3 * config/tc-pdp11.c (md_apply_fix): Cast first argument to
4 md_chars_to_numbers to an unsigned pointer in order to avoid a
9 * config/tc-ppc.c (ppc_target_format): Add VxWorks.
13 * config/tc-ms1.c: New.
14 * config/tc-ms1.h: New.
15 * testsuite/gas/ms1/allinsn.d: New.
16 * testsuite/gas/ms1/allinsn.s: New.
17 * testsuite/gas/ms1/badinsn.s: New.
18 * testsuite/gas/ms1/badinsn1.s: New.
19 * testsuite/gas/ms1/badoffsethigh.s: New.
20 * testsuite/gas/ms1/badoffsetlow.s: New.
21 * testsuite/gas/ms1/badorder.s: New.
22 * testsuite/gas/ms1/badreg.s: New.
23 * testsuite/gas/ms1/badsignedimmhigh.s: New.
24 * testsuite/gas/ms1/badsignedimmlow.s: New.
25 * testsuite/gas/ms1/badsyntax.s: New.
26 * testsuite/gas/ms1/badsyntax1.s: New.
27 * testsuite/gas/ms1/badunsignedimmhigh.s: New.
28 * testsuite/gas/ms1/badunsignedimmlow.s: New.
29 * testsuite/gas/ms1/errors.exp: New.
30 * testsuite/gas/ms1/ldst.s: New.
31 * testsuite/gas/ms1/misc.d: New.
32 * testsuite/gas/ms1/misc.s: New.
33 * testsuite/gas/ms1/ms1-16-003.d: New.
34 * testsuite/gas/ms1/ms1-16-003.s: New.
35 * testsuite/gas/ms1/ms1.exp: New.
36 * testsuite/gas/ms1/msys.d: New.
37 * testsuite/gas/ms1/msys.s: New.
38 * testsuite/gas/ms1/relocs.d: New.
39 * testsuite/gas/ms1/relocs.exp: New.
40 * testsuite/gas/ms1/relocs1.s: New.
41 * testsuite/gas/ms1/relocs2.s: New.
45 * config/tc-i386.h (CpuSVME): New.
46 (CpuUnknownFlags): Include CpuSVME.
47 * config/tc-i386.c (cpu_arch): Add .pacifica and .svme. Add opteron
48 as alias of sledgehammer.
49 (md_assemble): Include invlpga in the check for insns with two source
51 (process_operands): Include SVME insns in the check for ignored
52 segment overrides. Adjust diagnostic.
53 (i386_index_check): Special-case SVME insns with memory operands.
58 * tc-arm.c (struct arm_it): Make operands.imm signed to match its
59 use an immediate value.
60 (parse_vfp_reg_list): Make the 2nd parameter an unsigned pointer
61 since the register field of the operands structure is unsigned.
62 (s_arm_unwind_save_vfp): Make "reg" unsigned.
63 (parse_operands): Make the 2ns parameter an unsigned pointer to
65 (do_ldrd): When using the imm field of the operands structure as a
66 second register field, treat it as unsigned.
71 * config/obj-elf.c (obj_elf_change_section): Use backend
76 * config/tc-ia64.c (line_separator_chars): Add '{' and '}'.
77 (output_spill_psprel, output_spill_psprel_p): Combine.
78 (output_spill_sprel, output_spill_sprel_p): Combine.
79 (output_spill_reg, output_spill_regp_p): Combine.
80 (process_one_record): Handle psp_psprel.
81 (parse_predicate_and_operand): New.
82 (convert_expr_to_ab_reg): Two new parameters. Return void. Always
83 initialize output values. Emit diagnostic case here.
84 (convert_expr_to_xy_reg): Likewise. Don't allow r0, f0, and f1.
85 (add_unwind_entry): New second parameter. Allow first parameter to
86 be NULL. Parse optional tag, emit warning about further support for
87 it otherwise being missing. Check end-of-line when requested.
88 (dot_fframe): Clear operand when wrong. Allow tag.
89 (dot_vframe): Likewise.
90 (dot_vframesp): Likewise. Rename parameter, issue warning when psp
92 (dot_vframepsp): Remove.
93 (dot_altrp): Clear operand when wrong. Allow tag.
94 (dot_save): Likewise. Let default case also go through
96 (dot_savemem): Likewise.
97 (dot_restore): Don't return when wrong operand. Allow tag.
98 (dot_spillreg, dot_spillreg_p): Combine. Simplify by using
99 parse_predicate_and_operand and the new arguments to
100 convert_expr_to_ab_reg and convert_expr_to_xy_reg. Don't return
101 when wrong operand. Allow tag.
102 (dot_restorereg, dot_restorereg_p): Likewise.
103 (dot_spillmem, dot_spillmem_p): Likewise.
104 (dot_saveg): Clear operand when wrong. Perform tighter operand
106 (dot_savef): Likewise.
107 (dot_saveb): Likewise.
108 (dot_savegf): Likewise.
109 (dot_spill): Remove end-of-line check. Combine. Simplify by using
110 parse_predicate_and_operand and the new arguments to
111 convert_expr_to_ab_reg and convert_expr_to_xy_reg. Don't return
112 when wrong operand. Allow tag.
114 (dot_label_state): Don't return when wrong operand.
115 (dot_copy_state): Likewise.
116 (dot_unwabi): Likewise. Check if in prologue.
117 (dot_body): Don't call demand_empty_rest_of_line.
118 (dot_prologue): Type of mask and grsave is unsigned. Perform tighter
120 (md_pseudo_table): Also use dot_restorereg for .restorereg.p. Also
121 use dot_spillreg for .spillreg.p. Also use dot_spillmem for
122 .spillpsp.p and .spillsp.p. Also use dot_vframesp for .vframepsp.
123 (parse_operand): New second parameter. Don't deal with '}' here
124 anymore. Don't advance past end-of-line.
125 (parse_operands): Pass second argument to parse_operand.
126 (ia64_start_line): Prevent out-of-bounds access through
127 input_line_pointer. Deal with '}' here.
128 (ia64_unrecognized_line): Don't deal with '}' here.
129 (dot_alias): Use ignore_rest_of_line not its deprecated alias
130 discard_rest_of_line.
134 * config/tc-arm.c (T_OPCODE_BRANCH, encode_arm_addr_mode_2)
135 (encode_arm_addr_mode_3, encode_arm_cp_address, do_blx, do_t_blx)
136 (do_t_branch, insns [b, bl]): Don't encode pipeline offset.
137 (do_branch): Always set inst.reloc.pc_rel.
138 (s_arm_elf_cons): Disallow use of (plt) suffix.
139 (do_adrl): Adjust X_add_number unconditionally.
140 (md_pcrel_from): Rename md_pcrel_from_section, add second segT
141 argument. Handle all adjustment for pipeline offset here.
142 (md_apply_fix): No need to undo work of md_pcrel_from. No
143 need to extract pre-encoded pipeline adjustments from various
144 branch instructions. Generally, assume instructions are already
145 all-bits-zero in the field being fixed up. Remove all OBJ_ELF
146 special cases. Handle BFD_RELOC_ARM_PLT32 like
147 BFD_RELOC_ARM_PCREL_BRANCH.
148 (tc_gen_reloc): Remove OBJ_ELF special case.
149 * config/tc-arm.c: Define MD_PCREL_FROM_SECTION.
153 * Makefile.am (check-DEJAGNU): Don't search for expect.
154 * Makefile.in: Regenerate.
158 * Makefile.am (EXPECT): Set to expect.
159 (RUNTEST): Likewise, set to runtest.
160 * Makefile.in: Regenerate.
164 * config/m68k-parse.h: Use ISO C90.
165 * config/m68k-parse.y: Likewise.
166 * config/tc-m68k.h: Likewise.
171 * config/tc-i386.c (md_assemble): Don't call optimize_disp on
173 (optimize_disp): Optimize only if possible. Don't use 64bit
174 displacement on non-constants and do same on constants if
179 * config/tc-i386.c (reloc): Also handle BFD_RELOC_64_PCREL.
180 (tc_i386_fix_adjustable): Include BFD_RELOC_X86_64_GOTOFF64,
181 BFD_RELOC_X86_64_DTPOFF64, and BFD_RELOC_X86_64_TPOFF64.
182 (output_disp): Do GOTPC conversion also for BFD_RELOC_X86_64_32S
183 and BFD_RELOC_32_PCREL. Use BFD_RELOC_X86_64_GOTPC32 instead of
185 (output_imm): Do GOTPC conversion also for BFD_RELOC_X86_64_32S.
186 Use BFD_RELOC_X86_64_GOTPC32 instead of aborting.
187 (tc_gen_reloc): Do GOTPC conversion also for BFD_RELOC_32_PCREL.
188 Use BFD_RELOC_X86_64_GOTPC32 instead of aborting. Also handle
189 BFD_RELOC_X86_64_GOTOFF64, BFD_RELOC_X86_64_GOTPC32,
190 BFD_RELOC_X86_64_DTPOFF64, and BFD_RELOC_X86_64_TPOFF64. Also
191 convert 8-byte pc-relative relocations.
192 (lex_got): Use BFD_RELOC_X86_64_GOTOFF64 for 64-bit @gotoff.
193 (i386_validate_fix): Likewise.
194 (x86_cons): Also handle quad values in 64-bit mode.
195 (i386_displacement): Also handle BFD_RELOC_X86_64_GOTOFF64.
196 (md_apply_fix): Include BFD_RELOC_X86_64_DTPOFF64 and
197 BFD_RELOC_X86_64_TPOFF64 in the TLS check. Also convert BFD_RELOC_64
198 to pc-relative variant. Also check for BFD_RELOC_64_PCREL.
202 * config/tc-arm.c (find_real_start): Check S_IS_LOCAL on
203 symbolP as well as for names with a leading dot. Use ACONCAT.
204 (md_apply_fix): For branch relocations, only replace value
205 with fixP->fx_offset (under #ifdef OBJ_ELF) when !fixP->fx_done.
206 (arm_force_relocation): Remove #ifdef OBJ_ELF case.
207 * config/tc-arm.h (LOCAL_LABEL): Remove unnecessary parentheses.
208 (LOCAL_LABEL_PREFIX): Don't define.
212 * config/tc-hppa.c (pa_block): Allocate just one byte for the
218 * config/tc-ia64.c (slot_index): Revert last change. If first_frag
219 is NULL, then emit a warning, and return the current index.
224 * config/tc-m68k.c (m68k_ip): Test for insn compatiblity using a
225 temporary copy of the operands array so that changes can be safely
226 backed out if the insn does not match.
227 (m68k_compare_opcode): Shortcut the test when the parameters are
228 the same. Return 1 if the names match but the second opcode is
229 further on in the array than the first.
234 * config/tc-ia64.c (slot_index): Check for a NULL first_frag.
238 * config/tc-m32r.c (use_parallel): Change default value from 1 to 0.
244 * configure.in: Add ms1 case.
246 * configure: Regenerate.
248 * configure.tgt: Add ms1 case.
252 * config/tc-xtensa.h (resource_table): Change units to unsigned chars.
253 * config/tc-xtensa.c (new_resource_table): Likewise.
254 (resize_resource_table): Likewise.
255 (release_resources): Fix assertion for unsigned values.
259 * cgen.c, cgen.h, tc.h, write.c, config/obj-coff.c
260 * config/tc-a29k.c, config/tc-alpha.c, config/tc-alpha.h
261 * config/tc-arc.c, config/tc-arc.h, config/tc-arm.c
262 * config/tc-arm.h, config/tc-avr.c, config/tc-avr.h
263 * config/tc-cris.c, config/tc-crx.c, config/tc-d10v.c
264 * config/tc-d10v.h, config/tc-d30v.c, config/tc-d30v.h
265 * config/tc-dlx.c, config/tc-dlx.h, config/tc-fr30.h
266 * config/tc-frv.c, config/tc-frv.h, config/tc-h8300.c
267 * config/tc-h8500.c, config/tc-hppa.c, config/tc-hppa.h
268 * config/tc-i370.c, config/tc-i370.h, config/tc-i386.c
269 * config/tc-i386.h, config/tc-i860.c, config/tc-i860.h
270 * config/tc-i960.c, config/tc-i960.h, config/tc-ia64.c
271 * config/tc-ip2k.c, config/tc-ip2k.h, config/tc-iq2000.c
272 * config/tc-iq2000.h, config/tc-m32r.c, config/tc-m32r.h
273 * config/tc-m68hc11.c, config/tc-m68hc11.h, config/tc-m68k.c
274 * config/tc-m68k.h, config/tc-m88k.c, config/tc-maxq.c
275 * config/tc-mcore.c, config/tc-mcore.h, config/tc-mips.c
276 * config/tc-mips.h, config/tc-mmix.c, config/tc-mn10200.c
277 * config/tc-mn10300.c, config/tc-msp430.c, config/tc-ns32k.c
278 * config/tc-openrisc.h, config/tc-or32.c, config/tc-or32.h
279 * config/tc-pdp11.c, config/tc-pj.c, config/tc-pj.h
280 * config/tc-ppc.c, config/tc-ppc.h, config/tc-s390.c
281 * config/tc-s390.h, config/tc-sh64.c, config/tc-sh.c
282 * config/tc-sh.h, config/tc-sparc.c, config/tc-sparc.h
283 * config/tc-tahoe.c, config/tc-tic30.c, config/tc-tic4x.c
284 * config/tc-tic54x.c, config/tc-tic80.c, config/tc-v850.c
285 * config/tc-v850.h, config/tc-vax.c, config/tc-vax.h
286 * config/tc-w65.c, config/tc-xstormy16.c, config/tc-xstormy16.h
287 * config/tc-xtensa.c, config/tc-z8k.c:
288 Replace all instances of the string "_apply_fix3" with
290 * po/POTFILES.in, po/gas.pot: Regenerate.
294 * Makefile.am: Run "make dep-am".
295 (POTFILES): Remove GAS_CFILES.
296 * Makefile.in: Regenerate.
300 * config/tc-mips.c (mips_cpu_info_table): Add cpu names m4k, 24k,
301 24kc, 24kf and 24kfx under MIPS32 release 2.
305 * config/tc-arm.c (CE, C3, CM, UE, UF): Redefine without reference
306 to their Thumb-enabled equivalents.
310 * config/tc-mips.c (load_register): Add leading "0x" to the
311 output of sprintf_vma().
316 * config/tc-arm.c (TxCE, TxC3, TxCM, TUE, TUF): Remove redundant
317 test for the presence of thumb version of the parsing functions
318 since they must always exist and the test generates a compile time
323 * config/tc-alpha.c (O_lituse_jsrdirect): New.
324 (alpha_reloc_op): Add it.
325 (debug_exp): Handle it.
326 (DUMMY_RELOC_LITUSE_JSRDIRECT): New.
327 (emit_insn): Handle it.
328 * doc/c-alpha.texi (Alpha-Relocs): Document lituse_jsrdirect.
332 * write.c (dump_section_relocs): Convert to ISO-C.
333 (write_relocs): Avoid signed/unsigned and fprintf argument
334 warnings in debug code.
338 * config/tc-arm.h (TC_FIX_TYPE): Change to int.
339 (TC_INIT_FIX_DATA): Initialize to 0, not NULL.
340 * config/tc-arm.c (fix_new_arm): Remove now-unnecessary cast.
341 (md_apply_fix3): Delete fix_is_thumb variable; refer to
342 fixP->tc_fix_data directly in the sole place it was used.
343 Explicitly truncate value, *valP, fixP->fx_addnumber, and
344 fixP->fx_offset to 32 bits, for consistent behavior between 32-
349 * config/tc-ia64.c (struct proc_pending): New.
350 (unwind): Replace proc_start with proc_pending.
351 (unwind_diagnostic): Check unwind.proc_pending.sym.
352 (dot_proc): Replace unwind.proc_start with unwind.proc_pending.sym.
353 Check if previous proc not closed. Record all entry points.
354 (dot_endp): Replace unwind.proc_start with unwind.proc_pending.sym.
355 Set symbol sizes for entry points recorded in dot_proc. Check
356 arguments for consistency with respective .proc's.
357 (md_assemble): Replace unwind.proc_start with
358 unwind.proc_pending.sym.
362 * config/tc-ia64.c (emit_one_bundle): Restrict scope of ptr, end_ptr,
363 and last_ptr. Check all in-use slots for first one with non-NULL
364 unwind_record. Don't reload end_ptr before second update round.
368 * config/tc-ia64.c (extra_goodness): Update comment.
369 (md_begin): Add debugging code to print best_template table.
373 * config/tc-ia64.c (md_begin): Don't try to match slot 2 of an MLX
378 * config/tc-ia64.c (ia64_gen_real_reloc_type): Also handle
379 BFD_RELOC_UNUSED when determining the width of the reloc.
383 * config/tc-ia64.c (dot_endp): Clear out all three pointers in unwind
388 * config/tc-ia64.c (dot_radix): Rewrite.
392 * config/tc-ia64.c (struct unw_rec_list): Remove next_slot_number
394 (alloc_record): Remove references to next_slot_number and
396 (emit_one_bundle): Likewise.
400 * config/tc-v850.c (md_apply_fix3): Pass the address of the
401 message buffer when invoking the insert function.
405 * config/tc-hppa.c (pa_ip): Promote architecture from PA 1.0 to 1.1
406 only if an instruction match is found.
410 * config/tc-xtensa.c (xg_assemble_vliw_tokens): Change subtraction
411 to addition in argument to xtensa_dwarf2_emit_insn.
415 * Makefile.am: Have 'all' depend on 'info'.
416 * Makefile.in: Regenerate.
420 * config/tc-ppc.c (ppc_force_relocation): Add BFD_RELOC_24_PLT_PCREL.
424 * config/tc-ppc.c (parse_cpu): Handle "-mpower5".
425 (md_show_usage): Document it.
426 (ppc_setup_opcodes): Insert POWER5 mnemonics.
427 * doc/c-ppc.texi (PowerPC-Opts): Document "-mpower5".
431 * config/tc-ia64.c (dot_endp): Don't use global symbol for unwind
432 relocations in unwind section.
436 * config/tc-v850.c (md_apply_fix3): Only use the insertion routine
437 if one exists. Ignore any error messages it may produce, just
438 allow it to perform the insertion.
442 * hash.c (hash_lookup): Add len parameter. All callers changed.
443 (hash_find_n): New interface.
444 * hash.h: Prototype hash_find_n.
445 * sb.c: Include as.h.
446 (scrub_from_sb, sb_to_scrub, scrub_position): New statics.
447 (sb_scrub_and_add_sb): New interface.
448 * sb.h: Prototype sb_scrub_and_add_sb.
449 * input-scrub.c (input_scrub_include_sb): Use sb_scrub_and_add_sb.
451 * config/tc-arm.h (TC_FORCE_RELOCATION_LOCAL): Remove
452 reference to BFD_RELOC_ARM_GOT12 which is never generated.
453 * config/tc-arm.c: Rewrite, adding Thumb-2 support.
457 * doc/Makefile.am (gasver.texi): Don't use $<.
458 * doc/Makefile.in: Regenerated.
463 * symbols.c (resolve_symbol_value): Do not move symbols whose
464 value expression has not been resolved or finalized into the
465 absolute section as this will confuse other parts of the assembler
466 into thinking that their value is zero.
470 * read.c (_find_end_of_line): New.
471 (find_end_of_line): New.
472 (HANDLE_CONDITIONAL_ASSEMBLY): Use it.
473 (read_a_source_file): Use it.
476 (get_line_sb): Use it.
477 (s_errwarn): Replace discard_rest_of_line by ignore_rest_of_line.
478 (s_comm_internal): Likewise.
481 (s_ignore): Use ignore_rest_of_line.
482 * read.h (find_end_of_line): Prototype.
483 (discard_rest_of_line): Remove prototype. #define to
488 * config/tc-v850,h (TC_FIX_TYPE): Define.
489 (TC_INIT_FIX_TYPE): Define.
490 * config/tc-v850.c (md_assemble): When creating a fix record the
491 operand in the tc_fix_data field.
492 (md_apply_fix3): When applying a resolved fix use the operand's
493 insertion procedure to store the value, if the operand has been
498 * Makefile.am (m68k-parse.c, itbl-parse.c): Update ylwrap
500 * Makefile.in: Regenerated.
505 * config/tc-xtensa.c (xtensa_insnbuf_set_operand): Clarify error
507 (xtensa_mark_zcl_first_insns): Fix incorrect nesting of conditional
508 for handling RELAX_CHECK_ALIGN_NEXT_OPCODE.
512 * config/tc-ppc.c (md_apply_fix3): Allow pcrel forms of BFD_RELOC_16,
513 BFD_RELOC_LO16, BFD_RELOC_HI16 and BFD_RELOC_HI16_S.
517 * frags.c (frag_grow): Don't be too greedy in allocating memory.
519 * config/tc-hppa.c (pa_block): Check arguments to .block[z].
524 * config/tc-mmix.c (mmix_handle_mmixal): Rearrange slightly.
525 Handle label-without-colon before ordinary dot-pseudo as an
526 ordinary label. Don't leak memory for label-without-colon alone
527 on a line. Don't mmixal-munge operands for dot-pseudos.
531 * macro.c (get_any_string): Remove the two last parameters. Replace
532 references to the former expand parameter by using macro_alternate.
533 Simplify loop condition for checking for end-of-string.
534 (get_string): Remove redunant call to sb_skip_white.
535 (do_formals): Remove two last arguments to get_any_string.
536 (macro_expand): Likewise.
537 (expand_irp): Likewise.
541 * read.c (s_macro): Move local variable 'local' to smaller scope.
542 Call sb_kill on it when done.
546 * config/tc-i386.c (optimize_disp): Discard displacement entirely
547 when zero and not required by encoding constraints.
552 * config/tc-sh64.c (sh64_fake_label): New.
554 * config/tc-sh64.h (TC_FAKE_LABEL): New.
556 * doc/internals.texi (TC_FAKE_LABEL): Document.
558 * write.c (TC_FAKE_LABEL): New.
559 (adjust_reloc_syms): Use it.
560 (write_object_file): Likewise.
564 * config/tc-i386.c (parse_insn): Disallow use of prefix separator
565 and comma in Intel mode.
569 * config/tc-i386.c (tc_x86_regname_to_dw2regnum): Correct 64-bit mode
570 names to match ABI. Add more registers for 32-bit and 64-bit modes.
571 Make name array static and const. Adjust lookup to account for NULL
572 entries (standing for unused register numbers).
576 * config/tc-i386.c (parse_insn): Consider all matching instructions
577 when checking for string instruction after string-only prefix.
582 * config/tc-ia64.c (start_unwind_section): Properly check
583 comdat group with SHF_GROUP.
587 * doc/c-xtensa.texi (Literal Directive): Spelling correction.
591 * config/tc-xtensa.c: Remove excess whitespace.
592 * config/tc-xtensa.h: Likewise.
593 * config/xtensa-istack.h: Likewise.
594 * config/xtensa-relax.c: Likewise.
595 * config/xtensa-relax.h: Likewise.
599 * sb.h: Fix formatting of comments.
600 * sb.c: Fix formatting of comments.
604 * sb.h (sb_add_buffer): Reintroduce.
605 * sb.c (sb_add_buffer): Likewise.
