ChangeLog-gcc-4.1.txt 15 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251
  1. Changelog of the TIGCC-local changes to GCC (4.1 branch):
  2. 2009-06-08 Lionel Debroux <lionel_debroux@yahoo.fr>
  3. * version.c (VERSUFFIX): Bump to "4.1.2-tigcc-4".
  4. * varasm.c: fix a 64-bit compatibility problem in the AMS float support,
  5. which triggered harmless warnings in the generated assembly code.
  6. 2007-10-08 Kevin Kofler <Kevin@tigcc.ticalc.org>
  7. PR preprocessor/23479
  8. * version.c (VERSUFFIX): Bump to "4.1.2-tigcc-3".
  9. Sync binary integer patch with upstream GCC 4.3 patch
  10. (2007-06-05 Joerg Wunsch <j.gnu@uriah.heep.sax.de>):
  11. * ../libcpp/expr.c (cpp_classify_number): Don't allow 0b for floats.
  12. Change pedwarn to say "GCC extension" rather than "TIGCC extension".
  13. (append_digit): Fix overflow check for binary integers.
  14. 2007-04-17 Kevin Kofler <Kevin@tigcc.ticalc.org>
  15. PR c/25509
  16. * version.c (VERSUFFIX): Bump to "4.1.2-tigcc-2".
  17. * gimplify.c (gimplify_expr): Don't strip casts to void for function calls.
  18. (Patch originally by Richard Guenther.)
  19. 2007-02-19 Kevin Kofler <Kevin@tigcc.ticalc.org>
  20. * Import GCC 4.1.2 release.
  21. * version.c (VERSUFFIX): Bump to "4.1.2-tigcc-1".
  22. * Revert because of a conflict with the generalized lvalues patch:
  23. 2006-08-05 Volker Reichelt <reichelt@igpm.rwth-aachen.de>
  24. PR c/27721
  25. * c-typeck.c (build_modify_expr): Test earlier for non-lvalues.
  26. * c-typeck.c (build_modify_expr): Don't call build_binary_op on a function (PR c/27721).
  27. * c-decl.c (push_parm_decl): Fix missing argument for format string in error message.
  28. 2007-02-09 Kevin Kofler <Kevin@tigcc.ticalc.org>
  29. * gcc.c: #define access to access_check for _WIN32.
  30. (access_check): Translate X_OK to R_OK on _WIN32.
  31. 2006-10-17 Kevin Kofler <Kevin@tigcc.ticalc.org>
  32. * version.c (VERSUFFIX): Bump to "4.1.2-pre10".
  33. 2006-09-16 Kevin Kofler <Kevin@tigcc.ticalc.org>
  34. * version.c (VERSUFFIX): Bump to "4.1.2-pre9".
  35. * config/m68k/m68k.md (m68k_init_cumulative_args): Mark STATIC_CHAIN_REGNUM as used for
  36. nested functions needing a static chain so register parameters don't conflict.
  37. 2006-08-01 Kevin Kofler <Kevin@tigcc.ticalc.org>
  38. * version.c (VERSUFFIX): Bump to "4.1.2-pre8".
  39. * config/m68k/m68k.md (pushqi1): Don't use the optimized pushqi1_reg on SUBREGs with even
  40. offsets. Pass the full SUBREG expression to pushqi1_reg.
  41. (pushqi1_reg): Handle SUBREGs here. Compute new SUBREG_BYTE when handling SUBREGs instead
  42. of hard-coding it as 0.
  43. 2006-07-29 Kevin Kofler <Kevin@tigcc.ticalc.org>
  44. * Import GCC 4.1 20060728 snapshot.
  45. * version.c (VERSUFFIX): Bump to "4.1.2-pre7".
  46. 2006-04-30 Kevin Kofler <Kevin@tigcc.ticalc.org>
  47. * Import GCC 4.1 20060428 snapshot.
  48. * version.c (VERSUFFIX): Bump to "4.1.1-pre6".
  49. * gcc/c.opt (-Wpointer-sign): Don't change the initializer (now set to -1 upstream).
  50. * c-opts.c (c_common_handle_option): Don't enable -Wpointer-sign for -Wall.
  51. 2006-01-05 Kevin Kofler <Kevin@tigcc.ticalc.org>
  52. * version.c (VERSUFFIX): Bump to "4.1.0-pre5".
