1 2016-09-21 Janne Blomqvist <jb@gcc.gnu.org>
3 * intrinsics/random.c (getosrandom): Use rand_s() on
4 MinGW-w64. Fix bounds overflow in fallback code.
6 2016-09-19 Andre Vehreschild <vehre@gcc.gnu.org>
8 * caf/libcaf.h: Add caf_reference_type.
9 * caf/mpi.c: Adapted signature of caf_register().
10 * caf/single.c (struct caf_single_token): Added to keep the pointer
11 to the memory registered and array descriptor.
12 (caf_internal_error): Added convenience interface.
13 (_gfortran_caf_register): Adapted to work with caf_single_token and
14 return memory in the array descriptor.
15 (_gfortran_caf_deregister): Same.
16 (assign_char1_from_char4): Fixed style.
17 (convert_type): Fixed incorrect conversion.
18 (_gfortran_caf_get): Adapted to work with caf_single_token.
19 (_gfortran_caf_send): Same.
20 (_gfortran_caf_sendget): Same.
21 (copy_data): Added to stop repeating it in all _by_ref functions.
22 (get_for_ref): Recursive getting of coarray data using a chain of
24 (_gfortran_caf_get_by_ref): Driver for computing the memory needed for
25 the get and checking properties of the operation.
26 (send_by_ref): Same as get_for_ref but for sending data.
27 (_gfortran_caf_send_by_ref): Same like caf_get_by_ref but for sending.
28 (_gfortran_caf_sendget_by_ref): Uses get_by_ref and send_by_ref to
29 implement sendget for reference chains.
30 (_gfortran_caf_atomic_define): Adapted to work with caf_single_token.
31 (_gfortran_caf_atomic_ref): Likewise.
32 (_gfortran_caf_atomic_cas): Likewise.
33 (_gfortran_caf_atomic_op): Likewise.
34 (_gfortran_caf_event_post): Likewise.
35 (_gfortran_caf_event_wait): Likewise.
36 (_gfortran_caf_event_query): Likewise.
37 (_gfortran_caf_lock): Likewise.
38 (_gfortran_caf_unlock): Likewise.
40 2016-09-09 Steven G. Kargl <kargl@gcc.gnu.org>
43 * ieee/ieee_arithmetic.F90 (IEEE_VALUE_4,IEEE_VALUE_8,IEEE_VALULE_10,
44 IEEE_VALUE_16): Use correct keyword.
46 2016-09-06 Jerry DeLisle <jvdelisle@gcc.gnu.org>
49 * io/write_float.def (build_float_string): Recognize when the
50 result will not fit in the user provided, star fill, and exit
53 2016-08-31 Jerry DeLisle <jvdelisle@gcc.gnu.org>
56 * io/write.c (kind_from_size): New function to calculate required buffer
57 size based on kind type. (select_buffer, select_string): Use new
58 function. (write_float_0, write_real, write_real_g0, write_complex):
59 Adjust calls to pass parameters needed by new function.
61 2016-08-31 Jerry DeLisle <jvdelisle@gcc.gnu.org>
62 Paul Thomas <pault@gcc.gnu.org>
65 * gfortran.map : Flag _st_set_nml_dtio_var and
66 _gfortran_transfer_derived.
67 * io/format.c (format_lex): Detect DTIO formatting.
68 (parse_format_list): Parse the DTIO format.
69 (next_format): Include FMT_DT.
70 * io/format.h : Likewise. Add structure 'udf' to structure
71 'fnode' to carry the IOTYPE string and the 'vlist'.
72 * io/io.h : Add prototypes for the two types of DTIO subroutine
73 and a typedef for gfc_class. Also, add to 'namelist_type'
74 fields for the pointer to the DTIO procedure and the vtable.
75 Add fields to struct st_parameter_dt for pointers to the two
76 types of DTIO subroutine. Add to gfc_unit DTIO specific fields.
