3 * config/obj-coff.c (c_dot_file_symbol): Add "app" param.
4 (coff_adjust_symtab): Adjust call.
5 (crawl_symbols): Likewise.
6 * config/obj-coff.h (c_dot_file_symbol): Add "app" param.
7 (obj_app_file): Adjust.
11 * configure.in: Enable bfd_assember for the MAXQ port.
12 * configure: Regenerate.
17 * config/tc-xtensa.c (finish_vinsn): Clear pending instruction if
19 (check_t1_t2_reads_and_writes): Check for both reads and writes to
20 interfaces that are related as determined by xtensa_interface_class_id.
24 * config/tc-mn10300.c (md_relax_table): Fix off by one negative
25 offsets for conditional branches.
29 * config/tc-xtensa.c (MAX_IMMED6): Change value to 65.
33 * config/tc-xtensa.c (update_next_frag_state): Always add a NOP if
34 relaxing at the end of a loop. Don't mark frags as UNREACHABLE or
36 (relax_frag_immed): Update call to update_next_frag_state.
40 * obj.h (struct format_ops <app_file>): Add int param.
41 * read.h (s_app_file_string): Likewise.
42 * read.c (s_app_file_string): Likewise.
43 (s_app_file): Adjust s_app_file_string call.
44 * config/tc-mips.c (s_mips_file): Likewise.
45 * config/obj-coff.h (obj_app_file): Add app param.
46 * config/obj-ecoff.h (obj_app_file): Likewise.
47 * config/obj-multi.h (obj_app_file): Likewise.
48 * config/obj-elf.h (elf_file_symbol): Likewise.
49 * config/obj-elf.c (elf_file_symbol): Only emit one file symbol
50 if called for # preprocessor lines.
55 * symbols.c (resolve_symbol_value): Convert weak symbols only
57 (symbol_equated_reloc_p): Don't equate weaks when relocating
58 only for Windows PECOFF.
62 * config/tc-crx.c (print_insn): Check and set insn_addr.
63 * config/tc-crx.h (md_frag_check): Define.
68 * configure.in: Add support for new target maxq-coff.
69 * configure: Regenerate.
70 * NEWS: Mention new support.
71 * config/tc-maxq.c: New file.
72 * config/tc-maxq.h: New file.
73 * config/obj-coff.h: Add support for maxq-coff.
77 * symbols.c (any_external_name): Define.
78 (resolve_symbol_value): Convert weak symbols.
79 (S_SET_EXTERNAL): Support any_external_name.
80 (S_SET_NAME): Qualify parameter const.
81 (symbol_equated_reloc_p): Equate to weaks when relocating.
82 * symbols.h (S_SET_NAME): Qualfiy parameter const.
83 * tc.h (any_external_name): Declare.
84 * config/obj-coff.c ("coff/pe.h"): Include for BFD
86 (weak_is_altname): Declare and define.
87 (weak_name2altname): Same.
88 (weak_altname2name): Same.
89 (weak_uniquify): Same.
90 (weak_altprefix): Define.
91 (obj_coff_weak): Change .weak syntax and handling.
92 (coff_frob_symbol): Fix PE weak symbol alternates.
93 * config/obj-coff.h (USE_UNIQUE): Define.
94 * config/tc-i386.c (md_apply_fix3): Assume weak symbols
95 are in another segment.
96 (tc_gen_reloc): Remove broken addend hack.
97 doc/as.texinfo: Update.
101 * config/tc-xtensa.c (total_frag_text_expansion): New.
102 (md_estimate_size_before_relax): Use it.
103 (find_address_of_next_align_frag): Likewise.
107 * config/tc-crx.c: Rename argument types.
108 (processing_arg_number): Rename to 'cur_arg_num'.
109 (get_number_of_bits): Rename to 'set_operand_size'.
110 (get_operandtype): Rename to 'parse_operand', totally rewrite.
111 (set_cons_rparams): Rename to 'set_operand', totally rewrite.
112 (set_indexmode_parameters): Remove function, integrate its code into
114 (set_operand_size): Get rid of 'Operand Number' function parameter -
115 use global variable 'cur_arg_num' instead.
116 Use a local 'argument' pointer to reference the current argument.
117 (parse_operand): Likewise.
118 (set_operand): Likewise.
119 (process_label_constant): Likewise.
123 * config/tc-xtensa.c: Remove XTENSA_SECTION_RENAME ifdefs.
124 (add_section_rename): Delete. Inlined into...
125 (build_section_rename): ...here. Use xstrdup instead of strdup.
126 (xtensa_section_rename): Drop "const" from argument and return types.
127 (md_show_usage): Indent to match show_usage().
128 * config/tc-xtensa.h: Remove XTENSA_SECTION_RENAME ifdefs.
129 (tc_canonicalize_section_name): Define.
130 (md_elf_section_rename): Remove unused macro.
131 * doc/as.texinfo (Overview): Document Xtensa --rename-section option.
132 * doc/c-xtensa.texi (Xtensa Options): Likewise.
133 (Frame Directive): Delete.
137 * configure.in: Remove arm-*-oabi and thumb-*-oabi.
138 * config/tc-arm.c (target_oabi): Delete.
139 (md_apply_fix3, elf32_arm_target_format): Remove target_oabi checks.
140 (arm_opts): Remove moabi.
141 * doc/as.texinfo (Overview): Remove documentation of -moabi.
142 * doc/c-arm.texi (ARM Options): Likewise.
143 * configure: Regenerated.
147 * configure.in (crisv32): Recognize. AC_DEFINE_UNQUOTED
148 DEFAULT_CRIS_ARCH. Handle crisv32-*-linux-gnu* like
149 cris-*-linux-gnu* and crisv32-*-* like cris-*-*.
150 * configure: Regenerate.
151 * config/tc-cris.c (enum cris_archs): New.
152 (cris_mach, cris_arch_from_string, s_cris_arch, get_sup_reg)
153 (cris_insn_ver_valid_for_arch): New functions.
154 (DEFAULT_CRIS_ARCH): New macro, default to cris_any_v0_v10.
155 (cris_arch): New variable.
156 (md_pseudo_table): New pseudo .arch.
157 (err_for_dangerous_mul_placement): Initialize according to
159 (STATE_COND_BRANCH): Renamed from STATE_CONDITIONAL_BRANCH.
161 (STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON)
162 (STATE_ABS_BRANCH_V32, STATE_LAPC, BRANCH_BF_V32, BRANCH_BB_V32)
163 (BRANCH_WF_V32, BRANCH_WB_V32): New.
164 (BRANCH_BF, BRANCH_BB, BRANCH_WF, BRANCH_WB): Don't undef after
165 use in md_cris_relax_table.
166 (md_cris_relax_table): Add entries for STATE_COND_BRANCH_V32,
167 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
168 Update and improve head comment.
169 (OPTION_PIC): Define in terms of previous option, OPTION_US.
170 (OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): Similar.
172 (md_longopts): New option --march=...
173 (cris_any_v0_v10_long_jump_size, crisv32_long_jump_size): New
175 (md_long_jump_size): Initialize in terms of DEFAULT_CRIS_ARCH.
176 (HANDLE_RELAXABLE): New macro.
177 (md_estimate_size_before_relax): Use HANDLE_RELAXABLE for common
178 cases. Check for weak symbols and assume not relaxable. Handle
179 STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON,
180 STATE_ABS_BRANCH_V32, STATE_LAPC. Use new variable symbolP, not
182 (md_convert_frag): Handle STATE_COND_BRANCH_V32,
183 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
184 (cris_create_short_jump): Adjust for CRISv32.
185 (cris_relax_frag): Handle new states.
186 (md_create_long_jump): Ditto. Emit error for common_v10_v32.
187 (md_begin): Define symbols "..asm.arch.cris.v32",
188 "..asm.arch.cris.v10", "..asm.arch.cris.common_v10_v32" and
189 "..asm.arch.cris.any_v0_v10". Use cris_insn_ver_valid_for_arch
190 when entering opcode table entry points.
191 (md_assemble): Adjust branch handling for CRISv32. Handle LAPC
192 relaxation. In fix_new_exp call for main insn, pass 1 for pcrel
193 parameter for 8, 16 and 32-bit pc-relative insns and LAPC.
194 (cris_process_instruction): Initialize out_insnp->insn_type to
195 CRIS_INSN_NONE, not CRIS_INSN_NORMAL.
196 <case ']', '[', 'A', 'd', 'Q', 'N', 'n', 'Y', 'U', 'u', 'T'>: New
198 <case 'm'>: Check that modified_char == '.'.
199 <invalid operands>: Consume the rest of the line.
200 When operands don't match, skip over subsequent insns with
201 non-matching version specifier but same mnemonic.
202 <immediate constant, case SIZE_SPEC_REG>: Immediate operands for
203 special registers in CRISv32 are always 32 bit long.
204 <immediate constant, case SIZE_FIELD_SIGNED, SIZE_FIELD_UNSIGNED>:
206 (get_gen_reg): Only recognize "PC" when followed by "+]" for v32
207 and compatible. Recognize "ACR" for v32, unless followed by "+".
208 (get_spec_reg): Consider cris_arch when looking up register.
209 (get_autoinc_prefix_or_indir_op): Don't recognize assignment for
211 (get_3op_or_dip_prefix_op): Check for ']' after seeing '[rN+'.
212 (cris_get_expression): Restore input_line_pointer if failing "early".
213 (get_flags): Consider cris_arch and recognize flags accordingly.
214 (branch_disp): Adjust for CRISv32.
215 (gen_cond_branch_32): Similar. Emit error for common_v10_v32.
216 (cris_number_to_imm): Use as_bad_where, not as_bad. Remove
217 related FIXME. Don't insist on BFD_RELOC_32_PCREL fixup to be
218 resolved. Don't enter zeros in object file for
220 <case BFD_RELOC_CRIS_LAPCQ_OFFSET, BFD_RELOC_CRIS_SIGNED_16>
221 <case BFD_RELOC_CRIS_SIGNED_8>: New case.
222 (md_parse_option): Break out "return 1".
223 <OPTION_ARCH> New case.
224 (tc_gen_reloc): <case BFD_RELOC_CRIS_LAPCQ_OFFSET>
225 <case BFD_RELOC_CRIS_SIGNED_16, BFD_RELOC_CRIS_SIGNED_8>
226 <case BFD_RELOC_CRIS_UNSIGNED_8, BFD_RELOC_CRIS_UNSIGNED_16>
227 <case BFD_RELOC_32_PCREL>: New cases.
228 Addends for non-zero fx_pcrel are too in fx_offset.
229 (md_show_usage): Show --march=<arch>.
230 (md_apply_fix3): Adjust val for BFD_RELOC_CRIS_LAPCQ_OFFSET.
231 (md_pcrel_from): BFD_RELOC_CRIS_LAPCQ_OFFSET is PC-relative too.
232 (s_syntax) <struct syntaxes>: Properly constify member operand.
233 * config/tc-cris.h (TARGET_MACH): Define.
234 (cris_mach): Declare.
235 * doc/as.texinfo (Overview) <CRIS>: Add --march=...
236 * doc/c-cris.texi (CRIS-Symbols): New node for built-in symbols.
237 (CRIS-Opts): Document --march=...
238 (CRIS-Pseudos): Document .arch.
242 * config/tc-i386.c (set_intel_syntax): Allow % in symbol names when
243 intel syntax and no register prefix, allow $ in symbol names when
245 (set_16bit_gcc_code_flag): Replace literal 'l' by LONG_MNEM_SUFFIX.
246 (intel_float_operand): Add fourth return value indicating math control
247 operations. Make classification more precise.
248 (md_assemble): Complain if memory operand of mov[sz]x has no size
250 (parse_insn): Translate word operands to floating point instructions
251 operating on integers as well as control instructions to short ones
252 as expected by AT&T syntax. Translate 'd' suffix to short one only for
253 floating point instructions operating on non-integer operands.
254 (match_template): Remove fldcw special case. Adjust q-suffix handling
255 to permit it on fild/fistp/fisttp in AT&T mode.
256 (process_suffix): Don't guess DefaultSize insns' suffix from
257 stackop_size for certain floating point control instructions. Guess
258 suffix for branch and [ls][gi]dt based on flag_code. Split error
259 messages for Intel and AT&T syntax, and make the condition more strict
260 for the former. Adjust suppressing of generation of operand size
262 (intel parser): Allow the full set of MASM operators. Add FWORD, TBYTE,
263 OWORD, and XMMWORD operand size specifiers (TBYTE replaces XWORD). Add
265 * config/tc-i386.h (BYTE_PTR WORD_PTR DWORD_PTR QWORD_PTR XWORD_PTR
266 SHORT OFFSET_FLAT FLAT NONE_FOUND): Remove unused defines.
270 * symbols.c (colon) [!WORKING_DOT_WORD]: Don't declare
271 md_short_jump_size, md_long_jump_size.
272 * write.c [!WORKING_DOT_WORD]: Ditto.
273 * tc.h [!WORKING_DOT_WORD]: Declare them here. Drop const
275 * config/tc-cris.h (md_short_jump_size, md_long_jump_size): Don't
277 * config/tc-cris.c (md_short_jump_size, md_long_jump_size): Drop
278 const qualifier in these definitions.
