4 * config/tc-tic80.c (md_pseudo_table): Add "sect" and "section"
6 * config/tc-tic80.c (md_begin): Declare external variable
7 coff_flags and insert an F_AR32WR bit into it.
12 * read.c (equals): Add reassign parameter. Change all callers.
13 * read.h (equals): Update declaration.
17 * config/tc-mips.c (mips16_extended_frag): Don't assume that we
18 can rely on the frag address to determine whether a frag is
23 * write.h (LOCAL_LABEL): Only define if not BFD_ASSEMBLER.
24 (S_LOCAL_NAME): Likewise.
25 (FAKE_LABEL_NAME): Define unconditionally.
26 * symbols.c (colon): Call bfd_is_local_label, not LOCAL_LABEL, if
28 (S_IS_LOCAL): Call bfd_is_local_label_name, not LOCAL_LABEL.
29 * config/tc-*.h: Only define LOCAL_LABEL if not BFD_ASSEMBLER.
30 Don't define FAKE_LABEL_NAME.
31 * config/te-ic960.h: Likewise.
32 * config/tc-mips.h (tc_frob_file_before_adjust): Define.
33 (mips_frob_file_before_adjust): Declare.
34 * config/tc-mips.c (mips_frob_file_before_adjust): New function.
35 (mips_local_label): Remove.
37 * config/te-sco386.h: Remove; not used.
42 * config/tc-tic80 (md_pseudo_table): Add align pseudo op to do
43 byte alignment rather than power-of-two alignment that is the
49 * config/tc-m68k.c (md_assemble): Handle a reloc width of 'W'.n
51 * gasp.c (hash_add_to_string_table): Correct misspelling in error
52 message, and add newline.
53 (process_file): Don't process assignments in the label if this is
54 a equ or assign pseudo-op.
55 (process_pseudo_op): Swap first argument to do_assign for K_ASSIGN
56 and K_EQU, to match documentation.
60 * config/obj-coff.c (obj_coff_section): Add 'r' section attribute
61 to denote read-only data sections.
65 * config/obj-elf.c (obj_elf_common): Set BSF_OBJECT in flags.
66 * config/tc-sparc.c (s_common): Likewise, if BFD_ASSEMBLER.
68 * expr.c (operand): Simplify 0b handling. Don't treat 0b as a
69 binary number if the next character is '+' or '-'.
74 * config/tc-tic80.c (FLT_CHARS): Change from "dD" to "fF".
75 (find_opcode): Match operands that can be floats.
76 (build_insn): Handle O_big (float) expressions and build
82 * configure.in (mips*-*-lnews*): New target, also make empty
83 emulation list for this target.
85 * tc-mips.c (ECOFF_LITTLE_FORMAT): Define.
86 (mips_target_format): Use.
87 * te-lnews.h: New file.
92 * config/tc-tic80.c (find_opcode, build_insn): Changes to match
93 operands with :m or :s modifiers and generate the right opcodes
99 * Makefile.in (itbl-parse.c itbl-parse.h): Use $(BISON) and
100 $(BISONFLAGS), not $(YACC) and $(YACCFLAGS).
104 * config/tc-m68k.c (instring): Useless local declaration of
105 crack_operand removed.
106 * expr.h (expressionS): Changed type of X_op field to operatorT if
112 * configure.in: Add i386coff and i386elf to emulation list.
113 * configure: Rebuild.
114 * as.c (i386coff, i386elf): Declare.
115 * obj.h (coff_format_ops): Declare.
116 * config/obj-coff.c (OBJ_HEADER): Define.
117 (coff_obj_symbol_new_hook): Rename from obj_symbol_new_hook.
118 (coff_obj_read_begin_hook): Rename from obj_read_begin_hook.
119 (obj_pseudo_table): Add "version".
120 (coff_pop_insert): New static function.
121 (coff_sec_sym_ok_for_reloc): New static function.
122 (no_func): New static function.
123 (coff_format_ops): New variable.
124 * config/obj-coff.h (coff_obj_symbol_new_hook): Declare.
125 (obj_symbol_new_hook): Define.
126 (coff_obj_read_begin_hook): Declare.
127 (obj_read_begin_hook): Define.
128 * config/tc-i386.h (i386_target_format): Declare.
129 * config/tc-i386.c: Check OBJ_MAYBE_ELF as well as OBJ_ELF; check
130 OUTPUT_FLAVOR when appropriate.
131 (i386_target_format): New function.
132 * Makefile.in (obj-coff.o): New target.
133 (e-i386coff.o, e-i386elf.o): New targets.
136 * config/tc-i960.h (TC_SYMFIELD_TYPE): Define if OBJ_COFF.
137 (_tc_get_bal_of_call): Don't declare.
138 (tc_get_bal_of_call): Declare as function, don't define as macro.
139 * config/tc-i960.c (tc_set_bal_of_call): If OBJ_COFF, store balP
140 in sy_tc field, not x_balntry field.
141 (tc_get_bal_of_call): Rename from _tc_get_bal_of_call. Change
142 return type to symbolS *. If OBJ_COFF, retrieve value from sy_tc
143 field, not x_balntry field.
145 * config/obj-elf.c (obj_elf_section): Permit a .note section to
146 have the SHF_ALLOC attribute.
148 * Makefile.in ($(OBJS)): Don't depend upon $(IT_HDRS).
149 (TARG_CPU_DEP_mips): Depend upon $(srcdir)/itbl-ops.h.
150 (itbl-lex.o): Depend upon itbl-parse.h.
152 * itbl-parse.y (yyerror): Change return type to int. Change to
153 use old style function declaration.
155 * Makefile.in (itbl-lex.o): Remove -Wall.
156 (itbl-parse.o): Likewise.
158 * cond.c (s_ifdef): If we should omit conditionals from listings,
160 (s_if, s_ifc, s_endif, s_else, s_ifeqs): Likewise.
161 * listing.c (list_info_struct): Add EDICT_NOLIST_NEXT.
162 (listing_listing): Handle EDICT_NOLIST_NEXT.
163 (listing_list): An argument of 2 means EDICT_NOLIST_NEXT.
164 * listing.h (LISTING_NOCOND): Define.
165 (LISTING_SKIP_COND): Define.
166 * as.c (show_usage): Mention c as a suboption of -a.
167 (parse_args): Handle c as a suboption of -a.
168 * doc/as.texinfo: Document -alc.
173 * config/tc-tic80.c (md_apply_fix): Handle R_ABS type fixups.
178 * doc/as.texinfo: Document .symver.
182 * config/tc-d10v.c: Change pre_defined_registers to
183 d10v_predefined_registers and reg_name_cnt to d10v_reg_name_cnt.
187 * config/obj-coff.c: Fix typo in comment section.
189 * config/tc-tic80.c (md_pseudo_table): Add entry for bss, which takes
190 an additional alignment argument.
191 (find_opcode): Allow O_symbol relocs for any 32 bit field, not just
193 (build_insn): Handle O_symbol relocs for any 32 bit field, not just
199 * Makefile.in: Remove dependancies on itbl-cpu.h.
200 * as.c: Define stubs for itbl_parse and itbl_init if HAVE_ITBL_CPU
205 * itbl-ops.h: Include as.h.
209 * as.c: Remove -t option.
210 * configure, configure.in: Move itbl-cpu.h to mips specific configure.
211 * itbl-ops.h: Include itbl-cpu.h only if HAVE_ITBL_CPU is defined.
212 * config/tc-mips.h: Define HAVE_ITBL_CPU.
216 * itbl-ops.c: Don't define DEBUG.
220 * Makefile.in: Update itbl-test.c to reflect its new location.
224 * itbl-ops.c: Add test for itbl_have_entries.
225 * config/tc-mips.c: Remove test for itbl_have_entries.
226 * config/tc-mips.h: Define tc_init_after_args to mips_init_after_args.
230 * Makefile.in (DISTSTUFF): Remove itbl-parse.y, itbl-lex.l, and
231 itbl-ops.c. Add itbl-parse.c and itbl-lex.c.
232 (LEX, LEXFLAGS): Define.
233 * itbl-ops.c (append_insns_as_macros): Remove bogus ASSERT.
237 * itbl-parse.y: Fix indentation mistakes from indent program.
238 * itbl-lex.l: Fix indentation mistakes from indent program.
239 * itbl-ops.h: Add include for ansidecl.h.
240 Add PARAMS around function arguments.
241 Add declaration for itbl_have_entries.
242 * itbl-ops.c: Add PARAMS around function arguments.
243 * Makefile.in: Add itbl build rules.
244 Add dependancies for itbl files to mips target.
245 * as.c: Add itbl support.
246 Add new option "--insttbl" for dynamically extending instruction set.
247 * as.h: Declare insttbl_file_name;
248 the name of file defining extensions to the basic instruction set
249 * configure.in, configure: Add itbl-parse.o, itbl-lex.o, and
250 itbl-ops.o to extra_objects for mips configuration.
251 Add include file link from itbl-cpu.h to
252 config/itbl-${target_cpu_type}.h.
253 * config/tc-mips.c: Allow copz instructions.
254 Add notes for future additions to the itbl support.
256 (macro): Call itbl_assemble to assemble itbl instructions.
257 See if an unknown register is specified in an itbl entry.
260 * doc/internals.texi (CPU backend): Fix typo in md_section_align
263 * config/tc-tic80.h (NEED_FX_R_TYPE): Define.
264 * config/tc-tic80.c (find_opcode): Add code to support O_symbol
266 (build_insn): Grab a frag early so we can use the address in
267 fixups. Take one's complement of BITNUM values before insertion
268 in opcode. Add code to support O_symbol operands.
269 (md_apply_fix): Replace unimplemented warning with implementation.
270 (md_pcrel_from): Ditto.
271 (tc_coff_fix2rtype): Ditto.
276 * config/tc-d30v.c (parallel_ok): New function.
277 * config/tc-d30v.h: Define TARGET_BYTES_BIG_ENDIAN.
279 * config/tc-d10v.c (md_pcrel_from_section): Return 0 if
280 relocation is in different section. Fixes PR11574.
284 * tc-mips.c (mips_ip): If configured for an embedded ELF system,
285 don't set the section alignment to 2**4.
289 * config/tc-m68k.c (line_comment_chars): Add '*'.
291 * app.c (LEX_IS_TWOCHAR_COMMENT_2ND): Don't define.
292 (do_scrub_begin): Don't set lex['*'].
293 (do_scrub_chars): When handling LEX_IS_TWOCHAR_COMMENT_1ST, don't
294 check for LEX_IS_TWOCHAR_COMMENT_2ND. Instead, just check for
297 * configure.in: Set em=svr4 for m68k-*-sysv4*.
298 * configure: Rebuild.
299 * config/te-svr4.h: New file.
300 * config/tc-m68k.c (m68k_comment_chars): Only include `#' if
305 * config/tc-mn10200.c (md_convert_frag): Create a fixup for the
306 short conditional branch around a long unconditional branch.
310 * config/obj-coff.c (obj_coff_ln [both versions]): Call
313 * config/tc-arm.c (fix_new_arm): Use make_expr_symbol to handle a
316 * symbols.c (resolve_symbol_value): If both left and right
317 operands are undefined, warn about both of them.
322 * config/obj-elf.c (elf_pseudo_table): Add "symver".
323 (obj_elf_symver): New static function.
324 * config/obj-elf.h (OBJ_COPY_SYMBOL_ATTRIBUTES): Copy the st_other
327 * write.c (relax_segment): Make type and printf format agree.
329 * read.c (get_line_sb): Don't end the line on a semicolon inside a
335 * config/tc-d30v.c, config/tc-d30v.h: New files.
337 * configure: Rebuilt.
339 * configure.in: Add case for d30v.
344 * config/tc-alpha.h (md_operand): Define with a null expansion,
345 like all the other targets.
346 * doc/internals.texi (CPU backend): Add missing word in
347 md_flush_pending_output description. Fix typo in md_convert_frag
350 * config/tc-tic80: Minor comment additions/changes.
355 * config/tc-m68k.c (LOCAL_LABEL): Macro redefined if TE_DELTA.
356 (tc_canonicalize_symbol_name): Macro defined if TE_DELTA.
357 * config/obj-coff.c (obj_coff_def): Use
358 tc_canonicalize_symbol_name if defined.
359 (obj_coff_tag, obj_coff_val): Likewise.
360 * expr.c (operand): Reject '~' as operator if is_name_beginner.
365 does not actually work, though:
366 * configure.in (i386-sequent-bsd*): New target.
367 * configure: Rebuild.
368 * config/tc-dynix.h: New file.
369 * config/tc-i386.h: Define TARGET_FORMAT if TE_DYNIX.
371 * read.c (do_align): Add max parameter. Change all callers.
372 Remove useless static variables.
373 (s_align): New static function. Do common portion of
374 s_align_bytes and s_align_ptwo.
375 (s_align_bytes, s_align_ptwo): Just call s_align.
376 * frags.c (frag_align): Add max parameter. Change all callers.
377 (frag_align_pattern): Likewise.
378 * frags.h (frag_align, frag_align_pattern): Update declarations.
379 * write.c (relax_segment): Limit alignment change to fr_subtype.
380 Fix some types to be addressT.
381 * config/obj-coff.c (size_section): Likewise.
382 * config/obj-ieee.c (size_section): Likewise.
383 * config/tc-d10v.h (md_do_align): Add max parameter.
384 * config/tc-i386.h (md_do_align): Likewise.
385 * config/tc-m88k.h (md_do_align): Likewise.
386 * config/tc-m88k.c (m88k_do_align): Likewise.
387 * config/tc-sh.h (md_do_align): Likewise.
388 * config/tc-sh.c (sh_do_align): Likewise.
389 * as.h: Improve comments on rs_align and rs_align_code.
390 * doc/as.texinfo: Document new alignment arguments.
391 * doc/internals.texi (Frags): Document use of fr_subtype field for
392 rs_align and rs_align_code.
396 * config/tc-mips.c: Changed opcode parsing.
400 * config/{tc-alpha.h, tc-arc.h, tc-d10v.h, tc-generic.h, tc-i960.h,
401 tc-mn10200.h, tc-mn10300.h, tc-sh.h, tc-v850.h, tc-vax.h, tc-w65.h}:
402 Add default definition of zero for TARGET_BYTES_BIG_ENDIAN.
403 * config/{tc-arm.h, tc-hppa.h, tc-i386.h, tc-mips.h, tc-ns32k.h,
404 tc-ppc.h, tc-sparc.h}: Move definition of TARGET_BYTES_BIG_ENDIAN
405 to a location consistent with the rest of the target include files.
406 * config/tc-i386.c: Remove misleading comment.
407 * doc/internals.texi (CPU backend): Add description of function
410 * config/tc-tic80.c: Add code to insert predefined symbols into the
411 symbol table so they can be parsed by the standard expression parser.
412 Remove custom code that use to parse them.
413 * config/tc-tic80.h: Move definition of TARGET_BYTES_BIG_ENDIAN
414 to a location consistent with the rest of the target include files.
419 * as.h: GNU c provides unlink() function.
421 Unify section handling on openVMS/Alpha:
422 * config/tc-alpha.c(s_alpha_link): Remove.
423 (s_alpha_section): New function.
424 Remove case-hacking of symbols
425 Add .code_address pseudo-op.
426 (BFD_RELOC_ALPHA_CODEADDR): New relocation.
427 (s_alpha_code_address): New function.
428 (alpha_ctors_section, alpha_dtors_section): New sections for C++
429 static constructors/destructors.
430 Add debug code for crash debugs, to be removed when traceback code
431 is added to object code.
432 (s_alpha_name): New function for .name pseudo-op.
433 (alpha_print_token): New function to print token expressions with
434 alpha specific extensions.
436 * makefile.vms: Allow compilation with current gcc snapshot.
440 * doc/Makefile.in (TEXI2DVI): Set to just name of program.
441 (DVIPS): Set to dvips.
442 (ps, as.ps, gasp.ps): New targets.
443 (internals.info, gasp.dvi, internals.dvi): Set both TEXINPUTS
444 and MAKEINFO env variables.
445 (internals.ps): Use DVIPS macro.
446 (clean): Remove core and backup files.
447 (distclean): Remove temporary files from building internals.
449 * doc/internals.texi (Frags): Fix typo.
450 (GAS processing): Ditto.
451 (CPU backend): Ditto.
452 * ecoff.c (init_file): Use TARGET_BYTES_BIG_ENDIAN value directly.
453 * mpw-config.in: Define TARGET_BYTES_BIG_ENDIAN as 1.
454 * read.c: Remove ugly hack that dealt with config files not
455 correctly defining TARGET_BYTES_BIG_ENDIAN.
456 (target_big_endian): Use TARGET_BYTES_BIG_ENDIAN directly.
457 * config/arm-big.mt: Define TARGET_BYTES_BIG_ENDIAN to 1.
458 * config/arm-lit.mt: Define TARGET_BYTES_BIG_ENDIAN to 0.
459 * config/mips-big.mt: Define TARGET_BYTES_BIG_ENDIAN to 1.
460 * config/mips-lit.mt: Define TARGET_BYTES_BIG_ENDIAN to 0.
461 * config/ppc-lit.mt: Define TARGET_BYTES_BIG_ENDIAN to 1.
462 * config/ppc-sol.mt: Replace TARGET_BYTES_LITTLE_ENDIAN
463 with TARGET_BYTES_BIG_ENDIAN defined to 0.
464 * config/tc-arm.h: Remove use of TARGET_BYTES_LITTLE_ENDIAN
465 and simplify. Test value of TARGET_BYTES_BIG_ENDIAN, not just
466 whether it is defined or not.
467 * config/tc-mips.h: Remove use of TARGET_BYTES_LITTLE_ENDIAN.
468 * config/tc-ppc.h: Remove use of TARGET_BYTES_LITTLE_ENDIAN
469 and simplify. Test value of TARGET_BYTES_BIG_ENDIAN, not just
470 whether it is defined or not.
472 * config/tic80.h (TARGET_FORMAT): Define to coff-tic80.
473 (TARGET_BYTES_BIG_ENDIAN): Define to 0.
478 * write.c (write_relocs): Correct text in as_fatal error message,
479 bfd_perform_relocation -> bfd_install_relocation.
483 * config/tc-m68k.c (LEX_TILDE): Define if TE_DELTA.
484 * read.c (LEX_TILDE): Define if not defined.
485 (lex_type): Use LEX_TILDE.
486 * expr.c (get_symbol_end): Check first char with is_name_beginner,
491 * config/tc-sparc.c (md_show_usage): Add missing backslash at end
494 * config/tc-mips.c (mips16_extended_frag): Correct base address
495 for an extended PC relative instruction.
496 (md_convert_frag): Likewise.
498 * config/tc-mips.c (prev_nop_frag): New static variable.
499 (prev_nop_frag_holds): New static variable.
500 (prev_nop_frag_required): New static variable.
501 (prev_nop_frag_since): New static variable.
502 (append_insn): If we aren't reordering, and prev_nop_frag is not
503 NULL, and we don't need any nops, then decrease the size of
504 prev_nop_frag. Don't insert nops because of instructions in
505 noreorder sections. Remember whether the previous instructions
506 where in noreorder sections even when not reordering.
507 (mips_no_prev_insn): Add preserver parameter. Change all
508 callers. Refer prev_nop_frag variables when appropriate.
509 (mips_emit_delays): Set up prev_nop_frag.
510 (s_mipsset): Clear prev_nop_frag if reordering.
514 * config/tc-mips.c (append_insn): Remove useless code which
515 handled swapping a mips16 jump with a mips16 instruction with a
518 * config/tc-mips.c (md_parse_option): When debugging, set
519 mips_optimize to 1, not 0.
521 * config/tc-mips.c (mips16_ip): Handle an extend operand.
523 * config/tc-mips.c (my_getExpression): In mips16 mode, if it looks
524 like the expression was based on `.', adjust the value of the
527 * config/tc-mips.c (append_insn): Warn about an attempt to put an
528 extended instruction in a delay slot when not reordering.
529 (md_convert_frag): Warn if an extended instruction appears in a
532 * config/tc-mips.c (mips_pseudo_table): Add "insn".
533 (s_insn): New static function.
534 * doc/c-mips.texi: Document .insn.
