28 14:06:22 EST 2003 stevenj * [project @ 2003-02-28 19:06:22 by stevenj] slight improvements M ./doc/fftw3.texi -2 +12 Fri Feb 28 00:55:50 EST 2003 stevenj * [project @ 2003-02-28 05:55:50 by stevenj] addition M ./doc/fftw3.texi -3 +5 Fri Feb 28 00:54:09 EST 2003 stevenj * [project @ 2003-02-28 05:54:09 by stevenj] clarification M ./doc/fftw3.texi -7 +8 Thu Feb 27 23:49:37 EST 2003 stevenj * [project @ 2003-02-28 04:49:37 by stevenj] fix M ./doc/fftw3.texi -2 +2 Thu Feb 27 23:43:56 EST 2003 stevenj * [project @ 2003-02-28 04:43:56 by stevenj] slight changes M ./doc/fftw3.texi -15 +21 Thu Feb 27 23:27:48 EST 2003 stevenj * [project @ 2003-02-28 04:27:48 by stevenj] added R{E,O}DFTab tutorial M ./doc/fftw3.texi -9 +105 Thu Feb 27 17:24:20 EST 2003 stevenj * [project @ 2003-02-27 22:24:20 by stevenj] fixes M ./doc/fftw3.texi -12 +11 Thu Feb 27 17:20:42 EST 2003 stevenj * [project @ 2003-02-27 22:20:42 by stevenj] fixes M ./doc/fftw3.texi -24 +26 Thu Feb 27 17:11:54 EST 2003 stevenj * [project @ 2003-02-27 22:11:22 by stevenj] slight change M ./doc/fftw3.texi -4 +4 Thu Feb 27 17:07:45 EST 2003 stevenj * [project @ 2003-02-27 22:07:45 by stevenj] documented r2hc/hc2r M ./doc/fftw3.texi -6 +86 Thu Feb 27 16:19:16 EST 2003 stevenj * [project @ 2003-02-27 21:19:16 by stevenj] minor changes M ./doc/fftw3.texi -24 +26 Thu Feb 27 13:54:06 EST 2003 stevenj * [project @ 2003-02-27 18:54:06 by stevenj] timed planner and unifying radix-2 butterfly loops are not critical for release M ./TODO -8 +8 Thu Feb 27 13:51:20 EST 2003 stevenj * [project @ 2003-02-27 18:51:20 by stevenj] reodft/verify.c no longer exists M ./TODO -2 Thu Feb 27 13:44:19 EST 2003 stevenj * [project @ 2003-02-27 18:44:19 by stevenj] optimization: REDFT00 of size 2 is same as R2HC M ./rdft/problem.c -1 +5 Thu Feb 27 12:35:33 EST 2003 stevenj * [project @ 2003-02-27 17:35:33 by stevenj] R{E,O}DFT01 of size-1 is identity M ./rdft/problem.c -3 +4 Thu Feb 27 12:15:10 EST 2003 stevenj * [project @ 2003-02-27 17:15:10 by stevenj] minor simplification M ./reodft/reodft11e-r2hc-odd.c -31 +31 Thu Feb 27 02:46:31 EST 2003 stevenj * [project @ 2003-02-27 07:46:31 by stevenj] fixed add count M ./reodft/reodft11e-r2hc-odd.c -2 +2 Thu Feb 27 02:25:04 EST 2003 stevenj * [project @ 2003-02-27 07:25:04 by stevenj] whoops M ./reodft/reodft11e-r2hc-odd.c -5 +5 Thu Feb 27 02:22:03 EST 2003 stevenj * [project @ 2003-02-27 07:22:03 by stevenj] another optimization M ./reodft/reodft11e-r2hc-odd.c -53 +53 Thu Feb 27 01:43:00 EST 2003 stevenj * [project @ 2003-02-27 06:43:00 by stevenj] added op counts M ./reodft/reodft11e-r2hc-odd.c -2 +4 M ./reodft/reodft11e-radix2.c -2 +9 Thu Feb 27 01:29:32 EST 2003 stevenj * [project @ 2003-02-27 06:29:32 by stevenj] cleanup M ./reodft/reodft11e-r2hc-odd.c -53 +55 Thu Feb 27 01:17:23 EST 2003 stevenj * [project @ 2003-02-27 06:17:23 by stevenj] typo in comment M ./reodft/reodft11e-r2hc-odd.c -2 +2 Thu Feb 27 01:13:49 EST 2003 stevenj * [project @ 2003-02-27 06:13:49 by stevenj] fixed comment M ./reodft/reodft11e-r2hc-odd.c -2 +2 Thu Feb 27 01:12:05 EST 2003 stevenj * [project @ 2003-02-27 06:12:05 by stevenj] use E instead of R M ./reodft/reodft11e-r2hc-odd.c -9 +9 Thu Feb 27 01:05:39 EST 2003 stevenj * [project @ 2003-02-27 06:05:39 by stevenj] more unrolling to eliminate if statements in loops, for speedups of 25-40% M ./reodft/reodft11e-r2hc-odd.c -25 +28 Thu Feb 27 00:27:00 EST 2003 stevenj * [project @ 2003-02-27 05:27:00 by stevenj] some loop splitting to touch each element of output buf only once and eliminate some conditionals...speeds up by 30-40% M ./reodft/reodft11e-r2hc-odd.c -31 +167 Wed Feb 26 17:48:26 EST 2003 stevenj * [project @ 2003-02-26 22:48:26 by stevenj] comma M ./reodft/reodft11e-r2hc-odd.c -2 +2 Wed Feb 26 17:46:17 EST 2003 stevenj * [project @ 2003-02-26 22:46:17 by stevenj] pointer to odd case M ./reodft/reodft11e-radix2.c -1 +3 Wed Feb 26 17:40:54 EST 2003 stevenj * [project @ 2003-02-26 22:40:54 by stevenj] precision -> accuracy (c.f. Kahan) M ./reodft/reodft11e-r2hc.c -2 +2 Wed Feb 26 17:36:13 EST 2003 stevenj * [project @ 2003-02-26 22:36:13 by stevenj] added time limit for wisdom generation M ./Makefile.am -1 +4 M ./libbench2/bench-user.h -1 +2 M ./libbench2/problem.c -1 +4 M ./tools/fftw-wisdom.c -16 +59 M ./tools/fftw_wisdom.1.in +6 Wed Feb 26 13:24:36 EST 2003 stevenj * [project @ 2003-02-26 18:24:36 by stevenj] caps M ./reodft/reodft11e-r2hc-odd.c -2 +2 Tue Feb 25 20:56:01 EST 2003 stevenj * [project @ 2003-02-26 01:56:01 by stevenj] another note M ./reodft/reodft11e-r2hc-odd.c -2 +3 Tue Feb 25 20:54:57 EST 2003 stevenj * [project @ 2003-02-26 01:54:57 by stevenj] note M ./reodft/reodft11e-r2hc-odd.c -4 +5 Tue Feb 25 20:42:08 EST 2003 stevenj * [project @ 2003-02-26 01:42:08 by stevenj] added new, more accurate (hopefully) reodft11 algorithms; added --disable-debug-malloc; added --impulse-accuracy-rounds=rounds flags to libbench2 for impulse-response accuracy tests A ./reodft/reodft11e-r2hc-odd.c A ./reodft/reodft11e-radix2.c M ./configure.ac +8 M ./kernel/alloc.c -2 +2 M ./kernel/ifftw.h -3 +3 M ./libbench2/bench-main.c -2 +8 M ./libbench2/bench.h -2 +2 M ./libbench2/verify-dft.c -3 +4 M ./libbench2/verify-lib.c -18 +76 M ./libbench2/verify-r2r.c -7 +39 M ./libbench2/verify-rdft2.c -3 +4 M ./libbench2/verify.c -7 +10 M ./libbench2/verify.h -4 +7 M ./reodft/Makefile.am -1 +4 M ./reodft/conf.c -2 +4 M ./reodft/reodft.h -1 +3 M ./reodft/reodft11e-r2hc-odd.c +273 M ./reodft/reodft11e-r2hc.c -2 +8 M ./reodft/reodft11e-radix2.c +506 M ./tests/bench.c -1 +1 Sun Feb 23 14:07:48 EST 2003 athena * [project @ 2003-02-23 19:07:48 by athena] fftw_wisdom.1 is in $builddir, not $srcdir M ./tools/Makefile.am -1 +1 Mon Feb 17 03:42:19 EST 2003 stevenj * [project @ 2003-02-17 08:42:19 by stevenj] pde M ./doc/fftw3.texi -2 +2 Mon Feb 17 03:40:19 EST 2003 stevenj * [project @ 2003-02-17 08:40:19 by stevenj] consistent number M ./doc/fftw3.texi -4 +4 Mon Feb 17 03:39:02 EST 2003 stevenj * [project @ 2003-02-17 08:39:02 by stevenj] started r2r doc M ./doc/fftw3.texi -1 +65 Mon Feb 17 02:31:51 EST 2003 stevenj * [project @ 2003-02-17 07:31:50 by stevenj] rfftwnd A ./doc/rfftwnd.fig A ./doc/rfftwnd.gif M ./doc/Makefile.am +11 M ./doc/fftw3.texi -26 +147 M ./doc/rfftwnd.fig +1148 M ./doc/rfftwnd.gif Sat Feb 15 17:02:07 EST 2003 stevenj * [project @ 2003-02-15 22:02:07 by stevenj] continued M ./doc/fftw3.texi -4 +32 Sat Feb 15 15:16:26 EST 2003 stevenj * [project @ 2003-02-15 20:16:26 by stevenj] started r2c/c2r docs M ./doc/fftw3.texi -8 +87 Sat Feb 15 01:12:52 EST 2003 stevenj * [project @ 2003-02-15 06:12:52 by stevenj] added r{e,o}dft11 accuracy test M ./libbench2/verify-r2r.c -23 +45 Sat Feb 15 00:42:48 EST 2003 stevenj * [project @ 2003-02-15 05:42:48 by stevenj] added more r2r accuracy checks M ./libbench2/verify-dft.c -2 +2 M ./libbench2/verify-lib.c -10 +4 M ./libbench2/verify-r2r.c -1 +127 M ./libbench2/verify-rdft2.c -2 +2 M ./libbench2/verify.h -2 +3 Fri Feb 14 19:19:54 EST 2003 athena * [project @ 2003-02-15 00:19:54 by athena] $< is a GNUism M ./tools/Makefile.am -1 +1 Wed Feb 12 21:02:16 EST 2003 stevenj * [project @ 2003-02-13 02:02:16 by stevenj] r2r test cases are in M ./TODO -1 +1 Wed Feb 12 21:01:28 EST 2003 stevenj * [project @ 2003-02-13 02:01:28 by stevenj] added vector radix to TODO M ./TODO +2 Wed Feb 12 17:21:33 EST 2003 stevenj * [project @ 2003-02-12 22:21:33 by stevenj] fixed cross-ref M ./tools/fftw_wisdom.1.in -1 +1 Wed Feb 12 17:19:56 EST 2003 stevenj * [project @ 2003-02-12 22:19:56 by stevenj] shorter synopsis M ./tools/fftw_wisdom.1.in -1 +1 Wed Feb 12 12:53:19 EST 2003 stevenj * [project @ 2003-02-12 17:53:19 by stevenj] obsolete R ./tests/debug.h Wed Feb 12 12:52:53 EST 2003 stevenj * [project @ 2003-02-12 17:52:53 by stevenj] removed old dotens R ./tests/dotens.c R ./tests/dotens2.c Wed Feb 12 12:52:16 EST 2003 stevenj * [project @ 2003-02-12 17:52:16 by stevenj] removed old verify files R ./tests/verify-dft.c R ./tests/verify-lib.c R ./tests/verify-rdft.c R ./tests/verify-reodft.c R ./tests/verify.h Wed Feb 12 12:37:17 EST 2003 stevenj * [project @ 2003-02-12 17:37:17 by stevenj] disable threads support by default M ./tools/fftw-wisdom.c -4 +11 M ./tools/fftw_wisdom.1.in -7 Wed Feb 12 11:03:28 EST 2003 athena * [project @ 2003-02-12 16:03:28 by athena] Removed old test program M ./tests/bench.c -390 Tue Feb 11 22:30:55 EST 2003 stevenj * [project @ 2003-02-12 03:30:55 by stevenj] joke M ./tools/fftw-wisdom-to-conf.in -1 +1 Tue Feb 11 22:27:44 EST 2003 stevenj * [project @ 2003-02-12 03:27:44 by stevenj] add --help and --version, to be GNU-lly correct M ./tools/fftw-wisdom-to-conf.1 +7 M ./tools/fftw-wisdom-to-conf.in +36 Tue Feb 11 22:27:18 EST 2003 stevenj * [project @ 2003-02-12 03:27:18 by stevenj] whoops M ./tools/fftw_wisdom.1.in -1 Tue Feb 11 22:17:35 EST 2003 stevenj * [project @ 2003-02-12 03:15:03 by stevenj] better help M ./tools/fftw-wisdom.c -10 +43 Tue Feb 11 21:47:35 EST 2003 stevenj * [project @ 2003-02-12 02:47:35 by stevenj] comma M ./tools/fftw-wisdom-to-conf.1 -2 +1 Tue Feb 11 21:46:12 EST 2003 stevenj * [project @ 2003-02-12 02:46:12 by stevenj] formatting M ./tools/fftw-wisdom-to-conf.1 -1 +1 Tue Feb 11 21:45:23 EST 2003 stevenj * [project @ 2003-02-12 02:45:23 by stevenj] man pages for tools A ./tools/fftw-wisdom-to-conf.1 A ./tools/fftw_wisdom.1.in M ./configure.ac +1 M ./tools/Makefile.am -1 +7 M ./tools/fftw-wisdom-to-conf.1 +85 M ./tools/fftw_wisdom.1.in +190 Tue Feb 11 19:07:12 EST 2003 stevenj * [project @ 2003-02-12 00:07:12 by stevenj] added -V M ./tools/fftw-wisdom.c +25 Tue Feb 11 18:42:17 EST 2003 stevenj * [project @ 2003-02-11 23:42:17 by stevenj] added install-wisdom target M ./Makefile.am +10 Tue Feb 11 18:23:02 EST 2003 stevenj * [project @ 2003-02-11 23:23:02 by stevenj] another note M ./NEWS -1 +2 Tue Feb 11 17:32:56 EST 2003 stevenj * [project @ 2003-02-11 22:32:56 by stevenj] started r2r accuracy tests (only three kinds covered so far) M ./libbench2/verify-dft.c -2 +2 M ./libbench2/verify-lib.c -10 +13 M ./libbench2/verify-r2r.c +103 M ./libbench2/verify-rdft2.c -2 +2 M ./libbench2/verify.c -2 +2 M ./libbench2/verify.h -1 +4 Mon Feb 10 22:04:18 EST 2003 stevenj * [project @ 2003-02-11 03:04:18 by stevenj] silence warning M ./kernel/ifftw.h -3 +3 Mon Feb 10 20:55:20 EST 2003 athena * [project @ 2003-02-11 01:55:20 by athena] gcc bug is now avoided. M ./TODO -5 Mon Feb 10 20:37:54 EST 2003 athena * [project @ 2003-02-11 01:37:54 by athena] Accuracy test A ./libbench2/mp.c M ./libbench2/Makefile.am -5 +6 M ./libbench2/bench-user.h -7 +1 M ./libbench2/mp.c +592 M ./libbench2/verify-dft.c -1 +22 M ./libbench2/verify-lib.c -16 +39 M ./libbench2/verify-r2r.c -17 +2 M ./libbench2/verify-rdft2.c -1 +23 M ./libbench2/verify.c -2 +21 M ./libbench2/verify.h +27 Mon Feb 10 07:59:57 EST 2003 athena * [project @ 2003-02-10 12:59:57 by athena] There is no point in precomputing strides for the long-double code, as multiplication by sizeof(long double) cannot be folded into the addressing mode. This change also fixes the gcc-2.95 bug that causes miscompilation of certain codelets. M ./kernel/ifftw.h -2 +2 Mon Feb 10 02:54:35 EST 2003 stevenj * [project @ 2003-02-10 07:54:35 by stevenj] added random r2r tests M ./tests/check.pl -3 +19 Mon Feb 10 02:44:58 EST 2003 stevenj * [project @ 2003-02-10 07:44:58 by stevenj] whoops, bugfix: missing stride for ro10 M ./reodft/reodft010e-r2hc.c -2 +2 Mon Feb 10 02:21:50 EST 2003 stevenj * [project @ 2003-02-10 07:21:50 by stevenj] formatting M ./api/mapflags.c -1 +2 Sun Feb 9 23:24:52 EST 2003 stevenj * [project @ 2003-02-10 04:24:52 by stevenj] flop counts for reodft M ./reodft/redft00e-r2hc.c -5 +12 M ./reodft/reodft010e-r2hc.c -3 +16 M ./reodft/reodft11e-r2hc.c -3 +10 M ./reodft/rodft00e-r2hc.c -3 +12 Sun Feb 9 23:22:15 EST 2003 stevenj * [project @ 2003-02-10 04:22:15 by stevenj] declare aligned_main M ./libbench2/bench.h -1 +2 Sun Feb 9 20:56:06 EST 2003 stevenj * [project @ 2003-02-10 01:56:06 by stevenj] corrected rader op counts M ./rdft/dht-rader.c -6 +5 M ./rdft/rader-hc2hc.c -2 +2 Sun Feb 9 20:25:32 EST 2003 stevenj * [project @ 2003-02-10 01:25:32 by stevenj] punctuation M ./TODO -1 +1 Sun Feb 9 20:25:17 EST 2003 stevenj * [project @ 2003-02-10 01:25:17 by stevenj] noted need for better estimator M ./TODO +4 Sun Feb 9 19:58:59 EST 2003 stevenj * [project @ 2003-02-10 00:58:59 by stevenj] noted F77 api fix for g77 mangling incompatibility M ./NEWS +3 Sun Feb 9 19:30:55 EST 2003 stevenj * [project @ 2003-02-10 00:30:55 by stevenj] build f77 header file of constants from fftw3.h M ./api/Makefile.am -1 +10 Sun Feb 9 19:04:53 EST 2003 stevenj * [project @ 2003-02-10 00:04:53 by stevenj] updates M ./TODO -7 +7 Sun Feb 9 19:03:34 EST 2003 stevenj * [project @ 2003-02-10 00:03:34 by stevenj] threads f77 api A ./api/x77.h A ./threads/f77api.c A ./threads/f77funcs.c M ./api/Makefile.am -1 +1 M ./api/f77api.c -12 +1 M ./api/x77.h +32 M ./threads/Makefile.am -1 +1 M ./threads/f77api.c +39 M ./threads/f77funcs.c +39 Sun Feb 9 18:54:00 EST 2003 stevenj * [project @ 2003-02-09 23:54:00 by stevenj] finished f77 serial api M ./api/f77api.c +15 M ./api/f77funcs.c +220 Sun Feb 9 18:32:26 EST 2003 stevenj * [project @ 2003-02-09 23:32:26 by stevenj] added flops, slight cleanups M ./api/f77api.c -8 +5 M ./api/f77funcs.c -21 +27 Sun Feb 9 18:11:48 EST 2003 athena * [project @ 2003-02-09 23:11:48 by athena] Oops, forgot #include M ./libbench2/aligned-main.c -1 +3 Sun Feb 9 18:08:26 EST 2003 athena * [project @ 2003-02-09 23:08:26 by athena] Removed duplication of stack-alignment code A ./libbench2/aligned-main.c M ./libbench2/Makefile.am -5 +5 M ./libbench2/aligned-main.c +81 M ./libbench2/bench-main.c -62 +2 M ./libbench2/main.c -2 +2 M ./tools/fftw-wisdom.c -61 +1 Sun Feb 9 15:48:15 EST 2003 stevenj * [project @ 2003-02-09 20:48:15 by stevenj] allow - to read problems from stdin M ./tools/fftw-wisdom.c -2 +14 Sun Feb 9 15:22:23 EST 2003 stevenj * [project @ 2003-02-09 20:22:23 by stevenj] added fftw-wisdom tool A ./tools/fftw-wisdom.c M ./tools/Makefile.am +15 M ./tools/fftw-wisdom.c +298 Sun Feb 9 15:06:38 EST 2003 stevenj * [project @ 2003-02-09 20:06:38 by stevenj] elim. warning M ./tests/bench.c -1 +1 Sun Feb 9 14:24:19 EST 2003 stevenj * [project @ 2003-02-09 19:24:19 by stevenj] destroy_input should not contaminate flags of other problems M ./tests/bench.c -1 +5 Sun Feb 9 13:06:11 EST 2003 stevenj * [project @ 2003-02-09 18:06:11 by stevenj] updated M ./ChangeLog -53 +1740 Sun Feb 9 13:01:45 EST 2003 stevenj * [project @ 2003-02-09 18:01:44 by stevenj] removed overzealous inplace check, which caused problems for rdft2 M ./dft/rank-geq2.c -12 +1 M ./rdft/rank-geq2-rdft2.c -3 +2 M ./rdft/rank-geq2.c -12 +1 Sun Feb 9 08:14:03 EST 2003 athena * [project @ 2003-02-09 13:14:03 by athena] Consistent syntax for RNK_MINFTY tensors M ./kernel/tensor.c -4 +4 Sun Feb 9 07:31:13 EST 2003 athena * [project @ 2003-02-09 12:31:13 by athena] lisply-correct tensor print. We no longer need to parse tensors. M ./kernel/tensor.c -3 +9 Sun Feb 9 03:35:56 EST 2003 stevenj * [project @ 2003-02-09 08:35:56 by stevenj] removed completed items M ./TODO -4 Sun Feb 9 03:27:56 EST 2003 stevenj * [project @ 2003-02-09 08:27:56 by stevenj] slight renaming M ./libbench2/verify-r2r.c -4 +4 Sun Feb 9 03:15:28 EST 2003 stevenj * [project @ 2003-02-09 08:15:28 by stevenj] multi-dimensional r2r verifier M ./libbench2/problem.c -2 +2 M ./libbench2/verify-r2r.c -81 +118 Sun Feb 9 02:40:22 EST 2003 stevenj * [project @ 2003-02-09 07:40:22 by stevenj] comments M ./libbench2/verify-r2r.c -1 +3 Sun Feb 9 02:38:26 EST 2003 stevenj * [project @ 2003-02-09 07:38:26 by stevenj] slight simplification M ./libbench2/verify-r2r.c -7 +2 Sun Feb 9 02:36:25 EST 2003 stevenj * [project @ 2003-02-09 07:36:25 by stevenj] added 1d r2r verifier (triple ugh) A ./libbench2/verify-r2r.c M ./libbench2/Makefile.am -2 +2 M ./libbench2/allocate.c -1 +17 M ./libbench2/bench-user.h -2 +10 M ./libbench2/mflops.c +1 M ./libbench2/problem.c -7 +68 M ./libbench2/verify-r2r.c +616 M ./libbench2/verify.c -2 +2 M ./libbench2/zero.c -1 +4 M ./tests/bench.c +101 Sat Feb 8 22:23:00 EST 2003 stevenj * [project @ 2003-02-09 03:23:00 by stevenj] added vector transforms to random tests M ./tests/check.pl -3 +11 Sat Feb 8 20:59:07 EST 2003 stevenj * [project @ 2003-02-09 01:59:07 by stevenj] whoops M ./rdft/direct2.c -2 +1 Sat Feb 8 19:52:58 EST 2003 stevenj * [project @ 2003-02-09 00:52:58 by stevenj] fixed interaction between dwims for sz/vecsz with rdft2 transforms M ./libbench2/problem.c -15 +17 Sat Feb 8 19:35:56 EST 2003 stevenj * [project @ 2003-02-09 00:35:56 by stevenj] added destroy_input flag/check M ./libbench2/bench-user.h -1 +2 M ./libbench2/problem.c -1 +3 M ./libbench2/verify-dft.c -1 +4 M ./libbench2/verify-lib.c -1 +35 M ./libbench2/verify-rdft2.c -1 +5 M ./libbench2/verify.h +2 M ./tests/bench.c +7 Sat Feb 8 19:11:58 EST 2003 stevenj * [project @ 2003-02-09 00:11:57 by stevenj] added rdft2 verifier A ./api/extract-reim.c A ./libbench2/aset.c A ./libbench2/verify-rdft2.c M ./api/Makefile.am -3 +3 R ./api/dfthelp.c M ./api/extract-reim.c +36 M ./api/plan-guru-dft-c2r.c -1 +2 M ./api/plan-many-dft-c2r.c +2 M ./libbench2/Makefile.am -5 +5 M ./libbench2/allocate.c -7 +44 M ./libbench2/aset.c +10 M ./libbench2/bench-user.h -1 +4 M ./libbench2/bench.h -1 +2 M ./libbench2/problem.c -10 +42 M ./libbench2/tensor.c -1 +47 M ./libbench2/verify-dft.c -1 +3 M ./libbench2/verify-lib.c -16 +13 M ./libbench2/verify-rdft2.c +255 M ./libbench2/verify.c -2 +2 M ./libbench2/verify.h -1 +1 M ./libbench2/zero.c -2 +8 M ./tests/bench.c -8 +189 M ./tests/check.pl -1 +7 Sat Feb 8 13:31:14 EST 2003 stevenj * [project @ 2003-02-08 18:31:14 by stevenj] an additional check for in-place case M ./rdft/rdft2-radix2.c -3 +7 Fri Feb 7 17:36:56 EST 2003 stevenj * [project @ 2003-02-07 22:36:56 by stevenj] slight fix: hc2r constraints are mostly determined by sub-plan M ./rdft/rank0-rdft2.c -6 +7 Fri Feb 7 16:28:55 EST 2003 stevenj * [project @ 2003-02-07 21:28:55 by stevenj] make radix2-dft inapplicable to in-place/split case (r == rio, iio >= rio + n/2+1 != r + 1) M ./rdft/rdft2-radix2.c -3 +22 Tue Feb 4 06:36:29 EST 2003 athena * [project @ 2003-02-04 11:36:29 by athena] Allow plnr->hook to be 0 M ./kernel/planner.c -12 +12 M ./tests/hook.c -7 +2 Tue Feb 4 03:25:36 EST 2003 stevenj * [project @ 2003-02-04 08:25:36 by stevenj] moved dft stuff into verify-dft M ./libbench2/bench-user.h -4 +2 M ./libbench2/verify-dft.c -9 +84 M ./libbench2/verify.c -84 +2 Tue Feb 4 03:25:00 EST 2003 stevenj * [project @ 2003-02-04 08:25:00 by stevenj] cruft M ./tests/hook.c -1 Tue Feb 4 03:18:28 EST 2003 stevenj * [project @ 2003-02-04 08:18:28 by stevenj] further unify libbench2 and paranoid verifiers M ./libbench2/bench-user.h -1 +6 M ./libbench2/problem.c -1 +2 M ./libbench2/verify.c -16 +21 M ./tests/bench.c -3 +5 M ./tests/hook.c -125 +58 Sun Feb 2 01:45:37 EST 2003 stevenj * [project @ 2003-02-02 06:45:37 by stevenj] typo in comment M ./api/import-wisdom-from-file.c -1 +1 Sat Feb 1 09:30:03 EST 2003 athena * [project @ 2003-02-01 14:30:03 by athena] Fixed p==2 case M ./kernel/primes.c -1 +4 Sat Feb 1 09:23:43 EST 2003 athena * [project @ 2003-02-01 14:23:43 by athena] Incorporated new find_generator by Greg Dionne. M ./kernel/primes.c -21 +35 Fri Jan 31 20:46:24 EST 2003 athena * [project @ 2003-02-01 01:46:24 by athena] Removed nonportable call to gettext() M ./libbench2/getopt.c +7 Wed Jan 29 19:03:43 EST 2003 athena * [project @ 2003-01-30 00:03:43 by athena] uintptr_t is in in openbsd M ./kernel/ifftw.h -1 +5 Wed Jan 29 15:41:56 EST 2003 athena * [project @ 2003-01-29 20:41:56 by athena] Huge speedups in wisdom I/O. M ./api/export-wisdom-to-string.c -2 +2 M ./api/export-wisdom.c -1 +1 M ./api/import-wisdom-from-file.c -4 +27 M ./api/mkprinter-file.c -5 +25 M ./kernel/debug.c -2 +2 M ./kernel/ifftw.h -2 +5 M ./kernel/planner.c -5 +9 M ./kernel/print.c -2 +7 M ./kernel/scan.c -4 +25 M ./tests/bench.c -1 +5 Tue Jan 28 19:36:51 EST 2003 athena * [project @ 2003-01-29 00:36:51 by athena] Added appropriate warning against likely future bug. M ./kernel/planner.c -1 +4 Tue Jan 28 19:00:24 EST 2003 athena * [project @ 2003-01-29 00:00:24 by athena] Don't attempt to remove bogus wisdom entries. M ./kernel/planner.c -10 +3 Tue Jan 28 18:16:24 EST 2003 athena * [project @ 2003-01-28 23:16:24 by athena] Fixed a couple of very very very nasty bugs---pointers became invalid after the hash table was relocated. M ./kernel/planner.c -19 +29 Tue Jan 28 07:34:10 EST 2003 athena * [project @ 2003-01-28 12:34:10 by athena] Read wisdom at can_do() time, otherwise wisdom is destroyed. M ./tests/bench.c -2 +5 Tue Jan 28 06:54:38 EST 2003 athena * [project @ 2003-01-28 11:54:38 by athena] More conservative inheritance of blessings M ./kernel/planner.c -19 +24 Tue Jan 28 06:50:20 EST 2003 athena * [project @ 2003-01-28 11:50:20 by athena] Print the same info as it is hashed M ./dft/problem.c -3 +4 Tue Jan 28 06:49:48 EST 2003 athena * [project @ 2003-01-28 11:49:48 by athena] Print name of executable when FAILURE M ./tests/check.pl -2 +2 Mon Jan 27 06:59:40 EST 2003 athena * [project @ 2003-01-27 11:59:40 by athena] New NO_SEARCH planner flag, which avoids searching altogether. A wisdom entry must lead to a NO_SEARCH-grade plan, or else the wisdom entry is bogus. M ./kernel/ifftw.h -5 +10 M ./kernel/planner.c -13 +30 Sun Jan 26 20:45:21 EST 2003 athena * [project @ 2003-01-27 01:45:21 by athena] Use cosl()/sinl() when appropriate M ./libbench2/verify-lib.c -3 +20 Sun Jan 26 16:29:18 EST 2003 athena * [project @ 2003-01-26 21:29:18 by athena] Use null pointers when estimating. The estimator should never time anything. M ./kernel/planner.c -1 +6 M ./libbench2/problem.c -2 +1 M ./libbench2/speed.c -1 +2 M ./libbench2/verify.c -1 +3 Sun Jan 26 15:19:01 EST 2003 stevenj * [project @ 2003-01-26 20:19:01 by stevenj] note M ./api/f77api.c -1 +1 Sun Jan 26 15:16:22 EST 2003 stevenj * [project @ 2003-01-26 20:16:22 by stevenj] support multiple mangling schemes with g77 A ./api/f77funcs.c M ./api/Makefile.am +4 M ./api/f77api.c -130 +41 M ./api/f77funcs.c +139 M ./configure.ac +6 Sun Jan 26 12:58:57 EST 2003 stevenj * [project @ 2003-01-26 17:58:57 by stevenj] fixed verbose, made random tests only use selected rank, use rank <= 4, fixed final flush_problems call M ./tests/check.pl -4 +9 Sun Jan 26 12:42:49 EST 2003 stevenj * [project @ 2003-01-26 17:42:49 by stevenj] fixed typo (count instead of maxcount) M ./tests/check.pl -2 +2 Sun Jan 26 12:12:07 EST 2003 stevenj * [project @ 2003-01-26 17:12:07 by stevenj] hypot is no longer used M ./configure.ac -1 +1 Sun Jan 26 12:07:43 EST 2003 stevenj * [project @ 2003-01-26 17:07:43 by stevenj] check for _alloca (MSVC) M ./configure.ac -1 +1 M ./kernel/ifftw.h -1 +5 Sun Jan 26 11:56:53 EST 2003 stevenj * [project @ 2003-01-26 16:56:53 by stevenj] slight fix in assert M ./kernel/alloc.c -2 +2 Sun Jan 26 11:55:39 EST 2003 athena * [project @ 2003-01-26 16:55:39 by athena] Allocate problem in all cases--- can_do may need correct pointers. M ./libbench2/problem.c -1 +2 M ./libbench2/speed.c -2 +1 M ./libbench2/verify.c -3 +1 M ./tests/bench.c -6 Sun Jan 26 11:51:27 EST 2003 athena * [project @ 2003-01-26 16:51:27 by athena] Nastier checks M ./tests/bench.c -1 +8 M ./tests/check.pl -9 +23 Sun Jan 26 11:51:16 EST 2003 athena * [project @ 2003-01-26 16:51:16 by athena] X(use_plan) is a relic. M ./kernel/ifftw.h -3 +1 M ./kernel/plan.c -11 +3 M ./kernel/planner.c -5 +2 Sun Jan 26 09:23:16 EST 2003 athena * [project @ 2003-01-26 14:23:16 by athena] Print full pathname of the bench executable, so that I don't get confused when running multiple tests for different configurations. M ./tests/Makefile.am -1 +1 Sun Jan 26 07:35:46 EST 2003 athena * [project @ 2003-01-26 12:35:46 by athena] Split done() into done() and cleanup(), in order to test multiple problems with the same planner from the command line. M ./libbench2/bench-main.c -1 +2 M ./libbench2/bench-user.h -1 +2 M ./tests/bench.c -1 +8 Sat Jan 25 20:44:49 EST 2003 athena * [project @ 2003-01-26 01:44:49 by athena] Improved readability M ./kernel/alloc.c -6 +16 Sat Jan 25 19:17:26 EST 2003 stevenj * [project @ 2003-01-26 00:17:26 by stevenj] comment M ./kernel/alloc.c -3 +3 Sat Jan 25 19:16:53 EST 2003 stevenj * [project @ 2003-01-26 00:16:53 by stevenj] added macos9 mpallocatealigned function M ./kernel/alloc.c -1 +19 Sat Jan 25 18:59:55 EST 2003 stevenj * [project @ 2003-01-25 23:59:55 by stevenj] sometimes __APPLE__ is defined instead of __MACOSX__ M ./kernel/alloc.c -2 +3 Sat Jan 25 18:54:39 EST 2003 stevenj * [project @ 2003-01-25 23:54:39 by stevenj] macos x malloc is already 16-byte aligned M ./kernel/alloc.c -2 +3 Sat Jan 25 13:38:32 EST 2003 athena * [project @ 2003-01-25 18:38:32 by athena] Include because uintptr_t is defined there on solaris. M ./kernel/ifftw.h -2 +6 Sat Jan 25 13:22:59 EST 2003 athena * [project @ 2003-01-25 18:22:59 by athena] Oops---forgot getopt_long A ./libbench2/getopt1.c M ./libbench2/Makefile.am -2 +2 M ./libbench2/getopt1.c +188 Sat Jan 25 13:17:29 EST 2003 athena * [project @ 2003-01-25 18:17:29 by athena] Include default includes when checking for uintptr_t. (Otherwise solaris breaks.) M ./configure.ac -1 +1 Sat Jan 25 12:39:52 EST 2003 athena * [project @ 2003-01-25 17:39:52 by athena] distribute check.pl M ./tests/Makefile.am +1 Sat Jan 25 12:38:34 EST 2003 athena * [project @ 2003-01-25 17:38:34 by athena] Check split format, too. M ./tests/check.pl +4 Sat Jan 25 11:48:19 EST 2003 athena * [project @ 2003-01-25 16:48:19 by athena] New tests, added make check M ./tests/Makefile.am -1 +2 M ./tests/check.pl -3 +66 Thu Jan 23 08:34:24 EST 2003 athena * [project @ 2003-01-23 13:34:24 by athena] More tests M ./tests/check.pl -17 +83 Tue Jan 21 20:32:12 EST 2003 athena * [project @ 2003-01-22 01:32:09 by athena] Deal with rnk(sz)=-infinity M ./api/mktensor-iodims.c -4 +9 M ./api/mktensor-rowmajor.c -4 +6 M ./libbench2/problem.c -5 +2 Tue Jan 21 10:07:16 EST 2003 athena * [project @ 2003-01-21 15:07:16 by athena] Crazy idea M ./TODO +3 Tue Jan 21 07:14:22 EST 2003 athena * [project @ 2003-01-21 12:14:22 by athena] Test program, still barely worthy of the name. A ./tests/check.pl Mon Jan 20 08:29:21 EST 2003 athena * [project @ 2003-01-20 13:29:21 by athena] Stylistic changes M ./libbench2/problem.c -15 +10 Mon Jan 20 07:03:38 EST 2003 athena * [project @ 2003-01-20 12:03:38 by athena] Implemented flops api A ./api/flops.c M ./api/Makefile.am -10 +10 M ./api/fftw3.h -2 +4 M ./api/flops.c +27 M ./tests/bench.c +3 Sun Jan 19 14:27:21 EST 2003 stevenj * [project @ 2003-01-19 19:27:21 by stevenj] cleanup M ./libbench2/problem.c -31 +24 Sun Jan 19 14:14:49 EST 2003 stevenj * [project @ 2003-01-19 19:14:49 by stevenj] 'v' syntax now defaults to an 'internal' (stride 1) vector, which is a more interesting case and corresponds more closely to the intuitive notion of a 'vector' transform, while '*' does the old 'external' (stride n) vector M ./libbench2/problem.c -4 +24 Sun Jan 19 13:55:35 EST 2003 stevenj * [project @ 2003-01-19 18:55:35 by stevenj] removed '/' overloading M ./libbench2/problem.c -2 +2 Sun Jan 19 13:52:09 EST 2003 stevenj * [project @ 2003-01-19 18:52:09 by stevenj] get rid of '*' and ',' synonyms for 'x' in problem parser; there's no need to clutter the namespace with syntax we never use M ./libbench2/problem.c -2 +2 Sun Jan 19 07:28:27 EST 2003 athena * [project @ 2003-01-19 12:28:27 by athena] Signed/unsigned fixes. M ./kernel/planner.c -4 +4 Sun Jan 19 07:09:54 EST 2003 athena * [project @ 2003-01-19 12:09:54 by athena] Test split arrays. M ./libbench2/bench-user.h -3 +9 M ./libbench2/verify-dft.c -14 +11 M ./libbench2/verify.c -13 +43 M ./libbench2/verify.h +1 M ./tests/bench.c -13 +50 M ./tests/hook.c -1 +4 Sat Jan 18 23:46:57 EST 2003 stevenj * [project @ 2003-01-19 04:46:57 by stevenj] clarification M ./doc/fftw3.texi -17 +16 Sat Jan 18 21:53:18 EST 2003 stevenj * [project @ 2003-01-19 02:53:18 by stevenj] caps M ./doc/fftw3.texi -2 +2 Sat Jan 18 21:52:51 EST 2003 stevenj * [project @ 2003-01-19 02:52:51 by stevenj] brackets M ./doc/fftw3.texi -2 +2 Sat Jan 18 21:52:32 EST 2003 stevenj * [project @ 2003-01-19 02:52:32 by stevenj] quote M ./doc/fftw3.texi -1 +7 Sat Jan 18 20:53:11 EST 2003 stevenj * [project @ 2003-01-19 01:53:11 by stevenj] referencing M ./doc/fftw3.texi -4 +7 Sat Jan 18 20:33:28 EST 2003 stevenj * [project @ 2003-01-19 01:33:28 by stevenj] fix M ./doc/fftw3.texi -2 +2 Sat Jan 18 20:31:41 EST 2003 stevenj * [project @ 2003-01-19 01:31:41 by stevenj] slight change M ./doc/fftw3.texi -8 +8 Sat Jan 18 20:31:22 EST 2003 athena * [project @ 2003-01-19 01:31:22 by athena] Print errors when --verify. M ./libbench2/verify-dft.c -7 +12 M ./libbench2/verify-lib.c -30 +38 M ./libbench2/verify.h -12 +14 M ./tests/bench.c -5 +5 Sat Jan 18 20:30:27 EST 2003 stevenj * [project @ 2003-01-19 01:30:27 by stevenj] improved description, noted that FFTW_ESTIMATE does not destroy arrays M ./doc/fftw3.texi -7 +8 Sat Jan 18 20:23:12 EST 2003 stevenj * [project @ 2003-01-19 01:23:12 by stevenj] FFTW_DEFAULTS isn't really needed M ./api/fftw3.h -2 +1 Sat Jan 18 20:21:09 EST 2003 stevenj * [project @ 2003-01-19 01:21:09 by stevenj] added FFTW_MEASURE synonym for FFTW_DEFAULTS M ./api/fftw3.h -1 +2 M ./doc/fftw3.texi -6 +6 Sat Jan 18 20:18:29 EST 2003 stevenj * [project @ 2003-01-19 01:18:29 by stevenj] slight change M ./kernel/alloc.c -2 +2 Sat Jan 18 20:16:08 EST 2003 athena * [project @ 2003-01-19 01:16:08 by athena] Clearer name M ./tests/bench.c -2 +2 Sat Jan 18 20:13:14 EST 2003 athena * [project @ 2003-01-19 01:13:14 by athena] Completed dft api test M ./api/fftw3.h -12 +12 M ./libbench2/tensor.c -3 +3 M ./tests/bench.c -8 +65 Sat Jan 18 20:07:33 EST 2003 stevenj * [project @ 2003-01-19 01:07:33 by stevenj] index M ./doc/fftw3.texi -1 +2 Sat Jan 18 20:05:50 EST 2003 stevenj * [project @ 2003-01-19 01:05:50 by stevenj] fix M ./doc/fftw3.texi -2 +2 Sat Jan 18 20:04:11 EST 2003 stevenj * [project @ 2003-01-19 01:04:11 by stevenj] parallel structure M ./doc/fftw3.texi -2 +2 Sat Jan 18 20:03:18 EST 2003 stevenj * [project @ 2003-01-19 01:03:18 by stevenj] fix M ./doc/fftw3.texi -4 +4 Sat Jan 18 20:00:24 EST 2003 stevenj * [project @ 2003-01-19 01:00:24 by stevenj] joke M ./doc/fftw3.texi -5 +6 Sat Jan 18 19:59:28 EST 2003 stevenj * [project @ 2003-01-19 00:59:28 by stevenj] recommendation to read tutorial in-order M ./doc/fftw3.texi -1 +6 Sat Jan 18 19:54:55 EST 2003 stevenj * [project @ 2003-01-19 00:54:55 by stevenj] expanded outline M ./doc/fftw3.texi -7 +42 Sat Jan 18 19:35:52 EST 2003 stevenj * [project @ 2003-01-19 00:35:52 by stevenj] clarification M ./doc/fftw3.texi -3 +5 Sat Jan 18 19:17:27 EST 2003 stevenj * [project @ 2003-01-19 00:17:27 by stevenj] draft complex-dft tutorial M ./doc/fftw3.texi -25 +183 Sat Jan 18 17:27:15 EST 2003 athena * [project @ 2003-01-18 22:27:15 by athena] Paranoid mode is back. Fixed dwim to do what I mean. A ./tests/hook.c M ./libbench2/allocate.c -2 +1 M ./libbench2/bench-main.c -2 +1 M ./libbench2/bench-user.h -8 +1 M ./libbench2/bench.h -2 +1 M ./libbench2/can-do.c -2 +1 M ./libbench2/dotens2.c -2 +2 M ./libbench2/info.c -2 +1 M ./libbench2/problem.c -18 +21 M ./libbench2/report.c -2 +1 M ./libbench2/speed.c -2 +1 M ./libbench2/timer.c -2 +1 M ./libbench2/util.c -1 M ./libbench2/verify.c -4 +3 M ./libbench2/verify.h +7 M ./libbench2/zero.c -2 +1 M ./tests/Makefile.am -1 +1 M ./tests/bench.c +4 M ./tests/hook.c +191 Sat Jan 18 17:13:51 EST 2003 stevenj * [project @ 2003-01-18 22:13:51 by stevenj] started tut. M ./doc/fftw3.texi -14 +46 Sat Jan 18 16:13:15 EST 2003 athena * [project @ 2003-01-18 21:13:15 by athena] Great renaming, so that we can include both bench-user.h and ifftw.h to implement the paranoid-mode hook. M ./libbench2/allocate.c -5 +5 M ./libbench2/bench-user.h -30 +28 M ./libbench2/bench.h -12 +12 M ./libbench2/can-do.c -2 +2 M ./libbench2/dotens2.c -3 +3 M ./libbench2/mflops.c -1 +1 M ./libbench2/problem.c -12 +12 M ./libbench2/report.c -10 +10 M ./libbench2/speed.c -2 +2 M ./libbench2/tensor.c -22 +22 M ./libbench2/verify-dft.c -2 +2 M ./libbench2/verify-lib.c -6 +6 M ./libbench2/verify.c -9 +9 M ./libbench2/verify.h -2 +2 M ./libbench2/zero.c -2 +2 M ./tests/bench.c -29 +29 Sat Jan 18 15:41:18 EST 2003 athena * [project @ 2003-01-18 20:41:18 by athena] Trying to tweak the verifier so that I can use it in bench.c for paranoid mode M ./libbench2/bench-user.h -1 +23 M ./libbench2/problem.c -3 +3 M ./libbench2/verify-dft.c -66 +9 M ./libbench2/verify-lib.c -21 +18 M ./libbench2/verify.c -5 +62 M ./libbench2/verify.h -18 +3 Sat Jan 18 10:24:05 EST 2003 athena * [project @ 2003-01-18 15:24:05 by athena] Added stride_factor for complex arrays. M ./tests/bench.c -5 +5 Sat Jan 18 10:02:11 EST 2003 athena * [project @ 2003-01-18 15:02:11 by athena] can_do now calls the planner. M ./tests/bench.c -1 +6 Sat Jan 18 09:59:24 EST 2003 athena * [project @ 2003-01-18 14:59:24 by athena] Call guru api in bench.c M ./api/plan-guru-dft.c -2 +1 M ./tests/bench.c -3 +47 Sat Jan 18 08:17:23 EST 2003 athena * [project @ 2003-01-18 13:17:23 by athena] Fixed prototype. M ./libbench2/bench.h -1 +3 M ./libbench2/zero.c -2 +2 Sat Jan 18 08:14:48 EST 2003 athena * [project @ 2003-01-18 13:14:48 by athena] Attempt to make the signed/unsigned use of flags consistent. M ./api/api.h -2 +2 M ./api/apiplan.c -1 +1 M ./api/fftw3.h -25 +25 M ./api/mapflags.c -8 +10 M ./api/plan-dft-1d.c -1 +1 M ./api/plan-dft-2d.c -1 +1 M ./api/plan-dft-3d.c -1 +1 M ./api/plan-dft-c2r-1d.c -1 +1 M ./api/plan-dft-c2r-2d.c -1 +1 M ./api/plan-dft-c2r-3d.c -1 +1 M ./api/plan-dft-c2r.c -1 +1 M ./api/plan-dft-r2c-1d.c -1 +1 M ./api/plan-dft-r2c-2d.c -1 +1 M ./api/plan-dft-r2c-3d.c -1 +1 M ./api/plan-dft-r2c.c -1 +1 M ./api/plan-dft.c -1 +1 M ./api/plan-guru-dft-c2r.c -1 +1 M ./api/plan-guru-dft-r2c.c -1 +1 M ./api/plan-guru-dft.c -1 +1 M ./api/plan-guru-r2r.c -1 +1 M ./api/plan-many-dft-c2r.c -1 +1 M ./api/plan-many-dft-r2c.c -1 +1 M ./api/plan-many-dft.c -1 +1 M ./api/plan-many-r2r.c -1 +1 M ./api/plan-r2r-1d.c -1 +1 M ./api/plan-r2r-2d.c -1 +1 M ./api/plan-r2r-3d.c -1 +1 M ./api/plan-r2r.c -1 +1 M ./kernel/ifftw.h -2 +2 Sat Jan 18 08:03:07 EST 2003 athena * [project @ 2003-01-18 13:03:07 by athena] Implemented useropt. A ./libbench2/useropt.c M ./libbench2/Makefile.am -2 +3 M ./libbench2/bench-main.c -1 +5 M ./libbench2/bench-user.h -1 +2 M ./libbench2/useropt.c +30 M ./tests/bench.c -7 +24 Sat Jan 18 08:02:05 EST 2003 athena * [project @ 2003-01-18 13:02:05 by athena] The first map_flags pass must be transitive, i.e., always use the latest flags value as opposed to the original value. (I think.) M ./api/mapflags.c -11 +12 Sat Jan 18 07:20:19 EST 2003 athena * [project @ 2003-01-18 12:20:18 by athena] Started working on verifier A ./libbench2/dotens2.c A ./libbench2/verify-dft.c A ./libbench2/verify-lib.c A ./libbench2/verify.h M ./libbench2/Makefile.am -3 +3 M ./libbench2/bench-user.h -1 +4 M ./libbench2/dotens2.c +55 M ./libbench2/tensor.c -1 +8 M ./libbench2/verify-dft.c +119 M ./libbench2/verify-lib.c +399 M ./libbench2/verify.c -2 +9 M ./libbench2/verify.h +65 M ./tests/Makefile.am -4 +1 Fri Jan 17 14:53:28 EST 2003 stevenj * [project @ 2003-01-17 19:53:28 by stevenj] added X(threads_cleanup) M ./api/fftw3.h -1 +2 M ./threads/api.c +9 M ./threads/threads.c -6 +19 M ./threads/threads.h -1 +2 Fri Jan 17 10:35:56 EST 2003 athena * [project @ 2003-01-17 15:35:56 by athena] Use C style for upper and lower array bounds. Free tensors properly. M ./libbench2/allocate.c -3 +5 M ./libbench2/tensor.c -2 +2 Fri Jan 17 08:50:42 EST 2003 athena * [project @ 2003-01-17 13:50:42 by athena] Fixed ambiguous syntax M ./libbench2/problem.c -2 +2 Fri Jan 17 08:20:57 EST 2003 athena * [project @ 2003-01-17 13:20:57 by athena] Parse minus sign, bugfixes M ./libbench2/problem.c -4 +17 Fri Jan 17 08:11:56 EST 2003 athena * [project @ 2003-01-17 13:11:56 by athena] Skeleton libbench2 implemented (probably still buggy) A ./libbench2/ A ./libbench2/Makefile.am A ./libbench2/allocate.c A ./libbench2/bench-main.c A ./libbench2/bench-user.h A ./libbench2/bench.h A ./libbench2/can-do.c A ./libbench2/caset.c A ./libbench2/getopt-utils.c A ./libbench2/getopt.c A ./libbench2/getopt.h A ./libbench2/info.c A ./libbench2/main.c A ./libbench2/mflops.c A ./libbench2/ovtpvt.c A ./libbench2/pow2.c A ./libbench2/problem.c A ./libbench2/report.c A ./libbench2/speed.c A ./libbench2/tensor.c A ./libbench2/timer.c A ./libbench2/util.c A ./libbench2/verify.c A ./libbench2/zero.c M ./Makefile.am -1 +1 M ./configure.ac -1 +1 M ./libbench2/Makefile.am +13 M ./libbench2/allocate.c +55 M ./libbench2/bench-main.c +250 M ./libbench2/bench-user.h +206 M ./libbench2/bench.h +67 M ./libbench2/can-do.c +33 M ./libbench2/caset.c +12 M ./libbench2/getopt-utils.c +104 M ./libbench2/getopt.c +1062 M ./libbench2/getopt.h +180 M ./libbench2/info.c +59 M ./libbench2/main.c +40 M ./libbench2/mflops.c +25 M ./libbench2/ovtpvt.c +13 M ./libbench2/pow2.c +6 M ./libbench2/problem.c +172 M ./libbench2/report.c +159 M ./libbench2/speed.c +73 M ./libbench2/tensor.c +173 M ./libbench2/timer.c +241 M ./libbench2/util.c +216 M ./libbench2/verify.c +53 M ./libbench2/zero.c +36 M ./tests/Makefile.am -2 +2 M ./tests/bench.c -56 +93 Fri Jan 17 04:23:37 EST 2003 athena * [project @ 2003-01-17 09:23:37 by athena] Formatting M ./kernel/tensor4.c -10 +10 Fri Jan 17 03:15:24 EST 2003 fftw * [project @ 2003-01-17 08:15:24 by fftw] slight updates M ./doc/fftw3.texi -27 +34 Fri Jan 17 01:44:44 EST 2003 stevenj * [project @ 2003-01-17 06:44:44 by stevenj] eliminated obsolete uimin/uimax M ./dft/vrank-geq1.c -2 +2 M ./kernel/buffered.c -2 +2 M ./kernel/ifftw.h -3 +1 M ./kernel/minmax.c -11 +1 M ./kernel/tensor4.c -4 +4 M ./rdft/buffered2.c -4 +4 M ./rdft/rdft2-inplace-strides.c -2 +2 M ./rdft/rdft2-tensor-max-index.c -4 +4 M ./rdft/vrank-geq1-rdft2.c -2 +2 M ./rdft/vrank-geq1.c -2 +2 Fri Jan 17 01:40:10 EST 2003 stevenj * [project @ 2003-01-17 06:40:10 by stevenj] threads needs to have its own library, lest all programs linking to libfftw3.so need -lpthread A ./threads/api.c M ./Makefile.am -1 M ./api/Makefile.am -4 +3 M ./api/configure.c -2 M ./api/fftw3.h -2 +3 R ./api/plan-with-nthreads.c M ./tests/Makefile.am -1 +8 M ./tests/bench.c +5 M ./threads/Makefile.am -9 +6 M ./threads/api.c +56 M ./threads/threads.c -1 +1 Thu Jan 16 19:53:46 EST 2003 stevenj * [project @ 2003-01-17 00:53:46 by stevenj] whoops M ./api/f77api.c -2 +2 Thu Jan 16 19:53:30 EST 2003 stevenj * [project @ 2003-01-17 00:53:30 by stevenj] better name M ./api/f77api.c -2 +2 Thu Jan 16 19:52:36 EST 2003 stevenj * [project @ 2003-01-17 00:52:36 by stevenj] added more functions M ./api/f77api.c +68 Thu Jan 16 16:57:06 EST 2003 stevenj * [project @ 2003-01-16 21:57:06 by stevenj] if 'long' is big enough, use it for mulmod in preference to 'long long' M ./kernel/ifftw.h -1 +3 Thu Jan 16 14:53:41 EST 2003 stevenj * [project @ 2003-01-16 19:53:41 by stevenj] use uintptr_t for pointer alignment arithmetic M ./configure.ac -2 +9 M ./kernel/align.c -2 +2 M ./kernel/ifftw.h -5 +22 Thu Jan 16 07:58:28 EST 2003 athena * [project @ 2003-01-16 12:58:28 by athena] More signed/unsigned cleanup M ./kernel/planner.c -10 +10 M ./kernel/print.c -3 +3 M ./kernel/tensor.c -2 +2 M ./kernel/twiddle.c -2 +2 M ./rdft/problem.c -2 +2 Thu Jan 16 07:57:40 EST 2003 athena * [project @ 2003-01-16 12:57:40 by athena] null function pointers are technically nonportable M ./kernel/solvtab.c -2 +2 Thu Jan 16 07:17:45 EST 2003 athena * [project @ 2003-01-16 12:17:45 by athena] Free short_options M ./libbench/bench-main.c -2 +3 Thu Jan 16 05:48:30 EST 2003 athena * [project @ 2003-01-16 10:48:30 by athena] Oops, forgot STACK_FREE M ./kernel/alloc.c -15 +17 M ./kernel/ifftw.h -2 +3 M ./tests/bench.c -1 +1 Thu Jan 16 05:40:39 EST 2003 athena * [project @ 2003-01-16 10:40:39 by athena] Do not require memalign() unless HAVE_SIMD M ./kernel/alloc.c -6 +15 M ./kernel/ifftw.h -13 +18 Thu Jan 16 01:03:31 EST 2003 stevenj * [project @ 2003-01-16 06:03:31 by stevenj] MS VC++ _aligned_malloc M ./kernel/alloc.c -1 +7 Thu Jan 16 00:44:45 EST 2003 stevenj * [project @ 2003-01-16 05:44:45 by stevenj] added api fftw_malloc/free M ./api/fftw3.h -2 +5 M ./kernel/alloc.c -2 +10 Thu Jan 16 00:43:48 EST 2003 stevenj * [project @ 2003-01-16 05:43:48 by stevenj] silence warning M ./api/map-r2r-kind.c -1 +1 Wed Jan 15 22:39:04 EST 2003 stevenj * [project @ 2003-01-16 03:39:04 by stevenj] send error output to stderr M ./tools/fftw-wisdom-to-conf.in -1 +1 Wed Jan 15 13:20:35 EST 2003 athena * [project @ 2003-01-15 18:20:35 by athena] Pure paranoia. M ./kernel/tensor7.c -3 +6 Wed Jan 15 06:51:34 EST 2003 athena * [project @ 2003-01-15 11:51:34 by athena] Fixed formatting that was messed up by the conversion uint->int. Ensure that iodims etc are kosher. A ./kernel/tensor9.c M ./api/api.h -13 +20 M ./api/apiplan.c -9 +9 M ./api/configure.c -5 +5 M ./api/dfthelp.c -1 +1 M ./api/execute-dft-c2r.c -1 +1 M ./api/execute-dft-r2c.c -1 +1 M ./api/execute-dft.c -1 +1 M ./api/execute-r2r.c -1 +1 M ./api/execute.c -1 +1 M ./api/export-wisdom-to-file.c -4 +4 M ./api/export-wisdom-to-string.c -6 +6 M ./api/export-wisdom.c -6 +6 M ./api/f77api.c -58 +46 M ./api/fftw3.h -14 +11 M ./api/forget-wisdom.c -2 +2 M ./api/import-system-wisdom.c -2 +2 M ./api/import-wisdom-from-file.c -5 +5 M ./api/import-wisdom-from-string.c -4 +4 M ./api/import-wisdom.c -6 +6 M ./api/map-r2r-kind.c -37 +17 M ./api/mapflags.c -4 +4 M ./api/mkprinter-file.c -2 +2 M ./api/mktensor-iodims.c -2 +23 M ./api/mktensor-rowmajor.c -4 +20 M ./api/plan-dft-1d.c -2 +2 M ./api/plan-dft-2d.c -2 +2 M ./api/plan-dft-3d.c -3 +3 M ./api/plan-dft-c2r-1d.c -2 +2 M ./api/plan-dft-c2r-2d.c -2 +2 M ./api/plan-dft-c2r-3d.c -3 +3 M ./api/plan-dft-c2r.c -3 +3 M ./api/plan-dft-r2c-1d.c -2 +2 M ./api/plan-dft-r2c-2d.c -2 +2 M ./api/plan-dft-r2c-3d.c -3 +3 M ./api/plan-dft-r2c.c -3 +5 M ./api/plan-dft.c -4 +6 M ./api/plan-guru-dft-c2r.c -10 +10 M ./api/plan-guru-dft-r2c.c -10 +11 M ./api/plan-guru-dft.c -10 +11 M ./api/plan-guru-r2r.c -14 +16 M ./api/plan-many-dft-c2r.c -34 +21 M ./api/plan-many-dft-r2c.c -34 +21 M ./api/plan-many-dft.c -18 +19 M ./api/plan-many-r2r.c -22 +21 M ./api/plan-r2r-1d.c -2 +2 M ./api/plan-r2r-2d.c -3 +3 M ./api/plan-r2r-3d.c -4 +4 M ./api/plan-r2r.c -4 +4 M ./api/plan-with-nthreads.c -3 +3 M ./api/print-plan.c -3 +3 M ./api/rdft2-pad.c -2 +2 M ./api/the-planner.c -5 +5 M ./dft/buffered.c -2 +2 M ./dft/ct.c -2 +2 M ./dft/direct.c -2 +2 M ./dft/generic.c -1 +1 M ./dft/problem.c -4 +5 M ./dft/rader.c -1 +1 M ./dft/vrank-geq1.c -2 +2 M ./dft/vrank2-transpose.c -2 +2 M ./dft/vrank3-transpose.c -2 +2 M ./kernel/Makefile.am -2 +2 M ./kernel/alloc.c -2 +2 M ./kernel/ifftw.h -1 +2 M ./kernel/planner.c -10 +10 M ./kernel/print.c -3 +3 M ./kernel/tensor.c -2 +4 M ./kernel/tensor9.c +37 M ./rdft/buffered.c -2 +2 M ./rdft/buffered2.c -2 +2 M ./rdft/dft-r2hc.c -2 +2 M ./rdft/dht-r2hc.c -2 +2 M ./rdft/dht-rader.c -1 +1 M ./rdft/direct.c -2 +2 M ./rdft/direct2.c -2 +2 M ./rdft/generic.c -1 +1 M ./rdft/hc2hc.c -2 +2 M ./rdft/problem.c -2 +6 M ./rdft/problem2.c -2 +5 M ./rdft/rader-hc2hc.c -1 +1 M ./rdft/rdft-dht.c -2 +2 M ./rdft/rdft2-radix2.c -2 +2 M ./rdft/vrank-geq1-rdft2.c -2 +2 M ./rdft/vrank-geq1.c -2 +2 M ./rdft/vrank2-transpose.c -2 +2 M ./rdft/vrank3-transpose.c -2 +2 M ./reodft/redft00e-r2hc.c -2 +2 M ./reodft/reodft010e-r2hc.c -2 +2 M ./reodft/reodft11e-r2hc.c -2 +2 M ./reodft/rodft00e-r2hc.c -2 +2 M ./threads/dft-vrank-geq1.c -2 +2 M ./threads/rdft-vrank-geq1.c -2 +2 M ./threads/vrank-geq1-rdft2.c -2 +2 Wed Jan 15 01:32:18 EST 2003 stevenj * [project @ 2003-01-15 06:32:18 by stevenj] added version stamp M ./tools/fftw-wisdom-to-conf.in -1 +2 Wed Jan 15 01:28:20 EST 2003 stevenj * [project @ 2003-01-15 06:28:20 by stevenj] added warning M ./tools/fftw-wisdom-to-conf.in +1 Wed Jan 15 01:23:25 EST 2003 stevenj * [project @ 2003-01-15 06:23:25 by stevenj] add fftw-wisdom-to-conf to BUILT_SOURCES M ./tools/Makefile.am -1 +1 Wed Jan 15 01:09:29 EST 2003 stevenj * [project @ 2003-01-15 06:09:29 by stevenj] added const M ./tools/fftw-wisdom-to-conf.in -2 +2 Wed Jan 15 01:04:10 EST 2003 stevenj * [project @ 2003-01-15 06:04:10 by stevenj] added wisdom-to-conf A ./tools/ A ./tools/Makefile.am A ./tools/fftw-wisdom-to-conf.in M ./Makefile.am -1 +1 M ./configure.ac +3 M ./tools/Makefile.am +3 M ./tools/fftw-wisdom-to-conf.in +37 Wed Jan 15 00:23:36 EST 2003 stevenj * [project @ 2003-01-15 05:23:36 by stevenj] include type prefix in wisdom preamble M ./kernel/planner.c -2 +2 Wed Jan 15 00:02:31 EST 2003 stevenj * [project @ 2003-01-15 05:02:31 by stevenj] updates M ./TODO -4 +2 Tue Jan 14 23:59:26 EST 2003 stevenj * [project @ 2003-01-15 04:59:26 by stevenj] check the_plan before printing M ./tests/bench.c -2 +3 Tue Jan 14 21:10:25 EST 2003 athena * [project @ 2003-01-15 02:10:25 by athena] Eliminated those unsigned values that would break LP64 machines. M ./ChangeLog -1 +1 M ./api/api.h -16 +16 M ./api/apiplan.c -19 +19 M ./api/configure.c -5 +5 M ./api/dfthelp.c -5 +5 M ./api/execute-dft-c2r.c -1 +1 M ./api/execute-dft-r2c.c -1 +1 M ./api/execute-dft.c -1 +1 M ./api/execute-r2r.c -1 +1 M ./api/execute.c -1 +1 M ./api/export-wisdom-to-file.c -4 +4 M ./api/export-wisdom-to-string.c -13 +13 M ./api/export-wisdom.c -7 +7 M ./api/f77api.c -71 +69 M ./api/fftw3.h -134 +136 M ./api/forget-wisdom.c -2 +2 M ./api/import-system-wisdom.c -8 +8 M ./api/import-wisdom-from-file.c -6 +6 M ./api/import-wisdom-from-string.c -6 +6 M ./api/import-wisdom.c -9 +11 M ./api/map-r2r-kind.c -17 +40 M ./api/mapflags.c -18 +18 M ./api/mkprinter-file.c -3 +3 M ./api/mktensor-iodims.c -9 +9 M ./api/mktensor-rowmajor.c -6 +5 M ./api/plan-dft-1d.c -3 +2 M ./api/plan-dft-2d.c -5 +5 M ./api/plan-dft-3d.c -5 +7 M ./api/plan-dft-c2r-1d.c -2 +2 M ./api/plan-dft-c2r-2d.c -5 +5 M ./api/plan-dft-c2r-3d.c -5 +7 M ./api/plan-dft-c2r.c -6 +3 M ./api/plan-dft-r2c-1d.c -2 +2 M ./api/plan-dft-r2c-2d.c -5 +5 M ./api/plan-dft-r2c-3d.c -5 +7 M ./api/plan-dft-r2c.c -6 +3 M ./api/plan-dft.c -6 +4 M ./api/plan-guru-dft-c2r.c -10 +10 M ./api/plan-guru-dft-r2c.c -10 +10 M ./api/plan-guru-dft.c -10 +10 M ./api/plan-guru-r2r.c -13 +14 M ./api/plan-many-dft-c2r.c -23 +36 M ./api/plan-many-dft-r2c.c -23 +36 M ./api/plan-many-dft.c -17 +18 M ./api/plan-many-r2r.c -18 +22 M ./api/plan-r2r-1d.c -3 +2 M ./api/plan-r2r-2d.c -7 +8 M ./api/plan-r2r-3d.c -8 +11 M ./api/plan-r2r.c -3 +4 M ./api/plan-with-nthreads.c -7 +7 M ./api/print-plan.c -3 +3 M ./api/rdft2-pad.c -10 +9 M ./api/the-planner.c -7 +7 M ./configure.ac -9 +3 M ./dft/buffered.c -11 +11 M ./dft/codelet-dft.h -11 +11 M ./dft/codelets/n.c -1 +1 M ./dft/codelets/t.c -1 +1 M ./dft/ct-dif.c -4 +4 M ./dft/ct-dit.c -4 +4 M ./dft/ct-ditbuf.c -6 +6 M ./dft/ct-ditf.c -2 +2 M ./dft/ct.c -5 +5 M ./dft/ct.h -2 +2 M ./dft/dft.h -2 +2 M ./dft/direct.c -3 +3 M ./dft/generic.c -5 +5 M ./dft/indirect.c -2 +2 M ./dft/k7/k7.c -9 +9 M ./dft/problem.c -3 +3 M ./dft/rader-omega.c -2 +2 M ./dft/rader.c -13 +13 M ./dft/rank-geq2.c -9 +9 M ./dft/rank0.c -7 +7 M ./dft/simd/n1b.c -1 +1 M ./dft/simd/n1f.c -1 +1 M ./dft/simd/t1b.c -1 +1 M ./dft/simd/t1f.c -1 +1 M ./dft/vrank-geq1.c -11 +11 M ./dft/vrank2-transpose.c -4 +4 M ./dft/vrank3-transpose.c -10 +10 M ./dft/zero.c -3 +3 M ./genfft/gen_hc2hc.ml -4 +4 M ./genfft/gen_hc2r.ml -4 +4 M ./genfft/gen_notw.ml -4 +4 M ./genfft/gen_notw_c.ml -4 +4 M ./genfft/gen_r2hc.ml -4 +4 M ./genfft/gen_twiddle.ml -4 +4 M ./genfft/gen_twiddle_c.ml -4 +4 M ./genfft/gen_twidsq.ml -4 +4 M ./kernel/align.c -5 +5 M ./kernel/buffered.c -3 +3 M ./kernel/ct.c -1 +1 M ./kernel/hash.c -3 +3 M ./kernel/iabs.c -3 +3 M ./kernel/ifftw.h -71 +70 M ./kernel/md5-1.c -3 +3 M ./kernel/md5.c -4 +4 M ./kernel/minmax.c -3 +3 M ./kernel/ops.c -4 +4 M ./kernel/pickdim.c -6 +6 M ./kernel/planner.c -32 +32 M ./kernel/primes.c -14 +14 M ./kernel/print.c -5 +5 M ./kernel/rader.c -3 +3 M ./kernel/scan.c -3 +3 M ./kernel/tensor.c -9 +9 M ./kernel/tensor1.c -2 +2 M ./kernel/tensor2.c -3 +3 M ./kernel/tensor4.c -10 +10 M ./kernel/tensor5.c -6 +6 M ./kernel/tensor7.c -4 +4 M ./kernel/trig.c -4 +4 M ./kernel/twiddle.c -14 +14 M ./libbench/acopy.c -2 +2 M ./libbench/allocate.c -2 +2 M ./libbench/ascale.c -2 +2 M ./libbench/aset.c -2 +2 M ./libbench/bench-user.h -23 +23 M ./libbench/bench.h -2 +2 M ./libbench/caadd.c -2 +2 M ./libbench/cacopy.c -2 +2 M ./libbench/cascale.c -2 +2 M ./libbench/caset.c -2 +2 M ./libbench/casub.c -2 +2 M ./libbench/copy-c2h-1d-fftpack.c -1 +1 M ./libbench/copy-c2h-1d-halfcomplex.c -1 +1 M ./libbench/copy-c2h-1d-packed.c -2 +2 M ./libbench/copy-c2h-1d-unpacked-ri.c -1 +1 M ./libbench/copy-c2h-unpacked.c -1 +1 M ./libbench/copy-c2r-packed.c -2 +2 M ./libbench/copy-c2r-unpacked.c -1 +1 M ./libbench/copy-c2ri.c -2 +2 M ./libbench/copy-h2c-1d-fftpack.c -1 +1 M ./libbench/copy-h2c-1d-halfcomplex.c -1 +1 M ./libbench/copy-h2c-1d-packed.c -2 +2 M ./libbench/copy-h2c-1d-unpacked-ri.c -1 +1 M ./libbench/copy-h2c-unpacked.c -9 +9 M ./libbench/copy-r2c-packed.c -2 +2 M ./libbench/copy-r2c-unpacked.c -1 +1 M ./libbench/copy-ri2c.c -2 +2 M ./libbench/getopt-utils.c -2 +2 M ./libbench/getopt.c -3 +3 M ./libbench/log2.c -2 +2 M ./libbench/mp.c -20 +20 M ./libbench/pow2.c -1 +1 M ./libbench/prime.c -4 +4 M ./libbench/problem.c -2 +2 M ./libbench/timer.c -2 +2 M ./libbench/verify.c -36 +36 M ./rdft/buffered.c -10 +10 M ./rdft/buffered2.c -22 +22 M ./rdft/codelet-rdft.h -17 +17 M ./rdft/codelets/hc2r.c -1 +1 M ./rdft/codelets/hfb.c -1 +1 M ./rdft/codelets/r2hc.c -1 +1 M ./rdft/dft-r2hc.c -4 +4 M ./rdft/dht-r2hc.c -3 +3 M ./rdft/dht-rader.c -6 +6 M ./rdft/direct.c -6 +6 M ./rdft/direct2.c -5 +5 M ./rdft/generic.c -8 +8 M ./rdft/hc2hc-buf.c -7 +7 M ./rdft/hc2hc-dif.c -4 +4 M ./rdft/hc2hc-dit.c -4 +4 M ./rdft/hc2hc.c -5 +5 M ./rdft/hc2hc.h -2 +2 M ./rdft/indirect.c -2 +2 M ./rdft/problem.c -11 +11 M ./rdft/problem2.c -3 +3 M ./rdft/rader-hc2hc.c -12 +12 M ./rdft/rank-geq2-rdft2.c -9 +9 M ./rdft/rank-geq2.c -9 +9 M ./rdft/rank0-rdft2.c -4 +4 M ./rdft/rank0.c -5 +5 M ./rdft/rdft-dht.c -5 +5 M ./rdft/rdft.h -4 +4 M ./rdft/rdft2-inplace-strides.c -4 +4 M ./rdft/rdft2-radix2.c -15 +15 M ./rdft/rdft2-tensor-max-index.c -4 +4 M ./rdft/vrank-geq1-rdft2.c -12 +12 M ./rdft/vrank-geq1.c -11 +11 M ./rdft/vrank2-transpose.c -4 +4 M ./rdft/vrank3-transpose.c -10 +10 M ./reodft/redft00e-r2hc.c -7 +7 M ./reodft/reodft010e-r2hc.c -16 +16 M ./reodft/reodft11e-r2hc.c -10 +10 M ./reodft/rodft00e-r2hc.c -7 +7 M ./simd/3dnow.c -5 +5 M ./simd/sse.c -3 +3 M ./simd/sse2.c -3 +3 M ./tests/bench.c -13 +7 M ./tests/dotens.c -3 +3 M ./tests/dotens2.c -3 +3 M ./tests/trigtest.c -7 +7 M ./tests/verify-dft.c -4 +4 M ./tests/verify-lib.c -41 +41 M ./tests/verify-rdft.c -13 +13 M ./tests/verify-reodft.c -48 +48 M ./tests/verify.h -15 +15 M ./threads/ct-dit.c -5 +5 M ./threads/dft-vrank-geq1.c -15 +15 M ./threads/hc2hc-dif.c -6 +6 M ./threads/hc2hc-dit.c -6 +6 M ./threads/rdft-vrank-geq1.c -15 +15 M ./threads/threads.c -4 +4 M ./threads/threads.h -3 +3 M ./threads/vrank-geq1-rdft2.c -15 +15 Tue Jan 14 15:14:29 EST 2003 stevenj * [project @ 2003-01-14 20:14:29 by stevenj] comments M ./kernel/primes.c -4 +4 Tue Jan 14 08:00:08 EST 2003 athena * [project @ 2003-01-14 13:00:08 by athena] Oops M ./dft/generic.c -1 +1 M ./rdft/generic.c -1 +1 Tue Jan 14 07:59:14 EST 2003 athena * [project @ 2003-01-14 12:59:14 by athena] int/uint confusion M ./dft/generic.c -1 +2 M ./rdft/generic.c -1 +2 Tue Jan 14 02:25:33 EST 2003 stevenj * [project @ 2003-01-14 07:25:33 by stevenj] updated introduction and some organization M ./doc/fftw3.texi -79 +117 Tue Jan 14 01:34:46 EST 2003 stevenj * [project @ 2003-01-14 06:34:46 by stevenj] whoops M ./api/f77api.c -18 +18 Tue Jan 14 01:33:04 EST 2003 stevenj * [project @ 2003-01-14 06:33:04 by stevenj] newline M ./Makefile.am +1 Tue Jan 14 00:23:04 EST 2003 stevenj * [project @ 2003-01-14 05:23:04 by stevenj] added win32 timer M ./libbench/timer.c -1 +23 Tue Jan 14 00:12:21 EST 2003 stevenj * [project @ 2003-01-14 05:12:21 by stevenj] sync with kernel/alloc.c M ./libbench/util.c -2 +5 Tue Jan 14 00:03:20 EST 2003 stevenj * [project @ 2003-01-14 05:03:20 by stevenj] handle missing F77_FUNC_ M ./api/f77api.c +4 Mon Jan 13 17:42:50 EST 2003 stevenj * [project @ 2003-01-13 22:42:50 by stevenj] used fint instead of int to make Fortran integer type easier to change M ./api/f77api.c -24 +31 Mon Jan 13 17:38:56 EST 2003 stevenj * [project @ 2003-01-13 22:38:56 by stevenj] slight abbreviation M ./api/f77api.c -4 +4 Mon Jan 13 17:35:20 EST 2003 stevenj * [project @ 2003-01-13 22:35:20 by stevenj] the great lengthening, part I: int -> long in api; mv mktensor-rowmajor to api A ./api/mktensor-rowmajor.c M ./api/Makefile.am -2 +2 M ./api/api.h -2 +5 M ./api/f77api.c -7 +7 M ./api/fftw3.h -44 +46 M ./api/mktensor-rowmajor.c +44 M ./api/plan-dft-1d.c -1 +1 M ./api/plan-dft-2d.c -2 +2 M ./api/plan-dft-3d.c -2 +2 M ./api/plan-dft-c2r-1d.c -1 +1 M ./api/plan-dft-c2r-2d.c -2 +2 M ./api/plan-dft-c2r-3d.c -2 +2 M ./api/plan-dft-c2r.c -1 +1 M ./api/plan-dft-r2c-1d.c -1 +1 M ./api/plan-dft-r2c-2d.c -2 +2 M ./api/plan-dft-r2c-3d.c -2 +2 M ./api/plan-dft-r2c.c -1 +1 M ./api/plan-dft.c -1 +1 M ./api/plan-many-dft-c2r.c -7 +7 M ./api/plan-many-dft-r2c.c -7 +7 M ./api/plan-many-dft.c -6 +6 M ./api/plan-many-r2r.c -6 +6 M ./api/plan-r2r-1d.c -1 +1 M ./api/plan-r2r-2d.c -2 +2 M ./api/plan-r2r-3d.c -2 +2 M ./api/plan-r2r.c -1 +1 M ./api/rdft2-pad.c -4 +4 M ./kernel/Makefile.am -3 +3 M ./kernel/ifftw.h -4 +1 R ./kernel/tensor3.c M ./tests/bench.c -1 +9 Mon Jan 13 15:23:22 EST 2003 stevenj * [project @ 2003-01-13 20:23:22 by stevenj] long types M ./configure.ac +4 Mon Jan 13 04:20:37 EST 2003 athena * [project @ 2003-01-13 09:20:36 by athena] Renamed fftw_malloc -> MALLOC, X(free) -> X(ifree), X(free0) -> X(ifree0), non_fftw_malloc -> NATIVE_MALLOC M ./api/apiplan.c -2 +2 M ./api/export-wisdom-to-string.c -1 +1 M ./api/f77api.c -8 +8 M ./api/map-r2r-kind.c -1 +1 M ./api/plan-guru-r2r.c -1 +1 M ./api/plan-many-dft-c2r.c -2 +2 M ./api/plan-many-dft-r2c.c -2 +2 M ./api/plan-many-r2r.c -1 +1 M ./api/rdft2-pad.c -1 +1 M ./dft/buffered.c -6 +6 M ./dft/generic.c -1 +1 M ./dft/problem.c -2 +2 M ./dft/rader-omega.c -1 +1 M ./dft/rader.c -10 +10 M ./kernel/alloc.c -6 +6 M ./kernel/ifftw.h -11 +11 M ./kernel/plan.c -3 +3 M ./kernel/planner.c -12 +12 M ./kernel/print.c -3 +3 M ./kernel/problem.c -2 +2 M ./kernel/rader.c -3 +3 M ./kernel/scan.c -3 +3 M ./kernel/solver.c -3 +3 M ./kernel/stride.c -3 +3 M ./kernel/tensor.c -9 +9 M ./kernel/twiddle.c -5 +5 M ./rdft/buffered.c -6 +6 M ./rdft/buffered2.c -8 +8 M ./rdft/dht-rader.c -6 +6 M ./rdft/generic.c -1 +1 M ./rdft/problem.c -4 +4 M ./rdft/problem2.c -2 +2 M ./rdft/rader-hc2hc.c -10 +10 M ./reodft/redft00e-r2hc.c -5 +5 M ./reodft/reodft010e-r2hc.c -11 +11 M ./reodft/reodft11e-r2hc.c -7 +7 M ./reodft/rodft00e-r2hc.c -5 +5 M ./tests/verify-dft.c -15 +15 M ./tests/verify-rdft.c -29 +29 M ./tests/verify-reodft.c -15 +15 M ./threads/dft-vrank-geq1.c -4 +4 M ./threads/rdft-vrank-geq1.c -4 +4 M ./threads/threads.c -1 +1 M ./threads/vrank-geq1-rdft2.c -4 +4 Mon Jan 13 02:37:22 EST 2003 stevenj * [project @ 2003-01-13 07:37:22 by stevenj] added beginning of Fortran interface A ./api/f77api.c M ./api/Makefile.am -1 +1 M ./api/f77api.c +145 Mon Jan 13 01:05:29 EST 2003 stevenj * [project @ 2003-01-13 06:05:29 by stevenj] add fortran mangling check M ./configure.ac +18 Mon Jan 13 00:33:28 EST 2003 stevenj * [project @ 2003-01-13 05:33:28 by stevenj] added guru r2r interface A ./api/execute-r2r.c A ./api/plan-guru-r2r.c M ./api/Makefile.am -3 +4 M ./api/execute-r2r.c +29 M ./api/fftw3.h -1 +8 M ./api/plan-guru-r2r.c +42 Mon Jan 13 00:23:26 EST 2003 stevenj * [project @ 2003-01-13 05:23:26 by stevenj] whoops M ./api/fftw3.h -5 +7 M ./api/plan-r2r-1d.c -2 +2 M ./api/plan-r2r-2d.c -1 +4 M ./api/plan-r2r-3d.c -2 +5 Mon Jan 13 00:16:20 EST 2003 stevenj * [project @ 2003-01-13 05:16:20 by stevenj] added r2r planner A ./api/map-r2r-kind.c A ./api/plan-many-r2r.c A ./api/plan-r2r-1d.c A ./api/plan-r2r-2d.c A ./api/plan-r2r-3d.c A ./api/plan-r2r.c M ./api/Makefile.am -2 +3 M ./api/fftw3.h -1 +29 M ./api/map-r2r-kind.c +47 M ./api/plan-many-r2r.c +48 M ./api/plan-r2r-1d.c +27 M ./api/plan-r2r-2d.c +29 M ./api/plan-r2r-3d.c +30 M ./api/plan-r2r.c +27 Sun Jan 12 22:58:18 EST 2003 stevenj * [project @ 2003-01-13 03:58:18 by stevenj] more long-double checks M ./configure.ac -2 +13 Sun Jan 12 20:01:51 EST 2003 stevenj * [project @ 2003-01-13 01:01:51 by stevenj] slight regrouping M ./kernel/planner.c -5 +7 Sun Jan 12 19:58:46 EST 2003 stevenj * [project @ 2003-01-13 00:58:46 by stevenj] added joke M ./kernel/planner.c -2 +2 Sun Jan 12 19:53:58 EST 2003 stevenj * [project @ 2003-01-13 00:53:58 by stevenj] simplified rdft2 padding A ./api/rdft2-pad.c M ./api/Makefile.am -7 +7 M ./api/api.h -4 +2 R ./api/mktensor-rowmajor-pad.c M ./api/plan-many-dft-c2r.c -19 +16 M ./api/plan-many-dft-r2c.c -6 +18 M ./api/rdft2-pad.c +40 Sun Jan 12 19:02:09 EST 2003 stevenj * [project @ 2003-01-13 00:02:09 by stevenj] added comment M ./api/fftw3.h -2 +2 Sun Jan 12 18:54:49 EST 2003 stevenj * [project @ 2003-01-12 23:54:49 by stevenj] use latest api M ./tests/bench.c -2 +1 Sun Jan 12 18:49:58 EST 2003 stevenj * [project @ 2003-01-12 23:49:58 by stevenj] nembed should only be in advanced (many) interface, not basic interface...only a handful of people over the years have ever requested that functionality. M ./api/fftw3.h -10 +4 M ./api/plan-dft-1d.c -1 +1 M ./api/plan-dft-2d.c -1 +1 M ./api/plan-dft-3d.c -1 +1 M ./api/plan-dft-c2r-1d.c -1 +1 M ./api/plan-dft-c2r-2d.c -1 +1 M ./api/plan-dft-c2r-3d.c -1 +1 M ./api/plan-dft-c2r.c -5 +3 M ./api/plan-dft-r2c-1d.c -1 +1 M ./api/plan-dft-r2c-2d.c -1 +1 M ./api/plan-dft-r2c-3d.c -1 +1 M ./api/plan-dft-r2c.c -5 +3 M ./api/plan-dft.c -5 +3 Sun Jan 12 18:41:57 EST 2003 stevenj * [project @ 2003-01-12 23:41:57 by stevenj] impatient is default; generalize mapping functions using xor trick M ./api/fftw3.h -2 +2 M ./api/mapflags.c -55 +61 Sun Jan 12 14:39:42 EST 2003 stevenj * [project @ 2003-01-12 19:39:41 by stevenj] use NULL nembed to signal padding M ./api/mktensor-rowmajor-pad.c -5 +9 M ./api/plan-dft-c2r-1d.c -1 +1 M ./api/plan-dft-c2r-2d.c -1 +1 M ./api/plan-dft-c2r-3d.c -1 +1 M ./api/plan-dft-r2c-1d.c -1 +1 M ./api/plan-dft-r2c-2d.c -1 +1 M ./api/plan-dft-r2c-3d.c -1 +1 Sun Jan 12 14:23:00 EST 2003 stevenj * [project @ 2003-01-12 19:23:00 by stevenj] accept NULL nembed M ./api/plan-many-dft.c -1 +4 Sun Jan 12 13:57:13 EST 2003 stevenj * [project @ 2003-01-12 18:57:13 by stevenj] added execute-dft-r2c/c2r A ./api/execute-dft-c2r.c A ./api/execute-dft-r2c.c M ./api/Makefile.am -5 +6 M ./api/execute-dft-c2r.c +29 M ./api/execute-dft-r2c.c +29 M ./api/fftw3.h -1 +4 Sun Jan 12 13:43:20 EST 2003 stevenj * [project @ 2003-01-12 18:43:20 by stevenj] don't need dft.h M ./api/plan-dft.c -1 Sun Jan 12 13:22:14 EST 2003 stevenj * [project @ 2003-01-12 18:22:14 by stevenj] tensors are compressed in the problem, duh M ./api/plan-many-dft-c2r.c -2 +1 M ./api/plan-many-dft-r2c.c -2 +1 Sun Jan 12 12:45:26 EST 2003 stevenj * [project @ 2003-01-12 17:45:26 by stevenj] noted that posix_memalign bug is now fixed, thanks to bug report by yours truly M ./kernel/alloc.c -2 +3 Sun Jan 12 12:44:43 EST 2003 athena * [project @ 2003-01-12 17:44:43 by athena] Bug: n[3] instead of n[2]. Bug was propagated by copy-and-paste. Grrr... M ./api/plan-dft-3d.c -1 +1 M ./api/plan-dft-c2r-3d.c -1 +1 M ./api/plan-dft-r2c-3d.c -1 +1 Sun Jan 12 12:41:43 EST 2003 athena * [project @ 2003-01-12 17:41:43 by athena] Express plan_dft() in terms of plan_many_dft() M ./api/plan-dft.c -11 +4 Sun Jan 12 12:19:53 EST 2003 stevenj * [project @ 2003-01-12 17:19:38 by stevenj] whoops A ./api/plan-guru-dft-c2r.c A ./api/plan-guru-dft-r2c.c A ./api/plan-guru-dft.c M ./api/plan-guru-dft-c2r.c +36 M ./api/plan-guru-dft-r2c.c +35 M ./api/plan-guru-dft.c +35 Sun Jan 12 06:00:46 EST 2003 athena * [project @ 2003-01-12 11:00:46 by athena] Manual skeleton. A ./doc/ A ./doc/Makefile.am A ./doc/fftw3.texi M ./Makefile.am -1 +1 M ./configure.ac +1 M ./doc/Makefile.am -1 +2 M ./doc/fftw3.texi +318 M ./genfft-k7/vK7Optimization.ml -1 +1 Sat Jan 11 23:46:34 EST 2003 stevenj * [project @ 2003-01-12 04:46:34 by stevenj] added r2c/c2r guru api M ./api/Makefile.am -1 +2 M ./api/fftw3.h -1 +12 Sat Jan 11 23:42:10 EST 2003 stevenj * [project @ 2003-01-12 04:42:10 by stevenj] FFTW_DESTROY_INPUT is default for c2r transforms M ./api/plan-many-dft-c2r.c -1 +1 Sat Jan 11 23:36:26 EST 2003 stevenj * [project @ 2003-01-12 04:36:26 by stevenj] added more of r2c/c2r api A ./api/plan-dft-c2r-1d.c A ./api/plan-dft-c2r-2d.c A ./api/plan-dft-c2r-3d.c A ./api/plan-dft-c2r.c A ./api/plan-dft-r2c-1d.c A ./api/plan-dft-r2c-2d.c A ./api/plan-dft-r2c-3d.c A ./api/plan-dft-r2c.c M ./api/Makefile.am -1 +3 M ./api/fftw3.h -69 +90 M ./api/plan-dft-c2r-1d.c +26 M ./api/plan-dft-c2r-2d.c +29 M ./api/plan-dft-c2r-3d.c +29 M ./api/plan-dft-c2r.c +32 M ./api/plan-dft-r2c-1d.c +26 M ./api/plan-dft-r2c-2d.c +29 M ./api/plan-dft-r2c-3d.c +29 M ./api/plan-dft-r2c.c +32 Sat Jan 11 21:09:41 EST 2003 stevenj * [project @ 2003-01-12 02:09:41 by stevenj] r2c doesn't have adjustible sign M ./api/fftw3.h -3 +3 M ./api/plan-many-dft-c2r.c -2 +2 M ./api/plan-many-dft-r2c.c -2 +2 Sat Jan 11 21:07:55 EST 2003 stevenj * [project @ 2003-01-12 02:07:55 by stevenj] note that copyright year is out of date M ./TODO -1 +1 Sat Jan 11 21:04:23 EST 2003 stevenj * [project @ 2003-01-12 02:04:23 by stevenj] updated api for r2c M ./api/fftw3.h -1 +17 Sat Jan 11 21:00:07 EST 2003 stevenj * [project @ 2003-01-12 02:00:07 by stevenj] removed annoying nophys == niphys case M ./api/mktensor-rowmajor-pad.c -11 Sat Jan 11 20:58:13 EST 2003 stevenj * [project @ 2003-01-12 01:58:13 by stevenj] added basic r2c/c2r planner A ./api/mktensor-rowmajor-pad.c A ./api/plan-many-dft-c2r.c A ./api/plan-many-dft-r2c.c M ./api/Makefile.am -2 +3 M ./api/api.h +5 M ./api/mktensor-rowmajor-pad.c +88 M ./api/plan-many-dft-c2r.c +59 M ./api/plan-many-dft-r2c.c +44 Sat Jan 11 19:34:14 EST 2003 stevenj * [project @ 2003-01-12 00:34:14 by stevenj] dist should be in terms of complex values M ./api/plan-many-dft.c -1 +1 Sat Jan 11 19:14:24 EST 2003 stevenj * [project @ 2003-01-12 00:14:24 by stevenj] added plan-with-nthreads M ./api/fftw3.h -2 +5 Sat Jan 11 19:12:51 EST 2003 stevenj * [project @ 2003-01-12 00:12:51 by stevenj] added function to set nthr A ./api/plan-with-nthreads.c M ./api/Makefile.am -1 +1 M ./api/plan-with-nthreads.c +39 Sat Jan 11 18:04:57 EST 2003 stevenj * [project @ 2003-01-11 23:04:57 by stevenj] slight cleanup M ./api/fftw3.h -4 +5 Sat Jan 11 17:57:29 EST 2003 stevenj * [project @ 2003-01-11 22:57:29 by stevenj] whoops A ./api/mktensor-iodims.c Sat Jan 11 17:55:39 EST 2003 stevenj * [project @ 2003-01-11 22:55:39 by stevenj] maxlen is maximum string length, not including null termination M ./kernel/scan.c -2 +2 Sat Jan 11 17:50:49 EST 2003 stevenj * [project @ 2003-01-11 22:50:49 by stevenj] imprt reverts hashtable on failure M ./kernel/planner.c -3 +14 Sat Jan 11 16:43:54 EST 2003 stevenj * [project @ 2003-01-11 21:43:54 by stevenj] slight move M ./api/fftw3.h -14 +14 Sat Jan 11 16:34:56 EST 2003 stevenj * [project @ 2003-01-11 21:34:56 by stevenj] stdio.h should be inlcuded outside of extern "C" M ./api/fftw3.h -3 +3 Sat Jan 11 16:26:35 EST 2003 stevenj * [project @ 2003-01-11 21:26:35 by stevenj] added guru planner API M ./api/Makefile.am -2 +3 M ./api/api.h +1 M ./api/fftw3.h -1 +15 Sat Jan 11 15:54:57 EST 2003 stevenj * [project @ 2003-01-11 20:54:57 by stevenj] added FFTW_FORWARD/BACKWARD M ./api/fftw3.h -1 +4 Sat Jan 11 15:52:17 EST 2003 stevenj * [project @ 2003-01-11 20:52:17 by stevenj] added plan_many_dft A ./api/plan-many-dft.c M ./api/Makefile.am -1 +1 M ./api/fftw3.h -1 +9 M ./api/plan-many-dft.c +43 Sat Jan 11 15:44:37 EST 2003 stevenj * [project @ 2003-01-11 20:44:37 by stevenj] indenting M ./kernel/tensor3.c -3 +3 Sat Jan 11 14:49:08 EST 2003 athena * [project @ 2003-01-11 19:49:08 by athena] Final \n M ./tests/bench.c -2 +4 Sat Jan 11 14:47:31 EST 2003 athena * [project @ 2003-01-11 19:47:31 by athena] Do not compile if not defined(FFTW_DEBUG), in order to avoid unused code in the shared library. M ./kernel/debug.c -1 +3 Sat Jan 11 14:45:56 EST 2003 athena * [project @ 2003-01-11 19:45:56 by athena] Implemented print_plan() A ./api/mkprinter-file.c A ./api/print-plan.c M ./api/Makefile.am -4 +4 M ./api/api.h +1 M ./api/export-wisdom-to-file.c -19 +1 M ./api/fftw3.h -2 +3 M ./api/mkprinter-file.c +39 M ./api/print-plan.c +29 M ./tests/bench.c -2 +3 Sat Jan 11 13:12:01 EST 2003 stevenj * [project @ 2003-01-11 18:12:01 by stevenj] changed the OOP-like plan_destroy to the more-grammatical destroy_plan M ./api/apiplan.c -1 +1 M ./api/fftw3.h -2 +2 M ./tests/bench.c -1 +1 Sat Jan 11 12:58:04 EST 2003 stevenj * [project @ 2003-01-11 17:55:49 by stevenj] added guru execute_dft A ./api/execute-dft.c M ./api/Makefile.am -4 +5 M ./api/execute-dft.c +29 M ./api/fftw3.h -2 +3 Sat Jan 11 12:38:40 EST 2003 stevenj * [project @ 2003-01-11 17:38:40 by stevenj] allow for malloc errors in wisdom string, since non-fftw-malloc M ./api/export-wisdom-to-string.c -4 +5 Sat Jan 11 12:16:05 EST 2003 stevenj * [project @ 2003-01-11 17:16:05 by stevenj] cleanup should reset plnr to zero so that fftw can be restarted M ./api/the-planner.c -1 +3 Sat Jan 11 12:13:18 EST 2003 stevenj * [project @ 2003-01-11 17:13:18 by stevenj] NO_UGLY is an internal planner flag M ./api/fftw3.h -4 +3 M ./api/mapflags.c -3 Sat Jan 11 11:23:13 EST 2003 athena * [project @ 2003-01-11 16:23:13 by athena] Written 1d api in terms of generic n-d api. The code is less compact but easier to test M ./api/plan-dft-1d.c -9 +1 M ./tests/bench.c -27 +36 Sat Jan 11 11:07:25 EST 2003 athena * [project @ 2003-01-11 16:07:24 by athena] Added wisdom to header file, made scanners/printer static. stdio.h no longer needed in fftw.h, removed. Probably the printer_file should be reintroduced in a separate file if we ever want to print plans... M ./api/export-wisdom-to-file.c -2 +2 M ./api/export-wisdom-to-string.c -4 +4 M ./api/fftw3.h -2 +13 M ./api/import-wisdom-from-file.c -2 +2 M ./api/import-wisdom-from-string.c -2 +2 M ./kernel/alloc.c -1 +3 M ./kernel/assert.c -2 +2 M ./kernel/debug.c -2 +22 M ./kernel/ifftw.h -15 +1 M ./kernel/print.c -1 +2 M ./kernel/scan.c -1 +2 M ./tests/bench.c -13 +49 M ./tests/verify-lib.c -1 +2 M ./tests/verify-reodft.c -1 +2 Sat Jan 11 09:49:30 EST 2003 athena * [project @ 2003-01-11 14:49:30 by athena] Implemented more APIs A ./api/plan-dft-2d.c A ./api/plan-dft-3d.c A ./api/plan-dft.c M ./api/Makefile.am -6 +5 M ./api/apiplan.c -1 +1 M ./api/fftw3.h -13 +25 M ./api/plan-dft-2d.c +30 M ./api/plan-dft-3d.c +30 M ./api/plan-dft.c +40 M ./tests/bench.c +18 Sat Jan 11 09:21:53 EST 2003 athena * [project @ 2003-01-11 14:21:53 by athena] Added cleanup() to API M ./api/fftw3.h -3 +3 M ./api/the-planner.c +6 M ./tests/bench.c +1 Sat Jan 11 09:17:34 EST 2003 athena * [project @ 2003-01-11 14:17:34 by athena] Started new bench.c. I had to rename plan_destroy -> plan_destroy_internal to avoid conflicts with API M ./api/api.h -1 M ./api/apiplan.c -3 +3 M ./api/fftw3.h -1 +3 M ./dft/buffered.c -7 +7 M ./dft/ct.c -2 +2 M ./dft/generic.c -2 +2 M ./dft/indirect.c -5 +5 M ./dft/rader.c -8 +8 M ./dft/rank-geq2.c -5 +5 M ./dft/vrank-geq1.c -2 +2 M ./kernel/ifftw.h -2 +2 M ./kernel/plan.c -2 +2 M ./kernel/planner.c -3 +3 M ./libbench/bench-user.h -1 +4 M ./libbench/bench.h -3 +1 M ./rdft/buffered.c -7 +7 M ./rdft/buffered2.c -5 +5 M ./rdft/dft-r2hc.c -2 +2 M ./rdft/dht-r2hc.c -2 +2 M ./rdft/dht-rader.c -6 +6 M ./rdft/generic.c -2 +2 M ./rdft/hc2hc.c -7 +7 M ./rdft/indirect.c -5 +5 M ./rdft/rader-hc2hc.c -7 +7 M ./rdft/rank-geq2-rdft2.c -5 +5 M ./rdft/rank-geq2.c -5 +5 M ./rdft/rank0-rdft2.c -2 +2 M ./rdft/rdft-dht.c -2 +2 M ./rdft/rdft2-radix2.c -2 +2 M ./rdft/vrank-geq1-rdft2.c -2 +2 M ./rdft/vrank-geq1.c -2 +2 M ./reodft/redft00e-r2hc.c -2 +2 M ./reodft/reodft010e-r2hc.c -2 +2 M ./reodft/reodft11e-r2hc.c -2 +2 M ./reodft/rodft00e-r2hc.c -2 +2 M ./tests/Makefile.am -1 +1 M ./tests/bench.c -2 +103 M ./threads/dft-vrank-geq1.c -3 +3 M ./threads/rdft-vrank-geq1.c -3 +3 M ./threads/vrank-geq1-rdft2.c -3 +3 Sat Jan 11 02:45:39 EST 2003 stevenj * [project @ 2003-01-11 07:45:39 by stevenj] fix types A ./api/import-system-wisdom.c M ./api/Makefile.am -3 +3 M ./api/export-wisdom.c -2 +2 M ./api/import-system-wisdom.c +46 M ./api/import-wisdom-from-file.c -2 +3 M ./api/import-wisdom-from-string.c -3 +4 M ./api/import-wisdom.c -3 +5 Sat Jan 11 02:13:25 EST 2003 stevenj * [project @ 2003-01-11 07:13:25 by stevenj] whoops M ./api/export-wisdom-to-string.c -1 +2 Sat Jan 11 02:10:50 EST 2003 stevenj * [project @ 2003-01-11 07:10:50 by stevenj] added wisdom api A ./api/export-wisdom-to-file.c A ./api/export-wisdom-to-string.c A ./api/export-wisdom.c A ./api/forget-wisdom.c A ./api/import-wisdom-from-file.c A ./api/import-wisdom-from-string.c A ./api/import-wisdom.c M ./api/Makefile.am -2 +6 M ./api/export-wisdom-to-file.c +47 M ./api/export-wisdom-to-string.c +80 M ./api/export-wisdom.c +44 M ./api/forget-wisdom.c +27 M ./api/import-wisdom-from-file.c +47 M ./api/import-wisdom-from-string.c +49 M ./api/import-wisdom.c +44 M ./kernel/Makefile.am -4 +4 M ./kernel/ifftw.h -1 +3 R ./kernel/printers.c R ./kernel/scanners.c Sat Jan 11 01:01:17 EST 2003 stevenj * [project @ 2003-01-11 06:01:17 by stevenj] grammar M ./api/mapflags.c -1 +1 Sat Jan 11 00:54:54 EST 2003 stevenj * [project @ 2003-01-11 05:54:54 by stevenj] slight change M ./api/mapflags.c -1 +1 Sat Jan 11 00:52:04 EST 2003 stevenj * [project @ 2003-01-11 05:52:04 by stevenj] implemented api/mapflags M ./api/fftw3.h -1 +23 M ./api/mapflags.c -1 +83 Sat Jan 11 00:48:27 EST 2003 stevenj * [project @ 2003-01-11 05:48:27 by stevenj] IMPATIENT is an api issue M ./kernel/ifftw.h -11 +1 Fri Jan 10 01:57:41 EST 2003 stevenj * [project @ 2003-01-10 06:57:41 by stevenj] removed un-needed headers M ./api/the-planner.c -4 Fri Jan 10 01:56:59 EST 2003 stevenj * [project @ 2003-01-10 06:56:59 by stevenj] mkplanner initializes nthr to 1 already M ./api/the-planner.c -1 Thu Jan 9 18:53:09 EST 2003 stevenj * [project @ 2003-01-09 23:53:09 by stevenj] boilerplate M ./api/fftw3.h -1 +14 Thu Jan 9 18:16:39 EST 2003 stevenj * [project @ 2003-01-09 23:16:39 by stevenj] fold vecloop into r{e,o}dft apply function to share buffer, etcetera M ./rdft/vrank-geq1.c -1 +6 M ./reodft/redft00e-r2hc.c -42 +45 M ./reodft/reodft010e-r2hc.c -138 +153 M ./reodft/reodft11e-r2hc.c -98 +107 M ./reodft/rodft00e-r2hc.c -39 +42 Thu Jan 9 18:10:19 EST 2003 stevenj * [project @ 2003-01-09 23:10:19 by stevenj] whoops, bugfix in impulse test for vecn > 1 M ./tests/verify-reodft.c -2 +2 Thu Jan 9 14:23:51 EST 2003 stevenj * [project @ 2003-01-09 19:23:51 by stevenj] bugfix, grr M ./rdft/hc2hc-buf.c -2 +2 Thu Jan 9 14:21:16 EST 2003 stevenj * [project @ 2003-01-09 19:21:16 by stevenj] fixed signed-ness enum problem M ./rdft/codelet-rdft.h -2 +5 Thu Jan 9 14:12:42 EST 2003 athena * [project @ 2003-01-09 19:12:42 by athena] Explicit cast M ./kernel/md5-1.c -1 +1 Thu Jan 9 13:41:51 EST 2003 athena * [project @ 2003-01-09 18:41:51 by athena] Added configure_planner(). mkplan() behaves properly when plan is null. A ./api/configure.c M ./api/Makefile.am -2 +2 M ./api/api.h +1 M ./api/apiplan.c -14 +21 M ./api/configure.c +33 M ./api/fftw3.h -3 +3 M ./api/the-planner.c -4 +1 Thu Jan 9 06:48:53 EST 2003 athena * [project @ 2003-01-09 11:48:53 by athena] More API work A ./api/apiplan.c A ./api/execute.c A ./api/mapflags.c M ./api/Makefile.am -4 +2 M ./api/api.h -2 +3 M ./api/apiplan.c +60 M ./api/execute.c +27 M ./api/fftw3.h -7 +3 M ./api/mapflags.c +26 M ./api/plan-dft-1d.c -11 +4 M ./tests/bench.c -1 +1 Thu Jan 9 05:40:34 EST 2003 athena * [project @ 2003-01-09 10:40:34 by athena] First skeleton of API infrastructure A ./api/ A ./api/Makefile.am A ./api/api.h A ./api/dfthelp.c A ./api/fftw3.h A ./api/plan-dft-1d.c A ./api/the-planner.c M ./Makefile.am -5 +7 M ./api/Makefile.am +10 M ./api/api.h +51 M ./api/dfthelp.c +37 M ./api/fftw3.h +67 M ./api/plan-dft-1d.c +43 M ./api/the-planner.c +42 M ./configure.ac +2 M ./kernel/ifftw.h -1 +2 M ./kernel/trig.c -3 +1 M ./tests/Makefile.am -1 +1 Thu Jan 9 03:19:35 EST 2003 stevenj * [project @ 2003-01-09 08:19:35 by stevenj] unsigned strikes again M ./rdft/rdft2-tensor-max-index.c -3 +3 Thu Jan 9 01:51:45 EST 2003 stevenj * [project @ 2003-01-09 06:51:45 by stevenj] put rdft2_inplace_strides and rdft2_tensor_max_index in their own files for tighter linking A ./rdft/rdft2-inplace-strides.c A ./rdft/rdft2-tensor-max-index.c M ./rdft/Makefile.am -1 +2 M ./rdft/problem2.c -46 +1 M ./rdft/rdft2-inplace-strides.c +68 M ./rdft/rdft2-tensor-max-index.c +47 M ./rdft/vrank-geq1-rdft2.c -25 +1 Thu Jan 9 01:43:13 EST 2003 stevenj * [project @ 2003-01-09 06:43:13 by stevenj] added rdft2_tensor_max_index...incorrect use of tensor_max_index was preventing proper loop ordering for rnk > 2 rdft2 M ./rdft/rank-geq2-rdft2.c -2 +3 M ./rdft/rdft.h -1 +2 M ./rdft/vrank-geq1-rdft2.c -4 +27 Thu Jan 9 00:44:45 EST 2003 stevenj * [project @ 2003-01-09 05:44:45 by stevenj] arbitrary spltrnk in rdft2 rank-geq2 M ./rdft/rank-geq2-rdft2.c -11 +53 Thu Jan 9 00:40:17 EST 2003 stevenj * [project @ 2003-01-09 05:40:17 by stevenj] don't mention wisdom when non-verbose M ./tests/bench.c -2 +4 Thu Jan 9 00:02:35 EST 2003 stevenj * [project @ 2003-01-09 05:02:35 by stevenj] bug fix: printing %T should pass tensor *, not tensor ** M ./dft/problem.c -3 +3 M ./rdft/problem.c -3 +3 M ./rdft/problem2.c -3 +3 Wed Jan 8 23:40:48 EST 2003 stevenj * [project @ 2003-01-09 04:40:48 by stevenj] correct(?) normalization for rodft00 ... all of the even/odd transforms should be normalized according to the expanded'' DFT of ~twice the length M ./reodft/rodft00e-r2hc.c -4 +4 M ./tests/verify-reodft.c -2 +2 Wed Jan 8 23:18:23 EST 2003 stevenj * [project @ 2003-01-09 04:18:23 by stevenj] fixed tests for n=1 M ./tests/verify-reodft.c -3 +6 Wed Jan 8 22:10:08 EST 2003 stevenj * [project @ 2003-01-09 03:10:08 by stevenj] fixed bug in vector tests for rdft(2) M ./tests/bench.c -3 +5 Wed Jan 8 20:12:00 EST 2003 stevenj * [project @ 2003-01-09 01:12:00 by stevenj] fixed handling when first rnk-1 dimensions compress to nothing (ugh) M ./rdft/problem2.c -3 +6 Wed Jan 8 20:02:35 EST 2003 stevenj * [project @ 2003-01-09 01:02:35 by stevenj] fixed incorrect/missing rdft2 rank-0 handling A ./rdft/rank0-rdft2.c M ./rdft/Makefile.am -2 +2 M ./rdft/conf.c -1 +2 M ./rdft/nop2.c -3 +6 M ./rdft/rank0-rdft2.c +194 M ./rdft/rdft.h -1 +2 Wed Jan 8 19:49:05 EST 2003 stevenj * [project @ 2003-01-09 00:49:05 by stevenj] bug fix: for rnk > 1, must compress rnk-1 dims separately (ugh) M ./rdft/problem2.c -2 +11 Wed Jan 8 17:39:14 EST 2003 stevenj * [project @ 2003-01-08 22:39:14 by stevenj] added trailing newline M ./configure.ac +1 Wed Jan 8 17:38:02 EST 2003 stevenj * [project @ 2003-01-08 22:38:02 by stevenj] updated M ./ChangeLog +66 Wed Jan 8 16:53:16 EST 2003 stevenj * [project @ 2003-01-08 21:53:16 by stevenj] got rid of compiler warning M ./rdft/problem.c -2 +2 Wed Jan 8 16:49:48 EST 2003 stevenj * [project @ 2003-01-08 21:49:48 by stevenj] whoops, test r2hc and not rodft00 by default M ./tests/bench.c -1 +1 Wed Jan 8 16:46:24 EST 2003 stevenj * [project @ 2003-01-08 21:46:24 by stevenj] got rid of real_n...use physical n everywhere in rdft; fixed rdft sz compression; fixed rodft00 verify bug M ./rdft/buffered.c -3 +2 M ./rdft/indirect.c -10 +6 M ./rdft/problem.c -44 +43 M ./rdft/rank-geq2.c -10 +6 M ./rdft/rdft.h -3 +1 M ./reodft/redft00e-r2hc.c -4 +5 M ./reodft/reodft010e-r2hc.c -4 +3 M ./reodft/reodft11e-r2hc.c -4 +3 M ./reodft/rodft00e-r2hc.c -4 +3 M ./tests/bench.c -2 +2 M ./tests/verify-reodft.c -33 +35 Wed Jan 8 07:20:47 EST 2003 athena * [project @ 2003-01-08 12:20:47 by athena] icc-6.0 bug workaround A ./simd/sse-aux.c A ./simd/sse2-aux.c M ./simd/Makefile.am -2 +2 M ./simd/sse-aux.c +33 M ./simd/sse.c -3 +1 M ./simd/sse2-aux.c +34 M ./simd/sse2.c -3 +1 Wed Jan 8 04:21:40 EST 2003 athena * [project @ 2003-01-08 09:21:40 by athena] Reclaimed the fftw_real identifier, because I need it for the API M ./kernel/ifftw.h -12 +9 M ./rdft/buffered2.c -5 +5 M ./rdft/rader-hc2hc.c -1 +1 M ./tests/bench.c +1 Wed Jan 8 04:14:55 EST 2003 athena * [project @ 2003-01-08 09:14:55 by athena] Use recommended AC_OUTPUT syntax M ./configure.ac -1 +2 Wed Jan 8 04:00:22 EST 2003 athena * [project @ 2003-01-08 09:00:22 by athena] Removed FFTW(foo) as a synonym for X(foo). This is an API issue. M ./kernel/ifftw.h -8 +7 M ./tests/bench.c +1 Tue Jan 7 17:45:52 EST 2003 stevenj * [project @ 2003-01-07 22:45:52 by stevenj] get rid of warning M ./simd/sse2.c -1 +2 Tue Jan 7 16:22:39 EST 2003 athena * [project @ 2003-01-07 21:22:39 by athena] Renamed conflicting files */codelet.h into dft/codelet-dft.h and rdft/codelet-rdft.h A ./dft/codelet-dft.h A ./rdft/codelet-rdft.h A ./support/codelet_prelude.dft A ./support/codelet_prelude.rdft M ./dft/Makefile.am -1 +1 M ./dft/codelet-dft.h +113 R ./dft/codelet.h M ./dft/codelets/inplace/Makefile.am -4 +4 M ./dft/codelets/n.c -1 +1 M ./dft/codelets/standard/Makefile.am -3 +3 M ./dft/codelets/t.c -1 +1 M ./dft/dft.h -2 +2 M ./dft/simd/codelets/Makefile.am -6 +6 M ./dft/simd/n1b.c -1 +1 M ./dft/simd/n1f.c -1 +1 M ./dft/simd/t1b.c -1 +1 M ./dft/simd/t1f.c -1 +1 M ./rdft/Makefile.am -5 +5 M ./rdft/codelet-rdft.h +190 R ./rdft/codelet.h M ./rdft/codelets/hc2r/Makefile.am -4 +4 M ./rdft/codelets/hc2r.c -1 +1 M ./rdft/codelets/hfb.c -1 +1 M ./rdft/codelets/r2hc/Makefile.am -4 +4 M ./rdft/codelets/r2hc.c -1 +1 M ./rdft/rdft.h -2 +2 M ./support/Makefile.am -1 +2 M ./support/Makefile.codelets -2 +4 R ./support/codelet_prelude M ./support/codelet_prelude.dft +8 M ./support/codelet_prelude.rdft +8 Tue Jan 7 16:21:16 EST 2003 stevenj * [project @ 2003-01-07 21:21:16 by stevenj] updated M ./ChangeLog +737 Tue Jan 7 15:47:24 EST 2003 athena * [project @ 2003-01-07 20:47:24 by athena] Silence warnings M ./simd/simd-3dnow.h -1 +1 M ./simd/simd-sse.h -1 M ./simd/simd-sse2.h -1 +1 M ./simd/sse2.c -2 +1 Tue Jan 7 15:00:14 EST 2003 stevenj * [project @ 2003-01-07 20:00:14 by stevenj] fftw2 used spltrnk=1 M ./dft/rank-geq2.c -2 +4 M ./rdft/rank-geq2.c -2 +4 Tue Jan 7 14:32:06 EST 2003 athena * [project @ 2003-01-07 19:32:06 by athena] Silence warning M ./dft/codelet.h -2 +1 M ./rdft/codelet.h -2 +1 M ./simd/simd-sse.h -1 +2 M ./simd/sse.c -2 +1 Tue Jan 7 12:13:50 EST 2003 stevenj * [project @ 2003-01-07 17:13:50 by stevenj] noted deficiency M ./TODO +2 Tue Jan 7 07:18:51 EST 2003 athena * [project @ 2003-01-07 12:18:51 by athena] Strengthened conditions for a problem to be POSSIBLY_UNALIGNED M ./rdft/vrank-geq1-rdft2.c -4 +2 M ./rdft/vrank-geq1.c -2 +2 Tue Jan 7 05:09:42 EST 2003 athena * [project @ 2003-01-07 10:09:42 by athena] Strengthened conditions for a plan to be POSSIBLY_UNALIGNED M ./dft/vrank-geq1.c -5 +2 M ./kernel/align.c -1 +6 M ./kernel/ifftw.h -1 +2 Sun Jan 5 02:43:45 EST 2003 stevenj * [project @ 2003-01-05 07:43:45 by stevenj] added copyright todo M ./TODO +2 Sun Jan 5 02:37:31 EST 2003 stevenj * [project @ 2003-01-05 07:37:31 by stevenj] modified comment M ./kernel/planner.c -2 +3 Sun Jan 5 02:34:36 EST 2003 stevenj * [project @ 2003-01-05 07:33:41 by stevenj] fixed comment M ./tests/verify-rdft.c -3 +3 Sun Jan 5 02:31:56 EST 2003 stevenj * [project @ 2003-01-05 07:31:56 by stevenj] implemented rdft2 verify M ./TODO -4 +2 M ./tests/verify-rdft.c -5 +15 Sat Jan 4 16:20:42 EST 2003 stevenj * [project @ 2003-01-04 21:20:42 by stevenj] fix --enable-single M ./configure.ac -1 +1 Wed Oct 23 12:59:12 EDT 2002 stevenj * [project @ 2002-10-23 16:59:12 by stevenj] slight fixes M ./threads/threads.c -20 +14 Wed Oct 23 12:42:39 EDT 2002 stevenj * [project @ 2002-10-23 16:42:39 by stevenj] typo M ./threads/threads.c -1 +1 Tue Oct 1 09:32:56 EDT 2002 athena * [project @ 2002-10-01 13:32:56 by athena] Experimental stuff M ./genfft/annotate.ml -2 +9 M ./genfft/annotate.mli -3 +2 M ./genfft/c.ml -4 +16 M ./genfft/genutil.ml -1 +11 M ./genfft/magic.ml -1 +5 Sat Sep 28 13:03:53 EDT 2002 athena * [project @ 2002-09-28 17:03:53 by athena] Experimental Franz mode M ./configure.ac +3 M ./dft/simd/codelets/Makefile.am -3 +68 M ./genfft/gen_notw_c.ml -7 +8 M ./genfft/gen_twiddle_c.ml -6 +7 M ./genfft/genutil.ml -2 +10 Thu Sep 26 15:14:38 EDT 2002 athena * [project @ 2002-09-26 19:14:38 by athena] const-correct M ./kernel/tensor.c -2 +2 Thu Sep 26 15:06:38 EDT 2002 athena * [project @ 2002-09-26 19:06:38 by athena] Reuse dimcmp routine for other purposes M ./dft/vrank2-transpose.c -4 +3 M ./dft/vrank3-transpose.c -4 +2 M ./kernel/ifftw.h -1 +2 M ./kernel/tensor7.c -5 +5 M ./rdft/vrank2-transpose.c -4 +2 M ./rdft/vrank3-transpose.c -4 +2 Wed Sep 25 07:37:38 EDT 2002 athena * [project @ 2002-09-25 11:37:38 by athena] Use tornk1 correctly. M ./dft/direct.c -3 +3 M ./kernel/ifftw.h -2 +2 M ./kernel/tensor.c -2 +3 M ./rdft/direct.c -3 +3 M ./rdft/direct2.c -3 +3 Wed Sep 25 07:36:38 EDT 2002 athena * [project @ 2002-09-25 11:36:38 by athena] Hmm... I thought I had fixed this before... M ./rdft/rdft2-radix2.c -2 +2 Tue Sep 24 21:27:49 EDT 2002 athena * [project @ 2002-09-25 01:27:49 by athena] Collect more common idioms M ./dft/buffered.c -5 +3 M ./dft/rank0.c -16 +3 M ./kernel/tensor.c -2 +2 M ./rdft/buffered.c -5 +2 M ./rdft/buffered2.c -5 +2 M ./rdft/rank0.c -16 +3 Tue Sep 24 21:15:57 EDT 2002 athena * [project @ 2002-09-25 01:15:57 by athena] Still collecting common idioms... M ./dft/direct.c -10 +3 M ./rdft/direct.c -10 +2 M ./rdft/direct2.c -9 +2 Tue Sep 24 21:13:00 EDT 2002 athena * [project @ 2002-09-25 01:13:00 by athena] More garbage collection. M ./dft/direct.c -6 +5 M ./rdft/direct.c -6 +5 M ./rdft/direct2.c -6 +5 Tue Sep 24 21:08:19 EDT 2002 athena * [project @ 2002-09-25 01:08:19 by athena] More compact code M ./dft/buffered.c -6 +3 Tue Sep 24 20:54:43 EDT 2002 athena * [project @ 2002-09-25 00:54:43 by athena] Collect common pattern if (foo) free(foo) ==> free0(foo) M ./dft/buffered.c -7 +6 M ./dft/generic.c -2 +1 M ./dft/rader.c -6 +3 M ./kernel/alloc.c -1 +7 M ./kernel/ifftw.h -1 +2 M ./kernel/planner.c -8 +4 M ./kernel/stride.c -3 +2 M ./kernel/tensor.c -5 +2 M ./rdft/buffered.c -3 +2 M ./rdft/buffered2.c -3 +2 M ./rdft/dht-rader.c -2 +1 M ./rdft/generic.c -2 +1 M ./rdft/problem.c -3 +2 M ./rdft/rader-hc2hc.c -6 +3 Tue Sep 24 20:08:44 EDT 2002 athena * [project @ 2002-09-25 00:08:44 by athena] Collect some common code in */buffered*.c A ./kernel/buffered.c M ./dft/buffered.c -18 +2 M ./kernel/Makefile.am -7 +6 M ./kernel/buffered.c +44 M ./kernel/ifftw.h -1 +2 M ./rdft/buffered.c -18 +3 M ./rdft/buffered2.c -18 +2 Tue Sep 24 19:39:22 EDT 2002 stevenj * [project @ 2002-09-24 23:39:22 by stevenj] use STRUCT_HACK #define to determing rdft kind[] allocation M ./rdft/problem.c -8 +23 M ./rdft/rdft.h -2 +8 Tue Sep 24 17:21:09 EDT 2002 stevenj * [project @ 2002-09-24 21:21:09 by stevenj] report total pcost of measured/estimated plans...epcost is especially useful to estimate the effects of various impatience flags on planning time for large transforms M ./kernel/ifftw.h -1 +2 M ./kernel/planner.c -1 +6 Mon Sep 23 18:49:10 EDT 2002 athena * [project @ 2002-09-23 22:49:10 by athena] Prevent unwanted inlining A ./kernel/trig1.c M ./kernel/Makefile.am -1 +1 M ./kernel/trig.c -54 +7 M ./kernel/trig1.c +70 Mon Sep 23 18:37:59 EDT 2002 athena * [project @ 2002-09-23 22:37:59 by athena] Space compaction M ./kernel/ifftw.h -2 +1 M ./kernel/trig.c -21 +27 Mon Sep 23 11:49:32 EDT 2002 athena * [project @ 2002-09-23 15:49:32 by athena] Still reducing size A ./kernel/hash.c M ./kernel/Makefile.am -1 +1 M ./kernel/hash.c +31 M ./kernel/ifftw.h -1 +2 M ./kernel/md5-1.c +1 M ./kernel/planner.c -12 +3 M ./kernel/scan.c -22 +1 Sun Sep 22 16:03:30 EDT 2002 athena * [project @ 2002-09-22 20:03:30 by athena] Saved another 5KB by redesigning opcnt protocol. (gasp!) M ./dft/buffered.c -5 +6 M ./dft/ct-dif.c -4 +3 M ./dft/ct-dit.c -4 +3 M ./dft/ct-ditbuf.c -6 +6 M ./dft/ct-ditf.c -4 +3 M ./dft/direct.c -2 +3 M ./dft/generic.c -4 +2 M ./dft/indirect.c -2 +2 M ./dft/nop.c -2 +2 M ./dft/rader.c -4 +3 M ./dft/rank-geq2.c -2 +2 M ./dft/rank0.c -2 +2 M ./dft/vrank-geq1.c -2 +3 M ./dft/vrank2-transpose.c -2 +2 M ./dft/vrank3-transpose.c -2 +2 M ./kernel/ifftw.h -6 +14 M ./kernel/ops.c -23 +28 M ./kernel/plan.c -2 +2 M ./rdft/buffered.c -5 +6 M ./rdft/buffered2.c -3 +3 M ./rdft/dht-rader.c -1 +1 M ./rdft/direct.c -3 +8 M ./rdft/direct2.c -3 +8 M ./rdft/generic.c -4 +3 M ./rdft/hc2hc-buf.c -10 +9 M ./rdft/hc2hc-dif.c -8 +8 M ./rdft/hc2hc-dit.c -8 +8 M ./rdft/indirect.c -2 +2 M ./rdft/nop.c -2 +2 M ./rdft/nop2.c -2 +2 M ./rdft/rader-hc2hc.c -8 +5 M ./rdft/rank-geq2-rdft2.c -2 +2 M ./rdft/rank-geq2.c -2 +2 M ./rdft/rank0.c -2 +2 M ./rdft/rdft2-radix2.c -5 +4 M ./rdft/vrank-geq1-rdft2.c -2 +3 M ./rdft/vrank-geq1.c -2 +3 M ./rdft/vrank2-transpose.c -2 +2 M ./rdft/vrank3-transpose.c -2 +2 M ./threads/ct-dit.c -4 +3 M ./threads/dft-vrank-geq1.c -4 +3 M ./threads/hc2hc-dif.c -8 +8 M ./threads/hc2hc-dit.c -8 +8 M ./threads/rdft-vrank-geq1.c -4 +3 M ./threads/vrank-geq1-rdft2.c -4 +3 Sun Sep 22 15:00:59 EDT 2002 athena * [project @ 2002-09-22 19:00:59 by athena] More code compression A ./kernel/tensor8.c M ./dft/buffered.c -4 +3 M ./dft/direct.c -3 +2 M ./dft/indirect.c -4 +3 M ./dft/problem.c -5 +3 M ./dft/rank-geq2.c -9 +3 M ./kernel/Makefile.am -2 +2 M ./kernel/ifftw.h -1 +5 M ./kernel/tensor1.c -1 +6 M ./kernel/tensor4.c -1 +6 M ./kernel/tensor8.c +35 M ./rdft/buffered.c -4 +3 M ./rdft/dft-r2hc.c -2 +2 M ./rdft/direct.c -3 +2 M ./rdft/hc2hc.c -7 +6 M ./rdft/indirect.c -4 +3 M ./rdft/problem.c -7 +4 M ./rdft/problem2.c -5 +3 M ./rdft/rank-geq2-rdft2.c -9 +3 M ./rdft/rank-geq2.c -13 +5 Sun Sep 22 13:27:46 EDT 2002 athena * [project @ 2002-09-22 17:27:46 by athena] Smaller code size. M ./kernel/ifftw.h -2 +4 M ./kernel/solver.c -1 +6 Sun Sep 22 12:50:36 EDT 2002 athena * [project @ 2002-09-22 16:50:36 by athena] Started unification of rader A ./dft/rader-omega.c M ./dft/Makefile.am -3 +3 M ./dft/dft.h -1 +5 M ./dft/rader-omega.c +57 M ./dft/rader.c -38 +3 M ./rdft/rader-hc2hc.c -37 +3 Sun Sep 22 12:35:30 EDT 2002 athena * [project @ 2002-09-22 16:35:30 by athena] Typo M ./rdft/rdft2-radix2.c -2 +2 Sun Sep 22 12:25:20 EDT 2002 athena * [project @ 2002-09-22 16:25:20 by athena] Changed protocol for destroy_plan so as to save space. M ./dft/buffered.c -8 +4 M ./dft/ct.c -2 +1 M ./dft/direct.c -2 +1 M ./dft/generic.c -3 +1 M ./dft/indirect.c -6 +3 M ./dft/nop.c -7 +2 M ./dft/rader.c -7 +3 M ./dft/rank-geq2.c -6 +3 M ./dft/rank0.c -7 +2 M ./dft/vrank-geq1.c -2 +1 M ./dft/vrank2-transpose.c -7 +2 M ./dft/vrank3-transpose.c -7 +2 M ./kernel/ifftw.h -2 +2 M ./kernel/plan.c -3 +11 M ./kernel/problem.c -2 +3 M ./rdft/buffered.c -8 +4 M ./rdft/buffered2.c -6 +3 M ./rdft/dft-r2hc.c -2 +1 M ./rdft/dht-r2hc.c -2 +1 M ./rdft/dht-rader.c -7 +3 M ./rdft/direct.c -2 +1 M ./rdft/direct2.c -2 +1 M ./rdft/generic.c -3 +1 M ./rdft/hc2hc.c -14 +7 M ./rdft/indirect.c -6 +3 M ./rdft/nop.c -7 +2 M ./rdft/nop2.c -7 +2 M ./rdft/rader-hc2hc.c -9 +4 M ./rdft/rank-geq2-rdft2.c -6 +3 M ./rdft/rank-geq2.c -6 +3 M ./rdft/rank0.c -7 +2 M ./rdft/rdft-dht.c -2 +1 M ./rdft/rdft2-radix2.c -2 +1 M ./rdft/vrank-geq1-rdft2.c -2 +1 M ./rdft/vrank-geq1.c -2 +1 M ./rdft/vrank2-transpose.c -7 +2 M ./rdft/vrank3-transpose.c -7 +2 M ./reodft/redft00e-r2hc.c -2 +1 M ./reodft/reodft010e-r2hc.c -2 +1 M ./reodft/reodft11e-r2hc.c -2 +1 M ./reodft/rodft00e-r2hc.c -2 +1 M ./threads/dft-vrank-geq1.c -4 +2 M ./threads/rdft-vrank-geq1.c -4 +2 M ./threads/vrank-geq1-rdft2.c -4 +2 Sun Sep 22 11:08:57 EDT 2002 athena * [project @ 2002-09-22 15:08:57 by athena] Introduced convenient function X(mkplan_d) M ./dft/buffered.c -31 +18 M ./dft/ct.c -5 +2 M ./dft/generic.c -10 +5 M ./dft/indirect.c -14 +8 M ./dft/rader.c -38 +20 M ./dft/rank-geq2.c -16 +12 M ./dft/vrank-geq1.c -9 +7 M ./kernel/ifftw.h -2 +4 M ./kernel/planner.c -1 +8 M ./rdft/buffered.c -30 +20 M ./rdft/buffered2.c -9 +3 M ./rdft/dft-r2hc.c -7 +5 M ./rdft/dht-r2hc.c -7 +4 M ./rdft/dht-rader.c -23 +12 M ./rdft/generic.c -4 +1 M ./rdft/hc2hc.c -16 +10 M ./rdft/indirect.c -14 +9 M ./rdft/rader-hc2hc.c -41 +21 M ./rdft/rank-geq2-rdft2.c -12 +8 M ./rdft/rank-geq2.c -16 +12 M ./rdft/rdft-dht.c -5 +3 M ./rdft/rdft2-radix2.c -7 +3 M ./rdft/vrank-geq1-rdft2.c -9 +7 M ./rdft/vrank-geq1.c -10 +7 M ./reodft/redft00e-r2hc.c -3 +2 M ./reodft/reodft010e-r2hc.c -3 +2 M ./reodft/reodft11e-r2hc.c -3 +2 M ./reodft/rodft00e-r2hc.c -3 +2 M ./threads/ct-dit.c -4 +4 M ./threads/dft-vrank-geq1.c -12 +10 M ./threads/hc2hc-dif.c -4 +4 M ./threads/hc2hc-dit.c -4 +4 M ./threads/rdft-vrank-geq1.c -11 +9 M ./threads/vrank-geq1-rdft2.c -11 +9 Sun Sep 22 10:21:36 EDT 2002 athena * [project @ 2002-09-22 14:21:36 by athena] Split tensor/md5 into separate files to allow independent linking and/or prevent undesidred inlining A ./kernel/md5-1.c A ./kernel/tensor1.c A ./kernel/tensor2.c A ./kernel/tensor3.c A ./kernel/tensor4.c A ./kernel/tensor5.c A ./kernel/tensor7.c M ./kernel/Makefile.am -3 +4 M ./kernel/md5-1.c +53 M ./kernel/md5.c -31 M ./kernel/tensor.c -270 +2 M ./kernel/tensor1.c +32 M ./kernel/tensor2.c +37 M ./kernel/tensor3.c +46 M ./kernel/tensor4.c +68 M ./kernel/tensor5.c +93 M ./kernel/tensor7.c +127 Sun Sep 22 09:49:09 EDT 2002 athena * [project @ 2002-09-22 13:49:08 by athena] Treat all tensors as dynamically allocated objects. They were dynamically allocated in part anyway, so there is no point in complicating the object code with the clumsy calling conventions for by-value structs. M ./dft/buffered.c -19 +19 M ./dft/ct-dif.c -5 +5 M ./dft/ct-dit.c -5 +5 M ./dft/ct-ditbuf.c -4 +5 M ./dft/ct-ditf.c -5 +5 M ./dft/ct.c -14 +14 M ./dft/dft.h -4 +4 M ./dft/direct.c -13 +13 M ./dft/generic.c -7 +7 M ./dft/indirect.c -18 +16 M ./dft/nop.c -5 +5 M ./dft/problem.c -12 +12 M ./dft/rader.c -12 +12 M ./dft/rank-geq2.c -22 +22 M ./dft/rank0.c -14 +14 M ./dft/vrank-geq1.c -11 +11 M ./dft/vrank2-transpose.c -9 +9 M ./dft/vrank3-transpose.c -12 +12 M ./dft/zero.c -18 +20 M ./kernel/ifftw.h -13 +22 M ./kernel/tensor.c -92 +102 M ./rdft/buffered.c -23 +21 M ./rdft/buffered2.c -24 +24 M ./rdft/dft-r2hc.c -11 +11 M ./rdft/dht-r2hc.c -6 +6 M ./rdft/dht-rader.c -7 +7 M ./rdft/direct.c -18 +18 M ./rdft/direct2.c -16 +16 M ./rdft/generic.c -8 +8 M ./rdft/hc2hc-buf.c -4 +5 M ./rdft/hc2hc-dif.c -5 +5 M ./rdft/hc2hc-dit.c -5 +5 M ./rdft/hc2hc.c -23 +23 M ./rdft/indirect.c -20 +18 M ./rdft/nop.c -5 +5 M ./rdft/nop2.c -4 +4 M ./rdft/problem.c -39 +40 M ./rdft/problem2.c -34 +34 M ./rdft/rader-hc2hc.c -11 +11 M ./rdft/rank-geq2-rdft2.c -23 +23 M ./rdft/rank-geq2.c -28 +28 M ./rdft/rank0.c -10 +10 M ./rdft/rdft-dht.c -10 +10 M ./rdft/rdft.h -9 +9 M ./rdft/rdft2-radix2.c -21 +21 M ./rdft/vrank-geq1-rdft2.c -11 +11 M ./rdft/vrank-geq1.c -11 +11 M ./rdft/vrank2-transpose.c -9 +9 M ./rdft/vrank3-transpose.c -12 +12 M ./reodft/redft00e-r2hc.c -10 +10 M ./reodft/reodft010e-r2hc.c -10 +10 M ./reodft/reodft11e-r2hc.c -10 +10 M ./reodft/rodft00e-r2hc.c -10 +10 M ./tests/debug.h -2 +2 M ./tests/dotens.c -4 +4 M ./tests/dotens2.c -5 +5 M ./tests/verify-dft.c -10 +11 M ./tests/verify-lib.c -15 +16 M ./tests/verify-rdft.c -43 +47 M ./tests/verify-reodft.c -28 +28 M ./tests/verify.h -3 +3 Sat Sep 21 18:24:55 EDT 2002 stevenj * [project @ 2002-09-21 22:24:55 by stevenj] typo M ./kernel/ifftw.h -2 +2 Sat Sep 21 18:10:07 EDT 2002 athena * [project @ 2002-09-21 22:10:07 by athena] Avoid generating NaN when n = 0. M ./tests/verify-lib.c -13 +16 Sat Sep 21 18:04:05 EDT 2002 athena * [project @ 2002-09-21 22:04:05 by athena] Saved more. M ./dft/dft.h -2 +2 M ./dft/problem.c -5 +5 M ./dft/rank-geq2.c -2 +2 M ./rdft/dft-r2hc.c -2 +2 M ./rdft/dht-r2hc.c -2 +2 M ./rdft/hc2hc.c -5 +7 M ./rdft/problem.c -11 +11 M ./rdft/problem2.c -5 +5 M ./rdft/rank-geq2.c -2 +2 M ./rdft/rdft-dht.c -3 +3 M ./rdft/rdft.h -4 +4 M ./reodft/redft00e-r2hc.c -2 +2 M ./reodft/reodft010e-r2hc.c -2 +2 M ./reodft/reodft11e-r2hc.c -2 +2 M ./reodft/rodft00e-r2hc.c -2 +2 M ./threads/dft-vrank-geq1.c -2 +2 Sat Sep 21 17:47:36 EDT 2002 athena * [project @ 2002-09-21 21:47:35 by athena] Save 1200 bytes of object code. Do not pass structs by value whenever practical, because the calling protocol generates clumsy code. M ./dft/buffered.c -6 +6 M ./dft/ct.c -5 +5 M ./dft/direct.c -3 +3 M ./dft/indirect.c -12 +12 M ./dft/nop.c -2 +2 M ./dft/problem.c -11 +11 M ./dft/rank-geq2.c -21 +21 M ./dft/vrank-geq1.c -6 +6 M ./dft/vrank3-transpose.c -8 +8 M ./kernel/ifftw.h -20 +22 M ./kernel/pickdim.c -9 +9 M ./kernel/print.c -2 +2 M ./kernel/tensor.c -69 +71 M ./rdft/buffered.c -6 +6 M ./rdft/buffered2.c -2 +2 M ./rdft/dft-r2hc.c -3 +3 M ./rdft/direct.c -3 +3 M ./rdft/hc2hc.c -9 +9 M ./rdft/indirect.c -14 +14 M ./rdft/nop.c -2 +2 M ./rdft/problem.c -15 +15 M ./rdft/problem2.c -15 +15 M ./rdft/rank-geq2-rdft2.c -19 +19 M ./rdft/rank-geq2.c -27 +27 M ./rdft/rdft-dht.c -3 +3 M ./rdft/rdft.h -2 +2 M ./rdft/rdft2-radix2.c -7 +7 M ./rdft/vrank-geq1-rdft2.c -6 +7 M ./rdft/vrank-geq1.c -6 +6 M ./rdft/vrank3-transpose.c -8 +8 M ./reodft/redft00e-r2hc.c -2 +2 M ./reodft/reodft010e-r2hc.c -2 +2 M ./reodft/reodft11e-r2hc.c -2 +2 M ./reodft/rodft00e-r2hc.c -2 +2 M ./tests/verify-dft.c -6 +6 M ./tests/verify-lib.c -2 +2 M ./tests/verify-rdft.c -21 +21 M ./tests/verify-reodft.c -10 +10 M ./threads/dft-vrank-geq1.c -3 +3 M ./threads/rdft-vrank-geq1.c -3 +3 M ./threads/vrank-geq1-rdft2.c -3 +4 Sat Sep 21 12:10:21 EDT 2002 athena * [project @ 2002-09-21 16:10:21 by athena] Do not allocate buffers for rader omegas. Let the planner do it if necessary. M ./rdft/dht-rader.c -15 +6 Sat Sep 21 12:03:46 EDT 2002 athena * [project @ 2002-09-21 16:03:46 by athena] Check rank *before* reading kind[0], which may be undefined if rnk < 1 M ./tests/verify-rdft.c -2 +2 M ./tests/verify-reodft.c -2 +2 Sat Sep 21 11:48:50 EDT 2002 athena * [project @ 2002-09-21 15:48:50 by athena] Second step towards rader unification. M ./dft/rader.c -17 +7 M ./rdft/rader-hc2hc.c -1 +1 Sat Sep 21 11:37:06 