279 * config/tc-i370.c, config/tc-m68k.c, config/tc-pdp11.c,
280 config/tc-s390.c, config/tc-tahoe.c, config/tc-vax.c: Ditto.
284 * dwarf2dbg.c (dwarf2_finish): Check for the existence of a file
285 table before deciding to produce a .debug_line section to match up
286 with a user provided .debug_info section.
290 * config/tc-crx.c (getreg_image): Bug fix, a return value was
291 mistakenly omitted from CRX_C_REGTYPE and CRX_CS_REGTYPE cases.
295 * config/tc-crx.c: Remove global variable 'post_inc_mode'.
296 (get_flags): New function.
297 (get_number_of_bits): Edit comments, update numeric values to
299 (process_label_constant): Don't support the colon
300 format (SYMBOL:[s|m|l]).
301 (set_cons_rparams): Support argument type 'arg_rbase'.
302 (get_operandtype): Bug fix in 'rbase' operand type parsing.
303 (handle_LoadStor): Bug fix, first handle post-increment mode.
304 (getreg_image): Remove redundant code, update according to latest
306 (print_constant): Bug fix relate to 3-word instructions.
307 (assemble_insn): Bug fix, when matching instructions, verify also
308 instruction type (not only mnemonic).
309 Add various error checking.
310 (preprocess_reglist): Support HI/LO and user registers.
314 * config/tc-ia64.c (fixup_unw_records): Don't let the "t" value
315 in an epilogue directive go negative.
320 * config/tc-ia64.c (emit_one_bundle): Decrement md.num_slots_in_use
321 after reporting template error during manual bundling. Reported
326 * Makefile.am: Run dep-am.
327 * aclocal.m4: Regenerate with automake 1.9.2.
328 * Makefile.in: Regenerate with automake 1.9.2.
329 * doc/Makefile.in: Likewise.
331 * config/tc-arm.c: Include "dw2gencfi.h".
332 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
334 * config/tc-arm.h (TARGET_USE_CFIPOP, DWARF2_DEFAULT_RETURN_COLUMN)
335 (DWARF2_CIE_DATA_ALIGNMENT, tc_regname_to_dw2regnum)
336 (tc_cfi_frame_initial_instructions): Define.
337 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
342 * config/tc-crx.c (assemble_insn): Check unsigned immediate
344 Update coprocessor id to be unsigned immediate.
348 * config/tc-i386.c (O_secrel): Delete.
349 (tc_pe_dwarf2_emit_offset): New function.
350 * config/tc-i386.h (O_secrel): Define as O_md1.
351 (TC_DWARF2_EMIT_OFFSET): Define.
355 * config/tc-xstormy16.c (xstormy16_cons_fix_new): Accept and
356 ignore @fptr() directives for 4-byte fixups.
360 * Makefile.am: Run "make dep-am"
361 * Makefile.in: Regenerate.
365 * doc/c-xtensa.texi (Xtensa Options, Absolute Literals Directive):
366 Remove comments about placement of literal pools.
367 (Literal Directive): Update description of literal placement.
368 (Literal Prefix Directive): Remove statement that this does not apply
369 to absolute-mode literals. Describe new section naming scheme.
373 * config/tc-xtensa.c (xtensa_frequency_pseudo): Use set_subseg_freq.
374 (is_entry_opcode, is_movi_opcode, is_the_loop_opcode, is_jx_opcode,
375 is_windowed_return_opcode): Delete.
376 (xtensa_frob_label): Use get_subseg_target_freq.
377 (md_assemble): Inline call to is_entry_opcode.
378 (xtensa_handle_align): Inline call to get_frag_is_literal.
379 (relaxation_requirements): Inline call to is_jx_opcode.
380 (emit_single_op): Inline call to is_movi_opcode.
381 (xg_assemble_vliw_tokens): Inline calls to get_frag_is_insn,
382 get_frag_is_no_transform, is_entry_opcode, and
383 set_frag_is_specific_opcode. Use get_subseg_total_freq.
384 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags,
385 xtensa_fix_close_loop_end_frags, relax_frag_immed, convert_frag_immed):
386 Inline calls to get_frag_is_no_transform.
387 (next_instrs_are_b_retw): Inline call to is_windowed_return_opcode.
388 (xtensa_fix_short_loop_frags): Inline calls to is_the_loop_opcode and
389 get_frag_is_no_transform.
390 (convert_frag_immed_finish_loop): Inline calls to get_expression_value
391 and set_frag_is_no_transform.
392 (get_expression_value): Delete.
393 (subseg_map struct): Rename cur_total_freq to total_freq. Rename
394 cur_target_freq to target_freq.
395 (get_subseg_info): Split out code to create a new map entry into ...
396 (add_subseg_info): ... this new function.
397 (get_last_insn_flags): Check if get_subseg_info succeeded.
398 (set_last_insn_flags): Call add_subseg_info if needed.
399 (get_subseg_total_freq, get_subseg_target_freq, set_subseg_freq): New.
400 (xtensa_reorder_segments): Compute last_sec while counting sections.
401 Remove call to get_last_sec.
402 (get_last_sec): Delete.
403 (cache_literal_section): Inline call to retrieve_literal_seg and its
404 callees, seg_present and add_seg_list.
405 (retrieve_literal_seg, seg_present, add_seg_list): Delete.
406 (get_frag_is_insn, get_frag_is_no_transform,
407 set_frag_is_specific_opcode, set_frag_is_no_transform): Delete.
408 * config/tc-xtensa.h (MAX_SLOTS): Reduce from 31 to 15.
412 * config/tc-xtensa.c: Use ISO C90 formatting.
413 * config/tc-xtensa.h: Likewise.
414 * config/xtensa-istack.h: Likewise.
415 * config/xtensa-relax.c: Likewise.
416 * config/xtensa-relax.h: Likewise.
420 * config/tc-arm.c (md_begin): Change EF_ARM_EABI_VER3 to
423 * doc/c-arm.texi: Document that we actually support -meabi=4, not
428 * doc/as.texinfo (VTableEntry, VTableInherit): Add "directive" to index
430 (Acknowledgements): Use "GAS" instead of AS variable.
434 * config/tc-i386.c: Include "elf/x86-64.h".
435 (i386_elf_section_type): New function.
436 * config/tc-i386.h (md_elf_section_type): Define.
437 (i386_elf_section_type): New prototype.
441 * config/m68k-parse.h (enum m68k_register): New control register,
443 * config/tc-m68k.c: Correct control register set for MCF5249.
448 * config/tc-xtensa.c (absolute_literals_supported): New global flag.
449 (UNREACHABLE_MAX_WIDTH): Define.
450 (XTENSA_FETCH_WIDTH): Delete.
451 (cur_vinsn, xtensa_fetch_width, xt_saved_debug_type, past_xtensa_end,
452 prefer_const16, prefer_l32r): New global variables.
453 (LIT4_SECTION_NAME): Define.
454 (lit4_state struct): Add lit4_seg_name and lit4_seg fields.
455 (XTENSA_PROP_*, GET_XTENSA_PROP_*, SET_XTENSA_PROP_*): Define.
456 (frag_flags struct): New.
457 (xtensa_block_info struct): Move from tc-xtensa.h. Add flags field.
458 (subseg_map struct): Add cur_total_freq and cur_target_freq fields.
459 (bitfield, bit_is_set, set_bit, clear_bit): Define.
460 (MAX_FORMATS): Define.
461 (op_placement_info struct, op_placement_table): New.
462 (O_pltrel, O_hi16, O_lo16): Define.
463 (directiveE enum): Rename directive_generics to directive_transform.
464 Delete directive_relax. Add directive_schedule,
465 directive_absolute_literals, and directive_last_directive.
466 (directive_info): Rename "generics" to "transform". Delete "relax".
467 Add "schedule" and "absolute-literals".
468 (directive_state): Adjust entries to match changes in directive_info.
469 (xtensa_relax_statesE, RELAX_IMMED_MAXSTEPS): Move to tc-xtensa.h.
470 (xtensa_const16_opcode, xtensa_movi_opcode, xtensa_movi_n_opcode,
471 xtensa_l32r_opcode, xtensa_nop_opcode, xtensa_rsr_lcount_opcode): New.
472 (xtensa_j_opcode, xtensa_rsr_opcode): Delete.
473 (align_only_targets, software_a0_b_retw_interlock,
474 software_avoid_b_j_loop_end, maybe_has_b_j_loop_end,
475 software_avoid_short_loop, software_avoid_close_loop_end,
476 software_avoid_all_short_loops, specific_opcode): Delete.
477 (warn_unaligned_branch_targets): New.
478 (workaround_a0_b_retw, workaround_b_j_loop_end, workaround_short_loop,
479 workaround_close_loop_end, workaround_all_short_loops): Default FALSE.
480 (option_[no_]link_relax, option_[no_]transform,
481 option_[no_]absolute_literals, option_warn_unaligned_targets,
482 option_prefer_l32r, option_prefer_const16, option_target_hardware):
484 (option_[no_]align_only_targets, option_literal_section_name,
485 option_text_section_name, option_data_section_name,
486 option_bss_section_name, option_eb, option_el): Delete.
487 (md_longopts): Add entries for: [no-]transform, [no-]absolute-literals,
488 warn-unaligned-targets, prefer-l32r, prefer-const16, [no-]link-relax,
489 and target-hardware. Delete entries for [no-]target-align-only,
490 literal-section-name, text-section-name, data-section-name, and
492 (md_parse_option): Handle new options and remove old ones. Accept but
493 ignore [no-]density options. Warn for [no-]generics and [no-]relax
494 and treat them as [no-]transform.
495 (md_show_usage): Add new options and remove old ones.
496 (xtensa_setup_hw_workarounds): New.
497 (md_pseudo_table): Change "word" entry to use xtensa_elf_cons. Add
498 "long", "short", "loc" and "frequency" entries.
499 (use_generics): Rename to ...
500 (use_transform): ... this function. Add past_xtensa_end check.
501 (use_longcalls): Add past_xtensa_end check.
502 (code_density_available, can_relax): Delete.
503 (do_align_targets): New.
504 (get_directive): Accept dashes in directive names. Warn about
505 [no-]generics and [no-]relax directives and treat them as
507 (xtensa_begin_directive): Call md_flush_pending_output only for some
508 directives. Check for directives inside instruction bundles. Warn
509 about deprecated ".begin literal" usage. Warn and ignore [no-]density
510 directives. Handle new directives. Check generating_literals flag
512 (xtensa_end_directive): Check for directives inside instruction
513 bundles. Warn and ignore [no-]density directives. Handle new
514 directives. Call xtensa_set_frag_assembly_state.
515 (xtensa_loc_directive_seen, xtensa_dwarf2_directive_loc,
516 xtensa_dwarf2_emit_insn): New.
517 (xtensa_literal_position): Call md_flush_pending_output. Do not check
518 use_literal_section flag.
519 (xtensa_literal_pseudo): Call md_flush_pending_output. Handle absolute
520 literals. Use xtensa_elf_cons to parse the expression.
521 (xtensa_literal_prefix): Do not check use_literal_section. Support
522 ".lit4" sections for absolute literals. Change prefix convention to
523 replace ".text" (or ".t" in a linkonce section). No need to call
525 (xtensa_frequency_pseudo, xtensa_elf_cons, xtensa_elf_suffix): New.
526 (expression_end): Handle closing braces and colons.
527 (PLT_SUFFIX, plt_suffix): Delete.
528 (expression_maybe_register): Use new xtensa-isa.h functions. Use
529 xtensa_elf_suffix instead of checking for plt suffix, and handle O_lo16
530 and O_hi16 expressions as well.
531 (tokenize_arguments): Handle closing braces and colons.
532 (parse_arguments): Use new xtensa-isa.h functions. Handle "invisible"
533 operands and paired register syntax.
534 (get_invisible_operands): New.
535 (xg_translate_sysreg_op): Handle new Xtensa LX RSR/WSR/XSR syntax. Use
536 new xtensa-isa.h functions.
537 (xtensa_translate_old_userreg_ops, xtensa_translate_zero_immed): New.
538 (xg_translate_idioms): Check if inside bundle. Use use_transform.
539 Handle new Xtensa LX RSR/WSR/XSR syntax. Remove code to widen density
540 instructions. Use xtensa_translate_zero_immed.
541 (operand_is_immed, operand_is_pcrel_label): Delete.
542 (get_relaxable_immed): Use new xtensa-isa.h functions.
543 (get_opcode_from_buf): Add slot parameter. Use new xtensa-isa.h
545 (xtensa_print_insn_table, print_vliw_insn): New.
546 (is_direct_call_opcode): Use new xtensa-isa.h functions.
547 (is_call_opcode, is_loop_opcode, is_conditional_branch_opcode,
548 is_branch_or_jump_opcode): Delete.
549 (is_movi_opcode, decode_reloc, encode_reloc, encode_alt_reloc): New.
550 (opnum_to_reloc, reloc_to_opnum): Delete.
551 (xtensa_insnbuf_set_operand, xtensa_insnbuf_get_operand): Use new
552 xtensa-isa.h functions. Operate on one slot of an instruction.