536 * config/tc-mips.c (md_begin): Add the general registers to the
538 (mips16_ip): First parse the expression, and then see whether it
539 came up with a register, rather than trying to first see whether
540 we are looking at a register.
545 * config/tc-tic80.c: Numerous changes and additions to flesh
546 out functions that were previously just stubs, and fix some
547 problems found using the new TIc80 testsuite cases.
552 * config/tc-mips.c (mips16_ip): Handle %gprel modifier.
553 (md_apply_fix): Handle BFD_RELOC_MIPS16_GPREL.
555 * config/tc-mips.c (append_insn): Output jump instruction as a
556 pair of 2 byte instructions, rather than as a single 4 byte
560 * config/tc-mips.c (md_begin): Correct test of mips_5900.
561 (mips_ip): Don't check INSN_ISA for a macro.
567 * config/tc-mips.c: added r5900.
572 * itbl-ops.c, itbl-lex.l, itbl-parse.y, itbl-ops.h,
573 config/itbl-mips.h: Add copyright message and fix indentation.
577 * itbl-ops.c: New file. Add support for dynamically read
578 instruction registers, opcodes and formats. Build internal table
579 for new instructions and provide callbacks for assembler and
581 * itbl-lex.l, itbl-parse.y: Lex and yacc parsers for instruction
583 * itbl-ops.h: New file. Header file for itbl support.
584 * config/itbl-mips.h: New file. Mips specific definitions for
589 * config/tc-mn10200.c (md_assemble): If a constant operand won't
590 fit into the constant field of a relaxable operand, then it does
595 * config/tc-mn10200.c (md_estimate_size_before_relax): Treat
596 a jsr target in a different section just like a jsr to
601 * config/tc-mips.c (mips_fix_adjustable): Don't adjust relocations
602 against any mips16 symbols, not just externally visible ones.
603 (md_apply_fix): Corresponding change.
607 * config/tc-mips.c (mips16_ip): Accept floating point registers in
608 the operand of the exit instruction.
612 * symbols.c (resolve_symbol_value): If we leave an equated symbol
613 as O_symbol, copy over the segment.
617 * config/tc-mips.c (md_apply_fix): If we aren't adjusting this
618 fixup to be against the section symbol, adjust the value
621 * symbols.c (resolve_symbol_value): Don't change X_add_number for
623 * write.c (write_relocs): Avoid looping on equated symbols.
624 Adjust fx_offset by X_add_number for each symbol.
625 * config/obj-coff.c (do_relocs_for): Avoid looping on equated
627 (fixup_segment): Add a loop to track down equated symbols and
628 adjust fx_offset appropriately.
632 * config/tc-mn10200.c (md_relax_table): Add entries to allow
634 (md_convert_frag): Handle jmp->bra relaxing.
635 (md_assemble): Handle jmp->bra relaxing.
636 (md_estimate_size_before_relax): Likewise.
640 * config/tc-i386.c (i386_align_code): Add comments explaining the
645 * config/tc-sparc.c (enforce_aligned_data): New static variable.
646 (sparc_cons_align): Don't do anything unless enforce_aligned_data
648 (md_longopts): Add "enforce-aligned-data".
649 (md_show_usage): Mention --enforce-aligned-data.
650 * doc/c-sparc.texi (Sparc-Aligned-Data): New node; document
651 enforce-aligned-data.
653 * config/tc-ppc.c (md_pseudo_table): If OBJ_XCOFF, add "long",
655 (ppc_xcoff_cons): New static function.
657 * write.c (relax_segment): Give an error if a .space symbol is
660 * read.c (read_a_source_file): Don't handle mri_pending_align if
661 the handler is s_globl or s_ignore.
665 * config/tc-d10v.c (find_opcode): Remove unused variable "numops".
667 * config/tc-tic80.c: Many additions to previous placeholder file.
668 * config/tc-tic80.h: Ditto.
673 * config/tc-i386.c (i386_align_code): Improve the nop patterns.
677 * config/tc-mips.c (mips_fix_adjustable): New function.
678 * config/tc-mips.h (tc_fix_adjustable): Call mips_fix_adjustable.
679 (mips_fix_adjustable): Declare.
682 * read.c (read_a_source_file): After doing an mri_pending_align,
683 adjust the line_label if there is one.
684 (s_space): Set mri_pending_align if an odd number of bytes were
689 * config/tc-d10v.h (md_do_align): Add this hook to call
690 d10v_cleanup() when a ".align" is detected. Fixes PR11487.
692 * config/tc-d10v.c (find_opcode): Correctly calculate
693 branch displacement when .aligns are present.
697 * config/tc-mn10200.c (md_relax_table): Define.
698 (md_convert_frag): Implement.
699 (md_assemble): Handle relaxable operands/instructions correctly.
700 (md_estimate_size_before_relax): Implement.
701 * config/tc-mn10200.h (TC_GENERIC_RELAX_TABLE): Define.
705 * config/tc-mips.c (append_insn): Give an error for jumps to a
707 (md_apply_fix): Make a branch to an odd address an error rather
710 * config/tc-mips.c (md_convert_frag): If the user explicitly
711 requested an extended opcode, pass warn as true to mips16_immed.
713 * config/tc-mips.c (mips16_ip): Handle a missing expression like
714 an explicit 0, so that explicitly extended instructions work
719 * ecoff.c (ecoff_build_symbols): Don't generate a local ECOFF
720 symbol for a common symbol.
724 Patch presumed to have been checked in awhile ago but wasn't.
726 * write.c: Delete "ifndef md_relax_frag" around is_dnrange.
727 (relax_segment, case rs_org): Move code inside braces. Move locals
728 target,after inside too.
729 (relax_segment, case rs_machine_dependent): Guts moved to ...
730 (relax_frag): New function.
731 Call md_prepare_relax_scan if defined.
735 * config/tc-m68k.c (m68k_ip): Reject pc-relative addresses for the
736 'p' operand specifier.
740 * config/tc-m68k.c (HAVE_LONG_BRANCH): New macro, returns true for
741 m68k family cpus which support long branch addressing modes.
742 (m68k_ip, md_convert_frag_1, md_estimate_size_before_relax,
743 md_create_long_jump): Use it.
747 * config/tc-mips.c (md_begin): Don't set SEC_ALLOC or SEC_LOAD for
748 the .reginfo or .MIPS.options section if configured for an
751 * config/tc-mips.c (md_begin): Don't set interlocks for
756 * read.c (read_a_source_file): Make sure the symbol ends with
757 whitespace before checking whether the next character is '='.
761 * config/tc-i386.c (sco_id): Moved from here...
762 * config/obj-elf.c (sco_id): ...to here. Adding the identifier
763 really is an SCO ELF specific thing, not just a SCO x86 specific
769 * configure.in (tic80-*-*): Don't require 'coff'.
770 * configure: Regenerate.
775 * read.c (emit_expr): Check for overflow of a negative value
777 * write.c (fixup_segment): Likewise.
778 * config/obj-coff.c (fixup_segment): Likewise.
780 * config/tc-m68k.c (struct label_line): Define.
781 (labels, current_label): New static variables.
782 (md_assemble): Mark current_label as text, and clear it.
783 (m68k_frob_label): New function.
784 (m68k_flush_pending_output): New function.
785 (m68k_frob_symbol): New function.
786 * config/tc-m68k.h (tc_frob_label): Define.
787 (md_flush_pending_output): Define.
788 (tc_frob_symbol): Don't warn, just call m68k_frob_symbol.
789 (tc_frob_coff_symbol): Likewise.
791 * read.c (read_a_source_file): When defining a macro in MRI mode,
792 don't add the symbol to the symbol table.
796 * config/tc-mn10300.c (tc_gen_reloc): Handle sym1-sym2 fixups
797 here since fixup_segment doesn't (linkrelax is set).
798 * config/tc-mn10200.c (tc_gen_reloc): Likewise.
802 * config/tc-mn10200.c (md_assemble): Tweak fx_offset for pc-relative
807 * config/tc-hppa.c (struct hppa_fix_struct): Tweak fx_r_field's type
808 to avoid warnings with the native HP compiler.
809 (fix_new_hppa): Similarly for the r_type argument.
810 (pa_build_unwind_subspace, hppa_elf_mark_end_of_function): Enclose
811 in an #if OBJ_ELF to keep gcc -Wall quiet.
812 (md_apply_fix): Always initialize "result".
814 * config/tc-mn10200.c (md_assemble): Generate relocations.
818 * config/tc-m68k.c (s_even): Adjust the alignment of the current
823 * config/obj-elf.c (elf_file_symbol): When using ECOFF debugging,
824 pass on the new file hook.
826 * config/tc-alpha.c (alpha_fix_adjustable): Not quite the same as
827 !alpha_force_relocation, as local LITERALs can be adjusted to be
828 relative to the section.
832 * config/obj-coff.c (yank_symbols): If tc_frob_coff_symbol is
834 * config/tc-m68k.h (tc_frob_symbol): Check whether text label is
835 aligned to odd boundary.
836 (tc_frob_coff_symbol): Define.
838 * doc/as.texinfo (Set): Change parenthesized @xref to @pxref.
840 * macro.c (macro_expand_body): In MRI mode, just copy a single &.
842 * config/tc-m68k.c (m68k_ip): Call frag_grow before adding a
845 * config/tc-m68k.c (m68k_ip): Accept 'B' as a size for an
847 (md_assemble): If the size is 'B', set fx_signed.
848 (md_apply_fix_2): Use fx_signed when checking for overflow.
850 * write.h (struct fix): Add fx_signed field.
851 * write.c (fix_new_internal): Initialize fx_no_overflow and
853 (fixup_segment): Use fx_signed when checking for overflow.
854 * config/obj-coff.c (fixup_segment): Check fx_no_overflow and
855 fx_signed when checking for overflow.
859 * NOTES, NOTES.config: Removed. These are rarely, if ever,
860 updated, and all the useful information is in doc/internals.texi.
863 * read.c (read_a_source_file): Check for conditional operators
864 before doing an MRI pending alignment.
865 * config/tc-m68k.h (m68k_conditional_pseudoop): Declare.
866 (tc_conditional_pseudop): Define.
867 * config/tc-m68k.c (m68k_conditional_pseudop): New function.
868 * doc/internals.texi (CPU backend): Describe
869 tc_conditional_pseudoop.
872 * config/tc-m68k.c (m68k_rel32_from_cmdline): New static
874 (md_begin): Check m68k_rel32_from_cmdline before setting
876 (m68k_mri_mode_change): Likewise.
877 (md_longopts): Add --disp-size-default-16 and
878 --disp-size-default-32.
879 (md_parse_option): Handle new options.
880 (md_show_usage): Mention new options.
881 * doc/c-m68k.texi (M68K-Opts): Document new options.
884 * config/tc-m68k.c (m68k_index_width_default): New static
886 (m68k_ip): Use m68k_index_width_default to set the size of a base
887 register whose size was not given.
888 (md_longopts): Add --base-size-default-16 and
889 --base-size-default-32.
890 (md_parse_option): Handle new options.
891 (md_show_usage): Mention new options.
892 * doc/c-m68k.texi (M68K-Opts): Document new options.
894 * doc/c-mips.texi: Mention ISA level 4, and the -mips16 option.
896 * configure.in: Recognize mips-*-linux* target.
897 * configure: Rebuild.
899 * config/tc-mips.c (load_register): Rewrite 64 bit handling to
900 work if valueT is only 32 bits.
902 * config/tc-mips.c: Throughout, check target_big_endian rather
904 (byte_order): Remove.
905 (mips_init_after_args): Remove.
906 * config/tc-mips.h (LITTLE_ENDIAN, BIG_ENDIAN): Don't define.
907 (mips_init_after_args): Don't declare.
908 (tc_init_after_args): Don't define.
910 * config/tc-mips.h (tc_frob_after_relocs): Define if
912 (mips_elf_final_processing): Likewise.
913 (ELF_TC_SPECIAL_SECTIONS): Likewise.
918 * config/tc-v850.c (md_assemble): If this is sst.{h,w} or
919 sld.{h,w} and the operand is relocatable, adjust the adend by
920 shifting it right one bit.
925 * read.c (read_a_source_file): Check mri_pending_align after
926 checking for a macro. From Ronald F. Guilmette
929 * Makefile.in (ALL_CFLAGS): Add -D_GNU_SOURCE.
931 * config/tc-sparc.c (md_apply_fix3): Rename from md_apply_fix, and
932 add segment argument. If OBJ_ELF, treat a relocation against a
933 symbol in a linkonce section like a relocation against an external
935 * config/tc-sparc.h (MD_APPLY_FIX3): Define.
939 * config/tc-mips.c (mips16_macro): Add case for M_ABS.
943 * NOTES.config (Implementation): as.h #define's "GAS" not "gas",
944 includes config.h instead of host.h, tc.h instead of tp.h, and
945 targ-env.h instead of target-environment.h.
946 Also, obj-format.h includes targ-cpu.h instead of
949 * configure.in (case ${generic_target}): Add tic80-*-coff entry.
950 * configure: Rebuild with autoconf.
951 * config/obj-coff.h (coff/tic80.h): Include if TC_TIC80 defined.
952 (TARGET_FORMAT): Define to "coff-tic80".
953 * config/tc-tic80.c: New file for TIc80 support.
954 * config/tc-tic80.h: New file for TIc80 support.
959 * doc/as.texinfo (M): Mention explicitly that -M changes macro
964 * write.c (adjust_reloc_syms): If the fixup symbol has been
965 equated to an undefined symbol, convert the fixup to being against
966 the target symbol. Remove obsolete code handling a special case
971 * mpw-make.sed: Use NewFolderRecursive for installation.
975 * config/tc-d10v.c (do_assemble): Correct previous bug fix.
979 * config/tc-d10v.c (md_assemble): Fix bug which caused
980 second instruction in a line to be case sensitize. PR11312
984 * config/tc-mn10200.c (mn10200_insert_operand): Don't
985 range check operands with MN10200_OPERAND_NOCHECK set.
986 (check_operand): Likewise.
990 * config/tc-mips.c: Undo part of last Friday's alignment changes.
991 (md_begin): Always align the text section to a four byte
993 (append_insn): Remove call to record_align.
995 * config/tc-mips.c (insn_label): Remove.
996 (struct insn_label_list): Define.
997 (insn_labels, free_insn_labels): New static variables.
998 (mips_clear_insn_labels): New static function.
999 (append_insn): Mark all mips16 text labels, and make them odd.
1000 Handle all labels after emitting a nop, not just one. Call
1001 mips_clear_insn_labels rather than just clearing insn_label.
1002 (mips_emit_delays): Add insns parameter, and use it to decide
1003 whether to mark mips16 labels. Handle all labels, not just one.
1004 Force mips16 labels to be odd. Change all callers.
1005 (mips16_immed): Don't check for an odd branch target.
1006 (md_apply_fix): Don't check mips16 mode for a branch reloc.
1007 (mips16_extended_frag): Ignore the low bit in a branch target.
1008 (md_convert_frag): Likewise.
1009 (mips_no_prev_insn): Call mips_clear_insn_labels rather than just
1010 clearing insn_label.
1011 (mips_align, mips_flush_pending_output, s_cons): Likewise.
1012 (s_float_cons, s_gpword): Likewise.
1013 (s_align): Use insn_labels rather than insn_label.
1014 (s_cons, s_float_cons, s_gpword): Likewise.
1015 (mips_frob_file_after_relocs): New function.
1016 (mips_define_label): Rewrite to add to insn_labels list.
1017 * config/tc-mips.h (tc_frob_file_after_relocs): Define.
1018 * ecoff.c (ecoff_build_symbols): If the size of a function comes
1019 out odd, increment it.
1021 * config/tc-mips.c (append_insn): Only update prev_insn when not
1022 reordering if place is NULL.
1024 * config/tc-mips.c (mips16_ip): Check for a missing expression
1025 when using the register indirect addressing mode.
1029 * config/tc-mn10200.c (mn10200_insert_operand): Don't
1030 check 24bit operands for overflow.
1031 (check_operand): Likewise.
1035 * doc/as.texinfo (Section): Document how to use the .section
1036 pseudo-op for COFF and ELF.
1040 * write.c (adjust_reloc_syms): Fix linkonce check for ELF.
1044 * config/tc-mips.c (prev_insn_reloc_type): New static variable.
1045 (RELAX_MIPS16_ENCODE): Add dslot and jal_dslot arguments, and
1046 store them. Adjust other RELAX_MIPS16 macros.
1047 (RELAX_MIPS16_DSLOT): Define.
1048 (RELAX_MIPS16_JAL_DSLOT): Define.
1049 (append_insn): Pass new arguments to RELAX_MIPS16_ENCODE. Correct
1050 handling of whether previous instruction has a fixup. Set
1051 prev_insn_reloc_type.
1052 (mips_no_prev_insn): Clear prev_insn_reloc_type.
1053 (mips16_extended_frag): Use the right base address for a PC
1054 relative add or load.
1055 (md_convert_frag): Likewise. If a PC relative add or load is
1056 used, record the alignment for the section.
1060 * write.c (adjust_reloc_syms): Don't reduce a reloc against a
1061 linkonce section into a reloc against the section symbol.
1063 * config/tc-mips.c (mips16_macro): Remove nop instructions after
1064 branch instructions.
1066 * config/tc-mips.c (md_begin): If configured for an embedded ELF
1067 system, don't set the section alignment to 2**4.
1068 (s_change_sec): Likewise.
1069 (append_insn): Call record_alignment for the section.
1070 (md_section_align): Don't align the section size for an embedded
1075 * write.c (adjust_reloc_syms): Make sure that symbols are
1076 resolved; expression symbols may have been skipped.
1077 * config/obj-coff.c (fixup_segment): Likewise.
1081 * config/tc-ppc.c (ppc_elf_suffix): Move @plt to
1082 BFD_RELOC_24_PLT_PCREL relocation.
1083 (md_apply_fix3): Support BFD_RELOC_24_PLT_PCREL.
1087 * config/tc-d10v.c (write_2_short): Remove code that called
1088 parallel_ok() when the programmer specified parallel instructions.
1092 * config/tc-mn10300.c (md_assemble): Update to handle endianness
1095 * config/tc-mn10200.c (md_assemble): Opcode 0x0 is valid!
1096 * config/tc-mn10300.c (md_assemble): Likewise.
1100 * config/tc-mips.c (append_insn): Make sure there is enough room
1101 in a frag after a mips16 instruction to switch it with a jump
1104 * config/tc-mips.c (mips16_extended_frag): Give an error for an
1105 attempt to use a non absolute symbol in an extending frag.
1109 * config/tc-mn10200.c: Flesh out assembler support for MN10200.
1110 * config/tc-mn10200.h: Likewise.
1114 * app.c (do_scrub_chars): At the end of a C comment, pass space to
1115 UNGET rather than PUT. Set old_state before setting state to -2.
1117 * config/tc-mips.c (mips16_extended_frag): Avoid an infinite loop
1118 when extending because the value is exactly maxtiny + 1.
1120 * config/tc-mips.c (RELAX_MIPS16_ENCODE): Add small and ext
1121 arguments, and store them. Adjust other RELAX_MIPS16 macros.
1122 (RELAX_MIPS16_USER_SMALL): Define.
1123 (RELAX_MIPS16_USER_EXT): Define.
1124 (mips16_small, mips16_ext): New static variables.
1125 (append_insn): Pass mips16_small and mips16_ext to
1126 RELAX_MIPS16_ENCODE.
1127 (mips16_ip): Set mips16_small and mips16_ext.
1128 (mips16_immed): Don't check mips16_autoextend.
1129 (mips16_extended_frag): Check USER_SMALL and USER_EXT.
1131 * write.c (write_relocs): Print an error for an out of range
1132 fixup, rather than calling abort.
1134 * as.c (main): Unlink the output file if there are errors while
1135 generating the fixups.
1139 * config/tc-mips.c (mips16_extended_frag): Don't call
1141 (md_convert_frag): Call resolve_symbol_value before calling
1142 S_GET_VALUE, and don't add in the frag address.
1144 * config/tc-mips.c (mips16_immed): Add file and line parameters,
1145 and use them when reporting errors. Change all callers.
1149 * config/tc-mn10300.c: Fix various gcc -Wall warnings.