EDT 2002 athena * [project @ 2002-09-21 15:37:06 by athena] First step towards unification of Rader code A ./kernel/rader.c M ./dft/rader.c -57 +12 M ./kernel/Makefile.am -3 +3 M ./kernel/ifftw.h -1 +9 M ./kernel/rader.c +68 M ./rdft/dht-rader.c -50 +6 M ./rdft/rader-hc2hc.c -57 +11 Sat Sep 21 07:58:11 EDT 2002 athena * [project @ 2002-09-21 11:58:11 by athena] Fix ugliness condition for cooley-tukey. A ./kernel/ct.c M ./dft/ct-dif.c -5 +3 M ./dft/ct-dit.c -4 +2 M ./dft/ct-ditbuf.c -5 +3 M ./kernel/Makefile.am -4 +5 M ./kernel/ct.c +31 M ./kernel/ifftw.h -1 +2 M ./kernel/planner.c -3 +2 M ./rdft/dht-r2hc.c -3 +3 M ./rdft/dht-rader.c -3 +3 M ./rdft/hc2hc-buf.c -9 +3 M ./rdft/hc2hc-dif.c -4 +2 M ./rdft/hc2hc-dit.c -4 +2 M ./rdft/rdft-dht.c -4 +2 M ./reodft/redft00e-r2hc.c -4 +2 M ./reodft/reodft010e-r2hc.c -4 +2 M ./reodft/reodft11e-r2hc.c -4 +2 M ./reodft/rodft00e-r2hc.c -4 +2 M ./threads/ct-dit.c -5 +3 M ./threads/hc2hc-dif.c -5 +3 M ./threads/hc2hc-dit.c -5 +3 Fri Sep 20 16:53:45 EDT 2002 athena * [project @ 2002-09-20 20:53:45 by athena] Removed RADER_MIN_GOOD and associated machinery M ./dft/rader.c -29 +6 M ./kernel/ifftw.h -2 +1 M ./rdft/dht-rader.c -14 +3 M ./rdft/rader-hc2hc.c -17 +5 Fri Sep 20 14:49:12 EDT 2002 athena * [project @ 2002-09-20 18:49:12 by athena] Proper cast M ./rdft/dht-r2hc.c -4 +3 Fri Sep 20 14:45:54 EDT 2002 athena * [project @ 2002-09-20 18:45:54 by athena] Typo M ./kernel/planner.c -2 +2 Fri Sep 20 14:38:13 EDT 2002 athena * [project @ 2002-09-20 18:38:13 by athena] Implemented NO_LARGE_GENERIC M ./dft/generic.c -1 +8 M ./kernel/ifftw.h -20 +23 M ./rdft/dht-rader.c -1 +1 M ./rdft/generic.c -2 +8 M ./rdft/rdft-dht.c -10 +2 M ./tests/bench.c +1 Thu Sep 19 07:48:25 EDT 2002 athena * [project @ 2002-09-19 11:48:24 by athena] Consistent macroization of NO_DHT_R2HC M ./kernel/ifftw.h -1 +2 M ./rdft/dht-r2hc.c -4 +4 Wed Sep 18 21:47:17 EDT 2002 athena * [project @ 2002-09-19 01:47:17 by athena] NO_DHT_R2HC is a planner flag, otherwise the EXHAUSTIVE planner loops. M ./kernel/ifftw.h -7 +6 M ./kernel/planner.c -10 +6 M ./rdft/dht-r2hc.c -3 +3 M ./tests/bench.c +1 Wed Sep 18 20:47:31 EDT 2002 athena * [project @ 2002-09-19 00:47:31 by athena] Resurrected NO_EXHAUSTIVE M ./kernel/ifftw.h -1 +3 M ./kernel/planner.c -8 +24 Wed Sep 18 19:31:57 EDT 2002 stevenj * [project @ 2002-09-18 23:31:57 by stevenj] au revoir, score() M ./threads/ct-dit.c -25 +22 M ./threads/dft-vrank-geq1.c -14 +10 M ./threads/hc2hc-dif.c -25 +21 M ./threads/hc2hc-dit.c -25 +21 M ./threads/rdft-vrank-geq1.c -14 +10 M ./threads/vrank-geq1-rdft2.c -14 +10 Wed Sep 18 19:31:05 EDT 2002 stevenj * [project @ 2002-09-18 23:31:05 by stevenj] eliminated unused M ./tests/bench.c +2 M ./tests/verify-reodft.c -7 +7 Wed Sep 18 18:28:44 EDT 2002 stevenj * [project @ 2002-09-18 22:28:44 by stevenj] capitalize and parenthesize SUBSUMES M ./kernel/planner.c -9 +8 Wed Sep 18 18:26:58 EDT 2002 stevenj * [project @ 2002-09-18 22:26:58 by stevenj] comment M ./kernel/ifftw.h -2 +2 Wed Sep 18 18:03:18 EDT 2002 athena * [project @ 2002-09-18 22:03:18 by athena] Use flags from wisdom if wisdom is applicable. M ./kernel/ifftw.h -2 +3 M ./kernel/planner.c -70 +32 Wed Sep 18 17:16:17 EDT 2002 athena * [project @ 2002-09-18 21:16:16 by athena] Removed score() machinery M ./dft/buffered.c -16 +11 M ./dft/ct-dif.c -23 +21 M ./dft/ct-dit.c -29 +25 M ./dft/ct-ditbuf.c -30 +24 M ./dft/ct-ditf.c -10 +2 M ./dft/direct.c -8 +2 M ./dft/generic.c -9 +7 M ./dft/indirect.c -9 +12 M ./dft/nop.c -8 +2 M ./dft/rader.c -23 +25 M ./dft/rank-geq2.c -14 +13 M ./dft/rank0.c -8 +2 M ./dft/vrank-geq1.c -30 +29 M ./dft/vrank2-transpose.c -8 +2 M ./dft/vrank3-transpose.c -13 +15 M ./kernel/ifftw.h -12 +2 M ./kernel/planner.c -31 +19 M ./rdft/buffered.c -18 +11 M ./rdft/buffered2.c -18 +11 M ./rdft/dft-r2hc.c -12 +14 M ./rdft/dht-r2hc.c -8 +8 M ./rdft/dht-rader.c -12 +12 M ./rdft/direct.c -9 +3 M ./rdft/direct2.c -9 +3 M ./rdft/generic.c -8 +7 M ./rdft/hc2hc-buf.c -30 +29 M ./rdft/hc2hc-dif.c -33 +25 M ./rdft/hc2hc-dit.c -32 +27 M ./rdft/indirect.c -9 +14 M ./rdft/nop.c -8 +2 M ./rdft/nop2.c -8 +2 M ./rdft/rader-hc2hc.c -13 +12 M ./rdft/rank-geq2-rdft2.c -15 +17 M ./rdft/rank-geq2.c -17 +19 M ./rdft/rank0.c -8 +2 M ./rdft/rdft-dht.c -11 +13 M ./rdft/rdft2-radix2.c -12 +2 M ./rdft/vrank-geq1-rdft2.c -32 +29 M ./rdft/vrank-geq1.c -31 +31 M ./rdft/vrank2-transpose.c -8 +2 M ./rdft/vrank3-transpose.c -18 +14 M ./reodft/redft00e-r2hc.c -6 +7 M ./reodft/reodft010e-r2hc.c -6 +7 M ./reodft/reodft11e-r2hc.c -6 +7 M ./reodft/rodft00e-r2hc.c -6 +7 M ./tests/bench.c -2 Wed Sep 18 14:12:21 EDT 2002 athena * [project @ 2002-09-18 18:12:21 by athena] Revised planner hack M ./kernel/planner.c -4 +7 Wed Sep 18 10:14:41 EDT 2002 athena * [project @ 2002-09-18 14:14:41 by athena] Fix warning M ./simd/simd-altivec.h +2 Tue Sep 17 17:54:07 EDT 2002 athena * [project @ 2002-09-17 21:54:07 by athena] Type qualifiers. M ./dft/indirect.c -2 +2 M ./rdft/indirect.c -2 +2 Tue Sep 17 16:17:55 EDT 2002 athena * [project @ 2002-09-17 20:17:55 by athena] ESTIMATE is no longer subsumed by everything else. M ./kernel/planner.c -2 +1 Tue Sep 17 10:55:15 EDT 2002 athena * [project @ 2002-09-17 14:55:15 by athena] NO_BUFFERING is a planner flag, not a problem flag M ./dft/indirect.c -2 +2 M ./rdft/indirect.c -2 +2 Tue Sep 17 09:36:16 EDT 2002 athena * [project @ 2002-09-17 13:36:16 by athena] Maintain flags in canonical form. M ./kernel/ifftw.h -4 +6 M ./kernel/planner.c -13 +15 Tue Sep 17 09:09:57 EDT 2002 athena * [project @ 2002-09-17 13:09:56 by athena] In dramatic break with tradition, SUBSUME is now a partial order. I swear. M ./kernel/ifftw.h -4 +1 M ./kernel/planner.c -15 +54 Tue Sep 17 07:29:00 EDT 2002 athena * [project @ 2002-09-17 11:29:00 by athena] Added comment M ./kernel/planner.c -1 +3 Tue Sep 17 07:27:17 EDT 2002 athena * [project @ 2002-09-17 11:27:17 by athena] Inverted ESTIMATE flag, renamed USE_SCORE for consistency with the convention that 0 subsumes 1. M ./kernel/ifftw.h -9 +6 M ./kernel/planner.c -10 +7 M ./tests/bench.c +2 Tue Sep 17 02:50:15 EDT 2002 stevenj * [project @ 2002-09-17 06:50:15 by stevenj] NO_INDIRECT -> NO_INDIRECT_OP (out-of-place only) M ./dft/indirect.c -4 +5 M ./kernel/ifftw.h -3 +3 M ./rdft/indirect.c -4 +5 M ./tests/bench.c -1 +1 Tue Sep 17 00:40:04 EDT 2002 stevenj * [project @ 2002-09-17 04:40:04 by stevenj] hpux needs -D_REENTRANT (thanks to Clinton Roy for the bug report) M ./acx_pthread.m4 -2 +2 Mon Sep 16 23:54:34 EDT 2002 athena * [project @ 2002-09-17 03:54:34 by athena] Oops. M ./kernel/planner.c -2 +2 Mon Sep 16 23:44:47 EDT 2002 athena * [project @ 2002-09-17 03:44:47 by athena] Yet another attempt at getting the planner right. M ./kernel/ifftw.h -3 +3 M ./kernel/planner.c -28 +27 Mon Sep 16 21:56:14 EDT 2002 athena * [project @ 2002-09-17 01:56:14 by athena] Better coding. M ./kernel/planner.c -21 +11 Mon Sep 16 21:51:06 EDT 2002 athena * [project @ 2002-09-17 01:51:06 by athena] NO_UGLY is no longer a flag, but a separate planner field that does not interfere with wisdom. M ./kernel/ifftw.h -3 +2 M ./kernel/planner.c -14 +15 Mon Sep 16 19:04:41 EDT 2002 athena * [project @ 2002-09-16 23:04:41 by athena] Did not compile without FFTW_DEBUG M ./tests/verify-reodft.c -3 +1 Mon Sep 16 18:37:06 EDT 2002 athena * [project @ 2002-09-16 22:37:06 by athena] Changed scoring mechanism. M ./kernel/ifftw.h -5 +4 M ./kernel/plan.c -5 +1 M ./kernel/planner.c -51 +28 M ./tests/bench.c -5 +2 Mon Sep 16 17:13:45 EDT 2002 athena * [project @ 2002-09-16 21:13:45 by athena] Count infeasible plans M ./kernel/planner.c -4 +12 Mon Sep 16 16:36:12 EDT 2002 athena * [project @ 2002-09-16 20:36:12 by athena] curse subsumed plans before export M ./kernel/planner.c -27 +35 Mon Sep 16 15:40:46 EDT 2002 stevenj * [project @ 2002-09-16 19:40:46 by stevenj] removed ESTIMATE_BIT vs. ESTIMATE... ESTIMATE | IMPATIENT is a UI issue M ./kernel/ifftw.h -6 +4 M ./kernel/planner.c -2 +2 Mon Sep 16 15:31:39 EDT 2002 stevenj * [project @ 2002-09-16 19:31:39 by stevenj] cleanup M ./rdft/buffered2.c -6 +3 Mon Sep 16 15:28:47 EDT 2002 stevenj * [project @ 2002-09-16 19:28:47 by stevenj] use CONSERVE_MEMORY flag to prevent buffered for large sizes M ./dft/buffered.c -4 +7 M ./rdft/buffered.c -4 +7 M ./rdft/buffered2.c -5 +9 Mon Sep 16 15:16:16 EDT 2002 stevenj * [project @ 2002-09-16 19:16:16 by stevenj] moved NO_DHT_R2HC back into planner flags: there's no reason we would want this flag to block plan reuse M ./kernel/ifftw.h -4 +4 Mon Sep 16 14:59:14 EDT 2002 stevenj * [project @ 2002-09-16 18:59:14 by stevenj] whoops, commas M ./kernel/ifftw.h -3 +3 Mon Sep 16 14:58:26 EDT 2002 stevenj * [project @ 2002-09-16 18:58:26 by stevenj] problem_flags == checked in applicable, planner_flags == checked in score M ./kernel/ifftw.h -5 +7 Mon Sep 16 14:53:16 EDT 2002 stevenj * [project @ 2002-09-16 18:53:16 by stevenj] ESTIMATE should not *include* all impatience flags, even if it subsumes them; some impatience flags, like NO_INDIRECT, might make a problem unsolvable M ./kernel/ifftw.h -6 +6 M ./kernel/planner.c -3 +4 Mon Sep 16 00:56:29 EDT 2002 stevenj * [project @ 2002-09-16 04:56:29 by stevenj] quotatio marks M ./kernel/planner.c -3 +3 Sun Sep 15 23:55:44 EDT 2002 stevenj * [project @ 2002-09-16 03:55:44 by stevenj] delete blank line M ./kernel/planner.c -2 +1 Sun Sep 15 23:51:14 EDT 2002 stevenj * [project @ 2002-09-16 03:51:14 by stevenj] substitution M ./kernel/planner.c -2 +2 Sun Sep 15 23:49:50 EDT 2002 stevenj * [project @ 2002-09-16 03:49:50 by stevenj] note that we are not GNUlly correct M ./kernel/planner.c -1 +10 Sun Sep 15 23:41:01 EDT 2002 stevenj * [project @ 2002-09-16 03:41:01 by stevenj] indenting M ./kernel/planner.c -3 +3 Sun Sep 15 23:37:46 EDT 2002 stevenj * [project @ 2002-09-16 03:37:46 by stevenj] more jokes M ./kernel/planner.c -2 +5 Sun Sep 15 23:20:14 EDT 2002 stevenj * [project @ 2002-09-16 03:20:14 by stevenj] NONTHREADED_ICKYP includes nthr > 1 check M ./dft/ct-dit.c -2 +2 M ./dft/vrank-geq1.c -2 +2 M ./kernel/ifftw.h -2 +3 M ./rdft/hc2hc-dif.c -2 +2 M ./rdft/hc2hc-dit.c -2 +2 M ./rdft/vrank-geq1-rdft2.c -2 +2 M ./rdft/vrank-geq1.c -2 +2 Sun Sep 15 22:56:44 EDT 2002 stevenj * [project @ 2002-09-16 02:56:44 by stevenj] use md5sig M ./kernel/md5.c -1 +1 Sun Sep 15 22:55:41 EDT 2002 stevenj * [project @ 2002-09-16 02:55:41 by stevenj] md5sig typedef M ./kernel/ifftw.h -2 +4 M ./kernel/planner.c -9 +9 Sun Sep 15 22:35:13 EDT 2002 stevenj * [project @ 2002-09-16 02:35:13 by stevenj] updated M ./ChangeLog +324 Sun Sep 15 22:30:26 EDT 2002 stevenj * [project @ 2002-09-16 02:30:26 by stevenj] partially-ordered impatience M ./dft/buffered.c -2 +2 M ./dft/ct-dif.c -3 +3 M ./dft/ct-dit.c -3 +3 M ./dft/ct-ditbuf.c -2 +2 M ./dft/ct.c -5 +1 M ./dft/indirect.c -4 +4 M ./dft/rank-geq2.c -3 +2 M ./dft/vrank-geq1.c -13 +3 M ./kernel/ifftw.h -18 +51 M ./kernel/planner.c -11 +49 M ./rdft/buffered.c -2 +2 M ./rdft/buffered2.c -2 +2 M ./rdft/dft-r2hc.c -2 +2 M ./rdft/dht-r2hc.c -3 +3 M ./rdft/hc2hc-buf.c -4 +3 M ./rdft/hc2hc-dif.c -4 +4 M ./rdft/hc2hc-dit.c -3 +3 M ./rdft/hc2hc.c -5 +1 M ./rdft/indirect.c -4 +4 M ./rdft/rank-geq2-rdft2.c -3 +2 M ./rdft/rank-geq2.c -3 +2 M ./rdft/rdft-dht.c -3 +3 M ./rdft/rdft2-radix2.c -2 +2 M ./rdft/vrank-geq1-rdft2.c -13 +3 M ./rdft/vrank-geq1.c -13 +3 M ./tests/bench.c -3 M ./threads/dft-vrank-geq1.c -13 +2 M ./threads/hc2hc-dif.c -2 +2 M ./threads/rdft-vrank-geq1.c -12 +2 M ./threads/vrank-geq1-rdft2.c -12 +2 Sat Sep 14 19:47:56 EDT 2002 athena * [project @ 2002-09-14 23:47:56 by athena] Removed all that planner inheritance crap. M ./kernel/Makefile.am -4 +3 M ./kernel/ifftw.h -15 +5 R ./kernel/planner-naive.c R ./kernel/planner-score.c M ./kernel/planner.c -35 +128 M ./tests/bench.c -2 +2 Sat Sep 14 16:35:28 EDT 2002 stevenj * [project @ 2002-09-14 20:35:28 by stevenj] string.h is used for more than strlen M ./kernel/planner.c -2 +2 Sat Sep 14 12:19:13 EDT 2002 athena * [project @ 2002-09-14 16:19:13 by athena] Reduced hashtable size by 1/6 (on 32-bit machines) at the expense of messier planner. M ./kernel/ifftw.h -13 +17 M ./kernel/planner.c -80 +87 Sat Sep 14 08:31:29 EDT 2002 athena * [project @ 2002-09-14 12:31:29 by athena] Only print wisdom if verbose > 3 M ./tests/bench.c -2 +4 Sat Sep 14 07:56:56 EDT 2002 athena * [project @ 2002-09-14 11:56:56 by athena] Changed syntax of temporaries to avoid shadowing library functions (which is harmless but I hate the warning) M ./genfft/variable.ml -2 +2 M ./genfft-k7/variable.ml -1 +1 Fri Sep 13 23:07:39 EDT 2002 stevenj * [project @ 2002-09-14 03:07:39 by stevenj] only add warnings in debug/maintainer mode, and add a few more warning flags; eliminate more warnings; add support for posix_memalign (broken in glibc, grrr) M ./acinclude.m4 -1 +1 M ./configure.ac -1 +9 M ./dft/rader.c -1 +1 M ./kernel/alloc.c -3 +16 M ./kernel/assert.c -2 +2 M ./kernel/ifftw.h -5 +5 M ./kernel/md5.c -1 +1 M ./kernel/planner-score.c -11 +13 M ./kernel/primes.c -4 +4 M ./kernel/scan.c -2 +2 M ./libbench/bench-user.h -2 +2 M ./libbench/bench.h -3 +4 M ./libbench/report.c -5 +5 M ./libbench/timer.c -1 +2 M ./libbench/util.c -14 +26 M ./libbench/verify.c -4 +1 M ./rdft/rader-hc2hc.c -6 +6 M ./tests/bench.c -3 M ./tests/verify-lib.c -1 +4 Fri Sep 13 21:57:50 EDT 2002 athena * [project @ 2002-09-14 01:57:50 by athena] Explicit cast M ./kernel/twiddle.c -2 +2 Fri Sep 13 21:54:50 EDT 2002 athena * [project @ 2002-09-14 01:54:50 by athena] Use double-hashing. This allows a slightly higher load factor at the expense of a messier computation of the hashtable size. M ./kernel/ifftw.h -1 +2 M ./kernel/planner.c -57 +63 M ./kernel/primes.c -1 +7 Fri Sep 13 17:53:13 EDT 2002 stevenj * [project @ 2002-09-13 21:53:13 by stevenj] typo M ./genfft/magic.ml -2 +2 Fri Sep 13 15:36:07 EDT 2002 athena * [project @ 2002-09-13 19:36:07 by athena] Slight change in hash table growth functions. M ./kernel/planner.c -3 +12 Fri Sep 13 14:58:22 EDT 2002 athena * [project @ 2002-09-13 18:58:22 by athena] More statistics. M ./kernel/ifftw.h -1 +2 M ./kernel/planner.c -1 +8 Fri Sep 13 10:13:02 EDT 2002 athena * [project @ 2002-09-13 14:13:02 by athena] Clearer logic. M ./kernel/planner.c -2 +3 Fri Sep 13 10:11:10 EDT 2002 athena * [project @ 2002-09-13 14:11:10 by athena] Oops. M ./kernel/planner.c -2 +1 Fri Sep 13 09:31:46 EDT 2002 athena * [project @ 2002-09-13 13:31:46 by athena] Cleaned up M ./kernel/planner.c -5 +6 Fri Sep 13 09:16:07 EDT 2002 athena * [project @ 2002-09-13 13:16:07 by athena] Deal properly with infeasible problems. M ./kernel/planner.c -4 +7 Fri Sep 13 07:15:06 EDT 2002 athena * [project @ 2002-09-13 11:15:06 by athena] Redundantly initialize hash table to prevent valgrind warnings. M ./kernel/planner.c -3 +7 Thu Sep 12 19:00:22 EDT 2002 athena * [project @ 2002-09-12 23:00:22 by athena] Removed relics from past. M ./kernel/md5.c -5 +3 Thu Sep 12 18:53:44 EDT 2002 athena * [project @ 2002-09-12 22:53:44 by athena] md5hash a problem only once. M ./kernel/ifftw.h -6 +11 M ./kernel/planner.c -39 +32 Thu Sep 12 16:33:49 EDT 2002 athena * [project @ 2002-09-12 20:33:49 by athena] Renamed k7 codelets M ./genfft-k7/genUtil.ml -1 +1 M ./genfft-k7/gen_notw.ml -3 +3 M ./genfft-k7/gen_twiddle.ml -4 +4 Thu Sep 12 16:32:03 EDT 2002 stevenj * [project @ 2002-09-12 20:32:03 by stevenj] FORBID_DHT_R2HC -> DHT_R2HC_VERBOTEN for consistency M ./kernel/ifftw.h -2 +2 M ./rdft/dht-r2hc.c -3 +3 Thu Sep 12 16:28:43 EDT 2002 stevenj * [project @ 2002-09-12 20:28:43 by stevenj] removed obsolete macro M ./kernel/ifftw.h -3 +1 Thu Sep 12 16:20:39 EDT 2002 athena * [project @ 2002-09-12 20:20:39 by athena] Split flags in SIMD code. M ./dft/simd/n1b.c -1 +1 M ./dft/simd/n1f.c -1 +1 M ./dft/simd/t1b.c -1 +1 M ./dft/simd/t1f.c -1 +1 Thu Sep 12 16:18:51 EDT 2002 athena * [project @ 2002-09-12 20:18:51 by athena] Forgot to fix threads M ./threads/dft-vrank-geq1.c -5 +6 M ./threads/hc2hc-dif.c -2 +2 M ./threads/rdft-vrank-geq1.c -5 +6 M ./threads/vrank-geq1-rdft2.c -5 +6 Thu Sep 12 16:10:05 EDT 2002 athena * [project @ 2002-09-12 20:10:05 by athena] Split flags into planner_flags and problem_flags M ./dft/buffered.c -2 +2 M ./dft/ct-dif.c -3 +3 M ./dft/ct-dit.c -3 +3 M ./dft/ct-ditbuf.c -2 +2 M ./dft/ct.c -3 +3 M ./dft/indirect.c -4 +4 M ./dft/rader.c -1 +1 M ./dft/rank-geq2.c -2 +3 M ./dft/rank0.c -2 +2 M ./dft/vrank-geq1.c -7 +8 M ./kernel/ifftw.h -22 +21 M ./kernel/planner-naive.c -3 +3 M ./kernel/planner-score.c -3 +3 M ./kernel/planner.c -33 +28 M ./rdft/buffered.c -2 +2 M ./rdft/buffered2.c -3 +3 M ./rdft/dft-r2hc.c -2 +2 M ./rdft/dht-r2hc.c -3 +5 M ./rdft/dht-rader.c -1 +1 M ./rdft/hc2hc-buf.c -3 +3 M ./rdft/hc2hc-dif.c -4 +4 M ./rdft/hc2hc-dit.c -3 +3 M ./rdft/hc2hc.c -3 +3 M ./rdft/indirect.c -4 +4 M ./rdft/rank-geq2-rdft2.c -2 +3 M ./rdft/rank-geq2.c -2 +3 M ./rdft/rdft-dht.c -3 +3 M ./rdft/rdft2-radix2.c -2 +2 M ./rdft/vrank-geq1-rdft2.c -7 +8 M ./rdft/vrank-geq1.c -7 +8 M ./tests/bench.c -8 +8 Thu Sep 12 15:46:56 EDT 2002 stevenj * [project @ 2002-09-12 19:46:56 by stevenj] tetrameter M ./kernel/planner.c -2 +3 Thu Sep 12 15:11:21 EDT 2002 athena * [project @ 2002-09-12 19:11:21 by athena] Overwrite less impatient solutions properly. M ./kernel/planner.c -23 +15 Thu Sep 12 11:29:16 EDT 2002 athena * [project @ 2002-09-12 15:29:16 by athena] Oops. M ./kernel/planner.c -3 +3 Thu Sep 12 10:58:56 EDT 2002 athena * [project @ 2002-09-12 14:58:56 by athena] Keep less impatient solution in case of conflict. Paranoid cast to uint in certain places. M ./kernel/planner.c -4 +5 Thu Sep 12 10:02:51 EDT 2002 athena * [project @ 2002-09-12 14:02:51 by athena] Complete reimplementation of planner hash table. M ./kernel/ifftw.h -6 +7 M ./kernel/planner.c -143 +146 M ./tests/bench.c -2 +4 Thu Sep 12 07:58:45 EDT 2002 athena * [project @ 2002-09-12 11:58:45 by athena] planner->cnt was not properly decremented. M ./kernel/planner.c -13 +14 Wed Sep 11 17:52:39 EDT 2002 stevenj * [project @ 2002-09-11 21:52:39 by stevenj] typo M ./NEWS -1 +1 Mon Sep 9 17:10:45 EDT 2002 athena * [project @ 2002-09-09 21:10:45 by athena] Simplified M ./kernel/planner.c -15 +5 Mon Sep 9 17:03:32 EDT 2002 athena * [project @ 2002-09-09 21:03:32 by athena] Always overwrite old wisdom with new, in case the old is corrupt/conclicting. M ./kernel/planner.c -16 +8 Mon Sep 9 16:56:03 EDT 2002 stevenj * [project @ 2002-09-09 20:56:03 by stevenj] added quote/joke M ./kernel/plan.c -1 +5 Mon Sep 9 15:04:47 EDT 2002 athena * [project @ 2002-09-09 19:04:47 by athena] Completed wisdom import M ./kernel/ifftw.h -10 +11 M ./kernel/md5.c -1 +1 M ./kernel/planner.c -37 +54 M ./kernel/print.c -2 +3 M ./kernel/scan.c -73 +25 M ./tests/bench.c -3 +3 Mon Sep 9 10:14:22 EDT 2002 athena * [project @ 2002-09-09 14:14:22 by athena] Slight cleanup of md5 interface. M ./dft/problem.c -2 +2 M ./kernel/ifftw.h -2 +4 M ./kernel/md5.c -12 +24 M ./rdft/problem.c -2 +2 M ./rdft/problem2.c -2 +2 Tue Sep 3 22:32:43 EDT 2002 athena * [project @ 2002-09-04 02:32:43 by athena] More consistent protocol between planner and inferior. M ./kernel/planner-naive.c -2 +5 M ./kernel/planner-score.c -1 +2 M ./kernel/planner.c -3 +1 Tue Sep 3 21:08:30 EDT 2002 athena * [project @ 2002-09-04 01:08:30 by athena] I can't think of any situation where saving infeasible problems would be desirable. Removed relevant code. M ./kernel/planner.c -9 +3 Tue Sep 3 20:57:03 EDT 2002 athena * [project @ 2002-09-04 00:57:03 by athena] Encoder registrar's names in wisdom. Remove export_conf, since a separate program can now generate it. M ./kernel/ifftw.h -5 +5 M ./kernel/planner.c -108 +32 M ./kernel/solvtab.c -1 +2 M ./tests/bench.c -3 +1 Tue Sep 3 15:11:06 EDT 2002 athena * [project @ 2002-09-03 19:11:06 by athena] Fixed typo M ./kernel/planner.c -2 +2 Tue Sep 3 14:52:45 EDT 2002 athena * [project @ 2002-09-03 18:52:45 by athena] Fixed broken trochaic meter. M ./kernel/planner.c -2 +2 Tue Sep 3 09:49:50 EDT 2002 athena * [project @ 2002-09-03 13:49:50 by athena] Initialize planner->score. It is correct to leave it uninitialized, but I don't want people to send reports about purify complaining. M ./kernel/planner.c -1 +2 Tue Sep 3 09:03:46 EDT 2002 athena * [project @ 2002-09-03 13:03:46 by athena] More latin silliness M ./kernel/planner.c -3 +7 Mon Sep 2 17:57:32 EDT 2002 stevenj * [project @ 2002-09-02 21:57:32 by stevenj] updated M ./ChangeLog +193 Mon Sep 2 17:33:49 EDT 2002 stevenj * [project @ 2002-09-02 21:33:49 by stevenj] added clock() getseconds timer M ./kernel/timer.c -1 +16 Mon Sep 2 16:16:58 EDT 2002 athena * [project @ 2002-09-02 20:16:58 by athena] Oops M ./rdft/indirect.c -2 +1 Mon Sep 2 15:58:19 EDT 2002 athena * [project @ 2002-09-02 19:58:19 by athena] Experimental INDIRECT_VERBOTEN flag (not used) M ./dft/indirect.c -1 +3 M ./kernel/ifftw.h -5 +6 M ./rdft/indirect.c -3 +4 Mon Sep 2 15:36:21 EDT 2002 athena * [project @ 2002-09-02 19:36:21 by athena] Do not allow buffering in children of indirect solvers. M ./dft/buffered.c -1 +4 M ./dft/indirect.c -1 +3 M ./kernel/ifftw.h -1 +2 M ./rdft/buffered.c -1 +4 M ./rdft/buffered2.c -1 +4 M ./rdft/indirect.c -1 +3 Mon Sep 2 15:02:11 EDT 2002 athena * [project @ 2002-09-02 19:02:11 by athena] Oops M ./kernel/planner.c -2 +2 Mon Sep 2 14:32:28 EDT 2002 athena * [project @ 2002-09-02 18:32:28 by athena] Hash sizeof(R) as part of wisdom. M ./kernel/planner.c -1 +2 Mon Sep 2 13:47:57 EDT 2002 stevenj * [project @ 2002-09-02 17:47:57 by stevenj] added --enable-float synonym for --enable-single (since with have --enable-long-double) M ./configure.ac +1 Mon Sep 2 13:46:08 EDT 2002 athena * [project @ 2002-09-02 17:46:08 by athena] zerotens is now in its own file, so it does not cause dft to be linked in if only rdft is used. A ./dft/zero.c M ./dft/Makefile.am -1 +1 M ./dft/problem.c -25 +1 M ./dft/zero.c +49 Mon Sep 2 11:56:37 EDT 2002 athena * [project @ 2002-09-02 15:56:37 by athena] Removed unused var. M ./kernel/planner.c -2 +1 Mon Sep 2 11:55:33 EDT 2002 athena * [project @ 2002-09-02 15:55:33 by athena] Split insert() in preparation for wisdom import M ./kernel/planner.c -5 +11 Mon Sep 2 11:46:57 EDT 2002 athena * [project @ 2002-09-02 15:46:57 by athena] Moved debugging infrastructure to test directory so that it is not linked into the shared library. A ./tests/debug.h A ./tests/dotens.c A ./tests/dotens2.c A ./tests/verify-dft.c A ./tests/verify-lib.c A ./tests/verify-rdft.c A ./tests/verify-reodft.c A ./tests/verify.h M ./dft/Makefile.am -1 +1 M ./dft/dft.h -4 +1 R ./dft/verify.c M ./kernel/Makefile.am -6 +5 R ./kernel/dotens.c R ./kernel/dotens2.c M ./kernel/ifftw.h -17 +1 R ./kernel/verify-lib.c R ./kernel/verify.h M ./rdft/Makefile.am -2 +2 M ./rdft/rdft.h -4 +1 R ./rdft/verify.c M ./reodft/Makefile.am -1 +1 M ./reodft/reodft.h -4 +1 R ./reodft/verify.c M ./tests/Makefile.am -2 +7 M ./tests/bench.c +3 M ./tests/debug.h +18 M ./tests/dotens.c +48 M ./tests/dotens2.c +56 M ./tests/verify-dft.c +131 M ./tests/verify-lib.c +386 M ./tests/verify-rdft.c +420 M ./tests/verify-reodft.c +536 M ./tests/verify.h +83 Mon Sep 2 11:04:54 EDT 2002 athena * [project @ 2002-09-02 15:04:53 by athena] Reactivated wisdom export M ./kernel/planner.c -42 +9 M ./kernel/print.c -1 +12 Sun Sep 1 21:30:58 EDT 2002 athena * [project @ 2002-09-02 01:30:58 by athena] Dump errors to stderr, not stdout. M ./kernel/verify-lib.c -2 +3 Sun Sep 1 21:26:38 EDT 2002 athena * [project @ 2002-09-02 01:26:38 by athena] Removed traverse.c. traverse.c is no longer need for plan blessing. I figured out a way to avoid using it in planner-score.c, so the file is now redundant. M ./kernel/Makefile.am -2 +2 M ./kernel/ifftw.h -9 +2 M ./kernel/planner-score.c -30 +37 R ./kernel/traverse.c M ./tests/bench.c -20 Sun Sep 1 19:51:50 EDT 2002 athena * [project @ 2002-09-01 23:51:50 by athena] Removed code made obsolete by new MD5 scheme: problem equality tests, scanners, and associated list of problem kinds. M ./dft/conf.c -3 +1 M ./dft/dft.h -3 +1 M ./dft/problem.c -58 +3 M ./kernel/align.c -7 +1 M ./kernel/ifftw.h -24 +4 M ./kernel/planner.c -25 +10 M ./kernel/problem.c -11 +2 M ./kernel/scan.c -44 +2 M ./kernel/scanners.c -5 +5 M ./kernel/tensor.c -46 +1 M ./rdft/conf.c -4 +1 M ./rdft/problem.c -78 +3 M ./rdft/problem2.c -58 +3 M ./rdft/rdft.h -4 +1 Sun Sep 1 19:22:54 EDT 2002 athena * [project @ 2002-09-01 23:22:53 by athena] Started md5 implementation A ./kernel/md5.c M ./dft/problem.c -9 +9 M ./kernel/Makefile.am -1 +1 M ./kernel/ifftw.h -3 +29 M ./kernel/md5.c +164 M ./kernel/planner-naive.c -3 +3 M ./kernel/planner-score.c -7 +11 M ./kernel/planner.c -27 +47 M ./kernel/tensor.c -9 +7 M ./rdft/problem.c -12 +11 M ./rdft/problem2.c -10 +11 Sat Aug 31 14:00:04 EDT 2002 athena * [project @ 2002-08-31 18:00:04 by athena] Keep track of hit rate M ./kernel/ifftw.h -1 +2 M ./kernel/planner.c -3 +9 Sat Aug 31 12:44:04 EDT 2002 athena * [project @ 2002-08-31 16:44:04 by athena] Only dump when verbose > 4 M ./kernel/planner.c -2 +2 Sat Aug 31 09:55:57 EDT 2002 athena * [project @ 2002-08-31 13:55:57 by athena] Debugging infrastructure M ./dft/indirect.c -2 +1 M ./kernel/ifftw.h -5 +6 M ./kernel/plan.c -1 +6 M ./kernel/planner.c -7 +12 M ./tests/bench.c -3 +7 Sat Aug 31 09:21:48 EDT 2002 athena * [project @ 2002-08-31 13:21:48 by athena] Use debug infrastructure to dump planner. M ./kernel/planner.c -19 +10 M ./kernel/print.c -2 +5 Fri Aug 30 21:29:10 EDT 2002 athena * [project @ 2002-08-31 01:29:10 by athena] Do not store plans in planner, plus general planner cleanup. M ./kernel/alloc.c -2 +2 M ./kernel/ifftw.h -34 +37 M ./kernel/plan.c -15 +1 M ./kernel/planner-naive.c -7 +7 M ./kernel/planner-score.c -8 +7 M ./kernel/planner.c -103 +80 M ./kernel/scan.c -3 +3 M ./kernel/scanners.c -3 +3 M ./tests/bench.c -3 +16 Fri Aug 30 18:07:52 EDT 2002 stevenj * [project @ 2002-08-30 22:07:52 by stevenj] renamed IN_DHT_R2HC to the more general FORBID_DHT_R2HC M ./kernel/ifftw.h -2 +2 M ./rdft/dht-r2hc.c -3 +3 Fri Aug 30 18:07:21 EDT 2002 stevenj * [project @ 2002-08-30 22:07:21 by stevenj] eliminated unused var M ./kernel/planner.c -2 +1 Fri Aug 30 12:09:48 EDT 2002 athena * [project @ 2002-08-30 16:09:48 by athena] Score planner was not working correctly when using wisdom. Fixed. M ./kernel/planner-naive.c -2 +8 M ./kernel/planner-score.c -2 +13 M ./kernel/planner.c -13 +18 Fri Aug 30 08:20:48 EDT 2002 athena * [project @ 2002-08-30 12:20:48 by athena] Use hash table in debug malloc M ./kernel/alloc.c -12 +25 M ./kernel/ifftw.h -1 +2 M ./kernel/stride.c -2 +2 Fri Aug 30 02:45:15 EDT 2002 stevenj * [project @ 2002-08-30 06:45:15 by stevenj] listed some good stuff M ./NEWS +47 Fri Aug 30 02:17:30 EDT 2002 stevenj * [project @ 2002-08-30 06:17:30 by stevenj] timed planner M ./TODO +4 Fri Aug 30 02:07:00 EDT 2002 stevenj * [project @ 2002-08-30 06:07:00 by stevenj] fma? M ./TODO +2 Fri Aug 30 02:05:55 EDT 2002 stevenj * [project @ 2002-08-30 06:05:55 by stevenj] update M ./TODO -3 +3 Fri Aug 30 01:31:47 EDT 2002 stevenj * [project @ 2002-08-30 05:31:47 by stevenj] rader-dht -> dht-rader A ./rdft/dht-rader.c M ./rdft/Makefile.am -4 +4 M ./rdft/conf.c -2 +2 M ./rdft/dht-rader.c +422 R ./rdft/rader-dht.c M ./rdft/rdft.h -2 +2 Fri Aug 30 01:21:37 EDT 2002 stevenj * [project @ 2002-08-30 05:21:37 by stevenj] add DHT solver, and break up rader-dht and r2hc-hc2r A ./rdft/dht-r2hc.c A ./rdft/rdft-dht.c M ./kernel/ifftw.h -2 +3 M ./rdft/Makefile.am -6 +6 M ./rdft/buffered2.c -2 +4 M ./rdft/conf.c -4 +5 M ./rdft/dht-r2hc.c +151 R ./rdft/r2hc-hc2r.c M ./rdft/rader-dht.c -73 +21 M ./rdft/rank-geq2.c -1 +12 M ./rdft/rdft-dht.c +229 M ./rdft/rdft.h -3 +4 Thu Aug 29 23:20:35 EDT 2002 stevenj * [project @ 2002-08-30 03:20:35 by stevenj] another option M ./tests/bench.c +1 Thu Aug 29 22:55:29 EDT 2002 stevenj * [project @ 2002-08-30 02:55:29 by stevenj] generalized indirect solvers for fftw2-like buffering and more M ./dft/indirect.c -22 +35 M ./kernel/ifftw.h -1 +3 M ./kernel/tensor.c -13 +17 M ./rdft/indirect.c -18 +35 Thu Aug 29 18:08:16 EDT 2002 stevenj * [project @ 2002-08-29 22:08:16 by stevenj] tensor_max_index and tensor_min_stride are now both unsigned M ./dft/vrank-geq1.c -2 +3 M ./kernel/ifftw.h -3 +3 M ./kernel/tensor.c -11 +7 M ./rdft/vrank-geq1-rdft2.c -2 +3 M ./rdft/vrank-geq1.c -2 +3 Thu Aug 29 17:58:35 EDT 2002 stevenj * [project @ 2002-08-29 21:58:35 by stevenj] added iabs.c, and tensor_min_stride returns min absolute value A ./kernel/iabs.c M ./kernel/Makefile.am -5 +5 M ./kernel/iabs.c +28 M ./kernel/ifftw.h -1 +5 M ./kernel/tensor.c -12 +6 M ./rdft/buffered2.c -9 +4 M ./rdft/problem2.c -8 +3 Thu Aug 29 17:31:39 EDT 2002 stevenj * [project @ 2002-08-29 21:31:39 by stevenj] bug fix in cldrest hc2c/c2hc copy loops M ./rdft/buffered2.c -9 +14 Thu Aug 29 13:45:08 EDT 2002 athena * [project @ 2002-08-29 17:45:08 by athena] Added things to do. M ./TODO -1 +1 Thu Aug 29 13:10:04 EDT 2002 stevenj * [project @ 2002-08-29 17:10:04 by stevenj] added automake prereq M ./configure.ac -1 +1 Thu Aug 29 08:36:36 EDT 2002 athena * [project @ 2002-08-29 12:36:36 by athena] Use indexed addressing M ./rdft/rdft2-radix2.c -61 +61 Thu Aug 29 08:20:55 EDT 2002 athena * [project @ 2002-08-29 12:20:55 by athena] Ooops M ./libbench/verify.c -3 +6 M ./rdft/rdft2-radix2.c -4 +7 Thu Aug 29 07:45:37 EDT 2002 athena * [project @ 2002-08-29 11:45:37 by athena] Oops M ./kernel/ifftw.h -10 +10 Thu Aug 29 02:32:13 EDT 2002 stevenj * [project @ 2002-08-29 06:32:13 by stevenj] updates to win32 threads code (ick) M ./threads/threads.c -4 +22 Thu Aug 29 01:44:33 EDT 2002 stevenj * [project @ 2002-08-29 05:44:33 by stevenj] added threaded version A ./acx_pthread.m4 A ./threads/ A ./threads/Makefile.am A ./threads/conf.c A ./threads/ct-dit.c A ./threads/dft-vrank-geq1.c A ./threads/hc2hc-dif.c A ./threads/hc2hc-dit.c A ./threads/rdft-vrank-geq1.c A ./threads/threads.c A ./threads/threads.h A ./threads/vrank-geq1-rdft2.c M ./Makefile.am -2 +3 M ./acx_pthread.m4 +226 M ./configure.ac +56 M ./dft/ct-dif.c -2 +2 M ./dft/ct-dit.c -2 +5 M ./dft/ct-ditbuf.c -2 +2 M ./dft/ct-ditf.c -2 +2 M ./dft/ct.c -2 +3 M ./dft/ct.h -1 +2 M ./dft/dft.h -1 +4 M ./dft/kdft-dif.c -1 +5 M ./dft/kdft-dit.c -1 +5 M ./dft/vrank-geq1.c -1 +4 M ./kernel/alloc.c -42 +56 M ./kernel/ifftw.h -7 +18 M ./kernel/planner.c -15 +24 M ./rdft/hc2hc-buf.c -1 +3 M ./rdft/hc2hc-dif.c -1 +5 M ./rdft/hc2hc-dit.c -1 +5 M ./rdft/hc2hc.c -2 +3 M ./rdft/hc2hc.h -1 +2 M ./rdft/khc2hc-dif.c -1 +5 M ./rdft/khc2hc-dit.c -1 +5 M ./rdft/rdft.h -1 +4 M ./rdft/vrank-geq1-rdft2.c -1 +4 M ./rdft/vrank-geq1.c -1 +4 M ./tests/Makefile.am -2 +4 M ./tests/bench.c -1 +5 M ./threads/Makefile.am +15 M ./threads/conf.c +41 M ./threads/ct-dit.c +151 M ./threads/dft-vrank-geq1.c +249 M ./threads/hc2hc-dif.c +168 M ./threads/hc2hc-dit.c +166 M ./threads/rdft-vrank-geq1.c +246 M ./threads/threads.c +467 M ./threads/threads.h +53 M ./threads/vrank-geq1-rdft2.c +267 Wed Aug 28 19:47:21 EDT 2002 stevenj * [project @ 2002-08-28 23:47:21 by stevenj] fix make dist M ./kernel/Makefile.am -2 +3 Wed Aug 28 15:09:03 EDT 2002 stevenj * [project @ 2002-08-28 19:09:03 by stevenj] whoops, bugfix for inverse M ./rdft/rank-geq2-rdft2.c -4 +9 Wed Aug 28 14:50:34 EDT 2002 athena * [project @ 2002-08-28 18:50:34 by athena] Use C9x convention for naming (fftwf etc.). Removed installable header files since they will be part of the API. M ./Makefile.am -5 +5 M ./configure.ac -2 +9 M ./kernel/Makefile.am -1 R ./kernel/dfftw3.h R ./kernel/fftw3.h M ./kernel/ifftw.h -2 +17 R ./kernel/lfftw3.h