553 (xtensa_insnbuf_set_immediate_field, is_negatable_branch,
554 xg_get_insn_size): Delete.
555 (xg_get_build_instr_size): Use xg_get_single_size.
556 (xg_is_narrow_insn, xg_is_single_relaxable_insn): Update calls to
557 xg_build_widen_table. Use xg_get_single_size.
558 (xg_get_max_narrow_insn_size): Delete.
559 (xg_get_max_insn_widen_size, xg_get_max_insn_widen_literal_size,
560 xg_is_relaxable_insn): Update calls to xg_build_widen_table. Use
562 (xg_build_to_insn): Record the loc field. Handle OP_OPERAND_HI16U and
563 OP_OPERAND_LOW16U. Check xg_valid_literal_expression.
564 (xg_expand_to_stack, xg_expand_narrow): Update calls to
565 xg_build_widen_table. Use xg_get_single_size.
566 (xg_immeds_fit): Use new xtensa-isa.h functions. Update call to
568 (xg_symbolic_immeds_fit): Likewise. Also handle O_lo16 and O_hi16, and
569 treat weak symbols conservatively.
570 (xg_check_operand): Use new xtensa-isa.h functions.
571 (is_dnrange): Delete.
572 (xg_assembly_relax): Inline previous calls to tinsn_copy.
573 (xg_finish_frag): Specify separate relax states for the frag and slot0.
574 (is_branch_jmp_to_next, xg_add_branch_and_loop_targets): Use new
575 xtensa-isa.h functions.
576 (xg_instruction_matches_option_term, xg_instruction_matches_or_options,
577 xg_instruction_matches_options): New.
578 (xg_instruction_matches_rule): Handle O_register expressions. Call
579 xg_instruction_matches_options.
580 (transition_rule_cmp): New.
581 (xg_instruction_match): Update call to xg_build_simplify_table.
582 (xg_build_token_insn): Record loc fields.
583 (xg_simplify_insn): Check is_specific_opcode field and
584 density_supported flag.
585 (xg_expand_assembly_insn): Skip checking code_density_available. Use
586 new xtensa-isa.h functions. Call use_transform instead of can_relax.
587 (xg_assemble_literal): Add error handling for O_big. Call
588 record_alignment. Handle O_pltrel.
589 (xg_valid_literal_expression): New.
590 (xg_assemble_literal_space): Add slot parameter. Remove call to
591 set_expr_symbol_offset. Add call to record_alignment. Update call to
593 (xg_emit_insn): Delete.
594 (xg_emit_insn_to_buf): Add format parameter. Update calls to
595 xg_add_opcode_fix and xtensa_insnbuf_to_chars.
596 (xg_add_opcode_fix): Change opcode parameter to tinsn and add format
597 and slot parameters. Handle new "alternate" relocations for absolute
598 literals and CONST16 instructions. Check for bad uses of O_lo16 and
599 O_hi16. Use new xtensa-isa.h functions.
600 (xg_assemble_tokens): Delete.
601 (is_register_writer): Use new xtensa-isa.h functions.
602 (is_bad_loopend_opcode): Check for xtensa_rsr_lcount_opcode instead of
603 old-style RSR from LCOUNT.
604 (next_frag_opcode): Delete.
605 (next_frag_opcode_is_loop, next_frag_format_size, frag_format_size,
606 update_next_frag_state): New.
607 (update_next_frag_nop_state): Delete.
608 (next_frag_pre_opcode_bytes): Use next_frag_opcode_is_loop.
609 (xtensa_mark_literal_pool_location): Check use_literal_section flag and
610 the state of the absolute-literals directive. Add calls to
611 record_alignment and xtensa_set_frag_assembly_state. Call
612 xtensa_switch_to_non_abs_literal_fragment instead of
613 xtensa_switch_to_literal_fragment.
615 (assemble_nop): Use build_nop. Update call to xtensa_insnbuf_to_chars.
616 (get_expanded_loop_offset): Change check for undefined opcode to an
618 (xtensa_set_frag_assembly_state, relaxable_section,
619 xtensa_find_unmarked_state_frags, xtensa_find_unaligned_branch_targets,
620 xtensa_find_unaligned_loops, xg_apply_tentative_value): New.
621 (md_begin): Update call to xtensa_isa_init. Initialize linkrelax to 1.
622 Set lit4_seg_name. Call xg_init_vinsn. Initialize new global opcodes.
623 Call init_op_placement_info_table and xtensa_set_frag_assembly_state.
624 (xtensa_init_fix_data): New.
625 (xtensa_frob_label): Reset label symbol to the current frag. Check
626 do_align_targets and generating_literals flag. Propagate frequency
627 info to new alignment frag. Call xtensa_set_frag_assembly_state.
628 (xtensa_unrecognized_line): New.
629 (xtensa_flush_pending_output): Check if inside a bundle. Add a call
630 to xtensa_set_frag_assembly_state.
631 (error_reset_cur_vinsn): New.
632 (md_assemble): Remove check for literal frag. Remove call to
633 istack_init. Call use_transform instead of use_generics. Parse
634 explicit instruction format specifiers. Move code for
635 a0_b_retw_interlock workaround to xg_assemble_vliw_tokens. Call
636 error_reset_cur_vinsn on errors. Add call to get_invisible_operands.
637 Add dwarf2_where call. Remote automatic alignment for ENTRY
638 instructions. Move call to xtensa_clear_insn_labels to the end.
639 Rearrange to handle bundles.
640 (xtensa_cons_fix_new): Delete.
641 (xtensa_handle_align): New.
642 (xtensa_frag_init): Call xtensa_set_frag_assembly_state. Remove
643 assignment to is_no_density field.
644 (md_pcrel_from): Use new xtensa-isa.h functions. Use decode_reloc
645 instead of reloc_to_opnum. Handle "alternate" relocations.
646 (xtensa_force_relocation, xtensa_check_inside_bundle,
647 xtensa_elf_section_change_hook): New.
648 (xtensa_symbol_new_hook): Delete.
649 (xtensa_fix_adjustable): Check for difference of symbols with an
650 offset. Check for external and weak symbols.
651 (md_apply_fix3): Remove cases for XTENSA_OP{0,1,2} relocs.
652 (md_estimate_size_before_relax): Return expansion for the first slot.
653 (tc_gen_reloc): Handle difference of symbols by producing
654 XTENSA_DIFF{8,16,32} relocs and by writing the value of the difference
655 into the output. Handle new XTENSA_SLOT*_OP relocs by storing the
656 tentative values into the output when linkrelax is set.
657 (XTENSA_PROP_SEC_NAME): Define.
658 (xtensa_post_relax_hook): Call xtensa_find_unmarked_state_frags.
659 Create literal tables only if using literal sections. Create new
660 property tables instead of old instruction tables. Check for unaligned
661 branch targets and loops.
662 (finish_vinsn, find_vinsn_conflicts, check_t1_t2_reads_and_writes,
663 new_resource_table, clear_resource_table, resize_resource_table,
664 resources_available, reserve_resources, release_resources,
665 opcode_funcUnit_use_unit, opcode_funcUnit_use_stage,
666 resources_conflict, xg_find_narrowest_format, relaxation_requirements,
667 bundle_single_op, emit_single_op, xg_assemble_vliw_tokens): New.
668 (xtensa_end): Call xtensa_flush_pending_output. Set past_xtensa_end
669 flag. Update checks for workaround options. Call
670 xtensa_mark_narrow_branches and xtensa_mark_zcl_first_insns.
671 (xtensa_cleanup_align_frags): Add special case for branch targets.
672 Check for and mark unreachable frags.
673 (xtensa_fix_target_frags): Remove use of align_only_targets flag.
674 Use RELAX_LOOP_END_BYTES in special case for negatable branch at the
675 end of a zero-overhead loop body.
676 (frag_can_negate_branch): Handle instructions with multiple slots.
677 Use new xtensa-isa.h functions
678 (xtensa_mark_narrow_branches, is_narrow_branch_guaranteed_in_range,
679 xtensa_mark_zcl_first_insns): New.
680 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags): Error if
681 transformations are disabled.
682 (next_instrs_are_b_retw): Use new xtensa-isa.h functions. Handle
683 multislot instructions.
684 (xtensa_fix_close_loop_end_frags, xtensa_fix_short_loop_frags):
685 Likewise. Also error if transformations are disabled.
686 (unrelaxed_frag_max_size): New.
687 (unrelaxed_frag_min_insn_count, unrelax_frag_has_b_j): Use new
688 xtensa-isa.h functions.
689 (xtensa_sanity_check, is_empty_loop, is_local_forward_loop): Use
690 xtensa_opcode_is_loop instead of is_loop_opcode.
691 (get_text_align_power): Replace as_fatal with assertion.
692 (get_text_align_fill_size): Iterate instead of using modulus when
694 (get_noop_aligned_address): Assert that this is for a machine-dependent
695 RELAX_ALIGN_NEXT_OPCODE frag. Use next_frag_opcode_is_loop,
696 xg_get_single_size, and frag_format_size.
697 (get_widen_aligned_address): Rename to ...
698 (get_aligned_diff): ... this function. Add max_diff parameter.
699 Remove handling of rs_align/rs_align_code frags. Use
700 next_frag_format_size, get_text_align_power, get_text_align_fill_size,
701 next_frag_opcode_is_loop, and xg_get_single_size. Compute max_diff
702 and pass it back to caller.
703 (xtensa_relax_frag): Use relax_frag_loop_align. Add code for new
704 RELAX_SLOTS, RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN,
705 RELAX_FILL_NOP, and RELAX_UNREACHABLE frag types. Check relax_seen.
706 (relax_frag_text_align): Rename to ...
707 (relax_frag_loop_align): ... this function. Assume loops can only be
708 in the first slot of an instruction.
709 (relax_frag_add_nop): Use assemble_nop instead of constructing an OR
710 instruction. Remove call to frag_wane.
711 (relax_frag_narrow): Rename to ...
712 (relax_frag_for_align): ... this function. Extend to handle
713 RELAX_FILL_NOP and RELAX_UNREACHABLE, as well as RELAX_SLOTS with
714 RELAX_NARROW for the first slot.
715 (find_address_of_next_align_frag, bytes_to_stretch): New.
716 (future_alignment_required): Use find_address_of_next_align_frag and
717 bytes_to_stretch. Look ahead to subsequent frags to make smarter
719 (relax_frag_immed): Add format, slot, and estimate_only parameters.
720 Check if transformations are enabled for b_j_loop_end workaround.
721 Use new xtensa-isa.h functions and handle multislot instructions.
722 Update call to xg_assembly_relax.
723 (md_convert_frag): Handle new RELAX_SLOTS, RELAX_UNREACHABLE,
724 RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN, and RELAX_FILL_NOP
726 (convert_frag_narrow): Add segP, format and slot parameters. Call
727 convert_frag_immed for branch instructions. Adjust calls to
728 tinsn_from_chars, tinsn_immed_from_frag, and xg_emit_insn_to_buf. Use
729 xg_get_single_size and xg_get_single_format.
730 (convert_frag_fill_nop): New.
731 (convert_frag_immed): Add format and slot parameters. Handle multislot
732 instructions and use new xtensa-isa.h functions. Update calls to
733 tinsn_immed_from_frag and xg_assembly_relax. Check if transformations
734 enabled for b_j_loop_end workaround. Use build_nop instead of
735 assemble_nop. Check is_specific_opcode flag. Check for unreachable
736 frags. Use xg_get_single_size. Handle O_pltrel.
737 (fix_new_exp_in_seg): Remove check for old plt flag.
738 (convert_frag_immed_finish_loop): Update calls to tinsn_from_chars and
739 xtensa_insnbuf_to_chars. Call tinsn_immed_from_frag. Change check
740 for loop opcode to an assertion. Mark all frags up to the end of the
741 loop as not transformable.
742 (get_last_insn_flags, set_last_insn_flags): Use get_subseg_info.
743 (get_subseg_info): New.
744 (xtensa_move_literals): Call xtensa_set_frag_assembly_state. Add null
746 (xtensa_switch_to_literal_fragment): Rewrite to handle absolute
747 literals and use xtensa_switch_to_non_abs_literal_fragment otherwise.
748 (xtensa_switch_to_non_abs_literal_fragment): New.
749 (cache_literal_section): Add is_code parameter and pass it through to
750 retrieve_literal_seg.
751 (retrieve_literal_seg): Add is_code parameter and use it to set the
752 flags on the literal section. Handle case where head parameter is 0.
753 (get_frag_is_no_transform, set_frag_is_specific_opcode,
754 set_frag_is_no_transform): New.
755 (xtensa_create_property_segments): Add end_property_function parameter
756 and pass it through to add_xt_block_frags. Call bfd_get_section_flags
757 and skip SEC_DEBUGGING and !SEC_ALLOC sections.
758 (xtensa_create_xproperty_segments, section_has_xproperty): New.
759 (add_xt_block_frags): Add end_property_function parameter and call it
760 if it is non-zero. Call xtensa_frag_flags_init.
761 (xtensa_frag_flags_is_empty, xtensa_frag_flags_init,
762 get_frag_property_flags, frag_flags_to_number,
763 xtensa_frag_flags_combinable, xt_block_aligned_size,
764 xtensa_xt_block_combine, add_xt_prop_frags,
765 init_op_placement_info_table, opcode_fits_format_slot,
766 xg_get_single_size, xg_get_single_format): New.