1150 Remove '$' prefixing for registers.
1154 * config/tc-d10v.c (md_assemble): Check to see if prev_seg
1155 is initialized before using it.
1156 (d10v_cleanup): No longer uses its argument, so make it void.
1158 * config/tc-d10v.h (d10v_cleanup): Change prototype.
1162 * write.c (fixup_segment): Don't discard the symbol for a PC
1163 relative fixup to an absolute symbol.
1167 * config/tc-d10v.c (md_assemble, d10v_cleanup): Fix bug
1168 with multiple sections.
1172 * config/tc-mips.c (md_longopts): Rename mips-16 to mips16, and
1173 no-mips-16 to no-mips16.
1174 (s_mipsset): Accept .set mips16 and .set nomips16.
1178 * config/tc-ppc.c (ppc_elf_suffix): Take expressionS pointer
1179 argument, and check for +/- constant following the suffix, folding
1180 it into the expression.
1181 (ppc_elf_cons): Change ppc_elf_suffix calls.
1182 (md_assemble): Ditto.
1183 (shlib): Replace boolean mrelocatable with enumeration shlib.
1184 (md_parse_option): Discriminate between PIC style shared libraries
1186 (ppc_elf_validate_fix): Don't report warnings for PIC style shared
1191 * config/tc-ppc.h ({tc,ppc}_comment_chars): Define, so that we can
1192 change the comment characters.
1194 * config/tc-ppc.c (comment_chars): Delete in favor of
1196 (ppc_{eabi,solaris}_comment_chars): Eabi and Solaris versions of
1198 (ppc_comment_chars): Select appropriate comment chars by default.
1199 (msolaris): New flag for -m{,no-}solaris.
1200 (md_parse_option): Recognize -K pic. Add support for
1202 (md_show_usage): Update.
1203 (md_begin): Do not set ELF flags if Solaris.
1204 (ppc_elf_suffix): @local sets R_PPC_LOCAL24PC relocation.
1205 (md_apply_fix3): Add support for R_PPC_LOCAL24PC.
1209 * as.c (main): Correct handling of flag_always_generate_output.
1213 * config/tc-mn10300.c (tc_gen_reloc): Get the addend from
1214 fx_offset, not fx_addnumber.
1216 * config/tc-mn10300.h (tc_fix_adjustable): Don't do any
1221 * config/tc-i386.c: Correct misspelling: balenced to balanced.
1225 * config/tc-mips.c (md_section_align): Check for an alignment of
1226 4, not an alignment of 16. Corrects August 7 patch.
1230 * configure, conf.in: Rebuild with autoconf 2.12.
1232 * config/tc-ppc.c (ppc_elf_lcomm): Don't give an error if no
1233 alignment is specified.
1235 Add support for mips16 (16 bit MIPS implementation):
1236 * config/tc-mips.c: Extensive additions for mips16 support, not
1238 (RELAX_OLD, RELAX_NEW): Use only 7 bits each.
1239 (insn_uses_reg): Change last parameter to an enum.
1240 * config/tc-mips.h (LOCAL_LABELS_DOLLAR): Define as 0.
1241 (md_relax_frag): Define as mips_relax_frag.
1242 (mips_relax_frag): Declare.
1243 (struct mips_cl_insn): Add use_extend and extend fields.
1244 (tc_fix_adjustable): Define.
1245 * config/obj-elf.h (S_GET_OTHER): Define.
1246 (S_SET_OTHER): Define.
1250 * config/tc-m68k.c (m68k_ip): Implement cases for new <, >, m, n,
1251 o and p operand specifiers.
1255 * write.c: Delete "ifndef md_relax_frag" around is_dnrange.
1256 (relax_segment, case rs_org): Move code inside braces. Move locals
1257 target,after inside too.
1258 (relax_segment, case rs_machine_dependent): Guts moved to ...
1259 (relax_frag): New function.
1260 Call md_prepare_relax_scan if defined.
1261 * config/tc-m68k.h (md_prepare_relax_scan): Renamed from
1266 * config/tc-mn10300.c (address_registers): Use '$' as register
1267 prefix instead of '%'.
1268 (data_registers, other_registers, md_assemble): Likewise.
1270 * config/tc-mn10300.c (address_registers): Use '%' prefix for regs.
1271 (data_registers, other_registers, md_assemble): Likewise.
1273 * config/tc-mn10300.c (md_assemble): Correctly determine the
1274 correct location and type for each relocation.
1275 (md_pcrel_from): Simplify.
1279 * config/tc-sh.c (md_convert_frag): Improve warning when branch is
1280 converted into branch around branch.
1284 * config/tc-mn10300.h (DIFF_EXPR_OK): Don't define this.
1285 (tc_fix_adjustable): Don't adjust relocs against weak symbols or
1287 * config/tc-mn10300.c (md_begin): Set linkrelax.
1288 (md_assemble): Create fixups as needed.
1289 (md_apply_fix3): Gut. It shouldn't ever get called anymore.
1293 * config/tc-d10v.c (parallel_ok): When automatically converting
1294 serial ops to parallel, do not consider a branch as the first
1299 * config/tc-mn10300.c (md_assemble): Handle MN10300_OPERAND_REG_LIST.
1303 * config/tc-mn10300.c (mn10300_insert_operand): Provide prototype
1305 (check_operand): Likewise.
1309 * config/tc-d10v.c (parallel_ok): Branch and link instructions
1311 (write_2_short): Call parallel_ok to check whether two short
1312 instructions the user requested execute in parallel, can be
1317 * config/tc-d10v.c (write_2_short): Fix bug that wouldn't
1318 allow a branch and link in parallel with an exe instruction.
1322 * doc/c-d10v.texi: Add info on @word modifier.
1326 * config/tc-mn10300.c (mn10300_insert_operand): MN10300_OPERAND_SPLIT
1327 operands are assumed to be 32bits. Use "bits" field to hold the
1328 number of bits in the main instruction word for MN10300_OPERAND_SPLIT.
1329 (mn10300_check_operand): MN10300_OPERAND_SPLIT operands are assumed
1332 * config/tc-mn10300.c (mn10300_insert_operand): Shift low part
1333 of a MN10300_OPERAND_SPLIT operand by operand->shift.
1335 * config/tc-mn10300.c (mn10300_insert_operand): Handle
1336 MN10300_OPERAND_SPLIT.
1340 * config/tc-mn10300.c (md_assemble): Insert operands into
1341 the extension part of the instruction if necessary.
1342 (mn10300_insert_operand): Accept pointer to extension word
1343 argument. Make insn a pointer argument too. Return type
1344 is now void. All callers changed.
1348 * config/tc-mn10300.c (mn10300_insert_operand): Handle
1349 repeated register operands.
1353 * doc/as.texinfo: Added section on reporting bugs.
1355 * config/tc-alpha.c: Change uses of void * to PTR. Change the
1356 alpha_macro emit field to expect a const argument, and change the
1357 arg field to be const. Fix some spacing to follow the GNU
1362 * config/tc-alpha.c (md_parse_option): Add knowledge of 21164pc
1363 (pca56) and 21264 (ev6) cpus.
1364 (md_apply_fix): Private relocation types are now negative.
1365 (alpha_force_relocation): Likewise.
1366 (tc_gen_reloc): Likewise.
1367 (emit_insn): Likewise.
1368 (emit_ldXu): Do the right thing when the hardware can do byte insns.
1369 (emit_stX): Likewise.
1370 (emit_sextX): Likewise.
1374 * config/obj-coff.c (do_relocs_for): Call resolve_symbol_value on
1375 a symbol found in a reloc.
1377 * symbols.c (resolve_symbol_value): Improve the error message if
1378 an undefined symbol is used in an expression.
1382 * doc/internals.texi: Rewrite, and add a lot of documentation.
1383 * doc/Makefile.in (internals.info): New target.
1388 * config/tc-v850.h (tc_fix_adjustable): Don't adjust relocs
1389 against weak symbols.
1393 * config/tc-v850.c (md_assemble): Don't lose for relaxable
1396 * config/tc-v850.c (md_convert_frag): Make sure we insert the
1397 fixup at the right address within the frag.
1399 * config/tc-v850.c (md_convert_frag): Don't set fragP->fr_fix
1400 to an absolute value, instead increment it as needed.
1402 * config/tc-v850.h (TC_GENERIC_RELAX_TABLE): Define.
1403 * config/tc-v850.c: Fix some indention problems.
1404 (md_relax_table): Define for D9->D99 branch displacement
1406 (md_convert_frag): Do something useful instead of aborting.
1407 (md_estimate_size_before_relax): Likewise.
1408 (md_assemble): Note if the matching instruction has a relaxable
1409 operand. If it does, allocate frag with frag_var and don't
1415 * config/tc-d10v.h (md_cleanup): New function. This is needed to
1416 write out any buffered instructions when a ".end" is found.
1420 * read.c (read_a_source_file): New hook md_cleanup().
1424 * write.c (fix_new_exp): Use make_expr_symbol to build an
1425 expression symbol for a complex fixup.
1430 * config/tc-v850.c (v850_reloc_prefix): Several disgusting
1431 hacks to improve parsing of complex hi, lo, zda, etc
1433 (md_assemble): Don't demand and eat a trailing ')' after finding
1434 a v850 relocation prefix. Sign extend the constant in a
1435 BFD_RELOC_LO16 expression. Do eat a trailing ')' after a complete
1437 (parse_cons_expression_v850): Don't eat a trailing ')' after
1438 finding a v850 relocation prefix.
1440 * config/tc-v850.h (TC_PARSE_CONS_EXPRESSION): Define.
1441 (TC_CONS_FIX_NEW): Likewise.
1442 * config/tc-v850.c (parse_cons_expression_v850): New function.
1443 (cons_fix_new_v850): Likewise.
1445 * config/tc-v850.h (tc_fix_adjustable): Don't adjust TDA relocs.
1450 * config/tc-ppc.c (md_apply_fix3): Give a better warning message
1451 for an unknown relocation type.
1456 * config/tc-v850.c (md_pseudo_table): Add .word; allocates
1461 * config/tc-v850.c (md_assemble): Handle TDAOFF relocs
1462 differently for movea & sst/sld insns.
1467 * config/tc-d10v.c (parallel_ok): Don't allow illegal combinations
1472 * obj.h (struct format_ops): Add frob_file_after_relocs field.
1473 * config/obj-multi.h (obj_frob_file_after_relocs): Define.
1474 * config/obj-ecoff.c (ecoff_format_ops): Initialize new
1475 frob_file_after_relocs field.
1476 * config/obj-elf.c (elf_format_ops): Likewise.
1477 * config/tc-mips.c: Undefine obj_frob_file_after_relocs before
1478 including obj-elf.h.
1482 * config/tc-mips.c (cons_fix_new_mips): Only treat 8 byte reloc
1483 specially if not ELF.
1484 (md_apply_fix): Handle BFD_RELOC_64.
1485 (tc_gen_reloc): Handle BFD_RELOC_64.
1487 * config/tc-i386.c (md_apply_fix3): Don't increment value for a PC
1488 relative reloc when BFD_ASSEMBLER and OBJ_AOUT (more ugly gas
1491 * config/obj-aout.h (S_IS_DEFINE): non BFD_ASSEMBLER version:
1492 Don't check S_GET_OTHER.
1496 * config/tc-mips.c (mips_ip): Accept an odd floating point
1497 register with l.s or s.s.
1499 * config/obj-aout.c (obj_pseudo_table): Use obj_aout_type for
1501 (obj_aout_type): New static function.
1505 * Makefile.in ($(OBJS)): Depend upon libiberty.h.
1510 * config/tc-v850.c (v850_reloc_prefix): Recognize zdaoff, tdaoff
1511 and sdaoff expressions.
1513 * write.c (fixup_segment): Don't add symbol value to addend if
1514 TC_V850 and OBJ_ELF.
1515 * config/tc-v850.h (tc_fix_adjustable): Don't adjust any
1518 * config/tc-v850.c (md_pcrel_from): Undo yesterday's changes.
1519 (md_pcrel_from_section): Likewise.
1520 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Likewise.
1524 * config/tc-v850.c (md_pcrel_from): Delete unused function.
1525 (md_pcrel_from_section): New function.
1526 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Define.
1531 * config/tc-mips.c (load_register): Add cast to offsetT when using
1536 * config/obj-elf.c (elf_frob_file): Move ECOFF debug processing to ...
1537 (elf_frob_file_after_relocs): ... here. New function.
1538 * config/obj-elf.h (obj_from_file_after_relocs): New macro.
1539 * write.c (write_object_file): Call *frob_after_relocs after the
1540 call to write_relocs.
1542 * config/tc-alpha.c: Use new BFD_RELOC_ALPHA_ELF_LITERAL reloc.
1544 * config/tc-alpha.c (load_expression): Don't SET_VALUE on the section
1545 symbol, as this messes up linking. Instead, expand the recursive call
1546 inline and change up the appropriate bits to get the 0x8000 offset
1547 in the reloc addend.
1551 * config/tc-sparc.h (tc_fix_adjustable): Permit the difference of
1552 two symbols in the same segment to be adjusted.
1554 * configure.in: Don't get confused by CPU-VENDOR-linux-gnu.
1555 * configure: Rebuild.
1559 * config/tc-ppc.c (ppc_insert_operand): Change most warnings into
1561 (ppc_elf_validate_fix): Ditto.
1562 (md_assemble): Ditto.
1564 (ppc_pe_section): Ditto.
1565 (ppc_frob_symbol): Ditto.
1569 * config/mn10300.c (md_assemble): Pass an extra shift count
1570 to mn10300_insert_operand based on the opcode format.
1571 (mn10300_insert_operand): Accept and use extra shift count
1574 * config/tc-mn10300.c (md_assemble): Use FMT_* macros for
1575 formats rather than hard-coded constants.
1577 * config/tc-mn10300.c (md_assemble): Format D5 instructions
1578 are 7 bytes long. Write out instructions in big-endian format.
1582 * config/tc-mn10300.c (md_assemble): Tweak further so
1583 that all instructions are parsed correctly.
1587 * as.h: Include libiberty.h.
1588 (xmalloc, xrealloc): Don't declare.
1589 * as.c: Don't include libiberty.h.
1590 * expr.c, read.c, stabs.c, config/obj-coff.c: Likewise.
1591 * config/tc-mips.c: Likewise.
1592 * messages.c: Likewise.
1593 (xstrerror): Don't declare.
1594 * xmalloc.c: Remove.
1598 * config/tc-mn10300.h (pre_defined_registers) Remove.
1599 (system_registers, cc_names): Likewise.
1600 (address_registers, data_registers, other_registers): New register
1602 (register_name, system_register_name, cc_name): Remove.
1603 (mn10300_reloc_prefix): Likewise.
1604 (data_register_name): New function.
1605 (address_register_name, other_register_name): Likewise.
1606 (md_assemble): Rough cut at parsing operands. Remove lots of
1608 (md_apply_fix3): Disable for now.
1612 * config/tc-m68k.c (select_control_regs): New function, extracted
1613 out of m68k_init_after_args.
1614 (m68k_init_after_args): Use it.
1615 (mri_chip): Use it here as well to update set of allowed control
1620 * config/obj-elf.c (elf_begin): New function.
1621 (obj_elf_section): Add the section symbol to the symbol table.
1622 * config/obj-elf.h (obj_begin): Define.
1623 (elf_begin): Declare.
1624 * as.c (perform_an_assembly_pass): Call obj_begin if it is
1629 * config/obj-coff.c (fixup_segment): Subtract the section address
1630 from a PC relative reloc if TC_M68K.
1634 * config/tc-sparc.c (md_pseudo_table): Make .uahalf, .uaword, and
1635 .uaxword available even if not OBJ_ELF.
1636 (md_atof): Remove unused local variable wordP.
1640 * config/tc-mn10x00.c, config/tc-mn10x00.h: New files
1641 for Matsushita MN10x00 support.
1642 * configure.in: Recognize mn10x00-*-*
1643 * configure: Rebuilt.
1647 * obj-evax.h: move openvms definitions from here to tc-alpha.c.
1648 * tc-alpha.c: add support for vms_case_hack like in vax/vms.
1649 (load_expression): track clobbering of base reg before jmp/jsr.
1650 (s_alpha_file): pass case_hack flags and source filename via
1651 symbol table to bfd.
1652 * tc-alpha.h (TC_CONS_FIX_NEW): define
1656 * configure.in (mips-*-rtems*): New target, like mips-*-elf*.
1657 * configure: Rebuild.
1661 * read.c (s_macro): Warn if a macro has the same name as a
1663 (s_space): In m68k MRI mode, align to a word boundary.
1664 * macro.c (define_macro): Add namep parameter. Change all
1666 * macro.h (define_macro): Update declaration.
1668 * as.c (show_usage): Print bug report address.
1669 (parse_args): Change version printing to match current GNU
1671 * gasp.c (show_usage): Print bug report address.
1672 (main): Change version printing to match current GNU standards.
1674 * config/tc-m68k.c (init_table): Correct access control unit
1677 * config/tc-alpha.c: Add some static function prototypes.
1678 (alpha_macros): Move to top of file. Make static.
1679 (alpha_num_macros): Move to top of file.
1684 * tc-v850.h: Define LOCAL_LABEL to recognise _.L_* symbols
1690 * listing.c (list_symbol_table): Remove bogus code in BFD64 case,
1691 and just call sprintf_vma.
1695 * expr.c (expr): Change >>= to >> (fix typo). (From meissner).
1699 * read.c (float_cons): Call md_flush_pending_output if it is
1704 * config/tc-d10v.c (md_operand): Created. Allows operands to
1706 * config/tc-d10v.h (md_operand): Undefined.
1710 * config/tc-m68k.c (add_fix): Treat a width of '3' like 'B'.
1711 (md_assemble): A fixup width of '3' means a 1 byte reloc.
1715 * config/obj-coff.c (fixup_segment): Don't adjust PC relative
1716 reloc for the i960 for a reloc in the same section. This undoes
1717 one of the two changes made Aug 19.
1721 * config/obj-coff.c (obj_coff_endef): Both versions: Move C_STAT
1722 symbols to the position of the debugging information.
1726 * expr.c (expr): Always use unsigned right shifts for >>.
1730 * config/tc-arm.c (md_apply_fix3): Update two thumb instruction
1731 slots when processing BL fixups.
1733 * config/tc-arm.c (output_inst): Ensure Thumb BL fixup is marked
1734 on the first half of the instruction.
1738 * ecoff.c (ecoff_stab): Create an expression symbol for a complex
1739 stabs expression, rather than giving an error.
1741 * ecoff.c (ecoff_new_file): Don't do anything if we are still in
1746 * config/tc-mips.c (append_insn): Fill in the value for a constant
1747 jump, rather than creating a reloc.
1751 * config/tc-mips.c (append_insn): Don't swap an instruction which
1752 sets a condition code with an instruction which uses a condition
1754 (mips_ip): In cases 'N' and 'M', look for $fccN rather than an
1757 * config/tc-mips.c (md_begin): Recognize r5000 for cpu.
1758 (mips_ip): Give a better error message if the ISA level is wrong.
1759 (md_parse_option): Recognize -mcpu=[v][r]5000.
1763 * config/tc-mips.c (COUNT_TOP_ZEROES): Added macro to count
1765 (load_register): Ensure hi32 bits are not lost during lo32bit
1766 processing. Fix shift offset that was overflowing into the next
1767 instruction field. Add code to generate shorter sequences for
1768 constants with a single contiguous seqeuence of ones.
1772 * config/tc-d10v.c (d10v_dot_word): New function to support
1773 "@word" with the word pseudo-op.
1774 (md_apply_fix3): Cleanup and changes to support correct sizes
1775 for 16 and 18-bit relocs.
1779 * configure.in (sparc-*-aout): Set `em'.
1780 * configure: Regenerated.
1781 * config/te-sparcaout.h: New file.
1782 * config/tc-sparc.h (TARGET_BYTES_BIG_ENDIAN): Define.
1783 Ifdef TE_SPARCOUT define TARGET_FORMAT and SPARC_BIENDIAN.
1784 * config/tc-sparc.c (INSN_BIG_ENDIAN): New macro.
1785 (SPECIAL_CASE_{SETSW,SETX}): Define.
