libgfortran/ChangeLog

   1 2020-05-28  Harald Anlauf  <anlauf@gmx.de>
   2
   3         PR libfortran/95104
   4         * io/unit.c (unlock_unit): Guard by check for NULL pointer.
   5
   6 2020-05-26  Harald Anlauf  <anlauf@gmx.de>
   7             Steven G. Kargl   <kargl@gcc.gnu.org>
   8
   9         PR libfortran/95104
  10         * io/transfer.c (st_wait_async): Do not dereference NULL pointer.
  11
  12 2020-05-26  Harald Anlauf  <anlauf@gmx.de>
  13
  14         PR fortran/95195
  15         * io/transfer.c (finalize_transfer): Generate runtime error for
  16         namelist input/output to unformatted file.
  17
  18 2020-05-23  Thomas Koenig  <tkoenig@gcc.gnu.org>
  19
  20         PR libfortran/95191
  21         * io/async.c (async_wait_id): Generate error if ID is higher
  22         than the highest current ID.
  23         * runtime/error.c (translate_error): Handle LIBERROR_BAD_WAIT_ID.
  24
  25 2020-05-21  H.J. Lu  <hongjiu.lu@intel.com>
  26
  27         * m4/matmul.m4: Don't include <config/i386/cpuinfo.h>.  Use
  28         __builtin_cpu_is/__builtin_cpu_supports
  29         * generated/matmul_c10.c: Regenerated.
  30         * generated/matmul_c16.c: Likewise.
  31         * generated/matmul_c4.c: Likewise.
  32         * generated/matmul_c8.c: Likewise.
  33         * generated/matmul_i1.c: Likewise.
  34         * generated/matmul_i16.c: Likewise.
  35         * generated/matmul_i2.c: Likewise.
  36         * generated/matmul_i4.c: Likewise.
  37         * generated/matmul_i8.c: Likewise.
  38         * generated/matmul_r10.c: Likewise.
  39         * generated/matmul_r16.c: Likewise.
  40         * generated/matmul_r4.c: Likewise.
  41         * generated/matmul_r8.c: Likewise.
  42
  43 2020-05-15  H.J. Lu  <hongjiu.lu@intel.com>
  44
  45         PR bootstrap/95147
  46         * configure: Regenerated.
  47
  48 2020-05-14  Thomas Koenig  <tkoenig@gcc.gnu.org>
  49
  50         PR libfortran/95119
  51         * io/close.c (close_status): Add CLOSE_INVALID.
  52         (st_close): Return early on invalid STATUS parameter.
  53
  54 2020-05-14  H.J. Lu  <hongjiu.lu@intel.com>
  55
  56         * configure: Regenerated.
  57
  58 2020-05-06  Uroš Bizjak  <ubizjak@gmail.com>
  59
  60         * config/fpu-387.h (__math_force_eval): Remove.
  61         (__math_force_eval_div): New define.
  62         (local_feraiseexcept): Use __math_force_eval_div to use
  63         generic division to generate INVALID, DIVZERO and INEXACT
  64         exceptions.
  65         (struct fenv): Define named struct instead of typedef.
  66
  67 2020-05-01  Uroš Bizjak  <ubizjak@gmail.com>
  68
  69         * config/fpu-387.h (__math_force_eval): New define.
  70         (local_feraiseexcept): Use __math_force_eval to evaluate
  71         generic division to generate INVALID and DIVZERO exceptions.
  72
  73 2020-04-22  Fritz Reese  <foreese@gcc.gnu.org>
  74
  75         * intrinsics/trigd.c, intrinsics/trigd_lib.inc, intrinsics/trigd.inc:
  76         Guard against unavailable math functions.
  77         Use suffixes from kinds.h based on the REAL kind.
  78
  79 2020-04-22  Jakub Jelinek  <jakub@redhat.com>
  80
  81         PR libfortran/94694
  82         PR libfortran/94586
  83         * configure.ac: Add math func checks for fmaf, fma and fmal.  Add
  84         HAVE_INLINE_BUILTIN_COPYSIGN check.
  85         * c99_protos.h (copysign, fmaf, fma, fmal): Provide fallback
  86         prototypes.
  87         (HAVE_COPYSIGN, HAVE_FMAF, HAVE_FMA, HAVE_FMAL): Define if not
  88         defined and fallback version is provided.
