Merge branch 'FFTW:master' into master

Author: sh-zheng

CMakeLists.txt

@@ -279,7 +279,7 @@ if (HAVE_AVX2)
   list (APPEND SOURCEFILES ${fftw_dft_simd_avx2_SOURCE} ${fftw_rdft_simd_avx2_SOURCE})
 endif ()
 
-set (FFTW_VERSION 3.3.9)
+set (FFTW_VERSION 3.3.10)
 
 set (PREC_SUFFIX)
 if (ENABLE_FLOAT)
@@ -314,19 +314,29 @@ if (MSVC AND NOT (CMAKE_C_COMPILER_ID STREQUAL "Intel"))
   target_compile_definitions (${fftw3_lib} PRIVATE /bigobj)
 endif ()
 if (HAVE_SSE)
-  target_compile_options (${fftw3_lib} PRIVATE ${SSE_FLAG})
+  set_source_files_properties (${fftw_dft_simd_sse2_SOURCE}
+                               ${fftw_rdft_simd_sse2_SOURCE}
+                               PROPERTIES COMPILE_FLAGS "${SSE_FLAG}")
 endif ()
 if (HAVE_SSE2)
-  target_compile_options (${fftw3_lib} PRIVATE ${SSE2_FLAG})
+  set_source_files_properties (${fftw_dft_simd_sse2_SOURCE}
+                               ${fftw_rdft_simd_sse2_SOURCE}
+                               PROPERTIES COMPILE_FLAGS "${SSE2_FLAG}")
 endif ()
 if (HAVE_AVX)
-  target_compile_options (${fftw3_lib} PRIVATE ${AVX_FLAG})
+  set_source_files_properties (${fftw_dft_simd_avx_SOURCE}
+                               ${fftw_rdft_simd_avx_SOURCE}
+                               PROPERTIES COMPILE_FLAGS "${AVX_FLAG}")
 endif ()
 if (HAVE_AVX2)
-  target_compile_options (${fftw3_lib} PRIVATE ${AVX2_FLAG})
+  set_source_files_properties (${fftw_dft_simd_avx2_SOURCE}
+                               ${fftw_rdft_simd_avx2_SOURCE}
+                               PROPERTIES COMPILE_FLAGS "${AVX2_FLAG}")
 endif ()
 if (HAVE_FMA)
-  target_compile_options (${fftw3_lib} PRIVATE ${FMA_FLAG})
+  set_source_files_properties (${fftw_dft_simd_avx2_SOURCE}
+                               ${fftw_rdft_simd_avx2_SOURCE}
+                               PROPERTIES COMPILE_FLAGS "${FMA_FLAG}")
 endif ()
 if (HAVE_LIBM)
   target_link_libraries (${fftw3_lib} m)

configure.ac

@@ -610,7 +610,7 @@ fi
 
 dnl add gcc warnings, in debug/maintainer mode only
 if test "$enable_debug" = yes || test "$USE_MAINTAINER_MODE" = yes; then
-if test "$ac_test_CFLAGS" != "set"; then
+if test "x$ac_test_CFLAGS" != "xset" -a "x$ac_test_CFLAGS" != "xy"; then
     if test $ac_cv_prog_gcc = yes; then
         CFLAGS="$CFLAGS -Wall -W -Wcast-qual -Wpointer-arith -Wcast-align -pedantic -Wno-long-long -Wshadow -Wbad-function-cast -Wwrite-strings -Wstrict-prototypes -Wredundant-decls -Wnested-externs" # -Wundef -Wconversion -Wmissing-prototypes -Wmissing-declarations
     fi

doc/reference.texi

@@ -103,7 +103,7 @@ Include the @emph{same} @code{<fftw3.h>} header file.
 Replace all lowercase instances of @samp{fftw_} with @samp{fftwf_} or
 @samp{fftwl_} for single or long-double precision, respectively.
 (@code{fftw_complex} becomes @code{fftwf_complex}, @code{fftw_execute}
-becomes @code{fftwf_execute}, etcetera.)
+becomes @code{fftwf_execute}, etc.)
 