1786 ({NOP,OR,FMOVS,SETHI,SLLX,SRA}_INSN): Define.
1787 (md_begin): Delete setting of `target_big_endian'.
1788 (output_insn): New function.
1789 (md_assemble): Rewrite. Add `setx' support.
1790 (sparc_ip): Handle `0' operand char. Recognize setuw, setsw, setx
1792 (md_atof): Add little endian support.
1793 (md_number_to_chars): Likewise.
1794 (md_apply_fix): Likewise.
1795 (md_longopts): Recognize -EL,-EB ifdef SPARC_BIENDIAN.
1796 (md_parse_option): Likewise.
1797 (md_show_usage): Print -EL, -EB ifdef SPARC_BIENDIAN.
1801 * ecoff.c (ecoff_new_file): New function.
1802 * ecoff.h (ecoff_new_file): Declare.
1803 * config/obj-ecoff.h (obj_app_file): Define.
1807 * config/tc-alpha.c (load_expression): Bias the .lit8 section
1808 symbol by 32k so that our 16-bit signed offset can address the
1813 * config/tc-mips.c (load_register): Remove unused variable tmp.
1817 * config/tc-mips.c (load_register): Remove unnecessary code that
1818 was causing the high 32bits of 64bit constants to be lost.
1822 * config/tc-d10v.c: Added changes to support function
1823 pointers and "@word" syntax.
1828 * config/tc-v850.c: Remove commented out and #if 0'd code.
1829 (v850_reloc_prefix): Provide prototype.
1830 (postfix, get_reloc, build_insn): Remove prototypes for nonexistant
1832 (md_begin, md_assemble, md_apply_fix3): Remove unused variables.
1833 (md_assemble): Add default to case statement.
1837 * config/tc-v850.c (md_assemble): Compute size of the instrction
1840 * config/tc-v850.c (md_apply_fix3): Do simple byte, short and
1845 * config/tc-v850.c (md_apply_fix3): Use little endian get/put
1846 routines to fetch/store the updated instruction from/to memory.
1847 (v850_insert_operand): If the operand has a specialized insert
1852 * config/tc-v850.c (reg_name_search): Align calling convention to
1853 be like identical function found in tc-ppc.c.
1854 (get_reloc): Removed.
1855 (v850_reloc_prefix): New function, parse lo(), hi() and hi0().
1856 (md_assemble): emit fixups.
1857 (md_pcrel_from): renamed from md_pcrel_from_section, emit proper
1859 (md_apply_fix3): handle fixups/relocs.
1860 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Removed definition.
1866 * configure.in (sh-*-elf*): New target.
1867 * config/tc-sh.h (TARGET_ARCH): Define.
1868 (WORKING_DOT_WORD): Define.
1869 (TC_COFF_FIX2RTYPE): Only define if OBJ_COFF.
1870 (BFD_ARCH, COFF_MAGIC, TC_COUNT_RELOC): Likewise.
1871 (TC_RELOC_MANGLE, tc_coff_symbol_emit_hook): Likewise.
1872 (DO_NOT_STRIP, NEED_FX_R_TYPE, TC_KEEP_FX_OFFSET): Likewise.
1873 (TC_COFF_SIZEMACHDEP, tc_frob_file): Likewise.
1874 (SUB_SEGMENT_ALIGN): Likewise.
1875 (RELOC_32): Don't define.
1876 (tc_frob_file_before_adjust): Define if BFD_ASSEMBLER.
1877 (target_big_endian): Declare if OBJ_ELF.
1878 (TARGET_FORMAT): Define if OBJ_ELF.
1879 * config/tc-sh.c: Use BFD reloc codes instead of SH COFF reloc
1881 (tc_crawl_symbol_chain): Only define if OBJ_COFF.
1882 (tc_headers_hook, tc_coff_sizemachdep): Likewise.
1883 (struct sh_count_relocs): Define.
1884 (sh_count_relocs): New static function, broken out of
1885 sh_frob_file. Add BFD_ASSEMBLER code.
1886 (sh_frob_section): Likewise.
1887 (sh_frob_file): Call sh_frob_section.
1888 (md_convert_frag): If BFD_ASSEMBLER, change type of headers, and
1889 call section_symbol rather than seg_info (seg)->dot.
1890 (md_section_align): Add OBJ_ELF version.
1891 (SWITCH_TABLE_CONS): Define.
1892 (SWITCH_TABLE): Use SWITCH_TABLE_CONS.
1893 (md_apply_fix): Change parameter types if BFD_ASSEMBLER. Only
1894 handle fx_r_type == 0 if not BFD_ASSEMBLER. Return 0 if
1896 (struct reloc_map): Define if not BFD_ASSEMBLER.
1897 (coff_reloc_map): Likewise.
1898 (sh_coff_reloc_mangle): Use coff_reloc_map to convert fx_r_type.
1899 (tc_gen_reloc): New function if BFD_ASSEMBLER.
1900 * write.c (write_relocs): Ifdef out fx_where test which triggers
1901 inappropriately for SH ELF.
1902 (write_object_file): Call tc_frob_file_before_adjust and
1903 obj_frob_file_before_adjust if they are defined.
1905 * write.c (write_object_file): Use BFD_RELOC_16, not
1906 BFD_RELOC_NONE, when calling fix_new_exp for a broken word.
1908 * read.c (emit_expr): Fix conversion of byte count to BFD reloc
1913 * config/tc-d10v.c (find_opcode): Fix problem with calculating
1914 branch sizes in across sections.
1919 * config/tc-850.c (md_assemble): Handle hi() correctly. Handle
1924 * config/tc-v850.c (md_begin): Deal with end of opcode
1930 * config/tc-d10v.c (find_opcode): Fix a bug which could generate
1931 the wrong opcode for cases like st2w where there are many forms
1932 of the same instruction.
1936 * expr.c (operand): If md_parse_name is defined, call it before
1937 calling symbol_find_or_make.
1938 * config/tc-ppc.h (md_parse_name): Define.
1939 (ppc_parse_name): Declare.
1940 * config/tc-ppc.c (reg_name_search): Add regs and regcount
1942 (register_name): Update call to reg_name_search.
1943 (cr_operand): New static variable.
1944 (cr_names): New static const array.
1945 (ppc_parse_name): New function.
1946 (md_assemble): If PPC_OPERAND_CR is set in the operand flags, set
1947 cr_operand before calling expression.
1951 * config/tc-hppa.c (tc_gen_reloc): Add new argument to
1952 hppa_gen_reloc_type call.
1956 * config/tc-d10v.c: Fixed ".word". Fixed problem with range checking
1957 on addresses. Improved error messages.
1958 * doc/c-d10v.texi: Added docs for register pairs.
1962 * config/tc-d10v.c (parallel_ok): Fix bug in parallel
1967 * ecoff.c (init_file): Initialize fMerge to 1.
1968 (add_file): Restore old file merging code, but only merge files if
1970 (ecoff_directive_loc): Clear fMerge field of current file.
1971 (ecoff_generate_asm_lineno): Likewise.
1975 * doc/c-d10v.texi: Fix typo.
1980 * config/tc-v850.c (md_assemble): Correct bit masking for
1981 hi and lo expressions.
1983 * config/tc-v850.c (md_assemble): Rough cut at demanding
1984 "ep" or "r30" in sst and sld instructions.
1985 (md_apply_fix3): Don't abort. Just warn that we don't
1988 * config/tc-v850.c (CC_NAME_CNT): Define.
1989 (cc_name): New function.
1990 (md_assemble): Handle V850_OPERAND_CC correctly.
1992 * config/tc-v850.c (md_assemble): Don't forget to initialize
1995 * config/tc-v850.c (reg_name_search): Generalize to search
1996 any given register table.
1997 (register_name): Pass appropriate table and size to reg_name_search.
1998 (system_register_name): New function.
1999 (SYSREG_NAME_CNT): Define.
2000 (md_assemble): Handle operands which are system registers.
2002 * config/tc-v850.c (md_assemble): If we find a register, but the
2003 opcode doesn't want a register, then we don't have a match.
2004 (md_assemble): Get size of the instruction from the opcode table.
2009 * configure.in: Set and substitute HLDENV.
2010 * configure: Rebuild.
2011 * Makefile.in (HLDENV): New variable.
2012 (as.new): Use $(HLDENV).
2014 * ecoff.c (ecoff_directive_endef): Avoid a division by zero error
2015 if an array dimension is not known.
2019 * config/tc-d10v.c: Fix a reloc bug caused by my last change.
2020 * doc/c-d10v.texi: Cleanup.
2025 * config/tc-v850.c: New file.
2026 * config/tc-v850.h: New file.
2027 * configure (v850-*-elf): New target.
2028 * configure.in (v850-*-elf): New target.
2033 * doc/c-d10v.texi: New file.
2034 * doc/all.texi: Added D10V stuff.
2035 * doc/as.texinfo: Added D10V stuff.
2039 * config/tc-d10v.c: All references to defined symbols should
2040 now use the optimal instruction. .float and .double now work.
2044 * config/obj-coff.c (fixup_segment): Adjust PC relative reloc by
2045 section address for the i960 as is done for the i386.
2049 * mpw-config.in: Add wildcards for config matching, add mips-*-*
2050 case, forward-include bfd/elf-bfd.h.
2054 * config/tc-d10v.c: Add additional information to the opcode
2055 table to help determinine which instructions can be done
2060 * config/tc-arm.c: Major changes to add Thumb support, with lots
2062 Reverted to INSN_SIZE macro, rather than insn_size variable.
2063 (insns): Added ARM "bx" instruction support.
2064 (tinsns): Added Thumb instruction definition structure.
2065 (arm_tops_hsh): Added hash structure for Thumb opcodes.
2066 (md_pseudo_table): Added ".arm", ".thumb" and ".code" pseudo-ops.
2067 (opcode_select,s_arm,s_thumb,s_code): Added.
2068 (decode_shift): Allow upper-case RRX.
2069 (do_ldst): Simpler halfword support.
2070 (do_ldmstm): Improved.
2071 (reg_list, do_bx, thumb_reg, thumb_add_sub, thumb_shift,
2072 thumb_mov_compare, thumb_load_store, do_t_arit, do_t_add,
2073 do_t_asr, do_t_branch, do_t_bx, do_t_compare, do_t_ldmstm,
2074 do_t_ldrb, do_t_ldrh, do_t_lds, do_t_lsl, do_t_lsr, do_t_mov,
2075 do_t_push_pop, do_t_str, do_t_strb, do_t_strh, do_t_sub, do_t_swi,
2077 (md_apply_fix3): Add support for BFD_RELOC_ARM_THUMB_* relocations.
2078 (md_parse_option): Add support for -mthumb.
2079 (md_show_usage): Updated to reflect new command line option.
2080 (arm_data_in_code, arm_canonicalize_symbol_name): Added.
2081 * config/tc-arm.h: Provide TC_FIX_TYPE to allow private ARM
2082 fragment information to be held.
2086 * tc-arm.c (md_apply_fix3): Also set fixP->fx_done if fx_addsy is
2087 non-null, but is a constant.
2088 (fix_new_arm): Call make_expr_symbol to make the expression symbol
2089 so that error reporting will work correctly.
2093 * config/tc-i386.c (tc_i386_fix_adjustable): Don't adjust relocs
2094 against weak symbols.
2098 * config/tc-ppc.h (TC_FORCE_RELOCTION): Define if OBJ_XCOFF.
2099 (ppc_force_relocation): Declare if OBJ_XCOFF.
2100 * config/tc-ppc.c (ppc_force_relocation): New function if
2105 * config/tc-mips.h (BYTE_ORDER): Don't define. No longer used.
2109 * config/tc-d10v.c: Fix problem with relocs.
2113 * config/tc-sh.c (sh_do_align): If not BFD_ASSEMBLER, always align
2114 with nops if not in data_section or bss_section.
2118 Add support for openVMS/Alpha.
2119 * as.h (PRINTF_LIKE): Don't define if VMS, for now.
2120 * config/obj-evax.c: New file.
2121 * config/obj-evax.h: New file.
2122 * config/tc-alpha.c: Add support for EVAX format if OBJ_EVAX is
2124 * config/tc-alpha.h: Add support for EVAX format if OBJ_EVAX is
2125 defined. Add case for bfd_target_evax_flavour.
2126 * config/vms-a-conf.h: New file.
2127 * conf-a-gas.com: New file.
2128 * configure.in: Add target alpha-*-*vms*.
2129 * configure: Rebuild.
2130 * makefile.vms: New file.
2131 * read.c (s_lcomm): Align bss_seg on 8 byte boundary if OBJ_EVAX.
2132 Don't call ffs on openVMS/Alpha.
2136 * configure.in: Make GAS_CHECK_DECL_NEEDED include <string.h> or
2137 <strings.h> if they exist. Call GAS_CHECK_DECL_NEEDED on strstr
2139 * acconfig.h (NEED_DECLARATION_STRSTR): New macro.
2140 (NEED_DECLARATION_SBRK): New macro.
2141 * configure, conf.in: Rebuild.
2142 * as.h: Only include <strings.h> if HAVE_STRINGS_H.
2143 (strstr): Declare if NEED_DECLARATION_STRSTR.
2144 * as.c: If HAVE_SBRK and NEED_DECLARATION_SBRK, declare sbrk.
2148 * symbols.c (resolve_symbol_value): Handle addition or subtraction
2149 by a constant before entering the main switch. Reject attempts to
2150 apply an arithmetic function to non-absolute symbols, except for
2151 the special case of subtraction of two symbols in the same
2154 * config/tc-mips.c (md_section_align): Do align if OBJ_ELF, but
2155 not to more than a 16 byte boundary.
2157 * config/tc-i386.c (tc_gen_reloc): Accept all relocs; remove
2159 (tc_gen_reloc): If out of memory call as_fatal rather than
2160 assert. If no howto found, call as_bad_where rather than
2161 as_fatal. Change the error message slightly. Set howto to a
2162 non-NULL value in order to keep going.
2166 * config/tc-d10v.c: Added code to support 32-bit fixups for stabs.
2170 * config/tc-h8300.c (get_specific): New operand "size" derived
2171 from ".b", ".w" and ".l" extensions. All callers changed. If
2172 the base instruction has no operands, then use the size to
2173 determine which specific instruction to use.
2177 * config/tc-i960.c (mem_fmt): Call parse_expr before emit.
2181 * config/tc-mips.c (md_section_align): Don't change addr if
2186 * config/tc-hppa.c: Revert yesterday's changes.
2190 * config/tc-d10v.c: Disable range checking on 16-bit values.
2194 * config/tc-m68k.c (m68k_ip): Set ok_arch for every instruction,
2195 not just the ones that don't match.
2199 * config/tc-d10v.c: Fixed bugs in short relocs and range checking.
2203 * config/tc-arm.c: Changed INSN_SIZE to variable insn_size, as
2204 pre-cursor to adding Thumb support. Also added cpu_variant flag
2205 information to each of the asm_flg structures.
2206 (md_parse_option): Updated ARM7 parsing to allow 't' for
2207 thumb/halfword support, aswell as 'm' for long multiply.
2208 (md_show_usage): Updated help message.
2209 (md_assemble): Check that instruction flags are applicated to the
2210 current cpu variant.
2211 (md_apply_fix3, tc_gen_reloc): Add BFD_RELOC_ARM_OFFSET_IMM8 and
2212 BFD_RELOC_ARM_HWLITERAL relocation support for new halfword and
2213 signextension instructions.
2214 (do_ldst): Generate halfword and signextension variants if
2215 mnemonic flags match.
2216 (ldst_extend): Do not allow shifts in the offset field of halfword
2217 or signextension instructions.
2218 (validate_offset_imm): Provide check on halfword and signextension
2220 (add_to_lit_pool): Merge identical literal pool values.
2224 * config/tc-hppa.c (selector_table): Add 'E' selector.
2225 (cons_fix_new_hppa): Don't coke on e_esel.
2226 (tc_gen_reloc, SOM version): Handle R_COMP2 when used
2227 to help generate exception handling tables.
2228 (md_apply_fix): Don't try to apply fixups with an e_esel
2230 (hppa_fix_adjustable): Fixups with e_esel selectors
2235 * config/tc-sparc.c (md_pseudo_table): Add 2byte, 4byte, and 8byte
2240 * config/tc-d10v.c: Added lots of error checking. Added hacks
2241 to support accumulator shifts.
2245 * symbols.c (S_SET_EXTERNAL): Let .weak override.
2246 (S_CLEAR_EXTERNAL): Likewise.
2247 (S_SET_WEAK): Remove error; just let .weak override.
2251 * config/tc-d10v.c (md_assemble): Now handles multiline
2256 * config/tc-d10v.c: Fix packaging bug. Added range checking.
2257 Added kludge for divs instruction. Fixed minor problem with
2258 multiple text sections.
2259 * config/tc-d10v.h (d10v_cleanup): Change prototype.
2263 * config/tc-d10v.c (md_apply_fix3): Fix all instruction
2264 addresses to be right-shifted by 2.
2268 * config/tc-d10v.c: Many changes to get relocs working.
2269 (register_name): No longer creates a symbol for register names.
2270 (pre_defined_registers): moved to opcodes/d10v-opc.c.
2271 (d10v_insert_operand): Now works correctly for either container.
2272 * config/tc-d10v.h (d10v_cleanup): Declare.
2276 * config/tc-mips.c (tc_gen_reloc): BFD_RELOC_PCREL_HI16_S and
2277 BFD_RELOC_PCREL_LO16 are expected to be PC relative.
2281 * tc-alpha.c: Patches to track current minimum alignment to reduce
2282 the number of fragments created with frag_align.
2283 (alpha_current_align): New static variable.
2284 (s_alpha_text): Reset alignment to 0.
2285 (s_alpha_data, s_alpha_rdata, s_alpha_sdata): Likewise.
2286 (s_alpha_stringer, s_alpha_space): New functions.
2287 (s_alpha_cons, alpha_flush_pending_output): Remove functions.
2288 (alpha_cons_align): New function to replace both of them.
2289 (emit_insn): Only align if alpha_current_align is less than 2;
2290 reset alpha_current_align to 2.
2291 (s_alpha_gprel32): Likewise.
2292 (s_alpha_section): New function. Basically duplicate the other
2293 alpha section change hooks. Only define for ELF.
2294 (s_alpha_float_cons): Simplify alignment handling.
2295 (md_pseudo_table): Only define "rdata" and "sdata" if OBJ_ECOFF.
2296 If OBJ_ELF, define "section", "section.s", "sect", and "sect.s".
2297 Don't define the s_alpha_cons pseudo-ops. Do define
2298 s_alpha_stringer and s_alpha_space pseudo-ops.
2299 (alpha_align): Skip if less than current default alignment. Set
2301 * tc-alpha.h (md_flush_pending_output): Remove.
2302 (md_cons_align): Add.
2304 * tc-alpha.c: Add oodles of function description comments.
2305 (md_bignum_to_chars): Remove; there are no callers.
2306 (md_show_usage): Mention some more variants.
2311 * configure.in (sparc-*-sysv4*): New target.
2312 * configure: Rebuild.
2314 * config/tc-sparc.c (md_pseudo_table): Change uahalf, uaword, and
2315 uaxword to use s_uacons.
2316 (sparc_no_align_cons): New static variable.
2317 (s_uacons): New static function.
2318 (sparc_cons_align): If sparc_no_align_cons is set, just clear it
2321 * config/tc-sparc.c (s_common): Remove unused label allocate_bss.
2323 * configure.in: Add mips-*-irix6* target. Handle Irix 6 like Irix
2324 5 with regard to shared libraries.
2325 * configure: Rebuild.
2327 * config/tc-m68k.c (m68k_ip): Use the correct length when
2328 allocating space for the unsupported architecture error message.
2332 * configure.in (d10v-*-*): Allow d10v-*-*, don't require d10v-*-elf*.
2336 * config/tc-d10v.c: New file.
2337 * config/tc-d10v.h: New file.
2338 * configure (d10v-*-elf): New target.
2339 * configure.in (d10v-*-elf): New target.
2343 * config/tc-ppc.c (md_parse_option): Recognize -K PIC.
2347 * config/tc-alpha.c (alpha_align): Change fill parameter
2348 to a pointer. Take NULL as 0 or nop depending on section. Change
2350 (s_alpha_align): Rename local variables.