767 (istack_push): Inline call to tinsn_copy.
768 (tinsn_copy): Delete.
769 (tinsn_has_invalid_symbolic_operands): Handle O_hi16 and O_lo16 and
770 CONST16 opcodes. Handle O_big, O_illegal, and O_absent.
771 (tinsn_has_complex_operands): Handle O_hi16 and O_lo16.
772 (tinsn_to_insnbuf): Use xg_get_single_format and new xtensa-isa.h
773 functions. Handle invisible operands.
774 (tinsn_to_slotbuf): New.
775 (tinsn_check_arguments): Use new xtensa-isa.h functions.
776 (tinsn_from_chars): Add slot parameter. Rewrite using xg_init_vinsn,
777 vinsn_from_chars, and xg_free_vinsn.
778 (tinsn_from_insnbuf): New.
779 (tinsn_immed_from_frag): Add slot parameter and handle multislot
780 instructions. Handle symbol differences.
781 (get_num_stack_text_bytes): Use xg_get_single_size.
782 (xg_init_vinsn, xg_clear_vinsn, vinsn_has_specific_opcodes,
783 xg_free_vinsn, vinsn_to_insnbuf, vinsn_from_chars, expr_is_register,
784 get_expr_register, set_expr_symbol_offset_diff): New.
785 * config/tc-xtensa.h (MAX_SLOTS): Define.
786 (xtensa_relax_statesE): Move from tc-xtensa.c. Add
787 RELAX_CHECK_ALIGN_NEXT_OPCODE, RELAX_MAYBE_DESIRE_ALIGN, RELAX_SLOTS,
788 RELAX_FILL_NOP, RELAX_UNREACHABLE, RELAX_MAYBE_UNREACHABLE, and
790 (RELAX_IMMED_MAXSTEPS): Move from tc-xtensa.c.
791 (xtensa_frag_type struct): Add is_assembly_state_set,
792 use_absolute_literals, relax_seen, is_unreachable, is_specific_opcode,
793 is_align, is_text_align, alignment, and is_first_loop_insn fields.
794 Replace is_generics and is_relax fields by is_no_transform field.
795 Delete is_text and is_longcalls fields. Change text_expansion and
796 literal_expansion to arrays of MAX_SLOTS entries. Add arrays of
797 per-slot information: literal_frags, slot_subtypes, slot_symbols,
798 slot_sub_symbols, and slot_offsets. Add fr_prev field.
799 (xtensa_fix_data struct): New.
800 (xtensa_symfield_type struct): Delete plt field.
801 (xtensa_block_info struct): Move definition to tc-xtensa.h. Add
802 forward declaration here.
803 (xt_section_type enum): Delete xt_insn_sec. Add xt_prop_sec.
804 (XTENSA_SECTION_RENAME): Undefine.
805 (TC_FIX_TYPE, TC_INIT_FIX_DATA, TC_FORCE_RELOCATION, NO_PSEUDO_DOT,
806 tc_unrecognized_line, md_do_align, md_elf_section_change_hook,
807 HANDLE_ALIGN, TC_LINKRELAX_FIXUP, SUB_SEGMENT_ALIGN): Define.
808 (TC_CONS_FIX_NEW, tc_symbol_new_hook): Delete.
809 (unit_num_copies_func, opcode_num_units_func,
810 opcode_funcUnit_use_unit_func, opcode_funcUnit_use_stage_func): New.
811 (resource_table struct): New.
812 * config/xtensa-istack.h (MAX_INSN_ARGS): Increase from 6 to 10.
813 (TInsn struct): Add keep_wide, loc, fixup, record_fix, subtype,
814 literal_space, symbol, sub_symbol, offset, and literal_frag fields.
815 (tinsn_copy): Delete prototype.
816 (vliw_insn struct): New.
817 * config/xtensa-relax.c (insn_pattern_struct): Add options field.
818 (widen_spec_list): Add option conditions for density and boolean
819 instructions. Add expansions using CONST16 and conditions for using
820 CONST16 vs. L32R. Use new Xtensa LX RSR/WSR syntax. Add entries for
822 (simplify_spec_list): Add option conditions for density instructions.
823 Add entry for NOP instruction.
824 (append_transition): Add cmp function pointer parameter and use it to
825 insert the new entry in order.
826 (operand_function_LOW16U, operand_function_HI16U): New.
827 (xg_has_userdef_op_fn, xg_apply_userdef_op_fn): Handle
828 OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
829 (enter_opname, split_string): Use xstrdup instead of strdup.
830 (init_insn_pattern): Initialize new options field.
831 (clear_req_or_option_list, clear_req_option_list,
832 clone_req_or_option_list, clone_req_option_list, parse_option_cond):
834 (parse_insn_pattern): Parse option conditions.
835 (transition_applies): New.
836 (build_transition): Use new xtensa-isa.h functions. Fix incorrectly
837 swapped last arguments in calls to append_constant_value_condition.
838 Call clone_req_option_list. Add warning about invalid opcode.
839 Handle LOW16U and HI16U function names.
840 (build_transition_table): Add cmp parameter and use it in calls to
841 append_transition. Use new xtensa-isa.h functions. Check
842 transition_applies before adding entries.
843 (xg_build_widen_table, xg_build_simplify_table): Add cmp parameter and
844 pass it through to build_transition_table.
845 * config/xtensa-relax.h (ReqOrOptionList, ReqOrOption, ReqOptionList,
846 ReqOption, transition_cmp_fn): New types.
847 (OpType enum): Add OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
848 (transition_rule struct): Add options field.
849 * doc/as.texinfo (Overview): Update Xtensa options.
850 * doc/c-xtensa.texi (Xtensa Options): Delete --[no-]density,
851 --[no-]relax, and --[no-]generics options. Update descriptions of
852 --text-section-literals and --[no-]longcalls. Add
853 --[no-]absolute-literals and --[no-]transform.
854 (Xtensa Syntax): Add description of syntax for FLIX instructions.
855 Remove use of "generic" and "specific" terminology for opcodes.
856 (Xtensa Registers): Generalize the syntax description to include
857 user-defined register files.
858 (Xtensa Automatic Alignment): Update.
859 (Xtensa Branch Relaxation): Mention limitation of unconditional jumps.
860 (Xtensa Call Relaxation): Linker can now remove most of the overhead.
861 (Xtensa Directives): Remove confusing rules about precedence.
862 (Density Directive, Relax Directive): Delete.
863 (Schedule Directive): New.
864 (Generics Directive): Rename to ...
865 (Transform Directive): ... this node.
866 (Literal Directive): Update for absolute literals. Missing
867 literal_position directive is now an error.
868 (Literal Position Directive): Update for absolute literals.
869 (Freeregs Directive): Delete.
870 (Absolute Literals Directive): New.
871 (Frame Directive): Minor editing.
872 * Makefile.am (DEPTC_xtensa_elf, DEPOBJ_xtensa_elf, DEP_xtensa_elf):
874 * Makefile.in: Regenerate.
878 * config/tc-mips.c (append_insn): Use fix_new rather than fix_new_exp
879 to build the second and third fixups for a composite relocation.
880 (macro_read_relocs): New function.
881 (macro_build): Use it.
882 (s_cpsetup): Pass all three composite relocation codes to macro_build.
883 Simplify fragging code accordingly.
884 (s_gpdword): Use fix_new rather than fix_new_exp for the second part
885 of the composite relocation. Set fx_tcbit in both fixups.
889 * config/tc-mips.c (append_insn): Set fx_tcbit for composite relocs.
890 (md_apply_fix3): Don't treat composite relocs as done.
894 * macro.c (macro_expand_body): When ELF, use .LL rather than LL as
895 prefix for symbol names generated from the LOCAL macro directive.
897 * dw2gencfi.c (select_cie_for_fde): When separating CIE out from
898 FDE, treat a DW_CFA_remember_state as we do a DW_CFA_advance_loc.
902 * config/tc-crx.c (preprocess_reglist): Handle Co-processor
904 (md_assemble): Add error checking for Co-Processor instructions.
905 (get_cinv_parameters): Add 'b' option to invalidate the
910 * config/tc-arm.c (unwind): New variable.
911 (vfp_sp_encode_reg): New function.
912 (vfp_sp_reg_required_here): Use it.
913 (vfp_sp_reg_list, vfp_dp_reg_list): Remove.
914 (vfp_parse_reg_list): New function.
915 (s_arm_unwind_fnstart, s_arm_unwind_fnend, s_arm_unwind_cantunwind,
916 s_arm_unwind_personality, s_arm_unwind_personalityindex,
917 s_arm_unwind_handlerdata, s_arm_unwind_save, s_arm_unwind_movsp,
918 s_arm_unwind_pad, s_arm_unwind_setfp, s_arm_unwind_raw): New
920 (md_pseudo_table): Add them.
921 (do_vfp_reg2_from_sp2): Use vfp_parse_reg_list and vfp_sp_encode_reg.
922 (do_vfp_sp2_from_reg2, vfp_sp_ldstm, vfp_dp_ldstm): Ditto.
923 (set_section, add_unwind_adjustsp, flush_pending_unwind,
924 finish_unwind_opcodes, start_unwind_section, create_unwind_entry,
925 require_hashconst, add_unwind_opcode): New functions.
926 * doc/tc-arm.text: Document unwinding opcodes.
927 * NEWS: Mention the new feature.
931 * config/tc-mips.c (md_apply_fix3): Remove erroneous assert.
935 * config/tc-ppc.c (md_apply_fix3): Call S_SET_THREAD_LOCAL for
937 * config/tc-s390.c (md_apply_fix3): Likewise.
938 * config/tc-sparc.c (md_apply_fix3): Likewise.
942 * config/tc-arm.c (arm_elf_section_type): New function.
943 (arm_elf_change_section): Set section link for exidx sections.
944 * config/tc-arm.h (arm_elf_section_type): Add prototype.
945 (md_elf_section_type): Define.
949 * config/tc-pdp11.c (md_apply_fix3): Change to sign of the SOB
950 instruction's offset.
954 * (TARGET_FORMAT): Remove LynxOS COFF definition.
958 * config/tc-arc.c (tc_gen_reloc): Don't assume fixP->fx_addsy is an
959 asymbol *, instead use symbol_get_bfdsym.
963 * config/tc-m68k.c (select_control_regs): Add mcf5249.
967 * config/tc-arm.c (do_smi, do_nop): New functions.
968 (insns): Add ARMv6ZK instructions.
969 (md_apply_fix3): Handle BFD_RELOC_ARM_SMI.
970 (tc_gen_reloc): Ditto.
971 (arm_cpus): Add mpcore and arm1176.
972 (arm_archs): Add armv6{k,z,zk}.
973 * doc/c-arm.texi: Document new cores and architectures.
977 * config/tc-arm.c: Use ISO C90 formatting.
981 * config/tc-arm.c (mav_reg_required_here): Allow REG_TYPE_CN
982 as alternative when REG_TYPE_MVF, REG_TYPE_MVD, REG_TYPE_MVFX or
983 REG_TYPE_MVDX is expected.
987 * doc/c-i386.texi (i386-Mnemonics): Fix typo.
991 * config/tc-ia64.c (ENCODED_PSP_OFFSET): New.
992 (output_rp_psprel, output_pfs_psprel, output_preds_psprel,
993 output_spill_base, output_unat_psprel, output_lc_psprel,
994 output_fpsr_psprel, output_priunat_psprel, output_bsp_psprel,
995 output_bsprestore_psprel, output_rnat_psprel, output_spill_psprel,
996 output_spill_psprel_p): Use it.
1000 * config/tc-crx.c (handle_LoadStor): New function.
1001 Handle load/stor unique instructions before parsing.
1005 * config/tc-arm.c (s_arm_rel31): New funciton.
1006 (md_pseudo_table): Add .rel31.
1007 (md_apply_fix3): Handle BFD_RELOC_ARM_TARGET2,
1008 BFD_RELOC_32_PCREL and BFD_RELOC_ARM_PREL31.
1009 (tc_gen_reloc): Handle BFD_RELOC_ARM_PREL31 and BFD_RELOC_ARM_TARGET2.
1010 (arm_fix_adjustable): Return 0 for BFD_RELOC_ARM_TARGET2.
1011 (arm_parse_reloc): Add (target2).
1015 * Makefile.am: Run "make dep-am".
1016 * Makefile.in: Regenerate.
1017 * aclocal.m4: Regenerate.
1018 * configure: Regenerate.
1019 * doc/Makefile.in: Regenerate.
1020 * po/POTFILES.in: Regenerate.
1021 * po/gas.pot: Regenerate.
1025 * config/tc-mmix.c [!LLONG_MIN]: Correct #elsif to #elif.
1026 [!LLONG_MAX]: Ditto.
1030 * config/tc-arm.c: Rename RELABS to TARGET1.
1034 * messages.c (as_internal_value_out_of_range): Cast values passed
1035 to as_bad_where or as_warn_where to proper type.
1039 * config/tc-avr.c: Add support for
1040 atmega48, atmega88, atmega168, attiny13, attiny2313, at90can128.