  53. * config/m68k/m68k.md (movsi): Don't allow writes of a label directly to memory under -mpcrel.
  54. (anonymous pattern for pcrel movsi): Split into case for address register and case for push.
  55. Only accept stack pushes, not arbitrary predecrements.
  56. 2005-12-24 Kevin Kofler <Kevin@tigcc.ticalc.org>
  57. * version.c (VERSUFFIX): Bump to "4.1.0-pre4".
  58. * toplev.c (process_options): Don't warn about -fomit-frame-pointer with unwind tables and
  59. don't force -fno-omit-frame-pointer.
  60. 2005-12-22 Kevin Kofler <Kevin@tigcc.ticalc.org>
  61. * version.c (VERSUFFIX): Bump to "4.1.0-pre3".
  62. * c-parser.c (c_parser_cast_expression): Allow all postfix operators, not just ++ and --.
  63. 2005-12-22 Kevin Kofler <Kevin@tigcc.ticalc.org>
  64. * version.c (VERSUFFIX): Bump to "4.1.0-pre2".
  65. * c-parser.c (c_parser_cast_expression): Rename to c_parser_cast_expression_1, don't handle
  66. Objective C here.
  67. (c_parser_cast_expression): New function. Handle Objective C and postfix operators.
  68. (c_parser_unary_expression): Add in_cast argument. Pass it to c_parser_postfix_expression.
  69. (c_parser_cast_expression_1, c_parser_sizeof_expression, c_parser_alignof_expression): Update
  70. calls to c_parser_unary_expression.
  71. (c_parser_postfix_expression): Add in_cast argument. Pass it to
  72. c_parser_postfix_expression_after_primary.
  73. (c_parser_postfix_expression_after_primary): Add in_cast argument. Only accept postincrement/
  74. postdecrement if it is false.
  75. (c_parser_cast_expression, c_parser_postfix_expression_after_paren_type): Update calls to
  76. postfix_expression_after_primary.
  77. 2005-12-22 Kevin Kofler <Kevin@tigcc.ticalc.org>
  78. * version.c (VERSUFFIX): Add "4.1.0-pre1" tag.
  79. 2005-12-18 - 2005-12-21 Kevin Kofler <Kevin@tigcc.ticalc.org>
  80. * Import GCC 4.1 20051216 snapshot.
  81. * Merge changes from GCC 4.0.2 (see ChangeLog-gcc-4.0.txt).
  82. * Drop obsolete backports:
  83. DWARF 2 patches/backports from Fedora GCC:
  84. - gcc4-dwarf2-pr20268.patch
  85. 2005-03-30 James E. Wilson <wilson@specifixinc.com>
  86. PR debug/20268
  87. * dwarf2out.c (add_high_low_attributes): New function, extracted from
  88. gen_lexical_block_die.
  89. (gen_lexical_block_die, gen_inlined_subroutine_die): Call it.
  90. - gcc4-dwarf2-inline-details.patch
  91. 2005-07-25 James E Wilson <wilson@specifixinc.com>
  92. * dwarf2out.c (add_call_src_coords_attributes): New.
  93. (gen_inlined_subroutine_die): Call it.
  94. (maybe_emit_file, init_file_table): Add comments.
  95. (prune_unused_types_walk_attribs): Pass DW_AT_call_file through
  96. maybe_emit_file.
  97. * tree-inline.c (remap_block): Copy BLOCK_SOURCE_LOCATION.
  98. (expand_call_inline): Set BLOCK_SOURCE_LOCATION.
  99. * tree.h (BLOCK_SOURCE_LOCATION): New.
  100. (struct tree_block): New field locus.
  101. - gcc4-dwarf2-usefbreg.patch
  102. 2005-08-10 Jakub Jelinek <jakub@redhat.com>
  103. * dwarf2out.c (concat_loc_descriptor): Add can_use_fbreg argument,
  104. pass it down to loc_descriptor.
  105. (loc_descriptor): Pass can_use_fbreg to concat_loc_descriptor.
  106. (containing_function_has_frame_base): Move earlier in the file.
  107. (loc_descriptor_from_tree_1): Use containing_function_has_frame_base
  108. instead of always assuming fbreg can't be used.