609 * macro.c (new_formal, del_formal): New.
610 (do_formals): Use new_formal. Check for and parse qualifier. Warn if
611 required argument has default value. Stop looking for more formal
612 when there was a vararg one.
613 (macro_expand_body): Use new_formal and del_formal.
614 (macro_expand): Likewise. Initialize local variable err. Don't
615 return immediately when encountering an error. Warn when keyword
616 argument already had a value assigned. Eliminate duplicate clearing
617 of argument value. When current positional argument matches parameter
618 of vararg type, assign to it all the remaining arguments. Issue error
619 when required parameter does not have value.
620 (free_macro): Use del_formal.
621 (expand_irp): Initialize formal type. Free buffers associated with
622 formal prior to returning.
623 * macro.h (struct formal_struct): Add new field 'type' with new
624 enumeration type 'formal_type'.
625 * doc/as.texinfo: Document macro parameter qualifiers.
626 * NEWS: Mention new functionality.
630 * cond.c (s_ifb): New.
631 * read.c (potable): Add s_ifb as handler for .ifb and .ifnb.
632 * read.h (s_ifb): Prototype.
633 * doc/as.texinfo: Document .ifb and .ifnb.
637 * config/tc-ia64.c (MIN): Undef.
641 * config/tc-i386.h (ELF_TARGET_FORMAT): Define for TE_VXWORKS.
645 * config/tc-vax.c (md_assemble): Don't assume a valueT is 4
650 * Update the address and phone number of the FSF organization in
651 the GPL notices in the following files:
652 COPYING, app.c, as.c, as.h, asintl.h, atof-generic.c, bignum.h,
653 bit_fix.h, cgen.c, cgen.h, cond.c, debug.c, depend.c, dw2gencfi.c,
654 dw2gencfi.h, dwarf2dbg.c, dwarf2dbg.h, ecoff.c, ecoff.h, ehopt.c,
655 emul-target.h, emul.h, expr.c, expr.h, flonum-copy.c,
656 flonum-konst.c, flonum.h, frags.c, frags.h, hash.c, hash.h,
657 input-file.c, input-file.h, input-scrub.c, itbl-lex.h, itbl-lex.l,
658 itbl-ops.c, itbl-ops.h, itbl-parse.y, listing.c, listing.h,
659 literal.c, macro.c, macro.h, messages.c, obj.h, output-file.c,
660 output-file.h, read.c, read.h, sb.c, sb.h, stabs.c,
661 struc-symbol.h, subsegs.c, subsegs.h, symbols.c, symbols.h, tc.h,
662 write.c, write.h, config/aout_gnu.h, config/atof-ieee.c,
663 config/atof-vax.c, config/itbl-mips.h, config/m68k-parse.h,
664 config/m68k-parse.y, config/m88k-opcode.h, config/obj-aout.c,
665 config/obj-aout.h, config/obj-bout.c, config/obj-bout.h,
666 config/obj-coff.c, config/obj-coff.h, config/obj-ecoff.c,
667 config/obj-ecoff.h, config/obj-elf.c, config/obj-elf.h,
668 config/obj-evax.c, config/obj-evax.h, config/obj-hp300.c,
669 config/obj-hp300.h, config/obj-ieee.c, config/obj-ieee.h,
670 config/obj-multi.h, config/obj-som.c, config/obj-som.h,
671 config/tc-a29k.c, config/tc-a29k.h, config/tc-alpha.c,
672 config/tc-alpha.h, config/tc-arc.c, config/tc-arc.h,
673 config/tc-arm.h, config/tc-avr.c, config/tc-avr.h,
674 config/tc-cris.c, config/tc-cris.h, config/tc-crx.c,
675 config/tc-crx.h, config/tc-d10v.c, config/tc-d10v.h,
676 config/tc-d30v.c, config/tc-d30v.h, config/tc-dlx.c,
677 config/tc-dlx.h, config/tc-fr30.c, config/tc-fr30.h,
678 config/tc-frv.c, config/tc-frv.h, config/tc-generic.c,
679 config/tc-generic.h, config/tc-h8300.c, config/tc-h8300.h,
680 config/tc-h8500.c, config/tc-h8500.h, config/tc-hppa.c,
681 config/tc-hppa.h, config/tc-i370.c, config/tc-i370.h,
682 config/tc-i386.c, config/tc-i386.h, config/tc-i860.c,
683 config/tc-i860.h, config/tc-i960.c, config/tc-i960.h,
684 config/tc-ia64.c, config/tc-ia64.h, config/tc-ip2k.c,
685 config/tc-ip2k.h, config/tc-iq2000.c, config/tc-iq2000.h,
686 config/tc-m32r.c, config/tc-m32r.h, config/tc-m68851.h,
687 config/tc-m68hc11.c, config/tc-m68hc11.h, config/tc-m68k.c,
688 config/tc-m68k.h, config/tc-m88k.c, config/tc-m88k.h,
689 config/tc-maxq.c, config/tc-maxq.h, config/tc-mcore.c,
690 config/tc-mcore.h, config/tc-mips.c, config/tc-mips.h,
691 config/tc-mmix.c, config/tc-mmix.h, config/tc-mn10200.c,
692 config/tc-mn10200.h, config/tc-mn10300.c, config/tc-mn10300.h,
693 config/tc-msp430.c, config/tc-msp430.h, config/tc-ns32k.c,
694 config/tc-ns32k.h, config/tc-openrisc.c, config/tc-openrisc.h,
695 config/tc-or32.c, config/tc-or32.h, config/tc-pdp11.c,
696 config/tc-pdp11.h, config/tc-pj.c, config/tc-pj.h,
697 config/tc-ppc.c, config/tc-ppc.h, config/tc-s390.c,
698 config/tc-s390.h, config/tc-sh.c, config/tc-sh.h,
699 config/tc-sh64.c, config/tc-sh64.h, config/tc-sparc.c,
700 config/tc-sparc.h, config/tc-tahoe.c, config/tc-tahoe.h,
701 config/tc-tic30.c, config/tc-tic30.h, config/tc-tic4x.c,
702 config/tc-tic4x.h, config/tc-tic54x.c, config/tc-tic54x.h,
703 config/tc-tic80.c, config/tc-tic80.h, config/tc-v850.c,
704 config/tc-v850.h, config/tc-vax.c, config/tc-vax.h,
705 config/tc-w65.c, config/tc-w65.h, config/tc-xstormy16.c,
706 config/tc-xstormy16.h, config/tc-xtensa.c, config/tc-xtensa.h,
707 config/tc-z8k.c, config/tc-z8k.h, config/te-386bsd.h,
708 config/te-freebsd.h, config/te-hp300.h, config/te-hppa.h,
709 config/te-ic960.h, config/te-irix.h, config/te-nbsd.h,
710 config/te-netware.h, config/te-sparcaout.h, config/te-sun3.h,
711 config/te-tmips.h, config/te-vxworks.h, config/vax-inst.h,
712 config/xtensa-istack.h, config/xtensa-relax.c,
713 config/xtensa-relax.h, doc/fdl.texi
717 * config/tc-arm.c (arm_opts): Make -mlittle-endian switch set
718 the target_big_endian variable to false.
722 * config/obj-ecoff.c (ecoff_frob_file_before_fix): Correct section
723 list traversal. Use bfd_section_list_prepend.
724 * config/tc-mmix.c (mmix_frob_file): Don't needlessly iterate
725 over the section list.
726 * config/tc-xtensa.c (xtensa_remove_section): Delete.
727 (xtensa_insert_section): Delete.
728 (xtensa_move_seg_list_to_beginning): Use bfd_section_list_remove
729 and bfd_section_list_prepend.
730 (xtensa_reorder_seg_list): Use bfd_section_list_remove and
731 bfd_section_list_insert_after.
735 * config/obj-ecoff.c (ecoff_frob_file_before_fix): Fix invocations
736 of bfd_section_list... macros.
737 * config/tc-mmix.c (mmix_frob_file): Likewise.
738 * config/tc-xtensa.c (xtensa_remove_section): Likewise.
739 (xtensa_insert_section): Likewise.
741 * macro.c (macro_hash): Remove static.
742 * macro.h (macro_hash): Provide an external declaration.
746 * write.c (write_object_file): Use bfd_section_double_list_remove
751 * doc/Makefile.am (gasver.texi): Correct quoting.
752 * doc/Makefile.in: Regenerated.
756 * configure.tgt: Add h8300*-*-rtemscoff.
757 Switch h8300*-*-rtems* to elf.
761 * hash.c: Undo the last change.
766 * doc/Makefile.am (gasver.texi): Depend on bfd/configure instead
768 (DISTCLEANFILES): Remove.
769 (MAINTAINERCLEANFILES): Add asconfig.texi.
770 * aclocal.m4, Makefile.in, doc/Makefile.in: Regenerated.
774 * Makefile.am (GAS_CFILES): Remove bignum-copy.c.
775 (GENERIC_OBJS): Likewise, remove bignum-copy.o.
776 (bignum-copy.o): Remove.
777 * Makefile.in: Regenerate.
778 * makefile.vms (OBJS): Remove bignum-copy.obj.
779 * symbols.h (local_symbol_make): Remove declaration.
780 (verify_symbol_chain_2): Likewise.
781 * symbols.c (local_symbol_make): Make static.
782 (max_indent_level): Likewise.
783 (verify_symbol_chain_2): Remove.
784 * macro.c (macro_hash): Make static.
785 * messages.c (fprint_value): Remove.
786 * read.h (get_absolute_expr): Remove.
787 (emit_leb128_expr): Likewise.
788 (do_s_func): Likewise.
789 * read.c (do_s_func): Make static.
790 (emit_leb128_expr): Likewise.
791 (get_absolute_expr): Likewise.
792 * as.h (as_howmuch): Remove declaration.
793 (fprint_value): Likewise.
794 * as.c (myname): Make static.
795 * input-scrub.c (as_howmuch): Remove.
796 (as_1_char): Likewise.
797 * input-file.h (input_file_is_open): Remove.
798 * input-file.c (input_file_is_open): Likewise.
799 * expr.h (expr_build_unary): Remove declaration.
800 (expr_build_binary): Likewise.
801 * expr.c (expr_build_unary): Remove.
802 (expr_build_binary): Likewise.
803 * hash.h (hash_replace): Remove declaration.
804 (hash_delete): Likewise.
805 * hash.c (hash_replace): Remove.
806 (hash_delete): Likewise.
807 * bignum-copy.c (bignum_copy): Move from here ..
808 * config/tc-vax.c (bignum_copy): .. to here.
809 * bignum.h (LOG_TO_BASE_2_OF_10): Remove.
810 (bignum_copy): Remove extern declaration.
811 * sb.h (string_count): Remove extern declaration.
812 (sb_build, sb_add_buffer, sb_print, sb_print_at): Likewise.
814 * sb.c (dsize): Replace preprocessor macro with static int.
815 (string_count): Make static.
816 (sb_build, sb_add_buffer, sb_print, sb_print_at): Likewise.
818 * config/obj-coff.c (dim_index): Make static.
819 * config/tc-i386.c (GOT_symbol): Likewise.
820 (output_invalid_buf): Likewise.
821 * doc/internals.texi (Warning and error messages): Remove the
822 prototype for fprint_value.
830 * config/obj-multi.h (FAKE_LABEL_NAME): Defined.
832 * read.c (pseudo_set): Disallow symbol set to common symbol.
835 * write.c (write_object_file): Report common symbol name when
836 disallowing local symbol set to common symbol.
837 (adjust_reloc_syms): Disallow local symbol set to undefined
842 * macro.c (macro_expand_body): Replace locals indicator parameters
843 with actual macro_entry. New local variables macro_line and err.
844 Don't return when encountering an error, just record the fact.
845 Detect local symbol name colliding with parameter. Track line number
846 inside of macro expansion.
847 (do_formals): Move local variable name to wider scope. Check
848 parameter of the same name doesn't already exist. In MRI mode, also
849 check it doesn't collide with the argument count pseudo-parameter).
850 (define_macro): Add file and line number parameters. Remove local
851 variable namestr. New local variable error. Initialize macro_entry
852 members file, line, and name. Don't return when encountering an
853 error, just record the fact. Use %s in some diagnostics for read.c
854 to insert the macro name. Free macro_entry on error.
855 (macro_expand): Pass macro_entry to macro_epand_body. Don't return
856 when encountering an error, just record the fact.
857 (expand_irp): Don't return when encountering an error, just record
859 * macro.h (macro_struct): New members name, file, and line.
860 (define_macro): Add file and line number parameters.
861 * read.c (s_macro): Pass file and line to define_macro. Tag warning
862 regarding pseudo-op redefinition with the file/line that macro
863 definition started at.
867 * config/tc-xtensa.c (LOOKAHEAD_ALIGNER): Delete macro.
868 (future_alignment_required): Remove ifdefs that use it.
872 * config/tc-xtensa.c (xg_get_build_instr_size): Remove.
873 (xg_is_narrow_insn, xg_expand_narrow): Remove. Merge into...
874 (xg_is_single_relaxable_insn): ...here. Add "targ" and "narrow_only"
876 (xg_assembly_relax, xg_find_narrowest_format, relaxation_requirements,
877 convert_frag_narrow): Use new version of xg_is_single_relaxable_insn.
881 * config/tc-z8k.c (md_assemble): Fix buffer overrun in operand[]
886 * config/obj-aout.h (S_IS_EXTERN): Removed.
887 * config/obj-bout.h (S_IS_EXTERN): Likewise.
888 * config/obj-coff.h (S_IS_EXTERN): Likewise.
889 * symbols.c (S_IS_EXTERN): Likewise.
890 * symbols.h (S_IS_EXTERN): Likewise.
892 * config/tc-alpha.c (tc_gen_reloc): Replace S_IS_EXTERN with
894 * config/tc-d10v.c (md_apply_fix3): Likewise.
895 * config/tc-ia64.c (ia64_fix_adjustable): Likewise.
896 * config/tc-iq2000.c (iq2000_fix_adjustable): Likewise.
897 * config/tc-m32r.c (m32r_fix_adjustable): Likewise.
898 * config/tc-mmix.c (mmix_adjust_symtab): Likewise.
899 * config/tc-sh64.c (shmedia_frob_file_before_adjust): Likewise.
900 (shmedia_md_convert_frag): Likewise.
901 * symbols.c (print_symbol_value_1): Likewise.
902 * write.c (write_object_file): Likewise.
906 * config/tc-xtensa.c (get_aligned_diff): Change type of
907 branch_align to offsetT so that its signedness matches that of
910 * config/tc-mips.c (macro): Use sprintf_vma to convert a > 32 bit
911 number into a readable string.
912 (load_register): Likewise.
916 * config/tc-ia64.c (specify_resource): Initialize all of tmpl.
920 * config/tc-sparc.c (md_assemble): If sparc_ip gives us a
921 NULL insn, exit early. Remove now spurious NULL checks.
922 (sparc_ip): Use as_bad for unknown opcode errors, set *pinsn
927 * symbols.h (symbol_find_base): Remove prototype.
928 * symbols.c (save_symbol_name): Remove code section conditional upon
930 (symbol_find): Remove.
931 (symbol_find_base): Rename to symbol_find.
932 * subsegs.c (section_symbol): Replace use of symbol_find_base with
934 * config/obj-coff.c (tag_insert): Remove code section conditional
935 upon STRIP_UNDERSCORE.
936 (obj_coff_def): Likewise.
937 (obj_coff_endef): Replace use of symbol_find_base with symbol_find.
938 (coff_frob_symbol): Likewise.
939 (yank_symbols): Likewise.
940 (c_section_symbol): Likewise.
941 * config/obj-coff.h (DO_NOT_STRIP): Remove.
942 * config/tc-arm.c (symbol_locate): Remove code section conditional
943 upon STRIP_UNDERSCORE.
944 * config/tc-h8300.h (DO_NOT_STRIP): Remove.
945 * config/tc-h8500.h (DO_NOT_STRIP): Remove.
946 * config/tc-sh.h (DO_NOT_STRIP): Remove.
947 * config/tc-w65.h (DO_NOT_STRIP): Remove.
948 * config/tc-z8k.h (DO_NOT_STRIP): Remove.
952 * config/tc-ia64.c (md_assemble): Fix error message for wrong
953 access to application registers.
957 * config/te-sysv32.h: Remove.
958 * Makefile.am (TARG_ENV_HFILES): Remove reference to
960 * Makefile.in (TARG_ENV_HFILES): Likewise.
965 * config/tc-ia64.c (ia64_canonicalize_symbol_name): Re-allow zero-
970 * config/tc-i386.c (md_begin): Allow hyphens in mnemonics.
974 * config/tc-mips.c (normalize_constant_expr): Don't check for
976 (check_absolute_expr): Only call normalize_constant_expr() if
980 * config/tc-mips.c (check_absolute_expr): Fix formatting.
984 * macro.c (free_token): New, freeing all the memory associated with a
986 (do_formals): Move initializers to ...
987 (define_macro): ... here.
988 (delete_macro): Convert passed in name to lower case. Warn when
989 purging macro that doesn't exist. Use hash_jam instead of hash_delete.
993 * config/tc-mips.c (normalize_constant_expr): Fix formatting.
997 * config/obj-elf.c (obj_elf_struct): New.
998 (elf_pseudo_table). Use it for .offset and .struct.
1003 * config/tc-xtensa.c (get_aligned_diff): Handle target_size larger
1004 than the section alignment.
1009 * config/tc-xtensa.h (struct xtensa_frag_type): Add lit_frchain field.
1010 * config/tc-xtensa.c (xg_translate_sysreg_op,
1011 xtensa_translate_old_userregs_ops,
1012 xtensa_find_unaligned_branch_targets,
1013 xtensa_find_unaligned_loops, xtensa_fix_close_loop_end_frags,
1014 relax_frag_add_nop): Support 64-bit host.
1015 (xtensa_mark_literal_pool_location, xtensa_move_literals): Use
1016 tc_frag_data lit_frchain and lit_seg fields instead of fr_var.
1020 * configure.tgt: Add support for OpenBSD/sparc ELF.
1022 * configure.tgt: Set emulation for mips-*-openbsd*.
1023 Remove broken mips-dec-openbsd* config.
1024 * configure.in: Set default ABI for mips64-*-openbsd*.
1025 * configure: Regenerate.
1029 * config/tc-mips.c (macro) [ldd_std]: Don't attempt the GP
1030 optimization for constant addresses.
1034 * as.c (main): Move parse_args before symbol_begin and frag_init
1035 so that the hash table size can be set before it is used.
1036 * hash.c: Use an unsigned long type for the size of the hash
1038 * hash.h (set_gas_hash_table_size): Update the prototype.
1042 * Makefile.am (NO_WERROR): Define. Use instead of -Wno-error.
1043 * acinclude.m4: Include ../bfd/warning.m4.
1044 * configure.in: Invoke AM_BINUTILS_WARNINGS.
1045 * Makefile.in: Regenerate.
1046 * configure: Regenerate.
1047 * doc/Makefile.in: Regenerate.
1051 * config/tc-mips.c (IS_ZEXT_32BIT_NUM): New macro.
1052 (normalize_address_expr): New function to sign-extend address
1053 offsets that fit into 32 bits in 32-bit mode.
1054 (macro_build_ldst_constoffset): Use normalize_address_expr()
1055 instead of a handcoded sequence.
1056 (load_register): Likewise. Report oversized numbers in a useful
1058 (macro) [ld_st, ldd_std]: Reject all oversized offsets, not only
1059 for constant addresses. Report oversized numbers in a useful way.
1060 (mips_ip): Use normalize_address_expr() for addresses.
1064 * config/tc-i386.c (output_insn): Handle VIA PadLock instructions
1065 similar to other instructions now that they're marked as ImmExt.
1069 * hash.c (DEFAULT_SIZE): Delete. Replace with:
1070 (gas_hash_table_size): New static variable.
1071 (set_gas_hash_table_size): New function: Records a requested size
1072 for the hash tables.
1073 (get_gas_hash_table_size): New function: Return a prime number
1074 near the requested size of the hash table.
1075 (hash_new): Use get_gas_hash_table_size.
1076 * hash.h: Add a prototype for set_gas_hash_table_size.
1077 * as.c (show_usage): Add description of new switches: --hash-size
1078 and --reduce-memory-overheads.
1079 (option_values): Add OPTION_HASH_TABLE_SIZE and
1080 OPTION_REDUCE_MEMORY_OVERHEADS.
1081 (std_longpopts): Add entries for the new options.
1082 (parse_args): Handle the new options.
1083 * Makefile.am: Add a dependency of as.c on hash.h.
1084 * Makefile.in: Regenerate.
1085 * doc/as.texinfo: Document the new switches.
1086 * NEWS: Mention the new switches.
1091 * config/tc-hppa.c (pre_defined_registers): Fix %farg[0-3]
1096 * Makefile.am: Run "make dep-am".
1097 * Makefile.in: Regenerate.
1102 * config/tc-xtensa.c (check_t1_t2_reads_and_writes): Fix typo.
1106 * configure.tgt (generic_target): Add support for OpenBSD/i386 ELF.
1110 * NEWS: Mention these changes and their effects.
1111 * macro.c (get_token): Use is_name_beginner/is_part_of_name/
1113 (check_macro): Likewise.
1114 (buffer_and_nest): Likewise. Permit multiple labels. Don't discard
1115 labels together with the closing pseudo-op.
1116 (macro_expand_body): Adjust comment. Range-check input before use.
1117 Adjust mis-spelled diagnostic. Use is_name_beginner.
1118 * read.c (try_macro): New.
1119 (read_a_source_file): New static variable last_eol. Don't list
1120 macro expansion lines more than once. Call try_macro.
1121 (s_macro): Set section of line_label to absolute instead of undefined.
1122 * doc/as.texinfo: Add information on the caveats of these changes.
1126 * symbols.c (symbol_X_add_number): Change return type to "offsetT *".
1127 * symbols.h (symbol_X_add_number): Update prototype.
1131 * symbols.c (symbol_X_add_number): Fix warning.
1135 * config/tc-m68k.c (md_begin): Support 64bit host.
1136 (get_num): Support 64bit BFD on 32bit host.
1140 * config/tc-mips.c (md_apply_fix3): Fix typos in BFD_RELOC_64.
1144 * read.c (assign_symbol): Remove const from parm. Fix xcalloc
1145 call. Don't do the COFF hacks for multi-emulation gas. Move
1146 demand_empty_rest_of_line back to caller.
1147 (s_set, equals): demand_empty_rest_of_line here.
1150 * as.h (rs_dummy): Define.
1151 * symbols.c (symbol_X_add_number): New function.
1152 * symbols.h (symbol_X_add_number): Declare.
1153 * stabs.c (aout_process_stab): Tidy symbol frag setting.
1154 * read.c (assign_symbol): New function, split out from s_set.
1155 Use symbol_find_or_make. Leave fr_type of dummy frag as rs_dummy.
1156 Fix COFF hacks for multi-emulation gas.
1157 (s_set): Call assign_symbol. Remove "register" keyword.
1158 (set_zero_frag): New function.
1159 (pseudo_set): Always check for assignment to section syms.
1160 Always set segment and frag of symbol, and likewise extern for
1161 aout/bout. Handle assignment of sym=sym+/-const specially. Don't
1162 special case exp.X_add_number non-zero for O_symbol expressions.