2352 * doc/as.texinfo (.align): Document action of omitted
2357 * config/tc-ppc.c (md_apply_fix3): Give a useful error message
2358 when an unsupported PC relative reloc is seen, rather than calling
2361 * app.c (do_scrub_chars): Remove not_cpp_line local variable.
2362 Instead, check state when '#' comment is seen.
2366 * config/tc-mips.c (mips_regmask_frag): Only define if OBJ_ELF or
2368 (tc_gen_reloc): If fixup was changed to be PC relative, change
2369 reloc type accordingly. Use name of reloc in error message.
2371 * as.h: Don't define const or volatile.
2372 * flonum.h: Don't define const.
2374 * config/tc-m68k.c (tc_gen_reloc): Change the code appropriately
2375 if fx_pcrel is set. Correct setting the addend case in the
2376 OBJ_ELF case (from Andreas Schwab
2378 (md_show_usage): Correct -mfc5200 to -m5200.
2382 * doc/c-m68k.texi: Document -m5200 flag.
2383 * doc/as.texinfo: Likewise.
2385 * config/tc-m68k.c (m68k_ip): The coldfire does not support 8x
2390 * symbols.c (S_SET_EXTERNAL): Change as_warn to as_bad.
2391 (S_CLEAR_EXTERNAL, S_SET_WEAK): Likewise.
2395 * Makefile.in (VERSION): Set to cygnus-2.7.1.
2397 * Released binutils 2.7.
2401 * config/tc-mips.c (mips_ip): Only perform range check when
2402 dealing with O_constant expressions.
2406 * m68k-parse.h (m68k_register): Add new coldfile control
2409 * config/tc-m68k.c (mcf5200_control_regs): New variable,
2410 array of control registers for the coldfire.
2411 (cpu_of_arch): Added mcf5200.
2412 (archs): Added mcf5200.
2413 (init_table): Add new control registers.
2414 (m68k_ip): Added support for new control registers.
2415 (m68k_init_after_args): Likewise.
2417 * config/tc-m68k.c (md_show_usage): Add -m5200 to usage text.
2421 * read.h (is_it_end_of_statement): Declare.
2422 * read.c (is_it_end_of_statement): Remove declaration.
2424 * config/tc-ppc.c (ppc_elf_suffix): Correct parenthesization of ||
2426 (md_assemble): Fix handling of @l with an unsigned constant. Add
2427 default case to reloc switch.
2429 * config/tc-i386.h (AOUT_MACHTYPE): Define as 0 if TE_386BSD.
2432 * ecoff.c (last_lineno): New static variable.
2433 (add_procedure): Set last_lineno.
2434 (ecoff_directive_loc): Likewise.
2435 (ecoff_generate_asm_lineno): Likewise.
2436 (ecoff_fix_loc): New function.
2437 * ecoff.h (ecoff_fix_loc): Declare.
2438 * config/tc-mips.c (append_insn): When inserting nops, and using
2439 ECOFF debugging, call ecoff_fix_loc.
2443 * config/tc-h8300.c (build_bytes): If an operand type is
2444 marked as SRC_IN_DST retrieve it from the "destination" op.
2448 * configure.in (arm-*-riscix*): Set emulation to riscix.
2449 * configure: Rebuild.
2450 * config/te-riscix.h: New file to define TE_RISCIX.
2452 * config/tc-sh.h (SUB_SEGMENT_ALIGN): Define.
2456 * Makefile.in (config.status): Just run config.status as other
2461 * mpw-config.in (TARGET_OS): Add definition to conf.
2465 * config/tc-mips.c (append_insn): Parenthesize
2466 cop_interlocks expressions.
2470 * listing.c (listing_print): Close the listing file if it is not
2471 stdout. Close the other files opened for the listing.
2473 * config/tc-sparc.h (md_cons_align): Define.
2474 (sparc_cons_align): Declare.
2475 (HANDLE_ALIGN): Define.
2476 (sparc_handle_align): Declare.
2477 * config/tc-sparc.c (sparc_cons_align): New function.
2478 (sparc_handle_align): New function.
2479 * read.c (cons_worker): Call md_cons_align if it is defined.
2481 * as.h (struct frag): Add fr_file and fr_line fields.
2482 * frags.c (frag_new): Set fr_file and fr_line.
2483 (frag_var): Likewise.
2484 (frag_variant): Likewise.
2486 * as.h (struct frag): Remove unused align_mask and align_offset
2489 * listing.c (calc_hex): Offset by fr_fix when examining fr_var.
2494 * configure.in (mips-*-osf*): New target.
2495 * configure: Rebuild.
2497 * config/tc-m68k.c: Add 68ec060 as a synonym for 68060.
2501 * config/tc-mips.c: Added cop_interlocks, to avoid NOP insertion
2502 between co-processor comparisons and branches for the VR4300.
2506 * Makefile.in (bindir, libdir, datadir, mandir, infodir, includedir,
2507 INSTALL_PROGRAM, INSTALL_DATA): Use autoconf-set values.
2509 * configure.in (AC_PREREQ): autoconf 2.5 or higher.
2510 * doc/Makefile.in (bindir, libdir, datadir, mandir, infodir,
2511 includedir): Use autoconf set values.
2516 * listing.c (listing_eject): Don't do anything if listing is 0.
2517 (listing_list): Likewise.
2518 (listing_source_line): Likewise.
2519 (listing_title): Don't save title if listing is 0.
2520 (listing_source_file): Check listing rather than listing_tail.
2522 * configure.in: On alpha*-*-osf*, link against libbfd.a if not
2523 using shared libraries.
2524 * configure: Rebuild.
2528 * config/tc-mips.c (mips_ip): In case 'i'/'j', don't require an
2529 absolute expression if a relocation type was specified.
2533 * configure.in: Add support for *-*-rtems* configurations.
2534 * configure: Rebuild.
2538 * configure.in: Add alpha-*-linuxecoff* target. Use elf for
2539 alpha-*-linux* target. Force bfd_gas for alpha-*. Require
2540 opcodes library for alpha.
2541 * configure: Rebuild with autoconf 2.10.
2542 * config/tc-alpha.c: Substantial rewrite to add ELF support and
2543 use new opcode table.
2544 * config/tc-alpha.h (md_undefined_symbol): Don't define.
2545 (LOCAL_LABEL): Define differently if OBJ_ELF.
2546 (FAKE_LABEL_NAME): Define if OBJ_ELF.
2547 * config/alpha-opcode.h: Remove.
2548 * config/obj-elf.h: If TC_ALPHA, define ECOFF_DEBUGGING.
2549 * Makefile.in (TARG_CPU_DEP_alpha): Depend upon
2550 include/opcode/alpha.h rather than config/alpha-opcode.h.
2554 * config/obj-aout.c (obj_emit_relocations): Give an error if the
2555 relocation symbol was not resolved.
2556 * config/obj-coff.c (do_relocs_for): Likewise.
2558 * write.c (adjust_reloc_syms): Refetch the symbol section after
2559 calling S_GET_VALUE, since it may have changed.
2561 * expr.c (struct expr_symbol_line): Define.
2562 (expr_symbol_lines): New static variable.
2563 (make_expr_symbol): Add entry to expr_symbol_lines.
2564 (expr_symbol_where): New function.
2565 * expr.h: Use extern on function declarations.
2566 (expr_symbol_where): Declare.
2567 * symbols.c (resolve_symbol_value): Try to use expr_symbol_where
2568 rather than printing the meaningless name of an expression
2573 * config/tc-i386.c (md_number_to_chars): Deleted.
2574 * config/tc-i386.h (md_number_to_chars): New macro.
2576 * config/tc-alpha.c (build_operate_n, build_mem): Moved earlier in
2578 (load_symbol_address, load_expression): Use build_mem.
2579 (build_operate): New function.
2580 (emit_addq_r): Use it.
2584 * symbols.c (colon): #if VMS, use S_SET_OTHER to store `const_flag'.
2588 * config/tc-vax.h (NOP_OPCODE): Define.
2592 * config/obj-vms.h (S_IS_COMMON): Define.
2593 (S_IS_LOCAL): Check for \002 as well as \001.
2594 (LONGWORD_ALIGNMENT): New macro.
2595 (SUB_SEGMENT_ALIGN): Use it.
2599 * config/vms-conf.h: Reconcile with conf.in.
2603 * write.c (is_dnrange): Only define if TC_GENERIC_RELAX_TABLE is
2606 * doc/as.texinfo: Document that any number of hex digits can
2609 * as.c (struct defsym_list): Define.
2610 (defsyms): New static variable.
2611 (parse_args): Just put --defsym arguments on defsyms list, rather
2613 (main): Define defsyms after output file is created.
2615 * config/tc-m68k.c (m68k_ip): Reject PRE and POST indexing mode on
2618 * config/tc-mips.c (mips_ip): In cases 'I', 'i', and 'j', set
2619 insn_error rather than calling check_absolute_expr.
2621 * as.c (emulation_name): Remove unused static variable.
2622 (default_emul_bfd_name): Add return NULL to avoid warning.
2623 * ecoff.c (ecoff_stab): Remove unused variables name and
2625 * frags.c (frag_new): Remove unused variable tmp.
2626 * hash.c (hash_grow): Parenthesize + within <<.
2627 (hash_print_statistics): Use %lu, not %d, to print unsigned
2629 * messages.c: Include "libiberty.h".
2630 (fprint_value): Add cast to avoid printf warning.
2631 (sprint_value): Likewise.
2632 * read.c: Include "ecoff.h".
2633 (emit_expr): Add casts to avoid printf warnings.
2634 * read.h: Use extern for function declarations.
2635 (pop_insert): Declare.
2636 * stabs.c: Include "ecoff.h".
2637 * subsegs.c (subseg_set_rest): Remove unused variables tmp,
2638 former_last_fragP, and new_fragP.
2639 * subsegs.h (subsegs_print_statistics): Declare.
2640 * symbols.c (debug_verify_symchain): Change macro to discard
2642 * write.c (dump_section_relocs): Likewise.
2643 * write.h: Use extern for function declarations.
2644 (write_print_statistics): Declare.
2645 * config/e-mipsecoff.c (mipsecoff_bfd_name): Return NULL to avoid
2647 * config/e-mipself.c (mipself_bfd_name): Likewise.
2648 * config/obj-elf.h (elf_ecoff_set_ext): Declare.
2650 * config/tc-sparc.h (TC_RELOC_RTSYM_LOC_FIXUP): If OBJ_ELF, always
2651 emit relocations against external symbols.
2653 * config/tc-alpha.c (tc_gen_reloc): Output a sensible error
2654 message if bfd_reloc_type_lookup fails, rather than calling
2657 * config/tc-alpha.c (alpha_force_relocation): Add
2658 BFD_RELOC_12_PCREL to switch.
2662 * config/tc-i386.h (LOCAL_LABEL,FAKE_LABEL_NAME): Use defaults for
2663 TE_PE (Lfoo, not .Lfoo).
2667 * read.c (s_fill): Don't warn about a zero repeat count.
2669 * config/tc-mips.c (mips_ip): Don't warn about using AT as a
2670 coprocessor register.
2672 * config/tc-i386.c (md_assemble): When checking the size of a
2673 register to set the size of an instruction, do a bitwise and with
2674 Reg8 and Reg16 rather than requiring the type to be exactly Reg8
2679 * config/tc-h8300.c (parse_reg): Tweak error messages.
2680 (build_bytes): Likewise.
2681 (skip_colonthing): Handle :32 suffix.
2682 (get_specific): Promote L_24 to L_32 if it makes a match.
2683 Don't always promote L_8 to L_16.
2684 (do_a_fix_imm): Clean up L_32 and L_24 handling.
2686 * config/tc-h8300.c (Smode): New variable.
2687 (h8300hmode): Turn off Hmode.
2688 (h8300smode): New function. Turn on Smode and Hmode.
2689 (md_pseudo_table): New ".h8300s" pseudo-op.
2690 (parse_reg): Handle "exr" register.
2691 (get_operand): Handle bizarre syntax for "stm.l" and "ldm.l".
2692 Handle "mach" and "machl" operands for ldmac.
2693 (get_specific): Handle "stm.l" and "ldm.l".
2694 (build_bytes): Handle "stm.l" and "ldm.l"; handle MACREG operands.
2695 * config/tc-h8300.h (COFF_MAGIC): Handle H8/S magic number.
2700 * doc/as.texinfo: Reorder chapter of machine dependent options so
2701 that it is sorted by chip name.
2703 * doc/as.texinfo: Use consistant spelling of Vax.
2704 * doc/c-vax.texi: Likewise.
2708 * config/tc-hppa.c (md_pseudo_table): Add ".begin_try" and ".end_try"
2710 (tc_gen_reloc, SOM version): Handle R_BEGIN_TRY and R_END_TRY.
2711 (md_apply_fix): Likewise.
2712 (pa_try): New function.
2713 (hppa_force_relocation): Force relocs for BEGIN_TRY and END_TRY.
2717 * config/tc-hppa.c (md_pseudo_table): Add ".level" pseudo op.
2718 (pa_level): New function.
2722 * listing.c (listing_newline): Don't do anything if listing is 0.
2726 * subsegs.c (section_symbol): If symbol_table_frozen is set, call
2727 symbol_create, not symbol_new.
2731 * write.c (adjust_reloc_syms): Don't set sy_used_in_reloc for an
2732 absolute symbol unless TC_FORCE_RELOCATION returns true.
2734 * config/obj-coff.c (previous_file_symbol): Remove BFD_ASSEMBLER
2736 (c_dot_file_symbol): BFD_ASSEMBLER version: Don't set the value of
2737 the symbol to a pointer. Don't set previous_file_symbol.
2738 Simplify symbol list rearrangement.
2739 (coff_frob_symbol): Don't do anything with C_FILE symbols.
2740 (coff_adjust_symtab): Don't check previous_file_symbol.
2744 * config/tc-ppc.c (ppc_elf_lcomm): New function for .lcomm
2746 (md_pseudo_table): Add ppc_elf_lcomm.
2750 * config/tc-m68k.c (m68k_ip): Accept ABSL for 'O', so that `bfextu
2751 d0{24:1},d0' works without an immediate prefix on the bit numbers.
2752 (md_begin): Add digits to alt_notend_table.
2753 (md_parse_option): Make s a const pointer.
2755 * config/tc-sparc.c (md_pseudo_table): Add "empty".
2756 (s_empty): New static function.
2758 * config/obj-coff.c (struct filename_list): Only define if not
2760 (filename_list_head, filename_list_tail): Likewise.
2761 (c_section_symbol): Remove unused BFD_ASSEMBLER version.
2762 (obj_coff_endef, BFD_ASSEMBLER version): Don't set the debugging
2763 flag for C_MOS, C_MOE, C_MOU, or C_EOS symbols, since they should
2764 have a section of N_ABS rather than N_DEBUG. If we do a merge,
2765 remove the new symbol from the list.
2766 (obj_coff_endef, both versions): Call tag_insert even if there is
2767 an old symbol with the same name, if the old symbol does not
2769 (coff_frob_symbol): Check SF_GET_TAG, C_EOF, and C_FILE outside of
2770 the SF_GET_DEBUG condition. Don't call SA_SET_SYM_ENDNDX with a
2771 symbol that will be moved to the end of the symbol list.
2772 (coff_adjust_section_syms): Always call section_symbol for .text,
2774 (coff_frob_section): Likewise. Also, remove unused variable
2777 * config/tc-ns32k.c (convert_iif): Call frag_grow rather than
2778 manipulating frags directly.
2779 (md_number_to_field): Adjust mem_ptr correctly if ENDIAN is
2782 * app.c (do_scrub_chars): If '/' is LINE_COMMENT_START, check
2783 whether the next character is '*' before checking whether we are
2784 at the start of a line. Permit LINE_COMMENT_START to start a
2785 comment in state 1 (seen some whitespace) as well, to match the
2788 * gasp.c (do_align): Permit a fill value for .align.
2792 * read.c (next_char_of_string): Warn if a newline is seen in the
2793 middle of a string. Call bump_line_counters when appropriate.
2797 * symbols.c (colon): Use LOCAL_LABEL.
2801 * Makefile.in (install): Don't check to see if tooldir exists.
2802 Make $(tooldir) and $(tooldir)/bin.
2806 * config/ppc-sol.mt (TDEFINES): Don't turn on -mregnames by
2811 * config/tc-mips.c (mips_ip): Don't call as_warn if we are setting
2812 insn_error. Don't put the string "ERROR" in insn_error. Set
2813 insn_error rather than calling as_warn for an unsupported opcode.
2817 * config/tc-mips.c (md_parse_option): Check for a 64 bit format
2818 before permitting -64.
2819 * output-file.c (output_file_create): Remove duplicate
2824 * config/tc-mips.c (md_begin): If -64, create a .MIPS.options
2825 section rather than a .reginfo section.
2826 (mips_elf_final_processing): If -64, write out 64 bit RegInfo
2829 * config/tc-mips.c (load_register): If mips_isa < 3, permit a 32
2830 bit value with the high bit set.
2834 * read.c (s_lcomm): Set section flags for .sbss section.
2836 * config/tc-mips.c (mips_64): New static variable.
2837 (mips_target_format): If mips_64, return elf64 targets rather than
2839 (md_longopts): Add "32" and "64".
2840 (md_parse_option): Handle -32 and -64.
2841 (md_show_usage): Mention -32 and -64.
2842 (cons_fix_new_mips): If mips_64, don't convert an 8 byte reloc to
2847 * config/tc-ppc.c (comment_chars): Make '!' a comment character
2848 for Solaris compatibility.
2850 * stabs.c (s_stab_generic): Under PowerPC Solaris, convert a
2851 .stabd with 4 arguments into a .stabn.
2855 * config/tc-mips.c (macro): When passing X_add_number to
2856 macro_build, cast it to int first.
2860 * config/tc-z8k.c (md_apply_fix): Handle fx_r_type of 0, as
2861 created by emit_expr.
2863 * symbols.c (symbol_create): If bfd_make_empty_symbol fails, call
2864 as_perror rather than assert.
2868 * config/tc-mips.c (mips_ip): Mark sections created to hold
2869 floating point information as read only.
2873 * config/tc-ppc.c (ppc_set_cpu): Change defaults to match AIX.
2877 * read.c (potable): Add .skip as a synonym for .space.
2879 * stabs.c (s_stab_generic): For PowerPC ELF, allow .stabd to take
2880 4 arguments, providing the 4th argument is 0, to allow
2881 compatibility with the Solaris assembler.
2885 * config/tc-sh.h (struct sh_segment_info_type): Define.
2886 (TC_SEGMENT_INFO_TYPE): Define.
2887 (sh_frob_label): Declare.
2888 (tc_frob_label): Define.
2889 (sh_flush_pending_output): Declare.
2890 (md_flush_pending_output): Define.
2891 * config/tc-sh.c (md_assemble): If relaxing, emit a R_SH_CODE
2892 reloc before the instruction if necessary.
2893 (sh_frob_label): New function.
2894 (sh_flush_pending_output): New function.
2895 (sh_coff_frob_file): Ignore ALIGN, CODE, DATA, and LABEL relocs
2896 when looking for the reloc for the target of .uses.
2897 (md_convert_frag): Fix printf format (%0xlx to 0x%lx).
2898 (sh_force_relocation): Force CODE, DATA, and LABEL relocs to be
2900 (md_apply_fix): Ignore CODE, DATA, and LABEL relocs.
2901 (sh_coff_reloc_mangle): Force CODE, DATA, and LABEL relocs to use
2902 the absolute symbol.
2904 * subsegs.h (segment_info_type): Add tc_segment_info_data field if
2905 TC_SEGMENT_INFO_TYPE is defined.
2909 * config/tc-i386.c (md_assemble): Make sure the opcode suffix
2910 matches the register size.
2914 * config/obj-coff.c (count_entries_in_chain): Ignore Fixups with
2916 (do_relocs_for): Likewise.
2917 (fixup_segment): Don't just quit if linkrelax is set. Try to
2918 apply non pc-relative sym1-sym2 fixups, even if linkrelax is
2923 * config/tc-ppc.c (ppc_elf_validate_fix): Allow GOT and section
2924 relative relocations with -mrelocatable. Also allow unfixed
2925 relocs in .ex_shared.
