3 * config/tc-mn10200.c (md_relax_table): Correct branch ranges.
7 * configure.in (i386*-go32-rtems*): Fix to be the same as
13 * config/tc-mips.c (macro): The 4650 doesn't permit M_LDC1_AB,
14 M_SDC1_AB, M_L_DOB, M_L_DAB, M_S_DAB, or M_S_DOB.
15 (mips_ip): Always check for FP_D, not just for instructions that
16 are not part of the regular ISA.
20 * config/tc-d10v.c (build_insn): Make `number' a long for 64-bit hosts.
24 * config/tc-alpha.c (cpu_types): 21164pc/pca56 does not have CIX.
28 * expr.c (integer_constant 32bit bignum): Mask off bits outside
29 the range we care about.
33 * config/tc-d30v.c (md_shortopts): Add 'n' and 'N' options.
34 (exec_type_enum): Enumeration giving all of the exec types.
35 (warn_nops): New static variable to give nop warning level.
36 ({cur,prev}_mul32_p): New static variable to keep track of whether
37 the current/previous instruction is a 32-bit multiply.
38 (Optimizing): Make static.
39 (NOP{2,_LEFT,_RIGHT}): Macros for word of nops and left/right
41 (d30v_insert_operand): Delete declaration of unused function.
42 (write_2_short): Make exec_type argument enum, not int.
44 (check_range): Delete unused variable(s).
47 (md_apply_fix3): Ditto.
48 (md_show_usage): Document -n and -N.
49 (md_parse_option): Parse -n and -N.
50 (write_1_short): If -n, warn about adding a nop. Use
52 (write_2_short): Use enumeration values instead of hard coded
53 integers. Reset exec_type for default operations. For explicit
54 parallel operations, call parallel_ok to make sure everything is
55 ok. If writing out a parallel operation, and the previous
56 instruction was a 32-bit multiply, indicate current instruction
58 (parallel_ok): Allow add/tx ... to be done in parallel with
59 another add/tx ... assuming the gpr registers don't overlap.
60 (md_assemble): Use exec type enumeration values, not hard coded
61 ints. Check for loads or 16-bit multiplies following in the next
62 cycle after a 32-bit multiply. Add nops if that is the case.
63 (do_assemble): Copy prev_mul32_p to cur_mul32_p, and set
64 cur_mul32_p if current instruction is a 32-bit multiply.
65 (find_format): Change spacing and layout.
69 * config/tc-tic80.c (tic80_relax): New static variable.
70 (md_longopts): Add new OPTION_RELAX and OPTION_NO_RELAX options.
71 (md_parse_option): Handle new relax options.
72 (md_show_usage): Document new relax options.
73 (find_opcode): Don't use short forms of PC relative branches if
78 * config/tc-d30v.c (parallel_ok): Remove non-register bits from
79 used/set flag fields. Make flag vars unsigned long. Use
80 FLAG_A{0,1} for accumulators. Allow any 2 insns to be done in
81 parallel if they use the same conditional flag with reversed
82 meaning. Allow 2 add/sub insns that set the carry or overflow
83 flags but do not query them to be done in parallel. Don't allow 2
84 word store operations to be done in parallel with ADDppp or
85 SUBppp. Don't allow loads to be done in parallel with 16 bit
90 * config/tc-arm.c: Prevent use of interworking support for
95 * doc/all.texi: Add M32R cpu.
97 * doc/as.texinfo: Add documentation of m32r processor.
99 * doc/c-m32r.texi: New file, documenting m32r specific features.
103 * config/tc-mips.c (mips_ip): Correctly insert 'P' operands into
108 * config/tc-tic80.c (build_insn): Handle instructions that have
109 long (32 bit) PC relative offsets. Fix places that previously
110 misused R_MPPCR for 15 bit offsets to use the new R_MPPCR15W type.
111 (md_apply_fix): Add case to handle long PC relative offsets.
115 * doc/c-arm.texi (ARM Options): Document support for new ARM
118 * config/tc-arm.c (md_parse_option): Add support for new ARM
123 * config/tc-m68k.c (m68k_ip): Don't overwrite opcode table data.
124 (insop, m68k_ip): Make `opcode' const so it doesn't happen again.
128 * config/tc-v850.c (md_assemble): Fix BFD_RELOC_32 against a
131 * config/tc-v850.h (ELF_TC_SPECIAL_SECTIONS): Use
132 SHT_V850_{S|T|Z}COMMON to mark special common sections.
136 * config/tc-v850.c: Brought up to date with the branch.
140 * config/tc-sh.c (SWITCH_TABLE_CONS): Handle (fix)->fx_size == 1.
141 (SWITCH_TABLE): Handle BFD_RELOC_8.
142 (md_apply_fix): #ifndef BFD_ASSEMBLER code: Handle fixP->fx_size == 1.
143 (coff_reloc_map): Add BFD_RELOC_8_PCREL entry.
144 (sh_coff_reloc_mangle): SWITCH_TABLE case: Handle BFD_RELOC_8.
148 * config/tc-alpha.c (range_signed_16, range_signed_32): Work around an
149 apparent bug in gcc's long long support crossing from x86.
153 * config/tc-arm.c: Brought up to date with latest changes on arm
158 * config-gas.com: Get version info from configure.in.
160 * makefile.vms: include depend.obj in OBJS.
162 * config/tc-alpha.c (s_alpha_section): Remove ".lcomm" handling.
164 * config/tc-alpha.c (alpha_basereg_clobbered): Remove variable and
165 all corresponding code.
169 * config/tc-arm.h (TARGET_FORMAT for generic a.out targets): Allow
170 run-time endian selection.
174 * config/tc-sh.c (parse_reg): Properly quote for fv4.
178 * symbols.c (resolve_symbol_value): Add missing breaks in case on
179 symbol value operator.
183 * config/tc-d10v.c (parallel_ok, find_opcode):
184 Split OPERAND_FLAG into OPERAND_FFLAG and OPERAND_CFLAG.
188 * config/obj-coff.c (fixup_segment): Cast second arg of
189 md_apply_fix3 call to type "valueT *".
193 * configure.in (emulations): Make FreeBSD an aout / i386bsd
195 * configure: Re-generate.
199 * config/tc-mips.c (macro_build): Use the membership field
205 * config/tc-d10v.c (find_opcode): For OPCODE_FAKE, add check for
206 first argument if it's supposed to be a register.
210 * app.c (do_scrub_chars): If d10v, re-insert a space before
211 a '#' when in state 10.
215 * config/tc-h8300.c: Include "subsegs.h".
216 (tc_reloc_mangle): Handle references to symbols which are not
217 being output, so that references to `.' work.
221 * config/tc-m68k.c (m68k_ip): Call add_fix when needed for '_'
224 * macro.c (sub_actual): If we don't find a parameter for an &,
229 * config/tc-mips.c (mips_ip): In default case, call as_bad
230 instead of fprintf, to get "assembler messages:" message output
231 before instead of after.
235 * frags.h: Handle multiple inclusion.
240 * as.h (struct frag, frag support): Moved from here.
242 (struct frag, member tc_frag_data): New member if TC_FRAG_TYPE
244 (struct frag, member fr_cgen): Renamed from fr_targ.cgen.
245 * cgen.c (cgen_asm_finish_insn): Update.
246 * config/tc-m32r.c (md_estimate_size_before_relax): Update.
247 * config/tc-m32r.h (TC_FRAG_INIT): Renamed from md_init_frag.
248 (md_convert_frag): Ditto.
249 * config/tc-ns32k.h (TC_FRAG_TYPE): Define.
250 (frag_opcode_frag,frag_opcode_offset,frag_bsr): Update.
251 (TC_FRAG_INIT): Update.
255 * write.c (print_fixup): Use TC_FIX_DATA_PRINT (if defined) to
256 print out MD fields of fix.
257 * frags.c (frag_var, frag_variant): Use TC_FRAG_INIT macro (if
258 defined) to initialize MD fields in frag.
259 * as.h (struct frag, ns32k support): Rename ns32k to fr_ns32k.
260 Delete pcrel_adjust. Add fr_opcode_fragP, fr_opcode_offset.
261 * config/tc-ns32k.h: Add comments. Remove obsolete
262 BFD_FAST_SECTION_FILL definition, change prototypes for
263 fix_new_ns32k and fix_new_ns32k_exp to add new arguments
264 opcode_frag and opcode_offset and remove pcrel_adjust.
265 (TC_FIX_TYPE): add opcode_fragP and opcode_offset fields.
266 (TC_FIX_DATA_PRINT): new macro to print out TC_FIX_TYPE.
267 (TC_FRAG_INIT): new macro to initialize machine dependent field in
269 (frag_opcode_frag, frag_opcode_offset, frag_bsr): macros to access
270 MD fields in frag structure.
271 (fix_im_disp, fix_bit_fixP, fix_opcode_frag, fix_opcode_offset,
272 fix_bsr): macros to access MD fields in fix structure.
273 * config/tc-ns32k.c: Avoid overlength lines. Align comments. Don't
274 use struct opcode_location as these fields are now in the frag
276 (convert_iif): Call frag_more as it is needed instead
277 of trying to allocate for the whole insn. Avoid call of frag_more
278 with negative argument.
279 (md_pcrel_adjust, md_fix_pcrel_adjust, md_apply_fix,
280 md_estimate_size_before_relax, md_pcrel_from,
281 tc_aout_fix_to_chars): use accessor macros to get md fields in fix
283 (fix_new_ns32k, fix_new_ns32k_exp): add new arguments opcode_frag and
284 opcode_offset and remove pcrel_adjust.
285 (convert_iif, cons_fix_new_ns32k): call fix_new_ns32k,
286 fix_new_ns32k_exp with changed arguments.
290 * config/tc-mips.c (md_begin): Reorganize setting of default values so
291 that mips_cpu depends on TARGET_CPU, and mips_opts.isa depends on
293 (md_parse_option): Remove all code that sets defaults; md_begin
294 handles all of this now.
298 * Makefile.am (STAGESTUFF): Change bin_PROGRAMS to
300 (bootstrap, bootstrap2, bootstrap3): Likewise.
301 * Makefile.in: Rebuild.
303 * config/tc-ppc.c (ppc_fix_adjustable): Don't adjust relocs in the
304 TOC section to be against the csect.
308 * config/tc-mips.c (validate_mips_insn): New function, checks
309 match versus mask bits, and also verifies that all bits to be
310 output are actually specified somewhere.
311 (md_begin): Call it for 32-bit instructions, instead of doing
312 match/mask check here. In case of failure, print a message, but
313 check the rest of the opcode table before exiting.
317 * config/tc-arm.c (md_apply_fix3): Fix thumb ADR pseudo op. Patch
322 * config/tc-d30v.c (build_insn): Allow odd registers for ld2w and
327 * config/tc-ppc.c (md_assemble): When handling @l, always sign
328 extend if the operand expects a signed value.
330 * config/tc-mips.h (LOCAL_LABELS_DOLLAR): Don't define; use
331 default which is to permit dollar labels.
335 * config/tc-sparc.c (sparc_memory_model): New variable.
336 (md_longopts): Add -TSO/-PSO/-RMO options.
337 (md_parse_options): Handle them.
338 (sparc_elf_final_processing): For 64 ELF, set required
339 memory ordering in e_flags. Default to RMO and let the user
340 override it through command line.
342 * config/tc-sparc.h (elf_tc_final_processing): Add.
346 * config/tc-sparc.c (v9a_asr_table): New variable.
347 (sparc_ip): Handle v9a asr's.
352 * config/tc-sparc.h (md_do_align): New macro.
353 * config/tc-sparc.c (sparc_handle_align): Handle rs_align_code.
358 * config/tc-sh.c (sh_small): New variable.
359 (OPTION_SMALL): Define.
360 (md_longopts): Add "small".
361 (md_parse_option): Handle OPTION_SMALL.
362 (md_show_usage): Mention -small.
363 * config/tc-sh.h (sh_small): Declare.
364 (SUB_SEGMENT_ALIGN): Handle sh_small.
365 * config/obj-coff.h (TARGET_FORMAT): Check sh_small in TC_SH
368 * config/tc-mips.c (macro): Correct handling of constant in M_LI_D
369 case in little endian mode.
373 * config/tc-sparc.c (md_apply_fix3, cases ..._H44, ..._HIX22): Leave
374 overflow signalling to linker.
378 * makefile.vms: Fix for dec c.
380 * config-gas.com: Give explanation for dec c setup in error
383 * config/tc-alpha.c (s_alpha_comm): Make .comm symbols separate
384 sections on openvms/alpha.
386 * config/obj-evax.c: support .weak pseudo-op
390 * config/tc-sparc.c (default_arch_size): New static local.
391 (struct sparc_arch): Rename arch_size to default_arch_size.
392 New member user_option_p.
393 (sparc_arch_table): Always include v9, v9a. New entry v9-64.
394 (init_default_arch): Check whether default arch is valid.
395 Set default_arch_size in addition to sparc_arch_size.
396 (OPTION_32,OPTION_64): Define.
397 (md_longopts): New entries for -32, -64.
398 (md_parse_option): Handle them.
399 (md_show_usage): Print them. Ensure init_default_arch called.
400 * configure.in (sparc64): Set arch to v9-64.
401 * configure: Regenerated.
405 * write.c (subsegs_finish): New function, broken out of
407 (write_object_file): Some code moves into subsegs_finish.
408 * write.c (subsegs_finish): Declare.
409 * as.c (main): Call subsegs_finish.
411 * read.c (s_include): Check for error return from
416 * read.c (get_line_sb): Accept any eol marker while scanning macros.
420 * config/tc-alpha.h (DIFF_EXPR_OK): Define.
421 * config/tc-i386.h (DIFF_EXPR_OK): Define.
422 * config/tc-alpha.c (md_apply_fix): Notice fx_pcrel and substitute
423 the correct relocation when it exists.
424 * config/tc-i386.c (md_apply_fix3): Likewise.
426 * config/tc-ppc.h: Correct typo in comment.
427 * config/tc-v850.h: Likewise.
431 * config/tc-d10v.c (parallel_ok): Allow parallel instruction issue
432 when second instruction is writing to first instructions inputs.
436 * ecoff.c (PAGE_SIZE): Double to 8k as a hack to allow some C++
437 templated programs to build with -g.
441 * config/tc-v850.c (md_relax_table): Add support for relaxing
442 unconditional branches. This patch is courtesy of Jim Wilson.
443 (md_convert_frag): Fix relaxing of branches. This patch is
444 courtesy of Jim Wilson.
445 (md_assemble): Create different fixups for conditional and
446 unconditional branches. This patch is courtesy of Jim Wilson.
447 (md_estimate_size_before_relax): Estimate size of variable part of
448 fixup based on whether it is for a conditional or an unconditional
449 branch. This patch is courtesy of Jim Wilson.
450 (v850_sdata, v850_tdata, v850_zdata, v850_sbss, v850_tbss,
451 v850_zbss, v850_rosdata, v850_rozdata, v850_bss): Add call to
452 obj_elf_section_change_hook().
453 (v850_comm): New function.
454 (md_pseudo_table): Add new pseudo ops .zcomm, .scomm and .tcomm.
455 (md_begin): Add bss flag to seg_info of bss sections.
457 Add support for .scommon, .tcommon and .zcommon sections.
459 * config/tc-v850.h (ELF_TC_SPECIAL_SECTIONS): Add .scommon,
460 .zcommon, .tbss, .call_table_data and .call_table_text.
464 * configure.in (sparc): Set DEFAULT_ARCH from correct target.
465 * configure: Regenerated.
469 * config/tc-d10v.c: Fixes to make sure the AT_WORD
470 expression is not confused with -1.
474 * config/tc-d10v.c (parallel_ok): Flag SP as modified for @-sp
475 operand - OPERAND_ATMINUS.
479 * config/tc-d10v.c (parallel_ok): Note that auto increment and
480 decrement modify the index register.
485 * config/tc-m68k.c (archs): Add 68306, 68307, 68322, 68356, 68334,
487 * doc/c-m68k.texi (M68K-Opts): Add -m68ec000 -m68hc000 -m68hc001
488 -m68306, -m68307, -m68322, -m68356, -m68ec020, -m68ec030,
489 -m68ec040, -m68ec060, -m68330, -m68334, -m68336, -m68341,
492 * doc/Makefile.am (CPU_DOCS): Define.
493 (as.info): Depend upon $(CPU_DOCS).
494 * doc/Makefile.in: Rebuild.
496 * configure.in: Remove AM_PROG_INSTALL; it's called by
498 * configure: Rebuild.
502 * config/tc-d10v.h (TC_START_LABEL): Don't define.
503 (tc_frob_label): Define.
507 * config/tc-d10v.c (write_2_short): Fix bug that wouldn't allow
508 to pair a branch and link with anything but an exe instruction.
512 * config/tc-alpha.c (load_expression): Disable the sym+const .got
513 optimization to reduce the alignment surprises for gcc.
517 * config/obj-coff.h (TC_SPARC): Don't define TARGET_FORMAT.
518 * config/tc-sparc.c (sparc_target_format): Handle coff here.
519 (sparc_ip): Add %hix,%lox.
520 (md_apply_fix3): Call as_bad_where, not as_bad.
521 Add support for BFD_RELOC_SPARC_{HIX22,LOX10}.
522 (tc_gen_reloc): Add support for BFD_RELOC_SPARC_{HIX22,LOX10}.
526 * configure.in: Change alpha-*-* to alpha*-*-*; config.guess now
527 recognizes alphaev5 etc.
528 * configure: Rebuild.
532 * config/tc-mips.c (md_begin): Replace the TARGET_CPU value
533 of mipsr3900 with mipstx39.
535 * config/tc-mips.c (mips_ip): Don't print the 'opcode requires
536 -mipsXX message' if the insn isn't an ISA insn.
540 * config/tc-sparc.h (TARGET_FORMAT support): Moved to tc-sparc.c.
541 Redefine TARGET_FORMAT to call sparc_target_format.
542 * config/tc-sparc.c (in_unsigned_range): New function.
543 (sparc_arch_size): Make static.
544 (sparc_target_format): New function.
545 (sparc_ip): Delete variable immediate_max. Rewrite %hi/etc reloc
546 handling. Add support for %hh,%hm,%lm,%h44,%m44,%l44.
547 (output_insn): Set `fx_no_overflow'.
548 (md_apply_fix3): Handle BFD_RELOC_SPARC_{7,H44,M44,L44}.
549 (tc_gen_reloc): Likewise.
553 * config/tc-v850.c (v850_section): Remove.
555 * config/obj-elf.c (obj_elf_section): Enhance error message.
559 * config/tc-mips.c: Undef OBJ_COPY_SYMBOL_ATTRIBUTES before
560 including obj-elf.h in OBJ_MAYBE_ELF case.
561 (mips_target_format): Return NULL after abort to avoid warning.
563 * ecoff.c (generate_ecoff_stab): Remove unused static function.
565 * expr.c (operator): Accept ==. From Anders Blomdell
568 * config/atof-ieee.c (gen_to_words): When generating a denormal
569 number, handle an overflow into the smallest normalized number.
573 * as.h, input-scrub.c (new_logical_line): New return value.
574 * read.c (s_app_file): Don't note the same file several times
579 * config/tc-m68k.c (m68k_ip): Remove ` operand specifier.
583 * configure.in (sh*-*-rtems*): New target, like sh-*-elf*.
584 * configure: Rebuild.
588 * config/tc-m68k.c (m68k_ip): Handle q and v operand specifiers.
590 * doc/c-i386.texi (i386-Float): Remove incorrect assertion that
591 fn* instructions do not insert implicit fwait. This was changed
594 * config/m68k-parse.y (yylex): Permit an expression to be used for
597 * Makefile.am (EXTRA_as_new_SOURCES): Set to config/m68k-parse.y,
599 * Makefile.in: Rebuild.
601 * aclocal.m4: Rebuild with new libtool.
602 * configure: Rebuild.
606 * app.c (do_scrub_chars): Clear mri_state at end of .mri
609 * config/tc-mips.c (hilo_interlocks): Change from a static
610 variable to a macro, so that it varies with the variables upon
612 (gpr_interlocks, cop_interlocks): Likewise.
613 (md_begin): Don't initialize them.
617 * config/tc-mn10300.c (md_assemble): Use strcasecomp instead
618 of strcmp where appropriate.
622 * config/tc-v850.c (md_assemble): Cope with a zero data area
623 relocation with a constant offset.
624 (md_assemble): Produce error message when special data area
625 relocations are used on instructions which do not support them.
626 (md_assemble): Reset processor mask if defined by command line
631 * config/tc-sparc.c: Reorganize file.
632 (parse_keyword_arg): Allow numbers in reg names.
633 (SPECIAL_CASE_NONE): New macro.
634 (md_assemble): Use it.
635 (lookup_arch,init_default_arch): New functions.
636 (default_arch,default_init_p,sparc_arch_table): New static locals.
637 (sparc_arch_size): New static local.
638 (max_architecture): Initialize in init_default_arch.
639 (md_parse_options): Call init_default_arch if necessary.
640 Rewrite -xarch/-A processing.
641 (md_show_usage): Print -A values from sparc_arch_table.
642 (md_begin): Call init_default_arch if necessary.
643 (sparc_md_end): Handle both 32 and 64 bit environments.
644 * config/tc-sparc.h (TARGET_FORMAT): Likewise.
645 * acconfig.h (SPARC_V9,SPARC_ARCH64): Delete.
647 * config.in: Regenerate.
648 * configure.in (sparc): Default DEFAULT_ARCH based on target cpu.
649 (SPARC_V9,SPARC_ARCH64): Delete.
650 * configure: Regenerate.
651 * config/vms-conf.h (SPARC_V9,SPARC_ARCH64): Delete.
655 * config/tc-v850.c (v850_reloc_prefix): Recoded to use CHECK_ ()
657 (handle_tdaoff, handle_zdaoff, handle_sdaoff): New functions.
659 * config/tc-v850.c (md_assemble): Corrected typo.
660 * config/tc-v850.c Add new sections: call_table_data and
662 (v850_reloc_prefix): Add support for ctoff() relocation prefix.
663 (handle_ctoff): New Function.
665 * doc/c-v850.texi (V850 Opcodes): Document call table relocations.
669 * config/tc-v850.c (v850_reloc_prefix): Add support for a 16 bit
670 displacement from the tiny data area pointer.
674 * config/tc-hppa.c (fix_new_hppa): Make declaration match
679 * config/tc-v850.c (processor_mask): New variable.
680 (set_machine, md_parse_option): Set processor_mask.
681 (md_assemble): Check that instruction is available to target
684 * config/tc-v850.h (TARGET_PROCESSOR): New constant.
688 Merge in work from Martin Hunt:
690 * config/tc-d30v.c (build_insn): For mvfsys and mvtsys,
691 CR is 0 for PSWL and PSWH.
693 * config/tc-d30v.c (do_assemble): Don't accept
694 illegal condition codes for cmpu instruction.
696 * config/tc-d30v.c: Add support for BFD_RELOC_D30V_9_PCREL
697 used in d*i instructions.
699 * config/tc-d30v.c (check_size): New function. Check
700 relocations for overflows.
701 (md_pcrel_from_section): Fix relocations between sections.
702 (md_apply_fix3): Use new relocation types for 15 and 21
703 bit relocations in the right container. Needed because
704 the address of the instruction is not eight-byte aligned
705 but the relocations must be.
707 * config/tc-d30v.c (md_apply_fix3): Check for overflow.
708 (find_format): If ".s" or ".l" are used, don't try
709 to compute branch sizes.
711 * config/tc-d30v.c (do_assemble): Check for ".s" or
712 ".l" extensions to opcode names.
713 (find_format): Generate the correct instructions when
714 ".s" or ".l" are used.
716 * config/tc-d30v.c (build_insn): Check for odd registers
717 on instructions that require even registers.
719 * config/tc-d30v.h (md_start_line_hook): Define.
720 * config/tc-d30v.c (md_start_line_hook): New hook.
721 Checks the beginning of each line for a ".". If it
722 finds one, assume a pseudo-op and flush any unwritten
725 * config/tc-d30v.c (md_apply_fix3): Fix problem
726 with determining when fixups were done.
728 * config/tc-d30v.c (build_insn): Fix bug where the numeric
729 part of a symbol (for example, "foo+8") was being written
730 into the instruction.
731 (md_pseudo_table): Change .word to be 32 bits and add
734 * config/tc-d30v.c (parallel_ok): Check to see if first
735 instruction is a jump.
737 * config/tc-d30v.c (parallel_ok): Major code reorganization.
741 * config/tc-v850.c (md_assemble): Corrected spelling mistake.
742 * configure.in (emulations): Add v850 emulation.
746 * Makefile.am (CPU_TYPES): Add arc.
747 (TARGET_CPU_CFILES): Add tc-arc.c.
748 (TARGET_CPU_HFILES): Add tc-arc.h.
749 (dependencies): Rebuild.
750 * Makefile.in: Rebuild.
751 * configure.in: Recognize arc-*-elf*.
752 * configure: Regenerated.
753 * config/tc-arc.[ch]: New files.
757 * doc/c-v850.texi (V850 Opcodes): Document hi0() reloc prefix.
758 Correct description of hi() reloc prefix.
760 * doc/c-v850.texi (V850 Opcodes): Document new reloc prefix.
761 * config/tc-v850.c (v850_reloc_prefix): Add hilo() reloc prefix.
762 * config/tc-v850.c (md_assemble): Add support for BFD_RELOC_32.
764 * doc/c-v850.texi: Document new pseudo ops and command line
767 * config/tc-v850.c (set_machine): New function.
768 * config/tc-v850.c (.v850): New pseudo op.
769 * config/tc-v850.c (.v850e): New pseudo op.
770 * config/tc-v850.c (.v850ea): New pseudo op.
775 Support -alh and -ald for DWARF 1:
776 * listing.c (struct list_info_struct): Add debugging field.
777 (listing_newline): Initialize the debugging field. If ELF, if the
778 section starts with .debug or .line, set the debugging field in
779 the listing structure.
780 (debugging_pseudo): Add list parameter. Change all callers. If
781 the debugging field is set, consider it to be a debugging pseudo.
782 If ELF, skip blank lines between debugging lines.
783 * read.c (emit_expr): If ELF, look for line numbers.