1044 * dw2gencfi.c (select_cie_for_fde): When separating CIE out
1045 from FDE, treat a CFI_escape as we do a DW_CFA_advance_loc.
1049 * config/obj-elf.c (obj_elf_section_type): Handle init_array,
1050 fini_array and preinit_array section types.
1051 * config/tc-ia64.c (ia64_elf_section_type): Remove init_array
1053 * doc/as.texinfo: Document extra section types.
1057 * Makefile.am (TARG_ENV_HFILES): Add te-symbian.h.
1058 * Makefile.in: Regenerated.
1059 * configure.in: Set em for arm*-*-symbianelf*.
1060 * configure: Regenerated.
1061 * config/tc-arm.c (elf32_arm_target_format): Use Symbian target
1062 vectors when appropriate.
1063 * config/te-symbian.h: New file.
1067 * config/tc-crx.c (gettrap): Exception vector can be case
1069 (process_label_constant): Fix a 32-bit displacement bug in branch
1071 (get_operandtype) : Bug fix, wrong operand was used.
1072 (process_label_constant): Initialize relocation type to
1077 * tc-arm.c (arm_cpus, arm_fpus): Allow <cpu>-s as well as <cpu>s
1078 for synthesizable cores.
1080 * doc/c-arm.texi (ARM Options): Document canonical names of CPUs.
1084 * config/tc-msp430.c: Clean-up the code.
1085 (md_relax_table): New relax table.
1086 (mcu_types): Sort MCU types.
1087 (md_pseudo_table): Add .profiler pseudo handler.
1088 (pow2value): New function.
1089 (msp430_profiler): New function.
1090 (msp430_operands): Add new insns handlers.
1091 (msp430_srcoperand): Add register operand handler, allow complex
1093 (md_estimate_size_before_relax): Rewritten.
1094 (md_convert_frag): Rewritten.
1095 (msp430_relax_frag): New function.
1096 * config/tc-msp430.h (md_relax_frag): define macro
1097 * doc/c-msp430.texi: Update information.
1101 * as.c (std_shortopts): Allow -g to take an optional argument.
1102 (parse_args): Pass any switch starting with -g on to the backend
1107 * configure.in (arm*-*-symbianelf*): New target.
1108 (arm*-*-eabi*): Likewise.
1109 * configure: Regenerated.
1112 * config/tc-mips.c (append_insn): Handle delay slots in branch likely
1117 * config/tc-ia64.c (start_unwind_section): Add linkonce_empty
1118 argument, don't do anything if current section is not
1119 .gnu.linkonce.t.* and linkonce_empty is set.
1120 (generate_unwind_image, dot_endp): Adjust callers, call
1121 start_unwind_section (*, 1) if nothing will be put into the
1126 * as.c (MD_DEBUG_FORMAT_SELECTOR): Provide default definition.
1127 (show_usage): Add -g.
1128 (std_longopts): Add --gen-debug. Alpha sort the table.
1129 (parse_args): Print an error message if a switch is not handled.
1130 Handle the -g switch, calling md_debug_format_selector() if
1132 * NEWS: Mention new feature.
1133 * doc/as.texinfo: Document new switch.
1134 * doc/internals.texi: Document behaviour of md_parse_option.
1136 * config/tc-arm.c (md_parse_option): Do not issue an error message
1137 if the switch is not recognised.
1138 * config/tc-m68k.c (md_parse_option): Likewise.
1139 * config/tc-pdp11.c (md_parse_option): Likewise.
1140 * config/tc-v850.c (md_parse_option): Likewise.
1142 * as.h: Fix up formatting.
1147 * macro.c (macro_set_alternate): Use ISO C90 formatting.
1149 * configure.in: Sort architecture based tables alphabetically.
1150 * configure: Regenerate.
1154 * config/tc-ppc.c (tc_ppc_regname_to_dw2regnum <regnames>): Replace
1155 { "cc", 68 }, with { "cr", 70 }.
1160 * as.c: Add and handle new --alternate command line option.
1161 * macro.c (macro_set_alternate): New.
1162 * macro.h (macro_set_alternate): Declare.
1163 * read.c: Add and handle new .altmacro and .noaltmacro directives.
1164 * doc/as.texinfo: Document new command line option and pseudo-ops
1165 as well as insert documentation originating from gasp about
1166 alternate macro syntax.
1167 * NEWS: Mention new command line option and pseudo-ops.
1171 * expr.c (operand): Handle the "~", "-", and "!" operators applied
1176 * config/tc-arm.c (md_apply_fix3, tc_gen_reloc, arm_parse_reloc):
1177 Handle new relocations.
1178 * include/elf/arm.h (elf_arm_reloc_type): Add new EABI relocations.
1182 * write.c (relax_segment): Use was_address instead of address when
1183 setting fr_fix field for align frag due to backwards .org.
1187 Introduce SH2a support.
1189 * config/tc-sh.c (get_specific): Change arch_sh2a_up to
1192 * config/tc-sh.c (md_parse_option): Add sh2a-nofpu ISA handling.
1194 * config/tc-sh.c (sh_elf_final_processing): Move sh2a recognition
1195 to end of conditional expression.
1197 * config/tc-sh.c: Add sh2a-nofpu support.
1199 * tc-sh.c: Add sh2a support.
1200 (parse_reg): Add tbr.
1201 (parse_at): Support @@(disp,tbr).
1202 (get_specific): Support sh2a opcodes.
1203 (insert4): New, for 4 byte relocs.
1204 (build_Mytes): Support sh2a opcodes.
1205 (md_apply_fix3_Mytes): Support sh2a opcodes.
1207 * config/tc-sh.c (md_parse_option): Handle sh2a.
1208 (sh_elf_final_processing): Ditto.
1212 * config/tc-hppa.h (TARGET_FORMAT): Set to "elf32-hppa-netbsd"
1217 * config/tc-ppc.c (ppc_frob_file_before_adjust): Warn if .toc too big.
1218 (ppc_arch): Expand comment.
1222 * config/tc-crx.c: Support evaluating the difference between two
1224 * config/tc-crx.h: Likewise.
1228 * config/tc-ia64.c (start_unwind_section): Set the linked-to
1230 (ia64_elf_section_change_hook): Set the linked-to section for
1235 * config/tc-msp430.c: Add new subtargets: msp430x1610,
1236 msp430x1611, msp430x1612, msp430x415, msp430x417, msp430xG437,
1237 msp430xG438, msp430xG439.
1241 * doc/as.texinfo (Section, PushSection): Correct documentation
1246 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
1247 permissible for the selected instruction suffix.
1248 (match_template): Don't permit 64-bit general purpose operands in
1250 (finalize_imm): Permit 64-bit immediates.
1251 (build_modrm_byte): Don't treat 32-bit addressing in 64-bit mode
1252 specially except for the width of the used base and/or index
1253 registers. For 32-bit displacements, use sign-extended
1254 relocations only when using 64-bit addressing.
1255 Force zero displacement on rip-relative addressing when there is
1256 no other displacement.
1257 (i386_index_check): Don't treat 32-bit addressing in 64-bit mode
1258 specially except for the width of the used base and/or index
1260 (parse_register): Disallow Reg64 registers in 32-bit mode.
1262 * config/tc-i386.c: For DefaultSize instructions, don't guess a 'q'
1263 suffix if the instruction doesn't support it.
1267 * config/tc-mips.c (append_insn): Handle constant expressions with
1268 no associated relocation.
1269 (mips_ip): Cancel the expression after use for the Q format
1271 (parse_relocation): Return no relocation for unsupported
1273 (my_getSmallExpression): Return no relocation if no relocation
1278 * config/obj-som.c (adjust_stab_sections): Add prototype.
1279 (obj_som_compiler, obj_som_version, obj_som_copyright,
1280 adjust_stab_sections): Add ATTRIBUTE_UNUSED to unused arguments.
1281 * config/tc-hppa.c (update_subspace): Likewise.
1282 (is_defined_subspace): Amplify comment.
1283 * config/obj-som.h (som_frob_file): Add prototype.
1288 * subsegs.c (section_symbol): Don't create a new segment when
1289 existing segment is undefined.
1293 * config/tc-arm.c: Include include/opcode/arm.h.
1294 (ARM_EXT_*, ARM_ARCH_*, ARM_ANY, ARM_ALL, COPROC_ANY): Delete.
1295 (FPU_FPA_EXT_* FPU_VFP_EXT_*, FPU_ANY, FPU_NONE, FPU_MAVERICK): Delete.
1296 (FPU_ARCH_*): Delete.
1297 * Makefile.am: Update dependencies.
1298 * Makefile.in: Regenerate.
1302 * configure.in: Accept armbe as a big-endian arm configuration.
1303 * configure: Regenerate.
1307 * config/tc-i386.c (T_SHIFTOP): New constant.
1308 (intel_e05_1): Handle '&', '|' and T_SHIFTOP.
1309 (intel_el1): Handle '~'.
1310 (intel_get_token): Handle '<>', '&', '|' and '~'.
1314 (md_assemble): Remove spurious newline from end of as_bad error
1316 (intel_e05_1): Likewise.
1317 (intel_e11): Likewise.
1318 (intel_match_token): Likewise.
1322 * config/tc-m68k.c: Convert to C90. Remove redundant
1323 declarations. Indentation fixup.
1324 [M68KCOFF]: Include "obj-coff.h" instead of declaring
1325 obj_coff_section ourselves.
1329 * config/tc-ia64.c (default_big_endian): New.
1330 (dot_byteorder, md_begin): Use it.
1331 (md_parse_option): Set it.
1335 * configure.in: Change sh-sybmian-elf to sh-*-symbianelf.
1336 * configure: Regenerate.
1337 * NEWS: Change sh-sybmian-elf to sh-*-symbianelf.
1338 * config/tc-sh.c (sh_elf_final_processing): Use renamed version of
1339 sh_find_elf_flags if necessary.
1343 * config/tc-mips.c (mips_fix_adjustable): If the full addend is
1344 going to be split into more than one in-place addend, return 0
1345 for relocations against mergeable sections. Associate comments
1350 * Makefile.am (CPU_TYPES): Add crx.
1351 (TARGET_CPU_CFILES): Add config/tc-crx.c.
1352 (TARGET_CPU_HFILES): Add config/tc-crx.h.
1353 (DEPTC_crx_elf): New target.
1354 (DEPOBJ_crx_elf): Likewise.
1355 (DEP_crx_elf): Likewise.
1356 * Makefile.in: Regenerate.
1357 * configure.in: Add crx* target.
1358 * configure: Regenerate.
1359 * config/tc-crx.c: New file.
1360 * config/tc-crx.h: New file.
1361 * NEWS: Mention new target.
1365 * config.in: Undefine TARGET_SYMBIAN by default.
1367 * configure: Regenerate. Add sh-symbian-elf target. If
1368 selected define TARGET_SYMBIAN.
1369 * config/tc-sh.h (TARGET_FORMAT): Select a Symbian target
1370 format if TARGET_SYMBIAN has been defined.
1372 * output-file.c (output_file_create): Report the target format
1373 chosen when bfd_openw reports that it is invalid.
1375 * config/obj-coff.c (coff_pseudo_table): Only define the weak
1376 pseudo for BFD based assemblers.
1381 * config/tc-sh.c (md_assemble): Change isspace to ISSPACE.
1382 (md_parse_option): Remove redundant -isa testing.
1383 Make bfd_arch variable const.
1384 (md_show_usage): Make bfd_arch variable const.
1388 * config/tc-ia64.c (emit_one_bundle): Check and set insn_addr.
1389 * config/tc-ia64.h (md_frag_check): Define.
1393 * config/obj-coff.c (obj_coff_weak): New .weak syntax for PE weak
1395 * doc/as.texinfo (Weak): Document PE weak symbols.
1399 * config/tc-mips.c (HAVE_IN_PLACE_ADDENDS): New macro.
1400 (reloc_needs_lo_p): Only return true if HAVE_IN_PLACE_ADDENDS.
1401 (mips_frob_file): Rework so that only a single pass through the
1402 relocs is needed. Allow %lo()s to have higher offsets than their
1403 corresponding %hi()s or %got()s.
1407 * config/tc-arm.c (md_apply_fix3:BFD_RELOC_ARM_IMMEDIATE): Do not
1408 allow values which have come from undefined symbols.
1409 Always consider this fixup to have been processed as a reloc
1410 cannot be generated for it.
1414 * frags.h (struct frag): Add has_code and insn_addr fields.
1415 * write.c (cvt_frag_to_fill): Invoke md_frag_check.
1416 * config/tc-ppc.c (md_assemble): Check and set insn_addr.
1417 * config/tc-ppc.h (md_frag_check): Define.
1421 * doc/Makefile.am (info): Rename goal to...
1422 (info-local): ... this, to preserve implicit dependencies.
1423 * doc/Makefile.in: Regenerate with automake 1.8.5.
1427 * config/tc-m32r.c (md_convert_frag): Changed for @PLT.
1428 (m32r_cgen_record_fixup_exp): Changed for @GOTOFF, @GOT.
1429 (m32r_fix_adjustable): Changed for @GOTOFF, @GOT, @PLT.
1430 (tc_gen_reloc): Likewise.