R ./kernel/sfftw3.h M ./tests/Makefile.am -1 +1 Tue Aug 27 23:34:00 EDT 2002 stevenj * [project @ 2002-08-28 03:34:00 by stevenj] allow _1 variants to accept rnk 0 (sz 1) problems M ./rdft/problem.c -3 +3 Tue Aug 27 15:56:09 EDT 2002 stevenj * [project @ 2002-08-27 19:56:09 by stevenj] updated M ./ChangeLog +619 Mon Aug 26 20:14:56 EDT 2002 athena * [project @ 2002-08-27 00:14:56 by athena] Loop unroll is useless M ./dft/rank0.c -28 +6 Mon Aug 26 20:00:41 EDT 2002 athena * [project @ 2002-08-27 00:00:41 by athena] Use indexed addressing M ./dft/ct-ditbuf.c -12 +9 Mon Aug 26 19:46:46 EDT 2002 athena * [project @ 2002-08-26 23:46:46 by athena] Use indexed addressing in transpose routines. (Seems to be slightly better on athlon.) M ./dft/vrank2-transpose.c -16 +8 M ./dft/vrank3-transpose.c -15 +8 Mon Aug 26 12:59:44 EDT 2002 stevenj * [project @ 2002-08-26 16:59:44 by stevenj] added comment about stability M ./reodft/redft00e-r2hc.c -1 +2 M ./reodft/reodft11e-r2hc.c -1 +3 M ./reodft/rodft00e-r2hc.c -1 +2 Mon Aug 26 07:43:53 EDT 2002 athena * [project @ 2002-08-26 11:43:53 by athena] Approximate opcount M ./rdft/rdft2-radix2.c -6 +10 Mon Aug 26 06:38:49 EDT 2002 athena * [project @ 2002-08-26 10:38:49 by athena] Finished rdft2 via dft/rdft M ./dft/rank-geq2.c -2 +2 M ./rdft/rank-geq2.c -2 +2 M ./rdft/rdft2-radix2.c -9 +149 Mon Aug 26 00:15:59 EDT 2002 stevenj * [project @ 2002-08-26 04:15:59 by stevenj] some updates M ./TODO -9 +9 Mon Aug 26 00:05:53 EDT 2002 stevenj * [project @ 2002-08-26 04:05:52 by stevenj] rdft kind is now per-dimension, added rdft/rank-geq2 A ./rdft/rank-geq2.c M ./rdft/Makefile.am -2 +3 M ./rdft/buffered.c -3 +4 M ./rdft/buffered2.c -5 +5 M ./rdft/conf.c -3 +2 M ./rdft/dft-r2hc.c -2 +2 M ./rdft/direct.c -6 +6 M ./rdft/generic.c -5 +5 M ./rdft/hc2hc.c -10 +10 M ./rdft/indirect.c -2 +2 M ./rdft/problem.c -21 +75 M ./rdft/r2hc-hc2r.c -4 +4 M ./rdft/rader-dht.c -4 +4 M ./rdft/rader-hc2hc.c -4 +4 M ./rdft/rank-geq2.c +231 M ./rdft/rdft.h -4 +9 M ./rdft/rdft2-radix2.c -2 +2 M ./rdft/verify.c -5 +5 M ./reodft/redft00e-r2hc.c -3 +3 M ./reodft/reodft010e-r2hc.c -6 +6 M ./reodft/reodft11e-r2hc.c -5 +5 M ./reodft/rodft00e-r2hc.c -3 +3 M ./reodft/verify.c -4 +4 M ./tests/bench.c -2 +4 Sun Aug 25 22:45:38 EDT 2002 stevenj * [project @ 2002-08-26 02:45:38 by stevenj] added note M ./rdft/problem.c -2 +3 Sun Aug 25 22:28:12 EDT 2002 stevenj * [project @ 2002-08-26 02:28:12 by stevenj] must zero real sz M ./rdft/problem.c -2 +4 Sun Aug 25 22:06:52 EDT 2002 stevenj * [project @ 2002-08-26 02:06:52 by stevenj] unified pickdim funcs A ./kernel/pickdim.c M ./dft/rank-geq2.c -32 +10 M ./dft/vrank-geq1.c -45 +3 M ./kernel/Makefile.am -1 +1 M ./kernel/ifftw.h -1 +6 M ./kernel/pickdim.c +82 M ./rdft/vrank-geq1-rdft2.c -45 +3 M ./rdft/vrank-geq1.c -45 +3 M ./tests/bench.c -1 +1 Sun Aug 25 14:10:55 EDT 2002 fftw * [project @ 2002-08-25 18:10:55 by fftw] silence warnings M ./libbench/mp.c -14 M ./rdft/codelet.h -2 +2 M ./rdft/indirect.c -3 +1 M ./rdft/rank-geq2-rdft2.c -5 +2 M ./rdft/verify.c -10 +7 M ./reodft/reodft010e-r2hc.c -2 +2 M ./reodft/reodft11e-r2hc.c -7 +2 M ./reodft/verify.c -2 +2 Sun Aug 25 13:16:49 EDT 2002 athena * [project @ 2002-08-25 17:16:49 by athena] I had to add another planner flag to record whether pointers could become unaligned because of vrank-geq1 solvers (these solvers only plan the first element of a vector problem, but the second element may have a different alignment). This addition is ugly, but I don't see any way around it. M ./dft/codelet.h -3 +5 M ./dft/codelets/n.c -2 +3 M ./dft/codelets/t.c -2 +2 M ./dft/ct-dif.c -3 +3 M ./dft/ct-dit.c -3 +3 M ./dft/ct-ditbuf.c -3 +3 M ./dft/ct-ditf.c -2 +2 M ./dft/direct.c -5 +6 M ./dft/simd/n1b.c -1 +3 M ./dft/simd/n1f.c -1 +3 M ./dft/simd/t1b.c -1 +3 M ./dft/simd/t1f.c -1 +3 M ./dft/vrank-geq1.c -2 +11 M ./kernel/ifftw.h -4 +8 M ./rdft/vrank-geq1-rdft2.c -2 +8 M ./rdft/vrank-geq1.c -2 +7 Sun Aug 25 10:18:25 EDT 2002 athena * [project @ 2002-08-25 14:18:25 by athena] Added thoughts M ./TODO -1 +2 Sun Aug 25 10:08:59 EDT 2002 athena * [project @ 2002-08-25 14:08:59 by athena] Implemented rdft2 via vector rdft + radix2 step A ./rdft/rdft2-radix2.c M ./rdft/Makefile.am -1 +1 M ./rdft/conf.c -2 +2 M ./rdft/rdft.h -2 +2 R ./rdft/rdft2-dft.c M ./rdft/rdft2-radix2.c +325 Sat Aug 24 17:43:54 EDT 2002 athena * [project @ 2002-08-24 21:43:54 by athena] Stylistic changes M ./rdft/rdft2-dft.c -5 +5 Sat Aug 24 11:19:30 EDT 2002 athena * [project @ 2002-08-24 15:19:30 by athena] Simplified mktwiddle interface M ./dft/ct.c -7 +3 M ./dft/generic.c -5 +2 M ./kernel/ifftw.h -1 +3 M ./kernel/twiddle.c -1 +11 M ./rdft/generic.c -6 +3 M ./rdft/rdft2-dft.c -11 +5 M ./reodft/redft00e-r2hc.c -6 +2 M ./reodft/reodft010e-r2hc.c -5 +2 M ./reodft/reodft11e-r2hc.c -8 +3 M ./reodft/rodft00e-r2hc.c -5 +2 Sat Aug 24 11:05:08 EDT 2002 athena * [project @ 2002-08-24 15:05:08 by athena] Unification of certain vector computations. rdft2-dft is now a vector transform. M ./dft/ct-dif.c -2 +2 M ./dft/ct-dit.c -2 +2 M ./dft/ct.c -16 +2 M ./dft/ct.h -4 +1 M ./kernel/ifftw.h -1 +2 M ./kernel/tensor.c -1 +17 M ./rdft/hc2hc-dif.c -2 +2 M ./rdft/hc2hc-dit.c -2 +2 M ./rdft/hc2hc.c -18 +2 M ./rdft/hc2hc.h -4 +1 M ./rdft/rdft2-dft.c -54 +66 Fri Aug 23 20:21:25 EDT 2002 athena * [project @ 2002-08-24 00:21:25 by athena] Intel compiler seems to be still buggy M ./configure.ac +1 M ./simd/sse.c -1 +3 M ./simd/sse2.c -1 +3 Fri Aug 23 16:07:12 EDT 2002 athena * [project @ 2002-08-23 20:07:12 by athena] Streamlined twiddle protocol M ./dft/ct-dif.c -2 +3 M ./dft/ct-dit.c -2 +3 M ./dft/ct-ditbuf.c -2 +2 M ./dft/ct-ditf.c -2 +2 M ./dft/ct.c -13 +6 M ./dft/ct.h -3 +2 M ./dft/generic.c -16 +6 M ./dft/indirect.c -3 +1 M ./kernel/ifftw.h -3 +3 M ./kernel/twiddle.c -7 +14 M ./rdft/generic.c -18 +8 M ./rdft/hc2hc.c -10 +6 M ./rdft/rdft2-dft.c -19 +10 M ./reodft/redft00e-r2hc.c -15 +6 M ./reodft/reodft010e-r2hc.c -19 +10 M ./reodft/reodft11e-r2hc.c -21 +9 M ./reodft/rodft00e-r2hc.c -15 +6 Fri Aug 23 13:22:17 EDT 2002 athena * [project @ 2002-08-23 17:22:17 by athena] Implemented rdft2 via dft (forward only for now) A ./rdft/rdft2-dft.c M ./libbench/verify.c -1 +10 M ./rdft/Makefile.am -1 +1 M ./rdft/conf.c -1 +2 M ./rdft/rdft.h -1 +2 M ./rdft/rdft2-dft.c +213 Thu Aug 22 11:29:29 EDT 2002 athena * [project @ 2002-08-22 15:29:29 by athena] More cleanup of verify M ./kernel/verify-lib.c -21 +13 M ./libbench/verify.c -45 +12 Thu Aug 22 11:16:03 EDT 2002 athena * [project @ 2002-08-22 15:16:03 by athena] Changed error criterion because old one was too strict M ./kernel/verify-lib.c -40 +21 Thu Aug 22 11:15:17 EDT 2002 athena * [project @ 2002-08-22 15:15:17 by athena] Disable shared M ./bootstrap.sh -1 +1 Thu Aug 22 09:19:12 EDT 2002 athena * [project @ 2002-08-22 13:19:12 by athena] Added thoughts M ./TODO +4 Thu Aug 22 09:17:28 EDT 2002 athena * [project @ 2002-08-22 13:17:28 by athena] Oops M ./dft/generic.c -1 +2 Thu Aug 22 09:11:34 EDT 2002 athena * [project @ 2002-08-22 13:11:34 by athena] Do not use inline. Minor changes. M ./dft/generic.c -1 +1 M ./kernel/alloc.c -2 +2 M ./kernel/planner-score.c -2 +2 M ./kernel/tensor.c -3 +3 Wed Aug 21 16:23:26 EDT 2002 stevenj * [project @ 2002-08-21 20:23:26 by stevenj] more commented flags M ./tests/bench.c +2 Tue Aug 20 19:44:43 EDT 2002 stevenj * [project @ 2002-08-20 23:44:43 by stevenj] added DCT-IV and DST-IV A ./reodft/reodft11e-r2hc.c M ./reodft/Makefile.am -1 +2 M ./reodft/conf.c -1 +2 M ./reodft/reodft11e-r2hc.c +297 M ./reodft/verify.c -1 +15 M ./tests/bench.c -1 +5 Tue Aug 20 16:01:36 EDT 2002 athena * [project @ 2002-08-20 20:01:36 by athena] Slight improvement in twiddle scheme M ./genfft/twiddle.ml -2 +9 Tue Aug 20 15:31:54 EDT 2002 stevenj * [project @ 2002-08-20 19:31:54 by stevenj] name fix M ./reodft/conf.c -2 +2 M ./reodft/reodft.h -6 +4 M ./reodft/reodft010e-r2hc.c -2 +2 Tue Aug 20 15:16:48 EDT 2002 stevenj * [project @ 2002-08-20 19:16:48 by stevenj] removed extraneous variable M ./reodft/reodft010e-r2hc.c -5 +1 Tue Aug 20 11:46:29 EDT 2002 athena * [project @ 2002-08-20 15:46:29 by athena] Oops M ./libbench/mp.c -33 +79 M ./libbench/verify.c -33 +25 Tue Aug 20 08:37:45 EDT 2002 athena * [project @ 2002-08-20 12:37:45 by athena] Still playing around M ./genfft/twiddle.ml -35 +62 M ./kernel/trig.c -1 +3 Mon Aug 19 19:56:29 EDT 2002 athena * [project @ 2002-08-19 23:56:29 by athena] Playing around with addition chain M ./TODO -6 +11 M ./genfft/algsimp.ml -3 +4 M ./genfft/expr.ml -1 +6 M ./genfft/expr.mli -1 +2 M ./genfft/twiddle.ml -46 +64 M ./support/addchain.c -8 +25 Mon Aug 19 19:48:56 EDT 2002 stevenj * [project @ 2002-08-19 23:48:56 by stevenj] comments M ./reodft/redft00e-r2hc.c -1 +4 M ./reodft/rodft00e-r2hc.c -1 +4 Mon Aug 19 19:45:35 EDT 2002 stevenj * [project @ 2002-08-19 23:45:35 by stevenj] comment fixes M ./reodft/reodft010e-r2hc.c -5 +2 Mon Aug 19 19:40:18 EDT 2002 stevenj * [project @ 2002-08-19 23:40:18 by stevenj] added reodft stuff A ./reodft/ A ./reodft/Makefile.am A ./reodft/conf.c A ./reodft/redft00e-r2hc.c A ./reodft/reodft.h A ./reodft/reodft010e-r2hc.c A ./reodft/rodft00e-r2hc.c A ./reodft/verify.c M ./Makefile.am -2 +3 M ./configure.ac +2 M ./dft/dft.h -1 +5 M ./rdft/rdft.h -1 +5 M ./reodft/Makefile.am +6 M ./reodft/conf.c +37 M ./reodft/redft00e-r2hc.c +204 M ./reodft/reodft.h +44 M ./reodft/reodft010e-r2hc.c +400 M ./reodft/rodft00e-r2hc.c +199 M ./reodft/verify.c +521 M ./tests/Makefile.am -1 +1 M ./tests/bench.c -2 +11 Sun Aug 18 19:44:14 EDT 2002 athena * [project @ 2002-08-18 23:44:14 by athena] Sync with nbenchfft M ./libbench/Makefile.am +4 M ./libbench/verify.c -4 +3 Sun Aug 18 16:02:37 EDT 2002 athena * [project @ 2002-08-18 20:02:37 by athena] Economy of thought M ./genfft/complex.ml -23 +12 M ./genfft/complex.mli -3 +3 M ./genfft/twiddle.ml -5 +5 Sat Aug 17 15:52:05 EDT 2002 stevenj * [project @ 2002-08-17 19:52:05 by stevenj] distribute addchain.c M ./support/Makefile.am -1 +1 Sat Aug 17 14:09:11 EDT 2002 athena * [project @ 2002-08-17 18:09:11 by athena] Nothing serious M ./support/addchain.c -4 +7 Sat Aug 17 10:47:59 EDT 2002 athena * [project @ 2002-08-17 14:47:59 by athena] New twiddle policy (disabled for now) A ./support/addchain.c M ./genfft/twiddle.ml -1 +98 M ./support/addchain.c +151 Fri Aug 16 23:44:28 EDT 2002 stevenj * [project @ 2002-08-17 03:44:28 by stevenj] bug fix for hc2r (must use inverse dft) M ./rdft/rank-geq2-rdft2.c -2 +2 Fri Aug 16 20:27:10 EDT 2002 athena * [project @ 2002-08-17 00:27:10 by athena] New log3 twiddle policy M ./dft/codelets/inplace/Makefile.am -2 +2 M ./dft/codelets/standard/Makefile.am -1 +1 M ./genfft/twiddle.ml -9 +88 M ./rdft/codelets/hc2r/Makefile.am +6 M ./rdft/codelets/r2hc/Makefile.am -1 +1 Fri Aug 16 18:10:33 EDT 2002 athena * [project @ 2002-08-16 22:10:33 by athena] More verify cleanup M ./dft/verify.c -20 +2 M ./kernel/verify-lib.c -1 +19 M ./kernel/verify.h +3 M ./rdft/verify.c -23 +5 Fri Aug 16 16:31:19 EDT 2002 athena * [project @ 2002-08-16 20:31:19 by athena] Oops M ./rdft/verify.c -3 +3 Fri Aug 16 15:22:36 EDT 2002 athena * [project @ 2002-08-16 19:22:36 by athena] Economy of thought (and code) A ./kernel/verify-lib.c A ./kernel/verify.h M ./dft/verify.c -305 +12 M ./kernel/Makefile.am -2 +2 M ./kernel/verify-lib.c +394 M ./kernel/verify.h +80 M ./rdft/verify.c -364 +23 Fri Aug 16 14:05:45 EDT 2002 athena * [project @ 2002-08-16 18:05:45 by athena] Added comment M ./TODO +2 Fri Aug 16 12:57:43 EDT 2002 athena * [project @ 2002-08-16 16:57:43 by athena] Cleaner rounding algorithm M ./libbench/mp.c -16 +27 Fri Aug 16 11:27:43 EDT 2002 athena * [project @ 2002-08-16 15:27:43 by athena] Can get away with shorter length in bluestein (I think). M ./libbench/mp.c -1 +1 Fri Aug 16 11:08:09 EDT 2002 athena * [project @ 2002-08-16 15:08:09 by athena] Portability improvements M ./libbench/mp.c -11 +4 Fri Aug 16 08:06:31 EDT 2002 athena * [project @ 2002-08-16 12:06:31 by athena] Optionally average accuracy test over many rounds M ./libbench/bench-main.c -2 +9 M ./libbench/bench.h -2 +2 M ./libbench/verify.c -29 +43 Fri Aug 16 07:50:24 EDT 2002 athena * [project @ 2002-08-16 11:50:24 by athena] More accurate formula for trig tables M ./dft/rader.c -3 +3 M ./rdft/rader-dht.c -3 +3 M ./rdft/rader-hc2hc.c -3 +3 Fri Aug 16 06:42:02 EDT 2002 athena * [project @ 2002-08-16 10:42:02 by athena] Implemented accuracy test for all integers M ./libbench/mp.c -6 +123 M ./libbench/verify.c -2 +1 Thu Aug 15 18:54:44 EDT 2002 athena * [project @ 2002-08-15 22:54:44 by athena] inv, neg: make static M ./libbench/mp.c -2 +2 Thu Aug 15 17:25:37 EDT 2002 athena * [project @ 2002-08-15 21:25:37 by athena] Verify was not complete for real transforms M ./libbench/verify.c -9 +34 Thu Aug 15 16:30:03 EDT 2002 athena * [project @ 2002-08-15 20:30:03 by athena] Oops M ./libbench/verify.c -3 +1 Thu Aug 15 16:29:16 EDT 2002 athena * [project @ 2002-08-15 20:29:16 by athena] Fixed hb codelets M ./genfft/gen_hc2hc.ml -3 +5 M ./libbench/verify.c -1 +4 Thu Aug 15 14:10:45 EDT 2002 athena * [project @ 2002-08-15 18:10:45 by athena] Changed twiddle policy M ./dft/codelets/inplace/Makefile.am -2 +2 M ./dft/codelets/standard/Makefile.am -1 +1 M ./rdft/codelets/r2hc/Makefile.am -1 +1 Thu Aug 15 13:32:24 EDT 2002 stevenj * [project @ 2002-08-15 17:32:24 by stevenj] whoops M ./rdft/direct2.c -3 +3 Thu Aug 15 11:01:04 EDT 2002 athena * [project @ 2002-08-15 15:01:04 by athena] No point in libbench being a shared library M ./libbench/Makefile.am -2 +2 M ./tests/Makefile.am -1 +1 Thu Aug 15 09:48:37 EDT 2002 athena * [project @ 2002-08-15 13:48:37 by athena] Moved accuracy test to libbench A ./libbench/mp.c M ./libbench/Makefile.am -1 +1 M ./libbench/bench-main.c -4 +13 M ./libbench/bench.h -1 +3 M ./libbench/mp.c +439 M ./libbench/util.c +8 M ./libbench/verify.c -3 +66 M ./tests/Makefile.am -5 +1 R ./tests/accuracy.c R ./tests/mp.c Wed Aug 14 19:48:23 EDT 2002 athena * [project @ 2002-08-14 23:48:23 by athena] Modified accuracy test M ./tests/accuracy.c -33 +20 Wed Aug 14 08:34:26 EDT 2002 athena * [project @ 2002-08-14 12:34:26 by athena] Fixes for long double M ./tests/accuracy.c -2 +3 M ./tests/mp.c -1 +1 Wed Aug 14 08:17:57 EDT 2002 athena * [project @ 2002-08-14 12:17:57 by athena] Normalize input M ./tests/accuracy.c -3 +17 Wed Aug 14 07:26:41 EDT 2002 athena * [project @ 2002-08-14 11:26:41 by athena] Oops M ./tests/accuracy.c +1 Wed Aug 14 07:25:34 EDT 2002 athena * [project @ 2002-08-14 11:25:34 by athena] Also compute relative error M ./tests/accuracy.c -2 +13 Wed Aug 14 07:08:20 EDT 2002 athena * [project @ 2002-08-14 11:08:20 by athena] Loop over N M ./tests/accuracy.c -24 +32 Wed Aug 14 06:54:50 EDT 2002 athena * [project @ 2002-08-14 10:54:50 by athena] simple-minded accuracy test A ./tests/accuracy.c A ./tests/mp.c M ./tests/Makefile.am -1 +6 M ./tests/accuracy.c +48 M ./tests/mp.c +434 Wed Aug 14 03:26:06 EDT 2002 stevenj * [project @ 2002-08-14 07:26:06 by stevenj] whoops A ./rdft/rank-geq2-rdft2.c Tue Aug 13 11:42:41 EDT 2002 athena * [project @ 2002-08-13 15:42:41 by athena] fma() stuff is too nonportable, removed M ./kernel/trig.c -45 +1 Mon Aug 12 14:07:44 EDT 2002 stevenj * [project @ 2002-08-12 18:07:44 by stevenj] slight fix M ./rdft/problem.c -2 +2 Mon Aug 12 14:07:18 EDT 2002 stevenj * [project @ 2002-08-12 18:07:18 by stevenj] use table for rdft_kind_str M ./rdft/problem.c -26 +12 Mon Aug 12 13:43:08 EDT 2002 stevenj * [project @ 2002-08-12 17:43:08 by stevenj] slight fixes M ./rdft/problem2.c -6 +11 Mon Aug 12 13:31:37 EDT 2002 stevenj * [project @ 2002-08-12 17:31:37 by stevenj] multidimensional rdft2 M ./kernel/ifftw.h -2 +3 M ./kernel/planner.c -1 +2 M ./kernel/tensor.c -5 +6 M ./rdft/Makefile.am -2 +2 M ./rdft/buffered2.c -13 +18 M ./rdft/conf.c -1 +2 M ./rdft/direct2.c -9 +14 M ./rdft/nop2.c -2 +12 M ./rdft/problem2.c -36 +41 M ./rdft/rdft.h -7 +9 M ./rdft/vrank-geq1-rdft2.c -6 +23 M ./tests/bench.c -11 +36 Sat Aug 10 19:33:23 EDT 2002 stevenj * [project @ 2002-08-10 23:33:23 by stevenj] use tensor_copy_inplace M ./rdft/indirect.c -13 +5 Sat Aug 10 19:32:03 EDT 2002 stevenj * [project @ 2002-08-10 23:32:03 by stevenj] bugfix, use tensor_copy_inplace M ./dft/rank-geq2.c -4 +10 Sat Aug 10 19:30:39 EDT 2002 stevenj * [project @ 2002-08-10 23:30:39 by stevenj] use tensor_copy_inplace M ./dft/indirect.c -13 +5 Sat Aug 10 19:28:07 EDT 2002 stevenj * [project @ 2002-08-10 23:28:07 by stevenj] added tensor_copy_inplace M ./kernel/ifftw.h -1 +3 M ./kernel/tensor.c -1 +19 Sat Aug 10 19:25:50 EDT 2002 stevenj * [project @ 2002-08-10 23:25:50 by stevenj] fixed trig-function table type M ./kernel/twiddle.c -2 +2 Sat Aug 10 14:41:04 EDT 2002 athena * [project @ 2002-08-10 18:41:04 by athena] Improved trig scheme M ./kernel/trig.c -8 +53 M ./tests/trigtest.c -8 +74 Fri Aug 9 21:05:01 EDT 2002 athena * [project @ 2002-08-10 01:05:01 by athena] Allow for testing using long double instead of pari M ./tests/trigtest.c -15 +27 Fri Aug 9 20:49:32 EDT 2002 athena * [project @ 2002-08-10 00:49:32 by athena] Yet another trig scheme. M ./kernel/trig.c -25 +24 M ./tests/trigtest.c -25 +22 Fri Aug 9 20:38:07 EDT 2002 athena * [project @ 2002-08-10 00:38:07 by athena] Yet another scheme M ./kernel/trig.c -4 +12 M ./tests/trigtest.c +7 Fri Aug 9 20:31:16 EDT 2002 athena * [project @ 2002-08-10 00:31:16 by athena] Careful with overflow M ./kernel/ifftw.h -4 +4 M ./kernel/trig.c -23 +23 M ./tests/trigtest.c -24 +17 Fri Aug 9 20:16:23 EDT 2002 athena * [project @ 2002-08-10 00:16:23 by athena] Avoid overflow M ./kernel/ifftw.h -4 +4 M ./kernel/trig.c -20 +24 M ./tests/trigtest.c -21 +43 Fri Aug 9 19:26:57 EDT 2002 athena * [project @ 2002-08-09 23:26:57 by athena] New(er) trig routines M ./dft/rader.c -8 +6 M ./dft/verify.c -4 +3 M ./kernel/ifftw.h -4 +4 M ./kernel/trig.c -22 +17 M ./kernel/twiddle.c -9 +7 M ./rdft/rader-dht.c -4 +3 M ./rdft/rader-hc2hc.c -8 +6 M ./rdft/verify.c -4 +3 M ./tests/trigtest.c -20 +19 Fri Aug 9 19:25:44 EDT 2002 athena * [project @ 2002-08-09 23:25:44 by athena] Oops M ./tests/bench.c -1 +1 Fri Aug 9 18:49:04 EDT 2002 athena * [project @ 2002-08-09 22:49:04 by athena] New file A ./tests/trigtest.c Fri Aug 9 13:04:00 EDT 2002 athena * [project @ 2002-08-09 17:04:00 by athena] Commented about likely gcc bug M ./TODO +4 Fri Aug 9 13:01:49 EDT 2002 athena * [project @ 2002-08-09 17:01:49 by athena] Improved accuracy of twiddle factors A ./kernel/trig.c M ./dft/rader.c -8 +8 M ./dft/verify.c -5 +4 M ./kernel/Makefile.am -1 +1 M ./kernel/ifftw.h -10 +7 M ./kernel/trig.c +79 M ./kernel/twiddle.c -8 +10 M ./rdft/rader-dht.c -4 +4 M ./rdft/rader-hc2hc.c -8 +8 M ./rdft/verify.c -5 +4 M ./tests/bench.c -1 +1 Thu Aug 8 06:36:23 EDT 2002 athena * [project @ 2002-08-08 10:36:23 by athena] Wrong comment M ./simd/simd-3dnow.h -1 Wed Aug 7 17:14:09 EDT 2002 athena * [project @ 2002-08-07 21:14:09 by athena] Experimental 3dnow port using gcc, to compare it with Stefan's stuff. A ./simd/3dnow.c A ./simd/simd-3dnow.h M ./configure.ac +6 M ./genfft/gen_notw_c.ml -3 +6 M ./genfft/gen_twiddle_c.ml -2 +4 M ./kernel/ifftw.h -2 +2 M ./simd/3dnow.c +66 M ./simd/Makefile.am -2 +2 M ./simd/simd-3dnow.h +164 M ./simd/simd-altivec.h +2 M ./simd/simd-sse.h +3 M ./simd/simd-sse2.h +3 M ./simd/simd.h +4 Wed Aug 7 12:58:10 EDT 2002 athena * [project @ 2002-08-07 16:58:10 by athena] End of AREF experiment M ./genfft/c.ml -2 +2 M ./kernel/ifftw.h -3 +1 Wed Aug 7 07:47:19 EDT 2002 athena * [project @ 2002-08-07 11:47:19 by athena] Oops M ./configure.ac -4 +1 Wed Aug 7 07:46:38 EDT 2002 athena * [project @ 2002-08-07 11:46:38 by athena] Pathetic attempt to reduce size of configure script M ./configure.ac -16 +8 Tue Aug 6 20:38:11 EDT 2002 athena * [project @ 2002-08-07 00:38:11 by athena] Changed array syntax for experiments. M ./genfft/c.ml -8 +10 M ./kernel/ifftw.h -1 +3 Tue Aug 6 19:58:20 EDT 2002 athena * [project @ 2002-08-06 23:58:20 by athena] Fix warning M ./simd/simd-sse2.h +2 Tue Aug 6 13:35:28 EDT 2002 athena * [project @ 2002-08-06 17:35:28 by athena] Move nonportable stuff in one place. M ./dft/problem.c -2 +2 M ./kernel/align.c -1 +8 M ./kernel/ifftw.h -1 +2 M ./rdft/problem.c -2 +2 M ./rdft/problem2.c -2 +2 Tue Aug 6 10:32:53 EDT 2002 athena * [project @ 2002-08-06 14:32:53 by athena] Economy of thought: I didn't like having two algorithms for removing solutions, both correct. At least now we have the same algorithm copied twice. M ./kernel/planner.c -12 +22 Tue Aug 6 09:12:21 EDT 2002 athena * [project @ 2002-08-06 13:12:21 by athena] Added things to do M ./TODO -1 +1 Mon Aug 5 19:54:31 EDT 2002 stevenj * [project @ 2002-08-05 23:54:31 by stevenj] improved interaction of planner with patience flags M ./kernel/ifftw.h -1 +2 M ./kernel/planner.c -12 +34 Mon Aug 5 14:17:58 EDT 2002 stevenj * [project @ 2002-08-05 18:17:58 by stevenj] set up for real-even/odd DFTs, where n is not the size of the data M ./rdft/buffered.c -3 +3 M ./rdft/codelet.h -7 +25 M ./rdft/indirect.c -4 +8 M ./rdft/problem.c -13 +58 M ./rdft/rader-hc2hc.c -1 +1 M ./rdft/rdft.h -1 +3 M ./tests/bench.c -2 +3 Sun Aug 4 23:57:51 EDT 2002 stevenj * [project @ 2002-08-05 03:57:51 by stevenj] DESTROY_INPUT flag M ./dft/ct-dif.c -4 +5 M ./dft/ct-dit.c -3 +5 M ./dft/ct-ditbuf.c -3 +5 M ./dft/ct-ditf.c -3 +5 M ./dft/ct.c -2 +2 M ./dft/ct.h -2 +3 M ./kernel/ifftw.h -2 +3 M ./rdft/hc2hc-buf.c -3 +6 M ./rdft/hc2hc-dif.c -3 +5 M ./rdft/hc2hc-dit.c -3 +5 M ./rdft/hc2hc.c -2 +2 M ./rdft/hc2hc.h -2 +3 M ./rdft/r2hc-hc2r.c -13 +20 M ./tests/bench.c +2 Sun Aug 4 22:50:19 EDT 2002 stevenj * [project @ 2002-08-05 02:50:19 by stevenj] CLASSIC -> IMPATIENT M ./dft/rank-geq2.c -2 +2 M ./dft/vrank-geq1.c -2 +2 M ./kernel/ifftw.h -4 +4 M ./kernel/planner.c -3 +3 M ./rdft/dft-r2hc.c -2 +2 M ./rdft/vrank-geq1-rdft2.c -2 +2 M ./rdft/vrank-geq1.c -2 +2 M ./tests/bench.c -1 +1 Sun Aug 4 19:05:43 EDT 2002 athena * [project @ 2002-08-04 23:05:43 by athena] Require make maintainer-clean to remove the generator, as opposed to make clean. In this way we can type make clean without regenerating all codelets. M ./genfft/Makefile.am -11 +11 M ./genfft-k7/Makefile.am -3 +3 Sun Aug 4 17:34:04 EDT 2002 stevenj * [project @ 2002-08-04 21:34:04 by stevenj] ESTIMATE plans are not blessed M ./kernel/planner.c -10 +13 Sun Aug 4 17:24:37 EDT 2002 stevenj * [project @ 2002-08-04 21:24:37 by stevenj] use flags in wisdom M ./kernel/ifftw.h -2 +4 M ./kernel/planner.c -9 +14 Sun Aug 4 17:03:45 EDT 2002 stevenj * [project @ 2002-08-04 21:03:45 by stevenj] score now takes plnr, not flags, as arg M ./dft/buffered.c -3 +3 M ./dft/ct-dif.c -4 +3 M ./dft/ct-dit.c -3 +3 M ./dft/ct-ditbuf.c -3 +3 M ./dft/ct-ditf.c -3 +3 M ./dft/direct.c -3 +3 M ./dft/generic.c -2 +2 M ./dft/indirect.c -3 +3 M ./dft/nop.c -3 +3 M ./dft/rader.c -4 +4 M ./dft/rank-geq2.c -3 +3 M ./dft/rank0.c -3 +3 M ./dft/vrank-geq1.c -4 +4 M ./dft/vrank2-transpose.c -3 +3 M ./dft/vrank3-transpose.c -3 +3 M ./kernel/ifftw.h -2 +2 M ./kernel/planner-score.c -4 +3 M ./kernel/tensor.c -3 +3 M ./rdft/buffered.c -3 +3 M ./rdft/buffered2.c -3 +3 M ./rdft/dft-r2hc.c -3 +3 M ./rdft/direct.c -3 +3 M ./rdft/direct2.c -3 +3 M ./rdft/generic.c -2 +2 M ./rdft/hc2hc-buf.c -3 +3 M ./rdft/hc2hc-dif.c -3 +3 M ./rdft/hc2hc-dit.c -3 +3 M ./rdft/indirect.c -3 +3 M ./rdft/nop.c -3 +3 M ./rdft/nop2.c -3 +3 M ./rdft/r2hc-hc2r.c -3 +3 M ./rdft/rader-dht.c -2 +2 M ./rdft/rader-hc2hc.c -2 +2 M ./rdft/rank0.c -3 +3 M ./rdft/vrank-geq1-rdft2.c -4 +4 M ./rdft/vrank-geq1.c -4 +4 M ./rdft/vrank2-transpose.c -3 +3 M ./rdft/vrank3-transpose.c -3 +3 M ./tests/bench.c -1 +1 Sun Aug 4 16:37:46 EDT 2002 stevenj * [project @ 2002-08-04 20:37:46 by stevenj] align initial stack in alignment check, which should now pass for gcc 3.1.1 M ./acinclude.m4 -2 +3 Sat Aug 3 20:04:57 EDT 2002 athena * [project @ 2002-08-04 00:04:57 by athena] Detect ultrasparc (sort of) M ./acinclude.m4 +3 Sat Aug 3 19:38:17 EDT 2002 stevenj * [project @ 2002-08-03 23:38:17 by stevenj] added solvtab_rdft_r2r placeholder M ./rdft/codelet.h -1 +3 Sat Aug 3 19:34:49 EDT 2002 athena * [project @ 2002-08-03 23:34:49 by athena] Damn solaris M ./support/Makefile.codelets -2 +2 Sat Aug 3 17:55:44 EDT 2002 stevenj * [project @ 2002-08-03 21:55:44 by stevenj] use E extended precision in solvers M ./rdft/problem.c -7 +9 Sat Aug 3 17:53:29 EDT 2002 stevenj * [project @ 2002-08-03 21:53:29 by stevenj] an alternative notation for D{C,S}T: DXTio, where i/o are {0,1} according to whether the input/output are shifted, respectively. Alternatively, io is the binary representation of the usual DXT-{I,II,III,IV} nomenclature, minus 1. M ./rdft/codelet.h -2 +2 Sat Aug 3 17:49:11 EDT 2002 stevenj * [project @ 2002-08-03 21:49:11 by stevenj] use E extended precision in solvers M ./dft/generic.c -5 +5 M ./dft/rader.c -2 +2 M ./rdft/generic.c -21 +21 M ./rdft/r2hc-hc2r.c -3 +3 M ./rdft/rader-dht.c -4 +4 Sat Aug 3 15:39:49 EDT 2002 athena * [project @ 2002-08-03 19:39:49 by athena] More portability fixes, compiler bugs workarounds, etc. M ./configure.ac -2 +4 M ./kernel/cycle.h -2 +2 M ./kernel/planner.c -2 +2 M ./rdft/problem2.c -2 +4 Sat Aug 3 15:09:56 EDT 2002 athena * [project @ 2002-08-03 19:09:56 by athena] More portability work M ./configure.ac -2 +7 M ./kernel/cycle.h -2 +2 M ./kernel/ifftw.h -9 +2 Sat Aug 3 14:33:40 EDT 2002 athena * [project @ 2002-08-03 18:33:40 by athena] Improved portability, removed gnu make dependencies M ./acinclude.m4 -1 +1 M ./configure.ac -4 +3 M ./kernel/cycle.h -2 +2 M ./kernel/ifftw.h -2 +2 M ./support/Makefile.codelets -14 +6 Sat Aug 3 13:48:53 EDT 2002 athena * [project @ 2002-08-03 17:48:53 by athena] Remember to thank XXX M ./TODO +8 Fri Aug 2 17:38:18 EDT 2002 athena * [project @ 2002-08-02 21:38:18 by athena] Multiplication on altivec requires FMA with -0.0 to be IEEE754 compliant. M ./simd/simd-altivec.h -1 +1 Fri Aug 2 15:26:37 EDT 2002 athena * [project @ 2002-08-02 19:26:37 by athena] Allow for extended precision in codelets M ./genfft/c.ml -3 +4 M ./kernel/ifftw.h -7 +9 Fri Aug 2 08:52:04 EDT 2002 athena * [project @ 2002-08-02 12:52:04 by athena] Shortened names M ./dft/codelets/inplace/Makefile.am -5 +5 Fri Aug 2 03:49:09 EDT 2002 stevenj * [project @ 2002-08-02 07:49:09 by stevenj] added infrastructure for future r2r transforms M ./TODO -3 +1 M ./rdft/codelet.h -3 +33 M ./rdft/problem.c -1 +8 Thu Aug 1 21:29:14 EDT 2002 athena * [project @ 2002-08-02 01:29:05 by athena] Version info M ./Makefile.am +1 M ./configure.ac +2 Thu Aug 1 21:06:22 EDT 2002 athena * [project @ 2002-08-02 01:06:22 by athena] Listened to one customer and added radix-12. Added radix-15 for consistency (whatever that is) M ./dft/codelets/inplace/Makefile.am -1 +1 M ./dft/codelets/standard/Makefile.am -1 +1 M ./dft/k7/codelets/Makefile.am -3 +4 M ./dft/simd/codelets/Makefile.am -2 +2 M ./kernel/align.c -2 +8 M ./rdft/codelets/hc2r/Makefile.am -3 +3 M ./rdft/codelets/r2hc/Makefile.am -3 +3 Thu Aug 1 19:50:53 EDT 2002 stevenj * [project @ 2002-08-01 23:50:53 by stevenj] whoops again, fixed the wrong line M ./kernel/cycle.h -3 +3 Thu Aug 1 19:50:16 EDT 2002 stevenj * [project @ 2002-08-01 23:50:16 by stevenj] whoops M ./kernel/cycle.h -2 +2 Thu Aug 1 16:01:15 EDT 2002 stevenj * [project @ 2002-08-01 20:01:15 by stevenj] use new AC_INIT and add VERSION to wisdom M ./configure.ac -3 +4 M ./kernel/planner.c -4 +6 Thu Aug 1 14:56:45 EDT 2002 stevenj * [project @ 2002-08-01 18:56:45 by stevenj] mygetR -> getR M ./kernel/scan.c -3 +3 Thu Aug 1 14:56:02 EDT 2002 stevenj * [project @ 2002-08-01 18:56:02 by stevenj] scanner cleanups: just return 0/1, simplify integer reads M ./dft/problem.c -6 +4 M ./kernel/planner.c -14 +12 M ./kernel/scan.c -78 +42 M ./kernel/tensor.c -9 +8 M ./rdft/problem.c -7 +5 M ./rdft/problem2.c -7 +5 M ./tests/bench.c +7 Thu Aug 1 08:04:01 EDT 2002 athena * [project @ 2002-08-01 12:04:01 by athena] Reverted back to casting pointer to ulong M ./kernel/align.c -2 +2 Thu Aug 1 08:03:46 EDT 2002 athena * [project @ 2002-08-01 12:03:46 by athena] Cast to unsigned long, not long M ./kernel/ifftw.h -6 +6 Thu Aug 1 03:14:50 EDT 2002 stevenj * [project @ 2002-08-01 07:14:50 by stevenj] additional comment M ./kernel/scan.c -2 +4 Thu Aug 1 03:12:37 EDT 2002 stevenj * [project @ 2002-08-01 07:12:37 by stevenj] added comment M ./kernel/scan.c -1 +6 Thu Aug 1 03:03:18 EDT 2002 stevenj * [project @ 2002-08-01 07:03:18 by stevenj] added wisdom import A ./kernel/printers.c A ./kernel/scan.c A ./kernel/scanners.c M ./dft/conf.c -1 +3 M ./dft/dft.h -1 +3 M ./dft/problem.c -2 +31 M ./dft/verify.c -2 +1 M ./kernel/Makefile.am -3 +4 M ./kernel/alloc.c -2 +1 M ./kernel/assert.c -2 +1 M ./kernel/debug.c -9 +2 M ./kernel/ifftw.h -6 +53 M ./kernel/planner.c -22 +106 M ./kernel/print.c -2 +1 M ./kernel/printers.c +80 M ./kernel/problem.c -1 +2 M ./kernel/scan.c +322 M ./kernel/scanners.c +61 M ./kernel/tensor.c -4 +30 M ./kernel/timer.c -2 +1 M ./rdft/conf.c -1 +4 M ./rdft/problem.c -2 +32 M ./rdft/problem2.c -2 +32 M ./rdft/rdft.h -1 +4 M ./rdft/verify.c -2 +1 M ./tests/bench.c -9 +31 Wed Jul 31 23:12:05 EDT 2002 stevenj * [project @ 2002-08-01 03:12:05 by stevenj] whoops M ./kernel/align.c -1 +2 Wed Jul 31 22:06:46 EDT 2002 stevenj * [project @ 2002-08-01 02:06:46 by stevenj] use %u for alignment_of M ./dft/problem.c -3 +3 M ./rdft/problem.c -2 +2 M ./rdft/problem2.c -2 +2 Wed Jul 31 21:47:15 EDT 2002 stevenj * [project @ 2002-08-01 01:47:15 by stevenj] ptrdiff_t form M ./kernel/align.c -2 +2 Wed Jul 31 21:33:35 EDT 2002 athena * [project @ 2002-08-01 01:33:35 by athena] Cast to avoid warning from C++ compiler M ./kernel/ifftw.h -2 +2 Wed Jul 31 18:57:04 EDT 2002 athena * [project @ 2002-07-31 22:57:04 by athena] Make problem equality depend on alignments. A ./kernel/align.c M ./dft/problem.c -4 +10 M ./kernel/Makefile.am -5 +6 M ./kernel/align.c +35 M ./kernel/ifftw.h -1 +2 M ./rdft/problem.c -4 +9 M ./rdft/problem2.c -8 +13 M ./simd/simd.h +2 Wed Jul 31 15:45:31 EDT 2002 athena * [project @ 2002-07-31 19:45:31 by athena] Shorter names M ./dft/simd/codelets/Makefile.am -22 +18 Wed Jul 31 14:38:00 EDT 2002 athena * [project @ 2002-07-31 18:38:00 by athena] Oops M ./simd/simd-sse.h -1 +1 Wed Jul 31 14:37:19 EDT 2002 athena * [project @ 2002-07-31 18:37:19 by athena] Fix warning M ./simd/simd-sse.h +2 Wed Jul 31 07:52:53 EDT 2002 athena * [project @ 2002-07-31 11:52:53 by athena] Removed silly abstraction barrier. Also, cons() terminology was no longer appropriate. M ./kernel/alloc.c -2 +2 M ./kernel/ifftw.h -16 +19 M ./kernel/planner-naive.c -2 +2 M ./kernel/planner-score.c -2 +2 M ./kernel/planner.c -26 +13 Tue Jul 30 22:35:24 EDT 2002 stevenj * [project @ 2002-07-31 02:35:24 by stevenj] removed register_registrar and solvtab_exec_reverse hacks M ./kernel/ifftw.h -4 +3 M ./kernel/planner.c -24 +15 M ./kernel/solvtab.c -19 +6 Tue Jul 30 19:54:41 EDT 2002 stevenj * [project @ 2002-07-30 23:54:41 by stevenj] register_registrar doesn't search whole solver list (maybe we should change register_solver instead) M ./kernel/planner.c -1 +3 Tue Jul 30 19:36:37 EDT 2002 stevenj * [project @ 2002-07-30 23:36:37 by stevenj] credit M ./kernel/cycle.h -2 +2 Tue Jul 30 19:34:16 EDT 2002 stevenj * [project @ 2002-07-30 23:34:16 by stevenj] added HP/UX ia64 support, courtesy of Teresa L. Johnson M ./kernel/cycle.h -1 +23 Tue Jul 30 13:28:33 EDT 2002 athena * [project @ 2002-07-30 17:28:33 by athena] Fixed alignment checks M ./dft/simd/n1b.c -1 +1 M ./dft/simd/n1f.c -1 +1 M ./dft/simd/t1b.c -1 +1 