784 (stringer): If ELF, look for file names.
788 * config/tc-v850.c (v850_insert_operand): Only test for overflow
789 if there is no insert function.
791 * config/tc-v850.h (TARGET_MACHINE): New constant.
793 * config/tc-v850.c (v850_insert_operand): Add
794 -mwarn_unsigned_overflow.
795 (md_begin): Set BFD machine number based on machine variable.
796 (md_parse_option): Add -mv850, -mv850e and -mv850ea options.
800 * as.h: Don't declare alloca if it is a macro.
805 * config/tc-alpha.c (md_parse_option): Move m[] out to top level and
806 rename to cpu_types[].
807 (s_alpha_arch): New function.
808 (md_pseudo_table): Add "arch".
810 * config/tc-alpha.c (md_begin): Merge the two loops through the
812 (s_alpha_proc): Add initial SKIP_WHITESPACE.
813 (s_alpha_set): Likewise. Use get_symbol_end instead local while loop.
817 * read.h (s_lcomm_bytes): Add prototype (for real this time).
821 * config/obj-elf.c (elf_frob_symbol): Only set BSF_OBJECT for
826 * config/tc-v850.c: Remove BFD_RELOC_V850_16_PCREL.
830 * config/tc-mn10200.c (md_convert_frag): PC relative instructions arex
831 relative to the next instruction, not the current instruction.
832 (md_assemble): Similarly.
836 * doc/c-v850.texi: Explanations of offsets in SDA/ZDA areas
839 * config/tc-v850.c: Add support for SDA/TDA/ZDA sections.
840 (v850_reloc_prefix): Duplicate code eliminated. Add code to
841 recognise special instructions.
842 (md_assemble): Calculation of the size of a fixups corrected.
844 * config/tc-v850.h (ELF_TC_SPECIAL_SECTIONS): Add SDA/TDA/ZDA
849 * config/tc-v850.c (md_assemble): Use opcode->name instead of
850 opcode->opcode as the sentinal. Zero is a valid opcode.
854 * doc/as.texinfo (Machine Dependencies): Add v850 to menu.
855 * doc/c-v850.texi: Change node name to match other chapter nodes.
859 * doc/c-v850.texi (V850 Opcodes): Correct name for tiny data area
864 * expr.c (integer_constant): If BFD64, don't make a bignum if the
865 number will fit in 64 bits.
867 * config/tc-alpha.c (load_expression): Check explicitly for O_big,
868 rather than calling abort.
870 * as.h: Don't define alloca if __GNUC__. Just declare it.
871 * macro.c: Copy alloca handling from as.h.
873 * config/tc-i386.c (i386_align_code): Correct 16 bit noops. From
876 * config/tc-i386.c (md_assemble): In JumpByte case, when looking
877 for a WORD_PREFIX_OPCODE, change it to ADDR_PREFIX_OPCODE if this
878 is jcxz or a loop instruction.
882 * config/tc-v850.c (pre_defined_registers): Add 'hp' as alias for
884 (md_begin): Set up machine architecture and type.
888 * symbols.c (resolve_symbol_value): Store the value back into the
889 symbol expression, to handle add or subtract simplification
890 correctly. Handle O_symbol_rva. Add default case.
892 * config/tc-ppc.c (ppc_change_csect): Temporarily lower the
893 chunksize while creating the new subsection.
894 * as.c (chunksize): Initialize to zero.
895 * subsegs.c (subseg_set_rest): Change 5000 to chunksize when
896 calling obstack_begin.
900 * config/tc-v850.c (md_assemble): Restore input_line_pointer upon
903 * config/tc-v850.c (parse_register_list): Support constant
904 expressions as register lists.
908 * doc/c-v850.texi: Change the major node to v850 Machine
913 * doc/as.texinfo: Add inclusion of c-v850.texi
915 * doc/c-v850.texi: New file.
917 * read.c (is_end_of_line): Make NUL character be considered to be
922 * config/tc-v850.c (parse_register_list): Add support for curly
924 (cc_names): Add "e" and "ne" conditions.
928 * app.c (do_scrub_chars): Support a double dash as starting a
929 comment that extends to end of line.
933 * config/tc-v850.c (v850_section, v850_bss, v850_offset): New
935 (md_pseudo_table): New pseudo ops: .bss, .offset, .section
939 * config/tc-m32r.c (md_estimate_size_before_relax): Update recorded
940 insn when changing to a different instruction.
944 * config/tc-sh.c (parse_reg, get_specific, build_Mytes): Add SH4
945 floating point extensions.
946 (parse_reg): parse sgr and dbr.
950 * config/tc-v850.c (system_register_name): Support numbers for
955 * read.c (s_lcomm_internal): Renamed from s_lcomm, added arg to
956 flag when alignment is in bytes instead of power of 2, and code to
957 use that flag to convert alignment to bytes.
958 (s_lcomm, s_lcomm_bytes): New helpers that call s_lcomm_internal.
959 * read.h (s_lcomm_bytes): Add prototype.
960 * config/obj-coff.c (write_object_file): If ALIGNMENT_IN_S_FLAGS is
961 defined, write alignment to alignment bits in section header s_flags
962 rather than the s_align field.
963 * config/obj-coff.h (ALIGNMENT_IN_S_FLAGS): Define for TC_TIC80.
964 * config/tc-tic80.c (md_pseudo_table): Use s_lcomm_bytes for bss
965 pseudo, instead of s_lcomm which wants a power of two for alignment.
969 * macro.c (check_macro): use alloca instead of xmalloc to plug leak.
973 * as.c (show_usage): Add -am.
974 * input-scrub.c (input_scrub_include_sb): Don't add leading \n
975 if we've already got one.
976 * listing.c (struct list_info_struct): Add line_contents.
977 (listing_newline): Put unused argument to work: if non-null, save it...
978 (listing_listing): ... and regurgitate during listing instead of line
980 * listing.h (LISTING_MACEXP): New define.
981 (LISTING_NEWLINE): Argument is NULL.
982 * read.c (read_a_source_file): If expanding macros, break up input
983 lines and pass them to listing_newline.
984 * doc/as.texinfo: Document -ac and -am.
986 * cond.c (s_ifc): Add missing demand_empty_rest_of_line.
990 * config/tc-v850.c (md_apply_fix3): Add support for new 16 bit PC
995 * config/tc-v850.c: Remove support_v850e flag and command line
998 * configure.in (emulations): Add support for v850e target
1000 * configure (emulations): Add support for v850e target
1004 * config/tc-v850.c: Remove support_v850ea flag and command line
1007 * configure.in (emulations): Add support for v850ea target
1009 * configure (emulations): Add support for v850ea target
1013 * Makefile.am (check-DEJAGNU): Don't cd into testsuite until after
1014 setting EXPECT and TCL_LIBRARY.
1015 * Makefile.in: Rebuild.
1017 * as.h (enum debug_info_type): Define.
1018 (debug_type): Declare.
1019 * as.c (debug_type): New global variable.
1020 (show_usage): Add --gstabs.
1021 (parse_args): Handle --gstabs.
1022 * read.c (generate_asm_lineno): Remove.
1023 (read_a_source_file): Output stabs debugging if appropriate.
1024 Change checks of generate_asm_lineno to check debug_type. Only
1025 generate ECOFF debugging if ECOFF_DEBUGGING is defined.
1026 * read.h (generate_asm_lineno): Don't declare.
1027 (stabs_generate_asm_lineno): Declare.
1028 * stabs.c (stabs_generate_asm_lineno): New function.
1029 * ecoff.c (add_file): Use debug_type, not generate_asm_lineno.
1030 Don't turn off debugging.
1031 (add_file): Remove old #if 0 code.
1032 (ecoff_new_file): Set debug_type, not generate_asm_lineno.
1033 (ecoff_directive_end): Don't generate stabs line symbols.
1034 (ecoff_generate_asm_lineno): Don't check stabs_seen. Don't set
1035 generate_asm_lineno.
1036 (line_label_cnt): Remove.
1037 (ecoff_generate_asm_line_stab): Remove.
1038 * ecoff.h (ecoff_generate_asm_line_stab): Don't declare.
1039 * doc/as.texinfo, doc/as.1: Document --gstabs.
1043 * config/tc-v850.c (md_assemble, md_show_usage, md_parse_option):
1044 Add support for v850ea instructions.
1046 * config/tc-v850.c (md_assemble, md_show_usage, md_parse_option):
1047 Add support for v850e instructions.
1049 * config/tc-v850.c (md_assemble): Fix error recovery to reload
1050 text of entire opcode.
1054 * doc/internals.texi: Document rs_leb128.
1058 * config/tc-m68k.c (m68k_ip): Give an error message for SIZE_BYTE
1059 in ABSL case, rather than calling abort.
1063 * as.h (enum _relax_state): Add rs_leb128.
1064 * read.c (potable): Add sleb128 and uleb128.
1065 (sizeof_*leb128, output_*leb128, emit_leb128_expr, s_leb128): New
1067 * read.h: Update prototypes.
1068 * symbols.c (resolve_symbol_value): Streamline quite a bit. Return
1069 the symbol value, add a second FINALIZE argument that prevents
1070 changes from being comitted. Update all callers.
1071 * write.c (cvt_frag_to_fill, relax_segment): Handle rs_leb128.
1072 * doc/as.texinfo: Document the new pseudos.
1076 * Makefile.am (MOSTLYCLEANFILES): Add site.bak, site.exp, stage,
1078 (DISTCLEANFILES): Define.
1079 * doc/Makefile.am (DISTCLEANFILES): Define.
1080 * Makefile.in, doc/Makefile.in: Rebuild.
1084 * configure.in: Define TARGET_BYTES_BIG_ENDIAN if endian is set.
1085 Don't set targ or gas_target. Define SCO_ELF and
1086 TARGET_SOLARIS_COMMENT when appropriate. Don't substitute for
1088 * Makefile.am: Remove @target_frag@.
1089 (INCLUDES): Remove $(INTERNAL_CFLAGS), $(CROSS), $(HDEFINES), and
1091 (dep-am): Mark as phony.
1092 * acconfig.h: Add TARGET_BYTES_BIG_ENDIAN, TARGET_SOLARIS_COMMENT,
1094 * config/arm-big.mt, config/arm-lit.mt: Remove.
1095 * config/mips-big.mt, config/mips-lit.mt: Remove.
1096 * config/ppc-big.mt, config/ppc-lit.mt: Remove.
1097 * config/ppc-sol.mt: Remove.
1098 * config/i386coff.mt, config/m68kcoff.mt: Remove.
1099 * config/m88kcoff.mt: Remove.
1100 * config/sco5.mt: Remove.
1101 * configure, config.in, Makefile.in: Rebuild.
1103 * Makefile.am ($(srcdir)/config/m68k-parse.h): New target, to
1104 further try to circumvent the .y.h rule.
1105 * Makefile.in: Rebuild.
1109 * Makefile.am: New file, based on old Makefile.in.
1110 * acinclude.m4: New file, from old aclocal.m4.
1111 * configure.in: Call AM_INIT_AUTOMAKE and AM_PROG_LIBTOOL. Remove
1112 shared library handling; now handled by libtool. Replace
1113 AC_CONFIG_HEADER with AM_CONFIG_HEADER. Call AC_PROG_YACC,
1114 AC_PROG_LEX, and AC_DECL_YYTEXT. Call AM_MAINTAINER_MODE,
1115 AM_CYGWIN32, and AM_EXEEXT. Don't call CY_CYGWIN32 or CY_EXEEXT.
1116 * config.in: New file, created by autoheader.
1118 * acconfig.h: Mention PACKAGE, VERSION, and USING_CGEN.
1119 * stamp-h.in: New file.
1120 * as.c (print_version_id): Change GAS_VERSION to VERSION.
1121 (parse_args): Likewise.
1122 * config/obj-vms.c: (Write_VMS_MHD_Records): Likewise.
1123 * doc/Makefile.am: New file, based on old doc/Makefile.in.
1124 * Makefile.in, doc/Makefile.in: Now built with automake.
1125 * aclocal.m4: Now built with aclocal.
1126 * configure: Rebuild.
1128 * cond.c (s_else): If not listing false conditionals, turn listing
1129 off in the false branch of the else.
1133 * config/tc-mips.c (macro): Fix handling of a double load from a
1134 symbol plus an offset.
1136 * ecoff.c (ecoff_build_symbols): Set fMerge to 0 for an FDR which
1137 has an associated external symbol.
1141 * config/tc-alpha.c (s_alpha_ucons): New function.
1142 (md_pseudo_table): Add unaligned data pseudos for DWARF.
1146 * config/tc-v850.c (md_assemble): Ignore the rest of the current
1147 line if we encounter an error.
1149 * config/tc-v850.c (md_assemble): Sign extend constants value
1150 for hi and hi0 expressions.
1151 (v850_insert_operand): Enable range checking for generic 16bit
1156 * config/tc-v850.c (md_assemble): Turn on fx_no_overflow for
1157 LO16, HI16 and HI16_S relocs.
1161 * configure.in: Use CYGWIN and EXEEXT autoconf macro to look for
1163 * configure: Regenerated with autoconf 2.12.
1164 * Makefile.in: Add $(EXEEXT) to all executables.
1168 * config/tc-hppa.c (md_apply_fix): Improve warnings for out of range
1169 unconditional branches.
1170 (hppa_fix_adjustable): Don't adjust anything with a RR% or LR%
1175 * config/tc-sparc.c (md_begin): Cast sparc_opcodes to PTR for hash_insert.
1179 * macro.c (define_macro): Make sure the index is in range before
1184 * config/tc-tic80.c (build_insn): Remove "extended" and replace with
1185 "fx" and "fxfrag". Add "ffrag". Change code to initialize and use
1186 the right f/ffrag and fx/fxfrag pairs since instruction may be split
1191 * config/te-go32.h (USE_ALIGN_PTWO): Define.
1192 * config/tc-i386.c (md_pseudo_table): If USE_ALIGN_PTWO is
1193 defined, use s_align_ptwo for .align.
1194 * configure.in (i386-*-msdosdjgpp*): New target.
1195 (i386-*-go32*): Set em to go32 and targ to coffgo32.
1196 * configure: Rebuild.
1200 * config/tc-sparc.c (last_opcode): New static local.
1201 (md_assemble): Don't issue "FP branch in delay slot" warning if
1202 the delay slot has been annulled.
1206 * config/tc-m68k.c (md_apply_fix_2): Check for PC relative reloc
1207 code if BFD_ASSEMBLER.
1211 * config/tc-v850.c (system_registers): Fix ordering of registers.
1215 * config/tc-tic80.c (build_insn): Initialize extended word to zero
1216 when it will be filled in later by relocation information.
1220 * config/tc-mips.c (macro_build): Restore check of fmt argument.
1221 (mips_ip): Fix ISA checks.
1225 * config/tc-tic80.c (build_insn): Fix endianness problem with
1230 * config/tc-mips.c (check_absolute_expr): Change warning to
1235 * config/tc-mips.c (macro_build): Refine code to check if an
1236 instruction is available on a particular cpu variant.
1237 (mips_ip): Likewise.
1241 * config/tc-i386.c (tc_i386_fix_adjustable): Change ifndef
1242 OBJ_AOUT to ifdef OBJ_ELF.
1243 (md_apply_fix3): When mangling 32 bit PC relative reloc for
1244 BFD_ASSEMBLER, handle one ELF case for COFF as well, and add a PE
1246 * write.c (fixup_segment): Change special case for i386-coff to
1247 not apply for i386-pe.
1248 * config/obj-coff.c (coff_adjust_section_syms): Only count fixups
1249 which were not done.
1250 (coff_frob_file_after_relocs): Rename from coff_frob_file.
1251 (coff_format_ops): Initialize frob_file_after_relocs field rather
1252 than frob_file field.
1253 * config/obj-coff.h (coff_frob_file): Don't declare.
1254 (coff_frob_file_after_relocs): Declare.
1255 (obj_frob_file): Don't define.
1256 (obj_frob_file_after_relocs): Define.
1257 * configure.in: Set bfd_gas to yes for i386-*-cygwin32.
1258 * configure: Rebuild.
1262 * config/obj-coff.c (fixup_segment): Never subtract section
1263 address from PC relative reloc which will be fully resolved.
1267 * ecoff.c (page_type): Renamed from page_t to avoid conflict
1268 with hpux10 header files.
1273 * read.c (do_align): If BFD_ASSEMBLER, only use NOP_OPCODE if
1275 * config/tc-i386.h (md_maybe_text): Define.
1276 (md_do_align): Use md_maybe_text.
1280 * config/tc-ppc.h (tc_fix_adjustable): Only check for GOT type
1281 relocations, don't check for symbol being external, weak, etc.
1285 * config/tc-ppc.h (tc_fix_adjustable): Don't let the assembler
1286 calculate relocations to any external symbol, because we might be
1287 linking a shared object and the symbol might be overriden or moved
1288 (for instance, moved into a static executable's .bss section).
1289 (GLOBAL_OFFSET_TABLE_NAME): Delete. This is an i386 wierdness.
1291 * config/tc-ppc.h (tc_fix_adjustable): GOT-based relocations can't
1292 be calculated by the assembler.
1294 * config/tc-ppc.c (md_apply_fix3): Handle @plt or @local branch
1295 whose destination lies in the same file, by ignoring the @plt or
1296 @local and aiming the branch at its destination.
1300 * symbols.c (copy_symbol_attributes): Copy BSF_OBJECT flag.
1301 * config/obj-elf.h (OBJ_COPY_SYMBOL_ATTRIBUTES): Copy size
1304 * config/obj-multi.h (OBJ_COPY_SYMBOL_ATTRIBUTES): Define instead
1305 of obj_copy_symbol_attributes.
1309 * config/tc-ppc.c (ppc_insert_operand): In 32 bit mode, with a
1310 signed operand, sign extend a 32 bit value to the host size.
1312 * Makefile.in (CFLAGS): Subsitute from configure script. From
1315 * config/tc-i386.c (i386_operand): Use alloca rather than a fixed
1316 buffer size to make a copy of the symbol.
1318 * Makefile.in (OBJS): Put @extra_objects@ on the same line as
1323 * write.c (write_object_file): In non BFD_ASSEMBLER code, as we
1324 step through the frags calling cvt_frag_to_fill, switch to
1325 SEG_DATA when we reach data_frag_root.
1329 * config/tc-mn10300.c (md_assemble): Allow an empty register
1330 list for instructions which use register lists.
1334 * config/tc-arm.c (md_apply_fix3): Make temp unsigned long.
1336 * config/tc-arm.c (arm_adjust_symtab): Only set storage classes if
1339 * config/tc-arm.c: Add prototypes for many static functions.
1340 (struct asm_opcode ): Add prototypes for parms field.
1341 (struct thumb_opcode ): Likewise.
1342 (fp_op2): Remove unused flags parameter.
1343 (output_inst): Make static.
1344 (arm_after_pass_hook): Remove unused ignore parameter.
1345 * config/tc-arm.h (arm_after_pass_hook): Declare.
1346 (arm_start_line_hook): Declare.
1347 (arm_frob_label): Declare.
1351 * depend.c (wrap_output): new prototype.
1355 * config/tc-sh.c (md_apply_fix): Check for overflow.
1357 * config/tc-m68k.c (md_section_align): If a.out and BFD, force
1358 section size to be aligned.
1362 * config/tc-sh.h (md_cons_align): Define.
1363 (sh_cons_align): Declare.
1364 * config/tc-sh.c (md_pseudo_table): Add .uaword and .ualong.
1365 (sh_no_align_cons): New static variable.
1366 (s_uacons): New static function.
1367 (sh_cons_align): New function.
1368 (sh_handle_align): Warn about misaligned data.
1369 * doc/c-sh.texi: Document .uaword and .ualong.
1373 * macro.c (macro_expand): In MRI mode, treat single quote as a
1374 separator character when checking for a positional argument.
1378 * config/tc-arm.c (md_parse_option): Merge in changes from
1381 * config/tc-arm.h: Merge in changes from armT-970328-branch.
1383 * configure.in (emulations): Add Thumb architecture support from
1384 armT-9703-28-branch.
1388 * doc/all.texi: Add enabling of ARM documentation.
1390 * doc/as.texinfo: Add ARM documentation from armT-970328-branch.
1394 * config/tc-mips.c: Added r3900 support.
1398 * as.c: (parse_args) `-t' option requires an argument.
1402 * config/tc-arm.c (md_begin): Change call to
1403 coff_arm_bfd_set_private_flags() to a call to
1404 bfd_set_private_flags().
1408 * Makefile.in: Rebuild dependencies.
1410 * config/tc-i386.c (tc_gen_reloc): Don't try to convert the type
1411 of a BFD_RELOC_RVA reloc.
1415 * config/tc-hppa.c (tc_fix_adjustable): Reject absolute calls/jumps.
1416 (hppa_force_relocation): Force a relocation for an absolute
1421 * doc/as.texinfo: Don't use @value in section names or index
1422 entries; it confuses texinfo.tex.
1426 * doc/as.texinfo: Updated for -MD option.
1427 * Makefile.in (CFILES): Added depend.c.
1428 (OBJS): Added depend.o.
1429 * as.h (start_dependencies, register_dependency,
1430 print_dependencies): New declarations.
1431 * depend.c: New file.
1432 * as.c (parse_args): Added -MD option.
1433 (main): Call print_dependencies.
1434 (show_usage): Added help for -MD.
1435 * read.c (s_app_file): Call register_dependency.
1436 (s_include): Call register_dependency when file is found.
1437 (read_a_source_file): Call register_dependency.
1441 * config/obj-coff.c (symbol_to_chars): If TE_PE, don't add the
1442 section address to the symbol value.
1446 * config/tc-mips.c (macro_build,mips_ip): Move the INSN_ISA field
1447 into the new membership field.
1451 * config/tc-arm.c (md_begin): If no cpu type is specified on the
1452 command line then the ARM7 is now chosen by default when setting
1453 the BFD machine and architecture.
1457 * config/tc-arm.c (global variables): Added 'uses_apcs_26' flag to
1458 hold APCS selection.
1459 (md_begin): Added code to generate flags to be set into the COFF
1460 header and the calls to the BFD functions to do this.
1461 (md_parse_option, md_show_usage): Added new command line
1462 options -mapcs-32, -mapcs-26, -marmv2, -marmv2a, -marmv3,
1463 -marmv3m, -marmv4, -marmv4t.
1465 * config/tc-arm.h (LOCAL_LABEL): Removed the definition of this macro
1466 as it is never used.
1470 * config/tc-mn10200.c (md_convert_frag): Prefix temporary
1471 label name with ".".
1472 * config/tc-mn10300.c (md_convert_frag): Likewise.
1476 * config/tc-mips.c (set_at): Check for bignum.
1477 (check_absolute_expr, macro, mips16_macro): Likewise.
1481 * config/tc-tic80.c (md_apply_fix): Check PC relative relocations
1482 for overflow/underflow, only insert lower 15 bits into instruction.
1486 * config/tc-i386.c (pi): Check for RegMMX.
1490 * expr.c (expr): When subtracting values in the same frag,
1491 subtract X_add_number rather than adding it.
1495 * config/obj-coff.c (write_object_file): Just pass NULL to
1496 md_do_align, not the address of a char holding NOP_OPCODE.
1498 * config/tc-mips.c (macro): Handle constants for M_LI_D and
1500 (mips_ip): For 'F', 'L', 'f', and 'l', generate a constant rather
1501 than an address if the floating point value looks sufficiently
1506 * config/tc-i386.c (md_section_align): If a.out and BFD, force
1507 section size to be aligned.
1511 * cond.c: Include "macro.h".
1512 (struct conditional_frame): Add macro_nest field.
1513 (initialize_cframe): Initialize macro_nest.
1514 (cond_finish_check): Add nest parameter. Change all callers.
1515 (cond_exit_macro): New function.
1516 * as.h (cond_finish_check): Update declaration.
1517 (cond_exit_macro): Declare.
1518 * input-scrub.c (macro_nest): Make globally visible.
1519 (input_scrub_next_buffer): Call cond_finish_check.
1520 * macro.h (macro_nest): Declare.
1521 * read.c (s_mexit): Call cond_exit_macro.
1523 * config/tc-i386.h (RegMMX): Define.
1524 * config/tc-i386.c (pi): Check for all register types.
1525 (type_names): Add RegMMX.
1526 (md_assemble): Handle RegMMX.
1530 * config/obj-coff.c (c_section_symbol): Clear the LOCAL bit #ifdef
1535 * config/tc-mips.c (nopic_need_relax): Add new parameter
1536 before_relaxing. Use it when testing ecoff_extern_size.
1537 (load_address, macro, md_estimate_size_before_relax): Fix all
1542 * config/obj-elf.c (elf_pseudo_table): Add "subsection".
1543 (obj_elf_subsection): New static function.
1547 * config/obj-coff.c (coff_header_append): Don't reset string_size
1548 each time through the loop.
1552 * Makefile.in (DISTSTUFF): Add itbl-parse.h.
1556 * doc/internals.texi (Porting GAS): Correct documentation for
1557 current configure handling of targ-cpu.h, et. al.
1558 (CPU backend): Document listing macros.
1560 * listing.c (data_buffer): Set size based on other listing macros,
1561 rather than always using 100.
1562 (data_buffer_size): Remove static variable.
1563 (calc_hex): Make data_buffer_size a local variable. Don't leave
1564 any slop when filling data_buffer.
1568 * doc/c-mips.texi: Document .set autoextend.
1572 * configure.in (i386-*-openbsd*, m68k-*-openbsd*,
1573 mips-dec-openbsd*, ppc-*-*bsd*, ns32k-pc532-openbsd*,
1574 sparc-*-openbsd*): New targets.
1575 * configure: Rebuild.
1579 * config/obj-elf.c (elf_frob_symbol): If TC_MIPS, set BSF_OBJECT
1580 for all undefined symbols.
1584 * config/tc-ppc.c (ppc_fix_adjustable): Handle zero length csects
1589 * configure.in (alpha*-*-openbsd*): New target.
1590 * configure: Rebuild.
1594 * configure.in (mips-*-openbsd*): New target.
1595 * configure: Rebuild.
1599 * config/tc-d30v.c (parallel_ok): Fix parallel checking
1600 for instructions using conditional execution.