1431 (m32r_end_of_match): Add for @GOTOFF, @GOT, @PLT.
1432 (m32r_parse_name): Likewise.
1433 (m32r_cgen_parse_fix_exp): Likewise.
1434 * config/tc-m32r.h (md_parse_name): Define for @GOTOFF, @GOT, @PLT.
1435 (O_PIC_reloc): Likewise.
1436 (TC_CGEN_PARSE_FIX_EXP): Likewise..
1437 * cgen.c (gas_cgen_parse_operand): Add TC_CGEN_PARSE_FIX_EXP
1438 for @GOTOFF, @GOT, @PLT.
1442 * gas/symbols.c: While discarding ordinary local absolute symbols
1443 when --strip-local-absolute is in effect, retain file symbols.
1447 * config/tc-m68k.c (mri_chip): Replace current_chip, not augment.
1448 (md_parse_option): Likewise.
1452 * config/tc-i386.c: Deal with LEX_QM the same way as with LEX_AT.
1453 * config/te-netware.h: New file.
1454 * config/te-ppcnw.h: Delete: Obsolete.
1455 * configure.in: Eliminate ill NetWare targets. Make generic
1456 NetWare target use proper emulation.
1457 * Makefile.am: Eliminate reference to obsolete te-ppcnw.h, add
1458 reference to new te-netware.h.
1459 * configure: Regenerate.
1460 * Makefile.in: Regenerate.
1464 * config/tc-s390.c (s390_insn): Avoid incorrect signed/unsigned
1465 comparison in .insn pseudo operation.
1469 * config/obj-coff.c (coff_adjust_section_syms): Use
1470 bfd_get_section_size instead of bfd_get_section_size_before_reloc.
1471 (coff_frob_section): Likewise.
1472 * config/tc-mips.c (md_apply_fix3): Likewise.
1473 * config/obj-elf.c (elf_frob_file): Use bfd_set_section_size.
1474 (elf_frob_file_after_relocs): Likewise.
1478 * config/tc-hppa.c (log2): Rename to exact_log2.
1479 (pa_next_subseg): Delete unused function.
1480 (create_new_space): Mark unused arguments with ATTRIBUTE_UNUSED.
1481 (create_new_subspace): Likewise.
1484 * config/tc-hppa.c (hppa_fix_adjustable): Allow reduction of fake
1485 labels. Fix warning.
1489 * config/tc-mn10300.h (tc_fix_adjustable): Define.
1490 * config/tc-mn10300.c (mn10300_fix_adjustable): Don't adjust debug
1491 or non-merged symbols.
1495 * config/tc-ia64.c (remove_marked_resource): Save, clear and
1496 restore the old slot when inserting srlz.i/srlz.d.
1500 * Makefile.am: Regenerate dependecies.
1501 * Makefile.in: Regenerate.
1502 * config/tc-sh.c (valid_arch): Make unsigned.
1503 (preset_target_arch): Likewise.
1504 (md_begin): Use new architecture flags system.
1505 (get_specific): Likewise.
1506 (assemble_ppi): Likewise.
1507 (md_assemble): Likewise. Also fix error check for bad opcodes.
1508 (md_parse_option): Likewise. Also generate -isa values according
1509 to the table in bfd/cpu-sh.c instead of just constants. Also
1510 allow <arch>-up ISA variants.
1511 (sh_elf_final_processing): Replace if-else chain with a call to
1512 sh_find_elf_flags().
1513 * testsuite/gas/sh/arch: New directory.
1514 * testsuite/gas/sh/arch/arch.exp: New test script.
1515 * testsuite/gas/sh/arch/arch_expected.txt: New file.
1516 * testsuite/gas/sh/arch/sh.s: New file.
1517 * testsuite/gas/sh/arch/sh2.s: New file.
1518 * testsuite/gas/sh/arch/sh-dsp.s: New file.
1519 * testsuite/gas/sh/arch/sh2e.s: New file.
1520 * testsuite/gas/sh/arch/sh3-nommu.s: New file.
1521 * testsuite/gas/sh/arch/sh3.s: New file.
1522 * testsuite/gas/sh/arch/sh3-dsp.s: New file.
1523 * testsuite/gas/sh/arch/sh3e.s: New file.
1524 * testsuite/gas/sh/arch/sh4-nommu-nofpu.s: New file.
1525 * testsuite/gas/sh/arch/sh4-nofpu.s: New file.
1526 * testsuite/gas/sh/arch/sh4.s: New file.
1527 * testsuite/gas/sh/arch/sh4a-nofpu.s: New file.
1528 * testsuite/gas/sh/arch/sh4al-dsp.s: New file.
1529 * testsuite/gas/sh/arch/sh4a.s: New file.
1533 * config/gc-m68k.c(m68k_ip): Convert mode 5 addressing
1534 with zero offset into mode 2 addressing to save a word.
1538 * config/tc-ia64.c (ar_is_in_integer_unit): Removed.
1539 (ar_is_only_in_integer_unit): New.
1540 (ar_is_only_in_memory_unit): New.
1541 (generate_unwind_image): Silence gcc on 32bit host.
1542 (md_assemble): Use ar_is_only_in_integer_unit instead of
1543 ar_is_in_integer_unit. Check AR access.
1547 * config/tc-m68k.c (md_begin): Sort the opcode table into
1549 (m68k_compare_opcode): New function to do the sorting.
1553 * config/m68k-parse.y(operand): Allow for MAC/EMAC mask
1554 addressing on MIT style operands.
1555 * config/m68k-parse.y(yylex): Allow '-&' for predecrement
1557 * config/tc-m68k.c(install_operand): Comment 'G' and 'H' type
1562 * expr.c (operand, operator): Don't reject '++' and '--'.
1566 * config/tc-mips.c (append_insn): Use ISA-encoded addresses in MIPS16
1571 * configure.in: Add ppc-*-lynxos*. Update i386-*-lynxos* to ELF.
1572 * configure: Regenerate.
1576 * dw2gencfi.c (output_cie): Handle dwarf3 format CIE entries.
1580 * configure.in (or32-*-rtems*): Switch to elf.
1581 * configure: Regenerate.
1585 * po/fr.po: Updated French translation.
1589 * doc/as.texinfo (Section): Document G and T flags to .section
1590 directive. Document the extra arguments that the G flag
1591 requires. Document the #tls flag.
1595 * subsegs.c (section_symbol): Create a new section symbol if
1596 the existing one doesn't match.
1597 * symbols.c (symbol_set_bfdsym): Don't reset BFD section symbol.
1601 * config/tc-mips.c (append_insn, mips_emit_delays): Extend -mfix-vr4120
1602 to cope with VR4181A errata MD(1) and MD(4).
1606 * NEWS: Mention .secrel32 for pe[i]-i386.
1610 * config/tc-frv.h (MAX_MEM_FOR_RS_ALIGN_CODE): New.
1611 (HANDLE_ALIGN): New.
1615 * Makefile.am (DIST_SUBDIRS): Define.
1616 * aclocal.m4: Regenerate with automake 1.8.4.
1617 * Makefile.in: Likewise.
1618 * doc/Makefile.in: Likewise.
1622 * config/tc-ia64.c (dot_serialize): Declare.
1623 (dot_serialize): New function.
1624 (md_pseudo_table): Add ".serialize.data" and
1625 ".serialize.instruction" directives.
1629 * messages (as_internal_value_out_of_range): Print a message about
1630 a value being out of range. Be consistent about whether the
1631 values are printed in decimal or hexadecimal.
1632 (as_warn_value_out_of_range): Generate a warning message about an
1634 (as_bad_value_out_of_range): Generate an error message about an
1636 * as.h: Prototype the new functions.
1637 * config/tc-alpha.c (insert_operand): Use new function.
1638 * config/tc-arc.c (arc_insert_operand): Likewise.
1639 * config/tc-mn10200.c (mn10200_insert_operand): Likewise.
1640 * config/tc-mn10300.c (mn10300_insert_operand): Likewise.
1641 * config/tc-ppc.c (ppc_insert_operand): Likewise.
1642 * config/tc-s390.c (s390_insert_operand): Likewise.
1643 * config/tc-v850.c (v850_insert_operand): Likewise.
1647 * configure.in: Set em=linux for frv-*-*linux*.
1648 * configure: Rebuilt.
1649 * config/tc-frv.h (TARGET_FORMAT): Use elf32-frvfdpic if...
1650 (frv_md_fdpic_enabled): New.
1651 * config/tc-frv.c (frv_md_fdpic_enabled): New.
1652 (DEFAULT_FDPIC): New.
1653 (frv_flags): Use DEFAULT_FDPIC.
1654 (frv_pic_flag): Likewise.
1655 (OPTION_NOPIC): New.
1656 (md_longopts): Add -mnopic.
1657 (md_parse_option): Handle it.
1658 (md_show_usage): Add -mfdpic and -mnopic.
1662 * config/tc-m68k.c: Add find_cf_chip to print list of valid
1663 chips for invalid coldfire instructions, rename selectors
1664 for ColdFire sub-variants, add 521x,5249,547x,548x and aliases,
1665 add current_chip to track which chip is referred to(including save/restore),
1666 use current_chip to select control registers, not current_arch.
1667 (md_show_usage): Add new chips.
1668 * doc/c-m68k.texi: Document new command line switches.
1672 * tc-s390.h (md_do_align, HANDLE_ALIGN): Remove.
1673 (NOP_OPCODE): Define.
1674 (s390_align_code): Remove prototype.
1675 * tc-s390.c (s390_align_code): Remove.
1679 * config/tc-ia64.c (make_unw_section_name): Removed.
1680 (start_unwind_section): New function.
1681 (generate_unwind_image): Take const segT instead of const
1683 (dot_handlerdata): Adjusted.
1684 (dot_endp): Likewise.
1688 * config/obj-elf.c (obj_elf_change_section): Allow the
1689 ".note.GNU-stack" section has SHF_EXECINSTR.
1693 * config/obj-elf.c (get_section): Return bfd_boolean.
1694 (obj_elf_change_section): Call bfd_get_section_by_name_if
1695 instead of bfd_map_over_sections.
1699 * config/obj-elf.c (get_section): New function.
1700 (obj_elf_change_section): Support multiple sections with same
1705 * config/tc-arm.c (create_register_alias): Fix typo checking for
1706 case sensitive register aliases.
1707 (co_proc_number): Use error message string in all_reg_maps[]
1709 (cp_reg_required_here): Likewise.
1710 (fp_reg_required_here): Likewise.
1714 * dwarf2dbg.c (dwarf2_finish): Add SEC_DEBUGGING to section flags.
1718 * config/tc-mips.c (HAVE_32BIT_ADDRESSES, append_insn, macro_build)
1719 (load_address, macro, mips_ip, md_parse_option)
1720 (mips_force_relocation, mips_validate_fix, md_apply_fix3)
1721 (s_change_sec, pic_need_relax, tc_gen_reloc): Remove all
1722 embedded-PIC handling, and update comments.
1723 (SWITCH_TABLE): Remove.
1724 * config/tc-mips.h (DIFF_EXPR_OK): Delete.
1725 (enum mips_pic_level): Remove EMBEDDED_PIC.
1726 (EXTERN_FORCE_RELOC): Remove embedded-PIC handling.
1727 (TC_FORCE_RELOCATION): Update comment.
1728 * ecoff.c (ecoff_build_lineno): Add comment about some code that
1729 might be safe to remove now that MIPS embedded-PIC is gone.
1733 * config/obj-som.c (obj_som_init_stab_section): Add new arguments in
1734 call to obj_set_subsection_attributes.
1735 (obj_som_init_stab_section): Likewise.
1736 * config/tc-hppa.c (default_subspace_dict): Add comdat field.
1737 (pa_def_subspaces): Provide comdat default.
1738 (pa_subspace): Handle new "comdat" parameter. Set SEC_LINK_ONCE and
1739 not SEC_IS_COMMON if section is comdat, common or dup_common. Update
1740 calls to create_new_subspace and update_subspace to pass comdat flag.
1741 (create_new_subspace, update_subspace): Add new comdat argument. Use
1742 it in calls to obj_set_subsection_attributes.
1743 * doc/c-hppa.texi (.subspa, .nsubspa): Document new comdat parameter
1744 and use of comdat, common and dup_comm parameters.
1748 * config/obj-elf.c (obj_elf_change_section): Check if the old
1749 group name is NULL before comparison.
1753 * config/tc-mips.h (mips_dwarf2_addr_size): Prototype.
1757 * config/tc-mips.c (s_mipsset): Set default CPU type for .set mipsN.
1761 * config/tc-mips.c (md_longopts): Remove -membedded-pic option.
1762 (OPTION_MEMBEDDED_PIC): Remove.
1763 (OPTION_TRAP, OPTION_BREAK, OPTION_EB, OPTION_EL)
1764 (OPTION_FP32, OPTION_GP32, OPTION_CONSTRUCT_FLOATS)
1765 (OPTION_NO_CONSTRUCT_FLOATS, OPTIONS_FP64, OPTION_GP64)
1766 (OPTION_RELAX_BRANCH, OPTION_NO_RELAX_BRANCH)
1767 (OPTION_ELF_BASE): Renumber.
1768 (md_parse_option): Remove OPTION_MEMBEDDED_PIC handling.