1163 (equals): Use assign_symbol.
1167 * config/tc-xtensa.c (xtensa_create_xproperty_segments): Skip
1172 * config/tc-i386.c (tc_gen_reloc): Don't turn
1173 BFD_RELOC_X86_64_32S into BFD_RELOC_32.
1177 * po/rw.po: New translation: Kinyarwanda
1178 * configure.in (ALL_LINGUAS): Add rw
1179 * configure: Regenerate.
1184 * config/tc-xtensa.c (branch_align_power): New.
1185 (xtensa_find_unaligned_branch_targets, get_aligned_diff,
1186 future_alignment_required): Use branch_align_power to check section
1187 alignment as well as xtensa_fetch_width when aligning branch targets.
1191 * config/tc-xtensa.c: Warning fixes throughout.
1192 (xtensa_fetch_width): Change to unsigned.
1193 (assemble_nop, xtensa_find_unaligned_branch_targets,
1194 xtensa_find_unaligned_loops, xg_assemble_vliw_tokens,
1195 is_narrow_branch_guaranteed_in_range, xtensa_fix_close_loop_end_frags,
1196 min_bytes_to_other_loop_end, unrelaxed_frag_min_size,
1197 unrelaxed_frag_max_size, xtensa_fix_short_loop_frags,
1198 count_insns_to_loop_end, unrelaxed_frag_min_insn_count,
1199 get_text_align_max_fill_size, get_text_align_nop_count,
1200 get_text_align_nth_nop_size, get_noop_aligned_address,
1201 get_aligned_diff, convert_frag_align_next_opcode,
1202 convert_frag_immed_finish_loop, xtensa_create_property_segments,
1203 xtensa_create_xproperty_segments, xt_block_aligned_size): Clean up
1204 types, avoiding size_t and using offsetT and addressT appropriately.
1205 (get_text_align_power): Clean up types. Avoid incorrect bound.
1206 (get_text_align_fill_size): Clean up types. Restructure for clarity.
1210 * config/tc-ia64.c (start_unwind_section): Undo the change
1212 (generate_unwind_image, dot_endp): Likewise.
1216 * config/tc-ia64.c (ia64_handle_align): Move le_nop and
1217 le_nop_stop arrays and initializers to file scope.
1218 (md_begin): When generating code for anything other than
1219 Itanium 1, use MMI instead of MFI NOP bundles as a filler.
1223 * config/tc-i386.c (output_imm): Also set sign flag for 64-bit push
1228 * NEWS: Mention [fdgh]_floating.
1229 * config/atof-vax.c: Fix some whitespace.
1230 * config/tc-vax.c (md_pseudo_table): Add "[fdgh]_floating".
1234 * configure.in: Add a check for <unistd.h> providing a prototype
1235 for getopt() which is compatible with the one in
1236 include/getopt.h. If so then define HAVE_DECL_GETOPT.
1237 * configure: Regenerate.
1238 * config.in (HAVE_DECL_GETOPT): Add.
1242 * config/tc-i386.c (s_bss): Call obj_elf_section_change_hook.
1246 * config/tc-i386.c (md_apply_fix3): Also handle BFD_RELOC_X86_64_32S.
1247 (tc_gen_reloc): Handle BFD_RELOC_X86_64_32S in the default case.
1251 * config/tc-arm.c (arm_adjust_symtab): Rename
1252 bfd_elf32_is_arm_mapping_symbol_name to bfd_is_arm_mapping_symbol_name.
1256 * config/tc-arm.c (mapping_state): Change documentation in function
1257 comment to cross-reference spec instead. Change type of mapping symbols
1259 (arm_adjust_symtab): Don't change type of mapping symbols here.
1263 * as.h [NEED_DECLARATION_FFS] (ffs): Prototype and alphabetize.
1264 * config.in: Regenerate.
1265 * configure: Regenerate.
1266 * configure.in: Check for ffs decl and alphabetize.
1271 * config/tc-arm.c (arm_parse_reloc): Add TLS relocations.
1272 (md_apply_fix3): Mark TLS symbols.
1273 (tc_gen_reloc): Handle TLS relocations.
1274 (arm_fix_adjustable): Ignore TLS relocations.
1275 (s_arm_elf_cons): Support expressions after decorated symbols.
1279 * config/tc-arm.c (marked_pr_dependency): New bitmap, bit N indicates
1280 whether personality routine index N has been output for this section.
1281 (mapping_state): tc_segment_info_data now struct not enum.
1282 (arm_elf_change_section): Likewise, and marked_pr_dependency is now
1283 handled on section change.
1284 (create_unwind_entry): Previous code to output dependency removed.
1285 (s_arm_unwind_fnend): Output dependency if it hasn't been done already
1287 * config/tc-arm.h (TC_SEGMENT_INFO_TYPE): Redefined as struct
1288 arm_segment_info_type.
1289 (arm_segment_info_type): New struct.
1294 * config/tc-xtensa.c (do_align_targets): Update comment.
1295 (xtensa_frob_label): Compute "freq" before possibly switching frags.
1296 Insert a LOOP_END frag before every loop target, and do not overload
1297 DESIRE_ALIGN_IF_TARGET frags with loop end information.
1298 (xg_assemble_vliw_tokens): Use do_align_targets.
1299 (xtensa_fix_target_frags): Remove code to convert a
1300 DESIRE_ALIGN_IF_TARGET frag to a LOOP_END frag when there is a
1301 negatable branch at the end of a loop.
1302 (frag_can_negate_branch): Delete.
1308 NEWS: Mention "-mtune=[itanium1|itanium2]".
1310 * config/tc-ia64.c (md): Add tune.
1311 (md_parse_option): Accepted "-mtune=[itanium1|itanium2]".
1312 (md_show_usage): Add "-mtune=[itanium1|itanium2]".
1313 (extra_goodness): Prefer M- and I-unit NOPs for itanium2. F and
1314 B unit NOPs are discouraged for McKinley-derived cores.
1315 (md_begin): Don't hardcode the "extra_goodness()" function in
1317 (ia64_init): Set md.tune to itanium2.
1319 * doc/as.texinfo: Add -mtune=[itanium1|itanium2]".
1320 * doc/c-ia64.texi: Likewise.
1324 * config/obj-coff.c (coff_frob_symbol): When crashing because of a
1325 bad C_EFCN symbol, print its name.
1329 * config/tc-xtensa.c (use_longcalls): Delete.
1330 (xg_symbolic_immeds_fit): Check for direct calls and return TRUE if
1331 the use_longcalls flag is set. Do this before checking the segment.
1332 (xg_expand_assembly_insn): Rearrange to use new do_expand flag. Never
1333 expand direct calls at this point.
1334 (xtensa_set_frag_assembly_state): Set use_longcalls flag.
1335 (xtensa_find_unmarked_state_frags): Likewise.
1336 (md_assemble): Do not disable longcalls by setting is_specific_opcode.
1337 (xg_assemble_vliw_tokens): Switch frags when use_longcalls changes.
1338 (convert_frag_immed): Remove unnecessary check of is_specific_opcode.
1339 * config/tc-xtensa.h (xtensa_frag_type): Add use_longcalls flag.
1343 * config/tc-mmix.c, config/tc-mmix.h: Convert to ISO C90.
1347 * config/tc-cris.h: Convert to ISO C90.
1348 * config/tc-cris.c: Ditto.
1349 (md_estimate_size_before_relax): Remove obsolete comment for
1350 parameter "segment_type".
1351 (md_begin): Document reason for cast of hash_insert argument.
1352 (md_atof): Correct type of parameter "type".
1356 * write.h (bit_fix_new): Remove redundant prototype.
1357 * config/atof-ieee.c: Convert to ISO-C90 and fix formatting.
1358 * config/obj-aout.c: Convert to ISO-C90 and fix formatting.
1359 * config/obj-aout.h: Convert to ISO-C90 and fix formatting.
1360 * config/obj-bout.c: Convert to ISO-C90 and fix formatting.
1361 * config/obj-bout.h: Convert to ISO-C90 and fix formatting.
1362 * config/obj-ecoff.c: Convert to ISO-C90 and fix formatting.
1363 * config/obj-ecoff.h: Convert to ISO-C90 and fix formatting.
1364 * config/obj-som.c: Convert to ISO-C90 and fix formatting.
1365 * config/obj-som.h: Convert to ISO-C90 and fix formatting.
1366 * config/tc-a29k.c: Convert to ISO-C90 and fix formatting.
1367 * config/tc-a29k.h: Convert to ISO-C90 and fix formatting.
1368 * config/tc-alpha.c: Convert to ISO-C90 and fix formatting.
1369 * config/tc-alpha.h: Convert to ISO-C90 and fix formatting.
1370 * config/tc-arc.c: Convert to ISO-C90 and fix formatting.
1371 * config/tc-arc.h: Convert to ISO-C90 and fix formatting.
1372 * config/tc-d10v.c: Convert to ISO-C90 and fix formatting.
1373 * config/tc-d10v.h: Convert to ISO-C90 and fix formatting.
1374 * config/tc-d30v.c: Convert to ISO-C90 and fix formatting.
1375 * config/tc-d30v.h: Convert to ISO-C90 and fix formatting.
1376 * config/tc-dlx.c: Convert to ISO-C90 and fix formatting.
1377 * config/tc-dlx.h: Convert to ISO-C90 and fix formatting.
1378 * config/tc-fr30.c: Convert to ISO-C90 and fix formatting.
1379 * config/tc-fr30.h: Convert to ISO-C90 and fix formatting.
1380 * config/tc-h8500.c: Convert to ISO-C90 and fix formatting.
1381 * config/tc-h8500.h: Convert to ISO-C90 and fix formatting.
1382 * config/tc-i370.c: Convert to ISO-C90 and fix formatting.
1383 * config/tc-i370.h: Convert to ISO-C90 and fix formatting.
1384 * config/tc-i960.c: Convert to ISO-C90 and fix formatting.
1385 * config/tc-ip2k.c: Convert to ISO-C90 and fix formatting.
1386 * config/tc-ip2k.h: Convert to ISO-C90 and fix formatting.
1387 * config/tc-m32r.c: Convert to ISO-C90 and fix formatting.
1388 * config/tc-m32r.h: Convert to ISO-C90 and fix formatting.
1389 * config/tc-m88k.c: Convert to ISO-C90 and fix formatting.
1390 * config/tc-m88k.h: Convert to ISO-C90 and fix formatting.
1391 * config/tc-mcore.c: Convert to ISO-C90 and fix formatting.
1392 * config/tc-mcore.h: Convert to ISO-C90 and fix formatting.
1393 * config/tc-mn10200.c: Convert to ISO-C90 and fix formatting.
1394 * config/tc-ns32k.c: Convert to ISO-C90 and fix formatting.
1395 * config/tc-ns32k.h: Convert to ISO-C90 and fix formatting.
1396 * config/tc-openrisc.c: Convert to ISO-C90 and fix formatting.
1397 * config/tc-openrisc.h: Convert to ISO-C90 and fix formatting.
1398 * config/tc-or32.c: Convert to ISO-C90 and fix formatting.
1399 * config/tc-or32.h: Convert to ISO-C90 and fix formatting.
1400 * config/tc-pdp11.h: Convert to ISO-C90 and fix formatting.
1401 * config/tc-pj.c: Convert to ISO-C90 and fix formatting.
1402 * config/tc-pj.h: Convert to ISO-C90 and fix formatting.
1403 * config/tc-tahoe.c: Convert to ISO-C90 and fix formatting.
1404 * config/tc-tic80.c: Convert to ISO-C90 and fix formatting.
1405 * config/tc-tic80.h: Convert to ISO-C90 and fix formatting.
1406 * config/tc-v850.c: Convert to ISO-C90 and fix formatting.
1407 * config/tc-v850.h: Convert to ISO-C90 and fix formatting.
1408 * config/tc-w65.c: Convert to ISO-C90 and fix formatting.
1409 * config/tc-w65.h: Convert to ISO-C90 and fix formatting.
1410 * config/tc-xstormy16.c: Convert to ISO-C90 and fix formatting.
1411 * config/tc-xstormy16.h: Convert to ISO-C90 and fix formatting.
1415 * config/tc-arm.c (arm_adjust_symtab): Fetch elf_sym's binding
1416 attributes properly.
1421 * configure.tgt: Accept any C library to accompany a GNU Linux
1422 implementation, not just the GNU C library.
1423 * configure.in: Likewise.
1424 * configure: Regenerate.
1428 * config/tc-tic30.c: Convert to ISO C90 formatting.
1429 * config/tc-tic30.h: Convert to ISO C90 formatting.
1430 * config/tc-pdp11.c: Convert to ISO C90 formatting.
1431 * config/atof-vax.c: Convert to ISO C90 formatting.
1435 * config/tc-mips.c (mips_frob_file): Sort BFD_RELOC_MIPS16_LO16
1436 relocations correctly as well.
1437 (mips_fix_adjustable): Don't make BFD_RELOC_MIPS16_LO16
1438 relocations in mergeable sections section-relative either.
1442 * config/tc-xtensa.c (md_apply_fix3): Recognize XTENSA_PLT relocations.
1446 * config/tc-sh.c (sh_elf_final_processing): Fix compile time
1447 warning by providing a prototype for sh_symbian_find_elf_flags.
1449 * cgen.c (gas_cgen_parse_operand): Fix typo introduced by
1454 * configure.tgt: Handle setting of bfd_gas for fmt=multi targets
1455 along with other formats that set bfd_gas. Remove unnecessary
1456 setting of bfd_gas. Delete strongarm cases in generic_target
1462 * config/tc-m68k.c (TRUNC, SEXT): Define.
1463 (issbyte, isubyte, issword, isuword, isbyte, isword): Use the above.
1464 (m68k_ip): Truncate or sign extend expressions as appropriate.
1465 (get_num): Likewise.
1466 (md_apply_fix3): Use SEXT.
1470 * Makefile.am (OBJ_FORMAT_CFILES): Prune config/obj-vms.c.
1471 (OBJ_FORMAT_HFILES): Prune config/obj-vms.h.
1472 (obj-vms.o): Delete rule.
1474 * Makefile.in: Regenerate.
1475 * aclocal.m4: Regenerate.
1476 * doc/Makefile.in: Regenerate.
1477 * po/POTFILES.in: Regenerate.
1481 * config/tc-m68k.c (m68k_elf_final_processing): Set file specific
1482 flag for coldfire v4e.
1486 * config/tc-xtensa.c (xg_apply_tentative_value): Rename to
1487 xg_apply_fix_value and return a value to indicate success.
1488 (md_pcrel_from): Skip check of fx_done. Return 0 if not PC-relative.
1489 (xtensa_force_relocation): Remove checks for VTABLE relocs.
1490 (xtensa_validate_fix_sub): New.
1491 (xtensa_fix_adjustable): Remove check for external or weak symbols.
1492 (tc_gen_reloc): Move code to handle difference of symbols and code to
1493 apply tentative fix values to ...
1494 (md_apply_fix3): ...here. Enable standard overflow checks for simple
1495 8, 16, and 32 bit relocations. Apply fixes for slot-specific
1496 relocations when linkrelax flag is not set.
1497 * config/tc-xtensa.h (xtensa_validate_fix_sub): Add prototype.
1498 (TC_FORCE_RELOCATION_SUB_SAME, TC_VALIDATE_FIX_SUB): Define.
1502 * config/tc-i386.c (i386_scale): Beautify error message.
1503 (Intel syntax comments): Update.
1504 (struct intel_parser_s): Add fields in_offset, in_bracket, and
1506 (intel_e04_1, intel_e05_1, intel_e05_1, intel_e09_1, intel_e10_1):
1507 Remove declarations.
1508 (intel_bracket_expr): Declare.
1509 (i386_intel_operand): Initialize new intel_parser fields. Wrap most
1510 of the function body in a loop allowing to split an operand into two.
1511 Replace calls to malloc and checks of it returning non-NULL with
1512 calls to xmalloc/xstrdup.
1513 (intel_expr): SHORT no longer handled here. Add comment indicating
1514 comparison ops need implementation.
1515 (intel_e04, intel_e04_1): Combine, replace recursion with loop.
1516 Check right operand of - does not specify a register when parsing
1517 the address of a memory reference.
1518 (intel_e05, intel_e05_1): Combine, replace recursion with loop.
1519 Check operands do not specify a register when parsing the address of
1521 (intel_e06, intel_e06_1): Likewise.
1522 (intel_e09, intel_e09_1): Combine, replace recursion with loop. Also
1523 handle SHORT as well as unary + and -. Don't accept : except for
1524 segment overrides or in direct far jump/call insns.
1525 (intel_brack_expr): New.
1526 (intel_e10, intel_e10_1): Combine, replace recursion with loop. Use
1528 (intel_e11): Replace chain of if/else-if by switch, alloing fall-
1529 through in certain cases. Use intel_brack_expr. Add new diagnostics.
1530 Allow symbolic constants as register scale value.
1531 (intel_get_token): Replace call to malloc and check of return value
1532 with call to xmalloc. Change handling for FLAT to match MASM's.
1533 (intel_putback_token): Don't try to back up/free current token if
1538 * configure.tgt: Set emulation for arm-*-eabi*.
1539 * config/tc-arm.c (meabi_flags): Check EABI_DEFAULT.
1540 * config/te-armeabi.h: New file.
1541 * config/te-armlinuxeabi.h (EABI_DEFAULT): Define.
1542 * config/te-symbian.h: Include "te-armeabi.h".
1546 * cgen.c (gas_cgen_parse_operand): Copy opinfo parameter into a
1547 local variable in case it is clobbered by the setjmp.
1552 * configure.in (werror): New switch: Add -Werror to the
1553 compiler command line. Enabled by default. Disable via
1555 * configure: Regenerate.
1559 * config/obj-coff.h: Convert to ISO C90 formatting.
1560 * config/obj-coff.c: Convert to ISO C90 formatting.
1564 * config/tc-arm.c (do_mla): Rename to do_mlas, take second
1565 is_mls parameter; do not diagnose Rm==Rd when is_mls.
1566 (do_mla, do_mls, five_bit_unsigned_immediate, bfci_lsb_and_width)
1567 (do_bfc, do_bfi, do_bfx, do_rbit, do_mov16, do_ldsttv4): New functions.
1568 (insns): Add ARMv6T2 instructions:
1569 bfc bfi mls movw movt rbit sbfx ubfx ldrht ldrsht ldrsbt strht.
1570 (arm_archs): Add V6T2 variants.
1574 * NEWS: Add cutoff for changes in 2.16 release.
1578 * expr.c (operand): Merge handling of unary + into that for unary
1583 * config/tc-mips.c: Include dw2gencfi.h.
1584 (mips_cfi_frame_initial_instructions): New.
1585 * config/tc-mips.h (TARGET_USE_CFIPOP): Define.
1586 (tc_cfi_frame_initial_instructions): Ditto.
1587 (DWARF2_DEFAULT_RETURN_COLUMN): Ditto.
1588 (DWARF2_CIE_DATA_ALIGNMENT): Ditto.
1589 * Makefile.am: Update dependencies.
1590 * Makefile.in: Regenerate.
1594 * po/es.po: Commit new Spanish translation.
1598 * po/tr.po: Commit new Turkish translation.
1602 * config/tc-arm.c (tinsns): Add ARMv6K instructions sev, wfe,
1607 * config-gas.com: Mark vax-vms as obsolete.
1608 * configure.in: Remove fmt=vms support.
1609 * config.in: Regenerate.
1610 * configure: Regenerate.
1611 * config/obj-vms.h, config/obj-vms.c, config/vms-conf.h: Remove.
1616 * config/tc-xtensa.c (finish_vinsn): Include the last instruction slot
1617 when checking if xg_resolve_literals needs to be called.
1618 * config/tc-xtensa.h: Fix spelling typo in a comment.
1622 * config/tc-tic54x.h (tic54x_macro_info): Change parameter type.
1623 * config/tc-tic54x.c (tic54x_macro_info): Likewise. Replace hand-
1624 crafted structure declarations with the types from macro.h.
1628 * config/tc-mips.c (s_cpsetup): Use '__gnu_local_gp' instead of '_gp'
1629 for -mno-shared optimization.
1633 * config/tc-mips.c (MAX_VR4130_NOPS, MAX_DELAY_NOPS): New macros.
1634 (MAX_NOPS): Bump to 4.
1635 (mips_fix_vr4130): New variable.
1636 (nops_for_vr4130): New function.
1637 (nops_for_insn): Use MAX_DELAY_NOPS rather than MAX_NOPS. Use
1638 nops_for_vr4130 if working around VR4130 errata.
1639 (OPTION_FIX_VR4130, OPTION_NO_FIX_VR4130): New macros.
1640 (md_longopts): Add -mfix-vr4130 and -mno-fix-vr4130.
1641 (md_parse_option): Handle them.
1642 (md_show_usage): Print them.
1643 * doc/c-mips.texi: Document -mfix-vr4130 and -mno-fix-vr4130.
1647 * config/tc-mips.c (append_insn): Remove cop_interlocks test from
1652 * config/tc-mips.h (mips_flush_pending_output): Delete.
1653 (mips_emit_delays): Declare.
1654 (md_flush_pending_output): Use mips_emit_delays.
1655 * config/tc-mips.c (mips_no_prev_insn): Remove parameter; always forget
1656 the previous instructions.
1657 (md_begin, append_insn, md_parse_option): Update callers.
1658 (mips_emit_delay): Remove parameter. Move INSNS != 0 code to
1660 (mips_align, s_change_sec, s_cons, s_float_cons, s_gpword)
1661 (s_gpdword): Update callers.
1662 (start_noreorder, end_noreorder): New functions.
1663 (macro, macro2, mips16_macro, s_mipsset): Use them instead of
1664 manipulating mips_opts or prev_nop_frag directly.
1665 (mips_flush_pending_output): Delete.
1669 * config/tc-mips.c (mips_move_labels): New function, taken from...
1670 (append_insn, mips_emit_delays): ...here.
1674 * config/tc-mips.c (MAX_NOPS): New macro.
1675 (history): Resize to 1 + MAX_NOPS.
1676 (fix_vr4120_class): New enumeration.
1677 (vr4120_conflicts): New variable.
1678 (init_vr4120_conflicts): New function.
1679 (md_begin): Call it.
1680 (insn_uses_reg): Constify first argument.
1681 (classify_vr4120_insn, insns_between, nops_for_insn, nops_for_sequence)
1682 (nops_for_insn_or_target): New functions.
1683 (append_insn): Use the new nops_for_* functions instead of inline
1684 delay checks. Generalize prev_nop_frag handling to handle an
1685 arbitrary history length. Insert nops into the history buffer
1686 once the number of nops in prev_nop_frag is fixed.
1687 (emit_delays): Use nops_for_insn instead of inline delay checks.
1691 * config/tc-mips.c (append_insn): Remove now-redundant nops != 0
1692 check from branch delay code. Remove unnecessary check for branches.
1696 * config/tc-mips.c (dummy_opcode): Delete.
1697 (nop_insn, mips16_nop_insn): New variables.
1698 (NOP_INSN): New macro.
1699 (insn_length, create_insn, install_insn, move_insn, add_fixed_insn)
1700 (add_relaxed_insn, insert_into_history, emit_nop): New functions.