2929 * config/obj-coff.c (yank_symbols): Check that FNAME_OFFSET is
2930 non-zero before assuming this is a long file name.
2931 (w_strings): Likewise.
2932 (c_dot_file_symbol): Set FNAME_OFFSET to 1 for a long file name.
2934 * config/obj-coff.c (w_strings): Move declaration of i inside
2935 #ifdef block which uses it.
2939 * config/tc-h8300.c (do_a_fix_imm): Rename last argument to
2940 "relaxmode". Output relocs which identify various relaxing
2941 possibilities for mov.[bwl] instructions.
2942 (build_bytes): Pass in a relaxing mode to do_a_fix_imm.
2946 * config/tc-arm.h (TC_HANDLES_FX_DONE): Define.
2947 (MD_APPLY_FIX3): Define.
2948 * config/tc-arm.c (my_get_expression): Only watch for bad segments
2950 (md_apply_fix3): Renamed from md_apply_fix.
2951 If pcrel reloc and symbol is in different section, undo effects
2956 * config/tc-hppa.c (hppa_fix_adjustable): Don't adjust
2957 any reloc with an LR% or RR% field selector for SOM.
2961 * Makefile.in: Add subsegs.h to appropriate TARG_CPU_DEP_*
2966 * config/obj-coff.c (coff_frob_symbol): Don't merge a symbol with
2968 (yank_symbols): Likewise.
2972 * subsegs.h (segment_info_type): If MANY_SEGMENTS and not
2973 BFD_ASSEMBLER, add name field.
2974 * config/obj-coff.c: Include "libiberty.h".
2975 (coff_header_append): Handle long section names.
2976 (crawl_symbols): Just use the name field for the symbol name,
2977 without worrying about null byte termination.
2978 (w_strings): Handle long section names.
2979 (write_object_file): Likewise. Also, use the name field, rather
2981 (obj_coff_add_segment): Permit long section names.
2982 (obj_coff_init_stab_section): Use the name field, rather than
2984 (adjust_stab_section): Likewise.
2985 * config/te-pe.h (COFF_LONG_SECTION_NAMES): Define.
2987 * config/tc-i960.c (brtab_emit): Don't set fx_im_disp field.
2988 (mem_fmt): Likewise.
2989 (md_apply_fix): Don't check fx_im_disp field.
2993 * configure.in: Add * after sparc*-*-vxworks.
2994 * configure: Rebuild.
2996 * app.c (do_scrub_begin): If tc_comment_chars is not defined,
2997 define it to comment_chars. Use tc_comment_chars rather than
2999 (do_scrub_chars): Use tc_comment_chars rather than comment_chars.
3000 * config/tc-m68k.h (tc_comment_chars): Define.
3001 (m68k_comment_chars): Declare.
3002 * config/tc-m68k.c (m68k_comment_chars): Rename from
3003 comment_chars. Change into a pointer rather than an array.
3004 (md_longopts): Add "bitwise-or".
3005 (md_parse_option): Handle OPTION_BITWISE_OR.
3006 (md_show_usage): Mention --bitwise-or.
3007 * doc/c-m68k.texi: Document --bitwise-or.
3011 * config/tc-m68k.c (m68k_ip): Prevent attempts to use long offsets
3014 * config/obj-coff.c (obj_coff_section): BFD_ASSEMBLER version:
3015 call demand_empty_rest_of_line. Non BFD_ASSEMBLER version:
3016 correct handling of input line pointer, and call
3017 demand_empty_rest_of_line.
3021 * config/tc-sparc.c (in_bitfield_range): New static function.
3022 (sparc_ip): New cases X,Y. Use SPARC_OPCODE_ARCH_V9_P.
3023 (md_apply_fix, cases BFD_RELOC_32_PCREL_S2,
3024 BFD_RELOC_SPARC_{WDISP16,WDISP19}): Fix undefined code.
3025 (md_apply_fix): New cases BFD_RELOC_SPARC_[56].
3026 (tc_gen_reloc): New cases BFD_RELOC_SPARC_[56].
3030 * config/obj-coff.c: BFD_ASSEMBLER:
3031 (coff_last_bf): New static variable.
3032 (coff_frob_symbol): Set endndx of a .bf symbol.
3034 (obj_coff_endef): Call SF_SET_PROCESS on a .bf symbol.
3035 (last_bfP): New static variable.
3036 (yank_symbols): Set endndx of a .bf symbol.
3040 * config/tc-ppc.c (md_parse_option): Add support for Solaris's -le
3041 and -s options. Add -be for good measure.
3045 * read.c (s_space): Support non-constant fill value. Handle fill
3046 value correctly for a size other than 1.
3050 * config/tc-arm.c (my_get_float_expression): Update call to
3051 gen_to_words, X_PRECISION changed from 6 to 5.
3055 * config/tc-ppc.c (register_name,reg_name_search): Move register
3056 name lookup from PE specific code to all targets. Add support for
3057 -mregnames/-mno-regnames to control whether register names are
3059 (md_assemble): Call register_name for all platforms.
3060 (md_parse_option): Add support for -mregnames/-mno-regnames.
3062 * configure.in (powerpcle*-*-solaris): Add support.
3063 (powerpc*-*-linux): Ditto.
3064 * configure: Regenerate.
3066 * config/ppc-sol.mt: New config file for PowerPC Solaris.
3070 * config/tc-mips.c (mips_frob_file): Permit multiple %hi relocs to
3071 be associated with a single %lo reloc.
3073 * config/tc-mips.c (load_address): Cast X_add_number to valueT
3074 before comparing against MAX_GPREL_OFFSET, so that negative
3075 numbers are handled correctly.
3080 * config/tc-sparc.c (last_insn): New static variable.
3081 (md_assemble): Warn about putting floating point branches in a
3082 delay slot. If architecture is less than v9, insert NOP
3083 instructions between floating point instructions and floating
3084 point branches. (The SunOS assembler does both these operations.)
3085 Save the last instruction opcode.
3086 (sparc_ip): Add pinsn parameter. Change caller.
3088 * config/tc-m68k.c (md_estimate_size_before_relax): Correct check
3089 for byte jump to next instruction to skip empty frags.
3093 * config/tc-alpha.c (alpha_ip): If we are going to call emit_add64
3094 for addq with a 16 bit signed value, just emit a lda instruction
3099 * config/tc-h8300.c (do_a_fix_imm): Don't cut off high bits
3104 * configure.in: Permit --enable-shared to specify a list of
3106 * configure: Rebuild.
3110 * config/tc-h8300.c (get_specific): Remove some #if 0 code.
3111 (build_bytes): Remove all ABSMOV related code; it's unnecessary.
3115 * config/atof-ieee.c: Fix handling of denormalized extended
3116 precision numbers and overflow/underflow detection.
3117 (MAX_PRECISION, X_PRECISION, P_PRECISION): Changed from 6 to 5, to
3118 not include the 16 bit gap in the m68k extended precision format.
3122 * configure.in: Add i386-*-freebsdelf* target; from John Polstra
3124 * configure: Rebuild.
3128 * config/tc-mips.c: Allow non-zero offsets from .sdata symbols to
3129 be accessed using the $gp register.
3130 * config/tc-mips.h (MAX_GPREL_OFFSET): Added.
3134 * config/tc-sparc.c (sparc_md_end): Set bfd machine number to
3135 bfd_mach_sparc_sparclet if current_architecture is sparclet.
3139 * read.c (get_line_sb): Bump line counters based on
3140 input_line_pointer[-1], not *input_line_pointer. Don't bother to
3141 call LISTING_NEWLINE.
3142 (s_macro): Don't call demand_empty_rest_of_line.
3143 * app.c (do_scrub_chars): When handling C style comments, unget
3148 * read.h (enum linkonce_type): Define.
3149 (s_linkonce): Declare.
3150 * read.c (potable): Add "linkonce".
3151 (s_linkonce): New function.
3152 * subsegs.h (segment_info_type): Add linkonce field to
3153 MANY_SEGMENTS && ! BFD_ASSEMBLER section.
3154 * config/obj-coff.h (obj_handle_link_once): Define if TE_PE.
3155 (obj_coff_pe_handle_link_once): Declare if TE_PE.
3156 * config/obj-coff.c: If TE_PE and not BFD_ASSEMBLER, #include
3158 (obj_coff_pe_handle_link_once): New function, defined if TE_PE.
3159 (c_section_symbol): If TE_PE, set the x_comdat field in the aux
3160 entry based on the linkonce field in segment_info.
3161 * doc/as.texinfo: Document .linkonce.
3165 * doc/as.1: Changed to be recognized by catman -w on Solaris.
3169 * stabs.c (s_stab_generic): Call the listing functions before
3170 doing the rest of the processing, which may involve freeing the
3171 string. Pass string, not string + stroff, to OBJ_PROCESS_STAB in
3172 SEPARATE_STAB_SECTIONS case.
3174 * config/tc-hppa.c: Remove nested comment.
3175 (tc_gen_reloc): Move label done inside the ifdef in which it is
3177 (md_apply_fix): Pass pointers to correct types to libhppa.h
3178 functions. Always return a value.
3180 * config/tc-mips.h (tc_frob_file): Define.
3181 (mips_frob_file): Declare.
3182 * config/tc-mips.c (struct mips_hi_fixup): Define.
3183 (mips_hi_fixup_list): New static variable.
3184 (imm_unmatched_hi): New static variable.
3185 (md_assemble): Clear imm_reloc, imm_unmatched_hi, and
3186 offset_reloc. Pass imm_unmatched_hi to append_insn.
3187 (append_insn): Add unmatched_hi parameter. If it is set, add the
3188 new fixup to mips_hi_fixup_list. Change all callers.
3189 (mips_ip): Set imm_unmatched_hi when appropriate.
3190 (mips_frob_file): New function.
3194 * configure.in (sparc-*-solaris2*): Renamed from sparc*-*-solaris2*.
3195 * configure: Regenerated.
3199 * as.c (main): Call bfd_set_error_program_name.
3203 * as.h (strdup): Don't declare.
3204 * stabs.c: Include libiberty.h
3205 (get_stab_string_offset): Use xstrdup rather than strdup.
3206 (s_stab_generic): Likewise.
3207 * as.c (parse_args): Likewise.
3208 * read.c (s_mri_sect): Likewise.
3210 * gasp.c (change_base): Recognize \(...) construct documented to
3211 pass through enclosed characters literally through to the output.
3212 (process_assigns): Likewise. Also, be more careful to avoid
3213 looking past the end of the buffer.
3217 * config/tc-i386.c (md_parse_option): If OBJ_ELF, ignore -k for
3222 * doc/as.texinfo, doc/c-i960.texi: Fix typos.
3226 * config/alpha-opcode.h: Added cvtst instruction.
3230 * ecoff.c (ecoff_stab): Don't try to make a symbol out of the stab
3231 string. Extract the addend from the result of expression.
3235 * app.c (do_scrub_chars): If whitespace is seen in state 11, and
3236 LABELS_WITHOUT_COLONS is not defined, and we are not in m68k MRI
3237 mode, change the state to 3 rather than 1.
3241 * config/obj-coff.h (C_REGISTER_SECTION): Change from 20 to 50, to
3242 correspond to 11 March change.
3246 * config/tc-h8300.c (build_bytes, MEMIND case): Generate
3247 an R_MEM_INDIRECT reloc rather than R_RELBYTE.
3251 * configure: Rebuild with autoconf 2.8.
3255 * config/atof-ieee.c (gen_to_words): Improve handling of
3256 X_PRECISION numbers. Based on patches from Andreas Schwab
3261 * as.h (SEG_NORMAL, SEG_LIST): Bump segment limit from 10 to 40.
3263 * subsegs.c (MANY_SEGMENTS): Increase segment limit.
3264 * obj-coff.c (seg_N_TYPE, seg_info_off_by_4): Likewise.
3265 (do_relocs_for, w_symbols, obj_coff_add_segment, do_linenos_for,
3266 crawl_symbols, coff_header_append): Loop to SEG_LAST rather than
3271 * config/tc-sparc.c (sparc_ip): Handle operand char 'O' (neg reg).
3275 * config/tc-hppa.c (SUBSEG_MILLI): Define.
3276 (pa_def_subspaces): Add $MILLICODE$.
3277 (pa_spaces_begin): Set section flags for $MILLICODE$.
3281 * config/obj-elf.c (obj_elf_section): Only SEC_LOAD if the type is
3282 not SHT_NOBITS. Don't tamper with flags based on type if a
3283 special section was found (revert Feb 29 change).
3285 * config/tc-sh.c (sh_do_align): Only align using the nop pattern
3286 if aligning to a longword boundary or greater.
3290 * config/tc-sh.c (sh_do_align): Pass 1 not 2 to frag_align.
3294 * configure.in (i386-*-cygwin32): Don't use bfd_gas.
3295 * configure: Regenerated.
3299 * config/tc-hppa.c: Add default definitions for R_N0SEL and
3300 R_N1SEL since they're not defined for old versions of hpux.
3302 * config/tc-hppa.c (tc_gen_reloc): Fix typo in R_COMP2 code.
3303 Set "sym_ptr_ptr" and "addend" fields to dummy values for
3304 R_N0SEL and R_N1SEL.
3308 * flonum-konst.c: Add two more constants for 1e+-2048 and
3309 1e+-4096, and correct the other constants.
3311 * symbols.c (resolve_symbol_value): Handle O_logical_not.
3315 * config/obj-elf.c (obj_elf_section): Allow predefined section
3316 types to set the nobits type. Avoid a shadowed declaration.
3320 * config/tc-hppa.c (hppa_fix_adjustable): For SOM, don't
3321 reduce relocs using e_nlrsel field selectors.
3323 * write.c (fix_new_exp): Don't use #elif. Some compilers
3326 * config/tc-hppa.c (selector_table): Add "n", "nl", and "nlr" to
3328 (pa_chk_field_selector): Handle new field selectors for SOM.
3332 * configure.in (m68k-*-linux*aout*, m68k-*-linux*): New targets.
3333 * configure: Rebuild.
3334 * config/te-linux.h (LOCAL_LABELS_FB): Define.
3335 * config/tc-m68k.h (TARGET_FORMAT) [TE_LINUX]: Define to
3337 * config/tc-m68k.c (comment_chars): Don't include '#' if TE_LINUX
3342 * mpw-make.sed: Update to handle shared library support.
3346 * config/tc-sparc.c (sparc_ip): Print all architectures that support
3347 the insn on mismatch.
3351 * configure.in: Add support for a29-coff.
3352 * configure: Rebuild.
3356 * config/tc-sh.c (sh_coff_frob_file): Don't consider the address
3357 of the section when looking for the R_SH_USES fixup, because the
3358 frag addresses have not yet been adjusted.
3360 * gdbinit.in: Set a breakpoint on as_warn_where.
3362 * config/tc-mips.c (macro): Add missing arguments to macro_build
3367 * config/tc-h8300.c (tc_reloc_mangle): Change reloc based on size
3368 if it is TC_CONS_RELOC. Set a size of 4 to R_RELLONG.
3372 * config/tc-sparc.c (sparc_ip): Recognize %asr0 for v8.
3376 * config/tc-sparc.c (parse_keyword_arg): Accept leading '%'.
3377 (sparc_ip): Accept %asr[1..31] for v8 and %asr[%16..31] for v9.
3378 Recognize [uU] format args as sparclet cpregs.
3382 * config/tc-sh.c (sh_handle_align): Don't emit R_SH_ALIGN relocs
3387 * config/tc-sparc.h (TC_RELOC_RTSYM_LOC_FIXUP): Check S_IS_WEAK as
3388 well as S_IS_EXTERNAL.
3389 (tc_fix_adjustable): Likewise.
3390 * config/tc-sparc.c (md_apply_fix): In OBJ_ELF case, check for
3391 S_IS_WEAK as well as S_IS_EXTERNAL when deciding whether to return
3393 (tc_gen_reloc): Check S_IS_WEAK as wel as S_IS_EXTERNAL when
3394 deciding whether to convert BFD_RELOC_32_PCREL_S2 if PIC.
3398 * config/tc-sparc.c (max_architecture): Change to sparclite for
3400 (default_compatible): Delete.
3401 (sparc_ffs): New function.
3402 (md_begin): Only call SPARC_OPCODE_CONFLICT_P once.
3403 (sparc_ip): Rewrite architecture match and bump logic.
3407 * configure.in: Check for 'do not mix' from native linker before
3408 trying to use -rpath.
3409 * configure: Rebuild.
3413 * config/obj-coff.h (SF_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
3414 (SF_GET_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
3415 (SF_SET_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
3416 * config/obj-coff.c (obj_coff_endef): Set ADJ_LNNOPTR when LNNOPTR
3418 (w_symbols): If ADJ_LNNOPTR is set, add the section lnnoptr field
3419 to the symbol lnnoptr field, to get the correct file offset.
3423 * config/obj-elf.c (elf_frob_symbol): On the PowerPC, force all
3424 symbols that are not function, file, or section symbols to be
3429 * configure: Set and substitute RPATH_ENVVAR.
3430 * configure: Rebuild.
3431 * Makefile.in (RPATH_ENVVAR): New variable.
3432 (check): Use $(RPATH_ENVVAR) rather than LD_LIBRARY_PATH.
3435 doesn't need opcodes. If configuring shared, opcodes needs bfd.
3436 * configure: Rebuild.
3440 * read.c (s_mri_sect): Don't return '\0' in type. Set all
3441 appropriate flags in BFD section.
3443 * configure.in (m68k-*-psos*): New target.
3444 * configure: Rebuild.
3445 * config/te-psos.h: New file.
3446 * config/tc-m68k.c (comment_chars): Don't include '#' if TE_PSOS
3452 * configure.in: Remove duplicate setting of cpu_type. Check
3453 whether opcodes library is required for on all targets, not just
3455 * configure: Rebuild.
3457 * config/tc-mips.c (mips_big_got): New static variable.
3458 (s_extern): Don't declare.
3459 (reg_needs_delay): New static function.
3460 (macro_build): Permit GOT/CALL_HI/LO relocs.
3461 (macro_build_lui): If place is not NULL, use the number in the
3463 (load_address): Handle mips_big_got case.
3464 (macro): Handle mips_big_got for M_LA_AB, M_JAL_A, and load and
3466 (OPTION_XGOT): Define.
3467 (md_longopts): Add "xgot" if OBJ_ELF.
3468 (md_parse_option): Handle -xgot.
3469 (md_show_usage): Mention -xgot.
3470 (md_apply_fix): Permit GOT/CALL_HI/LO relocs.
3471 (tc_gen_reloc): Handle GOT/CALL_HI/LO relocs.
3475 * config/tc-m68k.c (m68k_ip) [operand kind '#']: When fixing
3476 the byte relocation, point it to the low byte of the word.
3480 * configure.in: Set HDLFLAGS for *-*-hpux with --enable-shared.
3481 * configure: Rebuild.
3485 * configure.in: Recognize any sparc* cpu.
3486 * configure: Regenerated.
3490 * read.c (potable): Add "mri" and ".mri".
3491 (s_mri): New function.
3492 * read.h (s_mri): Declare.
3493 * app.c (scrub_m68k_mri): New static variable.
3494 (mri_pseudo): New static variable.
3495 (do_scrub_begin): Add m68k_mri parameter. Use it rather than
3496 flag_m68k_mri. Initialize scrub_m68k_mri.
3497 (mri_state, mri_last_ch): New static variables.
3498 (struct app_save): Add scrub_m68k_mri, mri_state, and mri_last_ch
3500 (app_push): Save new fields.
3501 (app_pop): Restore new fields.
3502 (do_scrub_chars): Check scrub_m68k_mri rather than flag_mri_mri.
3503 If TC_M68K, use a trivial state machine to look for occurrences of
3504 the .mri pseudo-op, and change the mode appropriately.
3505 * as.h (do_scrub_begin): Update prototype.
3506 * input-scrub.c (input_scrub_begin): Pass flag_m68k_mri to
3508 * config/tc-m68k.c (reg_prefix_optional_seen): New static
3510 (m68k_mri_mode_change): New function.
3511 (md_parse_option): Set reg_prefix_optional_seen.