1604 * config/tc-mips.c (insn_uses_reg): Correct test for fpr pairs.
1608 * Makefile.in (srcroot): Remove.
1609 (INSTALL): Set to @INSTALL@.
1610 (INSTALL_XFORM, INSTALL_XFORM1): Remove.
1611 (all, dvi): Don't set srcroot.
1612 (install): Depend upon as.new, gasp.new, and installdirs. Use
1613 $(program_transform_name) directly, rather than using
1614 $(INSTALL_XFORM) and $(INSTALL_XFORM1).
1615 (installdirs): New target.
1616 * doc/Makefile.in (INSTALL_XFORM1): Remove.
1617 (install): Depend upon installdirs. Use $(program_transform_name)
1618 directly, rather than using $(INSTALL_XFORM) and
1620 (installdirs): New target.
1621 (install-info-as): Run mkinstalldirs.
1622 (install-info-gasp): Likewise.
1626 * Makefile.in (INSTALL): Change install.sh to install-sh.
1628 * symbols.c (resolve_symbol_value): Check for division by zero.
1631 * Makefile.in: Always use $(SHELL) when running move-if-change.
1632 * configure.in: Use ${CONFIG_SHELL} when running $ac_config_sub.
1633 * configure: Rebuild.
1637 * cgen.c (cgen_parse_operand): Renamed from cgen_asm_parse_operand.
1638 New argument `want'. Update enum cgen_parse_operand_result values.
1639 Initialize if CGEN_PARSE_OPERAND_INIT.
1640 * config/tc-m32r.c (md_begin): Set cgen_parse_operand_fn.
1641 (md_assemble): Call cgen_asm_init_parse.
1642 Update call to m32r_cgen_assemble_insn, call as_bad if assembly failed.
1646 * config/tc-m68k.c (m68k_ip): Handle #j.
1650 * config/tc-mn10300.c (md_convert_frag): Create fixup at the
1651 right address for call label:32,regs,imm.
1655 * config/tc-hppa.c (pa_subspace_start): If OBJ_ELF, then always return
1657 * config/tc-hppa.h (tc_frob_symbol): Don't reset the value of the
1658 symbol for OBJ_ELF anymore.
1662 * Makefile.in: Regenerate dependencies.
1663 (TARG_CPU): New variable.
1664 (cgen.o): Depend on cgen.h, $(TARG_CPU)-opc.h.
1665 (.dep1): Delete creating of cgen-opc.h.
1666 (.tcdep): Put proper contents in cgen-opc.h.
1667 * configure.in (m32r): Delete setting of extra_files, extra_links.
1668 (AC_OUTPUT): Create cgen-opc.h.
1669 * configure: Regenerated.
1673 * makefile.vms: Update to build gasp.exe.
1677 * write.c (relax_frag): Make non-static.
1678 * write.h (relax_frag): Add prototype for.
1679 * config/tc-m32r.h (md_do_align): New arg `max'.
1680 * config/tc-m32r.c (m32r_do_align): Likewise.
1681 Update calls to frag_align, frag_align_pattern.
1682 (fill_insn): Update call to m32r_do_align.
1683 (m32r_scomm): Update call to frag_align.
1685 * config/tc-m32r.[ch]: New files.
1687 * Makefile.in (CPU_TYPES): Add m32r.
1688 (TARGET_CPU_CFILES): Add tc-m32r.c.
1689 (TARGET_CPU_HFILES): Add tc-m32r.h.
1690 (DISTCLEAN_HERE): Add cgen-opc.h.
1691 (.dep1,.tcdep): Create empty cgen-opc.h.
1692 (cgen.o): Add dependencies.
1693 (dependencies): Regenerate.
1694 * as.h (struct frag): New member fr_targ.
1695 (fr_pcrel_adjust,fr_bsr): Move into union fr_targ.ns32k.
1696 * conf.in (USING_CGEN): New macro.
1697 * configure.in (m32r-*-*): Add entry for.
1698 Add cgen.o to extra_objects.
1699 * configure: Regenerate.
1700 * frags.c (frag_var): fr_pcrel_adjust renamed to
1701 fr_targ.ns32k.pcrel_adjust. fr_bsr renamed to fr_targ.ns32k.bsr.
1702 (frag_variant): Likewise.
1703 * write.c (relax_frag): Likewise.
1704 * config/tc-ns32k.c (*): Likewise.
1708 * config/tc-hppa.h (TC_EOL_IN_INSN): Check explicitly for '!',
1709 rather than for any end of line character.
1711 * config/tc-hppa.c (tc_gen_reloc): If hppa_ren_reloc_type fails,
1712 call abort (i.e., as_abort) rather than crashing.
1714 * config/tc-mips.c: Protect uses of STO_MIPS16 with an ifdef of
1715 OBJ_ELF, rather than of S_GET_OTHER.
1717 * Makefile.in (DISTCLEAN_HERE): Add site.exp and site.bak.
1721 * Makefile.in (VERSION): Set to 2.8.1.
1723 * Branched binutils 2.8.
1727 * COPYING: Update FSF address.
1729 * config/tc-mips.c (mips16_macro): Handle M_DMUL and M_MUL.
1733 * config/tc-mips.c (md_begin): Don't set interlocks for 4100.
1737 * config-gas.com: Update to handle both vax and alpha.
1738 * makefile.vms: Update to use config-gas.
1739 * conf-a-gas.com: Remove file.
1743 * Makefile.in: Remove unnecessary itbl-parse.h, ibtl-parse.c, and
1744 itbl-lex.c dependencies. Remove rules for itbl-lex.o,
1745 itbl-parse.o, and itbl-ops.o; just use the normal .c.o rule.
1749 * config/tc-tic80.c (line_comment_char): Make '#' start comments
1750 at the beginning of a line for compatibility with .S files where
1751 cpp leaves the filename transitions beginning with '#'.
1755 * config/tc-m68k.c: Only compile tc_coff_symbol_emit_hook and
1756 tc_coff_sizemachdep if OBJ_COFF.
1760 * config/tc-ppc.c (register_name): Declare.
1764 * configure.in (hppa*-*-rtems*): New target, like hppa-*-*elf*.
1765 * configure: Rebuild.
1769 * config/tc-mips.c (mips_pseudo_table): Add "stabn".
1770 (mips16_mark_labels): New static function.
1771 (append_insn): Call mips16_mark_labels.
1772 (mips_emit_delays): Likewise.
1773 (s_insn): Likewise. Don't call mips_clear_insn_labels.
1774 (s_mips_stab): New static function.
1776 * configure.in: Use ELF for mips-*-gnu*.
1777 * configure: Rebuild.
1781 * config/tc-m68k.h (TARGET_FORMAT): Set to "coff-m68k-sysv" if
1786 * configure.in: Add AC_ARG_ENABLE for commonbfdlib. If it is set,
1787 set OPCODES_LIB to empty.
1788 * configure: Rebuild.
1792 * configure.in (sparc-*-linux*aout*, sparc-*-linux*): New
1794 * configure: Rebuild.
1798 * itbl-parse.y (yyerror): Make static. Declare.
1801 * configure.in: Set emulations for mips-*-linux*-*.
1802 * configure: Rebuild.
1804 * config/tc-mips.c (struct mips_set_options): Define.
1805 (mips_opts): New static variable.
1806 (mips_isa): Remove. Now a field in mips_opts. Change all
1808 (mips16, mips16_autoextend, mips_warn_about_macros): Likewise.
1809 (mips_noreorder, mips_nomove, mips_noat, mips_nobopt): Likewise.
1810 (struct mips_option_stack): Define.
1811 (mips_opts_stack): New static variable.
1812 (s_mipsset): Add support for .set push and .set pop.
1813 * doc/c-mips.texi: Document .set push and .set pop.
1815 * config/obj-elf.c (obj_elf_section_change_hook): New function.
1816 * config/obj-elf.h (obj_elf_section_change_hook): Declare it.
1817 * config/tc-mips.c (s_change_sec): Call it if OBJ_ELF.
1821 * as.c (parse_args): Update copyright date in version message.
1823 * Makefile.in (clean-here): Remove dependency files.
1825 * read.c (s_comm): Check S_IS_COMMON as well as S_IS_DEFINED.
1826 (s_mri_common): Check S_IS_COMMON unconditionally.
1827 * symbols.c (colon): Check S_IS_COMMON as well as S_IS_DEFINED.
1828 * config/tc-alpha.c (s_alpha_comm): Likewise.
1829 * config/tc-mips.c (nopic_need_relax): Likewise.
1830 * config/tc-ppc.c (ppc_elf_lcomm): Likewise.
1831 (ppc_pe_comm): Likewise.
1832 * config/obj-elf.c (obj_elf_common): Likewise. Set segment of
1833 common symbol to bfd_com_section_ptr.
1834 * config/tc-sparc.c (s_common): Likewise.
1835 (tc_gen_reloc): Likewise.
1839 * config/tc-d30v.c (md_apply_fix3): Get the relocs right.
1843 * config/tc-i386.c (tc_i386_fix_adjustable): Only define if
1848 * input-scrub.c (input_scrub_next_buffer): Handle very long input
1851 * listing.c (print_lines): Add lineno parameter. Change all
1853 (listing_listing): Only call calc_hex for the right line.
1854 (listing_list): Set the new edict based on the current edict, in
1855 order to handle listing commands in macros correctly.
1857 * config/tc-mips.c (insn_uses_reg): Map register numbers in mips16
1860 * cond.c (cond_finish_check): New function.
1861 * as.h (cond_finish_check): Declare.
1862 * as.c (main): Call cond_finish_check.
1866 * config/tc-d30v.c (md_assemble): If two instructions
1867 are supposed to be assembled in parallel and the first one is
1868 long, print an error and stop.
1869 (md_apply_fix3): Don't calculate absolute relocs. Just write
1874 * config/tc-i386.h (iclrKludge): Define.
1875 * config/tc-i386.c (md_assemble): Handle iclrKludge.
1877 * config/tc-alpha.h (tc_frob_file_before_adjust): Define if
1879 (alpha_frob_file_before_adjust): Declare if OBJ_ECOFF.
1880 * config/tc-alpha.c (alpha_debug): New static variable.
1881 (md_parse_option): Set alpha_debug if -g is seen.
1882 (alpha_frob_file_before_adjust): New function if OBJ_ECOFF.
1886 * config/tc-d30v.c (build_insn): Enable range-checking code.
1887 (postfix): Stop at space or comma.
1888 (md_assemble): Change error message.
1892 * Makefile.in: Added automatic dependency building.
1893 * dep-in.sed: New file.
1897 * config/obj-ieee.c (segment_name): Don't define function if this
1900 * config/obj-coff.h (DO_STRIP): Don't define.
1901 * config/tc-h8300.h (DO_STRIP): Don't define.
1902 * config/tc-h8500.h (DO_STRIP): Don't define.
1903 * config/tc-w65.h (DO_STRIP): Don't define.
1904 * config/tc-z8k.h (DO_STRIP): Don't define.
1906 * symbols.c (colon): Call obj_frob_label if it is defined.
1907 * config/obj-vms.h (obj_frob_label): Rename from tc_frob_label.
1909 * configure.in: Don't set files and links. Don't call
1910 AC_LINK_FILES. Substitute te_file. Create targ-cpu.h,
1911 obj-format.h, targ-env.h, and itbl-cpu.h in AC_OUTPUT.
1912 * configure: Rebuild.
1913 * Makefile.in (TARG_CPU_C): New variable.
1914 (TARG_CPU_O, TARG_CPU_H): New variables.
1915 (OBJ_FORMAT_C, OBJ_FORMAT_O, OBJ_FORMAT_H): New variables.
1916 (TARG_ENV_H, ATOF_TARG_C, ATOF_TARG_O): New variables.
1917 (SOURCES): Rename from REAL_SOURCES. Delete old definition.
1918 (LINKED_SOURCES): Remove.
1919 (HEADERS): Rename from REAL_HEADERS. Delete old definition.
1920 (LINKED_HEADERS): Remove.
1921 (OBJS): Use $(TARG_CPU_O), etc., rather than targ-cpu.o, etc.
1922 ($(OBJS)): Depend upon $(TARG_ENV_H), etc., rather than
1924 ($(TARG_CPU_O), $(OBJ_FORMAT_O) $(ATOF_TARG_O)): New targets.
1925 (targ-cpu.o, obj-format.o, atof-targ.o): Remove targets.
1926 (itbl-cpu.h): Remove target.
1927 (DISTCLEAN_HERE): Remove targ-cpu.c, obj-format.c, atof-targ.c,
1932 * doc/as.texinfo (Symbol Names): Don't use obsolete @ctrl macro.
1936 * config/tc-m68k.c (mri_chip): Replace calls to get_symbol_end by
1937 open coded loop that does not require the name to start with a
1942 * frags.c (frag_var): Change offset parameter to offsetT.
1943 (frag_variant): Likewise.
1944 * frags.h (frag_variant, frag_var): Update declarations.
1945 * config/tc-m68k.c (struct m68k_it): Change foff field to
1947 (add_frag): Change off parameter to offsetT.
1948 * Several files: Add casts to calls to frag_var.
1950 * Makefile.in (m68k-parse.c): Depend upon itbl-parse.c, to
1951 serialize a parallel make.
1952 (itbl-parse.h): Split target out from itbl-parse.c.
1956 * config/m68k-parse.y (motorola_operand): Allow (zdireg,EXPR).
1958 * config/te-delta.h (COFF_COMMON_ADDEND): Define.
1959 * config/obj-coff.c (fixup_segment): Check COFF_COMMON_ADDEND when
1960 storing the value of a common symbol.
1964 * config/obj-coff.c (glue_symbols): Unused variable symbolP
1966 (crawl_symbols): Do not modify symbol_rootP and symbol_lastP here;
1967 that is done by symbol_remove and symbol_insert.
1969 * config/obj-coff.h (S_IS_LOCAL): Return 0 for a debugging
1974 * config/tc-mips.c (load_register): In 32 bit mode, when not
1975 dealing with a 64 bit number, permit the upper 32 bits to be set
1976 even if bit 31 is not set.
1980 * read.c (potable): Add "equiv".
1981 (s_set): Handle .equiv based on argument.
1982 * doc/as.texinfo (Equiv): New node to document .equiv.
1983 (Err): New node to document .err.
1987 * Many files: Add function prototypes.
1988 * as.c (show_usage, parse_args): Make static.
1989 * frags.h (frag_alloc): Declare.
1990 * subsegs.c (subseg_set_rest): Don't declare frag_alloc.
1991 * symbols.c (dollar_label_instance): Change return type to long.
1992 * symbols.h (print_symbol_value): Declare.
1993 (print_expr, print_expr_1, print_symbol_value_1): Declare.
1994 * write.c (fix_new_exp): Don't declare make_expr_symbol.
1995 (remove_subsegs, relax_frag): Make static.
1996 * config/atof-vax.c (atof_vax_sizeof): Change letter to int.
1997 (what_kind_of_float): Likewise.
1998 (atof_vax): Make static. Change what_kind to int.
1999 (md_atof): Change what_statement_type to int.
2000 * config/obj-ecoff.h (obj_ecoff_set_ext): Declare.
2001 * config/tc-alpha.c (vax_md_atof): Declare.
2002 (md_atof): Don't declare atof_ieee and vax_md_atof.
2003 * config/tc-i386.c (set_16bit_code_flag): Make static.
2004 * config/tc-i386.h (tc_i386_fix_adjustable): Declare.
2005 * config/tc-m68k.c (add_fix): Change width to int.
2006 (insert_reg): Change regname to const.
2007 (md_atof): Don't declare atof_ieee.
2008 (demand_empty_rest_of_line): Don't declare.
2009 * config/tc-m88k.c (md_atof): Don't declare atof_ieee.
2010 * config/tc-sparc.c (cmp_reg_entry): Change args to const PTR.
2011 (parse_keyword_arg): Change lookup_fn to take const arg.
2012 (md_atof): Don't declare atof_ieee.
2013 * config/tc-sparc.h: Add ifdef for multiple inclusion.
2014 (tc_aout_pre_write_hook): Don't declare.
2018 * as.h (bfd_alloc_by_size_t): Don't declare.
2019 * Many files: Use xmalloc rather than bfd_alloc_by_size_t.
2023 * symbols.c (symbol_new): Don't call debug_verify_symchain.
2024 (symbol_append): Set sy_next and sy_previous when adding a single
2025 symbol to an empty list. Call debug_verify_symchain.
2026 (verify_symbol_chain): Use assert, not know.
2030 * NEWS: Note BeOS support.
2031 * configure.in: (ppc-*-beos): New target, use coff as object format.
2032 * configure: Regenerate with autoconf.
2036 * config/tc-mips.c (md_apply_fix): Improve error message for out
2039 * Makefile.in: Add dependencies on obstack.h where needed.
2043 * config/tc-mips.c (md_estimate_size_before_relax): Handle the
2044 case of a symbol equated to another symbol when using SVR4_PIC.
2046 * Makefile.in (TARG_CPU_DEP_sparc): Add opcode/sparc.h.
2050 * read.c (read_a_source_file): Call LISTING_NEWLINE before
2051 HANDLE_CONDITIONAL_ASSEMBLY when handling an MRI line label.
2053 * config/obj-elf.c (obj_elf_data): Call md_flush_pending_output
2054 and md_elf_section_change_hook if they are defined.
2055 (obj_elf_text, obj_elf_previous): Likewise.
2059 * config/obj-multi.h (struct elf_obj_sy): Define if
2061 (OBJ_SYMFIELD_TYPE): Define as struct elf_obj_sy if
2063 * config/obj-elf.h (struct elf_obj_sy): Don't define if
2064 OBJ_SYMFIELD_TYPE is defined.
2066 * doc/as.texinfo (bss): Improve description of .bss section. In
2067 ELF or COFF, you are permitted to switch into the section.
2068 (Comm): Rewrite description of common symbols.
2069 (Lcomm): Mention that some targets permit a third argument.
2073 * config/tc-ppc.c (ppc_elf_lcomm): Don't call S_CLEAR_EXTERNAL.
2075 * symbols.c (colon): Change type of local to int. From Alan Modra
2078 * config/tc-m88k.c (m88k_do_align): Don't use a special nop
2079 alignment if a zero fill pattern was explicitly specified.
2080 * config/tc-sh.c (sh_do_align): Likewise.
2082 * read.c (equals): Always permit register names to be redefined.
2084 * config/tc-mips.c (mips_fix_adjustable): Permit a reloc against a
2085 mips16 symbol to be adjusted if a symbol is being subtracted from
2089 * config/obj-elf.c (obj_elf_symver): Check for duplicate or
2090 illegal symbol version names.
2091 (elf_frob_symbol): Check for external default versions.
2096 * config/obj-elf.h (struct elf_obj_sy): Define.
2097 (OBJ_SYMFIELD_TYPE): Define to elf_obj_sy struct. Change all
2099 * config/obj-elf.c (obj_elf_symver): Just record the name.
2100 (obj_symbol_new_hook): Initialized versioned_name field.
2101 (elf_frob_symbol): If there is a versioned_name, either rename the
2102 symbol, or add an alias with that name.
2106 * config/tc-mn10300.c (md_relax_table): Define.
2107 (md_convert_frag): Implement.
2108 (md_assemble): Handle relaxable operands/instructions correctly.
2109 (md_estimate_size_before_relax): Implement.
2110 * config/tc-mn10300.h (TC_GENERIC_RELAX_TABLE): Define.
2112 * config/tc-mn10200.c (md_relax_table): Fix typos.
2114 * config/tc-mn10300.c (md_assemble): Don't use any MN10300 specific
2115 relocs anymore. Tweak fx_offset for pc-relative relocs.
2119 * cond.c (s_ifc): Call mri_comment_field and mri_comment_end when
2124 * config/tc-tic80.c (md_pseudo_table): Add "sect" and "section"
2126 * config/tc-tic80.c (md_begin): Declare external variable
2127 coff_flags and insert an F_AR32WR bit into it.
2131 * read.c (equals): Add reassign parameter. Change all callers.
2132 * read.h (equals): Update declaration.
2136 * config/tc-mips.c (mips16_extended_frag): Don't assume that we
2137 can rely on the frag address to determine whether a frag is
2142 * write.h (LOCAL_LABEL): Only define if not BFD_ASSEMBLER.
2143 (S_LOCAL_NAME): Likewise.
2144 (FAKE_LABEL_NAME): Define unconditionally.
2145 * symbols.c (colon): Call bfd_is_local_label, not LOCAL_LABEL, if
2147 (S_IS_LOCAL): Call bfd_is_local_label_name, not LOCAL_LABEL.
2148 * config/tc-*.h: Only define LOCAL_LABEL if not BFD_ASSEMBLER.
2149 Don't define FAKE_LABEL_NAME.
2150 * config/te-ic960.h: Likewise.
2151 * config/tc-mips.h (tc_frob_file_before_adjust): Define.
2152 (mips_frob_file_before_adjust): Declare.
2153 * config/tc-mips.c (mips_frob_file_before_adjust): New function.
2154 (mips_local_label): Remove.
2156 * config/te-sco386.h: Remove; not used.
2160 * config/tc-tic80 (md_pseudo_table): Add align pseudo op to do
2161 byte alignment rather than power-of-two alignment that is the
2166 * config/tc-m68k.c (md_assemble): Handle a reloc width of 'W'.
2168 * gasp.c (hash_add_to_string_table): Correct misspelling in error
2169 message, and add newline.
2170 (process_file): Don't process assignments in the label if this is
2171 a equ or assign pseudo-op.
2172 (process_pseudo_op): Swap first argument to do_assign for K_ASSIGN
2173 and K_EQU, to match documentation.
2177 * config/obj-coff.c (obj_coff_section): Add 'r' section attribute
2178 to denote read-only data sections.
2182 * config/obj-elf.c (obj_elf_common): Set BSF_OBJECT in flags.
2183 * config/tc-sparc.c (s_common): Likewise, if BFD_ASSEMBLER.
2185 * expr.c (operand): Simplify 0b handling. Don't treat 0b as a
2186 binary number if the next character is '+' or '-'.
2190 * config/tc-tic80.c (FLT_CHARS): Change from "dD" to "fF".
2191 (find_opcode): Match operands that can be floats.
2192 (build_insn): Handle O_big (float) expressions and build
2197 * configure.in (mips*-*-lnews*): New target, also make empty
2198 emulation list for this target.
2199 * configure: Update.
2200 * config/tc-mips.c (ECOFF_LITTLE_FORMAT): Define.
2201 (mips_target_format): Use.
2202 * config/te-lnews.h: New file.
2206 * config/tc-tic80.c (find_opcode, build_insn): Changes to match
2207 operands with :m or :s modifiers and generate the right opcodes
2212 * Makefile.in (itbl-parse.c itbl-parse.h): Use $(BISON) and
2213 $(BISONFLAGS), not $(YACC) and $(YACCFLAGS).
2217 * config/tc-m68k.c (instring): Useless local declaration of
2218 crack_operand removed.
2219 * expr.h (expressionS): Changed type of X_op field to operatorT if
2225 * configure.in: Add i386coff and i386elf to emulation list.
2226 * configure: Rebuild.
2227 * as.c (i386coff, i386elf): Declare.
2228 * obj.h (coff_format_ops): Declare.
2229 * config/obj-coff.c (OBJ_HEADER): Define.
2230 (coff_obj_symbol_new_hook): Rename from obj_symbol_new_hook.
2231 (coff_obj_read_begin_hook): Rename from obj_read_begin_hook.
2232 (obj_pseudo_table): Add "version".
2233 (coff_pop_insert): New static function.
2234 (coff_sec_sym_ok_for_reloc): New static function.
2235 (no_func): New static function.
2236 (coff_format_ops): New variable.
2237 * config/obj-coff.h (coff_obj_symbol_new_hook): Declare.
2238 (obj_symbol_new_hook): Define.
2239 (coff_obj_read_begin_hook): Declare.
2240 (obj_read_begin_hook): Define.
2241 * config/tc-i386.h (i386_target_format): Declare.
2242 * config/tc-i386.c: Check OBJ_MAYBE_ELF as well as OBJ_ELF; check
2243 OUTPUT_FLAVOR when appropriate.
2244 (i386_target_format): New function.
2245 * Makefile.in (obj-coff.o): New target.
2246 (e-i386coff.o, e-i386elf.o): New targets.
2249 * config/tc-i960.h (TC_SYMFIELD_TYPE): Define if OBJ_COFF.
2250 (_tc_get_bal_of_call): Don't declare.
2251 (tc_get_bal_of_call): Declare as function, don't define as macro.
2252 * config/tc-i960.c (tc_set_bal_of_call): If OBJ_COFF, store balP
2253 in sy_tc field, not x_balntry field.
2254 (tc_get_bal_of_call): Rename from _tc_get_bal_of_call. Change
2255 return type to symbolS *. If OBJ_COFF, retrieve value from sy_tc
2256 field, not x_balntry field.
2258 * config/obj-elf.c (obj_elf_section): Permit a .note section to
2259 have the SHF_ALLOC attribute.
2261 * Makefile.in ($(OBJS)): Don't depend upon $(IT_HDRS).
2262 (TARG_CPU_DEP_mips): Depend upon $(srcdir)/itbl-ops.h.
2263 (itbl-lex.o): Depend upon itbl-parse.h.
2265 * itbl-parse.y (yyerror): Change return type to int. Change to
2266 use old style function declaration.
2268 * Makefile.in (itbl-lex.o): Remove -Wall.
2269 (itbl-parse.o): Likewise.
2271 * cond.c (s_ifdef): If we should omit conditionals from listings,
2273 (s_if, s_ifc, s_endif, s_else, s_ifeqs): Likewise.
2274 * listing.c (list_info_struct): Add EDICT_NOLIST_NEXT.
2275 (listing_listing): Handle EDICT_NOLIST_NEXT.
2276 (listing_list): An argument of 2 means EDICT_NOLIST_NEXT.
2277 * listing.h (LISTING_NOCOND): Define.
2278 (LISTING_SKIP_COND): Define.
2279 * as.c (show_usage): Mention c as a suboption of -a.
2280 (parse_args): Handle c as a suboption of -a.
2281 * doc/as.texinfo: Document -alc.
2285 * config/tc-tic80.c (md_apply_fix): Handle R_ABS type fixups.
2289 * doc/as.texinfo: Document .symver.