1701 (md_begin): Initialize nop_insn and mips16_nop_insn.
1702 (append_insn): Use the new emit_nop function to add nops, recording
1703 them in the history buffer. Use add_fixed_insn or add_relaxed_insn
1704 to reserve room for the instruction and install_insn to install the
1705 final form. Use insert_into_history to record the instruction in
1706 the history buffer. Use move_insn to do delay slot filling.
1707 (mips_emit_delays): Use add_fixed_insn instead of the emit_nop macro.
1708 (macro_build, mips16_macro_build, macro_build_lui, mips_ip)
1709 (mips16_ip): Use create_insn to initialize mips_cl_insns.
1713 * config/tc-mips.c (INSERT_BITS, EXTRACT_BITS, INSERT_OPERAND)
1714 (EXTRACT_OPERAND, MIPS16_INSERT_OPERAND, MIPS16_EXTRACT_OPERAND): New.
1715 (insn_uses_reg, reg_needs_delay, append_insn, macro_build)
1716 (mips16_macro_build, macro_build_lui, mips16_macro, mips_ip)
1717 (mips16_ip): Use the new macros instead of explicit masks and shifts.
1721 * config/tc-mips.c (mips_cl_insn): Replace the valid_p, delay_slot_p
1722 and extended_p fields with a single fixed_p field.
1723 (append_insn, mips_no_prev_insn): Adjust accordingly.
1727 * config/tc-mips.c (mips_cl_insn): Replace reloc_type array with
1728 a single mips16_absolute_jump_p bit.
1729 (append_insn): Adjust accordingly.
1733 * config/tc-mips.h (mips_cl_insn): Move definition to...
1734 * config/tc-mips.c (mips_cl_insn): ...here. Add new fields:
1735 frag, where, fixp, reloc_type, valid_p, noreorder_p, delay_slot_p
1737 (history): New variable.
1738 (prev_insn, prev_prev_insn, prev_insn_valid, prev_insn_frag)
1739 (prev_insn_where, prev_insn_reloc_type, prev_insn_fixp)
1740 (prev_insn_is_delay_slot, prev_insn_unreordered, prev_insn_extended)
1741 (prev_prev_insn_unreordered): Delete.
1742 (reg_needs_delay, append_insn, mips_no_prev_insn, mips_emit_delays)
1743 (macro_start): Replace uses of prev_insn* with the equivalent history[]
1748 * doc/Makefile.am: Update as.info dependencies.
1749 * aclocal.m4, Makefile.in, doc/Makefile.in: Regenerated.
1753 * doc/as.texinfo: Add sentence to indicate redefining a macro is an
1754 error, and point to .purgem documentation if someone really needs
1756 * NEWS: Mention macro redefinition is now an error.
1760 * config/tc-ia64.c (emit_one_bundle): Track last slot user insn was
1761 emitted to. Add more precise diagnostics for non-fitting insns based
1762 on that. Eliminate now superfluous special casing of MLX. Clear out
1763 slot information when dropping an insn.
1767 * config/tc-ia64.c (parse_section_name): Rename to...
1768 (cross_section): In addition to separating the name from the rest of
1769 the arguments, also carry out the operation.
1770 (dot_xdata): Use cross_section.
1771 (dot_float_cons): Likewise.
1772 (dot_xstringer): Likewise.
1773 (dot_xdata_ua): Likewise.
1774 (dot_float_cons_ua): Likewise. Pass float_cons, not stmt_float_cons.
1778 * po/gas.pot: Regenerate.
1782 * config/tc-mips.c (macro_build_lui): Use '__gnu_local_gp'
1783 instead of '_gp' for -mno-shared optimization.
1785 (s_abicalls): Document it in the comment.
1786 (md_show_usage): Document the -mno-shared option.
1790 * config/tc-mips.c (mips_set_options): Add sym32 field.
1791 (mips_opts): Initialize it.
1792 (HAVE_32BIT_ADDRESSES): Set to true if pointers are 32 bits wide.
1793 (HAVE_64BIT_ADDRESSES): Redefine as !HAVE_32BIT_ADDRESSES.
1794 (HAVE_32BIT_SYMBOLS, HAVE_64BIT_SYMBOLS): New macros.
1795 (load_address): Use HAVE_64BIT_SYMBOLS instead of HAVE_64BIT_ADDRESSES
1796 when deciding whether to use a symbolic %highest/%higher expansion.
1797 (macro): Likewise. Remove o64/n32 linux hack. Always use
1798 ADDRESS_ADD*_INSN for address addition in the expansion of "dla"
1799 and "la". Handle constants separately from symbolic expressions in
1800 the "ld_st:" case, using 64-bit arithmetic if HAVE_64BIT_ADDRESSES
1801 and using load_register to load the high part of the address.
1802 (OPTION_MSYM32, OPTION_NO_MSYM32): New macros.
1803 (OPTION_ELF_BASE): Bump by 2.
1804 (md_longopts): Add entries for -msym32 and -mno-sym32.
1805 (md_parse_option): Handle them.
1806 (usage): Document them.
1807 (s_mipsset): Handle ".set sym32" and ".set nosym32".
1808 (s_cpload, s_cpsetup): Use HAVE_64BIT_SYMBOLS instead of
1809 HAVE_64BIT_ADDRESSES to detect 64-bit values of "_gp".
1810 * doc/c-mips.texi: Document ".set sym32", ".set nosym32",
1811 -msym32 and -mno-sym32.
1815 * config/tc-mips.c (load_address): Implement GP optimization
1816 for 64bit address space non-PIC. Fix formatting.
1817 (macro): Likewise. Simplify code.
1818 (md_parse_option): Don't bail out if -G 0 is set for PIC code.
1819 (mips_after_parse_args): Simplify code.
1823 * expr.c (operand): Remove redundant code enclosed by #ifdef
1824 RELAX_PAREN_GROUPING....#endif.
1826 * config/tc-mn10200.c (tc_gen_reloc): Handle the case where the
1827 reloc is the difference of two symbols defined in the same
1830 * config/tc-iq2000.c (line_comment_chars): Include the # character
1831 as otherwise this breaks #APP/#NO_APP processing.
1835 * config/tc-arc.c(md_assemble): Remove dead code for handling
1836 immediate indexing of ld and st .
1841 * config/tc-mips.c (percent_op): Add %tlsgd, %tlsldm, %dtprel_hi,
1842 %dtprel_lo, %tprel_hi, %tprel_lo, and %gottprel.
1843 (parse_relocation): Check for a word break after a relocation
1845 (md_apply_fix3): Handle TLS relocations, and mark thread-local
1850 * config/tc-ppc.c (ppc_fix_adjustable <ELF>): Remove bogus checks.
1854 * as.c (main): Use unlink_if_ordinary instead of unlink.
1855 * messages.c (as_fatal): Likewise.
1859 * config/tc-i386.c (build_modrm_byte): Add lock prefix for cr8...15
1861 (parse_register): Allow cr8...15 in all modes.
1865 * config/tc-i386.c (intel_e11): If not followed by T_PTR, treat T_BYTE
1866 etc. like normal symbol references (T_ID).
1870 * symbols.c (fb_label_name): Fix silly thinko in last change.
1874 * expr.c (integer_constant): Remove TARGET_WORD_SIZE hack.
1875 * config/tc-m68k.h (TARGET_WORD_SIZE): Delete.
1877 * symbols.c (fb_label_name): Allow an augend of 2 for mmix.
1882 * config/tc-arc.c (md_assemble): Initialize suffix for extension
1887 * config/obj-coff.c (fixup_segment): Delete sy_mri_common assertion.
1889 * as.h (assert): Warning fix.
1890 * expr.c (expr): Correct assertion.
1891 * read.c (s_comm_internal): Remove assertion.
1892 * write.c (relax_segment): Enable vma assertion only for BFD_ASSEMBLER.
1893 (fixup_segment): Remove assertion.
1894 * config/tc-dlx.c (machine_ip): Remove untrue assertions.
1895 (md_apply_fix3): Likewise.
1896 * config/tc-i370.c (md_begin): Correct assertion.
1897 (i370_macro): Warning fix for assertion.
1901 * configure.in (AC_C_BIGENDIAN): Invoke.
1902 * configure: Regenerate.
1903 * write.c (write_object_file <!BFD_ASSEMBLER>): Don't use sizeof
1904 host variable to set string header size.
1905 * config/obj-aout.c (obj_header_append): Don't use host structs.
1906 (obj_symbol_to_chars): Likewise.
1907 (obj_emit_strings): Likewise. Use the passed in output pointer.
1908 * config/obj-aout.h (H_GET_FILE_SIZE): Include H_GET_LINENO_SIZE.
1909 * config/obj-bout.c (obj_emit_relocations): Use md_reloc_size,
1910 not sizeof host struct.
1911 (obj_header_append, obj_symbol_to_chars): Don't use host structs.
1912 (obj_emit_strings): Likewise.
1913 * config/obj-bout.h (EXEC_BYTES_SIZE): Define.
1914 (N_TXTOFF, H_GET_FILE_SIZE, H_GET_HEADER_SIZE): Use instead of
1916 (H_SET_SYMBOL_TABLE_SIZE): Hard code sym size rather than using
1918 (host_number_to_chars): Define.
1919 * config/obj-hp300.c (hp300_header_append): Don't use sizeof
1920 host internal struct to set header sizes.
1921 * config/tc-i960.c (md_number_to_field): Warning fix.
1922 (md_ri_to_chars): Use host byte order.
1923 (get_cdisp, md_apply_fix3): Warning fix.
1924 * config/tc-m68k.c (md_assemble): Don't use sizeof host short.
1928 * doc/c-arc.texi: Update documentation about ARC's extension
1933 * config/tc-tic4x.c (tic4x_gen_to_words): Changed mail
1938 * cgen.c: Warning fixes.
1939 * config/tc-arc.c: Likewise.
1940 * config/tc-arm.c: Likewise.
1941 * config/tc-avr.c: Likewise.
1942 * config/tc-d10v.c: Likewise.
1943 * config/tc-d30v.c: Likewise.
1944 * config/tc-frv.c: Likewise.
1945 * config/tc-frv.h: Likewise.
1946 * config/tc-h8300.c: Likewise.
1947 * config/tc-h8500.c: Likewise.
1948 * config/tc-i370.c: Likewise.
1949 * config/tc-i960.c: Likewise.
1950 * config/tc-ia64.c: Likewise.
1951 * config/tc-ip2k.c: Likewise.
1952 * config/tc-m68hc11.c: Likewise.
1953 * config/tc-maxq.c: Likewise.
1954 * config/tc-mcore.c: Likewise.
1955 * config/tc-mips.c: Likewise.
1956 * config/tc-msp430.c: Likewise.
1957 * config/tc-pj.c: Likewise.
1958 * config/tc-ppc.c: Likewise.
1959 * config/tc-ppc.h: Likewise.
1960 * config/tc-s390.c: Likewise.
1961 * config/tc-sh.c: Likewise.
1962 * config/tc-sh64.c: Likewise.
1963 * config/tc-tic4x.c: Likewise.
1964 * config/tc-tic80.c: Likewise.
1965 * config/tc-v850.c: Likewise.
1966 * config/tc-vax.c: Likewise.
1967 * config/tc-w65.c: Likewise.
1968 * config/tc-xstormy16.c: Likewise.
1969 * config/tc-z8k.c: Likewise.
1973 * read.c (read_a_source_file): Reinstate TC_EQUAL_IN_INSN test.
1974 * doc/internals.texi (TC_EQUAL_IN_INSN): Reinstate.
1978 * config/tc-mips.c (struct proc): Change isym to
1979 func_sym. New member func_end_sym.
1980 (s_mips_ent): Update.
1981 (s_mips_end): Ditto. Add code to compute function size.
1985 * read.c: Warning fixes.
1986 * config/obj-elf.c: Likewise.
1990 * config/tc-mips.c (append_insn): Call dwarf2_emit_insn() before
1995 * config/obj-coff.c (obj_coff_section): Replace SEC_SHARED with
1998 * config/tc-tic54x.c (tic54x_bss): Replace SEC_BLOCK with
2000 (demand_empty_rest_of_line): Likewise.
2001 (tic54x_sblock): Likewise.
2002 (tic54x_clink): Replace with SEC_CLINK with SEC_TIC54X_CLINK.
2006 * read.c (address_bytes): New function.
2007 (TC_ADDRESS_BYTES): Default for BSD_ASSEMBLER to address_bytes.
2008 (potable): Add "dc.a".
2009 (cons_worker): Handle "dc.a".
2010 * doc/internals.texi (TC_ADDRESS_BYTES): Document.
2014 * input-file.c (input_file_open): Rearrange to avoid warning.
2018 * config/tc-hppa.h (TC_EQUAL_IN_INSN): Delete.
2019 * read.c (read_a_source_file): Remove TC_EQUAL_IN_INSN test.
2020 * doc/internals.texi (TC_EQUAL_IN_INSN): Delete.
2024 * config/tc-mips.c (macro_build_ldst_constoffset): Fail on $at
2025 uses after .set noat.
2026 (load_address): Likewise.
2027 (macro): Likewise. Don't try to avoid $at use by sacrificing
2028 the target register before it is stored, it won't work.
2032 * config/tc-ia64.c (emit_one_bundle): Stop filling a bundle if we
2033 see an instruction that specifies a template.
2037 * config/tc-openrisc.c (openrisc_relax_frag): Delete unused function.
2038 * config/tc-sparc.c (sparc_ip): Make op_exp static to silence warnings.
2039 * config/tc-tic80.c (build_insn): Init insn[1] to silence warning.
2043 * NEWS: Mention "-mhint.b=[ok|warning|error]".
2045 * config/tc-ia64.c (md): Add hint_b.
2046 (emit_one_bundle): Handle md.hint_b for "hint".
2047 (md_parse_option): Accepted "-mhint.b=[ok|warning|error]".
2048 (md_show_usage): Add "-mhint.b=[ok|warning|error]".
2049 (ia64_init): Set md.hint_b to error.
2050 (md_assemble): Handle md.hint_b for "hint.b".
2052 * doc/as.texinfo: Add "-mhint.b=[ok|warning|error]".
2053 * doc/c-ia64.texi: Likewise.
2057 * tc.h (struct relax_type, relax_typeS): Move from here..
2058 * as.h: ..to here. Make rlx_forward and rlx_backward an offsetT.
2059 * ecoff.c (ecoff_new_file): Add appfile param.
2060 * ecoff.h (ecoff_new_file): Likewise.
2061 * itbl-lex.h: New file.
2062 * itbl-lex.l: Include itbl-lex.h.
2063 * itbl-parse.y: Likewise.
2064 (insntbl_line, yyparse, yylex): Move to itbl-lex.h.
2065 * read.c (s_app_file_string): Mark appfile possibly unused.
2066 * subsegs.c (seg_not_empty_p): Make sec possibly unused.
2067 * subsegs.h (struct seg_info_trash): Delete.
2068 (seg_info): Use segment_info_type instead.
2069 * config/obj-coff.c (struct filename_list): Make filename const char *.
2070 * config/obj-ecoff.h (obj_app_file): Pass app to ecoff_new_file.
2071 * config/obj-elf.c (elf_file_symbol): Similarly.
2072 * config/tc-a29k.c (md_apply_fix3): Make val a valueT. Don't use
2074 * config/tc-arc.c (md_operand): Warning fix.
2075 * config/tc-arm.c (arm_parse_reloc): Only define when OBJ_ELF.
2076 (md_begin): Rearrange #if defined OBJ_COFF || defined OBJ_ELF.
2077 * config/tc-cris.h (TC_IMPLICIT_LCOMM_ALIGNMENT): Use do while.
2078 * config/tc-frv.c (frv_force_relocation): Warning fix.
2079 * config/tc-m68k.c (md_parse_option): Delete unused var.
2080 * config/tc-mcore.c (mylog2): Rename from log2 throughout.
2081 * config/tc-sparc.c: Likewise.
2082 (s_common): Warning fix.
2083 * config/tc-mips.c (append_insn): Use unsigned long long expressions.
2084 * config/tc-mmix.c (PUSHJSTUB_MAX, PUSHJSTUB_MIN): Define from
2086 * config/tc-s390.c (s390_insn): Delete test of unsigned >= 0.
2087 * config/tc-sh.c (sh_cfi_frame_initial_instructions,
2088 sh_regname_to_dw2regnum): Only define for OBJ_ELF.
2089 * config/tc-tic4x.c (tic4x_insert_reg): Use ISLOWER.
2090 (tic4x_do_align): Use TIC_NOP_OPCODE.
2091 * config/tc-tic4x.h (TIC_NOP_OPCODE): Rename from NOP_OPCODE.
2092 * config/tc-vax.c: Include netinet/in.h.
2093 (tc_headers_hook): Formatting.
2094 * config/tc-xstormy16.c (md_pcrel_from_section): Correct parens.
2098 * config/tc-ia64.c (ia64_parse_name): Don't advance 'name' when
2099 parsing inN, locN, outN. Set 'idx' to offset register number starts
2100 at. Don't handle numbers with leading zeroes or beyond 95. Remove
2105 * config/tc-mips.c (load_address): Fix formatting.
2106 (macro): Don't use AT if .set noat is in effect. Fix formatting.
2107 Catch macros which are unexpandable without AT. Remove duplicate
2109 (macro2): Remove duplicate zeroing of used_at.
2113 * dw2gencfi.c (output_cie, output_fde): Use DW_CFA_nop rather
2119 * config/tc-mips.c (reloc_needs_lo_p): Handle
2120 BFD_RELOC_MIPS16_HI16_S.
2121 (fixup_has_matching_lo_p): Handle BFD_RELOC_MIPS16_LO16.
2122 (append_insn): Add BFD_RELOC_MIPS16_GPREL, BFD_RELOC_MIPS16_HI16_S
2123 and BFD_RELOC_MIPS16_LO16 to relocs to suppress overflow
2125 (mips16_ip): Resolve BFD_RELOC_MIPS16_HI16_S,
2126 BFD_RELOC_MIPS16_HI16 and BFD_RELOC_MIPS16_LO16 for constants.
2127 Call my_getSmallExpression() to parse percent operators.
2128 (percent_op_match, mips_percent_op): Separate definitions.
2129 (mips16_percent_op): Define percent operators for the MIPS16 mode.
2130 (parse_relocation): Handle the MIPS16 mode using
2132 (md_apply_fix3): Handle BFD_RELOC_MIPS16_HI16,
2133 BFD_RELOC_MIPS16_HI16_S and BFD_RELOC_MIPS16_LO16.
2137 * config/tc-ia64.c (md_apply_fix3): Call ia64_gen_real_reloc_type
2138 instead of explicitly dealing with the translation; exclude
2139 relocations that are already pcrel, however.
2143 * config/tc-ia64.c: Include limits.h (if available).
2144 (gr_values[0]): Set path to INT_MAX.
2145 (dot_reg_val): Don't allow changing value of r0. Limit range of
2146 general registers at r127.
2147 (specify_resource): Default resource index is -1. Don't set resource
2148 index (in case IA64_RS_RSE) without setting the specific flag.
2149 (note_register_values): Check operand is O_constant before tracking
2150 input value of moves. Add tracking for dep.z with constant inputs.
2151 (print_dependency): Resource index of specific resource may be zero.
2152 (check_dependencies): Likewise.
2156 * config/tc-ia64.c (parse_operands): New local variables reg1, reg2,
2157 reg_class. Check operands and emit diagnostics for illegal use of
2162 * config/tc-ia64.c (ia64_gen_real_reloc_type): Define and initialize
2163 new variables type, suffix, and width. Handle
2164 BFD_RELOC_IA64_DIR(32|64)[LM]SB in FUNC_LT_FPTR_RELATIVE case.
2165 Handle BFD_RELOC_IA64_DIR64[LM]SB in FUNC_TP_RELATIVE case. Add
2166 FUNC_DTP_MODULE case. Handle BFD_RELOC_IA64_DIR32[LM]SB in
2167 FUNC_DTP_RELATIVE case. Return incoming relocation type if
2168 BFD_RELOC_IA64_IPLT[LM]SB in FUNC_IPLT_RELOC case. Generate warning
2169 if unable to translate relocation type, using the new variables.
2173 * config/tc-ia64.h (ia64_symbol_chars): Declare.
2174 (ty_symbol_chars): Define.
2175 * config/tc-ia64.c (ia64_symbol_chars): Define.
2179 * config/tc-ia64.c (ia64_parse_name): Only update next character if
2180 input_line_pointer was advanced.
2184 * config/tc-sh.c (md_apply_fix3): Add parentheses around &
2189 * config/tc-ia64.c (md_parse_option): Handle -xnone and -xdebugn.
2190 (md_show_usage): Add -xnone, -xdebugn, and -xdebugx. Relocate default
2192 (ia64_init): Set md.detect_dv.
2193 (ia64_start_line): New static variable warned. Warn only once when
2194 encountering explicit stops in automatic mode.
2195 * doc/c-ia64.texi: Describe -xnone, -xdebugn, and -xdebugx.
2196 * NEWS: Mention new default mode.
2200 * config/tc-ia64.c (dot_rot): Add comment that name strings should
2201 be freed when wiping out previous state. Canonicalize names before
2202 use. Free name string when detecting redefinition.
2203 (dot_pred_rel): Call generic expression parser to process arguments.
2204 Handle O_register case for individual predicates and O_subtract for
2206 (ia64_parse_name): Canonicalize name before looking it up in dynamic
2208 (ia64_canonicalize_symbol_name): Strip off all trailing # characters.
2209 Warn if multiple found, issue error if resulting symbol name has zero
2211 (dot_alias): Canonicalize name before use.
2215 * config/tc-ia64.c (unwind_diagnostic): Return -1 for warning
2217 (in_procedure): Return -1 for warning.
2218 (in_prologue): Likewise.
2219 (in_body): Likewise.
2223 * config/tc-ia64.c (dot_xdata): Undo the last change.
2224 (dot_float_cons): Likewise.
2225 (dot_xstringer): Likewise.
2226 (dot_xdata_ua): Likewise.
2227 (dot_float_cons_ua): Likewise.
2231 * NEWS: Mention "-munwind-check=[warning|error]".
2233 * config/tc-ia64.c (md): Add unwind_check.
2234 (unwind_diagnostic): New.
2235 (in_procedure): Call unwind_diagnostic when a directive isn't
2237 (in_prologue): Call unwind_diagnostic when a directive isn't in
2239 (in_body): Call unwind_diagnostic when a directive isn't in
2241 (dot_endp): Set md.unwind_check to error before calling
2242 in_procedure and restore it after. When the name is missing or
2243 couldn't be found, use the one from the last .proc if
2244 md.unwind_check isn't error. Warn if md.unwind_check is
2246 (md_parse_option): Handle "-munwind-check=[warning|error]".
2247 (md_show_usage): Add "-munwind-check=[warning|error]".
2248 (ia64_init): Set md.unwind_check to warning.
2250 * doc/as.texinfo: Add "-munwind-check=[none|warning|error]".
2251 * doc/c-ia64.texi: Likewise.
2255 * config/tc-ia64.h (LEX_AT): Include LEX_BEGIN_NAME.
2257 (ia64_parse_name): New third parameter.