1769 (md_show_usage): Remove mention of -membedded-pic.
1770 * doc/as.texinfo: Remove mention of -membedded-pic.
1774 * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Remove.
1775 * config/tc-mips.c (RDATA_SECTION_NAME, mips_target_format): Remove
1777 (md_begin, mips_ip, md_parse_option, s_change_sec, s_option,
1778 s_abicalls, nopic_need_relax, tc_gen_reloc): Remove uses of
1779 USE_GLOBAL_POINTER_OPT.
1783 * config/tc-mips.c (macro): One more use of load_delay_nop.
1787 * config/tc-mips.c (load_delay_nop): New function.
1788 (load_address, macro): Use load_delay_nop() to build a nop
1789 which can be omitted with gpr_interlocks.
1793 * config/tc-mips.c (hilo_interlocks, gpr_interlocks,
1794 cop_interlocks): Remove superfluous CPU entries.
1798 * config/tc-arm.c (mav_parse_offset): Value must be multiple of 4.
1802 * NEWS: Added support for EMAC instructions and MAC/EMAC
1804 * config/m68k-parse.h: Add ACC[123], ACCEXT{01,23}, MAC/EMAC
1805 scale factor tokens, trailing_ampersand to mark mask addressing
1806 for MAC/EMAC instructions.
1807 * config/m68k-parse.y: Add options_ampersand clause, '<<',
1809 (yylex): Handle '>', '<', and '&' following '+'.
1810 * config/tc-m68k.c: Set mcfmac/mcfemac on appropriate ColdFire
1811 architectures in archs[].
1812 (m68k-ip): Add '4', 'e', 'g', 'i', cases to handle mask addressing
1813 for MAC/EMAC instructions, ACC[0123], ACCEXT{01,23}, and '<<'/'>>'
1815 (m68k_ip): Handle trailing '&' on MAC/EMAC insns.
1816 (install_operand): Fix 'n' case, Add 'F', 'f', 'G', 'H', 'I', ']'
1818 Add EMAC operands to init_table[].
1822 * config/tc-arm.h (enum mstate): Move here, add MAP_UNDEFINED
1824 (TC_SEGMENT_INFO_TYPE): Define to enum mstate.
1825 * config/tc-arm.c (enum mstate): Delete from here.
1826 (mapping_state): Remove the static mapstate variable and instead
1827 store the state in the segment. This allows a per-section mapping
1828 state. Handle and ignore MAP_UNDEFINED states.
1829 (arm_elf_change_section): Get the current mapping state from the
1831 (s_ltorg): Set the mapping state to MAP_DATA.
1832 (arm_cleanup): Use arm_elf_change_section to get the mapping state
1833 for each pool as it is emitted.
1837 * config/tc-arm.h: Formatting tidy ups.
1841 * NEWS: Note that MIPS -membedded-pic option is deprecated.
1845 * config/tc-i386.h [TE_PE] (TC_CONS_FIX_NEW): Define.
1846 * config/tc-i386.c (md_pseudo_table) [TE_PE]: Add "secrel32".
1847 [TE_PE] (O_secrel): Define.
1848 [TE_PE] (x86_pe_cons_fix_new): New.
1849 [TE_PE] (pe_directive_secrel): Likewise.
1850 (tc_gen_reloc) [TE_PE]: Support BFD_RELOC_32_SECREL.
1854 * config/tc-mips.c (mips_dwarf2_addr_size): Revert part
1855 of previous patch for fix in gcc.
1859 * config/tc-xtensa.c (xg_assembler_literal): Fix a typo.
1863 * read.c (do_align): Call md_flush_pending_output, if defined.
1867 * expr.c (operand): Correct checks for ++ and --.
1871 * config/tc-generic.c: Add some comments.
1875 * doc/c-mips.texi (-m{no-,}fix-vr4120): Renamed from
1876 -{no-}mfix-vr4122-bugs.
1877 * config/tc-mips.c (mips_fix_vr4120): Renamed from mips_fix_4122_bugs.
1878 (append_insn, mips_emit_delays): Update accordingly.
1879 (OPTION_FIX_VR4120, OPTION_NO_FIX_VR4120): Renamed from *VR4122.
1880 (md_longopts): Change -{no-,}mfix-vr4122-bugs to -m{no-,}fix-vr4120.
1881 (md_parse_option): Update after above changes.
1882 (md_show_usage): Add -mfix-vr4120.
1886 * doc/as.texinfo (Sub-Sections): Conditionalize COFF-specific use
1887 of .section directive; add a reference to the ELF .subsection
1892 * config/tc-m32r.c (md_assemble): Fixed infinite loop bug
1897 * Makefile.am: Remove mips from aout targets.
1898 * Makefile.in: Regenerate.
1899 * configure.in: Remove mips-dec-bsd* target.
1900 * configure: Regenerate.
1905 * config/tc-ppc.c (ppc_elf_suffix): Add valid32 and valid64 fields
1906 to struct map_bfd. Adjust MAP macro, and define MAP32, MAP64.
1907 Update "mapping". Restrict some @ modifiers to 32 bit.
1912 * config/tc-sh.c (dont_adjust_reloc_32): New variable.
1913 (sh_fix_adjustable): Avoid adjusting BFD_RELOC_32 when
1914 dont_adjust_reloc_32 is set.
1915 (md_longopts): Add option -renesas.
1916 (md_parse_option, md_show_usage): Likewise.
1917 * doc/c-sh.texi: Likewise.
1921 * config/tc-dlx.c (md_assemble): set fx_no_overflow flag for
1922 hi16 and lo16 fixS structs.
1923 (md_assemble): generate bit_fixS for RELOC_DLX_LO16 in
1924 exactly the same way as for RELOC_DLX_REL16.
1925 (machine_ip): properly respect LO flag in the_insn and
1926 output RELOC_DLX_LO16 rather than RELOC_DLX_16.
1927 (md_apply_fix3): apply RELOC_DLX_LO16.
1931 Remove long-obsolete MPW support.
1932 * mpw-config.in, mpw-make.sed, mac-as.r: Remove files.
1933 * configure.in: Remove mention of ppc-*-mpw* config.
1934 * configure.in: Likewise.
1938 * config/tc-arm.c (meabi_flags): Make its use conditional upon
1939 OBJ_ELF being defined.
1943 * config/obj-aout.c (obj_aout_type): Remove #ifdef BFD_ASSEMBLER code.
1947 * config/tc-arm.c (meabi_flags): New variable.
1948 (arm_parse_eabi): New function.
1949 (md_begin): Set flags for EABI v3.
1951 (arm_long_opts): Add meabi.
1952 * doc/as.texinf <ARM>: Document -meabi.
1953 * doc/c-arm.texi: Ditto.
1957 * config/tc-xtensa.c (xtensa_post_relax_hook): Create literal
1958 tables even when use_literal_section flag is not set.
1962 * config/tc-sh.c: Remove trailing whitespace.
1966 * doc/c-cris.texi (CRIS-Opts): Document --no-mul-bug-abort,
1967 --mul-bug-abort and the default behavior.
1968 * config/tc-cris.c (cris_insn_kind): New member CRIS_INSN_MUL.
1969 (err_for_dangerous_mul_placement): New variable.
1970 (STATE_MUL, OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): New
1972 (md_cris_relax_table): Have placeholder for STATE_MUL.
1973 (md_longopts): New options --mul-bug-abort and --no-mul-bug-abort.
1974 (cris_relax_frag) <case ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: New
1976 (md_estimate_size_before_relax) <case ENCODE_RELAX (STATE_MUL,
1977 STATE_BYTE)>: Ditto.
1978 (md_convert_frag) <ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: Check
1979 alignment and position of this frag, emit error message if
1981 (md_assemble): For a multiply insn and when checking it,
1982 transform the current frag into a special frag for that purpose.
1983 (md_parse_option) <case OPTION_MULBUG_ABORT_OFF, case
1984 OPTION_MULBUG_ABORT_ON>: Handle new options.
1988 * config/tc-xtensa.c (mark_literal_frags): New function.
1989 (xtensa_move_literals): Call mark_literal_frags for all literal
1990 segments, including init and fini literal segments.
1991 (xtensa_post_relax_hook): Swap use of xt_insn_sec and xt_literal_sec.
1995 * tc-hppa.c (cons_fix_new_hppa): Check for PC relative base type.
1996 (pa_comm): Set BSF_OBJECT in symbol flags.
2000 * Makefile.am: Run "make dep-am".
2001 * Makefile.in: Regenerate.
2002 * doc/Makefile.in: Regenerate.
2003 * config.in: Regenerate.
2004 * po/gas.pot: Regenerate.
2008 * read.c (read_a_source_file): Use demand_empty_rest_of_line.
2009 (demand_empty_rest_of_line): Issue an error here.
2010 (ignore_rest_of_line): Silently skip to end.
2011 (demand_copy_string): Issue an error, not warning.
2013 * config/obj-elf.c (obj_elf_section_name): Likewise.
2014 (obj_elf_section): Likewise.
2015 * config/tc-arc.c (arc_extoper): Remove bogus NULL checks.
2016 (arc_extinst): Likewise.
2017 * config/tc-ia64.c (dot_saveb): Use demand_empty_rest_of_line.
2018 (dot_spill): Likewise.
2019 (dot_unwabi): Likewise.
2020 (dot_prologue): Likewise.
2022 * expr.c (operand): Reject ++ and --.
2023 (operator): Likewise.
2027 * config/tc-sh.c: Include dw2gencfi.h.
2028 (sh_cfi_frame_initial_instructions): New function.
2029 (sh_regname_to_dw2regnum): Likewise.
2030 * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Move to the end of
2032 (TARGET_USE_CFIPOP): Define.
2033 (tc_cfi_frame_initial_instructions): Likewise.
2034 (tc_regname_to_dw2regnum): Likewise.
2035 (DWARF2_DEFAULT_RETURN_COLUMN, DWARF2_CIE_DATA_ALIGNMENT): Likewise.
2036 * Makefile.am: Update dependencies.
2037 * Makefile.in: Regenerate.
2041 * configure.in: Switch sh-*-rtems* to ELF. Add sh-*-rtemscoff*.
2042 * configure: Regenerate.
2046 * read.c (s_leb128): Call md_flush_pending_output.
2050 * config/tc-i386.c (output_insn): Handle PadLock instructions.
2051 * config/tc-i386.h (CpuPadLock): New define.
2052 (CpuUnknownFlags): Added CpuPadLock.
2056 * doc/c-hppa.texi (HPPA Directives): Fix typo.
2060 * dw2gencfi.c (output_cie): Align length to 4 byte boundary.
2061 (cfi_finish): Likewise for fde.
2065 * config/tc-ia64.c (md_assemble): Properly handle NULL
2067 (ia64_handle_align): Don't abort if failed to add a stop bit.
2071 * Makefile.in: Regenerated.
2072 * aclocal.m4: Likewise.
2073 * configure: Likewise.
2074 * doc/Makefile.in: Likewise.
2078 * config/tc-ia64.c (dot_align): New.
2079 (ia64_do_align): Make it static.
2080 (md_pseudo_table): Use "dot_align" for "align".
2081 (ia64_md_do_align): Don't set align_frag here.
2082 (ia64_handle_align): Add a stop bit to the previous bundle if
2085 * config/tc-ia64.h (ia64_do_align): Removed.
2089 * config/tc-sh.c (md_parse_option): Add -isa=sh4-nofpu and
2090 -isa=sh4-nommu-nofpu options. Adjust help messages accordingly.
2091 (sh_elf_final_processing): Output BFD type sh4_nofpu if that is
2092 the most general type or the user specifically requested it.
2093 (md_assemble): Add a new error message for when an instruction
2094 is understood, but is not allowed due to an -isa option.
2098 * config/tc-ia64.c (align_frag): New.
2099 (md_assemble): Set the tc_frag_data field in align_frag for
2100 IA64_OPCODE_FIRST instructions.
2101 (ia64_md_do_align): Set align_frag.
2102 (ia64_handle_align): Add a stop bit if needed.
2104 * config/tc-ia64.h (TC_FRAG_TYPE): New.
2105 (TC_FRAG_INIT): New.
2109 * config/tc-frv.c (fr400_audio): New variable.
2110 (md_parse_option, md_show_usage): Add -mcpu=fr405 and -mcpu=fr450.
2111 (md_parse_option): Set fr400_audio for -mcpu=fr400 and -mcpu=fr405.
2112 (target_implements_insn_p): New function.
2113 (md_assemble): Report an error if the processor doesn't implement
2118 * config/tc-m32r.c (md_longopts): Added -no-bitinst option.
2119 (md_parse_option): Ditto.
2120 (OPTION_NO_SPECIAL_M32R): Added.
2121 (md_show_usage): Document it.
2122 (enable_speial_m32r): Changed a default value from 0 to 1.
2123 * doc/c-m32r.texi: Document the -no-bitinst option.
2127 * config/tc-sh.c (get_operand): Revert previous delta.
2128 (tc_gen_reloc): Check for an unknown reloc type before processing
2133 * config/tc-s390.c (s390_insn): Correct range check for opcode in
2134 .insn pseudo operation.