2293 * config/tc-d10v.c: Change pre_defined_registers to
2294 d10v_predefined_registers and reg_name_cnt to d10v_reg_name_cnt.
2298 * config/obj-coff.c: Fix typo in comment section.
2299 * config/tc-tic80.c (md_pseudo_table): Add entry for bss, which takes
2300 an additional alignment argument.
2301 (find_opcode): Allow O_symbol relocs for any 32 bit field, not just
2303 (build_insn): Handle O_symbol relocs for any 32 bit field, not just
2308 * Makefile.in: Remove dependancies on itbl-cpu.h.
2309 * as.c: Define stubs for itbl_parse and itbl_init if HAVE_ITBL_CPU
2314 * itbl-ops.h: Include as.h.
2318 * as.c: Remove -t option.
2319 * configure, configure.in: Move itbl-cpu.h to mips specific configure.
2320 * itbl-ops.h: Include itbl-cpu.h only if HAVE_ITBL_CPU is defined.
2321 * config/tc-mips.h: Define HAVE_ITBL_CPU.
2325 * itbl-ops.c: Don't define DEBUG.
2329 * Makefile.in: Update itbl-test.c to reflect its new location.
2333 * itbl-ops.c: Add test for itbl_have_entries.
2334 * config/tc-mips.c: Remove test for itbl_have_entries.
2335 * config/tc-mips.h: Define tc_init_after_args to mips_init_after_args.
2339 * Makefile.in (DISTSTUFF): Remove itbl-parse.y, itbl-lex.l, and
2340 itbl-ops.c. Add itbl-parse.c and itbl-lex.c.
2341 (LEX, LEXFLAGS): Define.
2342 * itbl-ops.c (append_insns_as_macros): Remove bogus ASSERT.
2346 * itbl-parse.y: Fix indentation mistakes from indent program.
2347 * itbl-lex.l: Fix indentation mistakes from indent program.
2348 * itbl-ops.h: Add include for ansidecl.h.
2349 Add PARAMS around function arguments.
2350 Add declaration for itbl_have_entries.
2351 * itbl-ops.c: Add PARAMS around function arguments.
2352 * Makefile.in: Add itbl build rules.
2353 Add dependancies for itbl files to mips target.
2354 * as.c: Add itbl support.
2355 Add new option "--insttbl" for dynamically extending instruction set.
2356 * as.h: Declare insttbl_file_name;
2357 the name of file defining extensions to the basic instruction set
2358 * configure.in, configure: Add itbl-parse.o, itbl-lex.o, and
2359 itbl-ops.o to extra_objects for mips configuration.
2360 Add include file link from itbl-cpu.h to
2361 config/itbl-${target_cpu_type}.h.
2362 * config/tc-mips.c: Allow copz instructions.
2363 Add notes for future additions to the itbl support.
2365 (macro): Call itbl_assemble to assemble itbl instructions.
2366 See if an unknown register is specified in an itbl entry.
2369 * doc/internals.texi (CPU backend): Fix typo in md_section_align
2371 * config/tc-tic80.h (NEED_FX_R_TYPE): Define.
2372 * config/tc-tic80.c (find_opcode): Add code to support O_symbol
2374 (build_insn): Grab a frag early so we can use the address in
2375 fixups. Take one's complement of BITNUM values before insertion
2376 in opcode. Add code to support O_symbol operands.
2377 (md_apply_fix): Replace unimplemented warning with implementation.
2378 (md_pcrel_from): Ditto.
2379 (tc_coff_fix2rtype): Ditto.
2383 * config/tc-d30v.c (parallel_ok): New function.
2384 * config/tc-d30v.h: Define TARGET_BYTES_BIG_ENDIAN.
2385 * config/tc-d10v.c (md_pcrel_from_section): Return 0 if
2386 relocation is in different section.
2390 * config/tc-mips.c (mips_ip): If configured for an embedded ELF system,
2391 don't set the section alignment to 2**4.
2395 * config/tc-m68k.c (line_comment_chars): Add '*'.
2397 * app.c (LEX_IS_TWOCHAR_COMMENT_2ND): Don't define.
2398 (do_scrub_begin): Don't set lex['*'].
2399 (do_scrub_chars): When handling LEX_IS_TWOCHAR_COMMENT_1ST, don't
2400 check for LEX_IS_TWOCHAR_COMMENT_2ND. Instead, just check for
2403 * configure.in: Set em=svr4 for m68k-*-sysv4*.
2404 * configure: Rebuild.
2405 * config/te-svr4.h: New file.
2406 * config/tc-m68k.c (m68k_comment_chars): Only include `#' if
2407 TE_SVR4 or TE_DELTA.
2411 * config/tc-mn10200.c (md_convert_frag): Create a fixup for the
2412 short conditional branch around a long unconditional branch.
2416 * config/obj-coff.c (obj_coff_ln [both versions]): Call
2419 * config/tc-arm.c (fix_new_arm): Use make_expr_symbol to handle a
2422 * symbols.c (resolve_symbol_value): If both left and right
2423 operands are undefined, warn about both of them.
2428 * config/obj-elf.c (elf_pseudo_table): Add "symver".
2429 (obj_elf_symver): New static function.
2430 * config/obj-elf.h (OBJ_COPY_SYMBOL_ATTRIBUTES): Copy the st_other
2433 * write.c (relax_segment): Make type and printf format agree.
2435 * read.c (get_line_sb): Don't end the line on a semicolon inside a
2440 * config/tc-d30v.c, config/tc-d30v.h: New files.
2442 * configure: Rebuilt.
2444 * configure.in: Add case for d30v.
2448 * config/tc-alpha.h (md_operand): Define with a null expansion,
2449 like all the other targets.
2450 * doc/internals.texi (CPU backend): Add missing word in
2451 md_flush_pending_output description. Fix typo in md_convert_frag
2453 * config/tc-tic80: Minor comment additions/changes.
2457 * config/tc-m68k.c (LOCAL_LABEL): Macro redefined if TE_DELTA.
2458 (tc_canonicalize_symbol_name): Macro defined if TE_DELTA.
2459 * config/obj-coff.c (obj_coff_def): Use
2460 tc_canonicalize_symbol_name if defined.
2461 (obj_coff_tag, obj_coff_val): Likewise.
2462 * expr.c (operand): Reject '~' as operator if is_name_beginner.
2467 does not actually work, though:
2468 * configure.in (i386-sequent-bsd*): New target.
2469 * configure: Rebuild.
2470 * config/tc-dynix.h: New file.
2471 * config/tc-i386.h: Define TARGET_FORMAT if TE_DYNIX.
2473 * read.c (do_align): Add max parameter. Change all callers.
2474 Remove useless static variables.
2475 (s_align): New static function. Do common portion of
2476 s_align_bytes and s_align_ptwo.
2477 (s_align_bytes, s_align_ptwo): Just call s_align.
2478 * frags.c (frag_align): Add max parameter. Change all callers.
2479 (frag_align_pattern): Likewise.
2480 * frags.h (frag_align, frag_align_pattern): Update declarations.
2481 * write.c (relax_segment): Limit alignment change to fr_subtype.
2482 Fix some types to be addressT.
2483 * config/obj-coff.c (size_section): Likewise.
2484 * config/obj-ieee.c (size_section): Likewise.
2485 * config/tc-d10v.h (md_do_align): Add max parameter.
2486 * config/tc-i386.h (md_do_align): Likewise.
2487 * config/tc-m88k.h (md_do_align): Likewise.
2488 * config/tc-m88k.c (m88k_do_align): Likewise.
2489 * config/tc-sh.h (md_do_align): Likewise.
2490 * config/tc-sh.c (sh_do_align): Likewise.
2491 * as.h: Improve comments on rs_align and rs_align_code.
2492 * doc/as.texinfo: Document new alignment arguments.
2493 * doc/internals.texi (Frags): Document use of fr_subtype field for
2494 rs_align and rs_align_code.
2498 * config/tc-mips.c: Changed opcode parsing.
2502 * config/{tc-alpha.h, tc-d10v.h, tc-generic.h, tc-i960.h,
2503 tc-mn10200.h, tc-mn10300.h, tc-sh.h, tc-vax.h, tc-w65.h}:
2504 Add default definition of zero for TARGET_BYTES_BIG_ENDIAN.
2505 * config/{tc-arm.h, tc-hppa.h, tc-i386.h, tc-mips.h, tc-ns32k.h,
2506 tc-ppc.h, tc-sparc.h}: Move definition of TARGET_BYTES_BIG_ENDIAN
2507 to a location consistent with the rest of the target include files.
2508 * config/tc-i386.c: Remove misleading comment.
2509 * doc/internals.texi (CPU backend): Add description of function
2510 md_undefined_symbol.
2511 * config/tc-tic80.c: Add code to insert predefined symbols into the
2512 symbol table so they can be parsed by the standard expression parser.
2513 Remove custom code that use to parse them.
2514 * config/tc-tic80.h: Move definition of TARGET_BYTES_BIG_ENDIAN
2515 to a location consistent with the rest of the target include files.
2519 * as.h: GNU c provides unlink() function.
2521 Unify section handling on openVMS/Alpha:
2522 * config/tc-alpha.c(s_alpha_link): Remove.
2523 (s_alpha_section): New function.
2524 Remove case-hacking of symbols
2525 Add .code_address pseudo-op.
2526 (BFD_RELOC_ALPHA_CODEADDR): New relocation.
2527 (s_alpha_code_address): New function.
2528 (alpha_ctors_section, alpha_dtors_section): New sections for C++
2529 static constructors/destructors.
2530 Add debug code for crash debugs, to be removed when traceback code
2531 is added to object code.
2532 (s_alpha_name): New function for .name pseudo-op.
2533 (alpha_print_token): New function to print token expressions with
2534 alpha specific extensions.
2536 * makefile.vms: Allow compilation with current gcc snapshot.
2540 * doc/Makefile.in (TEXI2DVI): Set to just name of program.
2541 (DVIPS): Set to dvips.
2542 (ps, as.ps, gasp.ps): New targets.
2543 (internals.info, gasp.dvi, internals.dvi): Set both TEXINPUTS
2544 and MAKEINFO env variables.
2545 (internals.ps): Use DVIPS macro.
2546 (clean): Remove core and backup files.
2547 (distclean): Remove temporary files from building internals.
2549 * doc/internals.texi (Frags): Fix typo.
2550 (GAS processing): Ditto.
2551 (CPU backend): Ditto.
2552 * ecoff.c (init_file): Use TARGET_BYTES_BIG_ENDIAN value directly.
2553 * mpw-config.in: Define TARGET_BYTES_BIG_ENDIAN as 1.
2554 * read.c: Remove ugly hack that dealt with config files not
2555 correctly defining TARGET_BYTES_BIG_ENDIAN.
2556 (target_big_endian): Use TARGET_BYTES_BIG_ENDIAN directly.
2557 * config/arm-big.mt: Define TARGET_BYTES_BIG_ENDIAN to 1.
2558 * config/arm-lit.mt: Define TARGET_BYTES_BIG_ENDIAN to 0.
2559 * config/mips-big.mt: Define TARGET_BYTES_BIG_ENDIAN to 1.
2560 * config/mips-lit.mt: Define TARGET_BYTES_BIG_ENDIAN to 0.
2561 * config/ppc-lit.mt: Define TARGET_BYTES_BIG_ENDIAN to 1.
2562 * config/ppc-sol.mt: Replace TARGET_BYTES_LITTLE_ENDIAN
2563 with TARGET_BYTES_BIG_ENDIAN defined to 0.
2564 * config/tc-arm.h: Remove use of TARGET_BYTES_LITTLE_ENDIAN
2565 and simplify. Test value of TARGET_BYTES_BIG_ENDIAN, not just
2566 whether it is defined or not.
2567 * config/tc-mips.h: Remove use of TARGET_BYTES_LITTLE_ENDIAN.
2568 * config/tc-ppc.h: Remove use of TARGET_BYTES_LITTLE_ENDIAN
2569 and simplify. Test value of TARGET_BYTES_BIG_ENDIAN, not just
2570 whether it is defined or not.
2571 * config/tic80.h (TARGET_FORMAT): Define to coff-tic80.
2572 (TARGET_BYTES_BIG_ENDIAN): Define to 0.
2576 * write.c (write_relocs): Correct text in as_fatal error message,
2577 bfd_perform_relocation -> bfd_install_relocation.
2581 * config/tc-m68k.c (LEX_TILDE): Define if TE_DELTA.
2582 * read.c (LEX_TILDE): Define if not defined.
2583 (lex_type): Use LEX_TILDE.
2584 * expr.c (get_symbol_end): Check first char with is_name_beginner,
2585 not is_part_of_name.
2589 * config/tc-sparc.c (md_show_usage): Add missing backslash at end
2592 * config/tc-mips.c (mips16_extended_frag): Correct base address
2593 for an extended PC relative instruction.
2594 (md_convert_frag): Likewise.
2596 * config/tc-mips.c (prev_nop_frag): New static variable.
2597 (prev_nop_frag_holds): New static variable.
2598 (prev_nop_frag_required): New static variable.
2599 (prev_nop_frag_since): New static variable.
2600 (append_insn): If we aren't reordering, and prev_nop_frag is not
2601 NULL, and we don't need any nops, then decrease the size of
2602 prev_nop_frag. Don't insert nops because of instructions in
2603 noreorder sections. Remember whether the previous instructions
2604 where in noreorder sections even when not reordering.
2605 (mips_no_prev_insn): Add preserver parameter. Change all
2606 callers. Refer prev_nop_frag variables when appropriate.
2607 (mips_emit_delays): Set up prev_nop_frag.
2608 (s_mipsset): Clear prev_nop_frag if reordering.
2612 * config/tc-mips.c (append_insn): Remove useless code which
2613 handled swapping a mips16 jump with a mips16 instruction with a
2616 * config/tc-mips.c (md_parse_option): When debugging, set
2617 mips_optimize to 1, not 0.
2619 * config/tc-mips.c (mips16_ip): Handle an extend operand.
2621 * config/tc-mips.c (my_getExpression): In mips16 mode, if it looks
2622 like the expression was based on `.', adjust the value of the
2625 * config/tc-mips.c (append_insn): Warn about an attempt to put an
2626 extended instruction in a delay slot when not reordering.
2627 (md_convert_frag): Warn if an extended instruction appears in a
2630 * config/tc-mips.c (mips_pseudo_table): Add "insn".
2631 (s_insn): New static function.
2632 * doc/c-mips.texi: Document .insn.
2634 * config/tc-mips.c (md_begin): Add the general registers to the
2636 (mips16_ip): First parse the expression, and then see whether it
2637 came up with a register, rather than trying to first see whether
2638 we are looking at a register.
2642 * config/tc-tic80.c: Numerous changes and additions to flesh
2643 out functions that were previously just stubs, and fix some
2644 problems found using the new TIc80 testsuite cases.
2648 * config/tc-mips.c (mips16_ip): Handle %gprel modifier.
2649 (md_apply_fix): Handle BFD_RELOC_MIPS16_GPREL.
2651 * config/tc-mips.c (append_insn): Output jump instruction as a
2652 pair of 2 byte instructions, rather than as a single 4 byte
2657 * itbl-ops.c, itbl-lex.l, itbl-parse.y, itbl-ops.h,
2658 config/itbl-mips.h: Add copyright message and fix indentation.
2662 * itbl-ops.c: New file. Add support for dynamically read
2663 instruction registers, opcodes and formats. Build internal table
2664 for new instructions and provide callbacks for assembler and
2666 * itbl-lex.l, itbl-parse.y: Lex and yacc parsers for instruction
2668 * itbl-ops.h: New file. Header file for itbl support.
2669 * config/itbl-mips.h: New file. Mips specific definitions for
2674 * config/tc-mn10200.c (md_assemble): If a constant operand won't
2675 fit into the constant field of a relaxable operand, then it does
2680 * config/tc-mn10200.c (md_estimate_size_before_relax): Treat
2681 a jsr target in a different section just like a jsr to
2682 an undefined target.
2686 * config/tc-mips.c (mips_fix_adjustable): Don't adjust relocations
2687 against any mips16 symbols, not just externally visible ones.
2688 (md_apply_fix): Corresponding change.
2692 * config/tc-mips.c (mips16_ip): Accept floating point registers in
2693 the operand of the exit instruction.
2697 * symbols.c (resolve_symbol_value): If we leave an equated symbol
2698 as O_symbol, copy over the segment.
2702 * config/tc-mips.c (md_apply_fix): If we aren't adjusting this
2703 fixup to be against the section symbol, adjust the value
2706 * symbols.c (resolve_symbol_value): Don't change X_add_number for
2708 * write.c (write_relocs): Avoid looping on equated symbols.
2709 Adjust fx_offset by X_add_number for each symbol.
2710 * config/obj-coff.c (do_relocs_for): Avoid looping on equated
2712 (fixup_segment): Add a loop to track down equated symbols and
2713 adjust fx_offset appropriately.
2717 * config/tc-mn10200.c (md_relax_table): Add entries to allow
2718 jmp -> bra relaxing.
2719 (md_convert_frag): Handle jmp->bra relaxing.
2720 (md_assemble): Handle jmp->bra relaxing.
2721 (md_estimate_size_before_relax): Likewise.
2725 * config/tc-i386.c (i386_align_code): Add comments explaining the
2730 * config/tc-sparc.c (enforce_aligned_data): New static variable.
2731 (sparc_cons_align): Don't do anything unless enforce_aligned_data
2733 (md_longopts): Add "enforce-aligned-data".
2734 (md_show_usage): Mention --enforce-aligned-data.
2735 * doc/c-sparc.texi (Sparc-Aligned-Data): New node; document
2736 enforce-aligned-data.
2738 * config/tc-ppc.c (md_pseudo_table): If OBJ_XCOFF, add "long",
2739 "word", and "short".
2740 (ppc_xcoff_cons): New static function.
2742 * write.c (relax_segment): Give an error if a .space symbol is
2743 common or undefined.
2745 * read.c (read_a_source_file): Don't handle mri_pending_align if
2746 the handler is s_globl or s_ignore.
2750 * config/tc-d10v.c (find_opcode): Remove unused variable "numops".
2751 * config/tc-tic80.c: Many additions to previous placeholder file.
2752 * config/tc-tic80.h: Ditto.
2756 * config/tc-i386.c (i386_align_code): Improve the nop patterns.
2760 * config/tc-mips.c (mips_fix_adjustable): New function.
2761 * config/tc-mips.h (tc_fix_adjustable): Call mips_fix_adjustable.
2762 (mips_fix_adjustable): Declare.
2765 * read.c (read_a_source_file): After doing an mri_pending_align,
2766 adjust the line_label if there is one.
2767 (s_space): Set mri_pending_align if an odd number of bytes were
2772 * config/tc-d10v.h (md_do_align): Add this hook to call
2773 d10v_cleanup() when a ".align" is detected.
2775 * config/tc-d10v.c (find_opcode): Correctly calculate
2776 branch displacement when .aligns are present.
2780 * config/tc-mn10200.c (md_relax_table): Define.
2781 (md_convert_frag): Implement.
2782 (md_assemble): Handle relaxable operands/instructions correctly.
2783 (md_estimate_size_before_relax): Implement.
2784 * config/tc-mn10200.h (TC_GENERIC_RELAX_TABLE): Define.
2788 * config/tc-mips.c (append_insn): Give an error for jumps to a
2790 (md_apply_fix): Make a branch to an odd address an error rather
2793 * config/tc-mips.c (md_convert_frag): If the user explicitly
2794 requested an extended opcode, pass warn as true to mips16_immed.
2796 * config/tc-mips.c (mips16_ip): Handle a missing expression like
2797 an explicit 0, so that explicitly extended instructions work
2802 * ecoff.c (ecoff_build_symbols): Don't generate a local ECOFF
2803 symbol for a common symbol.
2807 Patch presumed to have been checked in awhile ago but wasn't.
2809 * write.c: Delete "ifndef md_relax_frag" around is_dnrange.
2810 (relax_segment, case rs_org): Move code inside braces. Move locals
2811 target,after inside too.
2812 (relax_segment, case rs_machine_dependent): Guts moved to ...
2813 (relax_frag): New function.
2814 Call md_prepare_relax_scan if defined.
2818 * config/tc-m68k.c (m68k_ip): Reject pc-relative addresses for the
2819 'p' operand specifier.
2823 * config/tc-m68k.c (HAVE_LONG_BRANCH): New macro, returns true for
2824 m68k family cpus which support long branch addressing modes.
2825 (m68k_ip, md_convert_frag_1, md_estimate_size_before_relax,
2826 md_create_long_jump): Use it.
2830 * config/tc-mips.c (md_begin): Don't set SEC_ALLOC or SEC_LOAD for
2831 the .reginfo or .MIPS.options section if configured for an
2834 * config/tc-mips.c (md_begin): Don't set interlocks for
2839 * read.c (read_a_source_file): Make sure the symbol ends with
2840 whitespace before checking whether the next character is '='.
2844 * config/tc-i386.c (sco_id): Moved from here...
2845 * config/obj-elf.c (sco_id): ...to here. Adding the identifier
2846 really is an SCO ELF specific thing, not just a SCO x86 specific
2851 * configure.in (tic80-*-*): Don't require 'coff'.
2852 * configure: Regenerate.
2856 * read.c (emit_expr): Check for overflow of a negative value
2858 * write.c (fixup_segment): Likewise.
2859 * config/obj-coff.c (fixup_segment): Likewise.
2861 * config/tc-m68k.c (struct label_line): Define.
2862 (labels, current_label): New static variables.
2863 (md_assemble): Mark current_label as text, and clear it.
2864 (m68k_frob_label): New function.
2865 (m68k_flush_pending_output): New function.
2866 (m68k_frob_symbol): New function.
2867 * config/tc-m68k.h (tc_frob_label): Define.
2868 (md_flush_pending_output): Define.
2869 (tc_frob_symbol): Don't warn, just call m68k_frob_symbol.
2870 (tc_frob_coff_symbol): Likewise.
2872 * read.c (read_a_source_file): When defining a macro in MRI mode,
2873 don't add the symbol to the symbol table.
2877 * config/tc-mn10300.c (tc_gen_reloc): Handle sym1-sym2 fixups
2878 here since fixup_segment doesn't (linkrelax is set).
2879 * config/tc-mn10200.c (tc_gen_reloc): Likewise.
2883 * config/tc-mn10200.c (md_assemble): Tweak fx_offset for pc-relative
2888 * config/tc-hppa.c (struct hppa_fix_struct): Tweak fx_r_field's type
2889 to avoid warnings with the native HP compiler.
2890 (fix_new_hppa): Similarly for the r_type argument.
2891 (pa_build_unwind_subspace, hppa_elf_mark_end_of_function): Enclose
2892 in an #if OBJ_ELF to keep gcc -Wall quiet.
2893 (md_apply_fix): Always initialize "result".
2895 * config/tc-mn10200.c (md_assemble): Generate relocations.
2899 * config/tc-m68k.c (s_even): Adjust the alignment of the current
2904 * config/obj-elf.c (elf_file_symbol): When using ECOFF debugging,
2905 pass on the new file hook.
2907 * config/tc-alpha.c (alpha_fix_adjustable): Not quite the same as
2908 !alpha_force_relocation, as local LITERALs can be adjusted to be
2909 relative to the section.
2913 * config/obj-coff.c (yank_symbols): If tc_frob_coff_symbol is
2915 * config/tc-m68k.h (tc_frob_symbol): Check whether text label is
2916 aligned to odd boundary.
2917 (tc_frob_coff_symbol): Define.
2919 * doc/as.texinfo (Set): Change parenthesized @xref to @pxref.
2921 * macro.c (macro_expand_body): In MRI mode, just copy a single &.
2923 * config/tc-m68k.c (m68k_ip): Call frag_grow before adding a
2926 * config/tc-m68k.c (m68k_ip): Accept 'B' as a size for an
2928 (md_assemble): If the size is 'B', set fx_signed.
2929 (md_apply_fix_2): Use fx_signed when checking for overflow.
2931 * write.h (struct fix): Add fx_signed field.
2932 * write.c (fix_new_internal): Initialize fx_no_overflow and
2934 (fixup_segment): Use fx_signed when checking for overflow.
2935 * config/obj-coff.c (fixup_segment): Check fx_no_overflow and
2936 fx_signed when checking for overflow.
2940 * NOTES, NOTES.config: Removed. These are rarely, if ever,
2941 updated, and all the useful information is in doc/internals.texi.
2944 * read.c (read_a_source_file): Check for conditional operators
2945 before doing an MRI pending alignment.
2946 * config/tc-m68k.h (m68k_conditional_pseudoop): Declare.
2947 (tc_conditional_pseudop): Define.
2948 * config/tc-m68k.c (m68k_conditional_pseudop): New function.
2949 * doc/internals.texi (CPU backend): Describe
2950 tc_conditional_pseudoop.
2953 * config/tc-m68k.c (m68k_rel32_from_cmdline): New static
2955 (md_begin): Check m68k_rel32_from_cmdline before setting
2957 (m68k_mri_mode_change): Likewise.
2958 (md_longopts): Add --disp-size-default-16 and
2959 --disp-size-default-32.
2960 (md_parse_option): Handle new options.
2961 (md_show_usage): Mention new options.
2962 * doc/c-m68k.texi (M68K-Opts): Document new options.
2965 * config/tc-m68k.c (m68k_index_width_default): New static
2967 (m68k_ip): Use m68k_index_width_default to set the size of a base
2968 register whose size was not given.
2969 (md_longopts): Add --base-size-default-16 and
2970 --base-size-default-32.
2971 (md_parse_option): Handle new options.
2972 (md_show_usage): Mention new options.
2973 * doc/c-m68k.texi (M68K-Opts): Document new options.
2975 * doc/c-mips.texi: Mention ISA level 4, and the -mips16 option.
2977 * configure.in: Recognize mips-*-linux* target.
2978 * configure: Rebuild.
2980 * config/tc-mips.c (load_register): Rewrite 64 bit handling to
2981 work if valueT is only 32 bits.
2983 * config/tc-mips.c: Throughout, check target_big_endian rather
2985 (byte_order): Remove.
2986 (mips_init_after_args): Remove.
2987 * config/tc-mips.h (LITTLE_ENDIAN, BIG_ENDIAN): Don't define.