  89         * intrinsics/c99_functions.c (copysign, fmaf, fma, fmal): Provide
  90         fallback implementations if possible
  91         * configure: Regenerated.
  92         * config.h.in: Regenerated.
  93
  94 2020-04-19  Uroš Bizjak  <ubizjak@gmail.com>
  95
  96         * config/fpu-387.h (local_feraiseexcept) [__SSE_MATH__]:
  97         Remove unneeded assignments to volatile memory.
  98
  99 2020-04-01  Fritz Reese  <foreese@gcc.gnu.org>
 100             Steven G. Kargl  <kargl@gcc.gnu.org>
 101
 102         PR fortran/93871
 103         * Makefile.am, Makefile.in: New make rule for intrinsics/trigd.c.
 104         * gfortran.map: New routines for {sind, cosd, tand}X{r4, r8, r10, r16}.
 105         * intrinsics/trigd.c, intrinsics/trigd_lib.inc, intrinsics/trigd.inc:
 106         New files. Defines native degree-valued trig functions.
 107
 108 2020-02-18  Thomas Koenig  <tkoenig@gcc.gnu.org>
 109
 110         PR fortran/93599
 111         * io/async.c (destroy_adv_cond): Do not destroy lock.
 112         (async_io): Make sure au->lock is locked for finishing of thread.
 113         Do not lock/unlock around signalling emptysignal. Unlock au->lock
 114         before return.
 115         (init_adv_cond): Do not initialize lock.
 116         (enqueue_transfer): Unlock after signal.
 117         (enqueue_done_id): Likewise.
 118         (enqueue_done): Likewise.
 119         (enqueue_close): Likewise.
 120         (enqueue_data_transfer): Likewise.
 121         (async_wait_id): Do not lock/unlock around signalling au->work.
 122         (async_wait): Unlock after signal.
 123         * io/async.h (SIGNAL): Add comment about needed au->lock.
 124         Remove locking/unlocking of advcond->lock.
 125         (WAIT_SIGNAL_MUTEX): Add comment. Remove locking/unlocking of
 126         advcond->lock.  Unlock mutex only at the end.  Loop on
 127         __ghread_cond_wait returning zero.
 128         (REVOKE_SIGNAL): Add comment. Remove locking/unlocking of
 129         advcond->lock.
 130         (struct adv_cond): Remove mutex from struct.
 131
 132 2020-02-12  Sandra Loosemore  <sandra@codesourcery.com>
 133
 134         PR libstdc++/79193
 135         PR libstdc++/88999
 136
 137         * configure: Regenerated.
 138
 139 2020-01-24  Maciej W. Rozycki  <macro@wdc.com>
 140
 141         * configure.ac: Handle `--with-toolexeclibdir='.
 142         * Makefile.in: Regenerate.
 143         * aclocal.m4: Regenerate.
 144         * configure: Regenerate.
 145
 146 2020-01-17  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 147
 148         PR libfortran/93234
 149         * io/unit.c (set_internal_unit): Set round and sign flags
 150         correctly.
 151
 152 2020-01-17  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 153
 154         PR libfortran/90374
 155         * io/format.c (parse_format_list): Zero width not allowed with
 156         FMT_D.
 157         * io/write_float.def (build_float_string): Include range of
 158         higher exponent values that require wider width.
 159
 160 2020-01-01  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 161
 162         PR libfortran/90374
 163         * io/format.c (parse_format_list): Implement the E0 exponent
 164         width to provide smallest possible width for exponent fields.
 165         Refactor code for correct parsing and better readability of the
 166         code.
 167         * io/io.h (write_real_w0): Change interface to pass in pointer
 168         to fnode.
 169         * io/transfer.c: Update all calls to write_real_w0 to use the
 170         new interface.
 171         * io/write.c ((write_real_w0): Use the new interface with fnode
 172         to access both the decimal precision and exponent widths used in
 173         build_float_string.
 174         * io/write_float.def (build_float_string): Use the passed in
 175         exponent width to calculate the used width in the case of E0.
 176
 177 2020-01-01  Jakub Jelinek  <jakub@redhat.com>
 178
 179         Update copyright years.
 180 \f
 181 Copyright (C) 2020 Free Software Foundation, Inc.
 182
 183 Copying and distribution of this file, with or without modification,
 184 are permitted in any medium without royalty provided the copyright
 185 notice and this notice are preserved.