  109. - gcc4-dwarf2-prefer-1elt-vartracking.patch
  110. 2005-08-10 Jakub Jelinek <jakub@redhat.com>
  111. * dwarf2out.c (add_location_or_const_value_attribute): Prefer locations
  112. gathered by var-tracking in single entry loc_list over loc_descriptor_from_tree.
  113. - gcc4-dwarf2-frame_base.patch
  114. 2005-08-22 Richard Henderson <rth@redhat.com>
  115. * function.c (ARG_POINTER_CFA_OFFSET): Move ...
  116. * defaults.h (ARG_POINTER_CFA_OFFSET): ... here.
  117. (INCOMING_FRAME_SP_OFFSET): Moved from ...
  118. * dwarf2out.c (INCOMING_FRAME_SP_OFFSET): ... here.
  119. (struct cfa_loc): Change reg to unsigned int,
  120. rearrange for better packing.
  121. (lookup_cfa_1): Remove inline marker.
  122. (cfa_equal_p): Split out of ...
  123. (def_cfa_1): ... here. Use INVALID_REGNUM.
  124. (build_cfa_loc): Handle !cfa->indirect.
  125. (frame_pointer_cfa_offset): New.
  126. (dbx_reg_number): Assert register elimination performed; do
  127. leaf register remapping.
  128. (reg_loc_descriptor): Avoid calling dbx_reg_number when unused.
  129. (eliminate_reg_to_offset): New.
  130. (based_loc_descr): Remove can_use_fbreg argument. Use fbreg only
  131. for verifiably local stack frame addresses; re-base to CFA.
  132. (mem_loc_descriptor): Remove can_use_fbreg argument.
  133. (concat_loc_descriptor, loc_descriptor): Likewise.
  134. (containing_function_has_frame_base): Remove.
  135. (rtl_for_decl_location): Don't do register elimination or
  136. leaf register remapping here.
  137. (secname_for_decl): Split out from ..
  138. (add_location_or_const_value_attribute): ... here.
  139. (convert_cfa_to_loc_list): New.
  140. (compute_frame_pointer_to_cfa_displacement): New.
  141. (gen_subprogram_die): Use them.
  142. * tree.h (frame_base_decl): Remove.
  143. * var-tracking.c (frame_base_decl, frame_stack_adjust): Remove.
  144. (prologue_stack_adjust): Remove.
  145. (vt_stack_adjustments): Use INCOMING_FRAME_SP_OFFSET.
  146. (adjust_stack_reference): Re-base memories to arg_pointer_rtx.
  147. (set_frame_base_location): Remove.
  148. (compute_bb_dataflow, emit_notes_in_bb): Don't call it.
  149. (dump_attrs_list, dump_dataflow_set): Use string concatenation.
  150. (vt_add_function_parameters): Don't eliminate_regs.
  151. (vt_initialize): Don't create frame_base_decl.
  152. 2005-09-21 Andreas Schwab <schwab@suse.de>
  153. PR rtl-optimization/23043
  154. * postreload-gcse.c (eliminate_partially_redundant_load): Fix typo
  155. when allocating a struct unoccr.
  156. 2005-09-29 Jakub Jelinek <jakub@redhat.com>
  157. * c-decl.c (c_write_global_declarations_1): If any
  158. wrapup_global_declaration_2 call returned true, restart the loop.
  159. * Drop no longer needed reversion of (part of larger patch):
  160. 2004-01-17 Joseph S. Myers <jsm@polyomino.org.uk>
  161. c-typeck.c (build_conditional_expr): Call non_lvalue instead of pedantic_non_lvalue.
  162. (The check has moved to fold-const.c and does call pedantic_non_lvalue now.)
  163. * Drop no longer needed patches:
  164. * collect2.c: Applied patch to allow collect2 to build on MinGW by Lev Serebryakov
  165. (2002-10-25).
  166. * expr.c (emit_push_insn): Don't take a modulus by 0. Set offset to constant 0 instead.
  167. 2004-07-25 Kevin Kofler <Kevin@tigcc.ticalc.org>
  168. * varasm.c (enum section_category): Move declaration up.
  169. (categorize_decl_for_section): Move prototype up.