2988 (mips_init_after_args): Don't declare.
2989 (tc_init_after_args): Don't define.
2991 * config/tc-mips.h (tc_frob_after_relocs): Define if
2993 (mips_elf_final_processing): Likewise.
2994 (ELF_TC_SPECIAL_SECTIONS): Likewise.
2998 * config/tc-v850.c (md_assemble): If this is sst.{h,w} or
2999 sld.{h,w} and the operand is relocatable, adjust the adend by
3000 shifting it right one bit.
3004 * read.c (read_a_source_file): Check mri_pending_align after
3005 checking for a macro. From Ronald F. Guilmette
3008 * Makefile.in (ALL_CFLAGS): Add -D_GNU_SOURCE.
3010 * config/tc-sparc.c (md_apply_fix3): Rename from md_apply_fix, and
3011 add segment argument. If OBJ_ELF, treat a relocation against a
3012 symbol in a linkonce section like a relocation against an external
3014 * config/tc-sparc.h (MD_APPLY_FIX3): Define.
3018 * config/tc-mips.c (mips16_macro): Add case for M_ABS.
3022 * NOTES.config (Implementation): as.h #define's "GAS" not "gas",
3023 includes config.h instead of host.h, tc.h instead of tp.h, and
3024 targ-env.h instead of target-environment.h.
3025 Also, obj-format.h includes targ-cpu.h instead of
3027 * configure.in (case ${generic_target}): Add tic80-*-coff entry.
3028 * configure: Rebuild with autoconf.
3029 * config/obj-coff.h (coff/tic80.h): Include if TC_TIC80 defined.
3030 (TARGET_FORMAT): Define to "coff-tic80".
3031 * config/tc-tic80.c: New file for TIc80 support.
3032 * config/tc-tic80.h: New file for TIc80 support.
3036 * doc/as.texinfo (M): Mention explicitly that -M changes macro
3041 * write.c (adjust_reloc_syms): If the fixup symbol has been
3042 equated to an undefined symbol, convert the fixup to being against
3043 the target symbol. Remove obsolete code handling a special case
3048 * mpw-make.sed: Use NewFolderRecursive for installation.
3052 * config/tc-d10v.c (do_assemble): Correct previous bug fix.
3056 * config/tc-d10v.c (md_assemble): Fix bug which caused
3057 second instruction in a line to be case sensitize.
3061 * config/tc-mn10200.c (mn10200_insert_operand): Don't
3062 range check operands with MN10200_OPERAND_NOCHECK set.
3063 (check_operand): Likewise.
3067 * config/tc-mips.c: Undo part of last Friday's alignment changes.
3068 (md_begin): Always align the text section to a four byte
3070 (append_insn): Remove call to record_align.
3072 * config/tc-mips.c (insn_label): Remove.
3073 (struct insn_label_list): Define.
3074 (insn_labels, free_insn_labels): New static variables.
3075 (mips_clear_insn_labels): New static function.
3076 (append_insn): Mark all mips16 text labels, and make them odd.
3077 Handle all labels after emitting a nop, not just one. Call
3078 mips_clear_insn_labels rather than just clearing insn_label.
3079 (mips_emit_delays): Add insns parameter, and use it to decide
3080 whether to mark mips16 labels. Handle all labels, not just one.
3081 Force mips16 labels to be odd. Change all callers.
3082 (mips16_immed): Don't check for an odd branch target.
3083 (md_apply_fix): Don't check mips16 mode for a branch reloc.
3084 (mips16_extended_frag): Ignore the low bit in a branch target.
3085 (md_convert_frag): Likewise.
3086 (mips_no_prev_insn): Call mips_clear_insn_labels rather than just
3087 clearing insn_label.
3088 (mips_align, mips_flush_pending_output, s_cons): Likewise.
3089 (s_float_cons, s_gpword): Likewise.
3090 (s_align): Use insn_labels rather than insn_label.
3091 (s_cons, s_float_cons, s_gpword): Likewise.
3092 (mips_frob_file_after_relocs): New function.
3093 (mips_define_label): Rewrite to add to insn_labels list.
3094 * config/tc-mips.h (tc_frob_file_after_relocs): Define.
3095 * ecoff.c (ecoff_build_symbols): If the size of a function comes
3096 out odd, increment it.
3098 * config/tc-mips.c (append_insn): Only update prev_insn when not
3099 reordering if place is NULL.
3101 * config/tc-mips.c (mips16_ip): Check for a missing expression
3102 when using the register indirect addressing mode.
3106 * config/tc-mn10200.c (mn10200_insert_operand): Don't
3107 check 24bit operands for overflow.
3108 (check_operand): Likewise.
3112 * doc/as.texinfo (Section): Document how to use the .section
3113 pseudo-op for COFF and ELF.
3117 * write.c (adjust_reloc_syms): Fix linkonce check for ELF.
3121 * config/tc-mips.c (prev_insn_reloc_type): New static variable.
3122 (RELAX_MIPS16_ENCODE): Add dslot and jal_dslot arguments, and
3123 store them. Adjust other RELAX_MIPS16 macros.
3124 (RELAX_MIPS16_DSLOT): Define.
3125 (RELAX_MIPS16_JAL_DSLOT): Define.
3126 (append_insn): Pass new arguments to RELAX_MIPS16_ENCODE. Correct
3127 handling of whether previous instruction has a fixup. Set
3128 prev_insn_reloc_type.
3129 (mips_no_prev_insn): Clear prev_insn_reloc_type.
3130 (mips16_extended_frag): Use the right base address for a PC
3131 relative add or load.
3132 (md_convert_frag): Likewise. If a PC relative add or load is
3133 used, record the alignment for the section.
3137 * write.c (adjust_reloc_syms): Don't reduce a reloc against a
3138 linkonce section into a reloc against the section symbol.
3140 * config/tc-mips.c (mips16_macro): Remove nop instructions after
3141 branch instructions.
3143 * config/tc-mips.c (md_begin): If configured for an embedded ELF
3144 system, don't set the section alignment to 2**4.
3145 (s_change_sec): Likewise.
3146 (append_insn): Call record_alignment for the section.
3147 (md_section_align): Don't align the section size for an embedded
3152 * write.c (adjust_reloc_syms): Make sure that symbols are
3153 resolved; expression symbols may have been skipped.
3154 * config/obj-coff.c (fixup_segment): Likewise.
3158 * config/tc-ppc.c (ppc_elf_suffix): Move @plt to
3159 BFD_RELOC_24_PLT_PCREL relocation.
3160 (md_apply_fix3): Support BFD_RELOC_24_PLT_PCREL.
3164 * config/tc-d10v.c (write_2_short): Remove code that called
3165 parallel_ok() when the programmer specified parallel instructions.
3169 * config/tc-mn10300.c (md_assemble): Update to handle endianness
3172 * config/tc-mn10200.c (md_assemble): Opcode 0x0 is valid!
3173 * config/tc-mn10300.c (md_assemble): Likewise.
3177 * config/tc-mips.c (append_insn): Make sure there is enough room
3178 in a frag after a mips16 instruction to switch it with a jump
3181 * config/tc-mips.c (mips16_extended_frag): Give an error for an
3182 attempt to use a non absolute symbol in an extending frag.
3186 * config/tc-mn10200.c: Flesh out assembler support for MN10200.
3187 * config/tc-mn10200.h: Likewise.
3191 * app.c (do_scrub_chars): At the end of a C comment, pass space to
3192 UNGET rather than PUT. Set old_state before setting state to -2.
3194 * config/tc-mips.c (mips16_extended_frag): Avoid an infinite loop
3195 when extending because the value is exactly maxtiny + 1.
3197 * config/tc-mips.c (RELAX_MIPS16_ENCODE): Add small and ext
3198 arguments, and store them. Adjust other RELAX_MIPS16 macros.
3199 (RELAX_MIPS16_USER_SMALL): Define.
3200 (RELAX_MIPS16_USER_EXT): Define.
3201 (mips16_small, mips16_ext): New static variables.
3202 (append_insn): Pass mips16_small and mips16_ext to
3203 RELAX_MIPS16_ENCODE.
3204 (mips16_ip): Set mips16_small and mips16_ext.
3205 (mips16_immed): Don't check mips16_autoextend.
3206 (mips16_extended_frag): Check USER_SMALL and USER_EXT.
3208 * write.c (write_relocs): Print an error for an out of range
3209 fixup, rather than calling abort.
3211 * as.c (main): Unlink the output file if there are errors while
3212 generating the fixups.
3216 * config/tc-mips.c (mips16_extended_frag): Don't call
3218 (md_convert_frag): Call resolve_symbol_value before calling
3219 S_GET_VALUE, and don't add in the frag address.
3221 * config/tc-mips.c (mips16_immed): Add file and line parameters,
3222 and use them when reporting errors. Change all callers.
3226 * config/tc-mn10300.c: Fix various gcc -Wall warnings.
3227 Remove '$' prefixing for registers.
3231 * config/tc-d10v.c (md_assemble): Check to see if prev_seg
3232 is initialized before using it.
3233 (d10v_cleanup): No longer uses its argument, so make it void.
3235 * config/tc-d10v.h (d10v_cleanup): Change prototype.
3239 * write.c (fixup_segment): Don't discard the symbol for a PC
3240 relative fixup to an absolute symbol.
3244 * config/tc-d10v.c (md_assemble, d10v_cleanup): Fix bug
3245 with multiple sections.
3249 * config/tc-mips.c (md_longopts): Rename mips-16 to mips16, and
3250 no-mips-16 to no-mips16.
3251 (s_mipsset): Accept .set mips16 and .set nomips16.
3255 * config/tc-ppc.c (ppc_elf_suffix): Take expressionS pointer
3256 argument, and check for +/- constant following the suffix, folding
3257 it into the expression.
3258 (ppc_elf_cons): Change ppc_elf_suffix calls.
3259 (md_assemble): Ditto.
3260 (shlib): Replace boolean mrelocatable with enumeration shlib.
3261 (md_parse_option): Discriminate between PIC style shared libraries
3263 (ppc_elf_validate_fix): Don't report warnings for PIC style shared
3268 * config/tc-ppc.h ({tc,ppc}_comment_chars): Define, so that we can
3269 change the comment characters.
3271 * config/tc-ppc.c (comment_chars): Delete in favor of
3273 (ppc_{eabi,solaris}_comment_chars): Eabi and Solaris versions of
3275 (ppc_comment_chars): Select appropriate comment chars by default.
3276 (msolaris): New flag for -m{,no-}solaris.
3277 (md_parse_option): Recognize -K pic. Add support for
3279 (md_show_usage): Update.
3280 (md_begin): Do not set ELF flags if Solaris.
3281 (ppc_elf_suffix): @local sets R_PPC_LOCAL24PC relocation.
3282 (md_apply_fix3): Add support for R_PPC_LOCAL24PC.
3286 * as.c (main): Correct handling of flag_always_generate_output.
3290 * config/tc-mn10300.c (tc_gen_reloc): Get the addend from
3291 fx_offset, not fx_addnumber.
3293 * config/tc-mn10300.h (tc_fix_adjustable): Don't do any
3298 * config/tc-i386.c: Correct misspelling: balenced to balanced.
3302 * config/tc-mips.c (md_section_align): Check for an alignment of
3303 4, not an alignment of 16. Corrects August 7 patch.
3307 * configure, conf.in: Rebuild with autoconf 2.12.
3309 * config/tc-ppc.c (ppc_elf_lcomm): Don't give an error if no
3310 alignment is specified.
3312 Add support for mips16 (16 bit MIPS implementation):
3313 * config/tc-mips.c: Extensive additions for mips16 support, not
3315 (RELAX_OLD, RELAX_NEW): Use only 7 bits each.
3316 (insn_uses_reg): Change last parameter to an enum.
3317 * config/tc-mips.h (LOCAL_LABELS_DOLLAR): Define as 0.
3318 (md_relax_frag): Define as mips_relax_frag.
3319 (mips_relax_frag): Declare.
3320 (struct mips_cl_insn): Add use_extend and extend fields.
3321 (tc_fix_adjustable): Define.
3322 * config/obj-elf.h (S_GET_OTHER): Define.
3323 (S_SET_OTHER): Define.
3327 * config/tc-m68k.c (m68k_ip): Implement cases for new <, >, m, n,
3328 o and p operand specifiers.
3332 * write.c: Delete "ifndef md_relax_frag" around is_dnrange.
3333 (relax_segment, case rs_org): Move code inside braces. Move locals
3334 target,after inside too.
3335 (relax_segment, case rs_machine_dependent): Guts moved to ...
3336 (relax_frag): New function.
3337 Call md_prepare_relax_scan if defined.
3338 * config/tc-m68k.h (md_prepare_relax_scan): Renamed from
3343 * config/tc-mn10300.c (address_registers): Use '$' as register
3344 prefix instead of '%'.
3345 (data_registers, other_registers, md_assemble): Likewise.
3347 * config/tc-mn10300.c (address_registers): Use '%' prefix for regs.
3348 (data_registers, other_registers, md_assemble): Likewise.
3350 * config/tc-mn10300.c (md_assemble): Correctly determine the
3351 correct location and type for each relocation.
3352 (md_pcrel_from): Simplify.
3356 * config/tc-sh.c (md_convert_frag): Improve warning when branch is
3357 converted into branch around branch.
3361 * config/tc-mn10300.h (DIFF_EXPR_OK): Don't define this.
3362 (tc_fix_adjustable): Don't adjust relocs against weak symbols or
3364 * config/tc-mn10300.c (md_begin): Set linkrelax.
3365 (md_assemble): Create fixups as needed.
3366 (md_apply_fix3): Gut. It shouldn't ever get called anymore.
3370 * config/tc-d10v.c (parallel_ok): When automatically converting
3371 serial ops to parallel, do not consider a branch as the first
3376 * config/tc-mn10300.c (md_assemble): Handle MN10300_OPERAND_REG_LIST.
3380 * config/tc-mn10300.c (mn10300_insert_operand): Provide prototype
3382 (check_operand): Likewise.
3386 * config/tc-d10v.c (parallel_ok): Branch and link instructions
3388 (write_2_short): Call parallel_ok to check whether two short
3389 instructions the user requested execute in parallel, can be
3394 * config/tc-d10v.c (write_2_short): Fix bug that wouldn't
3395 allow a branch and link in parallel with an exe instruction.
3399 * doc/c-d10v.texi: Add info on @word modifier.
3403 * config/tc-mn10300.c (mn10300_insert_operand): MN10300_OPERAND_SPLIT
3404 operands are assumed to be 32bits. Use "bits" field to hold the
3405 number of bits in the main instruction word for MN10300_OPERAND_SPLIT.
3406 (mn10300_check_operand): MN10300_OPERAND_SPLIT operands are assumed
3409 * config/tc-mn10300.c (mn10300_insert_operand): Shift low part
3410 of a MN10300_OPERAND_SPLIT operand by operand->shift.
3412 * config/tc-mn10300.c (mn10300_insert_operand): Handle
3413 MN10300_OPERAND_SPLIT.
3417 * config/tc-mn10300.c (md_assemble): Insert operands into
3418 the extension part of the instruction if necessary.
3419 (mn10300_insert_operand): Accept pointer to extension word
3420 argument. Make insn a pointer argument too. Return type
3421 is now void. All callers changed.
3425 * config/tc-mn10300.c (mn10300_insert_operand): Handle
3426 repeated register operands.
3430 * doc/as.texinfo: Added section on reporting bugs.
3432 * config/tc-alpha.c: Change uses of void * to PTR. Change the
3433 alpha_macro emit field to expect a const argument, and change the
3434 arg field to be const. Fix some spacing to follow the GNU
3439 * config/tc-alpha.c (md_parse_option): Add knowledge of 21164pc
3440 (pca56) and 21264 (ev6) cpus.
3441 (md_apply_fix): Private relocation types are now negative.
3442 (alpha_force_relocation): Likewise.
3443 (tc_gen_reloc): Likewise.
3444 (emit_insn): Likewise.
3445 (emit_ldXu): Do the right thing when the hardware can do byte insns.
3446 (emit_stX): Likewise.
3447 (emit_sextX): Likewise.
3451 * config/obj-coff.c (do_relocs_for): Call resolve_symbol_value on
3452 a symbol found in a reloc.
3454 * symbols.c (resolve_symbol_value): Improve the error message if
3455 an undefined symbol is used in an expression.
3459 * doc/internals.texi: Rewrite, and add a lot of documentation.
3460 * doc/Makefile.in (internals.info): New target.
3464 * config/tc-v850.h (tc_fix_adjustable): Don't adjust relocs
3465 against weak symbols.
3469 * config/tc-v850.c (md_assemble): Don't lose for relaxable
3472 * config/tc-v850.c (md_convert_frag): Make sure we insert the
3473 fixup at the right address within the frag.
3475 * config/tc-v850.c (md_convert_frag): Don't set fragP->fr_fix
3476 to an absolute value, instead increment it as needed.
3478 * config/tc-v850.h (TC_GENERIC_RELAX_TABLE): Define.
3479 * config/tc-v850.c: Fix some indention problems.
3480 (md_relax_table): Define for D9->D99 branch displacement
3482 (md_convert_frag): Do something useful instead of aborting.
3483 (md_estimate_size_before_relax): Likewise.
3484 (md_assemble): Note if the matching instruction has a relaxable
3485 operand. If it does, allocate frag with frag_var and don't
3490 * config/tc-d10v.h (md_cleanup): New function. This is needed to
3491 write out any buffered instructions when a ".end" is found.
3495 * read.c (read_a_source_file): New hook md_cleanup().
3499 * write.c (fix_new_exp): Use make_expr_symbol to build an
3500 expression symbol for a complex fixup.
3504 * config/tc-v850.c (v850_reloc_prefix): Several disgusting
3505 hacks to improve parsing of complex hi, lo, zda, etc
3507 (md_assemble): Don't demand and eat a trailing ')' after finding
3508 a v850 relocation prefix. Sign extend the constant in a
3509 BFD_RELOC_LO16 expression. Do eat a trailing ')' after a complete
3511 (parse_cons_expression_v850): Don't eat a trailing ')' after
3512 finding a v850 relocation prefix.
3514 * config/tc-v850.h (TC_PARSE_CONS_EXPRESSION): Define.
3515 (TC_CONS_FIX_NEW): Likewise.
3516 * config/tc-v850.c (parse_cons_expression_v850): New function.
3517 (cons_fix_new_v850): Likewise.
3519 * config/tc-v850.h (tc_fix_adjustable): Don't adjust TDA relocs.
3523 * config/tc-ppc.c (md_apply_fix3): Give a better warning message
3524 for an unknown relocation type.
3528 * config/tc-v850.c (md_pseudo_table): Add .word; allocates
3533 * config/tc-v850.c (md_assemble): Handle TDAOFF relocs
3534 differently for movea & sst/sld insns.
3538 * config/tc-d10v.c (parallel_ok): Don't allow illegal combinations
3543 * obj.h (struct format_ops): Add frob_file_after_relocs field.
3544 * config/obj-multi.h (obj_frob_file_after_relocs): Define.
3545 * config/obj-ecoff.c (ecoff_format_ops): Initialize new
3546 frob_file_after_relocs field.
3547 * config/obj-elf.c (elf_format_ops): Likewise.
3548 * config/tc-mips.c: Undefine obj_frob_file_after_relocs before
3549 including obj-elf.h.
3553 * config/tc-mips.c (cons_fix_new_mips): Only treat 8 byte reloc
3554 specially if not ELF.
3555 (md_apply_fix): Handle BFD_RELOC_64.
3556 (tc_gen_reloc): Handle BFD_RELOC_64.
3558 * config/tc-i386.c (md_apply_fix3): Don't increment value for a PC
3559 relative reloc when BFD_ASSEMBLER and OBJ_AOUT (more ugly gas
3562 * config/obj-aout.h (S_IS_DEFINE): non BFD_ASSEMBLER version:
3563 Don't check S_GET_OTHER.
3567 * config/tc-mips.c (mips_ip): Accept an odd floating point
3568 register with l.s or s.s.
3570 * config/obj-aout.c (obj_pseudo_table): Use obj_aout_type for
3572 (obj_aout_type): New static function.
3576 * Makefile.in ($(OBJS)): Depend upon libiberty.h.
3580 * config/tc-v850.c (v850_reloc_prefix): Recognize zdaoff, tdaoff
3581 and sdaoff expressions.
3583 * write.c (fixup_segment): Don't add symbol value to addend if
3584 TC_V850 and OBJ_ELF.
3585 * config/tc-v850.h (tc_fix_adjustable): Don't adjust any
3588 * config/tc-v850.c (md_pcrel_from): Undo yesterday's changes.
3589 (md_pcrel_from_section): Likewise.
3590 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Likewise.
3594 * config/tc-v850.c (md_pcrel_from): Delete unused function.
3595 (md_pcrel_from_section): New function.
3596 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Define.
3600 * config/tc-mips.c (load_register): Add cast to offsetT when using
3605 * config/obj-elf.c (elf_frob_file): Move ECOFF debug processing to ...
3606 (elf_frob_file_after_relocs): ... here. New function.
3607 * config/obj-elf.h (obj_from_file_after_relocs): New macro.
3608 * write.c (write_object_file): Call *frob_after_relocs after the
3609 call to write_relocs.
3611 * config/tc-alpha.c: Use new BFD_RELOC_ALPHA_ELF_LITERAL reloc.
3613 * config/tc-alpha.c (load_expression): Don't SET_VALUE on the section
3614 symbol, as this messes up linking. Instead, expand the recursive call
3615 inline and change up the appropriate bits to get the 0x8000 offset
3616 in the reloc addend.
3620 * config/tc-sparc.h (tc_fix_adjustable): Permit the difference of
3621 two symbols in the same segment to be adjusted.
3623 * configure.in: Don't get confused by CPU-VENDOR-linux-gnu.
3624 * configure: Rebuild.
3628 * config/tc-ppc.c (ppc_insert_operand): Change most warnings into
3630 (ppc_elf_validate_fix): Ditto.
3631 (md_assemble): Ditto.
3633 (ppc_pe_section): Ditto.
3634 (ppc_frob_symbol): Ditto.
3638 * config/mn10300.c (md_assemble): Pass an extra shift count
3639 to mn10300_insert_operand based on the opcode format.
3640 (mn10300_insert_operand): Accept and use extra shift count
3643 * config/tc-mn10300.c (md_assemble): Use FMT_* macros for
3644 formats rather than hard-coded constants.
3646 * config/tc-mn10300.c (md_assemble): Format D5 instructions
3647 are 7 bytes long. Write out instructions in big-endian format.
3651 * config/tc-mn10300.c (md_assemble): Tweak further so
3652 that all instructions are parsed correctly.
3656 * as.h: Include libiberty.h.
3657 (xmalloc, xrealloc): Don't declare.
3658 * as.c: Don't include libiberty.h.
3659 * expr.c, read.c, stabs.c, config/obj-coff.c: Likewise.
3660 * config/tc-mips.c: Likewise.
3661 * messages.c: Likewise.
3662 (xstrerror): Don't declare.
3663 * xmalloc.c: Remove.
3667 * config/tc-mn10300.h (pre_defined_registers) Remove.
3668 (system_registers, cc_names): Likewise.
3669 (address_registers, data_registers, other_registers): New register
3671 (register_name, system_register_name, cc_name): Remove.
3672 (mn10300_reloc_prefix): Likewise.
3673 (data_register_name): New function.
3674 (address_register_name, other_register_name): Likewise.
3675 (md_assemble): Rough cut at parsing operands. Remove lots of
3677 (md_apply_fix3): Disable for now.
3681 * config/tc-m68k.c (select_control_regs): New function, extracted
3682 out of m68k_init_after_args.
3683 (m68k_init_after_args): Use it.
3684 (mri_chip): Use it here as well to update set of allowed control
3689 * config/obj-elf.c (elf_begin): New function.
3690 (obj_elf_section): Add the section symbol to the symbol table.
3691 * config/obj-elf.h (obj_begin): Define.
3692 (elf_begin): Declare.
3693 * as.c (perform_an_assembly_pass): Call obj_begin if it is
3698 * config/obj-coff.c (fixup_segment): Subtract the section address
3699 from a PC relative reloc if TC_M68K.
3703 * config/tc-sparc.c (md_pseudo_table): Make .uahalf, .uaword, and
3704 .uaxword available even if not OBJ_ELF.
3705 (md_atof): Remove unused local variable wordP.
3709 * config/tc-mn10x00.c, config/tc-mn10x00.h: New files
3710 for Matsushita MN10x00 support.
3711 * configure.in: Recognize mn10x00-*-*
3712 * configure: Rebuilt.
3716 * config/obj-evax.h: move openvms definitions from here to tc-alpha.c.
3717 * config/tc-alpha.c: add support for vms_case_hack like in vax/vms.
3718 (load_expression): track clobbering of base reg before jmp/jsr.
3719 (s_alpha_file): pass case_hack flags and source filename via
3720 symbol table to bfd.
3721 * config/tc-alpha.h (TC_CONS_FIX_NEW): define
3725 * configure.in (mips-*-rtems*): New target, like mips-*-elf*.
3726 * configure: Rebuild.
3730 * read.c (s_macro): Warn if a macro has the same name as a
3732 (s_space): In m68k MRI mode, align to a word boundary.
3733 * macro.c (define_macro): Add namep parameter. Change all
3735 * macro.h (define_macro): Update declaration.
3737 * as.c (show_usage): Print bug report address.
3738 (parse_args): Change version printing to match current GNU
3740 * gasp.c (show_usage): Print bug report address.
3741 (main): Change version printing to match current GNU standards.
3743 * config/tc-m68k.c (init_table): Correct access control unit
3746 * config/tc-alpha.c: Add some static function prototypes.
3747 (alpha_macros): Move to top of file. Make static.
3748 (alpha_num_macros): Move to top of file.
3752 * config/tc-v850.h: Define LOCAL_LABEL to recognise _.L_* symbols
3757 * listing.c (list_symbol_table): Remove bogus code in BFD64 case,
3758 and just call sprintf_vma.
3762 * expr.c (expr): Change >>= to >> (fix typo). (From meissner).
3766 * read.c (float_cons): Call md_flush_pending_output if it is
3771 * config/tc-d10v.c (md_operand): Created. Allows operands to
3773 * config/tc-d10v.h (md_operand): Undefined.