 @item
 Uppercase names, i.e. names beginning with @samp{FFTW_}, remain the
@@ -176,7 +176,7 @@ fftw_complex *fftw_alloc_complex(size_t n);
 @findex fftw_alloc_complex
 
 The equivalent functions in other precisions allocate arrays of @code{n}
-elements in that precision.  e.g. @code{fftwf_alloc_real(n)} is
+elements in that precision, e.g. @code{fftwf_alloc_real(n)} is
 equivalent to @code{(float *) fftwf_malloc(sizeof(float) * n)}.
 @cindex precision
 
@@ -234,7 +234,7 @@ accumulating wisdom information again.
 memory leaks, you must still call @code{fftw_destroy_plan} before
 executing @code{fftw_cleanup}.
 
-Occasionally, it may useful to know FFTW's internal ``cost'' metric
+Occasionally, it may be useful to know FFTW's internal ``cost'' metric
 that it uses to compare plans to one another; this cost is
 proportional to an execution time of the plan, in undocumented units,
 if the plan was created with the @code{FFTW_MEASURE} or other
@@ -283,7 +283,7 @@ char *fftw_sprint_plan(const fftw_plan plan);
 @findex fftw_print_plan
 
 This outputs a ``nerd-readable'' representation of the @code{plan} to
-the given file, to @code{stdout}, or two a newly allocated
+the given file, to @code{stdout}, or to a newly allocated
 NUL-terminated string (which the caller is responsible for deallocating
 with @code{free}), respectively.
 
@@ -890,7 +890,7 @@ introduction to these transform kinds, see @ref{More DFTs of Real Data}.
 For dimension of size @code{n}, there is a corresponding ``logical''
 dimension @code{N} that determines the normalization (and the optimal
 factorization); the formula for @code{N} is given for each kind below.
-Also, with each transform kind is listed its corrsponding inverse
+Also, with each transform kind is listed its corresponding inverse
 transform.  FFTW computes unnormalized transforms: a transform followed
 by its inverse will result in the original data multiplied by @code{N}
 (or the product of the @code{N}'s for each dimension, in
@@ -1044,7 +1044,7 @@ row-major subarrays of larger rank-@code{rank} arrays, described by
 and @code{n} should be elementwise less than or equal to
 @{@code{i},@code{o}@}@code{nembed}.  Passing @code{NULL} for an
 @code{nembed} parameter is equivalent to passing @code{n} (i.e. same
-physical and logical dimensions, as in the basic interface.)
+physical and logical dimensions, as in the basic interface).
 
 The @code{stride} parameters indicate that the @code{j}-th element of
 the input or output arrays is located at @code{j*istride} or
@@ -1125,7 +1125,7 @@ Like @code{fftw_plan_many_dft}, these two functions add @code{howmany},
 @code{fftw_plan_dft_r2c} and @code{fftw_plan_dft_c2r} functions, but
 otherwise behave the same as the basic interface.
 
-The interpretation of @code{howmany}, @code{stride}, and @code{dist} are
+The interpretation of @code{howmany}, @code{stride}, and @code{dist} is
 the same as for @code{fftw_plan_many_dft}, above.  Note that the
 @code{stride} and @code{dist} for the real array are in units of
 @code{double}, and for the complex array are in units of
@@ -1162,10 +1162,10 @@ fftw_plan fftw_plan_many_r2r(int rank, const int *n, int howmany,
 @end example
 @findex fftw_plan_many_r2r
 
-Like @code{fftw_plan_many_dft}, this functions adds @code{howmany},
+Like @code{fftw_plan_many_dft}, this function adds @code{howmany},
 @code{nembed}, @code{stride}, and @code{dist} parameters to the
-@code{fftw_plan_r2r} function, but otherwise behave the same as the
-basic interface.  The interpretation of those additional parameters are
+@code{fftw_plan_r2r} function, but otherwise behaves the same as the
+basic interface.  The interpretation of those additional parameters is
 the same as for @code{fftw_plan_many_dft}.  (Of course, the
 @code{stride} and @code{dist} parameters are now in units of
 @code{double}, not @code{fftw_complex}.)
@@ -1282,7 +1282,7 @@ A row-major multidimensional array with dimensions @code{n[rank]}
 (@pxref{Row-major Format}) corresponds to @code{dims[i].n} =
 @code{n[i]} and the recurrence @code{dims[i].is} = @code{n[i+1] *
 dims[i+1].is} (similarly for @code{os}).  The stride of the last
-(@code{i=rank-1}) dimension is the overall stride of the array.
+(@code{i=rank-1}) dimension is the overall stride of the array,
 e.g. to be equivalent to the advanced complex-DFT interface, you would
 have @code{dims[rank-1].is} = @code{istride} and
 @code{dims[rank-1].os} = @code{ostride}.
@@ -1566,7 +1566,7 @@ detailed below, provided that the following conditions are met:
 @itemize @bullet
 
 @item
-The array size, strides, etcetera are the same (since those are set by
+The array size, strides, etc. are the same (since those are set by
 the plan).
 