  170. 2005-05-05 Kevin Kofler <Kevin@tigcc.ticalc.org>
  171. * dwarf2out.c (dwarf2out_start_source_file): Use the correct file number.
  172. * version.c: Don't hand-maintain version numbers, use (TIGCC) suffix instead.
  173. * Moved AMS-floating-point-related patch from config/m68k/m68k.c to config/m68k/predicates.md:
  174. (const_uint32_operand): Don't handle CONST_DOUBLE.
  175. Check for "INTVAL (op) >= 0" (possibly wrong?).
  176. * config/m68k/m68k.c (extended_address_operand): Move to config/m68k/predicates.md.
  177. * config/m68k/m68k.h: Don't add extended_address_operand to PREDICATE_CODES anymore.
  178. Don't define MASK_* and TARGET_* (except MASK_REG_RELATIVE,
  179. TARGET_REG_RELATIVE) anymore.
  180. (TARGET_DIRECTFLOAT): Define to always 0 (not supported yet).
  181. (TARGET_SWITCHES, TARGET_OPTIONS): Move options to...
  182. * config/m68k/m68k.opt: ... here. Remove "-mnoalign-int", "-mnopcrel", "-mnotios", "-mnoregparm",
  183. "-mnomerge-sections", "-mnomerge-to-data", "-mnorodata-to-text" (use
  184. "-mno-*" with the dash instead). Enable negative for "-mcoff-abslines".
  185. * Revert broken part of PR debug/24070 fix:
  186. 2005-10-06 Richard Henderson <rth@redhat.com>
  187. PR debug/24070
  188. * dwarf2out.c (lookup_filename): Return the result of maybe_emit_file.
  189. * builtins.c: Also disable expand_builtin_int_roundingfn, fold_builtin_int_roundingfn and
  190. stack-smashing protection builtins.
  191. (fold_builtins_1): Disable stack-smashing protection builtins.
  192. * builtins.def: Likewise.
  193. * Drop changes to c-parse.in (replaced by c-parser.c).
  194. * c-parser.c (c_parser_parameter_declaration): Handle an optional asmspec before the attributes
  195. to support register parameters. Pass it to
  196. build_c_parm (or NULL if there is none).
  197. (c_parser_compound_statement_nostart): Don't error about labels at end of compound statements,
  198. only pedwarn if pedantic.
  199. * config/m68k/m68k.c (output_move_double, floating_exact_log2): Use gcc_unreachable instead of
  200. abort.
  201. * config/smapbcd.h (REAL_VALUE_TO_TARGET_SINGLE, REAL_VALUE_TO_TARGET_DOUBLE): Likewise.
  202. (UNSIGNED_INF, POSITIVE_INF, NEGATIVE_INF, NAN, REAL_VALUE_LDEXP, REAL_VALUE_UNSIGNED_FIX,
  203. REAL_VALUE_RNDZINT, REAL_VALUE_TO_TARGET_LONG_DOUBLE, REAL_VALUE_TO_TARGET_SMAP_BCD):
  204. Add ull suffix to big numbers to get rid of warnings.
  205. * c-decl.c (c_insert_default_attributes): Use ISO C90 prototype.
  206. * c-typeck.c (build_modify_expr): Update parameters for default_function_array_conversion call.
  207. * expr.c (emit_block_move_via_libcall, clear_storage_via_libcall): Fix use of uninitialized
  208. variable.
  209. * real.c (real_value_dtof): Add ull suffix to big number to get rid of warning.
  210. (real_value_htof): Delete unused variable "exp2".
  211. (real_from_integer): Delete unused variable "i".
  212. * simplify-rtx.c (simplify_const_binary_operation): Declare tmp1 pointer const.
  213. (simplify_immed_subreg): Declare tmp pointer const.
  214. * tree-object-size.c (alloc_object_size): Don't handle BUILT_IN_CALLOC.
  215. (pass_through_call): Don't handle BUILT_IN_*_CHK.
  216. * optabs.c (lowpart_subreg_maybe_copy, expand_copysign_absneg): Disable.
  217. * config/m68k/m68k.c (m68k_handle_stkparm_attribute, m68k_handle_regparm_attribute): Update
  218. calls to warning().
  219. (m68k_save_reg, m68k_output_function_prologue): Fix pointer signedness mismatch.