3777 * config/tc-m68k.c (add_fix): Treat a width of '3' like 'B'.
3778 (md_assemble): A fixup width of '3' means a 1 byte reloc.
3782 * config/obj-coff.c (fixup_segment): Don't adjust PC relative
3783 reloc for the i960 for a reloc in the same section. This undoes
3784 one of the two changes made Aug 19.
3788 * config/obj-coff.c (obj_coff_endef): Both versions: Move C_STAT
3789 symbols to the position of the debugging information.
3793 * expr.c (expr): Always use unsigned right shifts for >>.
3797 * config/tc-arm.c (md_apply_fix3): Update two thumb instruction
3798 slots when processing BL fixups.
3800 * config/tc-arm.c (output_inst): Ensure Thumb BL fixup is marked
3801 on the first half of the instruction.
3805 * ecoff.c (ecoff_stab): Create an expression symbol for a complex
3806 stabs expression, rather than giving an error.
3808 * ecoff.c (ecoff_new_file): Don't do anything if we are still in
3813 * config/tc-mips.c (append_insn): Fill in the value for a constant
3814 jump, rather than creating a reloc.
3818 * config/tc-mips.c (append_insn): Don't swap an instruction which
3819 sets a condition code with an instruction which uses a condition
3821 (mips_ip): In cases 'N' and 'M', look for $fccN rather than an
3824 * config/tc-mips.c (md_begin): Recognize r5000 for cpu.
3825 (mips_ip): Give a better error message if the ISA level is wrong.
3826 (md_parse_option): Recognize -mcpu=[v][r]5000.
3830 * config/tc-mips.c (COUNT_TOP_ZEROES): Added macro to count
3832 (load_register): Ensure hi32 bits are not lost during lo32bit
3833 processing. Fix shift offset that was overflowing into the next
3834 instruction field. Add code to generate shorter sequences for
3835 constants with a single contiguous seqeuence of ones.
3839 * config/tc-d10v.c (d10v_dot_word): New function to support
3840 "@word" with the word pseudo-op.
3841 (md_apply_fix3): Cleanup and changes to support correct sizes
3842 for 16 and 18-bit relocs.
3846 * configure.in (sparc-*-aout): Set `em'.
3847 * configure: Regenerated.
3848 * config/te-sparcaout.h: New file.
3849 * config/tc-sparc.h (TARGET_BYTES_BIG_ENDIAN): Define.
3850 Ifdef TE_SPARCOUT define TARGET_FORMAT and SPARC_BIENDIAN.
3851 * config/tc-sparc.c (INSN_BIG_ENDIAN): New macro.
3852 (SPECIAL_CASE_{SETSW,SETX}): Define.
3853 ({NOP,OR,FMOVS,SETHI,SLLX,SRA}_INSN): Define.
3854 (md_begin): Delete setting of `target_big_endian'.
3855 (output_insn): New function.
3856 (md_assemble): Rewrite. Add `setx' support.
3857 (sparc_ip): Handle `0' operand char. Recognize setuw, setsw, setx
3859 (md_atof): Add little endian support.
3860 (md_number_to_chars): Likewise.
3861 (md_apply_fix): Likewise.
3862 (md_longopts): Recognize -EL,-EB ifdef SPARC_BIENDIAN.
3863 (md_parse_option): Likewise.
3864 (md_show_usage): Print -EL, -EB ifdef SPARC_BIENDIAN.
3868 * ecoff.c (ecoff_new_file): New function.
3869 * ecoff.h (ecoff_new_file): Declare.
3870 * config/obj-ecoff.h (obj_app_file): Define.
3874 * config/tc-alpha.c (load_expression): Bias the .lit8 section
3875 symbol by 32k so that our 16-bit signed offset can address the
3880 * config/tc-mips.c (load_register): Remove unused variable tmp.
3884 * config/tc-mips.c (load_register): Remove unnecessary code that
3885 was causing the high 32bits of 64bit constants to be lost.
3889 * config/tc-d10v.c: Added changes to support function
3890 pointers and "@word" syntax.
3894 * config/tc-v850.c: Remove commented out and #if 0'd code.
3895 (v850_reloc_prefix): Provide prototype.
3896 (postfix, get_reloc, build_insn): Remove prototypes for nonexistant
3898 (md_begin, md_assemble, md_apply_fix3): Remove unused variables.
3899 (md_assemble): Add default to case statement.
3903 * config/tc-v850.c (md_assemble): Compute size of the instrction
3906 * config/tc-v850.c (md_apply_fix3): Do simple byte, short and
3911 * config/tc-v850.c (md_apply_fix3): Use little endian get/put
3912 routines to fetch/store the updated instruction from/to memory.
3913 (v850_insert_operand): If the operand has a specialized insert
3918 * config/tc-v850.c (reg_name_search): Align calling convention to
3919 be like identical function found in tc-ppc.c.
3920 (get_reloc): Removed.
3921 (v850_reloc_prefix): New function, parse lo(), hi() and hi0().
3922 (md_assemble): emit fixups.
3923 (md_pcrel_from): renamed from md_pcrel_from_section, emit proper
3925 (md_apply_fix3): handle fixups/relocs.
3926 * config/tc-v850.h (MD_PCREL_FROM_SECTION): Removed definition.
3931 * configure.in (sh-*-elf*): New target.
3932 * config/tc-sh.h (TARGET_ARCH): Define.
3933 (WORKING_DOT_WORD): Define.
3934 (TC_COFF_FIX2RTYPE): Only define if OBJ_COFF.
3935 (BFD_ARCH, COFF_MAGIC, TC_COUNT_RELOC): Likewise.
3936 (TC_RELOC_MANGLE, tc_coff_symbol_emit_hook): Likewise.
3937 (DO_NOT_STRIP, NEED_FX_R_TYPE, TC_KEEP_FX_OFFSET): Likewise.
3938 (TC_COFF_SIZEMACHDEP, tc_frob_file): Likewise.
3939 (SUB_SEGMENT_ALIGN): Likewise.
3940 (RELOC_32): Don't define.
3941 (tc_frob_file_before_adjust): Define if BFD_ASSEMBLER.
3942 (target_big_endian): Declare if OBJ_ELF.
3943 (TARGET_FORMAT): Define if OBJ_ELF.
3944 * config/tc-sh.c: Use BFD reloc codes instead of SH COFF reloc
3946 (tc_crawl_symbol_chain): Only define if OBJ_COFF.
3947 (tc_headers_hook, tc_coff_sizemachdep): Likewise.
3948 (struct sh_count_relocs): Define.
3949 (sh_count_relocs): New static function, broken out of
3950 sh_frob_file. Add BFD_ASSEMBLER code.
3951 (sh_frob_section): Likewise.
3952 (sh_frob_file): Call sh_frob_section.
3953 (md_convert_frag): If BFD_ASSEMBLER, change type of headers, and
3954 call section_symbol rather than seg_info (seg)->dot.
3955 (md_section_align): Add OBJ_ELF version.
3956 (SWITCH_TABLE_CONS): Define.
3957 (SWITCH_TABLE): Use SWITCH_TABLE_CONS.
3958 (md_apply_fix): Change parameter types if BFD_ASSEMBLER. Only
3959 handle fx_r_type == 0 if not BFD_ASSEMBLER. Return 0 if
3961 (struct reloc_map): Define if not BFD_ASSEMBLER.
3962 (coff_reloc_map): Likewise.
3963 (sh_coff_reloc_mangle): Use coff_reloc_map to convert fx_r_type.
3964 (tc_gen_reloc): New function if BFD_ASSEMBLER.
3965 * write.c (write_relocs): Ifdef out fx_where test which triggers
3966 inappropriately for SH ELF.
3967 (write_object_file): Call tc_frob_file_before_adjust and
3968 obj_frob_file_before_adjust if they are defined.
3970 * write.c (write_object_file): Use BFD_RELOC_16, not
3971 BFD_RELOC_NONE, when calling fix_new_exp for a broken word.
3973 * read.c (emit_expr): Fix conversion of byte count to BFD reloc
3978 * config/tc-d10v.c (find_opcode): Fix problem with calculating
3979 branch sizes in across sections.
3983 * config/tc-850.c (md_assemble): Handle hi() correctly. Handle
3988 * config/tc-v850.c (md_begin): Deal with end of opcode
3993 * config/tc-d10v.c (find_opcode): Fix a bug which could generate
3994 the wrong opcode for cases like st2w where there are many forms
3995 of the same instruction.
3999 * expr.c (operand): If md_parse_name is defined, call it before
4000 calling symbol_find_or_make.
4001 * config/tc-ppc.h (md_parse_name): Define.
4002 (ppc_parse_name): Declare.
4003 * config/tc-ppc.c (reg_name_search): Add regs and regcount
4005 (register_name): Update call to reg_name_search.
4006 (cr_operand): New static variable.
4007 (cr_names): New static const array.
4008 (ppc_parse_name): New function.
4009 (md_assemble): If PPC_OPERAND_CR is set in the operand flags, set
4010 cr_operand before calling expression.
4014 * config/tc-hppa.c (tc_gen_reloc): Add new argument to
4015 hppa_gen_reloc_type call.
4019 * config/tc-d10v.c: Fixed ".word". Fixed problem with range checking
4020 on addresses. Improved error messages.
4021 * doc/c-d10v.texi: Added docs for register pairs.
4025 * config/tc-d10v.c (parallel_ok): Fix bug in parallel
4030 * ecoff.c (init_file): Initialize fMerge to 1.
4031 (add_file): Restore old file merging code, but only merge files if
4033 (ecoff_directive_loc): Clear fMerge field of current file.
4034 (ecoff_generate_asm_lineno): Likewise.
4038 * doc/c-d10v.texi: Fix typo.
4042 * config/tc-v850.c (md_assemble): Correct bit masking for
4043 hi and lo expressions.
4045 * config/tc-v850.c (md_assemble): Rough cut at demanding
4046 "ep" or "r30" in sst and sld instructions.
4047 (md_apply_fix3): Don't abort. Just warn that we don't
4050 * config/tc-v850.c (CC_NAME_CNT): Define.
4051 (cc_name): New function.
4052 (md_assemble): Handle V850_OPERAND_CC correctly.
4054 * config/tc-v850.c (md_assemble): Don't forget to initialize
4057 * config/tc-v850.c (reg_name_search): Generalize to search
4058 any given register table.
4059 (register_name): Pass appropriate table and size to reg_name_search.
4060 (system_register_name): New function.
4061 (SYSREG_NAME_CNT): Define.
4062 (md_assemble): Handle operands which are system registers.
4064 * config/tc-v850.c (md_assemble): If we find a register, but the
4065 opcode doesn't want a register, then we don't have a match.
4066 (md_assemble): Get size of the instruction from the opcode table.
4070 * configure.in: Set and substitute HLDENV.
4071 * configure: Rebuild.
4072 * Makefile.in (HLDENV): New variable.
4073 (as.new): Use $(HLDENV).
4075 * ecoff.c (ecoff_directive_endef): Avoid a division by zero error
4076 if an array dimension is not known.
4080 * config/tc-d10v.c: Fix a reloc bug caused by my last change.
4081 * doc/c-d10v.texi: Cleanup.
4085 * config/tc-v850.c: New file.
4086 * config/tc-v850.h: New file.
4087 * configure (v850-*-elf): New target.
4088 * configure.in (v850-*-elf): New target.
4092 * doc/c-d10v.texi: New file.
4093 * doc/all.texi: Added D10V stuff.
4094 * doc/as.texinfo: Added D10V stuff.
4098 * config/tc-d10v.c: All references to defined symbols should
4099 now use the optimal instruction. .float and .double now work.
4103 * config/obj-coff.c (fixup_segment): Adjust PC relative reloc by
4104 section address for the i960 as is done for the i386.
4108 * mpw-config.in: Add wildcards for config matching, add mips-*-*
4109 case, forward-include bfd/elf-bfd.h.
4113 * config/tc-d10v.c: Add additional information to the opcode
4114 table to help determinine which instructions can be done
4119 * config/tc-arm.c: Major changes to add Thumb support, with lots
4121 Reverted to INSN_SIZE macro, rather than insn_size variable.
4122 (insns): Added ARM "bx" instruction support.
4123 (tinsns): Added Thumb instruction definition structure.
4124 (arm_tops_hsh): Added hash structure for Thumb opcodes.
4125 (md_pseudo_table): Added ".arm", ".thumb" and ".code" pseudo-ops.
4126 (opcode_select,s_arm,s_thumb,s_code): Added.
4127 (decode_shift): Allow upper-case RRX.
4128 (do_ldst): Simpler halfword support.
4129 (do_ldmstm): Improved.
4130 (reg_list, do_bx, thumb_reg, thumb_add_sub, thumb_shift,
4131 thumb_mov_compare, thumb_load_store, do_t_arit, do_t_add,
4132 do_t_asr, do_t_branch, do_t_bx, do_t_compare, do_t_ldmstm,
4133 do_t_ldrb, do_t_ldrh, do_t_lds, do_t_lsl, do_t_lsr, do_t_mov,
4134 do_t_push_pop, do_t_str, do_t_strb, do_t_strh, do_t_sub, do_t_swi,
4136 (md_apply_fix3): Add support for BFD_RELOC_ARM_THUMB_* relocations.
4137 (md_parse_option): Add support for -mthumb.
4138 (md_show_usage): Updated to reflect new command line option.
4139 (arm_data_in_code, arm_canonicalize_symbol_name): Added.
4140 * config/tc-arm.h: Provide TC_FIX_TYPE to allow private ARM
4141 fragment information to be held.
4145 * config/tc-arm.c (md_apply_fix3): Also set fixP->fx_done if fx_addsy is
4146 non-null, but is a constant.
4147 (fix_new_arm): Call make_expr_symbol to make the expression symbol
4148 so that error reporting will work correctly.
4152 * config/tc-i386.c (tc_i386_fix_adjustable): Don't adjust relocs
4153 against weak symbols.
4157 * config/tc-ppc.h (TC_FORCE_RELOCTION): Define if OBJ_XCOFF.
4158 (ppc_force_relocation): Declare if OBJ_XCOFF.
4159 * config/tc-ppc.c (ppc_force_relocation): New function if
4164 * config/tc-mips.h (BYTE_ORDER): Don't define. No longer used.
4168 * config/tc-d10v.c: Fix problem with relocs.
4172 * config/tc-sh.c (sh_do_align): If not BFD_ASSEMBLER, always align
4173 with nops if not in data_section or bss_section.
4177 Add support for openVMS/Alpha.
4178 * as.h (PRINTF_LIKE): Don't define if VMS, for now.
4179 * config/obj-evax.c: New file.
4180 * config/obj-evax.h: New file.
4181 * config/tc-alpha.c: Add support for EVAX format if OBJ_EVAX is
4183 * config/tc-alpha.h: Add support for EVAX format if OBJ_EVAX is
4184 defined. Add case for bfd_target_evax_flavour.
4185 * config/vms-a-conf.h: New file.
4186 * conf-a-gas.com: New file.
4187 * configure.in: Add target alpha-*-*vms*.
4188 * configure: Rebuild.
4189 * makefile.vms: New file.
4190 * read.c (s_lcomm): Align bss_seg on 8 byte boundary if OBJ_EVAX.
4191 Don't call ffs on openVMS/Alpha.
4195 * configure.in: Make GAS_CHECK_DECL_NEEDED include <string.h> or
4196 <strings.h> if they exist. Call GAS_CHECK_DECL_NEEDED on strstr
4198 * acconfig.h (NEED_DECLARATION_STRSTR): New macro.
4199 (NEED_DECLARATION_SBRK): New macro.
4200 * configure, conf.in: Rebuild.
4201 * as.h: Only include <strings.h> if HAVE_STRINGS_H.
4202 (strstr): Declare if NEED_DECLARATION_STRSTR.
4203 * as.c: If HAVE_SBRK and NEED_DECLARATION_SBRK, declare sbrk.
4207 * symbols.c (resolve_symbol_value): Handle addition or subtraction
4208 by a constant before entering the main switch. Reject attempts to
4209 apply an arithmetic function to non-absolute symbols, except for
4210 the special case of subtraction of two symbols in the same
4213 * config/tc-mips.c (md_section_align): Do align if OBJ_ELF, but
4214 not to more than a 16 byte boundary.
4216 * config/tc-i386.c (tc_gen_reloc): Accept all relocs; remove
4218 (tc_gen_reloc): If out of memory call as_fatal rather than
4219 assert. If no howto found, call as_bad_where rather than
4220 as_fatal. Change the error message slightly. Set howto to a
4221 non-NULL value in order to keep going.
4225 * config/tc-d10v.c: Added code to support 32-bit fixups for stabs.
4229 * config/tc-h8300.c (get_specific): New operand "size" derived
4230 from ".b", ".w" and ".l" extensions. All callers changed. If
4231 the base instruction has no operands, then use the size to
4232 determine which specific instruction to use.
4236 * config/tc-i960.c (mem_fmt): Call parse_expr before emit.
4240 * config/tc-mips.c (md_section_align): Don't change addr if
4245 * config/tc-hppa.c: Revert yesterday's changes.
4249 * config/tc-d10v.c: Disable range checking on 16-bit values.
4253 * config/tc-m68k.c (m68k_ip): Set ok_arch for every instruction,
4254 not just the ones that don't match.
4258 * config/tc-d10v.c: Fixed bugs in short relocs and range checking.
4262 * config/tc-arm.c: Changed INSN_SIZE to variable insn_size, as
4263 pre-cursor to adding Thumb support. Also added cpu_variant flag
4264 information to each of the asm_flg structures.
4265 (md_parse_option): Updated ARM7 parsing to allow 't' for
4266 thumb/halfword support, aswell as 'm' for long multiply.
4267 (md_show_usage): Updated help message.
4268 (md_assemble): Check that instruction flags are applicated to the
4269 current cpu variant.
4270 (md_apply_fix3, tc_gen_reloc): Add BFD_RELOC_ARM_OFFSET_IMM8 and
4271 BFD_RELOC_ARM_HWLITERAL relocation support for new halfword and
4272 signextension instructions.
4273 (do_ldst): Generate halfword and signextension variants if
4274 mnemonic flags match.
4275 (ldst_extend): Do not allow shifts in the offset field of halfword
4276 or signextension instructions.
4277 (validate_offset_imm): Provide check on halfword and signextension
4279 (add_to_lit_pool): Merge identical literal pool values.
4283 * config/tc-hppa.c (selector_table): Add 'E' selector.
4284 (cons_fix_new_hppa): Don't coke on e_esel.
4285 (tc_gen_reloc, SOM version): Handle R_COMP2 when used
4286 to help generate exception handling tables.
4287 (md_apply_fix): Don't try to apply fixups with an e_esel
4289 (hppa_fix_adjustable): Fixups with e_esel selectors
4294 * config/tc-sparc.c (md_pseudo_table): Add 2byte, 4byte, and 8byte
4299 * config/tc-d10v.c: Added lots of error checking. Added hacks
4300 to support accumulator shifts.
4304 * symbols.c (S_SET_EXTERNAL): Let .weak override.
4305 (S_CLEAR_EXTERNAL): Likewise.
4306 (S_SET_WEAK): Remove error; just let .weak override.
4310 * config/tc-d10v.c (md_assemble): Now handles multiline
4315 * config/tc-d10v.c: Fix packaging bug. Added range checking.
4316 Added kludge for divs instruction. Fixed minor problem with
4317 multiple text sections.
4318 * config/tc-d10v.h (d10v_cleanup): Change prototype.
4322 * config/tc-d10v.c (md_apply_fix3): Fix all instruction
4323 addresses to be right-shifted by 2.
4327 * config/tc-d10v.c: Many changes to get relocs working.
4328 (register_name): No longer creates a symbol for register names.
4329 (pre_defined_registers): moved to opcodes/d10v-opc.c.
4330 (d10v_insert_operand): Now works correctly for either container.
4331 * config/tc-d10v.h (d10v_cleanup): Declare.
4335 * config/tc-mips.c (tc_gen_reloc): BFD_RELOC_PCREL_HI16_S and
4336 BFD_RELOC_PCREL_LO16 are expected to be PC relative.
4340 * config/tc-alpha.c: Patches to track current minimum alignment to reduce
4341 the number of fragments created with frag_align.
4342 (alpha_current_align): New static variable.
4343 (s_alpha_text): Reset alignment to 0.
4344 (s_alpha_data, s_alpha_rdata, s_alpha_sdata): Likewise.
4345 (s_alpha_stringer, s_alpha_space): New functions.
4346 (s_alpha_cons, alpha_flush_pending_output): Remove functions.
4347 (alpha_cons_align): New function to replace both of them.
4348 (emit_insn): Only align if alpha_current_align is less than 2;
4349 reset alpha_current_align to 2.
4350 (s_alpha_gprel32): Likewise.
4351 (s_alpha_section): New function. Basically duplicate the other
4352 alpha section change hooks. Only define for ELF.
4353 (s_alpha_float_cons): Simplify alignment handling.
4354 (md_pseudo_table): Only define "rdata" and "sdata" if OBJ_ECOFF.
4355 If OBJ_ELF, define "section", "section.s", "sect", and "sect.s".
4356 Don't define the s_alpha_cons pseudo-ops. Do define
4357 s_alpha_stringer and s_alpha_space pseudo-ops.
4358 (alpha_align): Skip if less than current default alignment. Set
4360 * config/tc-alpha.h (md_flush_pending_output): Remove.
4361 (md_cons_align): Add.
4363 * config/tc-alpha.c: Add oodles of function description comments.
4364 (md_bignum_to_chars): Remove; there are no callers.
4365 (md_show_usage): Mention some more variants.
4370 * configure.in (sparc-*-sysv4*): New target.
4371 * configure: Rebuild.
4373 * config/tc-sparc.c (md_pseudo_table): Change uahalf, uaword, and
4374 uaxword to use s_uacons.
4375 (sparc_no_align_cons): New static variable.
4376 (s_uacons): New static function.
4377 (sparc_cons_align): If sparc_no_align_cons is set, just clear it
4380 * config/tc-sparc.c (s_common): Remove unused label allocate_bss.
4382 * configure.in: Add mips-*-irix6* target. Handle Irix 6 like Irix
4383 5 with regard to shared libraries.
4384 * configure: Rebuild.
4386 * config/tc-m68k.c (m68k_ip): Use the correct length when
4387 allocating space for the unsupported architecture error message.
4391 * configure.in (d10v-*-*): Allow d10v-*-*, don't require d10v-*-elf*.
4395 * config/tc-d10v.c: New file.
4396 * config/tc-d10v.h: New file.
4397 * configure (d10v-*-elf): New target.
4398 * configure.in (d10v-*-elf): New target.
4402 * config/tc-ppc.c (md_parse_option): Recognize -K PIC.
4406 * config/tc-alpha.c (alpha_align): Change fill parameter
4407 to a pointer. Take NULL as 0 or nop depending on section. Change
4409 (s_alpha_align): Rename local variables.
4411 * doc/as.texinfo (.align): Document action of omitted
4416 * config/tc-ppc.c (md_apply_fix3): Give a useful error message
4417 when an unsupported PC relative reloc is seen, rather than calling
4420 * app.c (do_scrub_chars): Remove not_cpp_line local variable.
4421 Instead, check state when '#' comment is seen.
4425 * config/tc-mips.c (mips_regmask_frag): Only define if OBJ_ELF or
4427 (tc_gen_reloc): If fixup was changed to be PC relative, change
4428 reloc type accordingly. Use name of reloc in error message.
4430 * as.h: Don't define const or volatile.
4431 * flonum.h: Don't define const.
4433 * config/tc-m68k.c (tc_gen_reloc): Change the code appropriately
4434 if fx_pcrel is set. Correct setting the addend case in the
4435 OBJ_ELF case (from Andreas Schwab
4437 (md_show_usage): Correct -mfc5200 to -m5200.
4441 * doc/c-m68k.texi: Document -m5200 flag.
4442 * doc/as.texinfo: Likewise.
4444 * config/tc-m68k.c (m68k_ip): The coldfire does not support 8x
4449 * symbols.c (S_SET_EXTERNAL): Change as_warn to as_bad.
4450 (S_CLEAR_EXTERNAL, S_SET_WEAK): Likewise.
4454 * Makefile.in (VERSION): Set to cygnus-2.7.1.
4456 * Released binutils 2.7.
4460 * config/tc-mips.c (mips_ip): Only perform range check when
4461 dealing with O_constant expressions.
4465 * m68k-parse.h (m68k_register): Add new coldfile control
4468 * config/tc-m68k.c (mcf5200_control_regs): New variable,
4469 array of control registers for the coldfire.
4470 (cpu_of_arch): Added mcf5200.
4471 (archs): Added mcf5200.
4472 (init_table): Add new control registers.
4473 (m68k_ip): Added support for new control registers.
4474 (m68k_init_after_args): Likewise.
4476 * config/tc-m68k.c (md_show_usage): Add -m5200 to usage text.
4480 * read.h (is_it_end_of_statement): Declare.
4481 * read.c (is_it_end_of_statement): Remove declaration.
4483 * config/tc-ppc.c (ppc_elf_suffix): Correct parenthesization of ||
4485 (md_assemble): Fix handling of @l with an unsigned constant. Add
4486 default case to reloc switch.
4488 * config/tc-i386.h (AOUT_MACHTYPE): Define as 0 if TE_386BSD.
4491 * ecoff.c (last_lineno): New static variable.
4492 (add_procedure): Set last_lineno.
4493 (ecoff_directive_loc): Likewise.
4494 (ecoff_generate_asm_lineno): Likewise.
4495 (ecoff_fix_loc): New function.
4496 * ecoff.h (ecoff_fix_loc): Declare.
4497 * config/tc-mips.c (append_insn): When inserting nops, and using
4498 ECOFF debugging, call ecoff_fix_loc.
4502 * config/tc-h8300.c (build_bytes): If an operand type is
4503 marked as SRC_IN_DST retrieve it from the "destination" op.
4507 * configure.in (arm-*-riscix*): Set emulation to riscix.
4508 * configure: Rebuild.
4509 * config/te-riscix.h: New file to define TE_RISCIX.
4511 * config/tc-sh.h (SUB_SEGMENT_ALIGN): Define.