77 (internal_proto): Add prototype for 'read_user_defined' and
79 * io/list_read.c (check_buffers): Use the 'current_unit' field.
80 (unget_char): Likewise.
81 (eat_spaces): Likewise.
82 (list_formatted_read_scalar): For case BT_CLASS, call the DTIO
84 (nml_get_obj_data): Likewise when DTIO procedure is present,.
85 * io/transfer.c : Export prototypes for 'transfer_derived' and
86 'transfer_derived_write'.
87 (unformatted_read): For case BT_CLASS, call the DTIO procedure.
88 (unformatted_write): Likewise.
89 (formatted_transfer_scalar_read): Likewise.
90 (formatted_transfer_scalar_write: Likewise.
91 (transfer_derived): New function.
92 (data_transfer_init): Set last_char if no child_dtio.
93 (finalize_transfer): Return if child_dtio set.
94 (st_write_done): Add condition for child_dtio not set.
95 Add extra arguments for st_set_nml_var prototype.
96 (set_nml_var): New function that contains the contents of the
97 old version of st_set_nml_var. Also sets the 'dtio_sub' and
98 'vtable' fields of the 'nml' structure.
99 (st_set_nml_var): Now just calls set_nml_var with 'dtio_sub'
101 (st_set_nml_dtio_var): New function that calls set_nml_var.
102 * io/unit.c (get_external_unit): If the found unit child_dtio
103 is non zero, don't do any mutex locking/unlocking. Just
105 * io/unix.c (tempfile_open): Revert to C style comment.
106 * io/write.c (list_formatted_write_scalar): Do the DTIO call.
107 (nml_write_obj): Add BT_CLASS and do the DTIO call.
109 2016-08-29 Nathan Sidwell <nathan@acm.org>
111 * configure.ac (nvptx-*): Hardwire newlib.
112 * configure: Rebuilt.
114 2016-08-19 Janne Blomqvist <jb@gcc.gnu.org>
116 * intrinsics/random.c (xor_keys): New array with "secret" keys.
117 (scramble_seed): XOR given seed with xor_keys array rather than
119 (unscramble_seed): Remove function.
120 (random_seed_i4): Use new scramble_seed.
121 (random_seed_i8): Likewise.
123 2016-08-19 Janne Blomqvist <jb@gcc.gnu.org>
125 * intrinsics/random.c (master_init): New variable.
126 (init_rand_state): Move below getosrandom (), maybe initialize
128 (random_seed_i4): If called with no arguments, set master_init to
129 false, and reinitialize. If called with PUT=, set master_init to
131 (random_seed_i8): Likewise.
133 2016-08-11 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
135 * intrinsics/random.c: Include <stdlib.h>.
137 2016-08-11 Janne Blomqvist <jb@gcc.gnu.org>
139 * intrinsics/random.c: Replace KISS with xorshift1024* using
141 * runtime/main.c (init): Don't call random_seed_i4.
143 2016-07-22 Andre Vehreschild <vehre@gcc.gnu.org>
145 * caf/libcaf.h: Add parameter stat to caf_get() and
146 caf_send()'s function prototypes.
147 * caf/single.c (_gfortran_caf_get): Implement reporting
148 error using stat instead of abort().
149 (_gfortran_caf_send): Same.
150 (_gfortran_caf_sendget): Use NULL for stat when calling
153 2016-06-23 Jerry DeLisle <jvdelisle@gcc.gnu.org>
156 * io/write.c: Cleaned up whitespace.
157 (write_d, write_e, write_f, write_es, write_en): Use new helper
158 function write_float_0.
159 (write_float_0): New helper function.
160 (get_precision, select_buffer, select_string, write_float_string): New
161 helper functions used in remaining float writing functions.
162 Helper function write_float_string now contains code for writing
163 to kind=4 character internal units.
164 (write_real): Modified to establish working buffers at this level
165 and to use new helper functions.
166 (write_real_g0): Likewise modified.