2138 * config/tc-sh.c (get_operand): In case of #Imm, check has been
2139 added for wrong syntax.
2143 * config/tc-mips.c (mips_dwarf2_addr_size): New.
2144 * config/tc-mips.h (DWARF2_ADDR_SIZE): Use.
2148 * config/tc-sh.c (build_Mytes): Add REG_N_D and REG_N_B01
2149 nibble types to assembler.
2153 * config/tc-iq2000.c: Add missing \n\ in multiline string literal.
2157 * config/tc-ia64.c (slot_index): New arg before_relax. Use instead of
2159 (fixup_unw_records): New arg before_relax. Pass to slot_index.
2160 (ia64_estimate_size_before_relax): New.
2161 (ia64_convert_frag): Pass 0 to fixup_unw_records. Add comment.
2162 (generate_unwind_image): Pass 1 to fixup_unw_records.
2163 * config/tc-ia64.h (ia64_estimate_size_before_relax): Declare.
2164 (md_estimate_size_before_relax): Call ia64_estimate_size_before_relax.
2168 * stabs.c (generate_asm_file): Avoid warning about use of
2169 uninitialized variable.
2173 * config/tc-ia64.c (ia64_flush_insns): In addition to prologue,
2174 body, and endp, allow unwind records which do not have a "t"
2175 (time/instruction) field.
2179 * config/tc-arm.c (do_mav_dspsc_1): Correct offset of CRn.
2180 (do_mav_dspsc_2): Likewise.
2181 Fix accumulator registers move opcodes.
2186 * dwarf2dbg.c (get_filenum): Do not read beyond allocated memory.
2190 * config/tc-ia64.h (ia64_frob_symbol): New declaration.
2191 (tc_frob_symbol): New macro definition.
2192 * config/tc-ia64.c (ia64_frob_symbol): New routine.
2196 * config/tc-arm.c (md_begin): Mark .note.gnu.arm.ident as
2201 * read.h (IGNORE_OPCODE_CASE): Do not define. Replace with ...
2202 (TC_CASE_SENSITIVE): ... this.
2203 * read.c: Replace IGNORE_OPCODE_CASE with TC_CASE_SENSITIVE.
2204 * doc/internals.texi (TC_CASE_SENSITIVE): Document.
2208 * config/tc-ia64.c (dot_endp): Delete call to output_endp.
2209 (generate_unwind_image): Re-add it here.
2213 * dwarf2dbg.c (DWARF2_ADDR_SIZE): Remove trailing ';'
2214 * read.h (SKIP_WHITESPACE): Turn into an expression.
2215 * read.c (read_a_source_file): A pseudo is removed by having a
2220 * config/tc-ia64.c (output_endp): New.
2221 (count_bits): Delete.
2222 (ia64_flush_insns, process_one_record, optimize_unw_records): Handle
2223 endp unwind records.
2224 (fixup_unw_records): Handle endp unwind records. Delete code for
2225 shortening prologue regions not followed by a body record.
2226 (dot_endp): Call add_unwind_entry to emit endp unwind record.
2227 * config/tc-ia64.h (unw_record_type): Add endp.
2231 * config/tc-ia64.c (ia64_convert_frag): Call md_number_to_chars to
2232 fill padding bytes with zeroes.
2233 (emit_one_bundle): New locals last_ptr, end_ptr. Rewrite code that
2234 sets unwind_record slot_number and slot_frag fields.
2238 * config/tc-mips.c (add_got_offset_hilo): New function.
2239 (macro): Use load_register() and add_got_offset_hilo() to load
2240 constants instead of hardcoding code sequences throughout.
2244 * config/tc-ia64.c (emit_one_bundle): Add proper indentation.
2248 * config/tc-m68k.h (EXTERN_FORCE_RELOC): Handle m68k-uclinux specially,
2250 * config/tc-m68k.c (RELAXABLE_SYMBOL): Use EXTERN_FORCE_RELOC instead
2251 of hard-coded test for TARGET_OS=elf.
2255 * config/tc-mips.c (hilo_interlocks): Change definition
2256 so that MIPS32, MIPS64 and later ISAs are included, along with
2257 the already-included machines. Update comments.
2261 * config/tc-arm.c (tc_gen_reloc): Improve error message for
2262 undefined local labels.
2266 * config/tc-mips.c (load_address, macro): Update comments about
2267 NewABI GP relaxation.
2271 * config/tc-mips.c (macro_build): Remove place and counter arguments.
2272 (mips_build_lui, macro_build_ldst_constoffset): Likewise.
2273 (mips16_macro_build, macro_build_jalr): Remove counter argument.
2274 (set_at, load_register, load_address, move_register): Likewise.
2275 (load_got_offset, add_got_offset): Likewise.
2276 Update all calls and tidy accordingly.
2280 * config/tc-mips.c (RELAX_ENCODE): Remove WARN argument.
2281 (RELAX_FIRST, RELAX_SECOND): Turn into 8-bit quantities.
2282 (RELAX_USE_SECOND): Bump to 0x10000.
2283 (RELAX_SECOND_LONGER, RELAX_NOMACRO, RELAX_DELAY_SLOT): New flags.
2284 (mips_macro_warning): New variable.
2285 (md_assemble): Wrap macro expansion in macro_start() and macro_end().
2286 (s_cpload, s_cpsetup, s_cprestore, s_cpreturn): Likewise.
2287 (relax_close_frag): Set mips_macro_warning.first_frag. Adjust use
2289 (append_insn): Update mips_macro_warning.sizes.
2290 (macro_start, macro_warning, macro_end): New functions.
2291 (macro_build): Don't emit warnings here.
2292 (macro_build_lui, md_estimate_size_before_relax): ...or here.
2293 (md_convert_frag): Check for cases where one macro alternative
2294 needs a warning and the other doesn't. Emit a warning if the
2295 longer sequence was chosen.
2299 * config/tc-mips.h (tc_frag_data_type, TC_FRAG_TYPE): Remove.
2300 * config/tc-mips.c (RELAX_ENCODE): Take three arguments: the size of
2301 the first sequence, the size of the second sequence, and a flag
2302 that says whether we should warn.
2303 (RELAX_OLD, RELAX_NEW, RELAX_RELOC[123]): Delete.
2304 (RELAX_FIRST, RELAX_SECOND): New.
2305 (mips_relax): New variable.
2306 (relax_close_frag, relax_start, relax_switch, relax_end): New fns.
2307 (append_insn): Remove "place" argument. Use mips_relax.sequence
2308 rather than "place" to check whether we're expanding the second
2309 alternative of a relaxable macro. Remove redundant check for
2310 branch relaxation. If generating a normal insn, and there
2311 is not enough room in the current frag, call relax_close_frag()
2312 to close it. Update mips_relax.sizes[]. Emit fixups for the
2313 second version of a relaxable macro. Record the first relaxable
2314 fixup in mips_relax. Remove tc_gen_reloc workaround.
2315 (macro_build): Remove all uses of "place". Use mips_relax.sequence
2316 in the same way as in append_insn.
2317 (mips16_macro_build): Remove "place" argument.
2318 (macro_build_lui): As for macro_build. Don't drop the add_symbol
2319 when generating the second version of a relaxable macro.
2320 (load_got_offset, add_got_offset): New functions.
2321 (load_address, macro): Use new relaxation machinery. Remove
2322 tc_gen_reloc workarounds.
2323 (md_estimate_size_before_relax): Set RELAX_USE_SECOND if the second
2324 version of a relaxable macro is needed. Return -RELAX_SECOND if the
2325 first version is needed.
2326 (tc_gen_reloc): Remove relaxation handling.
2327 (md_convert_frag): Go through the fixups for a relaxable macro and
2328 mark those that belong to the unneeded alternative as done. If the
2329 second alternative is needed, adjust the fixup addresses to account
2330 for the deleted first alternative.
2334 * frags.h (frag_room): Declare.
2335 * frags.c (frag_room): New function.
2336 * doc/internals.texi: Document it.
2340 * config/tc-mips.c (append_insn): Don't do r3900 interlock
2341 optimization for -mtune=r3900, as this will break on other CPUs.
2345 * config/tc-m68hc11.c (build_indexed_byte): movb and movw cannot
2346 be relaxed, use fixup.
2347 (md_apply_fix3): Use 5 bit reloc from movb and movw fixup.
2351 * config/tc-sparc.c (sparc_ip): Disallow %f32-%f63 for single
2356 * config/tc-mips.c (append_insn): Properly detect variant frags
2357 that preclude swapping of relaxed branches. Correctly swap
2358 instructions between frags when dealing with relaxed branches.
2362 * acinclude.m4: Quote names of macros to be defined by AC_DEFUN
2364 * aclocal.m4: Regenerate.
2365 * configure: Regenerate.
2369 * config/tc-h8300.c (build_bytes): Apply relaxation to bit
2374 * config/tc-mips.c (macro_build_jalr): When adding an R_MIPS_JALR
2375 reloc, reserve space for the delay slot as well as the jalr itself.
2379 * config/tc-arm.c (do_vfp_reg2_from_sp2): Rename from do_vfp_sp_reg2.
2380 (do_vfp_sp2_from_reg2): New function.
2382 (do_vfp_dp_from_reg2): Check return values properly.
2386 * config/tc-mips.c (warn_nops): Remove static variable.
2387 (macro): Remove test of warn_nops.
2388 (md_shortops): Remove 'n'.
2389 (md_parse_option): Remove 'n' case.
2390 (md_show_usage): Remove -n.
2391 * doc/as.texinfo (Overview): Remove MIPS -n option.
2392 * doc/c-mips.texi (MIPS Opts): Remove mention -n.
2393 * NEWS: Mention removal of MIPS -n option.
2395 * config/tc-mips.c (ISA_HAS_COPROC_DELAYS): Remove.
2396 (cop_interlocks): Check ISA level.
2397 (cop_mem_interlocks): Define.
2398 (reg_needs_delay): Check cop_interlocks rather than
2399 ISA_HAS_COPROC_DELAYS.
2400 (append_insn): Likewise. Use cop_mem_interlocks rather than
2401 directly checking mips_opts.isa.
2402 (mips_emit_delays): Likewise.
2406 * config/tc-ia64.c (unwind): Move next_slot_number and
2407 next_slot_frag to ...
2408 (unw_rec_list): Here.
2409 (free_list_records): Removed.
2410 (output_unw_records): Likewise.
2411 (generate_unwind_image): Make it void.
2412 (alloc_record): Initialize next_slot_number and next_slot_frag.
2413 (slot_index): Take .org, .space and .align into account.
2414 (fixup_unw_records): Don't set slot_number to 0. Use
2415 list->next_slot_number and list->next_slot_frag instead of
2416 unwind.next_slot_number and unwind.next_slot_frag.
2417 (ia64_convert_frag): New.
2418 (generate_unwind_image): Generate a rs_machine_dependent frag
2420 (emit_one_bundle): Use list->next_slot_number and
2421 list->next_slot_frag instead of unwind.next_slot_number and
2422 unwind.next_slot_frag.
2424 * config/tc-ia64.h (md_convert_frag): Defined as
2426 (md_estimate_size_before_relax): Defined as (f)->fr_var.
2431 * config/tc-frv.h (md_apply_fix3): Don't define.
2432 * config/tc-frv.c (md_apply_fix3): New. Shift/truncate %hi/%lo
2434 * config/tc-frv.h (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
2436 * config/tc-frv.c (line_separator_chars): Add `!'.
2438 * config/tc-frv.c (md_assemble): Clear insn upfront.
2440 * config/tc-frv.c (OPTION_FDPIC): New macro.
2441 (md_longopts): Add mfdpic.
2442 (md_parse_option): Handle it.
2444 * config/tc-frv.c (md_cgen_lookup_reloc) <FRV_OPERAND_D12,
2445 FRV_OPERAND_S12>: Use reloc type encoded in fix-up.
2446 (frv_pic_ptr): Parse funcdesc.
2450 * doc/as.texinfo: Let texi2pod parse asconfig.texi and
2451 gasver.texi. Remove duplicate symbol definitions for texi2pod.
2455 * Makefile.am (Makefile): Move the dependency on
2456 $(BFDDIR)/configure.in to...
2457 (CONFIG_STATUS_DEPENDENCIES): ... here.
2458 (AUTOMAKE_OPTIONS): Require automake 1.8.
2459 * Makefile.in: Regenerate.
2460 * doc/Makefile.am (BASEDIR, BFDDIR): Define.
2461 (CONFIG_STATUS_DEPENDENCIES): Add a dependency on
2462 $(BFDDIR)/configure.in here as well.
2463 * doc/Makefile.in: Regenerate.
2467 * Makefile.am (install, install-info, RECURSIVE_TARGETS): Remove.
2468 * Makefile.in: Regenerate.
2469 * aclocal.m4: Regenerate.
2470 * doc/Makefile.am (install, install-info): Remove.
2471 (install-data-local): A new hook for install-info.
2472 (AUTOMAKE_OPTIONS): Require automake 1.8.
2473 * doc/Makefile.in: Regenerate.
2477 * doc/c-sh.texi: Update description about floating point behavior
2482 * configure.in: Add m68k-uClinux target.
2483 * configure: Regenerate.
2485 For older changes see ChangeLog-0203
2491 version-control: never