4515 * Makefile.in (config.status): Just run config.status as other
4520 * mpw-config.in (TARGET_OS): Add definition to conf.
4524 * config/tc-mips.c (append_insn): Parenthesize
4525 cop_interlocks expressions.
4529 * listing.c (listing_print): Close the listing file if it is not
4530 stdout. Close the other files opened for the listing.
4532 * config/tc-sparc.h (md_cons_align): Define.
4533 (sparc_cons_align): Declare.
4534 (HANDLE_ALIGN): Define.
4535 (sparc_handle_align): Declare.
4536 * config/tc-sparc.c (sparc_cons_align): New function.
4537 (sparc_handle_align): New function.
4538 * read.c (cons_worker): Call md_cons_align if it is defined.
4540 * as.h (struct frag): Add fr_file and fr_line fields.
4541 * frags.c (frag_new): Set fr_file and fr_line.
4542 (frag_var): Likewise.
4543 (frag_variant): Likewise.
4545 * as.h (struct frag): Remove unused align_mask and align_offset
4548 * listing.c (calc_hex): Offset by fr_fix when examining fr_var.
4553 * configure.in (mips-*-osf*): New target.
4554 * configure: Rebuild.
4556 * config/tc-m68k.c: Add 68ec060 as a synonym for 68060.
4560 * config/tc-mips.c: Added cop_interlocks, to avoid NOP insertion
4561 between co-processor comparisons and branches for the VR4300.
4565 * Makefile.in (bindir, libdir, datadir, mandir, infodir, includedir,
4566 INSTALL_PROGRAM, INSTALL_DATA): Use autoconf-set values.
4568 * configure.in (AC_PREREQ): autoconf 2.5 or higher.
4569 * doc/Makefile.in (bindir, libdir, datadir, mandir, infodir,
4570 includedir): Use autoconf set values.
4575 * listing.c (listing_eject): Don't do anything if listing is 0.
4576 (listing_list): Likewise.
4577 (listing_source_line): Likewise.
4578 (listing_title): Don't save title if listing is 0.
4579 (listing_source_file): Check listing rather than listing_tail.
4581 * configure.in: On alpha*-*-osf*, link against libbfd.a if not
4582 using shared libraries.
4583 * configure: Rebuild.
4587 * config/tc-mips.c (mips_ip): In case 'i'/'j', don't require an
4588 absolute expression if a relocation type was specified.
4592 * configure.in: Add support for *-*-rtems* configurations.
4593 * configure: Rebuild.
4597 * configure.in: Add alpha-*-linuxecoff* target. Use elf for
4598 alpha-*-linux* target. Force bfd_gas for alpha-*. Require
4599 opcodes library for alpha.
4600 * configure: Rebuild with autoconf 2.10.
4601 * config/tc-alpha.c: Substantial rewrite to add ELF support and
4602 use new opcode table.
4603 * config/tc-alpha.h (md_undefined_symbol): Don't define.
4604 (LOCAL_LABEL): Define differently if OBJ_ELF.
4605 (FAKE_LABEL_NAME): Define if OBJ_ELF.
4606 * config/alpha-opcode.h: Remove.
4607 * config/obj-elf.h: If TC_ALPHA, define ECOFF_DEBUGGING.
4608 * Makefile.in (TARG_CPU_DEP_alpha): Depend upon
4609 include/opcode/alpha.h rather than config/alpha-opcode.h.
4613 * config/obj-aout.c (obj_emit_relocations): Give an error if the
4614 relocation symbol was not resolved.
4615 * config/obj-coff.c (do_relocs_for): Likewise.
4617 * write.c (adjust_reloc_syms): Refetch the symbol section after
4618 calling S_GET_VALUE, since it may have changed.
4620 * expr.c (struct expr_symbol_line): Define.
4621 (expr_symbol_lines): New static variable.
4622 (make_expr_symbol): Add entry to expr_symbol_lines.
4623 (expr_symbol_where): New function.
4624 * expr.h: Use extern on function declarations.
4625 (expr_symbol_where): Declare.
4626 * symbols.c (resolve_symbol_value): Try to use expr_symbol_where
4627 rather than printing the meaningless name of an expression
4632 * config/tc-i386.c (md_number_to_chars): Deleted.
4633 * config/tc-i386.h (md_number_to_chars): New macro.
4635 * config/tc-alpha.c (build_operate_n, build_mem): Moved earlier in
4637 (load_symbol_address, load_expression): Use build_mem.
4638 (build_operate): New function.
4639 (emit_addq_r): Use it.
4643 * symbols.c (colon): #if VMS, use S_SET_OTHER to store `const_flag'.
4647 * config/tc-vax.h (NOP_OPCODE): Define.
4651 * config/obj-vms.h (S_IS_COMMON): Define.
4652 (S_IS_LOCAL): Check for \002 as well as \001.
4653 (LONGWORD_ALIGNMENT): New macro.
4654 (SUB_SEGMENT_ALIGN): Use it.
4658 * config/vms-conf.h: Reconcile with conf.in.
4662 * write.c (is_dnrange): Only define if TC_GENERIC_RELAX_TABLE is
4665 * doc/as.texinfo: Document that any number of hex digits can
4668 * as.c (struct defsym_list): Define.
4669 (defsyms): New static variable.
4670 (parse_args): Just put --defsym arguments on defsyms list, rather
4672 (main): Define defsyms after output file is created.
4674 * config/tc-m68k.c (m68k_ip): Reject PRE and POST indexing mode on
4677 * config/tc-mips.c (mips_ip): In cases 'I', 'i', and 'j', set
4678 insn_error rather than calling check_absolute_expr.
4680 * as.c (emulation_name): Remove unused static variable.
4681 (default_emul_bfd_name): Add return NULL to avoid warning.
4682 * ecoff.c (ecoff_stab): Remove unused variables name and
4684 * frags.c (frag_new): Remove unused variable tmp.
4685 * hash.c (hash_grow): Parenthesize + within <<.
4686 (hash_print_statistics): Use %lu, not %d, to print unsigned
4688 * messages.c: Include "libiberty.h".
4689 (fprint_value): Add cast to avoid printf warning.
4690 (sprint_value): Likewise.
4691 * read.c: Include "ecoff.h".
4692 (emit_expr): Add casts to avoid printf warnings.
4693 * read.h: Use extern for function declarations.
4694 (pop_insert): Declare.
4695 * stabs.c: Include "ecoff.h".
4696 * subsegs.c (subseg_set_rest): Remove unused variables tmp,
4697 former_last_fragP, and new_fragP.
4698 * subsegs.h (subsegs_print_statistics): Declare.
4699 * symbols.c (debug_verify_symchain): Change macro to discard
4701 * write.c (dump_section_relocs): Likewise.
4702 * write.h: Use extern for function declarations.
4703 (write_print_statistics): Declare.
4704 * config/e-mipsecoff.c (mipsecoff_bfd_name): Return NULL to avoid
4706 * config/e-mipself.c (mipself_bfd_name): Likewise.
4707 * config/obj-elf.h (elf_ecoff_set_ext): Declare.
4709 * config/tc-sparc.h (TC_RELOC_RTSYM_LOC_FIXUP): If OBJ_ELF, always
4710 emit relocations against external symbols.
4712 * config/tc-alpha.c (tc_gen_reloc): Output a sensible error
4713 message if bfd_reloc_type_lookup fails, rather than calling
4716 * config/tc-alpha.c (alpha_force_relocation): Add
4717 BFD_RELOC_12_PCREL to switch.
4721 * config/tc-i386.h (LOCAL_LABEL,FAKE_LABEL_NAME): Use defaults for
4722 TE_PE (Lfoo, not .Lfoo).
4726 * read.c (s_fill): Don't warn about a zero repeat count.
4728 * config/tc-mips.c (mips_ip): Don't warn about using AT as a
4729 coprocessor register.
4731 * config/tc-i386.c (md_assemble): When checking the size of a
4732 register to set the size of an instruction, do a bitwise and with
4733 Reg8 and Reg16 rather than requiring the type to be exactly Reg8
4738 * config/tc-h8300.c (parse_reg): Tweak error messages.
4739 (build_bytes): Likewise.
4740 (skip_colonthing): Handle :32 suffix.
4741 (get_specific): Promote L_24 to L_32 if it makes a match.
4742 Don't always promote L_8 to L_16.
4743 (do_a_fix_imm): Clean up L_32 and L_24 handling.
4745 * config/tc-h8300.c (Smode): New variable.
4746 (h8300hmode): Turn off Hmode.
4747 (h8300smode): New function. Turn on Smode and Hmode.
4748 (md_pseudo_table): New ".h8300s" pseudo-op.
4749 (parse_reg): Handle "exr" register.
4750 (get_operand): Handle bizarre syntax for "stm.l" and "ldm.l".
4751 Handle "mach" and "machl" operands for ldmac.
4752 (get_specific): Handle "stm.l" and "ldm.l".
4753 (build_bytes): Handle "stm.l" and "ldm.l"; handle MACREG operands.
4754 * config/tc-h8300.h (COFF_MAGIC): Handle H8/S magic number.
4759 * doc/as.texinfo: Reorder chapter of machine dependent options so
4760 that it is sorted by chip name.
4762 * doc/as.texinfo: Use consistant spelling of Vax.
4763 * doc/c-vax.texi: Likewise.
4767 * config/tc-hppa.c (md_pseudo_table): Add ".begin_try" and ".end_try"
4769 (tc_gen_reloc, SOM version): Handle R_BEGIN_TRY and R_END_TRY.
4770 (md_apply_fix): Likewise.
4771 (pa_try): New function.
4772 (hppa_force_relocation): Force relocs for BEGIN_TRY and END_TRY.
4776 * config/tc-hppa.c (md_pseudo_table): Add ".level" pseudo op.
4777 (pa_level): New function.
4781 * listing.c (listing_newline): Don't do anything if listing is 0.
4785 * subsegs.c (section_symbol): If symbol_table_frozen is set, call
4786 symbol_create, not symbol_new.
4790 * write.c (adjust_reloc_syms): Don't set sy_used_in_reloc for an
4791 absolute symbol unless TC_FORCE_RELOCATION returns true.
4793 * config/obj-coff.c (previous_file_symbol): Remove BFD_ASSEMBLER
4795 (c_dot_file_symbol): BFD_ASSEMBLER version: Don't set the value of
4796 the symbol to a pointer. Don't set previous_file_symbol.
4797 Simplify symbol list rearrangement.
4798 (coff_frob_symbol): Don't do anything with C_FILE symbols.
4799 (coff_adjust_symtab): Don't check previous_file_symbol.
4803 * config/tc-ppc.c (ppc_elf_lcomm): New function for .lcomm
4805 (md_pseudo_table): Add ppc_elf_lcomm.
4809 * config/tc-m68k.c (m68k_ip): Accept ABSL for 'O', so that `bfextu
4810 d0{24:1},d0' works without an immediate prefix on the bit numbers.
4811 (md_begin): Add digits to alt_notend_table.
4812 (md_parse_option): Make s a const pointer.
4814 * config/tc-sparc.c (md_pseudo_table): Add "empty".
4815 (s_empty): New static function.
4817 * config/obj-coff.c (struct filename_list): Only define if not
4819 (filename_list_head, filename_list_tail): Likewise.
4820 (c_section_symbol): Remove unused BFD_ASSEMBLER version.
4821 (obj_coff_endef, BFD_ASSEMBLER version): Don't set the debugging
4822 flag for C_MOS, C_MOE, C_MOU, or C_EOS symbols, since they should
4823 have a section of N_ABS rather than N_DEBUG. If we do a merge,
4824 remove the new symbol from the list.
4825 (obj_coff_endef, both versions): Call tag_insert even if there is
4826 an old symbol with the same name, if the old symbol does not
4828 (coff_frob_symbol): Check SF_GET_TAG, C_EOF, and C_FILE outside of
4829 the SF_GET_DEBUG condition. Don't call SA_SET_SYM_ENDNDX with a
4830 symbol that will be moved to the end of the symbol list.
4831 (coff_adjust_section_syms): Always call section_symbol for .text,
4833 (coff_frob_section): Likewise. Also, remove unused variable
4836 * config/tc-ns32k.c (convert_iif): Call frag_grow rather than
4837 manipulating frags directly.
4838 (md_number_to_field): Adjust mem_ptr correctly if ENDIAN is
4841 * app.c (do_scrub_chars): If '/' is LINE_COMMENT_START, check
4842 whether the next character is '*' before checking whether we are
4843 at the start of a line. Permit LINE_COMMENT_START to start a
4844 comment in state 1 (seen some whitespace) as well, to match the
4847 * gasp.c (do_align): Permit a fill value for .align.
4851 * read.c (next_char_of_string): Warn if a newline is seen in the
4852 middle of a string. Call bump_line_counters when appropriate.
4856 * symbols.c (colon): Use LOCAL_LABEL.
4860 * Makefile.in (install): Don't check to see if tooldir exists.
4861 Make $(tooldir) and $(tooldir)/bin.
4865 * config/ppc-sol.mt (TDEFINES): Don't turn on -mregnames by
4870 * config/tc-mips.c (mips_ip): Don't call as_warn if we are setting
4871 insn_error. Don't put the string "ERROR" in insn_error. Set
4872 insn_error rather than calling as_warn for an unsupported opcode.
4876 * config/tc-mips.c (md_parse_option): Check for a 64 bit format
4877 before permitting -64.
4878 * output-file.c (output_file_create): Remove duplicate
4883 * config/tc-mips.c (md_begin): If -64, create a .MIPS.options
4884 section rather than a .reginfo section.
4885 (mips_elf_final_processing): If -64, write out 64 bit RegInfo
4888 * config/tc-mips.c (load_register): If mips_isa < 3, permit a 32
4889 bit value with the high bit set.
4893 * read.c (s_lcomm): Set section flags for .sbss section.
4895 * config/tc-mips.c (mips_64): New static variable.
4896 (mips_target_format): If mips_64, return elf64 targets rather than
4898 (md_longopts): Add "32" and "64".
4899 (md_parse_option): Handle -32 and -64.
4900 (md_show_usage): Mention -32 and -64.
4901 (cons_fix_new_mips): If mips_64, don't convert an 8 byte reloc to
4906 * config/tc-ppc.c (comment_chars): Make '!' a comment character
4907 for Solaris compatibility.
4909 * stabs.c (s_stab_generic): Under PowerPC Solaris, convert a
4910 .stabd with 4 arguments into a .stabn.
4914 * config/tc-mips.c (macro): When passing X_add_number to
4915 macro_build, cast it to int first.
4919 * config/tc-z8k.c (md_apply_fix): Handle fx_r_type of 0, as
4920 created by emit_expr.
4922 * symbols.c (symbol_create): If bfd_make_empty_symbol fails, call
4923 as_perror rather than assert.
4927 * config/tc-mips.c (mips_ip): Mark sections created to hold
4928 floating point information as read only.
4932 * config/tc-ppc.c (ppc_set_cpu): Change defaults to match AIX.
4936 * read.c (potable): Add .skip as a synonym for .space.
4938 * stabs.c (s_stab_generic): For PowerPC ELF, allow .stabd to take
4939 4 arguments, providing the 4th argument is 0, to allow
4940 compatibility with the Solaris assembler.
4944 * config/tc-sh.h (struct sh_segment_info_type): Define.
4945 (TC_SEGMENT_INFO_TYPE): Define.
4946 (sh_frob_label): Declare.
4947 (tc_frob_label): Define.
4948 (sh_flush_pending_output): Declare.
4949 (md_flush_pending_output): Define.
4950 * config/tc-sh.c (md_assemble): If relaxing, emit a R_SH_CODE
4951 reloc before the instruction if necessary.
4952 (sh_frob_label): New function.
4953 (sh_flush_pending_output): New function.
4954 (sh_coff_frob_file): Ignore ALIGN, CODE, DATA, and LABEL relocs
4955 when looking for the reloc for the target of .uses.
4956 (md_convert_frag): Fix printf format (%0xlx to 0x%lx).
4957 (sh_force_relocation): Force CODE, DATA, and LABEL relocs to be
4959 (md_apply_fix): Ignore CODE, DATA, and LABEL relocs.
4960 (sh_coff_reloc_mangle): Force CODE, DATA, and LABEL relocs to use
4961 the absolute symbol.
4963 * subsegs.h (segment_info_type): Add tc_segment_info_data field if
4964 TC_SEGMENT_INFO_TYPE is defined.
4968 * config/tc-i386.c (md_assemble): Make sure the opcode suffix
4969 matches the register size.
4973 * config/obj-coff.c (count_entries_in_chain): Ignore Fixups with
4975 (do_relocs_for): Likewise.
4976 (fixup_segment): Don't just quit if linkrelax is set. Try to
4977 apply non pc-relative sym1-sym2 fixups, even if linkrelax is
4982 * config/tc-ppc.c (ppc_elf_validate_fix): Allow GOT and section
4983 relative relocations with -mrelocatable. Also allow unfixed
4984 relocs in .ex_shared.
4988 * config/obj-coff.c (yank_symbols): Check that FNAME_OFFSET is
4989 non-zero before assuming this is a long file name.
4990 (w_strings): Likewise.
4991 (c_dot_file_symbol): Set FNAME_OFFSET to 1 for a long file name.
4993 * config/obj-coff.c (w_strings): Move declaration of i inside
4994 #ifdef block which uses it.
4998 * config/tc-h8300.c (do_a_fix_imm): Rename last argument to
4999 "relaxmode". Output relocs which identify various relaxing
5000 possibilities for mov.[bwl] instructions.
5001 (build_bytes): Pass in a relaxing mode to do_a_fix_imm.
5005 * config/tc-arm.h (TC_HANDLES_FX_DONE): Define.
5006 (MD_APPLY_FIX3): Define.
5007 * config/tc-arm.c (my_get_expression): Only watch for bad segments
5009 (md_apply_fix3): Renamed from md_apply_fix.
5010 If pcrel reloc and symbol is in different section, undo effects
5015 * config/tc-hppa.c (hppa_fix_adjustable): Don't adjust
5016 any reloc with an LR% or RR% field selector for SOM.
5020 * Makefile.in: Add subsegs.h to appropriate TARG_CPU_DEP_*
5025 * config/obj-coff.c (coff_frob_symbol): Don't merge a symbol with
5027 (yank_symbols): Likewise.
5031 * subsegs.h (segment_info_type): If MANY_SEGMENTS and not
5032 BFD_ASSEMBLER, add name field.
5033 * config/obj-coff.c: Include "libiberty.h".
5034 (coff_header_append): Handle long section names.
5035 (crawl_symbols): Just use the name field for the symbol name,
5036 without worrying about null byte termination.
5037 (w_strings): Handle long section names.
5038 (write_object_file): Likewise. Also, use the name field, rather
5040 (obj_coff_add_segment): Permit long section names.
5041 (obj_coff_init_stab_section): Use the name field, rather than
5043 (adjust_stab_section): Likewise.
5044 * config/te-pe.h (COFF_LONG_SECTION_NAMES): Define.
5046 * config/tc-i960.c (brtab_emit): Don't set fx_im_disp field.
5047 (mem_fmt): Likewise.
5048 (md_apply_fix): Don't check fx_im_disp field.
5052 * configure.in: Add * after sparc*-*-vxworks.
5053 * configure: Rebuild.
5055 * app.c (do_scrub_begin): If tc_comment_chars is not defined,
5056 define it to comment_chars. Use tc_comment_chars rather than
5058 (do_scrub_chars): Use tc_comment_chars rather than comment_chars.
5059 * config/tc-m68k.h (tc_comment_chars): Define.
5060 (m68k_comment_chars): Declare.
5061 * config/tc-m68k.c (m68k_comment_chars): Rename from
5062 comment_chars. Change into a pointer rather than an array.
5063 (md_longopts): Add "bitwise-or".
5064 (md_parse_option): Handle OPTION_BITWISE_OR.
5065 (md_show_usage): Mention --bitwise-or.
5066 * doc/c-m68k.texi: Document --bitwise-or.
5070 * config/tc-m68k.c (m68k_ip): Prevent attempts to use long offsets
5073 * config/obj-coff.c (obj_coff_section): BFD_ASSEMBLER version:
5074 call demand_empty_rest_of_line. Non BFD_ASSEMBLER version:
5075 correct handling of input line pointer, and call
5076 demand_empty_rest_of_line.
5080 * config/tc-sparc.c (in_bitfield_range): New static function.
5081 (sparc_ip): New cases X,Y. Use SPARC_OPCODE_ARCH_V9_P.
5082 (md_apply_fix, cases BFD_RELOC_32_PCREL_S2,
5083 BFD_RELOC_SPARC_{WDISP16,WDISP19}): Fix undefined code.
5084 (md_apply_fix): New cases BFD_RELOC_SPARC_[56].
5085 (tc_gen_reloc): New cases BFD_RELOC_SPARC_[56].
5089 * config/obj-coff.c: BFD_ASSEMBLER:
5090 (coff_last_bf): New static variable.
5091 (coff_frob_symbol): Set endndx of a .bf symbol.
5093 (obj_coff_endef): Call SF_SET_PROCESS on a .bf symbol.
5094 (last_bfP): New static variable.
5095 (yank_symbols): Set endndx of a .bf symbol.
5099 * config/tc-ppc.c (md_parse_option): Add support for Solaris's -le
5100 and -s options. Add -be for good measure.
5104 * read.c (s_space): Support non-constant fill value. Handle fill
5105 value correctly for a size other than 1.
5109 * config/tc-arm.c (my_get_float_expression): Update call to
5110 gen_to_words, X_PRECISION changed from 6 to 5.
5114 * config/tc-ppc.c (register_name,reg_name_search): Move register
5115 name lookup from PE specific code to all targets. Add support for
5116 -mregnames/-mno-regnames to control whether register names are
5118 (md_assemble): Call register_name for all platforms.
5119 (md_parse_option): Add support for -mregnames/-mno-regnames.
5121 * configure.in (powerpcle*-*-solaris): Add support.
5122 (powerpc*-*-linux): Ditto.
5123 * configure: Regenerate.
5125 * config/ppc-sol.mt: New config file for PowerPC Solaris.
5129 * config/tc-mips.c (mips_frob_file): Permit multiple %hi relocs to
5130 be associated with a single %lo reloc.
5132 * config/tc-mips.c (load_address): Cast X_add_number to valueT
5133 before comparing against MAX_GPREL_OFFSET, so that negative
5134 numbers are handled correctly.
5139 * config/tc-sparc.c (last_insn): New static variable.
5140 (md_assemble): Warn about putting floating point branches in a
5141 delay slot. If architecture is less than v9, insert NOP
5142 instructions between floating point instructions and floating
5143 point branches. (The SunOS assembler does both these operations.)
5144 Save the last instruction opcode.
5145 (sparc_ip): Add pinsn parameter. Change caller.
5147 * config/tc-m68k.c (md_estimate_size_before_relax): Correct check
5148 for byte jump to next instruction to skip empty frags.
5152 * config/tc-alpha.c (alpha_ip): If we are going to call emit_add64
5153 for addq with a 16 bit signed value, just emit a lda instruction
5158 * config/tc-h8300.c (do_a_fix_imm): Don't cut off high bits
5163 * configure.in: Permit --enable-shared to specify a list of
5165 * configure: Rebuild.
5169 * config/tc-h8300.c (get_specific): Remove some #if 0 code.
5170 (build_bytes): Remove all ABSMOV related code; it's unnecessary.
5174 * config/atof-ieee.c: Fix handling of denormalized extended
5175 precision numbers and overflow/underflow detection.
5176 (MAX_PRECISION, X_PRECISION, P_PRECISION): Changed from 6 to 5, to
5177 not include the 16 bit gap in the m68k extended precision format.
5181 * configure.in: Add i386-*-freebsdelf* target; from John Polstra
5183 * configure: Rebuild.
5187 * config/tc-mips.c: Allow non-zero offsets from .sdata symbols to
5188 be accessed using the $gp register.
5189 * config/tc-mips.h (MAX_GPREL_OFFSET): Added.
5193 * config/tc-sparc.c (sparc_md_end): Set bfd machine number to
5194 bfd_mach_sparc_sparclet if current_architecture is sparclet.
5198 * read.c (get_line_sb): Bump line counters based on
5199 input_line_pointer[-1], not *input_line_pointer. Don't bother to
5200 call LISTING_NEWLINE.
5201 (s_macro): Don't call demand_empty_rest_of_line.
5202 * app.c (do_scrub_chars): When handling C style comments, unget
5207 * read.h (enum linkonce_type): Define.
5208 (s_linkonce): Declare.
5209 * read.c (potable): Add "linkonce".
5210 (s_linkonce): New function.
5211 * subsegs.h (segment_info_type): Add linkonce field to
5212 MANY_SEGMENTS && ! BFD_ASSEMBLER section.
5213 * config/obj-coff.h (obj_handle_link_once): Define if TE_PE.
5214 (obj_coff_pe_handle_link_once): Declare if TE_PE.
5215 * config/obj-coff.c: If TE_PE and not BFD_ASSEMBLER, #include
5217 (obj_coff_pe_handle_link_once): New function, defined if TE_PE.
5218 (c_section_symbol): If TE_PE, set the x_comdat field in the aux
5219 entry based on the linkonce field in segment_info.
5220 * doc/as.texinfo: Document .linkonce.
5224 * doc/as.1: Changed to be recognized by catman -w on Solaris.
5228 * stabs.c (s_stab_generic): Call the listing functions before
5229 doing the rest of the processing, which may involve freeing the
5230 string. Pass string, not string + stroff, to OBJ_PROCESS_STAB in
5231 SEPARATE_STAB_SECTIONS case.
5233 * config/tc-hppa.c: Remove nested comment.
5234 (tc_gen_reloc): Move label done inside the ifdef in which it is
5236 (md_apply_fix): Pass pointers to correct types to libhppa.h
5237 functions. Always return a value.
5239 * config/tc-mips.h (tc_frob_file): Define.
5240 (mips_frob_file): Declare.
5241 * config/tc-mips.c (struct mips_hi_fixup): Define.
5242 (mips_hi_fixup_list): New static variable.
5243 (imm_unmatched_hi): New static variable.
5244 (md_assemble): Clear imm_reloc, imm_unmatched_hi, and
5245 offset_reloc. Pass imm_unmatched_hi to append_insn.