167 (write_complex): Likewise modified. Gets both float strings before
168 output so that final lengths can be determined which allows right
169 justifying the complex number with no intervening spaces.
170 * io/write_float.def (build_float_string): Renamed from previosly
171 output_float, modified to use buffers passed in from higher functions,
172 builds a null terminated string of the floating point value. Character
173 kind=4 code eliminated.
174 (write_infnan): Likewise modified to use incoming buffers and eliminate
176 (OUTPUT_FLOAT_FMT_G): Deleted, functionality moved into FORMAT_FLOAT.
177 (FORMAT_FLOAT): Renamed macro from WRITE_FLOAT. Use build_float_string.
178 (get_float_string): Renamed from write_float, uses FORMAT_FLOAT macro.
179 Buffer allocation removed, now at higher level.
181 2016-05-23 Jerry DeLisle <jvdelisle@gcc.gnu.org>
184 * io/list_read (eat_spaces): Eat '\r' as part of spaces.
186 2016-04-19 Jerry DeLisle <jvdelisle@gcc.gnu.org>
189 * io/list_read (check_buffers): Add '\r' to check for end of line.
191 2016-03-30 Jerry DeLisle <jvdelisle@gcc.gnu.org>
192 Dominique d'Humieres <dominiq@lps.ens.fr>
195 * io/write_float.def: Fix PF format for negative values of the scale
198 2016-03-28 Alessandro Fanfarillo <fanfarillo.gcc@gmail.com>
200 * caf/libcaf.h: caf_stop_numeric and caf_stop_str prototype.
201 * caf/single.c: _gfortran_caf_stop_numeric and
202 _gfortran_caf_stop_str implementation.
204 2016-02-23 Jerry DeLisle <jvdelisle@gcc.gnu.org>
207 * io/list_read.c (read_real): If digit is missing from exponent issue
208 an error. (parse_real): Likewise and adjusted error message to clarify
209 it is part of a complex number.
210 (nml_read_obj): Bump item count and add comment that this is used to
211 identify which item in a namelist read has a problem.
213 2016-02-17 Jerry DeLisle <jvdelisle@gcc.gnu.org>
216 * io/list_read.c (push_char4): Fix the pointer usage for xrealloc.
218 2016-02-15 Jerry DeLisle <jvdelisle@gcc.gnu.org>
221 * io/list_read.c: Entire file trailing spaces removed.
222 (CASE_SEPARATORS): Remove '!'.
223 (is_separator): Add namelist mode as condition with '!'.
224 (push_char): Remove un-needed memset. (push_char4): Likewise and remove
225 'new' pointer. (eat_separator): Remove un-needed use of notify_std.
226 (read_logical): If '!' bang encountered when not in namelist mode got
227 bad_logical to give an error. (read_integer): Likewise reject '!'.
228 (read_character): Remove condition testing c = '!' which is now inside
229 the is_separator macro. (parse_real): Reject '!' unless in namelist mode.
230 (read_complex): Reject '!' unless in namelist mode. (read_real): Likewise
233 2016-02-12 Jerry DeLisle <jvdelisle@gcc.gnu.org>
236 * io/list_read.c (read_character): Remove code related to DELIM_NONE.
238 2016-01-23 John David Anglin <danglin@gcc.gnu.org>
241 * runtime/backtrace.c: Include gthr.h.
242 (show_backtrace): Use __gthread_active_p() to determine whether threads
243 are active. Return if lbstate is NULL.
245 2016-01-15 Jakub Jelinek <jakub@redhat.com>
247 * intrinsics/execute_command_line.c (set_cmdstat): Use "%s", msg
248 instead of msg to avoid -Wformat-security warning.
250 2016-01-04 Jakub Jelinek <jakub@redhat.com>
252 Update copyright years.
254 Copyright (C) 2016 Free Software Foundation, Inc.
256 Copying and distribution of this file, with or without modification,
257 are permitted in any medium without royalty provided the copyright
258 notice and this notice are preserved.