M ./dft/simd/t1f.c -1 +1 M ./kernel/alloc.c -1 +2 Tue Jul 30 01:20:11 EDT 2002 stevenj * [project @ 2002-07-30 05:20:11 by stevenj] ugh, wisdom id fixes in exprt_conf M ./kernel/ifftw.h -2 +4 M ./kernel/planner.c -2 +4 M ./kernel/solvtab.c -5 +18 Tue Jul 30 00:41:15 EDT 2002 stevenj * [project @ 2002-07-30 04:41:15 by stevenj] exprt_registrars -> exprt_conf, added missing SOLVTAB_END M ./kernel/ifftw.h -2 +2 M ./kernel/planner.c -5 +6 M ./tests/bench.c -1 +1 Tue Jul 30 00:36:26 EDT 2002 stevenj * [project @ 2002-07-30 04:36:26 by stevenj] exprt_registrars should output self-contained configuration M ./kernel/planner.c -9 +20 Mon Jul 29 23:52:07 EDT 2002 stevenj * [project @ 2002-07-30 03:52:07 by stevenj] added exprt_registrars M ./dft/conf.c -12 +12 M ./kernel/ifftw.h -2 +7 M ./kernel/planner.c -4 +91 M ./kernel/solvtab.c -3 +5 M ./rdft/conf.c -18 +18 M ./support/Makefile.codelets -3 +3 M ./tests/bench.c +2 Mon Jul 29 23:42:27 EDT 2002 stevenj * [project @ 2002-07-30 03:42:27 by stevenj] whoops M ./kernel/print.c -2 +3 Mon Jul 29 21:24:51 EDT 2002 athena * [project @ 2002-07-30 01:24:51 by athena] More stringent requirements on strides for SIMD codelets M ./dft/simd/n1b.c -4 +4 M ./dft/simd/n1b.h +3 M ./dft/simd/n1f.c -4 +4 M ./dft/simd/n1f.h +3 M ./dft/simd/t1b.c -2 +2 M ./dft/simd/t1b.h -1 +6 M ./dft/simd/t1f.c -2 +2 M ./dft/simd/t1f.h +5 M ./simd/simd-altivec.h +2 M ./simd/simd-sse.h +2 M ./simd/simd-sse2.h +2 M ./simd/simd.h -2 +3 Mon Jul 29 21:05:49 EDT 2002 stevenj * [project @ 2002-07-30 01:05:49 by stevenj] remove warning M ./rdft/buffered2.c -2 +3 Mon Jul 29 20:51:19 EDT 2002 stevenj * [project @ 2002-07-30 00:51:01 by stevenj] use %td for ptrdiff_t and %T for tensors M ./dft/problem.c -2 +2 M ./kernel/print.c -3 +13 M ./kernel/traverse.c -2 +7 M ./rdft/problem.c -2 +2 M ./rdft/problem2.c -2 +2 Mon Jul 29 16:17:11 EDT 2002 athena * [project @ 2002-07-29 20:17:11 by athena] Fix for SIMD M ./dft/buffered.c -1 +5 Mon Jul 29 15:40:53 EDT 2002 athena * [project @ 2002-07-29 19:40:53 by athena] Missing lfftw_mkstride and lfftw_stride_destroy M ./kernel/ifftw.h -1 +3 Mon Jul 29 14:34:46 EDT 2002 athena * [project @ 2002-07-29 18:34:46 by athena] Implement LDA/STA M ./simd/simd-altivec.h +10 Mon Jul 29 14:19:21 EDT 2002 athena * [project @ 2002-07-29 18:19:21 by athena] More SIMD work M ./dft/simd/n1b.c -4 +4 M ./dft/simd/n1f.c -4 +4 M ./dft/simd/t1b.c -2 +2 M ./dft/simd/t1f.c -2 +2 M ./simd/simd-altivec.h -3 +2 M ./simd/simd-sse.h -3 +2 M ./simd/simd-sse2.h -2 +1 M ./simd/simd.h +3 Mon Jul 29 13:16:12 EDT 2002 athena * [project @ 2002-07-29 17:16:12 by athena] Cleanup M ./simd/simd-altivec.h -9 +9 M ./simd/simd-sse.h -16 +28 M ./simd/simd-sse2.h -13 +25 Mon Jul 29 13:02:38 EDT 2002 stevenj * [project @ 2002-07-29 17:02:38 by stevenj] update M ./ChangeLog +461 Mon Jul 29 12:45:33 EDT 2002 athena * [project @ 2002-07-29 16:45:33 by athena] Also check strides in SIMD codelets M ./dft/simd/n1b.c +4 M ./dft/simd/n1f.c +4 M ./dft/simd/t1b.c +2 M ./dft/simd/t1f.c +2 M ./simd/simd-altivec.h -1 +2 M ./simd/simd-sse.h +2 M ./simd/simd-sse2.h +2 Mon Jul 29 11:26:08 EDT 2002 athena * [project @ 2002-07-29 15:26:08 by athena] Minor changes, mostly for consistency with the big-endian processor M ./simd/simd-altivec.h -18 +18 Mon Jul 29 00:50:06 EDT 2002 stevenj * [project @ 2002-07-29 04:50:06 by stevenj] added comment M ./rdft/rader-dht.c +1 Sun Jul 28 21:19:35 EDT 2002 stevenj * [project @ 2002-07-29 01:19:35 by stevenj] added code for icc's _mm_malloc (memalign replacement) M ./configure.ac -1 +1 M ./kernel/alloc.c -4 +11 Sun Jul 28 17:33:07 EDT 2002 stevenj * [project @ 2002-07-28 21:33:07 by stevenj] slight fixes M ./rdft/problem2.c -6 +11 M ./rdft/verify.c -4 +7 M ./tests/bench.c -1 +2 Sun Jul 28 16:28:43 EDT 2002 stevenj * [project @ 2002-07-28 20:28:43 by stevenj] whoops M ./rdft/problem2.c -3 +3 Sun Jul 28 16:13:19 EDT 2002 athena * [project @ 2002-07-28 20:13:19 by athena] Use vec_xor to change sign M ./simd/altivec.c -2 +4 M ./simd/simd-altivec.h -3 +10 Sun Jul 28 16:10:59 EDT 2002 stevenj * [project @ 2002-07-28 20:10:59 by stevenj] added rdft2 A ./rdft/buffered2.c A ./rdft/direct2.c A ./rdft/nop2.c A ./rdft/plan2.c A ./rdft/problem2.c A ./rdft/solve2.c A ./rdft/vrank-geq1-rdft2.c M ./rdft/Makefile.am -2 +4 M ./rdft/buffered2.c +451 M ./rdft/conf.c -1 +7 M ./rdft/direct2.c +221 M ./rdft/khc2r.c -3 +3 M ./rdft/kr2hc.c -3 +3 M ./rdft/nop2.c +88 M ./rdft/plan2.c +33 M ./rdft/problem2.c +175 M ./rdft/rdft.h -1 +53 M ./rdft/solve2.c +31 M ./rdft/verify.c -17 +195 M ./rdft/vrank-geq1-rdft2.c +259 M ./tests/bench.c -6 +35 Sun Jul 28 15:45:54 EDT 2002 athena * [project @ 2002-07-28 19:45:54 by athena] Optimized M ./simd/simd-altivec.h -17 +10 Sun Jul 28 15:11:14 EDT 2002 athena * [project @ 2002-07-28 19:11:14 by athena] Changed ALIGNMENT M ./simd/simd-altivec.h -1 +1 Sun Jul 28 15:09:40 EDT 2002 athena * [project @ 2002-07-28 19:09:40 by athena] alignment := 8 M ./simd/simd-sse.h -1 +1 Sun Jul 28 14:57:22 EDT 2002 athena * [project @ 2002-07-28 18:57:22 by athena] Avoid warning M ./simd/simd-altivec.h -1 +1 Sun Jul 28 14:53:03 EDT 2002 athena * [project @ 2002-07-28 18:53:03 by athena] Oops M ./simd/simd-sse2.h -2 +2 Sun Jul 28 14:50:09 EDT 2002 athena * [project @ 2002-07-28 18:50:09 by athena] New altivec experiment M ./genfft/annotate.ml -5 +5 M ./genfft/genutil.ml -7 +7 M ./genfft/simd.ml -3 +6 M ./genfft/variable.ml -38 +16 M ./genfft/variable.mli -8 +4 M ./simd/simd-altivec.h -15 +15 M ./simd/simd-sse.h -3 +5 M ./simd/simd-sse2.h -2 +2 Sun Jul 28 13:48:20 EDT 2002 athena * [project @ 2002-07-28 17:48:20 by athena] Nothing M ./simd/simd-altivec.h -6 +6 Sun Jul 28 13:47:50 EDT 2002 athena * [project @ 2002-07-28 17:47:50 by athena] Oops M ./simd/simd-altivec.h -26 +19 Sun Jul 28 13:44:28 EDT 2002 athena * [project @ 2002-07-28 17:44:28 by athena] Nothing M ./simd/simd-altivec.h -19 +26 Sun Jul 28 10:38:10 EDT 2002 athena * [project @ 2002-07-28 14:38:10 by athena] Constants are now in separate file. A ./simd/altivec.c M ./simd/Makefile.am -1 +1 M ./simd/altivec.c +38 M ./simd/simd-altivec.h -20 +13 Sun Jul 28 07:58:37 EDT 2002 athena * [project @ 2002-07-28 11:58:37 by athena] More precise comment M ./simd/simd-altivec.h -1 +1 Sun Jul 28 07:56:40 EDT 2002 athena * [project @ 2002-07-28 11:56:40 by athena] gcc-3.1 bug workaround M ./simd/simd-altivec.h -42 +24 Sun Jul 28 01:39:54 EDT 2002 stevenj * [project @ 2002-07-28 05:39:54 by stevenj] slight optimization, and exported zerotens functions M ./dft/buffered.c -4 +4 M ./dft/dft.h -1 +2 M ./dft/problem.c -4 +4 M ./rdft/buffered.c -4 +4 M ./rdft/problem.c -4 +4 M ./rdft/rdft.h -1 +2 Sun Jul 28 00:54:59 EDT 2002 stevenj * [project @ 2002-07-28 04:54:59 by stevenj] should be a plan_dft, not a plan_rdft M ./rdft/dft-r2hc.c -2 +2 Sat Jul 27 21:36:46 EDT 2002 athena * [project @ 2002-07-28 01:36:46 by athena] Optimizations. Make it work with vanilla non-Apple gcc. M ./simd/simd-altivec.h -21 +82 Sat Jul 27 19:20:09 EDT 2002 stevenj * [project @ 2002-07-27 23:20:09 by stevenj] whoops M ./rdft/generic.c -5 Sat Jul 27 18:54:01 EDT 2002 stevenj * [project @ 2002-07-27 22:54:01 by stevenj] added hc2r (dif) M ./rdft/generic.c -13 +122 Sat Jul 27 18:31:43 EDT 2002 stevenj * [project @ 2002-07-27 22:31:43 by stevenj] add hc2r (dif) case M ./rdft/rader-hc2hc.c -2 +158 Sat Jul 27 15:09:40 EDT 2002 athena * [project @ 2002-07-27 19:09:40 by athena] Altivec port M ./simd/simd-altivec.h -38 +131 M ./support/Makefile.codelets -3 +3 Sat Jul 27 15:06:21 EDT 2002 athena * [project @ 2002-07-27 19:06:21 by athena] Fixed signed/unsigned bug. M ./kernel/twiddle.c -2 +3 Thu Jul 25 20:11:26 EDT 2002 athena * [project @ 2002-07-26 00:11:26 by athena] Make rank0 unapplicable to in-place problems. M ./dft/rank0.c -1 +2 M ./rdft/rank0.c -1 +2 Thu Jul 25 17:10:52 EDT 2002 stevenj * [project @ 2002-07-25 21:10:52 by stevenj] only works for r odd M ./rdft/generic.c -1 +1 Thu Jul 25 15:30:06 EDT 2002 athena * [project @ 2002-07-25 19:30:06 by athena] Reinserted much better timing-avoidance heuristic M ./kernel/planner-score.c -2 +7 Thu Jul 25 15:21:13 EDT 2002 athena * [project @ 2002-07-25 19:21:13 by athena] Score is now a property of the plan, not of the solver. Revised representation of closures. M ./dft/buffered.c -5 +1 M ./kernel/ifftw.h -2 +7 M ./kernel/plan.c -4 +6 M ./kernel/planner-score.c -17 +30 M ./kernel/traverse.c -17 +16 M ./rdft/buffered.c -5 +1 M ./tests/bench.c -7 +14 Thu Jul 25 06:36:51 EDT 2002 athena * [project @ 2002-07-25 10:36:51 by athena] Cosmetic changes. Added hc2r_128.c M ./genfft/gen_hc2r.ml -3 +3 M ./genfft/gen_r2hc.ml -3 +3 M ./rdft/codelets/hc2r/Makefile.am -1 +1 Thu Jul 25 01:37:53 EDT 2002 stevenj * [project @ 2002-07-25 05:37:53 by stevenj] added hc2r M ./rdft/rader-dht.c -16 +55 Thu Jul 25 00:51:45 EDT 2002 stevenj * [project @ 2002-07-25 04:51:45 by stevenj] added hc2hc-difbuf A ./rdft/hc2hc-buf.c M ./rdft/Makefile.am -3 +3 M ./rdft/hc2hc-buf.c +248 R ./rdft/hc2hc-ditbuf.c M ./rdft/khc2hc-dif.c -3 +1 Thu Jul 25 00:25:06 EDT 2002 stevenj * [project @ 2002-07-25 04:25:06 by stevenj] added rdft-dif A ./rdft/hc2hc-dif.c M ./rdft/Makefile.am -3 +3 M ./rdft/hc2hc-dif.c +128 M ./rdft/hc2hc.c -3 +5 M ./rdft/khc2hc-dif.c -4 +3 M ./rdft/rdft.h -1 +3 Thu Jul 25 00:22:36 EDT 2002 stevenj * [project @ 2002-07-25 04:22:36 by stevenj] whoops, hc2r must be conjugated to have right sign M ./rdft/verify.c -3 +3 Wed Jul 24 23:27:45 EDT 2002 stevenj * [project @ 2002-07-25 03:27:45 by stevenj] slight change M ./dft/ct-dif.c -4 +6 Wed Jul 24 23:24:24 EDT 2002 stevenj * [project @ 2002-07-25 03:24:24 by stevenj] whoops M ./rdft/verify.c -3 +2 Wed Jul 24 22:46:39 EDT 2002 stevenj * [project @ 2002-07-25 02:46:39 by stevenj] support hc2r codelets A ./rdft/direct.c M ./rdft/Makefile.am -5 +5 M ./rdft/codelet.h -1 +3 R ./rdft/direct-r2hc.c M ./rdft/direct.c +227 M ./rdft/khc2r.c -5 +2 M ./rdft/rdft.h -2 +2 Wed Jul 24 22:01:53 EDT 2002 stevenj * [project @ 2002-07-25 02:01:53 by stevenj] use vector plan for r/i instead of two separate plans M ./rdft/dft-r2hc.c -36 +20 Wed Jul 24 20:36:34 EDT 2002 stevenj * [project @ 2002-07-25 00:36:34 by stevenj] hack to allow rader/generic to work in-place for small prime sizes, instead of always using buffered M ./dft/buffered.c -1 +5 M ./dft/rader.c -2 +2 M ./kernel/ifftw.h -1 +3 M ./rdft/buffered.c -1 +5 M ./rdft/rader-dht.c -1 +1 M ./rdft/rader-hc2hc.c -1 +1 Wed Jul 24 18:04:41 EDT 2002 stevenj * [project @ 2002-07-24 22:04:41 by stevenj] added rdft-generic A ./rdft/generic.c M ./rdft/Makefile.am -4 +4 M ./rdft/conf.c -2 +2 M ./rdft/generic.c +281 Wed Jul 24 17:27:34 EDT 2002 stevenj * [project @ 2002-07-24 21:27:34 by stevenj] fixed add count M ./dft/generic.c -1 +1 Wed Jul 24 14:52:26 EDT 2002 stevenj * [project @ 2002-07-24 18:52:26 by stevenj] again M ./rdft/rader-hc2hc.c -2 +2 Wed Jul 24 14:51:58 EDT 2002 stevenj * [project @ 2002-07-24 18:51:58 by stevenj] slight fix M ./rdft/rader-hc2hc.c -4 +5 Wed Jul 24 14:51:07 EDT 2002 stevenj * [project @ 2002-07-24 18:51:07 by stevenj] fixed comment M ./rdft/rader-hc2hc.c -3 +3 Wed Jul 24 14:41:24 EDT 2002 stevenj * [project @ 2002-07-24 18:41:24 by stevenj] whoops M ./tests/bench.c -2 Wed Jul 24 14:38:15 EDT 2002 stevenj * [project @ 2002-07-24 18:38:15 by stevenj] added rader-hc2hc A ./rdft/rader-hc2hc.c M ./rdft/Makefile.am -2 +2 M ./rdft/conf.c -2 +2 M ./rdft/rader-hc2hc.c +482 M ./rdft/rdft.h -2 +2 M ./tests/bench.c +2 Wed Jul 24 00:07:59 EDT 2002 stevenj * [project @ 2002-07-24 04:07:59 by stevenj] whoops, initialize W M ./dft/rader.c +1 Tue Jul 23 23:03:09 EDT 2002 stevenj * [project @ 2002-07-24 03:03:09 by stevenj] strides should not be unsigned M ./rdft/rader-dht.c -2 +4 Tue Jul 23 23:02:08 EDT 2002 stevenj * [project @ 2002-07-24 03:02:08 by stevenj] more stride sign fixes M ./dft/rader.c -3 +5 Tue Jul 23 23:01:04 EDT 2002 stevenj * [project @ 2002-07-24 03:01:04 by stevenj] strides should not be unsigned! M ./dft/rader.c -1 +2 Tue Jul 23 14:55:25 EDT 2002 stevenj * [project @ 2002-07-23 18:55:25 by stevenj] added comment M ./rdft/dft-r2hc.c -1 +4 Tue Jul 23 14:52:04 EDT 2002 stevenj * [project @ 2002-07-23 18:52:04 by stevenj] another fix to op count M ./rdft/r2hc-hc2r.c -1 +4 Tue Jul 23 14:51:01 EDT 2002 stevenj * [project @ 2002-07-23 18:51:01 by stevenj] whoops M ./rdft/r2hc-hc2r.c -2 +2 Tue Jul 23 14:49:43 EDT 2002 stevenj * [project @ 2002-07-23 18:49:43 by stevenj] slight fix to op counts M ./rdft/dft-r2hc.c -3 +3 M ./rdft/r2hc-hc2r.c -3 +3 Tue Jul 23 14:09:19 EDT 2002 stevenj * [project @ 2002-07-23 18:09:18 by stevenj] added dft-r2hc A ./rdft/dft-r2hc.c M ./rdft/Makefile.am -5 +6 M ./rdft/conf.c -2 +3 M ./rdft/dft-r2hc.c +187 M ./rdft/rdft.h -1 +2 Tue Jul 23 02:50:12 EDT 2002 stevenj * [project @ 2002-07-23 06:50:12 by stevenj] better comment and var. name M ./rdft/rader-dht.c -7 +7 Tue Jul 23 02:39:11 EDT 2002 stevenj * [project @ 2002-07-23 06:39:11 by stevenj] fixed tests for hc2r, and added r2hc-hc2r A ./rdft/r2hc-hc2r.c M ./rdft/Makefile.am -3 +3 M ./rdft/conf.c -1 +2 M ./rdft/r2hc-hc2r.c +182 M ./rdft/rdft.h -1 +2 M ./rdft/verify.c -1 +46 M ./tests/bench.c -1 +6 Tue Jul 23 00:45:23 EDT 2002 stevenj * [project @ 2002-07-23 04:45:23 by stevenj] added rader-dht A ./rdft/rader-dht.c M ./rdft/Makefile.am -2 +3 M ./rdft/conf.c -1 +2 M ./rdft/rader-dht.c +433 M ./rdft/rdft.h -1 +2 Mon Jul 22 21:05:12 EDT 2002 athena * [project @ 2002-07-23 01:05:12 by athena] Added r2hc_128, what the hell. M ./rdft/codelets/r2hc/Makefile.am -1 +1 Mon Jul 22 20:48:59 EDT 2002 athena * [project @ 2002-07-23 00:48:59 by athena] Added codelets that compute twiddle factors M ./rdft/codelets/r2hc/Makefile.am -1 +7 Mon Jul 22 19:57:16 EDT 2002 stevenj * [project @ 2002-07-22 23:57:16 by stevenj] added rdft-buffered A ./rdft/buffered.c M ./rdft/Makefile.am -2 +2 M ./rdft/buffered.c +330 M ./rdft/conf.c -2 +2 Mon Jul 22 19:43:39 EDT 2002 stevenj * [project @ 2002-07-22 23:43:39 by stevenj] added hc2hc-ditbuf A ./rdft/hc2hc-ditbuf.c M ./rdft/Makefile.am -3 +3 M ./rdft/hc2hc-ditbuf.c +189 M ./rdft/khc2hc-dit.c -3 +1 Mon Jul 22 14:29:04 EDT 2002 stevenj * [project @ 2002-07-22 18:29:04 by stevenj] use STACK_MALLOC (alloca), since generic radix is always small M ./dft/generic.c -3 +3 Mon Jul 22 14:22:43 EDT 2002 stevenj * [project @ 2002-07-22 18:22:43 by stevenj] small cleanup M ./rdft/hc2hc-dit.c -2 +2 Mon Jul 22 07:42:13 EDT 2002 athena * [project @ 2002-07-22 11:42:13 by athena] What the hell was I thinking? M ./rdft/problem.c -18 +8 Mon Jul 22 07:37:12 EDT 2002 athena * [project @ 2002-07-22 11:37:12 by athena] Reduced code size by using table instead of switch statement. M ./rdft/problem.c -12 +16 Mon Jul 22 07:27:06 EDT 2002 athena * [project @ 2002-07-22 11:27:06 by athena] Changed hash function to avoid collisions with DFT. M ./rdft/problem.c -4 +4 Mon Jul 22 01:37:06 EDT 2002 stevenj * [project @ 2002-07-22 05:37:06 by stevenj] added missing file, whoops A ./rdft/hc2hc-dit.c Mon Jul 22 01:24:17 EDT 2002 stevenj * [project @ 2002-07-22 05:24:17 by stevenj] whoops, generate enough twiddles for odd m M ./rdft/hc2hc.c -2 +2 Mon Jul 22 01:10:21 EDT 2002 stevenj * [project @ 2002-07-22 05:10:21 by stevenj] don't try to verify R2HCII or HC2RIII plans M ./rdft/verify.c -5 +7 Mon Jul 22 01:05:00 EDT 2002 stevenj * [project @ 2002-07-22 05:05:00 by stevenj] recursive case now works, I think M ./rdft/hc2hc.c -9 +4 Mon Jul 22 01:04:40 EDT 2002 stevenj * [project @ 2002-07-22 05:04:40 by stevenj] add extra impulse test for debugging M ./rdft/verify.c -1 +4 Mon Jul 22 01:02:38 EDT 2002 stevenj * [project @ 2002-07-22 05:02:38 by stevenj] whoops, multiply ios offset by stride (and rename to ioffset) M ./rdft/direct-r2hc.c -4 +5 Mon Jul 22 00:22:02 EDT 2002 stevenj * [project @ 2002-07-22 04:22:02 by stevenj] whoops M ./rdft/verify.c -2 +2 Sun Jul 21 23:58:14 EDT 2002 stevenj * [project @ 2002-07-22 03:58:14 by stevenj] whoops M ./rdft/verify.c -2 +2 Sun Jul 21 23:43:03 EDT 2002 stevenj * [project @ 2002-07-22 03:43:03 by stevenj] added hc2hc-dit M ./genfft/gen_hc2hc.ml -3 +3 M ./rdft/Makefile.am -2 +2 M ./rdft/hc2hc.c -10 +21 M ./rdft/khc2hc-dit.c -4 +3 Sun Jul 21 23:15:12 EDT 2002 stevenj * [project @ 2002-07-22 03:15:12 by stevenj] twiddles can be shared with smaller m's M ./kernel/twiddle.c -4 +4 Sun Jul 21 22:34:28 EDT 2002 stevenj * [project @ 2002-07-22 02:34:28 by stevenj] preparing for recursive rdft... A ./rdft/hc2hc.c A ./rdft/hc2hc.h M ./rdft/Makefile.am -3 +3 M ./rdft/codelet.h -1 +2 M ./rdft/codelets/hfb.c -2 +2 M ./rdft/hc2hc.c +249 M ./rdft/hc2hc.h +69 Sun Jul 21 19:31:22 EDT 2002 stevenj * [project @ 2002-07-21 23:31:22 by stevenj] slight fix, to match libbench/verify.c M ./rdft/verify.c -3 +3 Sun Jul 21 18:43:12 EDT 2002 stevenj * [project @ 2002-07-21 22:43:12 by stevenj] r2hcII has imag parts offset by n-1, not n. We can also allocate fewer strides. M ./rdft/direct-r2hc.c -4 +7 Sun Jul 21 18:27:09 EDT 2002 stevenj * [project @ 2002-07-21 22:27:09 by stevenj] delete unused var M ./rdft/rank0.c -2 +2 Sun Jul 21 02:06:53 EDT 2002 stevenj * [project @ 2002-07-21 06:06:53 by stevenj] added some rdft solvers A ./rdft/direct-r2hc.c A ./rdft/indirect.c A ./rdft/nop.c A ./rdft/rank0.c A ./rdft/vrank-geq1.c A ./rdft/vrank2-transpose.c A ./rdft/vrank3-transpose.c M ./rdft/Makefile.am -2 +4 M ./rdft/codelet.h -3 +8 M ./rdft/codelets/hc2r.c -2 +2 M ./rdft/codelets/r2hc.c -2 +2 M ./rdft/conf.c -4 +4 M ./rdft/direct-r2hc.c +174 M ./rdft/indirect.c +235 R ./rdft/khc2rIII.c M ./rdft/kr2hc.c -5 +2 R ./rdft/kr2hcII.c M ./rdft/nop.c +97 M ./rdft/problem.c -3 +20 M ./rdft/rank0.c +217 M ./rdft/rdft.h -5 +4 M ./rdft/vrank-geq1.c +260 M ./rdft/vrank2-transpose.c +130 M ./rdft/vrank3-transpose.c +171 Sun Jul 21 01:52:54 EDT 2002 stevenj * [project @ 2002-07-21 05:52:54 by stevenj] pass identifier in FFTW() through another macro so that the mangled name can itself be a preprocessor symbol M ./kernel/fftw3.h -4 +5 Sun Jul 21 01:05:21 EDT 2002 stevenj * [project @ 2002-07-21 05:05:21 by stevenj] fix in comment M ./dft/vrank-geq1.c -2 +2 Sun Jul 21 00:47:03 EDT 2002 stevenj * [project @ 2002-07-21 04:47:03 by stevenj] bench tests rdft plans M ./Makefile.am -2 +5 M ./rdft/rdft.h -2 +3 M ./tests/bench.c -16 +36 Sun Jul 21 00:22:14 EDT 2002 stevenj * [project @ 2002-07-21 04:22:14 by stevenj] make rdft.h and dft.h compatible M ./rdft/codelet.h -4 +4 M ./tests/Makefile.am -1 +1 M ./tests/bench.c -1 +1 Sun Jul 21 00:12:19 EDT 2002 stevenj * [project @ 2002-07-21 04:12:19 by stevenj] first-draft rdft verify A ./rdft/verify.c M ./rdft/Makefile.am -1 +1 M ./rdft/problem.c -8 +8 M ./rdft/rdft.h -5 +5 M ./rdft/verify.c +554 Sat Jul 20 22:09:15 EDT 2002 stevenj * [project @ 2002-07-21 02:09:15 by stevenj] got rid of annoying warnings M ./rdft/khc2hc-dif.c -1 +2 M ./rdft/khc2hc-dit.c -1 +2 M ./rdft/khc2r.c -1 +2 M ./rdft/khc2rIII.c -1 +2 M ./rdft/kr2hc.c -1 +2 M ./rdft/kr2hcII.c -1 +2 Sat Jul 20 22:07:37 EDT 2002 stevenj * [project @ 2002-07-21 02:07:37 by stevenj] added stub codelet registration for linking purposes A ./rdft/khc2hc-dif.c A ./rdft/khc2hc-dit.c A ./rdft/khc2r.c A ./rdft/khc2rIII.c A ./rdft/kr2hc.c A ./rdft/kr2hcII.c M ./rdft/Makefile.am -1 +2 M ./rdft/khc2hc-dif.c +31 M ./rdft/khc2hc-dit.c +31 M ./rdft/khc2r.c +31 M ./rdft/khc2rIII.c +31 M ./rdft/kr2hc.c +31 M ./rdft/kr2hcII.c +31 M ./rdft/rdft.h -2 +2 Sat Jul 20 21:46:03 EDT 2002 stevenj * [project @ 2002-07-21 01:46:03 by stevenj] basic rdft stuff A ./rdft/conf.c A ./rdft/plan.c A ./rdft/problem.c A ./rdft/rdft.h A ./rdft/solve.c M ./rdft/Makefile.am -1 +1 M ./rdft/conf.c +47 M ./rdft/plan.c +33 M ./rdft/problem.c +145 M ./rdft/rdft.h +82 M ./rdft/solve.c +31 Sat Jul 20 21:06:50 EDT 2002 stevenj * [project @ 2002-07-21 01:06:50 by stevenj] rdft codelets now compile A ./rdft/Makefile.am A ./rdft/codelet.h A ./rdft/codelets/Makefile.am A ./rdft/codelets/hb.h A ./rdft/codelets/hc2r.c A ./rdft/codelets/hc2r.h A ./rdft/codelets/hc2rIII.h A ./rdft/codelets/hf.h A ./rdft/codelets/hfb.c A ./rdft/codelets/r2hc.c A ./rdft/codelets/r2hc.h A ./rdft/codelets/r2hcII.h M ./Makefile.am -1 +1 M ./configure.ac +2 M ./dft/codelet.h -33 +1 M ./genfft/gen_hc2hc.ml -3 +3 M ./kernel/ifftw.h -1 +35 M ./rdft/Makefile.am +6 M ./rdft/codelet.h +133 M ./rdft/codelets/Makefile.am +7 M ./rdft/codelets/hb.h +23 M ./rdft/codelets/hc2r/Makefile.am -3 +3 M ./rdft/codelets/hc2r.c +44 M ./rdft/codelets/hc2r.h +23 M ./rdft/codelets/hc2rIII.h +23 M ./rdft/codelets/hf.h +23 M ./rdft/codelets/hfb.c +41 M ./rdft/codelets/r2hc/Makefile.am -2 +2 M ./rdft/codelets/r2hc.c +44 M ./rdft/codelets/r2hc.h +23 M ./rdft/codelets/r2hcII.h +23 Sat Jul 20 18:40:31 EDT 2002 athena * [project @ 2002-07-20 22:40:31 by athena] Oops, was generating rdfts instead of hdfts M ./genfft/gen_hc2r.ml -3 +3 Sat Jul 20 18:25:47 EDT 2002 athena * [project @ 2002-07-20 22:25:47 by athena] Added hc2r codelets A ./rdft/codelets/hc2r/ A ./rdft/codelets/hc2r/Makefile.am M ./TODO +4 M ./configure.ac +1 M ./genfft/twiddle.ml -7 +14 M ./genfft-k7/twiddle.ml -12 +7 M ./kernel/twiddle.c -26 +39 M ./rdft/codelets/hc2r/Makefile.am -1 +42 Sat Jul 20 17:54:39 EDT 2002 athena * [project @ 2002-07-20 21:54:39 by athena] return W in hc2hc codelets M ./genfft/gen_hc2hc.ml -3 +4 Sat Jul 20 17:51:06 EDT 2002 athena * [project @ 2002-07-20 21:51:06 by athena] Some work on rdft codelets A ./rdft/ A ./rdft/codelets/ A ./rdft/codelets/r2hc/ A ./rdft/codelets/r2hc/Makefile.am M ./configure.ac +2 M ./dft/codelets/inplace/Makefile.am +4 M ./dft/codelets/standard/Makefile.am +3 M ./dft/simd/codelets/Makefile.am -1 +1 M ./genfft/gen_hc2hc.ml -62 +18 M ./genfft/gen_hc2r.ml -5 +24 M ./genfft/gen_r2hc.ml -5 +21 M ./genfft/trig.ml -22 +22 M ./rdft/codelets/r2hc/Makefile.am -1 +42 M ./support/Makefile.codelets -9 +6 Tue Jul 16 13:55:50 EDT 2002 athena * [project @ 2002-07-16 17:55:50 by athena] fix const M ./kernel/fftw3.h -3 +4 Tue Jul 16 07:00:10 EDT 2002 athena * [project @ 2002-07-16 11:00:10 by athena] Separate CFLAGS in codelets. Fix const in certain places. M ./acinclude.m4 -2 +6 M ./configure.ac +1 M ./dft/codelets/inplace/Makefile.am +1 M ./dft/codelets/standard/Makefile.am +1 M ./dft/indirect.c -2 +2 M ./dft/rank0.c -2 +2 M ./dft/simd/codelets/Makefile.am +1 M ./kernel/version.c -3 +4 M ./tests/bench.c -2 +9 Mon Jul 15 21:10:42 EDT 2002 stevenj * [project @ 2002-07-16 01:10:42 by stevenj] note buffering problem M ./TODO +2 Mon Jul 15 20:27:51 EDT 2002 athena * [project @ 2002-07-16 00:27:51 by athena] Removed unpredictable branch from inner loop M ./dft/generic.c -2 +2 Mon Jul 15 19:35:04 EDT 2002 stevenj * [project @ 2002-07-15 23:35:04 by stevenj] update M ./TODO -2 Mon Jul 15 19:31:39 EDT 2002 stevenj * [project @ 2002-07-15 23:31:39 by stevenj] optimization M ./dft/generic.c -1 +2 Mon Jul 15 19:28:30 EDT 2002 stevenj * [project @ 2002-07-15 23:28:30 by stevenj] added generic dit A ./dft/generic.c M ./dft/Makefile.am -4 +4 M ./dft/conf.c -1 +2 M ./dft/dft.h -1 +2 M ./dft/generic.c +219 M ./kernel/ifftw.h -2 +3 M ./kernel/twiddle.c -1 +13 Mon Jul 15 17:03:53 EDT 2002 stevenj * [project @ 2002-07-15 21:03:53 by stevenj] whoops, mksolver should be static M ./dft/rader.c -2 +2 Mon Jul 15 16:46:36 EDT 2002 athena * [project @ 2002-07-15 20:46:35 by athena] First implementation of gen_hc2hc, probably still buggy. A ./genfft/gen_hc2hc.ml M ./genfft/Makefile.am -9 +15 M ./genfft/algsimp.ml -4 +5 M ./genfft/c.ml -3 +10 M ./genfft/c.mli -1 +2 M ./genfft/gen_hc2hc.ml +220 M ./genfft/gen_hc2r.ml -24 +69 M ./genfft/gen_r2hc.ml -22 +66 M ./genfft/genutil.ml -13 +17 Mon Jul 15 16:40:23 EDT 2002 stevenj * [project @ 2002-07-15 20:40:23 by stevenj] don't count loading of twiddle factors in ops.other, since it isn't counted for the codelets M ./dft/rader.c -1 Mon Jul 15 15:13:19 EDT 2002 stevenj * [project @ 2002-07-15 19:13:19 by stevenj] plan_destroy puts plan to sleep before deallocating it, to eliminate duplicate free calls in solvers M ./dft/ct.c -2 +1 M ./dft/rader.c -2 M ./kernel/plan.c -2 +5 Mon Jul 15 15:07:41 EDT 2002 stevenj * [project @ 2002-07-15 19:07:41 by stevenj] fftw2-like vector recursion flag M ./dft/ct-dif.c -1 +6 M ./dft/ct-dit.c -2 +2 M ./dft/ct-ditbuf.c -2 +2 M ./dft/ct.c -1 +5 M ./dft/vrank-geq1.c -2 +10 M ./kernel/ifftw.h -2 +6 M ./tests/bench.c -2 +1 Sun Jul 14 21:01:44 EDT 2002 athena * [project @ 2002-07-15 01:01:44 by athena] More jokes M ./kernel/planner.c -1 +2 Sun Jul 14 20:36:01 EDT 2002 athena * [project @ 2002-07-15 00:36:01 by athena] Bless plan for testing purposes M ./tests/bench.c +1 Sun Jul 14 20:35:49 EDT 2002 athena * [project @ 2002-07-15 00:35:49 by athena] Canonical linked-list deletion (hope it is right) M ./kernel/planner.c -12 +13 Sun Jul 14 18:26:19 EDT 2002 stevenj * [project @ 2002-07-14 22:26:19 by stevenj] use estimating planner for cld_omega M ./dft/rader.c -1 +1 Sun Jul 14 18:10:56 EDT 2002 stevenj * [project @ 2002-07-14 22:10:56 by stevenj] better internal naming M ./dft/rader.c -3 +3 Sun Jul 14 18:10:01 EDT 2002 stevenj * [project @ 2002-07-14 22:10:01 by stevenj] printing should really be fixed now, grrr M ./dft/rader.c -10 +12 Sun Jul 14 17:57:12 EDT 2002 stevenj * [project @ 2002-07-14 21:57:12 by stevenj] print all distinct child plans M ./dft/rader.c -2 +7 Sun Jul 14 17:49:21 EDT 2002 stevenj * [project @ 2002-07-14 21:49:21 by stevenj] whoops M ./tests/bench.c -1 +1 Sun Jul 14 17:45:54 EDT 2002 stevenj * [project @ 2002-07-14 21:45:54 by stevenj] whoops, destroy should delete twiddle/omega from list M ./dft/rader.c -4 +2 Sun Jul 14 17:33:02 EDT 2002 stevenj * [project @ 2002-07-14 21:33:02 by stevenj] whoops M ./kernel/planner.c -3 +6 Sun Jul 14 17:12:14 EDT 2002 stevenj * [project @ 2002-07-14 21:12:14 by stevenj] added plan_bless and FORGET_ACCURSED M ./kernel/ifftw.h -2 +6 M ./kernel/plan.c -1 +14 M ./kernel/planner.c -13 +24 Sun Jul 14 16:15:43 EDT 2002 stevenj * [project @ 2002-07-14 20:15:43 by stevenj] further cleanup M ./kernel/traverse.c -6 +4 Sun Jul 14 16:14:15 EDT 2002 stevenj * [project @ 2002-07-14 20:14:15 by stevenj] slight cleanup M ./kernel/traverse.c -8 +7 Sun Jul 14 16:09:17 EDT 2002 stevenj * [project @ 2002-07-14 20:09:17 by stevenj] added traverse_plan via print (ugh) A ./kernel/traverse.c M ./kernel/Makefile.am -3 +3 M ./kernel/ifftw.h -1 +5 M ./kernel/traverse.c +115 M ./tests/bench.c +12 Sun Jul 14 15:08:29 EDT 2002 stevenj * [project @ 2002-07-14 19:08:29 by stevenj] added TW_FULL, and additional n parameter for twiddles M ./dft/ct.c -3 +4 M ./kernel/ifftw.h -5 +5 M ./kernel/twiddle.c -18 +34 Sun Jul 14 15:03:51 EDT 2002 stevenj * [project @ 2002-07-14 19:03:51 by stevenj] whoops M ./kernel/planner.c -2 +4 Sun Jul 14 13:49:20 EDT 2002 stevenj * [project @ 2002-07-14 17:49:20 by stevenj] save flags before invoking solver mkplan M ./kernel/ifftw.h -1 +2 M ./kernel/planner-naive.c -2 +2 M ./kernel/planner-score.c -2 +2 M ./kernel/planner.c -3 +10 Sun Jul 14 09:28:37 EDT 2002 athena * [project @ 2002-07-14 13:28:37 by athena] M ./TODO -6 +1 Sat Jul 13 22:17:29 EDT 2002 stevenj * [project @ 2002-07-14 02:17:29 by stevenj] added support for UNICOS _rtc() real-time-clock intrinsic function M ./configure.ac -1 +10 M ./kernel/cycle.h -2 +20 Sat Jul 13 22:06:35 EDT 2002 stevenj * [project @ 2002-07-14 02:06:35 by stevenj] fixed typo: HAVE_TIME_H should include time.h, not sys/time.h M ./kernel/timer.c -2 +2 Sat Jul 13 21:46:02 EDT 2002 stevenj * [project @ 2002-07-14 01:46:02 by stevenj] support AIX read_real_time timer M ./configure.ac +1 M ./kernel/cycle.h -1 +24 Sat Jul 13 17:02:51 EDT 2002 stevenj * [project @ 2002-07-13 21:02:51 by stevenj] use && instead of the (sigh) unportable -a M ./configure.ac -1 +1 Sat Jul 13 16:38:18 EDT 2002 stevenj * [project @ 2002-07-13 20:38:18 by stevenj] use AC_HELP_STRING M ./configure.ac -11 +11 Sat Jul 13 16:05:43 EDT 2002 stevenj * [project @ 2002-07-13 20:05:43 by stevenj] support long-double precision A ./kernel/lfftw3.h M ./configure.ac -2 +22 M ./dft/codelet.h -2 +6 M ./dft/verify.c -10 +29 M ./kernel/Makefile.am -3 +3 M ./kernel/dfftw3.h -1 +2 M ./kernel/fftw3.h -2 +5 M ./kernel/ifftw.h -9 +14 M ./kernel/lfftw3.h +25 M ./kernel/sfftw3.h -1 +2 M ./libbench/bench-main.c -3 +5 M ./libbench/bench-user.h -4 +8 M ./libbench/info.c -2 +4 M ./libbench/verify.c -10 +38 M ./simd/simd-sse2.h -1 +1 Sat Jul 13 15:48:10 EDT 2002 stevenj * [project @ 2002-07-13 19:48:10 by stevenj] whoops whoops M ./dft/rader.c -2 +2 Sat Jul 13 15:47:39 EDT 2002 stevenj * [project @ 2002-07-13 19:47:39 by stevenj] whoops M ./dft/rader.c -1 +1 Sat Jul 13 14:13:42 EDT 2002 stevenj * [project @ 2002-07-13 18:13:42 by stevenj] buffered solver strides have been fixed M ./TODO -7 Sat Jul 13 13:48:13 EDT 2002 stevenj * [project @ 2002-07-13 17:48:13 by stevenj] convention M ./dft/rader.c -1 +1 Sat Jul 13 12:50:06 EDT 2002 stevenj * [project @ 2002-07-13 16:50:06 by stevenj] share twiddle arrays in Rader M ./TODO -2 M ./dft/rader.c -5 +68 Sat Jul 13 12:48:10 EDT 2002 stevenj * [project @ 2002-07-13 16:48:10 by stevenj] call done() after verify M ./libbench/verify.c -1 +2 Fri Jul 12 15:42:04 EDT 2002 stevenj * [project @ 2002-07-12 19:42:04 by stevenj] output planner time with -v M ./tests/bench.c +7 Fri Jul 12 15:40:14 EDT 2002 stevenj * [project @ 2002-07-12 19:40:14 by stevenj] support double outputs M ./kernel/print.c -2 +9 Fri Jul 12 15:09:19 EDT 2002 stevenj * [project @ 2002-07-12 19:09:19 by stevenj] removed extraneous parens M ./dft/vrank-geq1.c -2 +2 Fri Jul 12 15:08:13 EDT 2002 stevenj * [project @ 2002-07-12 19:08:13 by stevenj] increase maxbufsz to 64k; makes a big difference for large 2d transforms M ./dft/buffered.c -2 +2 Fri Jul 12 05:59:26 EDT 2002 athena * [project @ 2002-07-12 09:59:26 by athena] Fix M ./dft/vrank-geq1.c -2 +2 Fri Jul 12 01:22:38 EDT 2002 stevenj * [project @ 2002-07-12 05:22:38 by stevenj] fix comment M ./dft/rank-geq2.c -2 +2 Fri Jul 12 00:59:29 EDT 2002 stevenj * [project @ 2002-07-12 04:59:29 by stevenj] fix in comment M ./kernel/tensor.c -2 +2 Fri Jul 12 00:13:13 EDT 2002 stevenj * [project @ 2002-07-12 04:13:13 by stevenj] updated M ./ChangeLog +873 Thu Jul 11 23:39:27 EDT 2002 stevenj * [project @ 2002-07-12 03:39:27 by stevenj] buffered malloc's buffers M ./TODO -3 Thu Jul 11 23:30:26 EDT 2002 stevenj * [project @ 2002-07-12 03:30:26 by stevenj] share more code between apply and apply_dit in Rader M ./TODO -2 M ./dft/rader.c -69 +30 Mon Jul 8 12:30:34 EDT 2002 athena * [project @ 2002-07-08 16:30:34 by athena] Polished M ./simd/simd-sse.h -18 +28 M ./simd/simd-sse2.h -25 +31 M ./simd/sse.c -4 +2 M ./simd/sse2.c -3 +2 Mon Jul 8 09:47:11 EDT 2002 athena * [project @ 2002-07-08 13:47:11 by athena] M ./support/Makefile.codelets -1 +1 Mon Jul 8 09:42:08 EDT 2002 athena * [project @ 2002-07-08 13:42:08 by athena] SIMD/FMA stuff M ./dft/simd/codelets/Makefile.am -1 +1 M ./genfft/c.ml -7 +2 M ./genfft/gen_notw_c.ml -2 +3 M ./genfft/simd.ml -2 +13 M ./genfft/to_alist.ml -1 +3 M ./genfft/to_alist.mli -1 +2 M ./simd/simd-sse.h +3 M ./simd/simd-sse2.h +3 M ./support/Makefile.codelets -1 +2 Mon Jul 8 07:43:51 EDT 2002 athena * [project @ 2002-07-08 11:43:51 by athena] Avoid code duplication M ./simd/simd-sse.h -12 +6 Sun Jul 7 20:56:15 EDT 2002 athena * [project @ 2002-07-08 00:56:15 by athena] Fixes for FMA+SIMD M ./genfft/Makefile.am -3 +3 M ./genfft/to_alist.ml -7 +14 Sun Jul 7 20:32:01 EDT 2002 athena * [project @ 2002-07-08 00:32:01 by athena] Major changes in SIMD fftw A ./dft/simd/n1b.c A ./dft/simd/n1b.h A ./dft/simd/n1f.c A ./dft/simd/n1f.h A ./dft/simd/t1b.c A ./dft/simd/t1b.h A ./dft/simd/t1f.c A ./dft/simd/t1f.h A ./genfft/gen_notw_c.ml A ./genfft/gen_twiddle_c.ml M ./dft/buffered.c -2 +2 M ./dft/codelets/standard/Makefile.am -6 +5 M ./dft/simd/Makefile.am -3 +1 R ./dft/simd/NAMING M ./dft/simd/codelets/Makefile.am -26 +23 M ./dft/simd/n1b.c +43 M ./dft/simd/n1b.h +24 M ./dft/simd/n1f.c +43 M ./dft/simd/n1f.h +24 R ./dft/simd/n2f.c R ./dft/simd/n2f.h R ./dft/simd/n3f.h R ./dft/simd/n4.c R ./dft/simd/n4.h M ./dft/simd/t1b.c +38 M ./dft/simd/t1b.h +25 M ./dft/simd/t1f.c +38 M ./dft/simd/t1f.h +25 R ./dft/simd/t2f.c R ./dft/simd/t2f.h R ./dft/simd/t3f.h R ./dft/simd/t4.c R ./dft/simd/t4.h M ./genfft/Makefile.am -29 +27 M ./genfft/algsimp.ml -16 +27 M ./genfft/annotate.ml -131 +11 M ./genfft/annotate.mli -17 +2 M ./genfft/c.ml -7 +23 M ./genfft/complex.ml -1 +3 M ./genfft/complex.mli -1 +2 M ./genfft/expr.ml -3 +21 M ./genfft/expr.mli -1 +6 M ./genfft/gen_athnotw.ml -3 +3 M ./genfft/gen_athtw.ml -3 +3 M ./genfft/gen_conv.ml -4 +4 M ./genfft/gen_hc2r.ml -4 +4 M ./genfft/gen_notw.ml -4 +4 M ./genfft/gen_notw_c.ml +155 M ./genfft/gen_r2hc.ml -4 +4 M ./genfft/gen_trig.ml -4 +4 M ./genfft/gen_twiddle.ml -4 +4 M ./genfft/gen_twiddle_c.ml +165 M ./genfft/gen_twidsq.ml -4 +4 M ./genfft/genutil.ml -4 +5 M ./genfft/oracle.ml -1 +2 M ./genfft/simd.ml -177 +95 M ./genfft/simd.mli -2 +2 M ./genfft/simdmagic.ml -47 +1 M ./genfft/to_alist.ml -3 +9 M ./genfft/trig.ml -1 +10 M ./genfft/trig.mli -1 +2 M ./genfft/twiddle.ml -5 +6 M ./kernel/ifftw.h -2 +2 M ./simd/simd-sse.h -61 +88 M ./simd/simd-sse2.h -67 +57 M ./simd/sse.c -1 +4 M ./simd/sse2.c -1 +4 M ./support/Makefile.codelets -67 +2 Fri Jul 5 17:32:09 EDT 2002 athena * [project @ 2002-07-05 21:32:09 by athena] Use unpck instructions instead of shuffles M ./dft/buffered.c -2 +2 M ./simd/simd-altivec.h -13 +13 M ./simd/simd-sse.h -23 +10 Fri Jul 5 15:49:14 EDT 2002 athena * [project @ 2002-07-05 19:49:14 by athena] Minor tweaks M ./dft/codelets/n.c +1 M ./dft/codelets/t.c +1 M ./dft/ct-ditbuf.c -2 +2 M ./dft/verify.c -4 +1 M ./kernel/ifftw.h -10 +9 M ./kernel/planner.c -5 +4 M ./tests/bench.c -2 +2 Fri Jul 5 15:02:54 EDT 2002 athena * [project @ 2002-07-05 19:02:54 by athena] Use score planner M ./tests/bench.c -1 +1 Fri Jul 5 14:49:59 EDT 2002 athena * [project @ 2002-07-05 18:49:59 by athena] Added verifier A ./dft/verify.c A ./kernel/dotens.c A ./kernel/dotens2.c M ./CONVENTIONS +1 M ./dft/Makefile.am -1 +1 M ./dft/dft.h -1 +4 M ./dft/verify.c +428 M ./kernel/Makefile.am -1 +2 M ./kernel/debug.c -2 +2 M ./kernel/dotens.c +47 M ./kernel/dotens2.c +55 M ./kernel/ifftw.h -8 +24 M ./tests/bench.c -8 +14 Wed Jul 3 20:32:28 EDT 2002 athena * [project @ 2002-07-04 00:32:28 by athena] More simd codelets A ./dft/simd/n2f.c A ./dft/simd/n2f.h A ./dft/simd/n3f.h A ./dft/simd/t2f.c A ./dft/simd/t2f.h A ./dft/simd/t3f.h M ./dft/buffered.c -3 +3 M ./dft/codelet.h -1 +3 M ./dft/codelets/n.c +1 M ./dft/codelets/t.c +1 M ./dft/ct-dif.c -1 +6 M ./dft/ct-dit.c -1 +6 M ./dft/ct-ditbuf.c -9 +19 M ./dft/ct.c -11 +17 M ./dft/ct.h -3 +5 M ./dft/simd/Makefile.am -1 +2 M ./dft/simd/NAMING -7 +20 M ./dft/simd/codelets/Makefile.am -4 +27 M ./dft/simd/n2f.c +43 M ./dft/simd/n2f.h +25 M ./dft/simd/n3f.h +1 M ./dft/simd/n4.c -2 +1 M ./dft/simd/t2f.c +40 M ./dft/simd/t2f.h +25 M ./dft/simd/t3f.h +1 M ./dft/simd/t4.c -1 +1 M ./genfft/annotate.ml -3 +3 M ./genfft/gen_notw.ml -9 +23 M ./genfft/gen_twiddle.ml -6 +14 M ./genfft/gen_twidsq.ml -3 +9 M ./genfft/simd.ml -43 +69 M ./genfft/simdmagic.ml -11 +36 M ./genfft-k7/gen_notw.ml -1 +2 M ./genfft-k7/gen_twiddle.ml -1 +2 M ./kernel/alloc.c -5 +2 M ./kernel/ifftw.h -3 +9 M ./simd/simd-sse.h -28 +106 M ./simd/simd-sse2.h -24 +47 Tue Jul 2 16:18:09 EDT 2002 athena * [project @ 2002-07-02 20:18:09 by athena] Oops M ./dft/rank-geq2.c -3 +4 Tue Jul 2 16:13:24 EDT 2002 athena * [project @ 2002-07-02 20:13:24 by athena] Fixed classic mode M ./dft/rank-geq2.c -6 +5 M ./dft/vrank-geq1.c -6 +6 Tue Jul 2 15:38:36 EDT 2002 athena * [project @ 2002-07-02 19:38:36 by athena] Use LDK for constants so that we can play games. M ./genfft/simd.ml -2 +2 M ./simd/simd-altivec.h +1 M ./simd/simd-sse.h -5 +5 M ./simd/simd-sse2.h -2 +5 Tue Jul 2 13:15:58 EDT 2002 athena * [project @ 2002-07-02 17:15:58 by athena] Improved support for fixed strides M ./dft/codelet.h -1 +2 M ./dft/codelets/n.c -1 +2 M ./dft/simd/n4.c +3 M ./dft/simd/t4.c +2 M ./genfft/gen_notw.ml -8 +21 M ./genfft/genutil.ml -1 +6 M ./genfft/simd.ml -7 +7 M ./genfft/simd.mli -3 +3 M ./genfft-k7/gen_notw.ml -1 +2 M ./simd/simd-sse.h -17 +54 Tue Jul 2 10:30:58 EDT 2002 athena * [project @ 2002-07-02 14:30:58 by athena] Changed accounting of flops M ./dft/codelet.h -12 +25 M ./dft/codelets/n.c -4 +5 M ./dft/codelets/n.h -6 +2 M ./dft/codelets/t.c -3 +5 M ./dft/codelets/t.h -3 +2 M ./dft/ct-dif.c -3 +5 M ./dft/ct-dit.c -3 +5 M ./dft/ct-ditbuf.c -3 +4 M ./dft/ct-ditf.c -3 +6 M ./dft/direct.c -5 +5 M ./dft/k7/k7.c -14 +23 M ./dft/simd/n4.c -3 +5 M ./dft/simd/n4.h -5 +2 M ./dft/simd/t4.c -3 +5 M ./dft/simd/t4.h -3 +2 M ./genfft/gen_notw.ml -3 +3 M ./genfft/gen_twiddle.ml -3 +3 M ./genfft/gen_twidsq.ml -3 +3 M ./genfft-k7/gen_notw.ml -2 +2 M ./genfft-k7/gen_twiddle.ml -2 +2 Tue Jul 2 08:51:38 EDT 2002 athena * [project @ 2002-07-02 12:51:38 by athena] Wrong code in non-fma mode M ./genfft/algsimp.ml -11 +15 M ./genfft-k7/algsimp.ml -8 +24 M ./genfft-k7/to_alist.ml -3 +3 M ./simd/simd-sse2.h -4 +9 Mon Jul 1 23:17:06 EDT 2002 athena * [project @ 2002-07-02 03:17:06 by athena] sse2 stuff A ./simd/simd-sse2.h A ./simd/sse2.c M ./genfft/simdmagic.ml -2 +2 M ./kernel/alloc.c -2 +2 M ./simd/Makefile.am -1 +2 M ./simd/simd-sse2.h +125 M ./simd/sse2.c +73 Mon Jul 1 14:05:56 EDT 2002 athena * [project @ 2002-07-01 18:05:56 by athena] Identify CPUs for special codelets A ./simd/sse.c M ./Makefile.am +1 M ./dft/ct.c -4 +4 M ./dft/direct.c -2 +2 M ./dft/k7/k7.c -5 +43 M ./dft/simd/n4.c -1 +1 M ./dft/simd/t4.c -1 +1 M ./kernel/alloc.c -2 +2 M ./simd/Makefile.am -1 +3 M ./simd/simd-altivec.h +2 M ./simd/simd-sse.h +3 M ./simd/sse.c +73 Mon Jul 1 09:26:42 EDT 2002 athena * [project @ 2002-07-01 13:26:42 by athena] Change split problem syntax M ./libbench/problem.c -2 +2 Mon Jul 1 09:11:39 EDT 2002 athena * [project @ 2002-07-01 13:11:39 by athena] Removed -fma flag M ./dft/simd/codelets/Makefile.am -1 +1 Sun Jun 30 20:08:26 EDT 2002 athena * [project @ 2002-07-01 00:08:26 by athena] Work around gcc bug M ./simd/simd-altivec.h -3 +17 Sun Jun 30 18:34:06 EDT 2002 athena * [project @ 2002-06-30 22:34:06 by athena] New simd stuff M ./genfft/algsimp.ml -8 +20 M ./genfft/magic.ml -1 +3 M ./genfft/oracle.ml -2 +2 M ./genfft/simd.ml -2 +4 M ./genfft/to_alist.ml -3 +3 Sun Jun 30 17:00:09 EDT 2002 athena * [project @ 2002-06-30 21:00:09 by athena] Added altivec support A ./simd/simd-altivec.h M ./dft/simd/codelets/Makefile.am -1 +1 M ./simd/Makefile.am -1 +1 M ./simd/simd-altivec.h +66 M ./simd/simd-sse.h -1 Sun Jun 30 14:47:47 EDT 2002 athena * [project @ 2002-06-30 18:47:47 by athena] Forgot file A ./dft/simd/t4.c Sun Jun 30 14:37:55 EDT 2002 athena * [project @ 2002-06-30 18:37:55 by athena] Progress towards simd implementation A ./dft/codelets/f.h A ./dft/codelets/n.c A ./dft/codelets/n.h A ./dft/codelets/q.h A ./dft/codelets/t.c A ./dft/codelets/t.h A ./dft/k7/k7.c A ./dft/simd/ A ./dft/simd/Makefile.am A ./dft/simd/NAMING A ./dft/simd/codelets/ A ./dft/simd/codelets/Makefile.am A ./dft/simd/n4.c A ./dft/simd/n4.h A ./dft/simd/t4.h A ./simd/ A ./simd/Makefile.am A ./simd/README A ./simd/simd-sse.h A ./simd/simd.h M ./Makefile.am -2 +5 M ./configure.ac -6 +32 M ./dft/Makefile.am -3 +3 R ./dft/codelet-k7.h M ./dft/codelet.h -11 +23 M ./dft/codelets/Makefile.am +4 M ./dft/codelets/f.h +1 M ./dft/codelets/inplace/Makefile.am -13 +14 M ./dft/codelets/n.c +34 M ./dft/codelets/n.h +27 M ./dft/codelets/q.h +1 M ./dft/codelets/standard/Makefile.am -13 +14 M ./dft/codelets/t.c +32 M ./dft/codelets/t.h +24 M ./dft/conf.c -2 +5 M ./dft/ct-dif.c -3 +3 M ./dft/ct-dit.c -5 +3 M ./dft/ct-ditbuf.c -2 +2 M ./dft/ct-ditf.c -4 +3 M ./dft/ct.c -6 +6 M ./dft/ct.h -3 +1 M ./dft/direct.c -6 +11 M ./dft/k7/Makefile.am -2 +1 M ./dft/k7/codelets/Makefile.am -33 +29 R ./dft/k7/ct-dif.c R ./dft/k7/ct-dit.c R ./dft/k7/ct-ditbuf.c R ./dft/k7/direct.c M ./dft/k7/k7.c +69 R ./dft/k7/kdft-dif.c R ./dft/k7/kdft-dit.c R ./dft/k7/kdft.c M ./dft/simd/Makefile.am +6 M ./dft/simd/NAMING +17 M ./dft/simd/codelets/Makefile.am +36 M ./dft/simd/n4.c +39 M ./dft/simd/n4.h +28 M ./dft/simd/t4.h +26 M ./genfft/annotate.ml -7 +7 M ./genfft/c.ml -1 +6 M ./genfft/c.mli -1 +2 M ./genfft/gen_notw.ml -12 +20 M ./genfft/gen_twiddle.ml -14 +18 M ./genfft/gen_twidsq.ml -7 +7 M ./genfft/genutil.ml -4 +17 M ./genfft/magic.ml -1 +3 M ./genfft/simd.ml -66 +34 M ./genfft/simd.mli -1 +7 M ./genfft/simdmagic.ml -2 +12 M ./genfft/twiddle.ml -1 +12 M ./genfft/twiddle.mli -1 +2 M ./genfft-k7/genUtil.ml -1 +1 M ./genfft-k7/gen_notw.ml -11 +18 M ./genfft-k7/gen_twiddle.ml -12 +17 M ./kernel/alloc.c -9 +28 M ./kernel/ifftw.h -10 +10 M ./libbench/bench-user.h -1 +2 M ./libbench/problem.c -1 +3 M ./libbench/util.c -1 M ./simd/Makefile.am +1 M ./simd/README +2 M ./simd/simd-sse.h +94 M ./simd/simd.h +33 M ./support/Makefile.codelets -33 +36 M ./tests/Makefile.am +1 M ./tests/bench.c -14 +63 Tue Jun 25 20:23:29 EDT 2002 athena * [project @ 2002-06-26 00:23:29 by athena] Add 128- codelet M ./dft/k7/codelets/Makefile.am -2 +3 Sat Jun 22 20:47:28 EDT 2002 athena * [project @ 2002-06-23 00:47:28 by athena] More simd changes. Ensure proper stack alignment in k7 codelets. M ./configure.ac +1 M ./genfft/c.ml -25 +4 M ./genfft/c.mli -2 +4 M ./genfft/expr.ml -1 +18 M ./genfft/expr.mli -1 +4 M ./genfft/gen_hc2r.ml -3 +3 M ./genfft/gen_notw.ml -3 +3 M ./genfft/gen_r2hc.ml -3 +3 M ./genfft/gen_trig.ml -3 +3 M ./genfft/gen_twiddle.ml -3 +3 M ./genfft/gen_twidsq.ml -3 +3 M ./genfft/genutil.ml -1 +13 M ./genfft/simd.ml -1 +9 M ./genfft/simd.mli -1 +2 M ./genfft-k7/genUtil.ml -34 +6 M ./genfft-k7/gen_notw.ml -2 +2 M ./genfft-k7/gen_twiddle.ml -2 +2 M ./libbench/bench-main.c -1 +11 Sat Jun 22 13:01:33 EDT 2002 athena * [project @ 2002-06-22 17:01:33 by athena] Fixed prototypes M ./kernel/ifftw.h -2 +2 M ./kernel/solvtab.c -2 +2 Sat Jun 22 12:53:26 EDT 2002 athena * [project @ 2002-06-22 16:53:26 by athena] Sparc cycle counter requires v9 M ./kernel/cycle.h -2 +2 Sat Jun 22 11:45:48 EDT 2002 athena * [project @ 2002-06-22 15:45:48 by athena] Minor fixes M ./configure.ac -1 +1 M ./kernel/cycle.h -15 +36 M ./kernel/ifftw.h -1 +5 Sat Jun 22 11:11:46 EDT 2002 athena * [project @ 2002-06-22 15:11:46 by athena] Fixed ev67 detection M ./acinclude.m4 -1 +1 Sat Jun 22 10:19:11 EDT 2002 athena * [project @ 2002-06-22 14:19:11 by athena] Print flops M ./tests/bench.c +2 Sat Jun 22 09:36:47 EDT 2002 athena * [project @ 2002-06-22 13:36:47 by athena] Nothing really M ./genfft/simd.ml -7 +3 Fri Jun 21 22:19:20 EDT 2002 athena * [project @ 2002-06-22 02:19:20 by athena] More simd work M ./dft/codelet-k7.h -2 +2 M ./dft/codelet.h -3 +3 M ./genfft/Makefile.am -2 +2 M ./genfft/c.ml -19 +1 M ./genfft/c.mli -2 +2 M ./genfft/gen_conv.ml -3 +3 M ./genfft/gen_hc2r.ml -3 +3 M ./genfft/gen_notw.ml -3 +3 M ./genfft/gen_r2hc.ml -3 +3 M ./genfft/gen_trig.ml -3 +3 M ./genfft/gen_twiddle.ml -3 +3 M ./genfft/gen_twidsq.ml -3 +3 M ./genfft/genutil.ml -2 +20 M ./genfft/magic.ml -1 +5 M ./genfft/simd.ml -47 +39 M ./genfft/simd.mli -1 +3 M ./genfft/simdmagic.ml -4 +2 M ./genfft/to_alist.ml -2 +5 M ./genfft-k7/magic.ml +4 M ./genfft-k7/to_alist.ml -2 +5 M ./genfft-k7/to_alist.mli -1 +1 M ./kernel/ifftw.h -2 +2 Thu Jun 20 21:22:41 EDT 2002 athena * [project @ 2002-06-21 01:22:41 by athena] More simd work A ./genfft/simd.mli A ./genfft/simdmagic.ml M ./genfft/Makefile.am -7 +8 M ./genfft/annotate.ml -13 +13 M ./genfft/annotate.mli -1 +2 M ./genfft/magic.ml -21 +1 M ./genfft/simd.ml -21 +10 M ./genfft/simd.mli +21 M ./genfft/simdmagic.ml +43 Thu Jun 20 18:51:33 EDT 2002 athena * [project @ 2002-06-20 22:51:33 by athena] More simd work A ./genfft/simd.ml M ./genfft/Makefile.am -23 +7 M ./genfft/annotate.ml -18 +2 M ./genfft/c.ml -6 +1 M ./genfft/c.mli -5 +8 M ./genfft/magic.ml -1 +6 M ./genfft/simd.ml +288 M ./genfft/variable.ml -1 +18 M ./genfft/variable.mli -1 +4 Thu Jun 20 15:04:37 EDT 2002 athena * [project @ 2002-06-20 19:04:37 by athena] Moving towards incorporation of simd stuff M ./genfft/annotate.ml -11 +144 M ./genfft/annotate.mli -2 +16 M ./genfft/gen_athnotw.ml -5 +5 M ./genfft/gen_athtw.ml -6 +6 M ./genfft/gen_conv.ml -3 +3 M ./genfft/gen_hc2r.ml -3 +3 M ./genfft/gen_notw.ml -3 +3 M ./genfft/gen_r2hc.ml -3 +3 M ./genfft/gen_trig.ml -3 +3 M ./genfft/gen_twiddle.ml -4 +4 M ./genfft/gen_twidsq.ml -6 +5 M ./genfft/genutil.ml -26 +17 M ./genfft/magic.ml -2 +17 M ./genfft/twiddle.ml -4 +4 M ./genfft/twiddle.mli -2 +2 M ./genfft/variable.ml -12 +22 M ./genfft/variable.mli -3 +6 Wed Jun 19 18:47:55 EDT 2002 athena * [project @ 2002-06-19 22:47:55 by athena] Reorganized k7 stuff into own directory A ./dft/k7/ A ./dft/k7/Makefile.am A ./dft/k7/codelets/ A ./dft/k7/codelets/Makefile.am A ./dft/k7/ct-dif.c A ./dft/k7/ct-dit.c A ./dft/k7/ct-ditbuf.c A ./dft/k7/direct.c A ./dft/k7/kdft-dif.c A ./dft/k7/kdft-dit.c A ./dft/k7/kdft.c M ./Makefile.am -1 +2 M ./configure.ac -2 +3 M ./dft/Makefile.am -6 +5 M ./dft/codelets/Makefile.am -1 +1 R ./dft/ct-dif-k7.c R ./dft/ct-dit-k7.c R ./dft/ct-ditbuf-k7.c R ./dft/direct-k7.c M ./dft/k7/Makefile.am +7 M ./dft/k7/codelets/Makefile.am +74 M ./dft/k7/ct-dif.c +129 M ./dft/k7/ct-dit.c +137 M ./dft/k7/ct-ditbuf.c +176 M ./dft/k7/direct.c +172 M ./dft/k7/kdft-dif.c +30 M ./dft/k7/kdft-dit.c +31 M ./dft/k7/kdft.c +31 R ./dft/kdft-dif-k7.c R ./dft/kdft-dit-k7.c R ./dft/kdft-k7.c Wed Jun 19 13:21:13 EDT 2002 athena * [project @ 2002-06-19 17:20:37 by athena] Minor experimental stuff M ./genfft/expr.ml -3 +2 M ./genfft/expr.mli -2 +2 M ./genfft/genutil.ml -2 +11 M ./genfft/magic.ml -1 +3 M ./genfft-k7/expr.ml -1 +3 M ./genfft-k7/expr.mli -1 +3 Wed Jun 19 11:20:29 EDT 2002 athena * [project @ 2002-06-19 15:20:29 by athena] Cosmetic changes M ./genfft/expr.ml -1 +4 M ./genfft/expr.mli -1 +3 M ./genfft/genutil.ml -9 +2 Wed Jun 19 01:43:31 EDT 2002 fftw * [project @ 2002-06-19 05:43:31 by fftw] allocate buffers on the fly M ./dft/buffered.c -18 +7 M ./dft/rader.c -17 +20 Tue Jun 18 17:48:41 EDT 2002 athena * [project @ 2002-06-18 21:48:41 by athena] Added ct-ditbuf-k7.c . Major changes required in generator. A ./dft/ct-ditbuf-k7.c M ./dft/Makefile.am -1 +1 M ./dft/codelet-k7.h -1 +2 M ./dft/ct-dif-k7.c -17 +3 M ./dft/ct-dif.c -16 +2 M ./dft/ct-dit-k7.c -18 +10 M ./dft/ct-dit.c -16 +2 M ./dft/ct-ditbuf-k7.c +176 M ./dft/ct-ditbuf.c -16 +2 M ./dft/ct.c -1 +30 M ./dft/ct.h -1 +3 M ./dft/kdft-dit-k7.c -1 +2 M ./dft/rader.c -1 +1 M ./genfft/number.ml -2 +2 M ./genfft/to_alist.ml -2 +2 M ./genfft-k7/Makefile.am -4 +4 M ./genfft-k7/assignmentsToVfpinstrs.ml -1 +4 M ./genfft-k7/gen_twiddle.ml -2 +5 M ./genfft-k7/k7Basics.ml -1 +1 M ./genfft-k7/k7Basics.mli +1 M ./genfft-k7/k7RegisterAllocationBasics.ml +3 M ./genfft-k7/k7RegisterAllocationBasics.mli -2 +4 M ./genfft-k7/k7RegisterAllocator.ml -15 +28 M ./genfft-k7/k7RegisterAllocatorInit.ml -6 +12 M ./genfft-k7/number.ml -2 +2 M ./genfft-k7/to_alist.ml -2 +2 M ./kernel/ifftw.h -2 +2 M ./kernel/planner.c -3 +5 M ./kernel/primes.c -2 +2 Tue Jun 18 11:55:57 EDT 2002 athena * [project @ 2002-06-18 15:55:57 by athena] Nothing, really M ./genfft-k7/gen_twiddle.ml -2 +2 M ./kernel/ifftw.h -18 +19 Tue Jun 18 11:19:59 EDT 2002 athena * [project @ 2002-06-18 15:19:59 by athena] !SINGLE ==> !K7_MODE (for some reason the contrapositive sounds wrong) M ./configure.ac +1 Tue Jun 18 11:07:13 EDT 2002 athena * [project @ 2002-06-18 15:07:13 by athena] Buffer is now symmetric wrt forward/backward transform M ./dft/buffered.c -12 +22 Tue Jun 18 10:33:58 EDT 2002 athena * [project @ 2002-06-18 14:33:58 by athena] Fixed applicable() in indirect.c A ./kernel/debug.c M ./dft/ct-dif.c -2 +1 M ./dft/indirect.c -3 +3 M ./dft/vrank2-transpose.c -10 +14 M ./dft/vrank3-transpose.c -10 +13 M ./kernel/Makefile.am -1 +1 M ./kernel/debug.c +39 M ./kernel/ifftw.h -1 +5 M ./kernel/print.c -4 +10 Tue Jun 18 08:41:18 EDT 2002 athena * [project @ 2002-06-18 12:41:18 by athena] Fixed attempt to free() uninitialized pointer. M ./dft/rader.c -2 +3 Tue Jun 18 08:13:55 EDT 2002 athena * [project @ 2002-06-18 12:13:55 by athena] Added reference counts for awake() M ./CONVENTIONS -1 +1 M ./TODO +5 M ./dft/rader.c -11 +3 M ./kernel/ifftw.h -4 +7 M ./kernel/plan.c -2 +15 M ./kernel/planner.c -3 +3 M ./tests/bench.c -2 +3 Mon Jun 17 20:49:05 EDT 2002 stevenj * [project @ 2002-06-18 00:49:05 by stevenj] updated comment M ./dft/rader.c -2 +2 Mon Jun 17 20:46:45 EDT 2002 stevenj * [project @ 2002-06-18 00:46:45 by stevenj] slight update M ./TODO -1 +3 Mon Jun 17 02:30:16 EDT 2002 fftw * [project @ 2002-06-17 06:30:16 by fftw] moved prime-number stuff into primes.c, so it can be shared with generic codelet and with rfftw rader A ./kernel/primes.c M ./dft/rader.c -104 +5 M ./kernel/Makefile.am -3 +3 M ./kernel/ifftw.h -1 +24 M ./kernel/primes.c +112 Mon Jun 17 02:01:58 EDT 2002 fftw * [project @ 2002-06-17 06:01:58 by fftw] added comment M ./dft/rader.c -6 +9 Mon Jun 17 01:39:55 EDT 2002 fftw * [project @ 2002-06-17 05:39:55 by fftw] added rader-dit M ./dft/rader.c -66 +308 M ./kernel/ifftw.h -1 +10 M ./kernel/twiddle.c -8 +1 Sun Jun 16 23:50:16 EDT 2002 fftw * [project @ 2002-06-17 03:50:16 by fftw] added initial Rader (no DIT yet) A ./dft/rader.c M ./configure.ac +8 M ./dft/Makefile.am -1 +1 M ./dft/conf.c -1 +2 M ./dft/dft.h -1 +2 M ./dft/rader.c +457 Sun Jun 16 22:29:51 EDT 2002 fftw * [project @ 2002-06-17 02:29:51 by fftw] don't warn about long long M ./acinclude.m4 -1 +1 Sun Jun 16 21:30:42 EDT 2002 athena * [project @ 2002-06-17 01:30:42 by athena] Added k7 DIF codelets A ./dft/ct-dif-k7.c A ./dft/kdft-dif-k7.c M ./dft/Makefile.am -5 +6 M ./dft/codelet-k7.h -5 +10 M ./dft/ct-dif-k7.c +143 M ./dft/ct.h -1 +2 M ./dft/kdft-dif-k7.c +30 M ./genfft-k7/gen_twiddle.ml -4 +25 M ./kernel/planner.c -3 +2 M ./tests/bench.c +1 Sun Jun 16 19:13:31 EDT 2002 athena * [project @ 2002-06-16 23:13:31 by athena] Added stuff to do M ./TODO +6 Sun Jun 16 19:05:58 EDT 2002 athena * [project @ 2002-06-16 23:05:58 by athena] Handle dual case R = I + 1 M ./dft/rank0.c -6 +30 Sun Jun 16 18:54:31 EDT 2002 athena * [project @ 2002-06-16 22:54:31 by athena] Removed useless flag M ./bootstrap.sh -1 +1 Sun Jun 16 18:30:32 EDT 2002 athena * [project @ 2002-06-16 22:30:32 by athena] Removed useless file R ./mkdist.sh Sun Jun 16 18:30:18 EDT 2002 athena * [project @ 2002-06-16 22:30:18 by athena] More k7 work. Switched to runtime CLASSIC mode. A ./dft/ct-dit-k7.c A ./dft/kdft-dit-k7.c R ./CLASSIC-MODE M ./Makefile.am -10 +1 M ./configure.ac -21 M ./dft/Makefile.am -5 +5 M ./dft/buffered.c -2 +4 M ./dft/codelet-k7.h -5 +7 M ./dft/codelet.h -1 +2 M ./dft/codelets/inplace/Makefile.am -2 M ./dft/ct-dif.c -3 +4 M ./dft/ct-dit-k7.c +145 M ./dft/ct-dit.c -6 +9 M ./dft/ct-ditbuf.c -9 +11 M ./dft/ct-ditf.c -2 +3 M ./dft/ct.h -5 +3 M ./dft/direct-k7.c -2 +3 M ./dft/direct.c -2 +3 M ./dft/indirect.c -2 +3 M ./dft/kdft-dit-k7.c +30 M ./dft/kdft-dit.c -4 +2 M ./dft/nop.c -2 +3 M ./dft/rank-geq2.c -8 +8 M ./dft/rank0.c -2 +3 M ./dft/vrank-geq1.c -6 +7 M ./dft/vrank2-transpose.c -3 +3 M ./dft/vrank3-transpose.c -3 +3 M ./genfft/gen_twiddle.ml -4 +4 M ./genfft/gen_twidsq.ml -4 +4 M ./genfft/twiddle.ml -7 +28 M ./genfft/twiddle.mli -2 +5 M ./genfft-k7/Makefile.am -25 +24 M ./genfft-k7/genUtil.ml -2 +3 M ./genfft-k7/gen_notw.ml -1 +2 M ./genfft-k7/gen_twiddle.ml -23 +29 M ./genfft-k7/twiddle.ml -7 +36 M ./genfft-k7/twiddle.mli -3 +7 M ./kernel/ifftw.h -15 +13 M ./kernel/planner-naive.c -3 +3 M ./kernel/planner-score.c -5 +6 M ./kernel/planner.c -6 +5 M ./kernel/version.c -7 +2 M ./support/Makefile.codelets +1 M ./tests/bench.c -12 +2 Sun Jun 16 17:15:18 EDT 2002 stevenj * [project @ 2002-06-16 21:15:18 by stevenj] spelling M ./kernel/tensor.c -2 +2 Sun Jun 16 15:51:44 EDT 2002 athena * [project @ 2002-06-16 19:51:44 by athena] Do not compile if not K7_MODE M ./dft/kdft-k7.c -1 +3 Sun Jun 16 15:35:02 EDT 2002 athena * [project @ 2002-06-16 19:35:02 by athena] Do not require K7 definitions to compile M ./dft/codelet-k7.h -1 +2 M ./dft/dft.h -2 +1 Sun Jun 16 08:05:17 EDT 2002 athena * [project @ 2002-06-16 12:05:17 by athena] More k7 stuff A ./dft/codelet-k7.h A ./genfft-k7/gen_twiddle.ml M ./dft/Makefile.am -4 +4 M ./dft/codelet-k7.h +40 M ./dft/codelet.h -20 +5 M ./dft/direct-k7.c -1 +5 M ./genfft-k7/Makefile.am -1 +9 M ./genfft-k7/complex.ml -1 +2 M ./genfft-k7/complex.mli -1 +2 M ./genfft-k7/genUtil.ml -1 +12 M ./genfft-k7/gen_notw.ml -5 +1 M ./genfft-k7/gen_twiddle.ml +149 Sat Jun 15 18:30:43 EDT 2002 athena * [project @ 2002-06-15 22:30:43 by athena] Try to be compatible with automake-1.6 M ./acinclude.m4 -1 +5 Sat Jun 15 18:23:40 EDT 2002 athena * [project @ 2002-06-15 22:23:40 by athena] More merging of Stefan's generator with main genfft branch A ./genfft-k7/algsimp.ml A ./genfft-k7/algsimp.mli A ./genfft-k7/assoctable.ml A ./genfft-k7/assoctable.mli A ./genfft-k7/littlesimp.ml A ./genfft-k7/littlesimp.mli A ./genfft-k7/monads.ml A ./genfft-k7/oracle.ml A ./genfft-k7/oracle.mli A ./genfft-k7/to_alist.ml A ./genfft-k7/to_alist.mli A ./genfft-k7/twiddle.mli A ./support/codelet_asmprelude M ./acinclude.m4 +8 M ./configure.ac +1 M ./genfft/number.ml -7 +8 M ./genfft-k7/Makefile.am -23 +27 M ./genfft-k7/algsimp.ml +517 M ./genfft-k7/algsimp.mli +24 M ./genfft-k7/assignmentsToVfpinstrs.ml -11 +10 M ./genfft-k7/assoctable.ml +66 M ./genfft-k7/assoctable.mli +30 M ./genfft-k7/complex.ml -4 +19 M ./genfft-k7/complex.mli -5 +6 M ./genfft-k7/expr.ml -27 +58 M ./genfft-k7/expr.mli -4 +11 R ./genfft-k7/exprdag.ml R ./genfft-k7/exprdag.mli M ./genfft-k7/genUtil.ml -13 +18 M ./genfft-k7/gen_notw.ml -16 +31 M ./genfft-k7/k7Unparsing.ml -2 +2 M ./genfft-k7/littlesimp.ml +72 M ./genfft-k7/littlesimp.mli +26 M ./genfft-k7/magic.ml -16 +41 M ./genfft-k7/monads.ml +76 M ./genfft-k7/number.ml -9 +5 M ./genfft-k7/number.mli -8 +4 M ./genfft-k7/oracle.ml +138 M ./genfft-k7/oracle.mli +25 M ./genfft-k7/to_alist.ml +258 M ./genfft-k7/to_alist.mli +24 M ./genfft-k7/twiddle.ml -5 +22 M ./genfft-k7/twiddle.mli +29 M ./genfft-k7/util.ml -4 +4 M ./genfft-k7/util.mli -2 +1 M ./genfft-k7/vFpUnparsing.ml -1 +1 M ./genfft-k7/vSimdBasics.ml -2 +3 M ./genfft-k7/vSimdUnparsing.ml -1 +1 M ./genfft-k7/variable.ml +1 M ./genfft-k7/variable.mli +1 M ./support/Makefile.am -1 +1 M ./support/Makefile.codelets -1 +13 M ./support/codelet_asmprelude +8 Sat Jun 15 13:51:39 EDT 2002 athena * [project @ 2002-06-15 17:51:39 by athena] Slowly merging genfft-k7 with main genfft branch A ./genfft-k7/gen_notw.ml M ./genfft/expr.ml -1 +3 M ./genfft/expr.mli -1 +2 M ./genfft/genutil.ml -1 +9 M ./genfft-k7/Makefile.am -26 +23 M ./genfft-k7/complex.ml -34 +92 M ./genfft-k7/complex.mli -1 +11 M ./genfft-k7/expr.ml -1 +19 M ./genfft-k7/expr.mli -4 +3 M ./genfft-k7/exprdag.ml -73 +103 M ./genfft-k7/exprdag.mli -1 M ./genfft-k7/fft.ml -167 +83 M ./genfft-k7/fft.mli -29 +1 M ./genfft-k7/genUtil.ml -56 +33 R ./genfft-k7/gen_hc2hc.ml R ./genfft-k7/gen_hc2real.ml M ./genfft-k7/gen_notw.ml +143 R ./genfft-k7/gen_notwiddle.ml R ./genfft-k7/gen_notwiddle_fixedstride.ml R ./genfft-k7/gen_real2hc.ml R ./genfft-k7/gen_realeven.ml R ./genfft-k7/gen_realeven2.ml R ./genfft-k7/gen_realodd.ml R ./genfft-k7/gen_realodd2.ml R ./genfft-k7/gen_twiddle.ml M ./genfft-k7/magic.ml -11 +35 R ./genfft-k7/symmetry.ml M ./genfft-k7/twiddle.ml -63 +95 M ./genfft-k7/util.ml +15 M ./genfft-k7/util.mli +3 M ./genfft-k7/variable.ml +12 M ./genfft-k7/variable.mli +3 M ./support/Makefile.codelets -1 +1 Fri Jun 14 21:33:02 EDT 2002 athena * [project @ 2002-06-15 01:33:02 by athena] Fixed, really M ./genfft-k7/Makefile.am -1 +1 M ./genfft-k7/genUtil.ml -25 M ./genfft-k7/magic.ml -1 +1 R ./genfft-k7/magic.mli M ./genfft-k7/twiddle.ml -2 +2 M ./support/Makefile.codelets -4 +8 Fri Jun 14 21:27:12 EDT 2002 athena * [project @ 2002-06-15 01:27:12 by athena] Oops... M ./support/Makefile.codelets -1 +1 Fri Jun 14 21:25:34 EDT 2002 athena * [project @ 2002-06-15 01:25:34 by athena] Work properly when $(ALL_CODELETS) = "" M ./support/Makefile.codelets -4 +4 Fri Jun 14 21:11:16 EDT 2002 athena * [project @ 2002-06-15 01:11:16 by athena] Fixed k7 build machinery M ./Makefile.am -1 +2 M ./configure.ac +8 M ./dft/codelet.h -1 +4 M ./dft/codelets/Makefile.am -1 +1 M ./dft/conf.c -1 +4 M ./genfft-k7/gen_notwiddle.ml -1 +3 M ./kernel/ifftw.h -1 +10 M ./support/Makefile.codelets +2 Fri Jun 14 17:42:35 EDT 2002 athena * [project @ 2002-06-14 21:42:35 by athena] More work on k7 stuff M ./Makefile.am -1 +1 M ./configure.ac +1 M ./dft/codelet.h -1 +4 M ./dft/direct-k7.c -3 +3 M ./genfft-k7/Makefile.am -24 +24 R ./genfft-k7/codeletMisc.ml R ./genfft-k7/codeletMisc.mli M ./genfft-k7/genUtil.ml -10 +128 R ./genfft-k7/genUtil.mli R ./genfft-k7/gen_hc2hc.mli R ./genfft-k7/gen_hc2real.mli M ./genfft-k7/gen_notwiddle.ml -3 +33 R ./genfft-k7/gen_notwiddle.mli R ./genfft-k7/gen_real2hc.mli R ./genfft-k7/gen_realeven.mli R ./genfft-k7/gen_realeven2.mli R ./genfft-k7/gen_realodd.mli R ./genfft-k7/gen_realodd2.mli R ./genfft-k7/gen_twiddle.mli R ./genfft-k7/genfft.ml M ./genfft-k7/k7Basics.ml +19 M ./genfft-k7/k7Basics.mli +2 M ./genfft-k7/k7Unparsing.ml -1 +1 M ./genfft-k7/magic.ml -1 M ./genfft-k7/magic.mli -1 M ./kernel/ifftw.h -1 +4 M ./libbench/bench-user.h -1 +3 M ./support/Makefile.am -1 +2 Fri Jun 14 15:54:29 EDT 2002 athena * [project @ 2002-06-14 19:54:29 by athena] More work on k7 stuff A ./dft/direct-k7.c A ./dft/kdft-k7.c M ./dft/Makefile.am -1 +2 M ./dft/codelet.h -1 +13 M ./dft/conf.c -9 +15 M ./dft/dft.h -1 +2 M ./dft/direct-k7.c +167 M ./dft/kdft-k7.c +29 M ./genfft-k7/codeletMisc.ml -91 M ./genfft-k7/codeletMisc.mli -13 +1 M ./genfft-k7/genUtil.ml -24 +19 M ./genfft-k7/gen_notwiddle.ml -2 +44 M ./genfft-k7/genfft.ml +4 M ./genfft-k7/magic.ml +1 M ./genfft-k7/magic.mli +1 M ./support/Makefile.codelets -1 +1 Fri Jun 14 14:18:15 EDT 2002 athena * [project @ 2002-06-14 18:18:15 by athena] Changed my mind again M ./dft/codelet.h -4 +1 M ./dft/direct.c -7 +4 M ./genfft/gen_notw.ml -5 +4 Fri Jun 14 11:53:09 EDT 2002 athena * [project @ 2002-06-14 15:53:09 by athena] Removed some useless stuff. M ./genfft-k7/gen_notwiddle.ml -8 +3 Fri Jun 14 11:01:39 EDT 2002 athena * [project @ 2002-06-14 15:01:39 by athena] Hmm... M ./genfft-k7/gen_notwiddle.ml -3 +8 Fri Jun 14 10:28:12 EDT 2002 athena * [project @ 2002-06-14 14:28:12 by athena] More work in preparation for k7 stuff M ./dft/codelet.h -1 +2 M ./dft/direct.c -3 +4 M ./genfft/gen_notw.ml -4 +5 M ./genfft-k7/gen_notwiddle.ml -18 +13 Fri Jun 14 07:25:28 EDT 2002 athena * [project @ 2002-06-14 11:25:28 by athena] Still preparing to include k7 stuff M ./TODO +4 M ./dft/codelet.h -4 +6 M ./dft/direct.c -1 +5 M ./genfft/gen_notw.ml -7 +9 Fri Jun 14 07:06:02 EDT 2002 athena * [project @ 2002-06-14 11:06:02 by athena] Create .depend M ./bootstrap.sh -2 +2 Fri Jun 14 06:56:15 EDT 2002 athena * [project @ 2002-06-14 10:56:14 by athena] Imported Stefan's K7 generator A ./genfft-k7/ A ./genfft-k7/Makefile.am A ./genfft-k7/assignmentsToVfpinstrs.ml A ./genfft-k7/assignmentsToVfpinstrs.mli A ./genfft-k7/balanceVfpinstrs.ml A ./genfft-k7/balanceVfpinstrs.mli A ./genfft-k7/codeletMisc.ml A ./genfft-k7/codeletMisc.mli A ./genfft-k7/complex.ml A ./genfft-k7/complex.mli A ./genfft-k7/expr.ml A ./genfft-k7/expr.mli A ./genfft-k7/exprdag.ml A ./genfft-k7/exprdag.mli A ./genfft-k7/fft.ml A ./genfft-k7/fft.mli A ./genfft-k7/genUtil.ml A ./genfft-k7/genUtil.mli A ./genfft-k7/gen_hc2hc.ml A ./genfft-k7/gen_hc2hc.mli A ./genfft-k7/gen_hc2real.ml A ./genfft-k7/gen_hc2real.mli A ./genfft-k7/gen_notwiddle.ml A ./genfft-k7/gen_notwiddle.mli A ./genfft-k7/gen_notwiddle_fixedstride.ml A ./genfft-k7/gen_real2hc.ml A ./genfft-k7/gen_real2hc.mli A ./genfft-k7/gen_realeven.ml A ./genfft-k7/gen_realeven.mli A ./genfft-k7/gen_realeven2.ml A ./genfft-k7/gen_realeven2.mli A ./genfft-k7/gen_realodd.ml A ./genfft-k7/gen_realodd.mli A ./genfft-k7/gen_realodd2.ml A ./genfft-k7/gen_realodd2.mli A ./genfft-k7/gen_twiddle.ml A ./genfft-k7/gen_twiddle.mli A ./genfft-k7/genfft.ml A ./genfft-k7/id.ml A ./genfft-k7/id.mli A ./genfft-k7/k7Basics.ml A ./genfft-k7/k7Basics.mli A ./genfft-k7/k7ExecutionModel.ml A ./genfft-k7/k7ExecutionModel.mli A ./genfft-k7/k7FlatInstructionScheduling.ml A ./genfft-k7/k7FlatInstructionScheduling.mli A ./genfft-k7/k7InstructionSchedulingBasics.ml A ./genfft-k7/k7InstructionSchedulingBasics.mli A ./genfft-k7/k7RegisterAllocationBasics.ml A ./genfft-k7/k7RegisterAllocationBasics.mli A ./genfft-k7/k7RegisterAllocator.ml A ./genfft-k7/k7RegisterAllocator.mli A ./genfft-k7/k7RegisterAllocatorEATranslation.ml A ./genfft-k7/k7RegisterAllocatorEATranslation.mli A ./genfft-k7/k7RegisterAllocatorInit.ml A ./genfft-k7/k7RegisterAllocatorInit.mli A ./genfft-k7/k7RegisterReallocation.ml A ./genfft-k7/k7RegisterReallocation.mli A ./genfft-k7/k7Translate.ml A ./genfft-k7/k7Translate.mli A ./genfft-k7/k7Unparsing.ml A ./genfft-k7/k7Unparsing.mli A ./genfft-k7/k7Vectorization.ml A ./genfft-k7/k7Vectorization.mli A ./genfft-k7/magic.ml A ./genfft-k7/magic.mli A ./genfft-k7/memoMonad.ml A ./genfft-k7/memoMonad.mli A ./genfft-k7/nonDetMonad.ml A ./genfft-k7/nonDetMonad.mli A ./genfft-k7/nullVectorization.ml A ./genfft-k7/nullVectorization.mli A ./genfft-k7/number.ml A ./genfft-k7/number.mli A ./genfft-k7/stateMonad.ml A ./genfft-k7/stateMonad.mli A ./genfft-k7/symmetry.ml A ./genfft-k7/twiddle.ml A ./genfft-k7/util.ml A ./genfft-k7/util.mli A ./genfft-k7/vAnnotatedScheduler.ml A ./genfft-k7/vAnnotatedScheduler.mli A ./genfft-k7/vDag.ml A ./genfft-k7/vDag.mli A ./genfft-k7/vFpBasics.ml A ./genfft-k7/vFpBasics.mli A ./genfft-k7/vFpUnparsing.ml A ./genfft-k7/vFpUnparsing.mli A ./genfft-k7/vImproveSchedule.ml A ./genfft-k7/vImproveSchedule.mli A ./genfft-k7/vK7Optimization.ml A ./genfft-k7/vK7Optimization.mli A ./genfft-k7/vScheduler.ml A ./genfft-k7/vScheduler.mli A ./genfft-k7/vSimdBasics.ml A ./genfft-k7/vSimdBasics.mli A ./genfft-k7/vSimdIndexing.ml A ./genfft-k7/vSimdIndexing.mli A ./genfft-k7/vSimdUnparsing.ml A ./genfft-k7/vSimdUnparsing.mli A ./genfft-k7/variable.ml A ./genfft-k7/variable.mli M ./AUTHORS -2 +5 M ./Makefile.am -1 +1 M ./bootstrap.sh +2 M ./configure.ac +1 M ./genfft-k7/Makefile.am +108 M ./genfft-k7/assignmentsToVfpinstrs.ml +225 M ./genfft-k7/assignmentsToVfpinstrs.mli +38 M ./genfft-k7/balanceVfpinstrs.ml +195 M ./genfft-k7/balanceVfpinstrs.mli +29 M ./genfft-k7/codeletMisc.ml +127 M ./genfft-k7/codeletMisc.mli +45 M ./genfft-k7/complex.ml +142 M ./genfft-k7/complex.mli +44 M ./genfft-k7/expr.ml +46 M ./genfft-k7/expr.mli -1 +34 M ./genfft-k7/exprdag.ml +879 M ./genfft-k7/exprdag.mli +42 M ./genfft-k7/fft.ml +311 M ./genfft-k7/fft.mli +50 M ./genfft-k7/genUtil.ml +377 M ./genfft-k7/genUtil.mli +41 M ./genfft-k7/gen_hc2hc.ml +133 M ./genfft-k7/gen_hc2hc.mli +25 M ./genfft-k7/gen_hc2real.ml +77 M ./genfft-k7/gen_hc2real.mli +24 M ./genfft-k7/gen_notwiddle.ml +69 M ./genfft-k7/gen_notwiddle.mli +25 M ./genfft-k7/gen_notwiddle_fixedstride.ml +45 M ./genfft-k7/gen_real2hc.ml +72 M ./genfft-k7/gen_real2hc.mli +24 M ./genfft-k7/gen_realeven.ml +55 M ./genfft-k7/gen_realeven.mli +24 M ./genfft-k7/gen_realeven2.ml +57 M ./genfft-k7/gen_realeven2.mli +24 M ./genfft-k7/gen_realodd.ml +60 M ./genfft-k7/gen_realodd.mli +24 M ./genfft-k7/gen_realodd2.ml +60 M ./genfft-k7/gen_realodd2.mli +24 M ./genfft-k7/gen_twiddle.ml +98 M ./genfft-k7/gen_twiddle.mli +25 M ./genfft-k7/genfft.ml +282 M ./genfft-k7/id.ml +61 M ./genfft-k7/id.mli +79 M ./genfft-k7/k7Basics.ml +524 M ./genfft-k7/k7Basics.mli +219 M ./genfft-k7/k7ExecutionModel.ml +111 M ./genfft-k7/k7ExecutionModel.mli +24 M ./genfft-k7/k7FlatInstructionScheduling.ml +177 M ./genfft-k7/k7FlatInstructionScheduling.mli +22 M ./genfft-k7/k7InstructionSchedulingBasics.ml +205 M ./genfft-k7/k7InstructionSchedulingBasics.mli +74 M ./genfft-k7/k7RegisterAllocationBasics.ml +162 M ./genfft-k7/k7RegisterAllocationBasics.mli +88 M ./genfft-k7/k7RegisterAllocator.ml +516 M ./genfft-k7/k7RegisterAllocator.mli +23 M ./genfft-k7/k7RegisterAllocatorEATranslation.ml +134 M ./genfft-k7/k7RegisterAllocatorEATranslation.mli +27 M ./genfft-k7/k7RegisterAllocatorInit.ml +97 M ./genfft-k7/k7RegisterAllocatorInit.mli +30 M ./genfft-k7/k7RegisterReallocation.ml +259 M ./genfft-k7/k7RegisterReallocation.mli +22 M ./genfft-k7/k7Translate.ml +142 M ./genfft-k7/k7Translate.mli +87 M ./genfft-k7/k7Unparsing.ml +326 M ./genfft-k7/k7Unparsing.mli +53 M ./genfft-k7/k7Vectorization.ml +739 M ./genfft-k7/k7Vectorization.mli +25 M ./genfft-k7/magic.ml +80 M ./genfft-k7/magic.mli +55 M ./genfft-k7/memoMonad.ml +32 M ./genfft-k7/memoMonad.mli +24 M ./genfft-k7/nonDetMonad.ml +136 M ./genfft-k7/nonDetMonad.mli +94 M ./genfft-k7/nullVectorization.ml +78 M ./genfft-k7/nullVectorization.mli +24 M ./genfft-k7/number.ml +158 M ./genfft-k7/number.mli +53 M ./genfft-k7/stateMonad.ml +71 M ./genfft-k7/stateMonad.mli +41 M ./genfft-k7/symmetry.ml +314 M ./genfft-k7/twiddle.ml +116 M ./genfft-k7/util.ml -1 +322 M ./genfft-k7/util.mli +124 M ./genfft-k7/vAnnotatedScheduler.ml +182 M ./genfft-k7/vAnnotatedScheduler.mli +39 M ./genfft-k7/vDag.ml +123 M ./genfft-k7/vDag.mli +52 M ./genfft-k7/vFpBasics.ml +203 M ./genfft-k7/vFpBasics.mli +106 M ./genfft-k7/vFpUnparsing.ml +79 M ./genfft-k7/vFpUnparsing.mli +28 M ./genfft-k7/vImproveSchedule.ml +155 M ./genfft-k7/vImproveSchedule.mli +22 M ./genfft-k7/vK7Optimization.ml +626 M ./genfft-k7/vK7Optimization.mli +22 M ./genfft-k7/vScheduler.ml +166 M ./genfft-k7/vScheduler.mli +31 M ./genfft-k7/vSimdBasics.ml +296 M ./genfft-k7/vSimdBasics.mli +247 M ./genfft-k7/vSimdIndexing.ml +119 M ./genfft-k7/vSimdIndexing.mli +68 M ./genfft-k7/vSimdUnparsing.ml +104 M ./genfft-k7/vSimdUnparsing.mli +33 M ./genfft-k7/variable.ml -1 +119 M ./genfft-k7/variable.mli +53 Thu Jun 13 15:30:41 EDT 2002 athena * [project @ 2002-06-13 19:30:41 by athena] Generator for real->halfcomplex and halfcomplex->real codelets A ./genfft/gen_hc2r.ml A ./genfft/gen_r2hc.ml M ./genfft/Makefile.am -9 +27 M ./genfft/c.ml -2 +2 M ./genfft/complex.ml -1 +8 M ./genfft/complex.mli -1 +3 M ./genfft/gen_hc2r.ml +121 M ./genfft/gen_r2hc.ml +121 M ./genfft/genutil.ml -8 +9 M ./genfft/trig.ml -2 +2 Thu Jun 13 11:54:02 EDT 2002 athena * [project @ 2002-06-13 15:54:02 by athena] Improved hash functions, printers M ./dft/problem.c -5 +10 M ./kernel/planner.c -11 +5 M ./kernel/tensor.c -7 +6 M ./tests/bench.c -1 +1 Thu Jun 13 11:17:31 EDT 2002 athena * [project @ 2002-06-13 15:17:31 by athena] Only regenerate codlist.c in maintainer mode M ./support/Makefile.codelets -18 +18 Thu Jun 13 11:04:24 EDT 2002 athena * [project @ 2002-06-13 15:04:24 by athena] Planner can export solution list M ./dft/problem.c -10 +21 M ./dft/rank-geq2.c -2 +2 M ./kernel/ifftw.h -11 +13 M ./kernel/planner-naive.c -7 +7 M ./kernel/planner-score.c -8 +7 M ./kernel/planner.c -42 +47 M ./kernel/print.c -9 +27 M ./kernel/tensor.c -4 +4 M ./tests/bench.c -5 +6 Thu Jun 13 08:59:53 EDT 2002 athena * [project @ 2002-06-13 12:59:53 by athena] Fixed for intel compiler M ./dft/ct-ditbuf.c -2 +3 M ./dft/dft.h -2 +1 M ./dft/direct.c -3 +1 M ./kernel/cycle.h -2 +2 M ./libbench/bench-user.h -12 +1 Thu Jun 13 08:48:51 EDT 2002 athena * [project @ 2002-06-13 12:48:51 by athena] Revised strategy for constants in codelets M ./dft/codelet.h -13 +2 M ./genfft/c.ml -50 +43 M ./genfft/c.mli -1 +4 M ./genfft/gen_notw.ml -5 +5 M ./genfft/gen_trig.ml -3 +3 M ./genfft/gen_twiddle.ml -3 +3 M ./genfft/gen_twidsq.ml -3 +3 M ./genfft/magic.ml -5 +1 Thu Jun 13 06:21:31 EDT 2002 athena * [project @ 2002-06-13 10:21:31 by athena] Enable score planner in classic mode, naive planner in pro mode. M ./tests/bench.c -3 +3 Wed Jun 12 19:18:18 EDT 2002 athena * [project @ 2002-06-12 23:18:18 by athena] Report classic/pro M ./tests/bench.c -2 +7 Wed Jun 12 19:07:48 EDT 2002 athena * [project @ 2002-06-12 23:07:48 by athena] Fixed behavior of buffered solver for large buffers. M ./dft/buffered.c -5 +5 M ./tests/bench.c -1 +1 Wed Jun 12 18:57:19 EDT 2002 athena * [project @ 2002-06-12 22:57:19 by athena] Make assumption COST(vector) = length * COST(scalar) in classic mode. M ./dft/rank-geq2.c -2 +2 M ./dft/vrank-geq1.c -1 +2 M ./kernel/ifftw.h -4 +7 M ./kernel/planner-naive.c -4 +2 M ./kernel/planner-score.c -5 +3 M ./kernel/planner.c -13 +20 M ./kernel/timer.c -8 +2 M ./libbench/timer.c -2 +2 M ./tests/bench.c -1 +1 Wed Jun 12 18:19:48 EDT 2002 athena * [project @ 2002-06-12 22:19:48 by athena] Revised planner implementation in preparation for wisdom. M ./kernel/ifftw.h -5 +6 M ./kernel/plan.c -1 +2 M ./kernel/planner-naive.c -3 +7 M ./kernel/planner-score.c -3 +7 M ./kernel/planner.c -46 +81 M ./support/Makefile.codelets -19 +19 Wed Jun 12 08:27:36 EDT 2002 athena * [project @ 2002-06-12 12:27:36 by athena] Manually hoist loop invariants. M ./dft/ct-ditbuf.c -5 +11 Wed Jun 12 07:47:41 EDT 2002 athena * [project @ 2002-06-12 11:47:41 by athena] Revised loop to compile better with gcc -O M ./dft/rank-geq2.c -16 +8 M ./dft/rank0.c -23 +22 M ./dft/vrank-geq1.c -2 +2 Tue Jun 11 17:24:09 EDT 2002 athena * [project @ 2002-06-11 21:24:09 by athena] Changed tensor syntax M ./kernel/tensor.c -4 +5 Tue Jun 11 16:39:45 EDT 2002 athena * [project @ 2002-06-11 20:39:45 by athena] Added stuff to do. M ./TODO +15 Tue Jun 11 16:28:14 EDT 2002 athena * [project @ 2002-06-11 20:28:14 by athena] Report classic/pro in version number M ./kernel/version.c -2 +7 Tue Jun 11 14:22:49 EDT 2002 athena * [project @ 2002-06-11 18:22:49 by athena] Renamed versions into classic/pro A ./CLASSIC-MODE M ./Makefile.am -7 +7 R ./RESEARCH-MODE M ./bootstrap.sh -1 +1 M ./configure.ac -13 +13 M ./dft/codelets/inplace/Makefile.am -1 +1 M ./dft/ct-dit.c -2 +2 M ./dft/ct-ditbuf.c -2 +2 M ./dft/kdft-dit.c -2 +2 M ./dft/rank-geq2.c -4 +4 M ./dft/vrank-geq1.c -4 +4 M ./kernel/ifftw.h -3 +3 M ./mkdist.sh -2 +2 M ./tests/bench.c -1 +1 Tue Jun 11 14:06:06 EDT 2002 athena * [project @ 2002-06-11 18:06:06 by athena] Revised planners, estimator M ./kernel/Makefile.am -4 +4 M ./kernel/ifftw.h -6 +7 R ./kernel/planner-estimate.c M ./kernel/planner-naive.c -4 +4 M ./kernel/planner-score.c -4 +4 M ./kernel/planner.c -2 +17 M ./tests/bench.c -5 +5 Tue Jun 11 11:45:41 EDT 2002 athena * [project @ 2002-06-11 15:45:41 by athena] I don't know what I am doing. M ./Makefile.am -2 +2 M ./dft/buffered.c -3 +2 M ./dft/ct-dit.c -4 +5 M ./dft/ct-ditbuf.c -4 +6 M ./dft/kdft-dif.c -3 +2 M ./dft/kdft-difsq.c -3 +2 M ./dft/kdft-dit.c -7 +4 M ./kernel/ifftw.h -3 +2 Tue Jun 11 10:35:52 EDT 2002 athena * [project @ 2002-06-11 14:35:52 by athena] Massive revision of estimator A ./kernel/ops.c M ./Makefile.am -1 +1 M ./dft/buffered.c -8 +7 M ./dft/codelet.h -3 +3 M ./dft/ct-dif.c -5 +4 M ./dft/ct-dit.c -5 +7 M ./dft/ct-ditbuf.c -4 +9 M ./dft/ct-ditf.c -4 +3 M ./dft/ct.c -6 +1 M ./dft/dft.h -1 +3 M ./dft/direct.c -3 +2 M ./dft/indirect.c -3 +2 M ./dft/nop.c -1 +3 M ./dft/rank-geq2.c -3 +7 M ./dft/rank0.c -3 +4 M ./dft/vrank-geq1.c -10 +18 M ./dft/vrank2-transpose.c -3 +3 M ./dft/vrank3-transpose.c -3 +4 M ./genfft/c.ml -2 +2 M ./kernel/Makefile.am -1 +1 M ./kernel/cycle.h -3 +3 R ./kernel/flops.c M ./kernel/ifftw.h -13 +21 M ./kernel/ops.c +58 M ./kernel/plan.c -2 +2 M ./kernel/planner-estimate.c -2 +9 M ./kernel/planner-naive.c -3 +3 M ./kernel/planner-score.c -4 +4 M ./tests/bench.c -1 +3 Tue Jun 11 07:32:20 EDT 2002 athena * [project @ 2002-06-11 11:32:20 by athena] Many changes A ./dft/vrank-geq1.c M ./dft/Makefile.am -1 +1 M ./dft/buffered.c -9 +8 M ./dft/codelets/inplace/Makefile.am -2 +3 M ./dft/codelets/standard/Makefile.am -2 M ./dft/conf.c -2 +2 M ./dft/ct-dit.c -2 +5 M ./dft/ct.c -2 +2 M ./dft/dft.h -6 +4 M ./dft/indirect.c -2 +2 M ./dft/problem.c -1 +13 M ./dft/rank-geq2.c -5 +4 R ./dft/vecloop.c M ./dft/vrank-geq1.c +243 M ./kernel/ifftw.h -2 +6 M ./kernel/planner-estimate.c -2 +2 M ./kernel/planner-naive.c -2 +2 M ./kernel/planner-score.c -4 +17 M ./kernel/planner.c -4 +6 M ./kernel/print.c -6 +27 M ./kernel/tensor.c -1 +15 M ./kernel/timer.c -2 +2 M ./libbench/bench-main.c -3 +6 M ./tests/bench.c -15 +15 Mon Jun 10 21:35:29 EDT 2002 athena * [project @ 2002-06-11 01:35:29 by athena] Keep it simple, stupid. M ./dft/ct-ditbuf.c -21 +4 Mon Jun 10 19:24:28 EDT 2002 athena * [project @ 2002-06-10 23:24:28 by athena] Fixed when #undef PRECOMPUTE_ARRAY_INDICES M ./kernel/ifftw.h -3 +5 Mon Jun 10 17:58:13 EDT 2002 athena * [project @ 2002-06-10 21:58:13 by athena] Minor changes M ./dft/vrank3-transpose.c -5 +3 M ./kernel/print.c -2 +2 Mon Jun 10 16:30:37 EDT 2002 athena * [project @ 2002-06-10 20:30:37 by athena] Added ct-ditbuf.c, many changes everywhere A ./dft/ct-ditbuf.c M ./CONVENTIONS -2 +3 M ./configure.ac +2 M ./dft/Makefile.am -3 +3 M ./dft/buffered.c -11 +11 M ./dft/ct-dif.c -5 +3 M ./dft/ct-dit.c -6 +4 M ./dft/ct-ditbuf.c +187 M ./dft/ct.c -3 +3 M ./dft/ct.h -1 +3 M ./dft/dft.h -1 +2 M ./dft/direct.c -6 +6 M ./dft/indirect.c -6 +6 M ./dft/kdft-dif.c -2 +2 M ./dft/kdft-difsq.c -2 +2 M ./dft/kdft-dit.c -3 +8 M ./dft/kdft.c -2 +2 M ./dft/nop.c -2 +2 M ./dft/rank-geq2.c -6 +6 M ./dft/rank0.c -2 +2 M ./dft/vecloop.c -10 +8 M ./dft/vrank2-transpose.c -2 +2 M ./dft/vrank3-transpose.c -2 +2 M ./kernel/ifftw.h -3 +25 M ./kernel/plan.c -5 +4 M ./kernel/problem.c -2 +2 M ./kernel/timer.c -3 +3 M ./tests/bench.c -7 +8 Mon Jun 10 10:55:40 EDT 2002 athena * [project @ 2002-06-10 14:55:40 by athena] More name mangling M ./kernel/ifftw.h -2 +2 M ./kernel/planner.c -2 +2 M ./tests/bench.c -58 +47 Mon Jun 10 10:08:27 EDT 2002 athena * [project @ 2002-06-10 14:08:27 by athena] Fixed build system for single/double precision M ./Makefile.am -4 +3 M ./acinclude.m4 -13 M ./configure.ac +4 M ./tests/Makefile.am -1 +1 Mon Jun 10 09:04:21 EDT 2002 athena * [project @ 2002-06-10 13:04:21 by athena] Massive renaming to support both single and double precision. (Must recompile everything twice). A ./kernel/dfftw3.h A ./kernel/fftw3.h A ./kernel/sfftw3.h M ./CONVENTIONS -3 +9 M ./configure.ac -1 +1 M ./dft/buffered.c -96 +102 M ./dft/codelet.h -16 +16 M ./dft/codelets/inplace/Makefile.am -1 +1 M ./dft/codelets/standard/Makefile.am -1 +1 M ./dft/conf.c -12 +12 M ./dft/ct-dif.c -31 +33 M ./dft/ct-dit.c -29 +31 M ./dft/ct-ditf.c -27 +29 M ./dft/ct.c -40 +41 M ./dft/ct.h -10 +12 M ./dft/dft.h -25 +27 M ./dft/direct.c -26 +26 M ./dft/indirect.c -50 +63 M ./dft/kdft-dif.c -3 +3 M ./dft/kdft-difsq.c -3 +3 M ./dft/kdft-dit.c -3 +3 M ./dft/kdft.c -3 +3 M ./dft/nop.c -11 +11 M ./dft/plan.c -3 +3 M ./dft/problem.c -39 +40 M ./dft/rank-geq2.c -57 +59 M ./dft/rank0.c -54 +62 M ./dft/solve.c -2 +2 M ./dft/vecloop.c -49 +49 M ./dft/vrank2-transpose.c -27 +28 M ./dft/vrank3-transpose.c -44 +47 M ./genfft/gen_notw.ml -3 +3 M ./genfft/gen_twiddle.ml -4 +4 M ./genfft/gen_twidsq.ml -4 +4 M ./genfft/genutil.ml -2 +2 M ./kernel/Makefile.am -1 +2 M ./kernel/alloc.c -76 +76 M ./kernel/assert.c -2 +2 M ./kernel/awake.c -2 +2 M ./kernel/cycle.h -11 +11 M ./kernel/dfftw3.h +24 R ./kernel/fftw.h M ./kernel/fftw3.h +39 M ./kernel/flops.c -4 +6 M ./kernel/ifftw.h -90 +89 M ./kernel/minmax.c -5 +5 M ./kernel/plan.c -6 +6 M ./kernel/planner-estimate.c -6 +6 M ./kernel/planner-naive.c -7 +7 M ./kernel/planner-score.c -11 +11 M ./kernel/planner.c -73 +71 M ./kernel/print.c -12 +14 M ./kernel/problem.c -5 +5 M ./kernel/sfftw3.h +24 M ./kernel/solver.c -5 +5 M ./kernel/solvtab.c -4 +4 M ./kernel/square.c -2 +2 M ./kernel/stride.c -6 +6 M ./kernel/tensor.c -96 +96 M ./kernel/timer.c -39 +40 M ./kernel/twiddle.c -33 +38 M ./kernel/version.c -3 +4 M ./support/Makefile.codelets -3 +3 M ./tests/bench.c -39 +53 Mon Jun 10 06:49:55 EDT 2002 athena * [project @ 2002-06-10 10:49:55 by athena] Preliminary crude support for vector transforms in benchmark library. M ./libbench/allocate.c -7 +8 M ./libbench/bench-user.h -2 +5 M ./libbench/mflops.c -2 +2 M ./libbench/problem.c -2 +31 M ./tests/bench.c -1 +2 Sun Jun 9 21:11:51 EDT 2002 athena * [project @ 2002-06-10 01:11:51 by athena] Wrong cast M ./kernel/tensor.c -3 +3 Sun Jun 9 16:48:54 EDT 2002 athena * [project @ 2002-06-09 20:48:54 by athena] Added things to do. M ./TODO +1 Sun Jun 9 16:07:12 EDT 2002 athena * [project @ 2002-06-09 20:07:12 by athena] twlen0: make static M ./kernel/twiddle.c -2 +2 Sun Jun 9 15:34:54 EDT 2002 athena * [project @ 2002-06-09 19:34:54 by athena] Nothing M ./dft/buffered.c -3 +3 Sun Jun 9 15:30:13 EDT 2002 athena * [project @ 2002-06-09 19:30:13 by athena] Forgot break in switch statement. M ./kernel/print.c -1 +2 Sun Jun 9 15:27:24 EDT 2002 athena * [project @ 2002-06-09 19:27:24 by athena] Fix for c++ compatibility M ./kernel/print.c -4 +6 Sun Jun 9 15:16:43 EDT 2002 athena * [project @ 2002-06-09 19:16:43 by athena] Added printer, changed everything A ./kernel/print.c M ./TODO -4 M ./dft/buffered.c -26 +22 M ./dft/ct.c -10 +5 M ./dft/direct.c -8 +3 M ./dft/indirect.c -20 +16 M ./dft/nop.c -4 +6 M ./dft/rank-geq2.c -7 +3 M ./dft/rank0.c -6 +3 M ./dft/vecloop.c -5 +3 M ./dft/vrank2-transpose.c -3 +3 M ./dft/vrank3-transpose.c -3 +3 M ./kernel/Makefile.am -2 +2 M ./kernel/ifftw.h -4 +14 M ./kernel/planner.c -3 +10 M ./kernel/print.c +122 M ./tests/bench.c -2 +22 Sun Jun 9 11:37:07 EDT 2002 athena * [project @ 2002-06-09 15:37:07 by athena] Removed redundant nop solver M ./dft/buffered.c -8 +3 M ./dft/nop.c -2 +12 M ./dft/rank0.c -25 +3 M ./tests/bench.c -1 +1 Sun Jun 9 11:06:31 EDT 2002 athena * [project @ 2002-06-09 15:06:31 by athena] More things to do M ./TODO +10 Sun Jun 9 11:01:41 EDT 2002 athena * [project @ 2002-06-09 15:01:41 by athena] Introduced idea of rank -infinity and associated NOP plans A ./TODO A ./dft/nop.c M ./TODO +4 M ./dft/Makefile.am -4 +4 M ./dft/buffered.c -88 +66 M ./dft/conf.c -1 +2 M ./dft/dft.h -1 +2 M ./dft/direct.c -2 +3 M ./dft/indirect.c -1 +2 M ./dft/nop.c +84 M ./dft/problem.c -2 +5 M ./dft/rank0.c -3 +12 M ./dft/vecloop.c -1 +2 M ./kernel/ifftw.h -1 +11 M ./kernel/tensor.c -11 +39 Sun Jun 9 08:36:27 EDT 2002 athena * [project @ 2002-06-09 12:36:27 by athena] Fixed comment M ./dft/buffered.c -2 +2 Sun Jun 9 08:20:13 EDT 2002 athena * [project @ 2002-06-09 12:20:13 by athena] Removed useless assertions. M ./kernel/tensor.c -4 +3 Sun Jun 9 08:19:26 EDT 2002 athena * [project @ 2002-06-09 12:19:26 by athena] Don't malloc(0). M ./kernel/tensor.c -3 +7 Sun Jun 9 08:08:13 EDT 2002 athena * [project @ 2002-06-09 12:08:13 by athena] Fixed signed/unsigned puns M ./dft/buffered.c -11 +11 Sun Jun 9 07:52:22 EDT 2002 athena * [project @ 2002-06-09 11:52:22 by athena] Added buffered.c A ./dft/buffered.c M ./dft/Makefile.am -1 +1 M ./dft/buffered.c +374 M ./dft/conf.c -1 +2 M ./dft/ct-dif.c -2 +2 M ./dft/ct-dit.c -2 +2 M ./dft/ct-ditf.c -2 +2 M ./dft/dft.h -1 +2 M ./dft/direct.c -2 +2 M ./dft/indirect.c -3 +3 M ./dft/rank-geq2.c -2 +2 M ./dft/rank0.c -6 +6 M ./dft/vecloop.c -2 +2 M ./dft/vrank2-transpose.c -2 +2 M ./dft/vrank3-transpose.c -2 +2 M ./libbench/bench-main.c -2 +7 M ./libbench/bench-user.h -1 +3 M ./tests/bench.c -1 +1 Sat Jun 8 16:57:54 EDT 2002 athena * [project @ 2002-06-08 20:57:54 by athena] Fixed printout M ./dft/ct.c -6 +5 Sat Jun 8 16:42:52 EDT 2002 athena * [project @ 2002-06-08 20:42:52 by athena] Fixed comment M ./dft/vrank3-transpose.c -2 +2 Sat Jun 8 16:40:58 EDT 2002 athena * [project @ 2002-06-08 20:40:58 by athena] Added vrank3-transpose, renamed vrank0-transpose -> vrank2-transpose A ./dft/vrank2-transpose.c A ./dft/vrank3-transpose.c M ./dft/Makefile.am -2 +2 M ./dft/conf.c -2 +3 M ./dft/dft.h -1 +3 R ./dft/vrank0-transpose.c M ./dft/vrank2-transpose.c +132 M ./dft/vrank3-transpose.c +173 M ./tests/bench.c -1 +1 Sat Jun 8 15:51:46 EDT 2002 athena * [project @ 2002-06-08 19:51:46 by athena] Added vrank0-transpose A ./dft/rank-geq2.c A ./dft/vrank0-transpose.c M ./bootstrap.sh -1 +1 M ./dft/Makefile.am -1 +2 M ./dft/conf.c -1 +2 M ./dft/direct.c -4 +4 M ./dft/rank-geq2.c +257 M ./dft/rank0.c -2 +2 R ./dft/rank_geq2.c M ./dft/vrank0-transpose.c +132 M ./tests/bench.c -1 +2 Sat Jun 8 15:11:09 EDT 2002 athena * [project @ 2002-06-08 19:11:09 by athena] Added planner-score.c A ./kernel/planner-score.c M ./dft/Makefile.am -2 +2 M ./dft/ct-dif.c -2 +2 M ./dft/ct-dit.c -2 +2 M ./dft/ct-ditf.c -2 +2 M ./dft/direct.c -2 +2 M ./dft/indirect.c -2 +2 M ./dft/rank0.c -2 +2 M ./dft/rank_geq2.c -2 +2 M ./dft/vecloop.c -2 +2 M ./kernel/Makefile.am -3 +3 M ./kernel/ifftw.h -4 +5 M ./kernel/planner-score.c +70 M ./kernel/planner.c -12 +13 M ./tests/bench.c -1 +4 Sat Jun 8 11:10:44 EDT 2002 athena * [project @ 2002-06-08 15:10:44 by athena] Added indirect.c A ./dft/indirect.c M ./dft/Makefile.am -1 +1 M ./dft/conf.c -1 +2 M ./dft/ct-dif.c -2 +2 M ./dft/ct-dit.c -2 +2 M ./dft/ct-ditf.c -4 +6 M ./dft/dft.h -1 +3 M ./dft/indirect.c +227 M ./dft/rank_geq2.c -5 +5 M ./dft/vecloop.c -2 +1 Sat Jun 8 09:34:58 EDT 2002 athena * [project @ 2002-06-08 13:34:58 by athena] dif, ditf solvers A ./mkdist.sh A ./dft/ct-dif.c A ./dft/ct-ditf.c A ./dft/kdft-dif.c A ./dft/kdft-difsq.c M ./Makefile.am -4 +6 M ./dft/Makefile.am -5 +3 M ./dft/codelet.h -2 +3 M ./dft/codelets/Makefile.am -5 M ./dft/codelets/inplace/Makefile.am -2 +4 M ./dft/conf.c -2 +3 M ./dft/ct-dif.c +120 M ./dft/ct-dit.c -2 +1 M ./dft/ct-ditf.c +106 M ./dft/dft.h -1 +2 M ./dft/direct.c -2 +2 M ./dft/kdft-dif.c +29 M ./dft/kdft-difsq.c +29 M ./mkdist.sh +6 M ./tests/Makefile.am -2 +2 Fri Jun 7 18:07:53 EDT 2002 athena * [project @ 2002-06-07 22:07:53 by athena] Implemented rank_geq2. Revised build system A ./RESEARCH-MODE A ./dft/rank_geq2.c M ./Makefile.am -3 +12 M ./RESEARCH-MODE +1 M ./bootstrap.sh -1 +1 M ./configure.ac -1 +10 M ./dft/Makefile.am -1 +3 M ./dft/conf.c -1 +2 M ./dft/dft.h -1 +2 M ./dft/rank_geq2.c +257 M ./dft/vecloop.c -42 +48 M ./kernel/ifftw.h -1 +3 M ./kernel/minmax.c -1 +11 M ./kernel/planner.c -1 +2 M ./support/Makefile.codelets -1 +1 Fri Jun 7 07:12:25 EDT 2002 athena * [project @ 2002-06-07 11:12:25 by athena] Fixed printout M ./kernel/alloc.c -2 +2 Fri Jun 7 07:07:46 EDT 2002 athena * [project @ 2002-06-07 11:07:46 by athena] Added rank0. Revised codelet organization. A ./dft/codelet.h A ./dft/conf.c A ./dft/codelets/ A ./dft/codelets/Makefile.am A ./dft/codelets/inplace/ A ./dft/codelets/inplace/Makefile.am A ./dft/codelets/standard/ A ./dft/codelets/standard/Makefile.am A ./dft/rank0.c M ./Makefile.am -3 +2 M ./bootstrap.sh -5 +1 M ./configure.ac -3 +18 M ./dft/Makefile.am -3 +4 M ./dft/codelet.h +118 M ./dft/codelets/Makefile.am +7 M ./dft/codelets/inplace/Makefile.am -1 +35 M ./dft/codelets/standard/Makefile.am -1 +37 M ./dft/conf.c +30 M ./dft/dft.h -1 +5 M ./dft/rank0.c +263 M ./genfft/gen_notw.ml -3 +3 M ./tests/Makefile.am -4 +5 M ./tests/bench.c -6 +1 Thu Jun 6 18:03:17 EDT 2002 athena * [project @ 2002-06-06 22:03:17 by athena] Added memoization M ./dft/ct.c -2 +2 M ./dft/vecloop.c -2 +2 M ./genfft/trig.ml -5 +5 M ./kernel/ifftw.h -2 +13 M ./kernel/planner-estimate.c -2 +2 M ./kernel/planner-naive.c -2 +2 M ./kernel/planner.c -12 +191 M ./libbench/bench-user.h -1 +2 M ./tests/bench.c -2 +16 Thu Jun 6 08:07:33 EDT 2002 athena * [project @ 2002-06-06 12:07:33 by athena] Added vecloop A ./dft/vecloop.c M ./dft/Makefile.am -1 +1 M ./dft/dft.h -1 +3 M ./dft/direct.c -9 +7 M ./dft/vecloop.c +242 M ./kernel/alloc.c -11 +38 M ./kernel/ifftw.h -2 +4 M ./kernel/planner.c -2 +2 M ./tests/bench.c +5 Wed Jun 5 19:02:56 EDT 2002 athena * [project @ 2002-06-05 23:02:56 by athena] First DIT solver/plan A ./dft/ct-dit.c M ./dft/Makefile.am -2 +2 M ./dft/ct-dit.c +118 M ./dft/ct.c -9 +16 M ./dft/ct.h -3 +14 M ./dft/dft.h -1 +2 M ./dft/direct.c -10 +6 M ./dft/kdft-dit.c -2 +3 M ./kernel/alloc.c -1 +2 M ./kernel/twiddle.c -3 +11 Wed Jun 5 16:03:44 EDT 2002 athena * [project @ 2002-06-05 20:03:44 by athena] More work on ct A ./dft/ct.c A ./dft/ct.h M ./dft/Makefile.am -2 +2 M ./dft/ct.c +166 M ./dft/ct.h +56 M ./kernel/ifftw.h -1 +2 M ./kernel/stride.c -2 +3 M ./kernel/twiddle.c -16 +26 Wed Jun 5 11:28:09 EDT 2002 athena * [project @ 2002-06-05 15:28:09 by athena] Only use cycle counters M ./kernel/ifftw.h -2 +2 M ./kernel/planner-naive.c -2 +2 M ./kernel/timer.c -166 +18 Tue Jun 4 20:22:23 EDT 2002 athena * [project @ 2002-06-05 00:22:23 by athena] Signed/unsigned fixup M ./CONVENTIONS +2 M ./bootstrap.sh -1 +1 M ./kernel/ifftw.h -7 +6 M ./kernel/twiddle.c -28 +16 Tue Jun 4 20:03:56 EDT 2002 athena * [project @ 2002-06-05 00:03:56 by athena] New file twiddle.c A ./kernel/twiddle.c M ./kernel/Makefile.am -1 +1 M ./kernel/ifftw.h -1 +24 M ./kernel/twiddle.c +137 Tue Jun 4 17:49:39 EDT 2002 athena * [project @ 2002-06-04 21:49:39 by athena] Made tensor ranks and vector lengths unsigned. Hopefully fixed all places where it matters. A ./dft/kdft-dit.c M ./configure.ac +1 M ./dft/Makefile.am -1 +2 M ./dft/direct.c -3 +4 M ./dft/kdft-dit.c +28 M ./dft/problem.c -5 +7 M ./genfft/gen_notw.ml -4 +4 M ./genfft/gen_twiddle.ml -4 +4 M ./genfft/gen_twidsq.ml -4 +4 M ./kernel/Makefile.am -2 +2 R ./kernel/codelet.h M ./kernel/flops.c -2 +2 M ./kernel/ifftw.h -17 +20 M ./kernel/tensor.c -35 +38 M ./kernel/timer.c -3 +2 M ./tests/bench.c +1 Tue Jun 4 16:28:58 EDT 2002 athena * [project @ 2002-06-04 20:28:58 by athena] System is in working state now (but very incomplete) A ./dft/kdft.c A ./kernel/planner-estimate.c A ./kernel/solvtab.c A ./tests/ A ./tests/Makefile.am A ./tests/bench.c M ./Makefile.am -1 +1 M ./configure.ac +2 M ./dft/Makefile.am -2 +2 M ./dft/dft.h -1 +10 M ./dft/direct.c -3 +2 M ./dft/kdft.c +29 M ./kernel/Makefile.am -3 +4 M ./kernel/codelet.h -1 +3 M ./kernel/fftw.h -1 +5 M ./kernel/ifftw.h -2 +21 M ./kernel/planner-estimate.c +55 M ./kernel/planner-naive.c -9 +9 M ./kernel/solvtab.c +30 M ./libbench/Makefile.am -2 +2 M ./support/Makefile.codelets +23 M ./tests/Makefile.am -1 +7 M ./tests/bench.c +92 Mon Jun 3 18:10:12 EDT 2002 athena * [project @ 2002-06-03 22:10:12 by athena] Started implementing planners A ./kernel/planner-naive.c A ./kernel/planner.c M ./CONVENTIONS +2 M ./kernel/Makefile.am -1 +1 M ./kernel/ifftw.h -6 +55 M ./kernel/planner-naive.c +58 M ./kernel/planner.c +113 Mon Jun 3 11:44:18 EDT 2002 athena * [project @ 2002-06-03 15:44:18 by athena] Imported libbench from the new benchfft. We will use libbench for benchmarking and testing. A ./libbench/ A ./libbench/Makefile.am A ./libbench/accopy-from.c A ./libbench/accopy-to.c A ./libbench/acopy.c A ./libbench/allocate.c A ./libbench/ascale.c A ./libbench/aset.c A ./libbench/bench-main.c A ./libbench/bench-user.h A ./libbench/bench.h A ./libbench/caadd.c A ./libbench/cacopy.c A ./libbench/can-do.c A ./libbench/cascale.c A ./libbench/caset.c A ./libbench/casub.c A ./libbench/ccopy-from.c A ./libbench/ccopy-to.c A ./libbench/copy-c2c-from.c A ./libbench/copy-c2c-to.c A ./libbench/copy-c2h-1d-fftpack.c A ./libbench/copy-c2h-1d-halfcomplex.c A ./libbench/copy-c2h-1d-packed.c A ./libbench/copy-c2h-1d-unpacked-ri.c A ./libbench/copy-c2h-unpacked.c A ./libbench/copy-c2h.c A ./libbench/copy-c2r-packed.c A ./libbench/copy-c2r-unpacked.c A ./libbench/copy-c2r.c A ./libbench/copy-c2ri.c A ./libbench/copy-h2c-1d-fftpack.c A ./libbench/copy-h2c-1d-halfcomplex.c A ./libbench/copy-h2c-1d-packed.c A ./libbench/copy-h2c-1d-unpacked-ri.c A ./libbench/copy-h2c-unpacked.c A ./libbench/copy-h2c.c A ./libbench/copy-r2c-packed.c A ./libbench/copy-r2c-unpacked.c A ./libbench/copy-r2c.c A ./libbench/copy-ri2c.c A ./libbench/deallocate.c A ./libbench/getopt-utils.c A ./libbench/getopt.c A ./libbench/getopt.h A ./libbench/getopt1.c A ./libbench/info.c A ./libbench/log2.c A ./libbench/main.c A ./libbench/mflops.c A ./libbench/ovtpvt.c A ./libbench/pow2.c A ./libbench/prime.c A ./libbench/problem.c A ./libbench/report.c A ./libbench/speed.c A ./libbench/timer.c A ./libbench/unnormalize.c A ./libbench/util.c A ./libbench/verify.c A ./libbench/zero.c M ./Makefile.am -1 +1 M ./configure.ac -1 +4 M ./libbench/Makefile.am +18 M ./libbench/accopy-from.c +10 M ./libbench/accopy-to.c +10 M ./libbench/acopy.c +11 M ./libbench/allocate.c +40 M ./libbench/ascale.c +13 M ./libbench/aset.c +10 M ./libbench/bench-main.c +218 M ./libbench/bench-user.h +247 M ./libbench/bench.h +66 M ./libbench/caadd.c +15 M ./libbench/cacopy.c +11 M ./libbench/can-do.c +33 M ./libbench/cascale.c +15 M ./libbench/caset.c +10 M ./libbench/casub.c +15 M ./libbench/ccopy-from.c +21 M ./libbench/ccopy-to.c +20 M ./libbench/copy-c2c-from.c +7 M ./libbench/copy-c2c-to.c +7 M ./libbench/copy-c2h-1d-fftpack.c +29 M ./libbench/copy-c2h-1d-halfcomplex.c +29 M ./libbench/copy-c2h-1d-packed.c +40 M ./libbench/copy-c2h-1d-unpacked-ri.c +25 M ./libbench/copy-c2h-unpacked.c +32 M ./libbench/copy-c2h.c +10 M ./libbench/copy-c2r-packed.c +11 M ./libbench/copy-c2r-unpacked.c +24 M ./libbench/copy-c2r.c +6 M ./libbench/copy-c2ri.c +15 M ./libbench/copy-h2c-1d-fftpack.c +31 M ./libbench/copy-h2c-1d-halfcomplex.c +34 M ./libbench/copy-h2c-1d-packed.c +44 M ./libbench/copy-h2c-1d-unpacked-ri.c +35 M ./libbench/copy-h2c-unpacked.c +51 M ./libbench/copy-h2c.c +10 M ./libbench/copy-r2c-packed.c +13 M ./libbench/copy-r2c-unpacked.c +26 M ./libbench/copy-r2c.c +6 M ./libbench/copy-ri2c.c +15 M ./libbench/deallocate.c +14 M ./libbench/getopt-utils.c +104 M ./libbench/getopt.c +1062 M ./libbench/getopt.h +180 M ./libbench/getopt1.c +188 M ./libbench/info.c +57 M ./libbench/log2.c +13 M ./libbench/main.c +40 M ./libbench/mflops.c +22 M ./libbench/ovtpvt.c +13 M ./libbench/pow2.c +6 M ./libbench/prime.c +49 M ./libbench/problem.c +126 M ./libbench/report.c +159 M ./libbench/speed.c +73 M ./libbench/timer.c +218 M ./libbench/unnormalize.c +13 M ./libbench/util.c +189 M ./libbench/verify.c +408 M ./libbench/zero.c +37 Mon Jun 3 09:18:46 EDT 2002 athena * [project @ 2002-06-03 13:18:46 by athena] Removed useless rand.c M ./kernel/Makefile.am -2 +2 M ./kernel/ifftw.h -4 +1 R ./kernel/rand.c M ./kernel/timer.c -2 +2 Mon Jun 3 08:09:05 EDT 2002 athena * [project @ 2002-06-03 12:09:05 by athena] Added timer A ./kernel/cycle.h A ./kernel/timer.c M ./CONVENTIONS +1 M ./dft/problem.c -2 +2 M ./kernel/Makefile.am -1 +1 M ./kernel/alloc.c -2 +2 M ./kernel/cycle.h +204 M ./kernel/ifftw.h -2 +10 M ./kernel/plan.c -2 +4 M ./kernel/timer.c +293 Sun Jun 2 21:03:51 EDT 2002 athena * [project @ 2002-06-03 01:03:51 by athena] Split codelets into standard and inplace M ./configure.ac -2 +4 Sun Jun 2 19:49:03 EDT 2002 athena * [project @ 2002-06-02 23:49:03 by athena] Many many changes A ./CONVENTIONS A ./dft/direct.c A ./dft/plan.c A ./dft/solve.c A ./kernel/awake.c A ./kernel/square.c M ./CONVENTIONS +49 M ./Makefile.am -4 +4 M ./dft/Makefile.am -1 +1 M ./dft/dft.h -3 +22 M ./dft/direct.c +171 M ./dft/plan.c +33 M ./dft/problem.c -10 +10 M ./dft/solve.c +31 M ./kernel/Makefile.am -1 +1 M ./kernel/awake.c +30 M ./kernel/ifftw.h -13 +33 M ./kernel/square.c +28 Sun Jun 2 15:00:11 EDT 2002 athena * [project @ 2002-06-02 19:00:11 by athena] Fixed anachronism M ./kernel/codelet.h -4 +4 Sat Aug 12 21:43:16 EDT 2006 Unknown tagger tagged bar Sun Jun 2 14:42:32 EDT 2002 athena * [project @ 2002-06-02 18:42:32 by athena] Initial revision A ./AUTHORS A ./COPYRIGHT A ./ChangeLog A ./Makefile.am A ./NEWS A ./README A ./acinclude.m4 A ./bootstrap.sh A ./configure.ac A ./dft/ A ./dft/Makefile.am A ./dft/dft.h A ./dft/problem.c A ./genfft/ A ./genfft/Makefile.am A ./genfft/algsimp.ml A ./genfft/algsimp.mli A ./genfft/annotate.ml A ./genfft/annotate.mli A ./genfft/assoctable.ml A ./genfft/assoctable.mli A ./genfft/c.ml A ./genfft/c.mli A ./genfft/complex.ml A ./genfft/complex.mli A ./genfft/conv.ml A ./genfft/conv.mli A ./genfft/dag.ml A ./genfft/dag.mli A ./genfft/expr.ml A ./genfft/expr.mli A ./genfft/fft.ml A ./genfft/fft.mli A ./genfft/gen_athnotw.ml A ./genfft/gen_athtw.ml A ./genfft/gen_conv.ml A ./genfft/gen_notw.ml A ./genfft/gen_trig.ml A ./genfft/gen_twiddle.ml A ./genfft/gen_twidsq.ml A ./genfft/genutil.ml A ./genfft/littlesimp.ml A ./genfft/littlesimp.mli A ./genfft/magic.ml A ./genfft/monads.ml A ./genfft/number.ml A ./genfft/number.mli A ./genfft/oracle.ml A ./genfft/oracle.mli A ./genfft/schedule.ml A ./genfft/schedule.mli A ./genfft/to_alist.ml A ./genfft/to_alist.mli A ./genfft/trig.ml A ./genfft/trig.mli A ./genfft/twiddle.ml A ./genfft/twiddle.mli A ./genfft/unique.ml A ./genfft/unique.mli A ./genfft/util.ml A ./genfft/util.mli A ./genfft/variable.ml A ./genfft/variable.mli A ./kernel/ A ./kernel/Makefile.am A ./kernel/alloc.c A ./kernel/assert.c A ./kernel/codelet.h A ./kernel/fftw.h A ./kernel/flops.c A ./kernel/ifftw.h A ./kernel/minmax.c A ./kernel/plan.c A ./kernel/problem.c A ./kernel/rand.c A ./kernel/solver.c A ./kernel/stride.c A ./kernel/tensor.c A ./kernel/version.c A ./support/ A ./support/Makefile.am A ./support/Makefile.codelets A ./support/codelet_prelude M ./AUTHORS +4 M ./COPYRIGHT +19 M ./Makefile.am +21 M ./acinclude.m4 +356 M ./bootstrap.sh +26 M ./configure.ac +92 M ./dft/Makefile.am +6 M ./dft/dft.h +30 M ./dft/problem.c +121 M ./genfft/Makefile.am +136 M ./genfft/algsimp.ml +517 M ./genfft/algsimp.mli +24 M ./genfft/annotate.ml +180 M ./genfft/annotate.mli +37 M ./genfft/assoctable.ml +66 M ./genfft/assoctable.mli +30 M ./genfft/c.ml +445 M ./genfft/c.mli +65 M ./genfft/complex.ml +198 M ./genfft/complex.mli +61 M ./genfft/conv.ml +130 M ./genfft/conv.mli +23 M ./genfft/dag.ml +110 M ./genfft/dag.mli +44 M ./genfft/expr.ml +93 M ./genfft/expr.mli +38 M ./genfft/fft.ml +227 M ./genfft/fft.mli +24 M ./genfft/gen_athnotw.ml +80 M ./genfft/gen_athtw.ml +108 M ./genfft/gen_conv.ml +89 M ./genfft/gen_notw.ml +126 M ./genfft/gen_trig.ml +159 M ./genfft/gen_twiddle.ml +144 M ./genfft/gen_twidsq.ml +172 M ./genfft/genutil.ml +256 M ./genfft/littlesimp.ml +72 M ./genfft/littlesimp.mli +26 M ./genfft/magic.ml +121 M ./genfft/monads.ml +76 M ./genfft/number.ml +153 M ./genfft/number.mli +49 M ./genfft/oracle.ml +138 M ./genfft/oracle.mli +25 M ./genfft/schedule.ml +186 M ./genfft/schedule.mli +30 M ./genfft/to_alist.ml +258 M ./genfft/to_alist.mli +24 M ./genfft/trig.ml +158 M ./genfft/trig.mli +35 M ./genfft/twiddle.ml +165 M ./genfft/twiddle.mli +29 M ./genfft/unique.ml +39 M ./genfft/unique.mli +25 M ./genfft/util.ml +177 M ./genfft/util.mli +50 M ./genfft/variable.ml +99 M ./genfft/variable.mli +35 M ./kernel/Makefile.am +5 M ./kernel/alloc.c +217 M ./kernel/assert.c +32 M ./kernel/codelet.h +126 M ./kernel/fftw.h +29 M ./kernel/flops.c +41 M ./kernel/ifftw.h +215 M ./kernel/minmax.c +33 M ./kernel/plan.c +50 M ./kernel/problem.c +47 M ./kernel/rand.c +36 M ./kernel/solver.c +43 M ./kernel/stride.c +41 M ./kernel/tensor.c +318 M ./kernel/version.c +26 M ./support/Makefile.am -1 +1 M ./support/Makefile.codelets +99 M ./support/codelet_prelude +8