2258 (md_parse_name): Pass third argument.
2259 * config/tc-ia64.c (pseudo_func): Placeholders use NULL as name.
2260 (md_operand): Handling of '@'-prefixed symbols moved from here...
2261 (ia64_parse_name): ...to here.
2265 * config/tc-ia64.c (md): Remove last_groups and group_idx.
2266 (errata_nop_necessary_p): Remove declaraction and definition.
2267 (emit_one_bundle): Don't call errata_nop_necessary_p. Don't
2268 update md.group_idx. Don't reset md.last_groups.
2272 * config/tc-ia64.c (parse_section_name): Handle non-quoted first
2274 (dot_xdata): Free section name after use.
2275 (dot_float_cons): Likewise.
2276 (dot_xstringer): Likewise.
2277 (dot_xdata_ua): Likewise.
2278 (dot_float_cons_ua): Likewise.
2279 (md_pseudo_table): Add xdata16 and xdata16.ua.
2283 * doc/all.texi: Add IA64.
2284 * doc/as.texinfo: Likewise.
2286 * doc/c-ia64.texi: Fix typos.
2290 * config/tc-arm.c (do_t_ldmstm): Change BFD_RELOC_NONE to
2292 (do_t_push_pop): Likewise.
2293 (md_assemble): Likewise.
2294 (md_apply_fix3): Handle BFD_RELOC_NONE correctly, make
2295 BFD_RELOC_UNUSED same as previous meaning of BFD_RELOC_NONE.
2296 (create_unwind_entry): Output dependency on the required personality
2301 * as.h (seg_not_empty_p): Return int, not bfd_boolean.
2302 * subsegs.c (seg_not_empty_p): Likewise.
2306 * config/tc-maxq.c (md_estimate_size_before_relax): Correct the
2307 relative jump calculation.
2308 <md_convert_frag) : Likewise.
2309 <output_disp): Likewise.
2313 * write.c (write_object_file): Recognize warning-symbol construct
2314 and skip object- and target- handling for the second symbol.
2318 * config/tc-ia64.c (dot_pred_rel): Update comment. Handle @-prefixed
2319 designators along with quoted ones. Free copy of quoted designator
2324 * config/atof-ieee.c, config/obj-coff.c, config/obj-elf.c,
2325 config/obj-ieee.c, config/obj-som.c, config/obj-vms.c,
2326 config/tc-a29k.c, config/tc-alpha.c, config/tc-arc.c,
2327 config/tc-arm.c, config/tc-d30v.c, config/tc-dlx.c,
2328 config/tc-fr30.c, config/tc-h8300.c, config/tc-h8500.c,
2329 config/tc-i370.c, config/tc-i386.c, config/tc-i960.c,
2330 config/tc-ia64.c, config/tc-m32r.c, config/tc-m32r.h,
2331 config/tc-m68hc11.c, config/tc-m68hc11.h, config/tc-mips.c,
2332 config/tc-mn10200.c, config/tc-msp430.c, config/tc-ns32k.c,
2333 config/tc-openrisc.c, config/tc-or32.c, config/tc-pdp11.c,
2334 config/tc-pj.c, config/tc-sparc.h, config/tc-tic54x.c,
2335 config/tc-tic80.c, config/tc-v850.c, config/tc-w65.c,
2336 config/tc-xtensa.c, config/tc-z8k.c, config/xtensa-relax.c: Remove
2337 #if 0'd code throughout.
2341 * as.c (parse_args): Bump copyright date reported by --version to
2346 * configure.tgt: Remove obsolete targets m68k-lynxos, sparc-lynxos
2351 * macro.c (buffer_and_nest): Allow 'from' being NULL; handle anything
2352 that can end with .endr in that case. Make requiring/permitting
2353 pseudo-ops without leading dot closer to the logic in read.c serving
2355 (expand_irp): Don't pass a mnemonic to buffer_and_nest as it will be
2360 * macro.c (do_formals): Adjust to no longer accept empty parameter
2362 (define_macro): Adjust to no longer accept empty macro name, garbage
2363 following the parameters, or macros that were previously defined.
2364 * read.c (s_bad_end): Declare.
2365 (potable): Add endm. Handler for endr and endm is s_bad_end.
2366 (s_bad_end): Rename from s_bad_endr. Adjust to handle both .endm
2368 * read.h (s_bad_endr): Remove.
2372 * config/tc-ia64.c (parse_operands): Parse all specified operands,
2373 immediately discarding (but counting) those exceeding the maximum
2374 possible amount. Track whether output and input operand counts ever
2375 matched, and use this to better indicate which of the operands/
2376 operand types was wrong; specifically don't default to pointing to
2381 * config/tc-ia64.c (unwind): Remove proc_end (now an automatic
2382 variable in dot_endp). Add body and insn. Make prologue,
2383 prologue_mask, body, and insn bitfields.
2384 (fixup_unw_records): Remove spurious new-lines from end of diagnostic
2386 (in_procedure, in_prologue, in_body): New.
2387 (dot_fframe, dot_vframe, dot_vframesp, dot_vframepsp, dot_save,
2388 dot_restore, dot_restorereg, dot_restorereg_p, dot_handlerdata,
2389 dot_unwentry, dot_altrp, dot_savemem, dot_saveg, dot_savef, dot_saveb,
2390 dot_savegf, dot_spill, dot_spillreg, dot_spillmem, dot_spillreg_p,
2391 dot_spillmem_p, dot_label_state, dot_copy_state, dot_unwabi,
2392 dot_personality): Use the appropriate one of the above.
2393 (dot_proc): Clear unwind.proc_start; set to current location only if
2394 none of the entry points were valid. Check for non-zero-length entry
2395 point names. Check that entry points aren't defined, yet. Clear
2396 unwind.prologue, unwind.body, and unwind.insn.
2397 (dot_body): Call in_procedure. Check that first directive in procedure
2398 had no insns emitted before. Set unwind.body.
2399 (dot_prologue): Call in_procedure. Check that not already in prologue.
2400 Check that first directive in procedure had no insns emitted before.
2402 (dot_endp): Call in_procedure. Declare proc_end. Check for non-zero-
2403 length entry point names. Check that entry points became defined.
2404 (md_assemble): Set unwind.insn once unwind.proc_start is defined.
2408 * config/tc-ia64.c (emit_one_bundle): Snapshot manual bundling state
2409 before actually using it. Don't generate an error in manual bundling
2410 mode when looking at an insn requiring slot 2 but not yet at slot 2.
2411 Don't generate an error in manual bundling mode when looking at an
2412 insn required to be last in its group but the required slot hasn't
2413 been reached, yet. Allow conversion from MII to MI;I for bundle
2414 consisting of only 2 insns with the stop between them. Suppress
2415 various meaningless errors resulting from detecting earlier ones.
2419 * config/tc-ia64.c (parse_operands): Also handle alloc without first
2424 * config/tc-z8k.c (md_assemble): Improve error detection.
2428 * config/tc-ia64.c (ia64_estimate_size_before_relax): Allocate space
2429 for personality routine pointer only if there is one.
2430 (ia64_convert_frag): Likewise.
2431 (generate_unwind_image): Likewise.
2435 * config/tc-z8k.c (INSERT): Remove, not used anywhere.
2436 (md_apply_fix3): Make relative branches out of range an error
2437 instead of a warning. Display correct line number for out of
2438 range branches/calls/memory accesses.
2442 * dwarf2dbg.c (dwarf2_finish): Correct logic for determining when
2443 to emit .debug_line and other debug sections.
2444 * as.h (seg_not_empty_p): Declare.
2445 * subsegs.c (seg_not_empty_p): New predicate.
2449 * configure: Regenerate to track ../gettext.m4 change.
2453 * config/tc-ia64.c (emit_one_bundle): Change "?imbf??" to "?ibmfxx".
2457 * config/tc-ia64.c (emit_one_bundle): Add late resolution of move
2458 to/from application registers dynamic insns.
2459 (md_assemble): Defer resolution of move to/from application registers
2460 dynamic insns when they can be issued on either the I- or M-units.
2464 * config/tc-frv.c (md_apply_fix3): Mark TLS symbols as such.
2466 * config/tc-frv.c (frv_pic_ptr): Add tlsmoff support.
2468 * cgen.c (gas_cgen_parse_operand): Handle
2469 CGEN_PARSE_OPERAND_SYMBOLIC.
2470 * config/tc-frv.c (md_cgen_lookup_reloc): Handle TLS relocations.
2471 (frv_force_relocation): Likewise. Fix handling of PIC
2473 (md_apply_fix3): Likewise.
2477 * as.h: Remove #if 0'd code.
2478 * atof-generic.c (atof_generic): Likewise.
2479 * ecoff.c (ecoff_directive_frame): Likewise.
2480 * frags.h (FRAG_APPEND_1_CHAR): Likewise.
2481 * itbl-ops.c (itbl_add_reg): Likewise.
2482 * listing.c (calc_hex): Likewise.
2483 * read.c (MASK_CHAR): Likewise.
2484 * subsegs.c (subsegs_print_statistics): Likewise.
2485 * symbols.c (indent): Likewise.
2486 * write.c (write_relocs): Likewise.
2487 (write_object_file): Likewise.
2488 (relax_frag): Likewise.
2492 * as.c (std_longopts): Add an entry for "--a" in order to prevent
2493 getopt_long_only() from considering -a as an abbreviation for
2495 (parse_args): Fix the parsing of -a=<file>.
2500 * read.c (s_incbin): Adjust default count for skip. Check validity
2501 of count and skip rigorously.
2505 * config/tc-mips.c (dummy_opcode): Add init for new struct member.
2509 * read.c (convert_to_bignum): New function, split out from...
2510 (emit_expr): ...here. Handle the case where X_add_number is
2511 positive and the input value is negative.
2512 (output_big_sleb128): Fix setting of continuation bit. Check whether
2513 the final byte needs to be sign-extended. Fix size-shrinking loop.
2514 (emit_leb128_expr): When generating a signed leb128, see whether the
2515 sign of an O_constant's X_add_number matches the sign of the input
2516 value. Use a bignum if not.
2520 * tc-sh.c (md_begin,md_parse_option): Change arch_sh1_up to
2521 arch_sh_up in order to match the external name and make the
2522 testsuite's job easier.
2527 * config/tc-i386.c (i386_scale): Disallow 0 scale.
2531 * config/tc-iq2000.c (s_iq2000_set): Fix thinko parsing
2532 ignored_arguments array.
2536 * write.c (write_object_file): Disallow a symbol equated to
2541 * tc-maxq.c: Replace constants 10 and 20 with bfd_mach_maxq10 and
2543 (md_pseudo_table): Add new pseudo ops for maxq10 and maxq20.
2544 (maxq_target): New function: Set the machine type.
2548 * config/tc-arm.c (FPU_DEFAULT): Define for TE_VXWORKS.
2549 (md_begin): Handle TE_VXWORKS for FP defaults.
2550 (md_apply_fix3): Correct rela offsets.
2551 (elf32_arm_target_format): Add VxWorks targets.
2555 * configure.tgt: Set em=vxworks for *-*-vxworks.
2556 * config/te-vxworks.h: New File.
2560 * config/tc-arm.c (arm_cpus): Correct arch field for arm1026ej-s.
2564 * config/tc-msp430.c (md_apply_fix3): Fix offset calculation for
2569 * config/tc-ia64.c (md): Add member "loc_directive_seen".
2570 (dot_loc): New function.
2571 (md_pseudo_table): Add entry to map .loc to dot_loc().
2572 (emit_one_bundle): Only call dwarf2_gen_line_info() if we have
2573 seen a .loc directive or we're generating DWARF2 debug info for
2579 * read.c (s_comm_internal): Don't zero end of name until size
2580 expression has been parsed.
2584 * config/tc-avr.c (mcu_types): Move attiny{13,2313} from avr4 to avr2.
2588 * config/tc-crx.c: Support 'bcop' relaxation (dealt as in 'cmp&branch'
2593 * configure.tgt: New.
2594 * configure.in: Move setting of cpu_type, fmt, etc., to
2596 * Makefile.am (CONFIG_STATUS_DEPENDENCIES): Add
2597 $(srcdir)/configure.tgt.
2598 * configure, Makefile.in: Rebuild.
2602 * config/tc-avr.c: Add support for the new R_AVR_LDI, R_AVR_6 and
2603 R_AVR_6_ADIW relocs for the LDI, ADIW/SBIW and LDD/STD
2605 (avr_offset_expression): New function to parse offsets for LDI
2607 (avr_operand): Use it.
2608 (md_apply_fix3): Generate the relocs.
2612 * config/tc-sh64.c (shmedia_md_apply_fix3): Add missing
2613 BFD_RELOC_SH_IMMS10BY8 relocation.
2615 * config/tc-sh64.c (shmedia_build_Mytes): Emit an error message rather
2616 than just ignoring bad code.
2620 * config/tc-v850.c (handle_lo16): New function.
2621 (v850_reloc_prefix): Use it to check lo().
2622 (md_assemble, md_apply_fix3): Handle BFD_RELOC_V850_LO16_SPLIT_OFFSET.
2626 * configure.in: Add s390x-ibm-tpf support.
2627 * configure: Regenerate.
2631 * config/obj-elf.c (obj_elf_change_section): Only set type and
2632 attributes on new sections. Emit warning when type of re-declared
2633 section doesn't match.
2637 * dw2gencfi.c (dot.cfi.startproc): Clear cur_cfa_offset so
2638 '.cfi_startproc simple' doesn't inherit the old value.
2642 * dw2gencfi.c (output_cfi_insn): Adjust DW_CFA_def_cfa_sf generation
2643 to emit a signed and factored offset. Adjust DW_CFA_def_cfa_offset_sf
2644 generation to emit a factored offset.
2648 * config/tc-mips.c (macro) [M_LA_AB]: Give an error for a offset
2649 which is too large in the case of NO_PIC without 64-bit
2652 * config/tc-mips.c (mips_in_shared): New static variable.
2653 (macro_build_lui): Permit "_gp" if !mips_in_shared.
2654 (md_longopts): Add -mshared and -mno-shared.
2655 (md_parse_option): Handle OPTION_MSHARED and OPTION_MNO_SHARED.
2656 (s_cpload): Implement !mips_in_shared case.
2657 (s_cpsetup): Likewise.
2658 * doc/c-mips.texi (MIPS Opts): Document -mno-shared.
2659 * NEWS: Mention -mno-shared.
2663 * config/tc-arm.c (s_arm_unwind_fnend): Use R_ARM_PREL31 relocation
2668 * config/tc-mips.c (append_insn): If we emit a nop during a relax
2669 sequence, increase the size of the sequence.
2671 * config/tc-mips.c (mips_cpu_info_table): Change "9000" entry to
2676 * read.c (s_align): Use an align_limit temporary to allay a GCC
2677 signed/unsigned comparison warning.
2681 * Makefile.am (TARG_ENV_HFILES): Add te-armlinuxeabi.h.
2682 * configure.in: Use it for arm*-*-linux-gnueabi*.
2683 * config/tc-arm.c: Allow emulation file to set FPU_DEFAULT.
2684 * config/te-armlinuxeabi.h: New file.
2685 * Makefile.in: Regenerated.
2686 * aclocal.m4: Likewise.
2687 * configure: Likewise.
2688 * doc/Makefile.in: Regenerated.
2692 * config/tc-xtensa.c (xtensa_switch_section_emit_state): Use subseg_set.
2693 (xtensa_restore_emit_state): Likewise.
2697 * read.c (ALIGN_LIMIT): Define, increasing limit for BFD_ASSEMBLER.
2702 * Makefile.am: Change LOCALEDIR to $(datadir)/share.
2703 * Makefile.in: Regenerate.
2707 * config/tc-crx.c: Major code cleanup. Remove unused variables and
2708 functions, give functions a meaningful name, add comments.
2709 (check_range): New function - Replace operand size calculation
2710 with range checking.
2711 (assemble_insn): Update Algorithm, improve error issuing.
2713 (process_label_constant): Bug fix regarding COP_BRANCH_INS relocation
2718 * config/tc-m32r.c (md_pcrel_from_section): Fixed a pcrel relocte
2719 miss between different section in the same module.
2720 (tc_gen_reloc): Likewise.
2724 * gas/config/tc-avr.c (mcu_types): Add support for atmega165,
2725 atmega325, atmega3250, atmega645 and atmega6450.
2729 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
2730 permissible for the selected instruction suffix.
2731 (process_suffix): For DefaultSize instructions, suppressing the
2732 guessing of a 'q' suffix if the instruction doesn't support it is
2733 pointless, because only an 'l' suffix can be guessed in this place.
2737 * config/tc-iq2000.c: Remove support for IQ10 processor.
2738 Convert to ISO C90 formatting.
2739 * config/tc-iq2000.h: Likewise.
2743 * config/tc-mn10300.c (md_relax_table): More fixes to the offsets
2744 in this table. They should be correct now.
2748 * config/tc-i386.h (CpuMMX2): Declare. Artificial classifier to
2749 indicate the MMX extensions added by both SSE and 3DNow!A.
2750 (Cpu3dnowA): Declare.
2751 (CpuUnknownFlags): Update.
2752 * config/tc-i386.c (cpu_sub_arch_name): Declare.
2753 (cpu_arch): i586 and pentium do not imply MMX. i686 and pentiumpro do
2754 neither imply SSE nor MMX. k6 implies MMX. k6_2 additionally implies
2755 3DNow!. Athlon additionally implies 3DNow!A. Several new
2756 entries (those starting with a dot are for sub-arch specification).
2757 (set_cpu_arch): Handle sub-arch specifications.
2758 (parse_insn): Distinguish between instructions not supported because
2759 of insufficient CPU features and because of 64-bit mode.
2760 * doc/c-i386.texi: Describe enhanced .arch directive.
2764 * config/tc-xtensa.c (xg_add_opcode_fix): Set fx_no_overflow.
2768 * dwarf2dbg.c (dwarf2_finish): Don't write a .debug_line section
2769 without a corresponding .debug_info section.
2773 * read.c (potable): Add "error" and "warning".
2774 (s_errwarn): New function.
2775 * read.h (s_errwarn): Declare.
2776 * doc/as.texinfo (Error, Warning): Document .error and .warning.
2780 * config/tc-tic54x.c (tic54x_adjust_symtab): Adjust call to
2785 * config/tc-msp430.c (struct rcodes_s, MSP430_RLC, msp430_rcodes,
2786 struct hcodes_s, msp430_hcodes): From include/opcode/msp430.h.
2790 * config/obj-coff.c (c_dot_file_symbol): Add "app" param.
2791 (coff_adjust_symtab): Adjust call.
2792 (crawl_symbols): Likewise.
2793 * config/obj-coff.h (c_dot_file_symbol): Add "app" param.
2794 (obj_app_file): Adjust.
2798 * configure.in: Enable bfd_assember for the MAXQ port.
2799 * configure: Regenerate.
2804 * config/tc-xtensa.c (finish_vinsn): Clear pending instruction if
2805 there is a conflict.
2806 (check_t1_t2_reads_and_writes): Check for both reads and writes to
2807 interfaces that are related as determined by xtensa_interface_class_id.
2811 * config/tc-mn10300.c (md_relax_table): Fix off by one negative
2812 offsets for conditional branches.
2816 * config/tc-xtensa.c (MAX_IMMED6): Change value to 65.
2820 * config/tc-xtensa.c (update_next_frag_state): Always add a NOP if
2821 relaxing at the end of a loop. Don't mark frags as UNREACHABLE or
2823 (relax_frag_immed): Update call to update_next_frag_state.
2827 * obj.h (struct format_ops <app_file>): Add int param.
2828 * read.h (s_app_file_string): Likewise.
2829 * read.c (s_app_file_string): Likewise.
2830 (s_app_file): Adjust s_app_file_string call.
2831 * config/tc-mips.c (s_mips_file): Likewise.
2832 * config/obj-coff.h (obj_app_file): Add app param.
2833 * config/obj-ecoff.h (obj_app_file): Likewise.
2834 * config/obj-multi.h (obj_app_file): Likewise.
2835 * config/obj-elf.h (elf_file_symbol): Likewise.
2836 * config/obj-elf.c (elf_file_symbol): Only emit one file symbol
2837 if called for # preprocessor lines.
2842 * symbols.c (resolve_symbol_value): Convert weak symbols only
2844 (symbol_equated_reloc_p): Don't equate weaks when relocating
2845 only for Windows PECOFF.
2849 * config/tc-crx.c (print_insn): Check and set insn_addr.
2850 * config/tc-crx.h (md_frag_check): Define.
2855 * configure.in: Add support for new target maxq-coff.
2856 * configure: Regenerate.
2857 * NEWS: Mention new support.
2858 * config/tc-maxq.c: New file.
2859 * config/tc-maxq.h: New file.
2860 * config/obj-coff.h: Add support for maxq-coff.
2864 * symbols.c (any_external_name): Define.
2865 (resolve_symbol_value): Convert weak symbols.
2866 (S_SET_EXTERNAL): Support any_external_name.
2867 (S_SET_NAME): Qualify parameter const.
2868 (symbol_equated_reloc_p): Equate to weaks when relocating.
2869 * symbols.h (S_SET_NAME): Qualfiy parameter const.
2870 * tc.h (any_external_name): Declare.
2871 * config/obj-coff.c ("coff/pe.h"): Include for BFD
2873 (weak_is_altname): Declare and define.
2874 (weak_name2altname): Same.
2875 (weak_altname2name): Same.
2876 (weak_uniquify): Same.
2877 (weak_altprefix): Define.
2878 (obj_coff_weak): Change .weak syntax and handling.
2879 (coff_frob_symbol): Fix PE weak symbol alternates.
2880 * config/obj-coff.h (USE_UNIQUE): Define.
2881 * config/tc-i386.c (md_apply_fix3): Assume weak symbols
2882 are in another segment.
2883 (tc_gen_reloc): Remove broken addend hack.
2884 doc/as.texinfo: Update.
2888 * config/tc-xtensa.c (total_frag_text_expansion): New.
2889 (md_estimate_size_before_relax): Use it.
2890 (find_address_of_next_align_frag): Likewise.
2894 * config/tc-crx.c: Rename argument types.
2895 (processing_arg_number): Rename to 'cur_arg_num'.
2896 (get_number_of_bits): Rename to 'set_operand_size'.
2897 (get_operandtype): Rename to 'parse_operand', totally rewrite.
2898 (set_cons_rparams): Rename to 'set_operand', totally rewrite.
2899 (set_indexmode_parameters): Remove function, integrate its code into
2901 (set_operand_size): Get rid of 'Operand Number' function parameter -
2902 use global variable 'cur_arg_num' instead.
2903 Use a local 'argument' pointer to reference the current argument.
2904 (parse_operand): Likewise.
2905 (set_operand): Likewise.
2906 (process_label_constant): Likewise.
2910 * config/tc-xtensa.c: Remove XTENSA_SECTION_RENAME ifdefs.
2911 (add_section_rename): Delete. Inlined into...
2912 (build_section_rename): ...here. Use xstrdup instead of strdup.
2913 (xtensa_section_rename): Drop "const" from argument and return types.
2914 (md_show_usage): Indent to match show_usage().
2915 * config/tc-xtensa.h: Remove XTENSA_SECTION_RENAME ifdefs.