 @item
@@ -1621,7 +1621,7 @@ if necessary).
 If you are tempted to use the new-array execute interface because you
 want to transform a known bunch of arrays of the same size, you should
 probably go use the advanced interface instead (@pxref{Advanced
-Interface})).
+Interface}).
 
 The new-array execute functions are:
 
@@ -1723,7 +1723,7 @@ convenience, the following three ``wrapper'' routines are provided:
 @code{filename} (which is created or overwritten), returning @code{1}
 on success and @code{0} on failure.  A lower-level function, which
 requires you to open and close the file yourself (e.g. if you want to
-write wisdom to a portion of a larger file) is
+write wisdom to a portion of a larger file), is
 @code{fftw_export_wisdom_to_file}.  This writes the wisdom to the
 current position in @code{output_file}, which should be open with
 write permission; upon exit, the file remains open and is positioned
@@ -1771,7 +1771,7 @@ the following three ``wrapper'' routines are provided:
 @code{fftw_import_wisdom_from_filename} reads wisdom from a file named
 @code{filename}.  A lower-level function, which requires you to open
 and close the file yourself (e.g. if you want to read wisdom from a
-portion of a larger file) is @code{fftw_import_wisdom_from_file}. This
+portion of a larger file), is @code{fftw_import_wisdom_from_file}. This
 reads wisdom from the current position in @code{input_file} (which
 should be open with read permission); upon exit, the file remains
 open, but the position of the read pointer is unspecified.
@@ -1812,7 +1812,7 @@ merely summarize them here, since they come with their own @code{man}
 pages for Unix and GNU systems (with HTML versions on our web site).
 
 The first program is @code{fftw-wisdom} (or @code{fftwf-wisdom} in
-single precision, etcetera), which can be used to create a wisdom file
+single precision, etc.), which can be used to create a wisdom file
 containing plans for any of the transform sizes and types supported by
 FFTW.  It is preferable to create wisdom directly from your executable
 (@pxref{Caveats in Using Wisdom}), but this program is useful for
@@ -2438,7 +2438,7 @@ peculiar array format described in more detail by @ref{Real-data DFT
 Array Format}.
 
 The multi-dimensional c2r transform is simply the unnormalized inverse
-of the r2c transform.  i.e. it is the same as FFTW's complex backward
+of the r2c transform, i.e. it is the same as FFTW's complex backward
 multi-dimensional DFT, operating on a Hermitian input array in the
 peculiar format mentioned above and outputting a real array (since the
 DFT output is purely real).

genfft/annotate.ml

@@ -85,7 +85,7 @@ let reorder l =
 	let b' = List.map (fun (a, _) -> a) c' in
 	a :: (loop b') in
   let l' = List.map (fun x -> x, uniq (find_block_vars x)) l in
-  (* start with smallest block --- does this matter ? *)
+  (* start with smallest block --- does this matter? *)
   match l' with
     [] -> []
   | _ ->  

genfft/c.ml

@@ -249,7 +249,7 @@ and unparse_function = function
 
 
 (*************************************************************
- * traverse a a function and return a list of all expressions,
+ * traverse a function and return a list of all expressions,
  * in the execution order
  **************************************************************)
 let rec fcn_to_expr_list = fun (Fcn (_, _, _, body)) -> ast_to_expr_list body 

genfft/number.ml

@@ -33,7 +33,7 @@
    unity, since the Num package only supplies simple arithmetic.  The
    arbitrary-precision operations in Num look like the normal
    operations except that they have an appended slash (e.g. +/ -/ */
-   // etcetera). *)
+   // etc.). *)
 
 open Num
 

tests/fftw-bench.c

@@ -291,13 +291,5 @@ void cleanup(void)
      FFTW(cleanup)();
 #endif
 
-#    ifdef FFTW_DEBUG_MALLOC
-     {
-	  /* undocumented memory checker */
-	  FFTW_EXTERN void FFTW(malloc_print_minfo)(int v);
-	  FFTW(malloc_print_minfo)(verbose);
-     }
-#    endif
-
      final_cleanup();
 }