5246 (append_insn): Add unmatched_hi parameter. If it is set, add the
5247 new fixup to mips_hi_fixup_list. Change all callers.
5248 (mips_ip): Set imm_unmatched_hi when appropriate.
5249 (mips_frob_file): New function.
5253 * configure.in (sparc-*-solaris2*): Renamed from sparc*-*-solaris2*.
5254 * configure: Regenerated.
5258 * as.c (main): Call bfd_set_error_program_name.
5262 * as.h (strdup): Don't declare.
5263 * stabs.c: Include libiberty.h
5264 (get_stab_string_offset): Use xstrdup rather than strdup.
5265 (s_stab_generic): Likewise.
5266 * as.c (parse_args): Likewise.
5267 * read.c (s_mri_sect): Likewise.
5269 * gasp.c (change_base): Recognize \(...) construct documented to
5270 pass through enclosed characters literally through to the output.
5271 (process_assigns): Likewise. Also, be more careful to avoid
5272 looking past the end of the buffer.
5276 * config/tc-i386.c (md_parse_option): If OBJ_ELF, ignore -k for
5281 * doc/as.texinfo, doc/c-i960.texi: Fix typos.
5285 * config/alpha-opcode.h: Added cvtst instruction.
5289 * ecoff.c (ecoff_stab): Don't try to make a symbol out of the stab
5290 string. Extract the addend from the result of expression.
5294 * app.c (do_scrub_chars): If whitespace is seen in state 11, and
5295 LABELS_WITHOUT_COLONS is not defined, and we are not in m68k MRI
5296 mode, change the state to 3 rather than 1.
5300 * config/obj-coff.h (C_REGISTER_SECTION): Change from 20 to 50, to
5301 correspond to 11 March change.
5305 * config/tc-h8300.c (build_bytes, MEMIND case): Generate
5306 an R_MEM_INDIRECT reloc rather than R_RELBYTE.
5310 * configure: Rebuild with autoconf 2.8.
5314 * config/atof-ieee.c (gen_to_words): Improve handling of
5315 X_PRECISION numbers. Based on patches from Andreas Schwab
5320 * as.h (SEG_NORMAL, SEG_LIST): Bump segment limit from 10 to 40.
5322 * subsegs.c (MANY_SEGMENTS): Increase segment limit.
5323 * config/obj-coff.c (seg_N_TYPE, seg_info_off_by_4): Likewise.
5324 (do_relocs_for, w_symbols, obj_coff_add_segment, do_linenos_for,
5325 crawl_symbols, coff_header_append): Loop to SEG_LAST rather than
5330 * config/tc-sparc.c (sparc_ip): Handle operand char 'O' (neg reg).
5334 * config/tc-hppa.c (SUBSEG_MILLI): Define.
5335 (pa_def_subspaces): Add $MILLICODE$.
5336 (pa_spaces_begin): Set section flags for $MILLICODE$.
5340 * config/obj-elf.c (obj_elf_section): Only SEC_LOAD if the type is
5341 not SHT_NOBITS. Don't tamper with flags based on type if a
5342 special section was found (revert Feb 29 change).
5344 * config/tc-sh.c (sh_do_align): Only align using the nop pattern
5345 if aligning to a longword boundary or greater.
5349 * config/tc-sh.c (sh_do_align): Pass 1 not 2 to frag_align.
5353 * configure.in (i386-*-cygwin32): Don't use bfd_gas.
5354 * configure: Regenerated.
5358 * config/tc-hppa.c: Add default definitions for R_N0SEL and
5359 R_N1SEL since they're not defined for old versions of hpux.
5361 * config/tc-hppa.c (tc_gen_reloc): Fix typo in R_COMP2 code.
5362 Set "sym_ptr_ptr" and "addend" fields to dummy values for
5363 R_N0SEL and R_N1SEL.
5367 * flonum-konst.c: Add two more constants for 1e+-2048 and
5368 1e+-4096, and correct the other constants.
5370 * symbols.c (resolve_symbol_value): Handle O_logical_not.
5374 * config/obj-elf.c (obj_elf_section): Allow predefined section
5375 types to set the nobits type. Avoid a shadowed declaration.
5379 * config/tc-hppa.c (hppa_fix_adjustable): For SOM, don't
5380 reduce relocs using e_nlrsel field selectors.
5382 * write.c (fix_new_exp): Don't use #elif. Some compilers
5385 * config/tc-hppa.c (selector_table): Add "n", "nl", and "nlr" to
5387 (pa_chk_field_selector): Handle new field selectors for SOM.
5391 * configure.in (m68k-*-linux*aout*, m68k-*-linux*): New targets.
5392 * configure: Rebuild.
5393 * config/te-linux.h (LOCAL_LABELS_FB): Define.
5394 * config/tc-m68k.h (TARGET_FORMAT) [TE_LINUX]: Define to
5396 * config/tc-m68k.c (comment_chars): Don't include '#' if TE_LINUX
5401 * mpw-make.sed: Update to handle shared library support.
5405 * config/tc-sparc.c (sparc_ip): Print all architectures that support
5406 the insn on mismatch.
5410 * configure.in: Add support for a29-coff.
5411 * configure: Rebuild.
5415 * config/tc-sh.c (sh_coff_frob_file): Don't consider the address
5416 of the section when looking for the R_SH_USES fixup, because the
5417 frag addresses have not yet been adjusted.
5419 * gdbinit.in: Set a breakpoint on as_warn_where.
5421 * config/tc-mips.c (macro): Add missing arguments to macro_build
5426 * config/tc-h8300.c (tc_reloc_mangle): Change reloc based on size
5427 if it is TC_CONS_RELOC. Set a size of 4 to R_RELLONG.
5431 * config/tc-sparc.c (sparc_ip): Recognize %asr0 for v8.
5435 * config/tc-sparc.c (parse_keyword_arg): Accept leading '%'.
5436 (sparc_ip): Accept %asr[1..31] for v8 and %asr[%16..31] for v9.
5437 Recognize [uU] format args as sparclet cpregs.
5441 * config/tc-sh.c (sh_handle_align): Don't emit R_SH_ALIGN relocs
5446 * config/tc-sparc.h (TC_RELOC_RTSYM_LOC_FIXUP): Check S_IS_WEAK as
5447 well as S_IS_EXTERNAL.
5448 (tc_fix_adjustable): Likewise.
5449 * config/tc-sparc.c (md_apply_fix): In OBJ_ELF case, check for
5450 S_IS_WEAK as well as S_IS_EXTERNAL when deciding whether to return
5452 (tc_gen_reloc): Check S_IS_WEAK as wel as S_IS_EXTERNAL when
5453 deciding whether to convert BFD_RELOC_32_PCREL_S2 if PIC.
5457 * config/tc-sparc.c (max_architecture): Change to sparclite for
5459 (default_compatible): Delete.
5460 (sparc_ffs): New function.
5461 (md_begin): Only call SPARC_OPCODE_CONFLICT_P once.
5462 (sparc_ip): Rewrite architecture match and bump logic.
5466 * configure.in: Check for 'do not mix' from native linker before
5467 trying to use -rpath.
5468 * configure: Rebuild.
5472 * config/obj-coff.h (SF_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
5473 (SF_GET_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
5474 (SF_SET_ADJ_LNNOPTR): Define (non BFD_ASSEMBLER).
5475 * config/obj-coff.c (obj_coff_endef): Set ADJ_LNNOPTR when LNNOPTR
5477 (w_symbols): If ADJ_LNNOPTR is set, add the section lnnoptr field
5478 to the symbol lnnoptr field, to get the correct file offset.
5482 * config/obj-elf.c (elf_frob_symbol): On the PowerPC, force all
5483 symbols that are not function, file, or section symbols to be
5488 * configure: Set and substitute RPATH_ENVVAR.
5489 * configure: Rebuild.
5490 * Makefile.in (RPATH_ENVVAR): New variable.
5491 (check): Use $(RPATH_ENVVAR) rather than LD_LIBRARY_PATH.
5494 doesn't need opcodes. If configuring shared, opcodes needs bfd.
5495 * configure: Rebuild.
5499 * read.c (s_mri_sect): Don't return '\0' in type. Set all
5500 appropriate flags in BFD section.
5502 * configure.in (m68k-*-psos*): New target.
5503 * configure: Rebuild.
5504 * config/te-psos.h: New file.
5505 * config/tc-m68k.c (comment_chars): Don't include '#' if TE_PSOS
5511 * configure.in: Remove duplicate setting of cpu_type. Check
5512 whether opcodes library is required for on all targets, not just
5514 * configure: Rebuild.
5516 * config/tc-mips.c (mips_big_got): New static variable.
5517 (s_extern): Don't declare.
5518 (reg_needs_delay): New static function.
5519 (macro_build): Permit GOT/CALL_HI/LO relocs.
5520 (macro_build_lui): If place is not NULL, use the number in the
5522 (load_address): Handle mips_big_got case.
5523 (macro): Handle mips_big_got for M_LA_AB, M_JAL_A, and load and
5525 (OPTION_XGOT): Define.
5526 (md_longopts): Add "xgot" if OBJ_ELF.
5527 (md_parse_option): Handle -xgot.
5528 (md_show_usage): Mention -xgot.
5529 (md_apply_fix): Permit GOT/CALL_HI/LO relocs.
5530 (tc_gen_reloc): Handle GOT/CALL_HI/LO relocs.
5534 * config/tc-m68k.c (m68k_ip) [operand kind '#']: When fixing
5535 the byte relocation, point it to the low byte of the word.
5539 * configure.in: Set HDLFLAGS for *-*-hpux with --enable-shared.
5540 * configure: Rebuild.
5544 * configure.in: Recognize any sparc* cpu.
5545 * configure: Regenerated.
5549 * read.c (potable): Add "mri" and ".mri".
5550 (s_mri): New function.
5551 * read.h (s_mri): Declare.
5552 * app.c (scrub_m68k_mri): New static variable.
5553 (mri_pseudo): New static variable.
5554 (do_scrub_begin): Add m68k_mri parameter. Use it rather than
5555 flag_m68k_mri. Initialize scrub_m68k_mri.
5556 (mri_state, mri_last_ch): New static variables.
5557 (struct app_save): Add scrub_m68k_mri, mri_state, and mri_last_ch
5559 (app_push): Save new fields.
5560 (app_pop): Restore new fields.
5561 (do_scrub_chars): Check scrub_m68k_mri rather than flag_mri_mri.
5562 If TC_M68K, use a trivial state machine to look for occurrences of
5563 the .mri pseudo-op, and change the mode appropriately.
5564 * as.h (do_scrub_begin): Update prototype.
5565 * input-scrub.c (input_scrub_begin): Pass flag_m68k_mri to
5567 * config/tc-m68k.c (reg_prefix_optional_seen): New static
5569 (m68k_mri_mode_change): New function.
5570 (md_parse_option): Set reg_prefix_optional_seen.
5571 * config/tc-m68k.h (m68k_mri_mode_change): Declare.
5572 (MRI_MODE_CHANGE): Define.
5573 * doc/as.texinfo: Document .mri pseudo-op.
5575 * app.c (do_scrub_chars): In MRI mode, don't treat '#' as a
5580 Support for OBJ_ELF on m68k, mostly inside #ifdef OBJ_ELF:
5581 * config/m68k-parse.h (enum pic_relocation): Define.
5582 (struct m68k_exp): Add pic_reloc field.
5583 * config/tc-m68k.h (TC_RELOC_RTSYM_LOC_FIXUP): Define.
5584 (tc_fix_adjustable): Define to call tc_m68k_fix_adjustable.
5585 (NO_RELOC): Define to BFD_RELOC_NONE if BFD_ASSEMBLER, to zero
5587 * config/tc-m68k.c: Delete definition of NO_RELOC.
5588 (struct m68k_it): Add pic_reloc field.
5589 (add_fix): Copy over pic_reloc field.
5590 (md_pseudo_table): Interpret .align parameter as byte count.
5591 (mote_pseudo_table): Likewise.
5592 (tc_m68k_fix_adjustable): New function.
5593 (get_reloc_code): New function.
5594 (md_assemble): Use it as last argument to fix_new_exp.
5595 (md_apply_fix_2): For a relocation against a symbol don't put the
5596 addend into the data.
5597 (tc_gen_reloc): Different addend computation for OBJ_ELF.
5598 (m68k_ip): Don't relax an operand that requires pic relocation.
5599 (md_begin): Align .text, .data and .bss on 4 byte boundary by
5601 * write.c (fixup_segment): Don't add symbol value to addend if
5602 TC_M68K and OBJ_ELF.
5603 * config/m68k-parse.y (yylex): Handle @PLTPC, etc.
5604 (motorola_operand): Add rule for `(zapc, EXPR)'.
5608 * ecoff.c (ecoff_directive_weakext): Fixed so that whitespace
5609 *really* is permissible before the comma.
5613 * config/tc-sh.c (sh_do_align): Align to a 2 byte boundary before
5614 inserting nop instructions.
5618 * config/te-aux.h: Change include of aux.h to aux-coff.h.
5622 * config/tc-i960.c (tc_coff_symbol_emit_hook): Correct storage
5623 class setting for a CALLNAME symbol in COFF.
5625 * read.c (potable): Pass negative numbers for new .balign[wl] and
5626 .p2align[wl] pseudo-ops.
5627 (s_align_bytes): Treat a negative argument as specifying the fill
5629 (s_align_ptwo): Likewise.
5633 * read.c (potable): Add balignw, balignl, p2alignw, and p2alignl.
5634 (do_align): Take new len parameter. Change all callers. Pass it
5636 (s_align_bytes): Arg now indicates the length of the fill pattern.
5637 (s_align_ptwo): Likewise.
5638 * config/obj-coff.c (write_object_file): Pass length to
5640 * config/tc-i386.h (md_do_align): Take new len parameter.
5641 * config/tc-m88k.h (md_do_align): Likewise.
5642 * config/tc-m88k.c (m88k_do_align): Likewise.
5643 * config/tc-sh.h (md_do_align): Likewise.
5644 * config/tc-sh.c (sh_do_align): Likewise.
5645 * doc/as.texinfo: Document new pseudo-ops.
5647 * config/obj-coff.c (fixup_mdeps): Divide offset by fr_var, as is
5648 done in cvt_frag_to_fill.
5650 * config/tc-sh.h (sh_do_align): Declare.
5651 (md_do_align): Define.
5652 * config/tc-sh.c (sh_do_align): New function.
5654 * ecoff.c (ecoff_build_lineno): Don't try to store the address
5655 difference if the next address is before the current one.
5657 * config/tc-m68k.c (struct m68k_cpu): Add alias field.
5658 (archs): Initialize new field.
5659 (m68k_ip): Don't list alias names when listing CPUs which support
5662 * as.c (main): Call parse_args before read_begin.
5663 * app.c (do_scrub_chars): If flag_m68k_mri, don't put a dot in
5664 front of generated pseudo-ops.
5665 * read.c (potable): Ignore "name".
5666 (s_app_file): Permit a single quote after the string, since one
5667 may appear in m68k MRI mode.
5669 * configure.in: Check for --enable-shared. If linking against
5670 shared BFD and opcodes, fix library name on SunOS, and try to set
5672 * configure: Rebuild.
5676 * as.h (flag_m68k_mri): Declare.
5677 * as.c (parse_args): If TC_M68K, set flag_m68k_mri for -M.
5678 * Many files: For MRI syntax that is specific to the m68k MRI
5679 assembler, check flag_m68k_mri rather than flag_mri or
5680 MRI_MODE_NEEDS_PSEUDO_DOT.
5684 * config/tc-i960.c (ARCH_HX): Define.
5686 (targ_has_sfr): Handle ARCH_HX.
5687 (targ_has_iclass): Handle ARCH_HX.
5688 (tc_coff_fix2rtype): Add return 0 to avoid warning.
5689 (tc_headers_hook): If the architecture was specified explicitly,
5690 use it when setting the flags. Set the extern variable coff_flags
5691 rather than headers->filehdr.f_flags, since the latter is set
5692 unconditionally in obj-coff.c.
5693 (i960_handle_align): Remove unused variable fixp.
5695 Support for building bfd and opcodes as shared libraries, based on
5697 * configure.in: Set OPCODES and BFD to search directories.
5698 Substitute OPCODES_DEP and BFDDEP. On SunOS, set HLDFLAGS.
5699 * configure: Rebuild.
5700 * Makefile.in (LDFLAGS, HLDFLAGS): New variables.
5701 (LIBDEPS): New variable.
5702 (as.new0: Depend upon $(LIBDEPS) rather than $(LIBS). Use
5703 $(HLDFLAGS) in link.
5704 (check): Set LD_LIBRARY_PATH in the environment.
5708 * config/tc-ppc.h (ELF_TC_SPECIAL_SECTIONS): Make .sdata2, .sbss2,
5709 .PPC.EMB.sdata0, and .PPC.EMB.sbss0 sections all default to
5710 read-only, not read/write.
5714 * Makefile.in (INSTALL_XFORM): Remove -e.
5718 * write.c (write_relocs): Use S_IS_DEFINED and S_IS_COMMON rather
5719 than comparing S_GET_SEGMENT to undefined_section.
5720 (write_object_file): Skip symbols which were equated to an
5721 undefined or common symbol.
5722 * symbols.c (resolve_symbol_value): Use S_IS_DEFINED and
5723 S_IS_COMMON rather than comparing S_GET_SEGMENT to
5725 (S_GET_VALUE): Likewise. Avoid recursion problems if S_IS_DEFINED
5726 or S_IS_COMMON call S_GET_VALUE.
5727 * config/obj-aout.h (S_IS_COMMON): Define if not BFD_ASSEMBLER.
5728 * config/obj-aout.c (obj_emit_relocations): If a reloc is equated
5729 to an undefined or common symbol, convert the reloc to be against
5731 (obj_crawl_symbol_chain): Skip symbols which were equated to an
5732 undefined or common symbol.
5733 * config/obj-bout.h (S_IS_COMMON): Define if not BFD_ASSEMBLER.
5734 * config/obj-bout.c (obj_emit_relocations): If a reloc is equated
5735 to an undefined or common symbol, convert the reloc to be against
5737 (obj_crawl_symbol_chain): Skip symbols which were equated to an
5738 undefined or common symbol.
5739 * config/obj-coff.c (do_relocs_for): Use S_IS_DEFINED and
5740 S_IS_COMMON rather than comparing S_GET_SEGMENT to
5742 (yank_symbols): Skip symbols which were equated to an undefined or
5747 * config/obj-aout.h (S_IS_LOCAL): Check for \002 as well as \001.
5748 * config/obj-bout.h (S_IS_LOCAL): Likewise.
5750 * configure.in: Make sure we only add m68k-parse.o to
5751 ${extra_objects} once, no matter how many m68k targets have been
5753 * configure: Rebuild.
5757 * configure.in (i386-*-cygwin32, ppc-*-cygwin32): New.
5758 * configure: Rebuild.
5762 * config/tc-m68k.c (md_pseudo_table): Add "extend" and "ldouble".
5763 * doc/c-m68k.texi: Document .extend and .ldouble.
5765 * configure.in (m68*-apple-aux*): New target.
5766 * config/te-aux.h: New file.
5767 * config/obj-coff.c (compare_external_relocs): New static function
5769 (do_relocs_for): Sort relocs if TE_AUX.
5770 (fixup_segment): If TE_AUX, store common symbol value in segment.
5771 * config/tc-m68k.h (TARGET_FORMAT): Define if TE_AUX.
5775 * config/obj-coff.h (S_IS_LOCAL): Check for \002 as well as \001.
5777 * config/tc-mips.c (s_mips_globl): Set BSF_OBJECT if it is not
5779 (s_cpload): Set BSF_OBJECT for _gp_disp symbol.
5780 * read.c (s_lcomm): If S_SET_SIZE is defined, set the size of the
5782 * ecoff.c (add_procedure): Set the BSF_FUNCTION flag.
5783 (ecoff_build_symbols): If S_SET_SIZE is defined, set the size of
5784 an undefined symbol and the size of a function symbol.
5785 * config/obj-elf.c (elf_frob_symbol): If TC_MIPS, set BSF_OBJECT
5786 for all common symbols.
5790 * config/tc-i960.c (parse_memop): In MRI mode, don't use implicit
5793 * expr.c (operand): Accept 0x hex constants in MRI mode if not on
5798 * config/obj-elf.c (obj_elf_type): Set BSF_OBJECT flag for a type
5801 * ecoff.c (localsym_t): Add addend field.
5802 (add_ecoff_symbol): Add addend argument. Change all callers.
5803 (coff_sym_value): Make static.
5804 (coff_sym_addend): New static variable.
5805 (ecoff_directive_def): Initialize coff_sym_addend.
5806 (ecoff_directive_val): Accept symbol + constant.
5807 (ecoff_directive_endef): Pass coff_sym_addend to add_ecoff_symbol.
5808 (ecoff_build_symbols): Include the addend in the symbol value.
5812 * config/tc-ppc.c (md_assemble): Ignore overflow on
5813 BFD_RELOC_16_GOTOFF and BFD_RELOC_PPC_TOC16.
5817 * config/tc-ppc.c (md_apply_fix3): SDA21 relocations are now 4
5818 bytes in size, so offset appropriately in big endian mode when
5819 writing the bottom 2 bytes.
5823 * config/tc-sparc.c (default_compatible): New static local.
5824 (md_begin): Initialize it. Rewrite warn_on_bump handling.
5825 (sparc_ip): If no architecture or -bump specified, don't mark as
5826 mismatched those in default_compatible.
5831 * configure.in (i386-*-sco*elf*): Use fmt elf, targ sco5.
5832 * configure: Rebuild.
5833 * config/sco5.mt: New file; set TDEFINES to -DSCO_ELF.
5834 * config/tc-i386.c (sco_id): New function, if SCO_ELF.
5835 * config/tc-i386.h (tc_init_after_args): Define if SCO_ELF.
5836 (sco_id): Declare if SCO_ELF.
5840 * config/tc-sparc.c (initial_architecture,can_bump_v9_p): Deleted.
5841 ({max,warn_after}_architecture): New static locals.
5842 (md_begin): Replace NUMOPCODES with sparc_num_opcodes.
5843 If both architecture and -bump requested, set max_architecture to max.
5844 (sparc_md_end): Simplify.
5845 (sparc_ip): Replace references to can_bump_v9_p with max_architecture.
5846 Rewrite code to bump architecture and check for conflicts.
5847 (md_longopts): Recognize -xarch={v8plus,v8plusa} for compatibility
5848 with Solaris assembler.
5849 (md_parse_option): Likewise. Call sparc_opcode_lookup_arch.
5850 (md_show_usage): Update.
5854 * Makefile.in (RUNTEST): Fix reference to $${srcdir}.
5858 * config/tc-sparc.h (TARGET_FORMAT): Use #ifdef SPARC_ARCH64 instead of
5859 #ifdef sparcv9 when choosing value.
5862 (sparc_md_end): Declare.
5863 * config/tc-sparc.c (SPARC_V9): Renamed from sparcv9.
5864 (initial_architecture): New static local.
5865 (can_bump_v9_p): Likewise.
5866 (NO_V9): Delete all occurrences.
5867 (sparc_md_end): New function.
5868 (sparc_ip): New local v9_arg_p. Rework fp reg number test.
5869 Don't bump architecture to v9 unless can_bump_v9_p set.
5870 (md_parse_option): -A<arch> passed, set can_bump_v9_p accordingly.
5871 * configure.in (sparc64 target cpu): Don't set obj_format here.
5872 (SPARC_V9): Renamed from sparcv9.
5873 (sparc64-*-elf*): Define SPARC_ARCH64.
5874 * configure: Regenerated.
5875 * acconfig.h (SPARC_V9): Renamed from sparcv9.
5876 (SPARC_ARCH64): Add.
5877 * conf.in: Regenerated.
5878 * config/vmsconf.h: Update.
5882 * config/tc-mips.c (load_register): Optimise "dli" loads.
5883 (md_show_usage): add "-mcpu=vr4100" to help text.
5887 * symbols.c (resolve_symbol_value): If a symbol is equated to an
5888 undefined symbol, preserve the X_op of O_symbol.
5889 (S_GET_VALUE): Fix check to permit this case.
5890 * write.c (write_relocs): If a reloc is against an undefined
5891 symbol equated to another symbol, change the reloc to be against
5893 * config/obj-coff.c (do_relocs_for): Likewise.
5895 * config/tc-ppc.c (ppc_csect): An unnamed csect is storage class
5900 * config/obj-elf.c (elf/ppc.h): Include elf/ppc.h if target
5901 computer is PowerPC.
5903 * config/tc-ppc.c (md_apply_fix3): Add more embedded relocations.
5905 * config/tc-ppc.h (ELF_TC_SPECIAL_SECTIONS): Add sections
5906 mentioned in the eabi.
5910 * config/tc-ppc.c (ppc_reldata): Changed alignement on reldata_section
5911 * config/tc-ppc.c (ppc_pdata): Changed the alignment on pdata_section
5915 * config/tc-ppc.c (mapping): Add more relocation suffixes.
5919 * config/tc-ppc.c (ppc_elf_validate_fix): Allow .gcc_except_table
5920 as a section it is ok to have unadorned -mrelocatable pointers in.
5924 * config/tc-ppc.c (ppc_section*): Wrap these functions inside
5929 * config/obj-elf.c (obj_elf_section): Add hooks so machine
5930 dependent section attributes can be handled.
5932 * config/tc-ppc.h: (md_elf_section_{letter,type,word,flags}): New
5933 macros to add support for exclude section flag and ordered section
5936 * config/tc-ppc.c (ppc_elf_section_{letter,type,word,flags}): New
5937 functions to add support for exclude section flag and ordered
5942 * subsegs.c (section_symbol): Don't try to look up the section
5943 symbol in the hash table. It should be possible to have a symbol
5944 with the same name as a section, but no connection to it.
5946 * read.c (cons_worker): Only call mri_comment_end from flag_mri.
5949 * expr.c (operand): Skip whitespace after a close parenthesis.
5954 * config/tc-sh.c (md_apply_fix): Call as_bad_where instead of
5955 as_warn_where for relocation overflow.
5956 (parse_reg): Accept register name only if next character is
5959 For older changes see ChangeLog-9295