2916 (tc_canonicalize_section_name): Define.
2917 (md_elf_section_rename): Remove unused macro.
2918 * doc/as.texinfo (Overview): Document Xtensa --rename-section option.
2919 * doc/c-xtensa.texi (Xtensa Options): Likewise.
2920 (Frame Directive): Delete.
2924 * configure.in: Remove arm-*-oabi and thumb-*-oabi.
2925 * config/tc-arm.c (target_oabi): Delete.
2926 (md_apply_fix3, elf32_arm_target_format): Remove target_oabi checks.
2927 (arm_opts): Remove moabi.
2928 * doc/as.texinfo (Overview): Remove documentation of -moabi.
2929 * doc/c-arm.texi (ARM Options): Likewise.
2930 * configure: Regenerated.
2934 * configure.in (crisv32): Recognize. AC_DEFINE_UNQUOTED
2935 DEFAULT_CRIS_ARCH. Handle crisv32-*-linux-gnu* like
2936 cris-*-linux-gnu* and crisv32-*-* like cris-*-*.
2937 * configure: Regenerate.
2938 * config/tc-cris.c (enum cris_archs): New.
2939 (cris_mach, cris_arch_from_string, s_cris_arch, get_sup_reg)
2940 (cris_insn_ver_valid_for_arch): New functions.
2941 (DEFAULT_CRIS_ARCH): New macro, default to cris_any_v0_v10.
2942 (cris_arch): New variable.
2943 (md_pseudo_table): New pseudo .arch.
2944 (err_for_dangerous_mul_placement): Initialize according to
2946 (STATE_COND_BRANCH): Renamed from STATE_CONDITIONAL_BRANCH.
2948 (STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON)
2949 (STATE_ABS_BRANCH_V32, STATE_LAPC, BRANCH_BF_V32, BRANCH_BB_V32)
2950 (BRANCH_WF_V32, BRANCH_WB_V32): New.
2951 (BRANCH_BF, BRANCH_BB, BRANCH_WF, BRANCH_WB): Don't undef after
2952 use in md_cris_relax_table.
2953 (md_cris_relax_table): Add entries for STATE_COND_BRANCH_V32,
2954 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
2955 Update and improve head comment.
2956 (OPTION_PIC): Define in terms of previous option, OPTION_US.
2957 (OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): Similar.
2959 (md_longopts): New option --march=...
2960 (cris_any_v0_v10_long_jump_size, crisv32_long_jump_size): New
2962 (md_long_jump_size): Initialize in terms of DEFAULT_CRIS_ARCH.
2963 (HANDLE_RELAXABLE): New macro.
2964 (md_estimate_size_before_relax): Use HANDLE_RELAXABLE for common
2965 cases. Check for weak symbols and assume not relaxable. Handle
2966 STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON,
2967 STATE_ABS_BRANCH_V32, STATE_LAPC. Use new variable symbolP, not
2969 (md_convert_frag): Handle STATE_COND_BRANCH_V32,
2970 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
2971 (cris_create_short_jump): Adjust for CRISv32.
2972 (cris_relax_frag): Handle new states.
2973 (md_create_long_jump): Ditto. Emit error for common_v10_v32.
2974 (md_begin): Define symbols "..asm.arch.cris.v32",
2975 "..asm.arch.cris.v10", "..asm.arch.cris.common_v10_v32" and
2976 "..asm.arch.cris.any_v0_v10". Use cris_insn_ver_valid_for_arch
2977 when entering opcode table entry points.
2978 (md_assemble): Adjust branch handling for CRISv32. Handle LAPC
2979 relaxation. In fix_new_exp call for main insn, pass 1 for pcrel
2980 parameter for 8, 16 and 32-bit pc-relative insns and LAPC.
2981 (cris_process_instruction): Initialize out_insnp->insn_type to
2982 CRIS_INSN_NONE, not CRIS_INSN_NORMAL.
2983 <case ']', '[', 'A', 'd', 'Q', 'N', 'n', 'Y', 'U', 'u', 'T'>: New
2985 <case 'm'>: Check that modified_char == '.'.
2986 <invalid operands>: Consume the rest of the line.
2987 When operands don't match, skip over subsequent insns with
2988 non-matching version specifier but same mnemonic.
2989 <immediate constant, case SIZE_SPEC_REG>: Immediate operands for
2990 special registers in CRISv32 are always 32 bit long.
2991 <immediate constant, case SIZE_FIELD_SIGNED, SIZE_FIELD_UNSIGNED>:
2993 (get_gen_reg): Only recognize "PC" when followed by "+]" for v32
2994 and compatible. Recognize "ACR" for v32, unless followed by "+".
2995 (get_spec_reg): Consider cris_arch when looking up register.
2996 (get_autoinc_prefix_or_indir_op): Don't recognize assignment for
2998 (get_3op_or_dip_prefix_op): Check for ']' after seeing '[rN+'.
2999 (cris_get_expression): Restore input_line_pointer if failing "early".
3000 (get_flags): Consider cris_arch and recognize flags accordingly.
3001 (branch_disp): Adjust for CRISv32.
3002 (gen_cond_branch_32): Similar. Emit error for common_v10_v32.
3003 (cris_number_to_imm): Use as_bad_where, not as_bad. Remove
3004 related FIXME. Don't insist on BFD_RELOC_32_PCREL fixup to be
3005 resolved. Don't enter zeros in object file for
3007 <case BFD_RELOC_CRIS_LAPCQ_OFFSET, BFD_RELOC_CRIS_SIGNED_16>
3008 <case BFD_RELOC_CRIS_SIGNED_8>: New case.
3009 (md_parse_option): Break out "return 1".
3010 <OPTION_ARCH> New case.
3011 (tc_gen_reloc): <case BFD_RELOC_CRIS_LAPCQ_OFFSET>
3012 <case BFD_RELOC_CRIS_SIGNED_16, BFD_RELOC_CRIS_SIGNED_8>
3013 <case BFD_RELOC_CRIS_UNSIGNED_8, BFD_RELOC_CRIS_UNSIGNED_16>
3014 <case BFD_RELOC_32_PCREL>: New cases.
3015 Addends for non-zero fx_pcrel are too in fx_offset.
3016 (md_show_usage): Show --march=<arch>.
3017 (md_apply_fix3): Adjust val for BFD_RELOC_CRIS_LAPCQ_OFFSET.
3018 (md_pcrel_from): BFD_RELOC_CRIS_LAPCQ_OFFSET is PC-relative too.
3019 (s_syntax) <struct syntaxes>: Properly constify member operand.
3020 * config/tc-cris.h (TARGET_MACH): Define.
3021 (cris_mach): Declare.
3022 * doc/as.texinfo (Overview) <CRIS>: Add --march=...
3023 * doc/c-cris.texi (CRIS-Symbols): New node for built-in symbols.
3024 (CRIS-Opts): Document --march=...
3025 (CRIS-Pseudos): Document .arch.
3029 * config/tc-i386.c (set_intel_syntax): Allow % in symbol names when
3030 intel syntax and no register prefix, allow $ in symbol names when
3032 (set_16bit_gcc_code_flag): Replace literal 'l' by LONG_MNEM_SUFFIX.
3033 (intel_float_operand): Add fourth return value indicating math control
3034 operations. Make classification more precise.
3035 (md_assemble): Complain if memory operand of mov[sz]x has no size
3037 (parse_insn): Translate word operands to floating point instructions
3038 operating on integers as well as control instructions to short ones
3039 as expected by AT&T syntax. Translate 'd' suffix to short one only for
3040 floating point instructions operating on non-integer operands.
3041 (match_template): Remove fldcw special case. Adjust q-suffix handling
3042 to permit it on fild/fistp/fisttp in AT&T mode.
3043 (process_suffix): Don't guess DefaultSize insns' suffix from
3044 stackop_size for certain floating point control instructions. Guess
3045 suffix for branch and [ls][gi]dt based on flag_code. Split error
3046 messages for Intel and AT&T syntax, and make the condition more strict
3047 for the former. Adjust suppressing of generation of operand size
3049 (intel parser): Allow the full set of MASM operators. Add FWORD, TBYTE,
3050 OWORD, and XMMWORD operand size specifiers (TBYTE replaces XWORD). Add
3051 more error checking.
3052 * config/tc-i386.h (BYTE_PTR WORD_PTR DWORD_PTR QWORD_PTR XWORD_PTR
3053 SHORT OFFSET_FLAT FLAT NONE_FOUND): Remove unused defines.
3057 * symbols.c (colon) [!WORKING_DOT_WORD]: Don't declare
3058 md_short_jump_size, md_long_jump_size.
3059 * write.c [!WORKING_DOT_WORD]: Ditto.
3060 * tc.h [!WORKING_DOT_WORD]: Declare them here. Drop const
3062 * config/tc-cris.h (md_short_jump_size, md_long_jump_size): Don't
3064 * config/tc-cris.c (md_short_jump_size, md_long_jump_size): Drop
3065 const qualifier in these definitions.
3066 * config/tc-i370.c, config/tc-m68k.c, config/tc-pdp11.c,
3067 config/tc-s390.c, config/tc-tahoe.c, config/tc-vax.c: Ditto.
3071 * dwarf2dbg.c (dwarf2_finish): Check for the existence of a file
3072 table before deciding to produce a .debug_line section to match up
3073 with a user provided .debug_info section.
3077 * config/tc-crx.c (getreg_image): Bug fix, a return value was
3078 mistakenly omitted from CRX_C_REGTYPE and CRX_CS_REGTYPE cases.
3082 * config/tc-crx.c: Remove global variable 'post_inc_mode'.
3083 (get_flags): New function.
3084 (get_number_of_bits): Edit comments, update numeric values to
3086 (process_label_constant): Don't support the colon
3087 format (SYMBOL:[s|m|l]).
3088 (set_cons_rparams): Support argument type 'arg_rbase'.
3089 (get_operandtype): Bug fix in 'rbase' operand type parsing.
3090 (handle_LoadStor): Bug fix, first handle post-increment mode.
3091 (getreg_image): Remove redundant code, update according to latest
3093 (print_constant): Bug fix relate to 3-word instructions.
3094 (assemble_insn): Bug fix, when matching instructions, verify also
3095 instruction type (not only mnemonic).
3096 Add various error checking.
3097 (preprocess_reglist): Support HI/LO and user registers.
3101 * config/tc-ia64.c (fixup_unw_records): Don't let the "t" value
3102 in an epilogue directive go negative.
3107 * config/tc-ia64.c (emit_one_bundle): Decrement md.num_slots_in_use
3108 after reporting template error during manual bundling. Reported
3113 * Makefile.am: Run dep-am.
3114 * aclocal.m4: Regenerate with automake 1.9.2.
3115 * Makefile.in: Regenerate with automake 1.9.2.
3116 * doc/Makefile.in: Likewise.
3118 * config/tc-arm.c: Include "dw2gencfi.h".
3119 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
3121 * config/tc-arm.h (TARGET_USE_CFIPOP, DWARF2_DEFAULT_RETURN_COLUMN)
3122 (DWARF2_CIE_DATA_ALIGNMENT, tc_regname_to_dw2regnum)
3123 (tc_cfi_frame_initial_instructions): Define.
3124 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
3129 * config/tc-crx.c (assemble_insn): Check unsigned immediate
3131 Update coprocessor id to be unsigned immediate.
3135 * config/tc-i386.c (O_secrel): Delete.
3136 (tc_pe_dwarf2_emit_offset): New function.
3137 * config/tc-i386.h (O_secrel): Define as O_md1.
3138 (TC_DWARF2_EMIT_OFFSET): Define.
3142 * config/tc-xstormy16.c (xstormy16_cons_fix_new): Accept and
3143 ignore @fptr() directives for 4-byte fixups.
3147 * Makefile.am: Run "make dep-am"
3148 * Makefile.in: Regenerate.
3152 * doc/c-xtensa.texi (Xtensa Options, Absolute Literals Directive):
3153 Remove comments about placement of literal pools.
3154 (Literal Directive): Update description of literal placement.
3155 (Literal Prefix Directive): Remove statement that this does not apply
3156 to absolute-mode literals. Describe new section naming scheme.
3160 * config/tc-xtensa.c (xtensa_frequency_pseudo): Use set_subseg_freq.
3161 (is_entry_opcode, is_movi_opcode, is_the_loop_opcode, is_jx_opcode,
3162 is_windowed_return_opcode): Delete.
3163 (xtensa_frob_label): Use get_subseg_target_freq.
3164 (md_assemble): Inline call to is_entry_opcode.
3165 (xtensa_handle_align): Inline call to get_frag_is_literal.
3166 (relaxation_requirements): Inline call to is_jx_opcode.
3167 (emit_single_op): Inline call to is_movi_opcode.
3168 (xg_assemble_vliw_tokens): Inline calls to get_frag_is_insn,
3169 get_frag_is_no_transform, is_entry_opcode, and
3170 set_frag_is_specific_opcode. Use get_subseg_total_freq.
3171 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags,
3172 xtensa_fix_close_loop_end_frags, relax_frag_immed, convert_frag_immed):
3173 Inline calls to get_frag_is_no_transform.
3174 (next_instrs_are_b_retw): Inline call to is_windowed_return_opcode.
3175 (xtensa_fix_short_loop_frags): Inline calls to is_the_loop_opcode and
3176 get_frag_is_no_transform.
3177 (convert_frag_immed_finish_loop): Inline calls to get_expression_value
3178 and set_frag_is_no_transform.
3179 (get_expression_value): Delete.
3180 (subseg_map struct): Rename cur_total_freq to total_freq. Rename
3181 cur_target_freq to target_freq.
3182 (get_subseg_info): Split out code to create a new map entry into ...
3183 (add_subseg_info): ... this new function.
3184 (get_last_insn_flags): Check if get_subseg_info succeeded.
3185 (set_last_insn_flags): Call add_subseg_info if needed.
3186 (get_subseg_total_freq, get_subseg_target_freq, set_subseg_freq): New.
3187 (xtensa_reorder_segments): Compute last_sec while counting sections.
3188 Remove call to get_last_sec.
3189 (get_last_sec): Delete.
3190 (cache_literal_section): Inline call to retrieve_literal_seg and its
3191 callees, seg_present and add_seg_list.
3192 (retrieve_literal_seg, seg_present, add_seg_list): Delete.
3193 (get_frag_is_insn, get_frag_is_no_transform,
3194 set_frag_is_specific_opcode, set_frag_is_no_transform): Delete.
3195 * config/tc-xtensa.h (MAX_SLOTS): Reduce from 31 to 15.
3199 * config/tc-xtensa.c: Use ISO C90 formatting.
3200 * config/tc-xtensa.h: Likewise.
3201 * config/xtensa-istack.h: Likewise.
3202 * config/xtensa-relax.c: Likewise.
3203 * config/xtensa-relax.h: Likewise.
3207 * config/tc-arm.c (md_begin): Change EF_ARM_EABI_VER3 to
3210 * doc/c-arm.texi: Document that we actually support -meabi=4, not
3215 * doc/as.texinfo (VTableEntry, VTableInherit): Add "directive" to index
3217 (Acknowledgements): Use "GAS" instead of AS variable.
3221 * config/tc-i386.c: Include "elf/x86-64.h".
3222 (i386_elf_section_type): New function.
3223 * config/tc-i386.h (md_elf_section_type): Define.
3224 (i386_elf_section_type): New prototype.
3228 * config/m68k-parse.h (enum m68k_register): New control register,
3230 * config/tc-m68k.c: Correct control register set for MCF5249.
3235 * config/tc-xtensa.c (absolute_literals_supported): New global flag.
3236 (UNREACHABLE_MAX_WIDTH): Define.
3237 (XTENSA_FETCH_WIDTH): Delete.
3238 (cur_vinsn, xtensa_fetch_width, xt_saved_debug_type, past_xtensa_end,
3239 prefer_const16, prefer_l32r): New global variables.
3240 (LIT4_SECTION_NAME): Define.
3241 (lit4_state struct): Add lit4_seg_name and lit4_seg fields.
3242 (XTENSA_PROP_*, GET_XTENSA_PROP_*, SET_XTENSA_PROP_*): Define.
3243 (frag_flags struct): New.
3244 (xtensa_block_info struct): Move from tc-xtensa.h. Add flags field.
3245 (subseg_map struct): Add cur_total_freq and cur_target_freq fields.
3246 (bitfield, bit_is_set, set_bit, clear_bit): Define.
3247 (MAX_FORMATS): Define.
3248 (op_placement_info struct, op_placement_table): New.
3249 (O_pltrel, O_hi16, O_lo16): Define.
3250 (directiveE enum): Rename directive_generics to directive_transform.
3251 Delete directive_relax. Add directive_schedule,
3252 directive_absolute_literals, and directive_last_directive.
3253 (directive_info): Rename "generics" to "transform". Delete "relax".
3254 Add "schedule" and "absolute-literals".
3255 (directive_state): Adjust entries to match changes in directive_info.
3256 (xtensa_relax_statesE, RELAX_IMMED_MAXSTEPS): Move to tc-xtensa.h.
3257 (xtensa_const16_opcode, xtensa_movi_opcode, xtensa_movi_n_opcode,
3258 xtensa_l32r_opcode, xtensa_nop_opcode, xtensa_rsr_lcount_opcode): New.
3259 (xtensa_j_opcode, xtensa_rsr_opcode): Delete.
3260 (align_only_targets, software_a0_b_retw_interlock,
3261 software_avoid_b_j_loop_end, maybe_has_b_j_loop_end,
3262 software_avoid_short_loop, software_avoid_close_loop_end,
3263 software_avoid_all_short_loops, specific_opcode): Delete.
3264 (warn_unaligned_branch_targets): New.
3265 (workaround_a0_b_retw, workaround_b_j_loop_end, workaround_short_loop,
3266 workaround_close_loop_end, workaround_all_short_loops): Default FALSE.
3267 (option_[no_]link_relax, option_[no_]transform,
3268 option_[no_]absolute_literals, option_warn_unaligned_targets,
3269 option_prefer_l32r, option_prefer_const16, option_target_hardware):
3271 (option_[no_]align_only_targets, option_literal_section_name,
3272 option_text_section_name, option_data_section_name,
3273 option_bss_section_name, option_eb, option_el): Delete.
3274 (md_longopts): Add entries for: [no-]transform, [no-]absolute-literals,
3275 warn-unaligned-targets, prefer-l32r, prefer-const16, [no-]link-relax,
3276 and target-hardware. Delete entries for [no-]target-align-only,
3277 literal-section-name, text-section-name, data-section-name, and
3279 (md_parse_option): Handle new options and remove old ones. Accept but
3280 ignore [no-]density options. Warn for [no-]generics and [no-]relax
3281 and treat them as [no-]transform.
3282 (md_show_usage): Add new options and remove old ones.
3283 (xtensa_setup_hw_workarounds): New.
3284 (md_pseudo_table): Change "word" entry to use xtensa_elf_cons. Add
3285 "long", "short", "loc" and "frequency" entries.
3286 (use_generics): Rename to ...
3287 (use_transform): ... this function. Add past_xtensa_end check.
3288 (use_longcalls): Add past_xtensa_end check.
3289 (code_density_available, can_relax): Delete.
3290 (do_align_targets): New.
3291 (get_directive): Accept dashes in directive names. Warn about
3292 [no-]generics and [no-]relax directives and treat them as
3294 (xtensa_begin_directive): Call md_flush_pending_output only for some
3295 directives. Check for directives inside instruction bundles. Warn
3296 about deprecated ".begin literal" usage. Warn and ignore [no-]density
3297 directives. Handle new directives. Check generating_literals flag
3299 (xtensa_end_directive): Check for directives inside instruction
3300 bundles. Warn and ignore [no-]density directives. Handle new
3301 directives. Call xtensa_set_frag_assembly_state.
3302 (xtensa_loc_directive_seen, xtensa_dwarf2_directive_loc,
3303 xtensa_dwarf2_emit_insn): New.
3304 (xtensa_literal_position): Call md_flush_pending_output. Do not check
3305 use_literal_section flag.
3306 (xtensa_literal_pseudo): Call md_flush_pending_output. Handle absolute
3307 literals. Use xtensa_elf_cons to parse the expression.
3308 (xtensa_literal_prefix): Do not check use_literal_section. Support
3309 ".lit4" sections for absolute literals. Change prefix convention to
3310 replace ".text" (or ".t" in a linkonce section). No need to call
3312 (xtensa_frequency_pseudo, xtensa_elf_cons, xtensa_elf_suffix): New.
3313 (expression_end): Handle closing braces and colons.
3314 (PLT_SUFFIX, plt_suffix): Delete.
3315 (expression_maybe_register): Use new xtensa-isa.h functions. Use
3316 xtensa_elf_suffix instead of checking for plt suffix, and handle O_lo16
3317 and O_hi16 expressions as well.
3318 (tokenize_arguments): Handle closing braces and colons.
3319 (parse_arguments): Use new xtensa-isa.h functions. Handle "invisible"
3320 operands and paired register syntax.
3321 (get_invisible_operands): New.
3322 (xg_translate_sysreg_op): Handle new Xtensa LX RSR/WSR/XSR syntax. Use
3323 new xtensa-isa.h functions.
3324 (xtensa_translate_old_userreg_ops, xtensa_translate_zero_immed): New.
3325 (xg_translate_idioms): Check if inside bundle. Use use_transform.
3326 Handle new Xtensa LX RSR/WSR/XSR syntax. Remove code to widen density
3327 instructions. Use xtensa_translate_zero_immed.
3328 (operand_is_immed, operand_is_pcrel_label): Delete.
3329 (get_relaxable_immed): Use new xtensa-isa.h functions.
3330 (get_opcode_from_buf): Add slot parameter. Use new xtensa-isa.h
3332 (xtensa_print_insn_table, print_vliw_insn): New.
3333 (is_direct_call_opcode): Use new xtensa-isa.h functions.
3334 (is_call_opcode, is_loop_opcode, is_conditional_branch_opcode,
3335 is_branch_or_jump_opcode): Delete.
3336 (is_movi_opcode, decode_reloc, encode_reloc, encode_alt_reloc): New.
3337 (opnum_to_reloc, reloc_to_opnum): Delete.
3338 (xtensa_insnbuf_set_operand, xtensa_insnbuf_get_operand): Use new
3339 xtensa-isa.h functions. Operate on one slot of an instruction.
3340 (xtensa_insnbuf_set_immediate_field, is_negatable_branch,
3341 xg_get_insn_size): Delete.
3342 (xg_get_build_instr_size): Use xg_get_single_size.
3343 (xg_is_narrow_insn, xg_is_single_relaxable_insn): Update calls to
3344 xg_build_widen_table. Use xg_get_single_size.
3345 (xg_get_max_narrow_insn_size): Delete.
3346 (xg_get_max_insn_widen_size, xg_get_max_insn_widen_literal_size,
3347 xg_is_relaxable_insn): Update calls to xg_build_widen_table. Use
3349 (xg_build_to_insn): Record the loc field. Handle OP_OPERAND_HI16U and
3350 OP_OPERAND_LOW16U. Check xg_valid_literal_expression.
3351 (xg_expand_to_stack, xg_expand_narrow): Update calls to
3352 xg_build_widen_table. Use xg_get_single_size.