3512 * config/tc-m68k.h (m68k_mri_mode_change): Declare.
3513 (MRI_MODE_CHANGE): Define.
3514 * doc/as.texinfo: Document .mri pseudo-op.
3516 * app.c (do_scrub_chars): In MRI mode, don't treat '#' as a
3521 Support for OBJ_ELF on m68k, mostly inside #ifdef OBJ_ELF:
3522 * config/m68k-parse.h (enum pic_relocation): Define.
3523 (struct m68k_exp): Add pic_reloc field.
3524 * config/tc-m68k.h (TC_RELOC_RTSYM_LOC_FIXUP): Define.
3525 (tc_fix_adjustable): Define to call tc_m68k_fix_adjustable.
3526 (NO_RELOC): Define to BFD_RELOC_NONE if BFD_ASSEMBLER, to zero
3528 * config/tc-m68k.c: Delete definition of NO_RELOC.
3529 (struct m68k_it): Add pic_reloc field.
3530 (add_fix): Copy over pic_reloc field.
3531 (md_pseudo_table): Interpret .align parameter as byte count.
3532 (mote_pseudo_table): Likewise.
3533 (tc_m68k_fix_adjustable): New function.
3534 (get_reloc_code): New function.
3535 (md_assemble): Use it as last argument to fix_new_exp.
3536 (md_apply_fix_2): For a relocation against a symbol don't put the
3537 addend into the data.
3538 (tc_gen_reloc): Different addend computation for OBJ_ELF.
3539 (m68k_ip): Don't relax an operand that requires pic relocation.
3540 (md_begin): Align .text, .data and .bss on 4 byte boundary by
3542 * write.c (fixup_segment): Don't add symbol value to addend if
3543 TC_M68K and OBJ_ELF.
3544 * config/m68k-parse.y (yylex): Handle @PLTPC, etc.
3545 (motorola_operand): Add rule for `(zapc, EXPR)'.
3549 * ecoff.c (ecoff_directive_weakext): Fixed so that whitespace
3550 *really* is permissible before the comma.
3554 * config/tc-sh.c (sh_do_align): Align to a 2 byte boundary before
3555 inserting nop instructions.
3559 * config/te-aux.h: Change include of aux.h to aux-coff.h.
3563 * config/tc-i960.c (tc_coff_symbol_emit_hook): Correct storage
3564 class setting for a CALLNAME symbol in COFF.
3566 * read.c (potable): Pass negative numbers for new .balign[wl] and
3567 .p2align[wl] pseudo-ops.
3568 (s_align_bytes): Treat a negative argument as specifying the fill
3570 (s_align_ptwo): Likewise.
3574 * read.c (potable): Add balignw, balignl, p2alignw, and p2alignl.
3575 (do_align): Take new len parameter. Change all callers. Pass it
3577 (s_align_bytes): Arg now indicates the length of the fill pattern.
3578 (s_align_ptwo): Likewise.
3579 * config/obj-coff.c (write_object_file): Pass length to
3581 * config/tc-i386.h (md_do_align): Take new len parameter.
3582 * config/tc-m88k.h (md_do_align): Likewise.
3583 * config/tc-m88k.c (m88k_do_align): Likewise.
3584 * config/tc-sh.h (md_do_align): Likewise.
3585 * config/tc-sh.c (sh_do_align): Likewise.
3586 * doc/as.texinfo: Document new pseudo-ops.
3588 * config/obj-coff.c (fixup_mdeps): Divide offset by fr_var, as is
3589 done in cvt_frag_to_fill.
3591 * config/tc-sh.h (sh_do_align): Declare.
3592 (md_do_align): Define.
3593 * config/tc-sh.c (sh_do_align): New function.
3595 * ecoff.c (ecoff_build_lineno): Don't try to store the address
3596 difference if the next address is before the current one.
3598 * config/tc-m68k.c (struct m68k_cpu): Add alias field.
3599 (archs): Initialize new field.
3600 (m68k_ip): Don't list alias names when listing CPUs which support
3603 * as.c (main): Call parse_args before read_begin.
3604 * app.c (do_scrub_chars): If flag_m68k_mri, don't put a dot in
3605 front of generated pseudo-ops.
3606 * read.c (potable): Ignore "name".
3607 (s_app_file): Permit a single quote after the string, since one
3608 may appear in m68k MRI mode.
3610 * configure.in: Check for --enable-shared. If linking against
3611 shared BFD and opcodes, fix library name on SunOS, and try to set
3613 * configure: Rebuild.
3617 * as.h (flag_m68k_mri): Declare.
3618 * as.c (parse_args): If TC_M68K, set flag_m68k_mri for -M.
3619 * Many files: For MRI syntax that is specific to the m68k MRI
3620 assembler, check flag_m68k_mri rather than flag_mri or
3621 MRI_MODE_NEEDS_PSEUDO_DOT.
3625 * config/tc-i960.c (ARCH_HX): Define.
3627 (targ_has_sfr): Handle ARCH_HX.
3628 (targ_has_iclass): Handle ARCH_HX.
3629 (tc_coff_fix2rtype): Add return 0 to avoid warning.
3630 (tc_headers_hook): If the architecture was specified explicitly,
3631 use it when setting the flags. Set the extern variable coff_flags
3632 rather than headers->filehdr.f_flags, since the latter is set
3633 unconditionally in obj-coff.c.
3634 (i960_handle_align): Remove unused variable fixp.
3636 Support for building bfd and opcodes as shared libraries, based on
3638 * configure.in: Set OPCODES and BFD to search directories.
3639 Substitute OPCODES_DEP and BFDDEP. On SunOS, set HLDFLAGS.
3640 * configure: Rebuild.
3641 * Makefile.in (LDFLAGS, HLDFLAGS): New variables.
3642 (LIBDEPS): New variable.
3643 (as.new0: Depend upon $(LIBDEPS) rather than $(LIBS). Use
3644 $(HLDFLAGS) in link.
3645 (check): Set LD_LIBRARY_PATH in the environment.
3649 * config/tc-ppc.h (ELF_TC_SPECIAL_SECTIONS): Make .sdata2, .sbss2,
3650 .PPC.EMB.sdata0, and .PPC.EMB.sbss0 sections all default to
3651 read-only, not read/write.
3655 * Makefile.in (INSTALL_XFORM): Remove -e.
3659 * write.c (write_relocs): Use S_IS_DEFINED and S_IS_COMMON rather
3660 than comparing S_GET_SEGMENT to undefined_section.
3661 (write_object_file): Skip symbols which were equated to an
3662 undefined or common symbol.
3663 * symbols.c (resolve_symbol_value): Use S_IS_DEFINED and
3664 S_IS_COMMON rather than comparing S_GET_SEGMENT to
3666 (S_GET_VALUE): Likewise. Avoid recursion problems if S_IS_DEFINED
3667 or S_IS_COMMON call S_GET_VALUE.
3668 * config/obj-aout.h (S_IS_COMMON): Define if not BFD_ASSEMBLER.
3669 * config/obj-aout.c (obj_emit_relocations): If a reloc is equated
3670 to an undefined or common symbol, convert the reloc to be against
3672 (obj_crawl_symbol_chain): Skip symbols which were equated to an
3673 undefined or common symbol.
3674 * config/obj-bout.h (S_IS_COMMON): Define if not BFD_ASSEMBLER.
3675 * config/obj-bout.c (obj_emit_relocations): If a reloc is equated
3676 to an undefined or common symbol, convert the reloc to be against
3678 (obj_crawl_symbol_chain): Skip symbols which were equated to an
3679 undefined or common symbol.
3680 * config/obj-coff.c (do_relocs_for): Use S_IS_DEFINED and
3681 S_IS_COMMON rather than comparing S_GET_SEGMENT to
3683 (yank_symbols): Skip symbols which were equated to an undefined or
3688 * config/obj-aout.h (S_IS_LOCAL): Check for \002 as well as \001.
3689 * config/obj-bout.h (S_IS_LOCAL): Likewise.
3691 * configure.in: Make sure we only add m68k-parse.o to
3692 ${extra_objects} once, no matter how many m68k targets have been
3694 * configure: Rebuild.
3698 * configure.in (i386-*-cygwin32, ppc-*-cygwin32): New.
3699 * configure: Rebuild.
3703 * config/tc-m68k.c (md_pseudo_table): Add "extend" and "ldouble".
3704 * doc/c-m68k.texi: Document .extend and .ldouble.
3706 * configure.in (m68*-apple-aux*): New target.
3707 * config/te-aux.h: New file.
3708 * config/obj-coff.c (compare_external_relocs): New static function
3710 (do_relocs_for): Sort relocs if TE_AUX.
3711 (fixup_segment): If TE_AUX, store common symbol value in segment.
3712 * config/tc-m68k.h (TARGET_FORMAT): Define if TE_AUX.
3716 * config/obj-coff.h (S_IS_LOCAL): Check for \002 as well as \001.
3718 * config/tc-mips.c (s_mips_globl): Set BSF_OBJECT if it is not
3720 (s_cpload): Set BSF_OBJECT for _gp_disp symbol.
3721 * read.c (s_lcomm): If S_SET_SIZE is defined, set the size of the
3723 * ecoff.c (add_procedure): Set the BSF_FUNCTION flag.
3724 (ecoff_build_symbols): If S_SET_SIZE is defined, set the size of
3725 an undefined symbol and the size of a function symbol.
3726 * config/obj-elf.c (elf_frob_symbol): If TC_MIPS, set BSF_OBJECT
3727 for all common symbols.
3731 * config/tc-i960.c (parse_memop): In MRI mode, don't use implicit
3734 * expr.c (operand): Accept 0x hex constants in MRI mode if not on
3739 * config/obj-elf.c (obj_elf_type): Set BSF_OBJECT flag for a type
3742 * ecoff.c (localsym_t): Add addend field.
3743 (add_ecoff_symbol): Add addend argument. Change all callers.
3744 (coff_sym_value): Make static.
3745 (coff_sym_addend): New static variable.
3746 (ecoff_directive_def): Initialize coff_sym_addend.
3747 (ecoff_directive_val): Accept symbol + constant.
3748 (ecoff_directive_endef): Pass coff_sym_addend to add_ecoff_symbol.
3749 (ecoff_build_symbols): Include the addend in the symbol value.
3753 * config/tc-ppc.c (md_assemble): Ignore overflow on
3754 BFD_RELOC_16_GOTOFF and BFD_RELOC_PPC_TOC16.
3758 * config/tc-ppc.c (md_apply_fix3): SDA21 relocations are now 4
3759 bytes in size, so offset appropriately in big endian mode when
3760 writing the bottom 2 bytes.
3764 * config/tc-sparc.c (default_compatible): New static local.
3765 (md_begin): Initialize it. Rewrite warn_on_bump handling.
3766 (sparc_ip): If no architecture or -bump specified, don't mark as
3767 mismatched those in default_compatible.
3772 * configure.in (i386-*-sco*elf*): Use fmt elf, targ sco5.
3773 * configure: Rebuild.
3774 * config/sco5.mt: New file; set TDEFINES to -DSCO_ELF.
3775 * config/tc-i386.c (sco_id): New function, if SCO_ELF.
3776 * config/tc-i386.h (tc_init_after_args): Define if SCO_ELF.
3777 (sco_id): Declare if SCO_ELF.
3781 * config/tc-sparc.c (initial_architecture,can_bump_v9_p): Deleted.
3782 ({max,warn_after}_architecture): New static locals.
3783 (md_begin): Replace NUMOPCODES with sparc_num_opcodes.
3784 If both architecture and -bump requested, set max_architecture to max.
3785 (sparc_md_end): Simplify.
3786 (sparc_ip): Replace references to can_bump_v9_p with max_architecture.
3787 Rewrite code to bump architecture and check for conflicts.
3788 (md_longopts): Recognize -xarch={v8plus,v8plusa} for compatibility
3789 with Solaris assembler.
3790 (md_parse_option): Likewise. Call sparc_opcode_lookup_arch.
3791 (md_show_usage): Update.
3795 * Makefile.in (RUNTEST): Fix reference to $${srcdir}.
3799 * config/tc-sparc.h (TARGET_FORMAT): Use #ifdef SPARC_ARCH64 instead of
3800 #ifdef sparcv9 when choosing value.
3803 (sparc_md_end): Declare.
3804 * config/tc-sparc.c (SPARC_V9): Renamed from sparcv9.
3805 (initial_architecture): New static local.
3806 (can_bump_v9_p): Likewise.
3807 (NO_V9): Delete all occurrences.
3808 (sparc_md_end): New function.
3809 (sparc_ip): New local v9_arg_p. Rework fp reg number test.
3810 Don't bump architecture to v9 unless can_bump_v9_p set.
3811 (md_parse_option): -A<arch> passed, set can_bump_v9_p accordingly.
3812 * configure.in (sparc64 target cpu): Don't set obj_format here.
3813 (SPARC_V9): Renamed from sparcv9.
3814 (sparc64-*-elf*): Define SPARC_ARCH64.
3815 * configure: Regenerated.
3816 * acconfig.h (SPARC_V9): Renamed from sparcv9.
3817 (SPARC_ARCH64): Add.
3818 * conf.in: Regenerated.
3819 * config/vmsconf.h: Update.
3823 * config/tc-mips.c (load_register): Optimise "dli" loads.
3824 (md_show_usage): add "-mcpu=vr4100" to help text.
3828 * symbols.c (resolve_symbol_value): If a symbol is equated to an
3829 undefined symbol, preserve the X_op of O_symbol.
3830 (S_GET_VALUE): Fix check to permit this case.
3831 * write.c (write_relocs): If a reloc is against an undefined
3832 symbol equated to another symbol, change the reloc to be against
3834 * config/obj-coff.c (do_relocs_for): Likewise.
3836 * config/tc-ppc.c (ppc_csect): An unnamed csect is storage class
3841 * config/obj-elf.c (elf/ppc.h): Include elf/ppc.h if target
3842 computer is PowerPC.
3844 * config/tc-ppc.c (md_apply_fix3): Add more embedded relocations.
3846 * config/tc-ppc.h (ELF_TC_SPECIAL_SECTIONS): Add sections
3847 mentioned in the eabi.
3851 * config/tc-ppc.c (ppc_reldata): Changed alignement on reldata_section
3852 * config/tc-ppc.c (ppc_pdata): Changed the alignment on pdata_section
3856 * config/tc-ppc.c (mapping): Add more relocation suffixes.
3860 * config/tc-ppc.c (ppc_elf_validate_fix): Allow .gcc_except_table
3861 as a section it is ok to have unadorned -mrelocatable pointers in.
3865 * config/tc-ppc.c (ppc_section*): Wrap these functions inside
3870 * config/obj-elf.c (obj_elf_section): Add hooks so machine
3871 dependent section attributes can be handled.
3873 * config/tc-ppc.h: (md_elf_section_{letter,type,word,flags}): New
3874 macros to add support for exclude section flag and ordered section
3877 * config/tc-ppc.c (ppc_elf_section_{letter,type,word,flags}): New
3878 functions to add support for exclude section flag and ordered
3883 * subsegs.c (section_symbol): Don't try to look up the section
3884 symbol in the hash table. It should be possible to have a symbol
3885 with the same name as a section, but no connection to it.
3887 * read.c (cons_worker): Only call mri_comment_end from flag_mri.
3890 * expr.c (operand): Skip whitespace after a close parenthesis.
3895 * config/tc-sh.c (md_apply_fix): Call as_bad_where instead of
3896 as_warn_where for relocation overflow.
3897 (parse_reg): Accept register name only if next character is
3902 * ecoff.c (ecoff_stab): Simplify. Correctly handle sym + offset
3903 addresses for static variables.
3907 * config/tc-ppc.c (mapping): Make @got give a real GOT relocation,
3908 and xgot give the old toc16 relocation.
3909 (md_apply_fix3): Support all GOT relocations.
3913 * config/tc-mips.c (load_address): Correctly handle a constant in
3918 * config/tc-sh.c (parse_reg): Recognize SH3 registers.
3919 (get_specific): Handle A_SSR, A_SPC and A_REG_B.
3920 (build_Mbytes): Handle REG_B.
3924 * ecoff.c (ecoff_build_aux): Use new bfd_big_endian macro.
3928 * mpw-make.sed: If linking, edit ALL_CFLAGS to CFLAGS.
3932 * config/obj-coff.c (write_object_file): Set the s_align field to
3933 the number of bytes, rather than to the power of 2.
3937 * Makefile.in (DISTCLEAN_HERE): New variable.
3938 (distclean): Use it.
3939 (maintainer-clean): Depend upon clean-here rather than clean,
3940 distclean, and clean-info. Run make maintainer-clean in doc.
3941 Remove files listed in DISTCLEAN_HERE.
3942 * doc/Makefile.in (maintainer-clean realclean): Split out from
3943 distclean. Depend upon clean-info and distclean.
3947 * mac-as.r: Fix copyright and version strings.
3948 (cfrg): Use PROG_NAME instead of literal name.
3952 * read.c (read_a_source_file): If tc_unrecognized_line is defined,
3954 * config/tc-a29k.h (tc_unrecognized_line): Define.
3955 * config/tc-a29k.c (a29k_unrecognized_line): New function.
3956 (md_operand): Handle a29k style local dollar labels.
3960 * config/obj-multi.h: If OBJ_MAYBE_ELF, define OBJ_SYMFIELD_TYPE.
3964 * read.c (s_fill): If md_flush_pending_output is defined, call
3969 * config/obj-coff.c (size_section, fill_section, fixup_mdeps):
3970 Treat rs_align_code like rs_align.
3974 * config/tc-arm.c (cp_address_required_here): Set pre_inc when
3975 converting an absolute address into a PC-relative one.
3979 * mpw-config.in: Don't always use te-generic.h for emulation.
3980 (powerpc-apple-macos): Use emulation te-macos.h.
3982 (mips-gm-magic): New configuration.
3984 * mpw-make.sed (install, install-only): Edit in Mac-specific
3989 * configure.in: Improve message about unsupported ELF targets.
3990 * configure: Rebuild.
3992 * config/tc-m88k.c (m88k_do_align): Correct check for whether fill
3993 pattern is zero. From Manfred Hollstein.
3997 * config/tc-ppc.c (ppc_pe_section): To get the alignment right for
3998 the various idata sections, we check the name on the .section pseudo.
4002 * config/obj-coff.c (fixup_segment): If TC_M88K is defined, do not
4003 add section's paddr to add_number; compatibility to native as and
4008 * configure.in: Treat m68k-sysv4 like m68k-elf, not m68k-sysv3.
4010 * hash.c (struct hash_entry): Moved here...
4011 * hash.h (struct hash_entry): ...from here.
4013 * config/obj-elf.c (elf_frob_symbol): Don't free and clear sy_obj
4014 if it's already known to be null.
4018 * config/tc-ppc.c (md_apply_fix3): Don't adjust the value for 32
4019 bit relocs converted to PC relative relocs. This turned out to
4020 add the offset from the beginning of .text twice.
4024 * stabs.c (s_stab_generic): In 's' case, free string from
4027 * config/obj-elf.h (ELF_TARGET_SYMBOL_FIELDS): Remove unused field
4029 * config/obj-multi.h (ELF_TARGET_SYMBOL_FIELDS) [OBJ_MAYBE_ELF]:
4032 * subsegs.h (segment_info_type): Make bitfields unsigned.
4034 * expr.h (struct expressionS): Make X_op and X_unsigned bitfields,
4035 and move them together. On most systems this will reduce the
4036 structure size by one word.
4037 (operatorT): Define O_max.
4038 * expr.c (expr_begin): Verify that X_op is wide enough to hold
4041 * read.c (pop_insert): Print error returned by hash table
4044 * as.c (dump_statistics): Split out from main; dump some hash
4045 table stats and target-specific stats.
4046 (start_time): No longer automatic to main.
4047 (main): Set file-level start_time and call dump_statistics at
4048 exit. Exit by calling xexit.
4049 (show_usage): Make --statistics description less specific.
4050 * subsegs.c (subsegs_print_statistics): New function.
4051 * write.c (write_print_statistics): New function.
4052 (n_fixups): New static variable.
4053 (fix_new_internal): Increment it.
4054 * read.c (read_print_statistics): New function.