3353 (xg_immeds_fit): Use new xtensa-isa.h functions. Update call to
3355 (xg_symbolic_immeds_fit): Likewise. Also handle O_lo16 and O_hi16, and
3356 treat weak symbols conservatively.
3357 (xg_check_operand): Use new xtensa-isa.h functions.
3358 (is_dnrange): Delete.
3359 (xg_assembly_relax): Inline previous calls to tinsn_copy.
3360 (xg_finish_frag): Specify separate relax states for the frag and slot0.
3361 (is_branch_jmp_to_next, xg_add_branch_and_loop_targets): Use new
3362 xtensa-isa.h functions.
3363 (xg_instruction_matches_option_term, xg_instruction_matches_or_options,
3364 xg_instruction_matches_options): New.
3365 (xg_instruction_matches_rule): Handle O_register expressions. Call
3366 xg_instruction_matches_options.
3367 (transition_rule_cmp): New.
3368 (xg_instruction_match): Update call to xg_build_simplify_table.
3369 (xg_build_token_insn): Record loc fields.
3370 (xg_simplify_insn): Check is_specific_opcode field and
3371 density_supported flag.
3372 (xg_expand_assembly_insn): Skip checking code_density_available. Use
3373 new xtensa-isa.h functions. Call use_transform instead of can_relax.
3374 (xg_assemble_literal): Add error handling for O_big. Call
3375 record_alignment. Handle O_pltrel.
3376 (xg_valid_literal_expression): New.
3377 (xg_assemble_literal_space): Add slot parameter. Remove call to
3378 set_expr_symbol_offset. Add call to record_alignment. Update call to
3380 (xg_emit_insn): Delete.
3381 (xg_emit_insn_to_buf): Add format parameter. Update calls to
3382 xg_add_opcode_fix and xtensa_insnbuf_to_chars.
3383 (xg_add_opcode_fix): Change opcode parameter to tinsn and add format
3384 and slot parameters. Handle new "alternate" relocations for absolute
3385 literals and CONST16 instructions. Check for bad uses of O_lo16 and
3386 O_hi16. Use new xtensa-isa.h functions.
3387 (xg_assemble_tokens): Delete.
3388 (is_register_writer): Use new xtensa-isa.h functions.
3389 (is_bad_loopend_opcode): Check for xtensa_rsr_lcount_opcode instead of
3390 old-style RSR from LCOUNT.
3391 (next_frag_opcode): Delete.
3392 (next_frag_opcode_is_loop, next_frag_format_size, frag_format_size,
3393 update_next_frag_state): New.
3394 (update_next_frag_nop_state): Delete.
3395 (next_frag_pre_opcode_bytes): Use next_frag_opcode_is_loop.
3396 (xtensa_mark_literal_pool_location): Check use_literal_section flag and
3397 the state of the absolute-literals directive. Add calls to
3398 record_alignment and xtensa_set_frag_assembly_state. Call
3399 xtensa_switch_to_non_abs_literal_fragment instead of
3400 xtensa_switch_to_literal_fragment.
3402 (assemble_nop): Use build_nop. Update call to xtensa_insnbuf_to_chars.
3403 (get_expanded_loop_offset): Change check for undefined opcode to an
3405 (xtensa_set_frag_assembly_state, relaxable_section,
3406 xtensa_find_unmarked_state_frags, xtensa_find_unaligned_branch_targets,
3407 xtensa_find_unaligned_loops, xg_apply_tentative_value): New.
3408 (md_begin): Update call to xtensa_isa_init. Initialize linkrelax to 1.
3409 Set lit4_seg_name. Call xg_init_vinsn. Initialize new global opcodes.
3410 Call init_op_placement_info_table and xtensa_set_frag_assembly_state.
3411 (xtensa_init_fix_data): New.
3412 (xtensa_frob_label): Reset label symbol to the current frag. Check
3413 do_align_targets and generating_literals flag. Propagate frequency
3414 info to new alignment frag. Call xtensa_set_frag_assembly_state.
3415 (xtensa_unrecognized_line): New.
3416 (xtensa_flush_pending_output): Check if inside a bundle. Add a call
3417 to xtensa_set_frag_assembly_state.
3418 (error_reset_cur_vinsn): New.
3419 (md_assemble): Remove check for literal frag. Remove call to
3420 istack_init. Call use_transform instead of use_generics. Parse
3421 explicit instruction format specifiers. Move code for
3422 a0_b_retw_interlock workaround to xg_assemble_vliw_tokens. Call
3423 error_reset_cur_vinsn on errors. Add call to get_invisible_operands.
3424 Add dwarf2_where call. Remote automatic alignment for ENTRY
3425 instructions. Move call to xtensa_clear_insn_labels to the end.
3426 Rearrange to handle bundles.
3427 (xtensa_cons_fix_new): Delete.
3428 (xtensa_handle_align): New.
3429 (xtensa_frag_init): Call xtensa_set_frag_assembly_state. Remove
3430 assignment to is_no_density field.
3431 (md_pcrel_from): Use new xtensa-isa.h functions. Use decode_reloc
3432 instead of reloc_to_opnum. Handle "alternate" relocations.
3433 (xtensa_force_relocation, xtensa_check_inside_bundle,
3434 xtensa_elf_section_change_hook): New.
3435 (xtensa_symbol_new_hook): Delete.
3436 (xtensa_fix_adjustable): Check for difference of symbols with an
3437 offset. Check for external and weak symbols.
3438 (md_apply_fix3): Remove cases for XTENSA_OP{0,1,2} relocs.
3439 (md_estimate_size_before_relax): Return expansion for the first slot.
3440 (tc_gen_reloc): Handle difference of symbols by producing
3441 XTENSA_DIFF{8,16,32} relocs and by writing the value of the difference
3442 into the output. Handle new XTENSA_SLOT*_OP relocs by storing the
3443 tentative values into the output when linkrelax is set.
3444 (XTENSA_PROP_SEC_NAME): Define.
3445 (xtensa_post_relax_hook): Call xtensa_find_unmarked_state_frags.
3446 Create literal tables only if using literal sections. Create new
3447 property tables instead of old instruction tables. Check for unaligned
3448 branch targets and loops.
3449 (finish_vinsn, find_vinsn_conflicts, check_t1_t2_reads_and_writes,
3450 new_resource_table, clear_resource_table, resize_resource_table,
3451 resources_available, reserve_resources, release_resources,
3452 opcode_funcUnit_use_unit, opcode_funcUnit_use_stage,
3453 resources_conflict, xg_find_narrowest_format, relaxation_requirements,
3454 bundle_single_op, emit_single_op, xg_assemble_vliw_tokens): New.
3455 (xtensa_end): Call xtensa_flush_pending_output. Set past_xtensa_end
3456 flag. Update checks for workaround options. Call
3457 xtensa_mark_narrow_branches and xtensa_mark_zcl_first_insns.
3458 (xtensa_cleanup_align_frags): Add special case for branch targets.
3459 Check for and mark unreachable frags.
3460 (xtensa_fix_target_frags): Remove use of align_only_targets flag.
3461 Use RELAX_LOOP_END_BYTES in special case for negatable branch at the
3462 end of a zero-overhead loop body.
3463 (frag_can_negate_branch): Handle instructions with multiple slots.
3464 Use new xtensa-isa.h functions
3465 (xtensa_mark_narrow_branches, is_narrow_branch_guaranteed_in_range,
3466 xtensa_mark_zcl_first_insns): New.
3467 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags): Error if
3468 transformations are disabled.
3469 (next_instrs_are_b_retw): Use new xtensa-isa.h functions. Handle
3470 multislot instructions.
3471 (xtensa_fix_close_loop_end_frags, xtensa_fix_short_loop_frags):
3472 Likewise. Also error if transformations are disabled.
3473 (unrelaxed_frag_max_size): New.
3474 (unrelaxed_frag_min_insn_count, unrelax_frag_has_b_j): Use new
3475 xtensa-isa.h functions.
3476 (xtensa_sanity_check, is_empty_loop, is_local_forward_loop): Use
3477 xtensa_opcode_is_loop instead of is_loop_opcode.
3478 (get_text_align_power): Replace as_fatal with assertion.
3479 (get_text_align_fill_size): Iterate instead of using modulus when
3481 (get_noop_aligned_address): Assert that this is for a machine-dependent
3482 RELAX_ALIGN_NEXT_OPCODE frag. Use next_frag_opcode_is_loop,
3483 xg_get_single_size, and frag_format_size.
3484 (get_widen_aligned_address): Rename to ...
3485 (get_aligned_diff): ... this function. Add max_diff parameter.
3486 Remove handling of rs_align/rs_align_code frags. Use
3487 next_frag_format_size, get_text_align_power, get_text_align_fill_size,
3488 next_frag_opcode_is_loop, and xg_get_single_size. Compute max_diff
3489 and pass it back to caller.
3490 (xtensa_relax_frag): Use relax_frag_loop_align. Add code for new
3491 RELAX_SLOTS, RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN,
3492 RELAX_FILL_NOP, and RELAX_UNREACHABLE frag types. Check relax_seen.
3493 (relax_frag_text_align): Rename to ...
3494 (relax_frag_loop_align): ... this function. Assume loops can only be
3495 in the first slot of an instruction.
3496 (relax_frag_add_nop): Use assemble_nop instead of constructing an OR
3497 instruction. Remove call to frag_wane.
3498 (relax_frag_narrow): Rename to ...
3499 (relax_frag_for_align): ... this function. Extend to handle
3500 RELAX_FILL_NOP and RELAX_UNREACHABLE, as well as RELAX_SLOTS with
3501 RELAX_NARROW for the first slot.
3502 (find_address_of_next_align_frag, bytes_to_stretch): New.
3503 (future_alignment_required): Use find_address_of_next_align_frag and
3504 bytes_to_stretch. Look ahead to subsequent frags to make smarter
3505 alignment decisions.
3506 (relax_frag_immed): Add format, slot, and estimate_only parameters.
3507 Check if transformations are enabled for b_j_loop_end workaround.
3508 Use new xtensa-isa.h functions and handle multislot instructions.
3509 Update call to xg_assembly_relax.
3510 (md_convert_frag): Handle new RELAX_SLOTS, RELAX_UNREACHABLE,
3511 RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN, and RELAX_FILL_NOP
3513 (convert_frag_narrow): Add segP, format and slot parameters. Call
3514 convert_frag_immed for branch instructions. Adjust calls to
3515 tinsn_from_chars, tinsn_immed_from_frag, and xg_emit_insn_to_buf. Use
3516 xg_get_single_size and xg_get_single_format.
3517 (convert_frag_fill_nop): New.
3518 (convert_frag_immed): Add format and slot parameters. Handle multislot
3519 instructions and use new xtensa-isa.h functions. Update calls to
3520 tinsn_immed_from_frag and xg_assembly_relax. Check if transformations
3521 enabled for b_j_loop_end workaround. Use build_nop instead of
3522 assemble_nop. Check is_specific_opcode flag. Check for unreachable
3523 frags. Use xg_get_single_size. Handle O_pltrel.
3524 (fix_new_exp_in_seg): Remove check for old plt flag.
3525 (convert_frag_immed_finish_loop): Update calls to tinsn_from_chars and
3526 xtensa_insnbuf_to_chars. Call tinsn_immed_from_frag. Change check
3527 for loop opcode to an assertion. Mark all frags up to the end of the
3528 loop as not transformable.
3529 (get_last_insn_flags, set_last_insn_flags): Use get_subseg_info.
3530 (get_subseg_info): New.
3531 (xtensa_move_literals): Call xtensa_set_frag_assembly_state. Add null
3533 (xtensa_switch_to_literal_fragment): Rewrite to handle absolute
3534 literals and use xtensa_switch_to_non_abs_literal_fragment otherwise.
3535 (xtensa_switch_to_non_abs_literal_fragment): New.
3536 (cache_literal_section): Add is_code parameter and pass it through to
3537 retrieve_literal_seg.
3538 (retrieve_literal_seg): Add is_code parameter and use it to set the
3539 flags on the literal section. Handle case where head parameter is 0.
3540 (get_frag_is_no_transform, set_frag_is_specific_opcode,
3541 set_frag_is_no_transform): New.
3542 (xtensa_create_property_segments): Add end_property_function parameter
3543 and pass it through to add_xt_block_frags. Call bfd_get_section_flags
3544 and skip SEC_DEBUGGING and !SEC_ALLOC sections.
3545 (xtensa_create_xproperty_segments, section_has_xproperty): New.
3546 (add_xt_block_frags): Add end_property_function parameter and call it
3547 if it is non-zero. Call xtensa_frag_flags_init.
3548 (xtensa_frag_flags_is_empty, xtensa_frag_flags_init,
3549 get_frag_property_flags, frag_flags_to_number,
3550 xtensa_frag_flags_combinable, xt_block_aligned_size,
3551 xtensa_xt_block_combine, add_xt_prop_frags,
3552 init_op_placement_info_table, opcode_fits_format_slot,
3553 xg_get_single_size, xg_get_single_format): New.
3554 (istack_push): Inline call to tinsn_copy.
3555 (tinsn_copy): Delete.
3556 (tinsn_has_invalid_symbolic_operands): Handle O_hi16 and O_lo16 and
3557 CONST16 opcodes. Handle O_big, O_illegal, and O_absent.
3558 (tinsn_has_complex_operands): Handle O_hi16 and O_lo16.
3559 (tinsn_to_insnbuf): Use xg_get_single_format and new xtensa-isa.h
3560 functions. Handle invisible operands.
3561 (tinsn_to_slotbuf): New.
3562 (tinsn_check_arguments): Use new xtensa-isa.h functions.
3563 (tinsn_from_chars): Add slot parameter. Rewrite using xg_init_vinsn,
3564 vinsn_from_chars, and xg_free_vinsn.
3565 (tinsn_from_insnbuf): New.
3566 (tinsn_immed_from_frag): Add slot parameter and handle multislot
3567 instructions. Handle symbol differences.
3568 (get_num_stack_text_bytes): Use xg_get_single_size.
3569 (xg_init_vinsn, xg_clear_vinsn, vinsn_has_specific_opcodes,
3570 xg_free_vinsn, vinsn_to_insnbuf, vinsn_from_chars, expr_is_register,
3571 get_expr_register, set_expr_symbol_offset_diff): New.
3572 * config/tc-xtensa.h (MAX_SLOTS): Define.
3573 (xtensa_relax_statesE): Move from tc-xtensa.c. Add
3574 RELAX_CHECK_ALIGN_NEXT_OPCODE, RELAX_MAYBE_DESIRE_ALIGN, RELAX_SLOTS,
3575 RELAX_FILL_NOP, RELAX_UNREACHABLE, RELAX_MAYBE_UNREACHABLE, and
3577 (RELAX_IMMED_MAXSTEPS): Move from tc-xtensa.c.
3578 (xtensa_frag_type struct): Add is_assembly_state_set,
3579 use_absolute_literals, relax_seen, is_unreachable, is_specific_opcode,
3580 is_align, is_text_align, alignment, and is_first_loop_insn fields.
3581 Replace is_generics and is_relax fields by is_no_transform field.
3582 Delete is_text and is_longcalls fields. Change text_expansion and
3583 literal_expansion to arrays of MAX_SLOTS entries. Add arrays of
3584 per-slot information: literal_frags, slot_subtypes, slot_symbols,
3585 slot_sub_symbols, and slot_offsets. Add fr_prev field.
3586 (xtensa_fix_data struct): New.
3587 (xtensa_symfield_type struct): Delete plt field.
3588 (xtensa_block_info struct): Move definition to tc-xtensa.h. Add
3589 forward declaration here.
3590 (xt_section_type enum): Delete xt_insn_sec. Add xt_prop_sec.
3591 (XTENSA_SECTION_RENAME): Undefine.
3592 (TC_FIX_TYPE, TC_INIT_FIX_DATA, TC_FORCE_RELOCATION, NO_PSEUDO_DOT,
3593 tc_unrecognized_line, md_do_align, md_elf_section_change_hook,
3594 HANDLE_ALIGN, TC_LINKRELAX_FIXUP, SUB_SEGMENT_ALIGN): Define.
3595 (TC_CONS_FIX_NEW, tc_symbol_new_hook): Delete.
3596 (unit_num_copies_func, opcode_num_units_func,
3597 opcode_funcUnit_use_unit_func, opcode_funcUnit_use_stage_func): New.
3598 (resource_table struct): New.
3599 * config/xtensa-istack.h (MAX_INSN_ARGS): Increase from 6 to 10.
3600 (TInsn struct): Add keep_wide, loc, fixup, record_fix, subtype,
3601 literal_space, symbol, sub_symbol, offset, and literal_frag fields.
3602 (tinsn_copy): Delete prototype.
3603 (vliw_insn struct): New.
3604 * config/xtensa-relax.c (insn_pattern_struct): Add options field.
3605 (widen_spec_list): Add option conditions for density and boolean
3606 instructions. Add expansions using CONST16 and conditions for using
3607 CONST16 vs. L32R. Use new Xtensa LX RSR/WSR syntax. Add entries for
3609 (simplify_spec_list): Add option conditions for density instructions.
3610 Add entry for NOP instruction.
3611 (append_transition): Add cmp function pointer parameter and use it to
3612 insert the new entry in order.
3613 (operand_function_LOW16U, operand_function_HI16U): New.
3614 (xg_has_userdef_op_fn, xg_apply_userdef_op_fn): Handle
3615 OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
3616 (enter_opname, split_string): Use xstrdup instead of strdup.
3617 (init_insn_pattern): Initialize new options field.
3618 (clear_req_or_option_list, clear_req_option_list,
3619 clone_req_or_option_list, clone_req_option_list, parse_option_cond):
3621 (parse_insn_pattern): Parse option conditions.
3622 (transition_applies): New.
3623 (build_transition): Use new xtensa-isa.h functions. Fix incorrectly
3624 swapped last arguments in calls to append_constant_value_condition.
3625 Call clone_req_option_list. Add warning about invalid opcode.
3626 Handle LOW16U and HI16U function names.
3627 (build_transition_table): Add cmp parameter and use it in calls to
3628 append_transition. Use new xtensa-isa.h functions. Check
3629 transition_applies before adding entries.
3630 (xg_build_widen_table, xg_build_simplify_table): Add cmp parameter and
3631 pass it through to build_transition_table.
3632 * config/xtensa-relax.h (ReqOrOptionList, ReqOrOption, ReqOptionList,
3633 ReqOption, transition_cmp_fn): New types.
3634 (OpType enum): Add OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
3635 (transition_rule struct): Add options field.
3636 * doc/as.texinfo (Overview): Update Xtensa options.
3637 * doc/c-xtensa.texi (Xtensa Options): Delete --[no-]density,
3638 --[no-]relax, and --[no-]generics options. Update descriptions of
3639 --text-section-literals and --[no-]longcalls. Add
3640 --[no-]absolute-literals and --[no-]transform.
3641 (Xtensa Syntax): Add description of syntax for FLIX instructions.
3642 Remove use of "generic" and "specific" terminology for opcodes.
3643 (Xtensa Registers): Generalize the syntax description to include
3644 user-defined register files.
3645 (Xtensa Automatic Alignment): Update.
3646 (Xtensa Branch Relaxation): Mention limitation of unconditional jumps.
3647 (Xtensa Call Relaxation): Linker can now remove most of the overhead.
3648 (Xtensa Directives): Remove confusing rules about precedence.
3649 (Density Directive, Relax Directive): Delete.
3650 (Schedule Directive): New.
3651 (Generics Directive): Rename to ...
3652 (Transform Directive): ... this node.
3653 (Literal Directive): Update for absolute literals. Missing
3654 literal_position directive is now an error.
3655 (Literal Position Directive): Update for absolute literals.
3656 (Freeregs Directive): Delete.
3657 (Absolute Literals Directive): New.
3658 (Frame Directive): Minor editing.
3659 * Makefile.am (DEPTC_xtensa_elf, DEPOBJ_xtensa_elf, DEP_xtensa_elf):
3660 Update dependencies.
3661 * Makefile.in: Regenerate.
3665 * config/tc-mips.c (append_insn): Use fix_new rather than fix_new_exp
3666 to build the second and third fixups for a composite relocation.
3667 (macro_read_relocs): New function.
3668 (macro_build): Use it.
3669 (s_cpsetup): Pass all three composite relocation codes to macro_build.
3670 Simplify fragging code accordingly.
3671 (s_gpdword): Use fix_new rather than fix_new_exp for the second part
3672 of the composite relocation. Set fx_tcbit in both fixups.
3676 * config/tc-mips.c (append_insn): Set fx_tcbit for composite relocs.
3677 (md_apply_fix3): Don't treat composite relocs as done.
3681 * macro.c (macro_expand_body): When ELF, use .LL rather than LL as
3682 prefix for symbol names generated from the LOCAL macro directive.
3684 * dw2gencfi.c (select_cie_for_fde): When separating CIE out from
3685 FDE, treat a DW_CFA_remember_state as we do a DW_CFA_advance_loc.
3689 * config/tc-crx.c (preprocess_reglist): Handle Co-processor
3691 (md_assemble): Add error checking for Co-Processor instructions.
3692 (get_cinv_parameters): Add 'b' option to invalidate the
3693 branch-target cache.
3697 * config/tc-arm.c (unwind): New variable.
3698 (vfp_sp_encode_reg): New function.
3699 (vfp_sp_reg_required_here): Use it.
3700 (vfp_sp_reg_list, vfp_dp_reg_list): Remove.
3701 (vfp_parse_reg_list): New function.
3702 (s_arm_unwind_fnstart, s_arm_unwind_fnend, s_arm_unwind_cantunwind,
3703 s_arm_unwind_personality, s_arm_unwind_personalityindex,
3704 s_arm_unwind_handlerdata, s_arm_unwind_save, s_arm_unwind_movsp,
3705 s_arm_unwind_pad, s_arm_unwind_setfp, s_arm_unwind_raw): New
3707 (md_pseudo_table): Add them.
3708 (do_vfp_reg2_from_sp2): Use vfp_parse_reg_list and vfp_sp_encode_reg.
3709 (do_vfp_sp2_from_reg2, vfp_sp_ldstm, vfp_dp_ldstm): Ditto.
3710 (set_section, add_unwind_adjustsp, flush_pending_unwind,
3711 finish_unwind_opcodes, start_unwind_section, create_unwind_entry,
3712 require_hashconst, add_unwind_opcode): New functions.
3713 * doc/c-arm.texi: Document unwinding opcodes.
3714 * NEWS: Mention the new feature.
3718 * config/tc-mips.c (md_apply_fix3): Remove erroneous assert.
3722 * config/tc-ppc.c (md_apply_fix3): Call S_SET_THREAD_LOCAL for
3724 * config/tc-s390.c (md_apply_fix3): Likewise.
3725 * config/tc-sparc.c (md_apply_fix3): Likewise.
3729 * config/tc-arm.c (arm_elf_section_type): New function.
3730 (arm_elf_change_section): Set section link for exidx sections.
3731 * config/tc-arm.h (arm_elf_section_type): Add prototype.
3732 (md_elf_section_type): Define.
3736 * config/tc-pdp11.c (md_apply_fix3): Change to sign of the SOB
3737 instruction's offset.