4055 * read.h (read_print_statistics): Declare.
4056 * symbol.c (symbol_print_statistics): New function.
4057 * symbol.h (symbol_print_statistics): Declare.
4058 * hash.c (hash_print_statistics): New function.
4059 * hash.h (hash_print_statistics): Declare.
4060 * config/tc-i386.c (i386_print_statistics): New function.
4061 * config/tc-i386.h (i386_print_statistics): Declare.
4062 (tc_print_statistics): New macro.
4063 * messages.c (as_fatal, as_assert, as_abort): Use xexit, not
4066 * hash.c (DELETED): Rewrite to use a valid but unique address.
4067 (START_POWER): Reduce to 10.
4068 (enum stat_enum): New enumerator, replacing STAT_* index macros.
4069 Add new values for counting strcmp calls.
4070 (GROW_FACTOR): New macro.
4071 (hash_grow): Use GROW_FACTOR. Rewrite for quick returns instead
4073 (FULL_VALUE): New macro. Use 1/4 of table size instead of 1/2.
4074 (hash_new): Use FULL_VALUE.
4075 (struct hash_control): Definition moved here.
4076 (hash_code): Don't mask to low bits.
4077 (hash_ask): Mask returned hash code here. Check hash value before
4078 calling strcmp; count strcmp calls.
4079 * hash.h (struct hash_control): Declare, don't define, here.
4080 (HASH_STATLENGTH): Deleted.
4081 (struct hash_entry): Add field for hash code.
4082 (hash_say, hash_apply): Don't declare.
4084 * hash.c (destroy): Return void.
4085 (applicatee): Ditto.
4086 (main): Fix declarations.
4087 (hash_apply): Return void. Argument `function' returns void. Put
4088 inside "#ifdef TEST".
4089 (hash_say): Define only if TEST is defined.
4090 * hash.h (hash_apply, hash_say): Declarations deleted.
4094 * configure: Rebuild with autoconf 2.7.
4098 * aclocal.m4 (AC_PROG_CC): Remove local definition.
4099 * configure: Rebuild with autoconf 2.6.
4103 * config/tc-ppc.c (ppc_debug_name_section_size): Remove.
4104 (ppc_stabx): Don't increment ppc_debug_name_section_size.
4106 (ppc_frob_file): Remove.
4107 * config/tc-ppc.h (tc_frob_file): Don't define.
4108 (ppc_frob_file): Don't declare.
4112 * Makefile.in (TARG_CPU_DEP_alpha): Mention alpha-opcode.h.
4113 * config/alpha-opcode.h: Include one-operand variants of jmp and
4116 * config/te-delt88.h: Renamed from te-delta88.h, to avoid conflict
4117 with te-delta.h in 8.3 file systems.
4118 * configure.in: Adjusted.
4122 * config/tc-mips.c (s_err): Remove; just use the one in read.c.
4124 * config/m68k-parse.y (yylex): In MRI mode, '@' can start an octal
4126 * expr.c (operand): Handle MRI suffixes after unadorned 0.
4130 Version 2.6 released.
4131 * Makefile.in (VERSION): Updated to 2.6.
4133 * config/obj-coff.c (write_object_file): Change use of md_do_align
4134 to pass a pointer rather than a fill value, to match other uses.
4138 * config/tc-ns32k.h (TC_FIX_TYPE): Add missing semicolon.
4140 * as.c (main): Move md_end call to just after call to
4141 perform_an_assembly_pass. Delete cpu-specific code here.
4142 * config/tc-i960.h (md_end): New macro, calls brtab_emit.
4143 * config/tc-arm.c (md_end): Unused function deleted.
4144 * config/tc-ns32k.c (md_end): Ditto.
4146 * config/tc-i386.c (i386_align_code): New function, moved here
4147 from HANDLE_ALIGN macro.
4148 * config/tc-i386.h (HANDLE_ALIGN): Call it.
4152 * config/tc-i386.h (md_do_align): cast fill and 0x90 to char
4157 * config/tc-i386.h (md_do_align): Make ".align n,0x90" generate
4158 multi-byte nops to avoid changing gcc. The necessary gcc change
4159 might break old assemblers.
4163 * config/tc-i386.h (md_do_align, HANDLE_ALIGN): Add macros to
4164 generate optimal multi-byte nop instructions for ".align n"
4165 ".align n,0x90", and aligns requiring more than 15 bytes of
4166 padding still generate multiple 0x90's as before.
4170 * config/tc-m68k.c (s_mri_until): Call pop_mri_control.
4174 * configure.in (ppc-*-macos*, ppc-*-mpw*): New configurations.
4175 * configure: Update.
4176 * mpw-make.sed: Reorder commands to make sed happier.
4177 * config/te-macos.h: New file.
4178 * config/tc-ppc.h (TARGET_FORMAT): Set correctly for PowerMac.
4182 * config/tc-hppa.c (pa_ip): Fix off-by-2 bug in length check for
4183 conditional branches.
4184 (md_apply_fix): Likewise.
4188 * config/tc-a29k.c (md_apply_fix): Warn if an attempt is made to
4189 generate a reloc which the linker will not handle correctly. Fix
4190 overflow checking--R_IREL is 18 bits, not 17.
4194 * config/obj-coff.c (fixup_segment): Don't subtract md_pcrel_from
4195 from a PC relative reloc if TC_A29K.
4197 * config/tc-a29k.c (line_separator_chars): Restore '@'. Existing
4198 code depends upon it.
4200 * config/tc-a29k.c (md_operand): Handle $float, $double, and
4201 $extend. Based on code from Eric Freudenthal
4203 * config/tc-a29k.h (LEX_DOLLAR): Define.
4204 * read.c (LEX_DOLLAR): Define if not defined.
4205 (lex_type): Use LEX_DOLLAR.
4209 * configure.in (a29k-nyu-sym1): New target, just like other a29k
4214 * config/obj-coff.c (c_dot_file_symbol): Cast xmalloc return.
4218 * config/tc-ppc.c (md_apply_fix3): Added BFD_RELOC_RVA. Currently
4219 used only by "dlltool.c".
4223 * config/tc-alpha.c: Undefine inline if not __GNUC__.
4224 (md_pseudo_table): Don't define "extern".
4228 * config/tc-ppc.c (ppc_biei): Force symbol into text_section.
4230 * config/tc-ppc.c (md_show_usage): Put backslash at end of line.
4234 * macro.c (macro_expand_body): Don't warn about == with a
4235 nonexistent parameter, in case it is in a comment field.
4237 * as.c (main): On TC_A29K, call macro_init with macro_alternate
4239 * macro.c (get_any_string): Don't keep quotes if macro_strip_at is
4240 set, even if macro_alternate is set.
4241 (get_apost_token): If macro_strip_at, only skip kind if it is '@'.
4242 (sub_actual): If macro_strip_at, and kind is '@', don't look up
4243 the token unless it ended in '@'.
4244 * config/tc-a29k.c (line_separator_chars): Remove '@'.
4245 * doc/c-a29k.texi: Document macro usage on A29K.
4249 * config/tc-m68k.c (m68k_ip): Handle new 'W' place, meaning a
4251 (install_operand): Likewise.
4253 * config/obj-elf.c (ecoff_debug_pseudo_table): Add "extern".
4257 * configure.in (m88k-motorola-sysv*): New target.
4258 * configure: Rebuild.
4259 * config/te-delta88.h: New file.
4260 * config/obj-coff.c (write_object_file): Use md_do_align if it is
4262 * config/tc-m88k.h (SUB_SEGMENT_ALIGN): Define.
4263 (md_do_align): Define.
4264 * config/tc-m88k.c: Include "subsegs.h".
4265 (m88k_do_align): New function.
4267 * config/te-delta.h (STRIP_UNDERSCORE): Don't define.
4268 (COFF_NOLOAD_PROBLEM): Define.
4269 (LOCAL_LABELS_DOLLAR, LOCAL_LABELS_FB): Define.
4273 * config/tc-i386.c (md_assemble): For a jump instruction with
4274 non-constant target, require 7 available bytes in the current
4279 * config/obj-elf.h: Include bfd/elf-bfd.h rather than
4284 * configure.in (alpha-*-linux*): Use ecoff.
4285 * configure: Rebuild.
4286 * ecoff.c (ecoff_directive_extern): New function.
4287 (ecoff_directive_weakext): New function.
4288 (ecoff_build_symbols): Handle weak symbols.
4289 (ecoff_setup_ext): Likewise.
4290 (ecoff_frob_symbol): Warn about weak common symbols.
4291 * ecoff.h (ecoff_directive_extern): Declare.
4292 (ecoff_directive_weakext): Declare.
4293 * symbols.c (S_IS_WEAK): New function.
4294 * symbols.h (S_IS_WEAK): Declare.
4295 * config/obj-ecoff.c (obj_pseudo_table): Add "extern" and
4297 * config/tc-mips.c (mips_pseudo_table): Remove "extern".
4302 * config/tc-ppc.c (ppc_lglobl): Do the right thing.
4304 * config/tc-ppc.c (ppc_bb): Call SF_SET_PROCESS.
4305 (ppc_eb): Likewise. Set the storage class to C_BLOCK, not C_FCN.
4306 (ppc_frob_symbol): Don't change C_BLOCK symbols to C_HIDEXT.
4307 * config/obj-coff.c (coff_frob_symbol): Don't call
4308 SA_SET_SYM_ENDNDX with the current symbol; call it with the next
4309 one. If OBJ_XCOFF, try to figure out whether the symbol is going
4312 * config/tc-ppc.c (md_pseudo_table): Add "bc" and "ec".
4313 (ppc_stab_symbol): New static variable.
4314 (ppc_change_csect): Check that ppc_toc_csect is not NULL.
4315 (ppc_stabx): Set ppc_stab_symbol around call to symbol_make. Set
4316 sy_tc.real_name to the stab string.
4317 (ppc_bc, ppc_ec): New static functions.
4318 (ppc_canonicalize_symbol_name): If ppc_stab_symbol is set, don't
4320 (ppc_symbol_new_hook): If ppc_stab_symbol is set, don't look for a
4322 (ppc_frob_symbol): Set BSF_NOT_AT_END for symbols with csect aux
4325 * input-scrub.c (input_scrub_push): Reset sb_index.
4329 * listing.c (listing_newline): Don't create a frag in the absolute
4334 * config/tc-ppc.c (md_pseudo_table): Add "data" and "text".
4335 (ppc_csect): Move most of the code to ppc_change_csect, and call
4337 (ppc_change_csect): New static function, taken from ppc_csect.
4338 (ppc_section): New static function.
4339 (ppc_saw_abs): New static varable.
4340 (ppc_frob_symbol): Create aux entry for absolute symbols. Warn if
4341 a symbol has no csect.
4342 (ppc_adjust_symtab): New function.
4343 * config/tc-ppc.h (tc_adjust_symtab): Define if OBJ_XCOFF.
4344 (ppc_adjust_symtab): Declare if OBJ_XCOFF.
4346 * write.c (write_object_file): If tc_adjust_symtab is defined,
4347 call it just before the call to obj_adjust_symtab.
4349 * symbols.c (symbol_find_or_make): Change name to be const.
4350 * symbols.h (symbol_find_or_make): Update declaration.
4354 * doc/as.texinfo (Align): Mention SH.
4355 * doc/c-m68k.texi (M68K-Directives, .even): Describe behavior, not
4357 * doc/c-z8k.texi (Z8000 Directives, global): Fix minor typo.
4358 (Z8000 Directives, even): Don't give numeric align value, instead
4363 * tc-arm.c (do_ldst): Assemble ldr/str r0, [r1] as a pre-increment
4368 * Makefile.in (diststuff): Also make info.
4369 (maintainer-clean realclean): Also make clean-info.
4373 * config/tc-hppa.c (md_pseudo_table): Add new ".nsubspa" opcode.
4374 (pa_subspace): For ".nsubspa", always create a new subspace
4375 with the given attributes, even if one already exists with the
4380 * config/tc-ppc.h (TC_FORCE_RELOCATION_SECTION): Rename from
4381 TC_FORCE_RELOCATION, taking an additional section argument. If
4382 the section of the target symbol is not the same as the current
4383 section, always force the relocation to be used.
4384 (MD_PCREL_FROM_SECTION): New macro to call md_pcrel_from_section.
4386 * config/tc-ppc.c (md_pcrel_from_section): Rename from the
4387 md_pcrel_from function, taking an additional section argument.
4388 Invoke TC_FORCE_RELOCATION_SECTION instead of TC_FORCE_RELOCATION.
4390 * write.c (TC_FORCE_RELOCATION_SECTION): Define in terms of the
4391 older TC_FORCE_RELOCATION if not defined.
4392 (MD_PCREL_FROM_SECTION): If not defined, invoke md_pcrel_from.
4393 (fixup_segment): Use MD_PCREL_FROM_SECTION instead of
4394 md_pcrel_from, and TC_FORCE_RELOCATION_SECTION instead of
4395 TC_FORCE_RELOCATION.
4399 * input-scrub.c (as_where): Set name to null pointer if we don't
4401 * messages.c (identify): Only print filename if non-null.
4402 (as_show_where): Ditto, for line number too.
4403 (as_warn_internal, as_bad_internal): Ditto.
4405 * input-file.c (input_file_open): If the input file can't be
4406 opened, consider it an error.
4410 * config/tc-mips.c: Added mips_4100 control, and support for
4411 accepting the 4100 as a MIPS architecture variant (md_begin,
4412 macro_build, mips_ip, md_parse_option). Adding suitable
4413 command-line OPTIONs, and updating the help text (md_show_usage).
4417 * subsegs.c (subseg_begin): Only set absolute_frchain.fix_* when
4418 BFD_ASSEMBLER is defined.
4420 * Use one active frag and one obstack per frag chain:
4421 * frags.c (frags): Variable deleted.
4422 (frag_alloc): New function.
4423 (frag_grow, frag_more, frag_variant, frag_now_fix,
4424 frag_append_1_char): Refer to frchain_now->frch_obstack instead of
4426 (frag_new): Ditto. Verify that frch_last and frag_now match on
4427 entry and exit, and that old frag_now has non-zero type. Replace
4428 "know" uses with "assert". Use frag_alloc instead of mucking with
4430 * frags.h (frags): Declaration deleted.
4431 * subsegs.h (struct frchain): Add new field frch_frag_now.
4432 * subsegs.c (frchains, dummy_frag, absolute_frchain): New static
4434 (subsegs_begin): Initialize frchains obstack. Under gcc, don't
4435 give it any stricter alignment than frchainS structures need. Do
4436 not initialize frags obstack. Set frag_now to point to
4437 dummy_obstack. Initialize absolute_frchain.
4438 (subseg_set_rest): Save and restore frag_now in frch_frag_now
4439 field of frchainS. Don't create new frags on section switch, and
4440 use frag_alloc when creating a new frag chain. For absolute
4441 section, set frchain_now to absolute_frchain. Verify that
4442 frch_last and frag_now match on entry and exit. Initialize
4443 per-chain obstack, and under gcc, set required alignment to that
4444 needed by fragS structure.
4446 * write.c (chain_frchains_together_1): Verify fr_type is nonzero.
4448 * stabs.c (get_stab_string_offset): Only copy input string if a
4449 fresh copy is needed, not if the section already exists.
4450 (s_stab_generic): Cache stab section name to bypass lookups, since
4451 usually it will match. Could be made faster still by changing the
4452 memory allocation rules.
4453 (s_xstab): Cache section name to bypass repeated string
4456 * frags.c (frag_new): Deleted register declarations.
4458 * listing.c (frag_now): Don't declare.
4460 * as.c (chunksize): New variable.
4461 (debug_memory): New variable.
4462 (main): If debug_memory is set, reduce chunksize and
4464 * as.h (chunksize): Declare it.
4465 * read.c (read_begin): Use it.
4467 * config/tc-alpha.c (md_shortopts): Include 'g'.
4468 (md_parse_option): Ignore it.
4470 * Makefile.in (distclean): Remove Makefile and config.status from
4471 testsuite directory.
4472 (clean-here): Don't delete testsuite. Instead, delete only the
4473 files within it that would be generated by running tests.
4475 * config/tc-hppa.c (hppa_elf_mark_end_of_function): Call
4476 frag_now_fix instead of accessing obstack info directly.
4477 * config/tc-arm.c (s_ltorg): Ditto.
4478 (md_assemble): Ditto.
4480 * config/tc-i386.c (md_assemble): Call frag_grow instead of
4485 * stabs.c (aout_process_stab): Insert debug symbol into symbol
4486 chain after parsing value expression, if any, to avoid separating
4487 continued .stabs lines.
4491 * config/tc-mips.c (mips_elf_pseudo_table): Remove.
4492 (mips_pop_insert): Don't call pop_insert on mips_elf_pseudo_table.
4496 * config/tc-ppc.c (md_begin): Use new flags PPC_OPCODE_COMMON for
4497 -mcom support and PPC_OPCODE_ANY for -many.
4498 (md_parse_option): Ditto.
4500 (md_begin): For duplicate instructions, print all duplicates
4505 * config/tc-ppc.c (md_parse_option): Support for -mcom to turn on
4506 common mode operation.
4507 (md_show_usage): Add -mcom to usage message.
4511 * expr.c (op_rank): Add O_symbol_rva.
4512 * expr.h (operatorT): Add O_symbol_rva.
4513 * read.c (cons_worker): Set O_symbol_rva when necessary.
4514 * write.c (fix_new_exp): Understand O_symbol_rva.
4518 * config/tc-mips.c: Correct s_cons arguments. From Michael
4523 * config/tc-ppc.c (ppc_macro): Make count unsigned.
4524 (ppc_biei): Set segment to now_seg and value to coff_n_line_nos.
4525 (ppc_frob_symbol): Handle C_BINCL and C_EINCL symbols by setting
4527 * config/obj-coff.c (coff_n_line_nos): Rename from n_line_nos, and
4528 make non-static. Change all users.
4529 * config/obj-coff.h (coff_n_line_nos): Declare.
4535 * configure.in (AC_ARG_WITH(bfd-assembler)): Fix help message.
4537 * config/obj-elf.c (obj_elf_common): Convert specified byte
4538 alignment to power of two. Set size of local bss symbol.
4540 * config/tc-m68k.c (tc_gen_reloc): Fix typo in variable name.
4544 * sb.c, macro.c: Decide whether to include <string.h> or
4545 <strings.h> just as as.h does.
4549 * Makefile.in (site.exp): Fix setting of $srcdir.
4551 * config/tc-arm.c (md_atof): Fix little-endian output.
4552 * config/tc-arm.h (ARM_BI_ENDIAN): Move definition so defined for
4557 * doc/as.texinfo: Split out the NS32k family documentation,
4558 despite its being commented out for now.
4559 * doc/c-ns32k.texi: New file.
4561 * sb.c, macro.c: Include string.h.
4563 * Makefile.in (comparison): Only check *.o; we don't care if
4564 timestamps inserted by the native linker differ.
4566 * config/tc-alpha.c (alpha_align): Only fill with a no-op pattern
4567 if alignment stricter than 4 bytes is requested; in that case,
4568 align to a 4-byte boundary first.
4572 * config/obj-vms.c (VMS_RSYM_Parse): eliminate "might be used
4573 uninitialized" warning for `Max_Source_Offset'.
4577 * config/tc-ppc.c (parse_toc_entry): New function to parse [toc]
4578 qualifiers and detect errors if present.
4579 (md_assemble): Add call to parse_toc_entry. Also added some support
4580 for the [tocv] qualifier.
4581 (ppc_pe_tocd): New function to support data in the toc section.
4585 * config/tc-ppc.c (ppc_frob_symbol): Don't create an aux entry for
4590 * config/tc-m68k.c (isword): Accept all values from -65536 to
4591 +65535, so ~VAL will not be rejected.
4593 * cond.c (s_endif): Call demand_empty_rest_of_line. In MRI mode,
4594 skip characters after the pseudo-op.
4596 * read.c (get_line_sb): Don't look past buffer_limit.
4597 (s_include): In MRI mode, skip characters after the file name.
4601 * config/m68k-parse.y (m68k_reg_parse): In MRI mode, permit
4602 periods in register names.
4604 For older changes see ChangeLog.1.