3741 * (TARGET_FORMAT): Remove LynxOS COFF definition.
3745 * config/tc-arc.c (tc_gen_reloc): Don't assume fixP->fx_addsy is an
3746 asymbol *, instead use symbol_get_bfdsym.
3750 * config/tc-m68k.c (select_control_regs): Add mcf5249.
3754 * config/tc-arm.c (do_smi, do_nop): New functions.
3755 (insns): Add ARMv6ZK instructions.
3756 (md_apply_fix3): Handle BFD_RELOC_ARM_SMI.
3757 (tc_gen_reloc): Ditto.
3758 (arm_cpus): Add mpcore and arm1176.
3759 (arm_archs): Add armv6{k,z,zk}.
3760 * doc/c-arm.texi: Document new cores and architectures.
3764 * config/tc-arm.c: Use ISO C90 formatting.
3768 * config/tc-arm.c (mav_reg_required_here): Allow REG_TYPE_CN
3769 as alternative when REG_TYPE_MVF, REG_TYPE_MVD, REG_TYPE_MVFX or
3770 REG_TYPE_MVDX is expected.
3774 * doc/c-i386.texi (i386-Mnemonics): Fix typo.
3778 * config/tc-ia64.c (ENCODED_PSP_OFFSET): New.
3779 (output_rp_psprel, output_pfs_psprel, output_preds_psprel,
3780 output_spill_base, output_unat_psprel, output_lc_psprel,
3781 output_fpsr_psprel, output_priunat_psprel, output_bsp_psprel,
3782 output_bsprestore_psprel, output_rnat_psprel, output_spill_psprel,
3783 output_spill_psprel_p): Use it.
3787 * config/tc-crx.c (handle_LoadStor): New function.
3788 Handle load/stor unique instructions before parsing.
3792 * config/tc-arm.c (s_arm_rel31): New funciton.
3793 (md_pseudo_table): Add .rel31.
3794 (md_apply_fix3): Handle BFD_RELOC_ARM_TARGET2,
3795 BFD_RELOC_32_PCREL and BFD_RELOC_ARM_PREL31.
3796 (tc_gen_reloc): Handle BFD_RELOC_ARM_PREL31 and BFD_RELOC_ARM_TARGET2.
3797 (arm_fix_adjustable): Return 0 for BFD_RELOC_ARM_TARGET2.
3798 (arm_parse_reloc): Add (target2).
3802 * Makefile.am: Run "make dep-am".
3803 * Makefile.in: Regenerate.
3804 * aclocal.m4: Regenerate.
3805 * configure: Regenerate.
3806 * doc/Makefile.in: Regenerate.
3807 * po/POTFILES.in: Regenerate.
3808 * po/gas.pot: Regenerate.
3812 * config/tc-mmix.c [!LLONG_MIN]: Correct #elsif to #elif.
3813 [!LLONG_MAX]: Ditto.
3817 * config/tc-arm.c: Rename RELABS to TARGET1.
3821 * messages.c (as_internal_value_out_of_range): Cast values passed
3822 to as_bad_where or as_warn_where to proper type.
3826 * config/tc-avr.c: Add support for
3827 atmega48, atmega88, atmega168, attiny13, attiny2313, at90can128.
3831 * dw2gencfi.c (select_cie_for_fde): When separating CIE out
3832 from FDE, treat a CFI_escape as we do a DW_CFA_advance_loc.
3836 * config/obj-elf.c (obj_elf_section_type): Handle init_array,
3837 fini_array and preinit_array section types.
3838 * config/tc-ia64.c (ia64_elf_section_type): Remove init_array
3840 * doc/as.texinfo: Document extra section types.
3844 * Makefile.am (TARG_ENV_HFILES): Add te-symbian.h.
3845 * Makefile.in: Regenerated.
3846 * configure.in: Set em for arm*-*-symbianelf*.
3847 * configure: Regenerated.
3848 * config/tc-arm.c (elf32_arm_target_format): Use Symbian target
3849 vectors when appropriate.
3850 * config/te-symbian.h: New file.
3854 * config/tc-crx.c (gettrap): Exception vector can be case
3856 (process_label_constant): Fix a 32-bit displacement bug in branch
3858 (get_operandtype) : Bug fix, wrong operand was used.
3859 (process_label_constant): Initialize relocation type to
3864 * tc-arm.c (arm_cpus, arm_fpus): Allow <cpu>-s as well as <cpu>s
3865 for synthesizable cores.
3867 * doc/c-arm.texi (ARM Options): Document canonical names of CPUs.
3871 * config/tc-msp430.c: Clean-up the code.
3872 (md_relax_table): New relax table.
3873 (mcu_types): Sort MCU types.
3874 (md_pseudo_table): Add .profiler pseudo handler.
3875 (pow2value): New function.
3876 (msp430_profiler): New function.
3877 (msp430_operands): Add new insns handlers.
3878 (msp430_srcoperand): Add register operand handler, allow complex
3880 (md_estimate_size_before_relax): Rewritten.
3881 (md_convert_frag): Rewritten.
3882 (msp430_relax_frag): New function.
3883 * config/tc-msp430.h (md_relax_frag): define macro
3884 * doc/c-msp430.texi: Update information.
3888 * as.c (std_shortopts): Allow -g to take an optional argument.
3889 (parse_args): Pass any switch starting with -g on to the backend
3894 * configure.in (arm*-*-symbianelf*): New target.
3895 (arm*-*-eabi*): Likewise.
3896 * configure: Regenerated.
3899 * config/tc-mips.c (append_insn): Handle delay slots in branch likely
3904 * config/tc-ia64.c (start_unwind_section): Add linkonce_empty
3905 argument, don't do anything if current section is not
3906 .gnu.linkonce.t.* and linkonce_empty is set.
3907 (generate_unwind_image, dot_endp): Adjust callers, call
3908 start_unwind_section (*, 1) if nothing will be put into the
3913 * as.c (MD_DEBUG_FORMAT_SELECTOR): Provide default definition.
3914 (show_usage): Add -g.
3915 (std_longopts): Add --gen-debug. Alpha sort the table.
3916 (parse_args): Print an error message if a switch is not handled.
3917 Handle the -g switch, calling md_debug_format_selector() if
3919 * NEWS: Mention new feature.
3920 * doc/as.texinfo: Document new switch.
3921 * doc/internals.texi: Document behaviour of md_parse_option.
3923 * config/tc-arm.c (md_parse_option): Do not issue an error message
3924 if the switch is not recognised.
3925 * config/tc-m68k.c (md_parse_option): Likewise.
3926 * config/tc-pdp11.c (md_parse_option): Likewise.
3927 * config/tc-v850.c (md_parse_option): Likewise.
3929 * as.h: Fix up formatting.
3934 * macro.c (macro_set_alternate): Use ISO C90 formatting.
3936 * configure.in: Sort architecture based tables alphabetically.
3937 * configure: Regenerate.
3941 * config/tc-ppc.c (tc_ppc_regname_to_dw2regnum <regnames>): Replace
3942 { "cc", 68 }, with { "cr", 70 }.
3947 * as.c: Add and handle new --alternate command line option.
3948 * macro.c (macro_set_alternate): New.
3949 * macro.h (macro_set_alternate): Declare.
3950 * read.c: Add and handle new .altmacro and .noaltmacro directives.
3951 * doc/as.texinfo: Document new command line option and pseudo-ops
3952 as well as insert documentation originating from gasp about
3953 alternate macro syntax.
3954 * NEWS: Mention new command line option and pseudo-ops.
3958 * expr.c (operand): Handle the "~", "-", and "!" operators applied
3963 * config/tc-arm.c (md_apply_fix3, tc_gen_reloc, arm_parse_reloc):
3964 Handle new relocations.
3965 * include/elf/arm.h (elf_arm_reloc_type): Add new EABI relocations.
3969 * write.c (relax_segment): Use was_address instead of address when
3970 setting fr_fix field for align frag due to backwards .org.
3974 Introduce SH2a support.
3976 * config/tc-sh.c (get_specific): Change arch_sh2a_up to
3979 * config/tc-sh.c (md_parse_option): Add sh2a-nofpu ISA handling.
3981 * config/tc-sh.c (sh_elf_final_processing): Move sh2a recognition
3982 to end of conditional expression.
3984 * config/tc-sh.c: Add sh2a-nofpu support.
3986 * tc-sh.c: Add sh2a support.
3987 (parse_reg): Add tbr.
3988 (parse_at): Support @@(disp,tbr).
3989 (get_specific): Support sh2a opcodes.
3990 (insert4): New, for 4 byte relocs.
3991 (build_Mytes): Support sh2a opcodes.
3992 (md_apply_fix3_Mytes): Support sh2a opcodes.
3994 * config/tc-sh.c (md_parse_option): Handle sh2a.
3995 (sh_elf_final_processing): Ditto.
3999 * config/tc-hppa.h (TARGET_FORMAT): Set to "elf32-hppa-netbsd"
4004 * config/tc-ppc.c (ppc_frob_file_before_adjust): Warn if .toc too big.
4005 (ppc_arch): Expand comment.
4009 * config/tc-crx.c: Support evaluating the difference between two
4011 * config/tc-crx.h: Likewise.
4015 * config/tc-ia64.c (start_unwind_section): Set the linked-to
4017 (ia64_elf_section_change_hook): Set the linked-to section for
4022 * config/tc-msp430.c: Add new subtargets: msp430x1610,
4023 msp430x1611, msp430x1612, msp430x415, msp430x417, msp430xG437,
4024 msp430xG438, msp430xG439.
4028 * doc/as.texinfo (Section, PushSection): Correct documentation
4033 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
4034 permissible for the selected instruction suffix.
4035 (match_template): Don't permit 64-bit general purpose operands in
4037 (finalize_imm): Permit 64-bit immediates.
4038 (build_modrm_byte): Don't treat 32-bit addressing in 64-bit mode
4039 specially except for the width of the used base and/or index
4040 registers. For 32-bit displacements, use sign-extended
4041 relocations only when using 64-bit addressing.
4042 Force zero displacement on rip-relative addressing when there is
4043 no other displacement.
4044 (i386_index_check): Don't treat 32-bit addressing in 64-bit mode
4045 specially except for the width of the used base and/or index
4047 (parse_register): Disallow Reg64 registers in 32-bit mode.
4049 * config/tc-i386.c: For DefaultSize instructions, don't guess a 'q'
4050 suffix if the instruction doesn't support it.
4054 * config/tc-mips.c (append_insn): Handle constant expressions with
4055 no associated relocation.
4056 (mips_ip): Cancel the expression after use for the Q format
4058 (parse_relocation): Return no relocation for unsupported
4060 (my_getSmallExpression): Return no relocation if no relocation
4065 * config/obj-som.c (adjust_stab_sections): Add prototype.
4066 (obj_som_compiler, obj_som_version, obj_som_copyright,
4067 adjust_stab_sections): Add ATTRIBUTE_UNUSED to unused arguments.
4068 * config/tc-hppa.c (update_subspace): Likewise.
4069 (is_defined_subspace): Amplify comment.
4070 * config/obj-som.h (som_frob_file): Add prototype.
4075 * subsegs.c (section_symbol): Don't create a new segment when
4076 existing segment is undefined.
4080 * config/tc-arm.c: Include include/opcode/arm.h.
4081 (ARM_EXT_*, ARM_ARCH_*, ARM_ANY, ARM_ALL, COPROC_ANY): Delete.
4082 (FPU_FPA_EXT_* FPU_VFP_EXT_*, FPU_ANY, FPU_NONE, FPU_MAVERICK): Delete.
4083 (FPU_ARCH_*): Delete.
4084 * Makefile.am: Update dependencies.
4085 * Makefile.in: Regenerate.
4089 * configure.in: Accept armbe as a big-endian arm configuration.
4090 * configure: Regenerate.
4094 * config/tc-i386.c (T_SHIFTOP): New constant.
4095 (intel_e05_1): Handle '&', '|' and T_SHIFTOP.
4096 (intel_el1): Handle '~'.
4097 (intel_get_token): Handle '<>', '&', '|' and '~'.
4101 (md_assemble): Remove spurious newline from end of as_bad error
4103 (intel_e05_1): Likewise.
4104 (intel_e11): Likewise.
4105 (intel_match_token): Likewise.
4109 * config/tc-m68k.c: Convert to C90. Remove redundant
4110 declarations. Indentation fixup.
4111 [M68KCOFF]: Include "obj-coff.h" instead of declaring
4112 obj_coff_section ourselves.
4116 * config/tc-ia64.c (default_big_endian): New.
4117 (dot_byteorder, md_begin): Use it.
4118 (md_parse_option): Set it.
4122 * configure.in: Change sh-sybmian-elf to sh-*-symbianelf.
4123 * configure: Regenerate.
4124 * NEWS: Change sh-sybmian-elf to sh-*-symbianelf.
4125 * config/tc-sh.c (sh_elf_final_processing): Use renamed version of
4126 sh_find_elf_flags if necessary.
4130 * config/tc-mips.c (mips_fix_adjustable): If the full addend is
4131 going to be split into more than one in-place addend, return 0
4132 for relocations against mergeable sections. Associate comments
4137 * Makefile.am (CPU_TYPES): Add crx.
4138 (TARGET_CPU_CFILES): Add config/tc-crx.c.
4139 (TARGET_CPU_HFILES): Add config/tc-crx.h.
4140 (DEPTC_crx_elf): New target.
4141 (DEPOBJ_crx_elf): Likewise.
4142 (DEP_crx_elf): Likewise.
4143 * Makefile.in: Regenerate.
4144 * configure.in: Add crx* target.
4145 * configure: Regenerate.
4146 * config/tc-crx.c: New file.
4147 * config/tc-crx.h: New file.
4148 * NEWS: Mention new target.
4152 * config.in: Undefine TARGET_SYMBIAN by default.
4154 * configure: Regenerate. Add sh-symbian-elf target. If
4155 selected define TARGET_SYMBIAN.
4156 * config/tc-sh.h (TARGET_FORMAT): Select a Symbian target
4157 format if TARGET_SYMBIAN has been defined.
4159 * output-file.c (output_file_create): Report the target format
4160 chosen when bfd_openw reports that it is invalid.
4162 * config/obj-coff.c (coff_pseudo_table): Only define the weak
4163 pseudo for BFD based assemblers.
4168 * config/tc-sh.c (md_assemble): Change isspace to ISSPACE.
4169 (md_parse_option): Remove redundant -isa testing.
4170 Make bfd_arch variable const.
4171 (md_show_usage): Make bfd_arch variable const.
4175 * config/tc-ia64.c (emit_one_bundle): Check and set insn_addr.
4176 * config/tc-ia64.h (md_frag_check): Define.
4180 * config/obj-coff.c (obj_coff_weak): New .weak syntax for PE weak
4182 * doc/as.texinfo (Weak): Document PE weak symbols.
4186 * config/tc-mips.c (HAVE_IN_PLACE_ADDENDS): New macro.
4187 (reloc_needs_lo_p): Only return true if HAVE_IN_PLACE_ADDENDS.
4188 (mips_frob_file): Rework so that only a single pass through the
4189 relocs is needed. Allow %lo()s to have higher offsets than their
4190 corresponding %hi()s or %got()s.
4194 * config/tc-arm.c (md_apply_fix3:BFD_RELOC_ARM_IMMEDIATE): Do not
4195 allow values which have come from undefined symbols.
4196 Always consider this fixup to have been processed as a reloc
4197 cannot be generated for it.
4201 * frags.h (struct frag): Add has_code and insn_addr fields.
4202 * write.c (cvt_frag_to_fill): Invoke md_frag_check.
4203 * config/tc-ppc.c (md_assemble): Check and set insn_addr.
4204 * config/tc-ppc.h (md_frag_check): Define.
4208 * doc/Makefile.am (info): Rename goal to...
4209 (info-local): ... this, to preserve implicit dependencies.
4210 * doc/Makefile.in: Regenerate with automake 1.8.5.
4214 * config/tc-m32r.c (md_convert_frag): Changed for @PLT.
4215 (m32r_cgen_record_fixup_exp): Changed for @GOTOFF, @GOT.
4216 (m32r_fix_adjustable): Changed for @GOTOFF, @GOT, @PLT.
4217 (tc_gen_reloc): Likewise.
4218 (m32r_end_of_match): Add for @GOTOFF, @GOT, @PLT.
4219 (m32r_parse_name): Likewise.
4220 (m32r_cgen_parse_fix_exp): Likewise.
4221 * config/tc-m32r.h (md_parse_name): Define for @GOTOFF, @GOT, @PLT.
4222 (O_PIC_reloc): Likewise.
4223 (TC_CGEN_PARSE_FIX_EXP): Likewise..
4224 * cgen.c (gas_cgen_parse_operand): Add TC_CGEN_PARSE_FIX_EXP
4225 for @GOTOFF, @GOT, @PLT.
4229 * gas/symbols.c: While discarding ordinary local absolute symbols
4230 when --strip-local-absolute is in effect, retain file symbols.
4234 * config/tc-m68k.c (mri_chip): Replace current_chip, not augment.
4235 (md_parse_option): Likewise.
4239 * config/tc-i386.c: Deal with LEX_QM the same way as with LEX_AT.
4240 * config/te-netware.h: New file.
4241 * config/te-ppcnw.h: Delete: Obsolete.
4242 * configure.in: Eliminate ill NetWare targets. Make generic
4243 NetWare target use proper emulation.
4244 * Makefile.am: Eliminate reference to obsolete te-ppcnw.h, add
4245 reference to new te-netware.h.
4246 * configure: Regenerate.
4247 * Makefile.in: Regenerate.
4251 * config/tc-s390.c (s390_insn): Avoid incorrect signed/unsigned
4252 comparison in .insn pseudo operation.
4256 * config/obj-coff.c (coff_adjust_section_syms): Use
4257 bfd_get_section_size instead of bfd_get_section_size_before_reloc.
4258 (coff_frob_section): Likewise.
4259 * config/tc-mips.c (md_apply_fix3): Likewise.
4260 * config/obj-elf.c (elf_frob_file): Use bfd_set_section_size.
4261 (elf_frob_file_after_relocs): Likewise.
4265 * config/tc-hppa.c (log2): Rename to exact_log2.
4266 (pa_next_subseg): Delete unused function.
4267 (create_new_space): Mark unused arguments with ATTRIBUTE_UNUSED.
4268 (create_new_subspace): Likewise.
4271 * config/tc-hppa.c (hppa_fix_adjustable): Allow reduction of fake
4272 labels. Fix warning.
4276 * config/tc-mn10300.h (tc_fix_adjustable): Define.
4277 * config/tc-mn10300.c (mn10300_fix_adjustable): Don't adjust debug
4278 or non-merged symbols.
4282 * config/tc-ia64.c (remove_marked_resource): Save, clear and
4283 restore the old slot when inserting srlz.i/srlz.d.
4287 * Makefile.am: Regenerate dependecies.
4288 * Makefile.in: Regenerate.
4289 * config/tc-sh.c (valid_arch): Make unsigned.
4290 (preset_target_arch): Likewise.
4291 (md_begin): Use new architecture flags system.
4292 (get_specific): Likewise.
4293 (assemble_ppi): Likewise.
4294 (md_assemble): Likewise. Also fix error check for bad opcodes.
4295 (md_parse_option): Likewise. Also generate -isa values according
4296 to the table in bfd/cpu-sh.c instead of just constants. Also
4297 allow <arch>-up ISA variants.
4298 (sh_elf_final_processing): Replace if-else chain with a call to
4299 sh_find_elf_flags().
4303 * config/gc-m68k.c(m68k_ip): Convert mode 5 addressing
4304 with zero offset into mode 2 addressing to save a word.
4308 * config/tc-ia64.c (ar_is_in_integer_unit): Removed.
4309 (ar_is_only_in_integer_unit): New.
4310 (ar_is_only_in_memory_unit): New.
4311 (generate_unwind_image): Silence gcc on 32bit host.
4312 (md_assemble): Use ar_is_only_in_integer_unit instead of
4313 ar_is_in_integer_unit. Check AR access.
4317 * config/tc-m68k.c (md_begin): Sort the opcode table into
4319 (m68k_compare_opcode): New function to do the sorting.
4323 * config/m68k-parse.y(operand): Allow for MAC/EMAC mask
4324 addressing on MIT style operands.
4325 * config/m68k-parse.y(yylex): Allow '-&' for predecrement
4327 * config/tc-m68k.c(install_operand): Comment 'G' and 'H' type
4332 * expr.c (operand, operator): Don't reject '++' and '--'.
4336 * config/tc-mips.c (append_insn): Use ISA-encoded addresses in MIPS16
4341 * configure.in: Add ppc-*-lynxos*. Update i386-*-lynxos* to ELF.
4342 * configure: Regenerate.
4346 * dw2gencfi.c (output_cie): Handle dwarf3 format CIE entries.
4350 * configure.in (or32-*-rtems*): Switch to elf.
4351 * configure: Regenerate.
4355 * po/fr.po: Updated French translation.
4359 * doc/as.texinfo (Section): Document G and T flags to .section
4360 directive. Document the extra arguments that the G flag
4361 requires. Document the #tls flag.
4365 * subsegs.c (section_symbol): Create a new section symbol if
4366 the existing one doesn't match.
4367 * symbols.c (symbol_set_bfdsym): Don't reset BFD section symbol.
4371 * config/tc-mips.c (append_insn, mips_emit_delays): Extend -mfix-vr4120
4372 to cope with VR4181A errata MD(1) and MD(4).
4376 * NEWS: Mention .secrel32 for pe[i]-i386.
4380 * config/tc-frv.h (MAX_MEM_FOR_RS_ALIGN_CODE): New.
4381 (HANDLE_ALIGN): New.
4385 * Makefile.am (DIST_SUBDIRS): Define.
4386 * aclocal.m4: Regenerate with automake 1.8.4.
4387 * Makefile.in: Likewise.
4388 * doc/Makefile.in: Likewise.
4392 * config/tc-ia64.c (dot_serialize): Declare.
4393 (dot_serialize): New function.
4394 (md_pseudo_table): Add ".serialize.data" and
4395 ".serialize.instruction" directives.
4399 * messages (as_internal_value_out_of_range): Print a message about
4400 a value being out of range. Be consistent about whether the
4401 values are printed in decimal or hexadecimal.
4402 (as_warn_value_out_of_range): Generate a warning message about an
4404 (as_bad_value_out_of_range): Generate an error message about an
4406 * as.h: Prototype the new functions.
4407 * config/tc-alpha.c (insert_operand): Use new function.
4408 * config/tc-arc.c (arc_insert_operand): Likewise.
4409 * config/tc-mn10200.c (mn10200_insert_operand): Likewise.
4410 * config/tc-mn10300.c (mn10300_insert_operand): Likewise.
4411 * config/tc-ppc.c (ppc_insert_operand): Likewise.
4412 * config/tc-s390.c (s390_insert_operand): Likewise.
4413 * config/tc-v850.c (v850_insert_operand): Likewise.
4417 * configure.in: Set em=linux for frv-*-*linux*.
4418 * configure: Rebuilt.
4419 * config/tc-frv.h (TARGET_FORMAT): Use elf32-frvfdpic if...
4420 (frv_md_fdpic_enabled): New.
4421 * config/tc-frv.c (frv_md_fdpic_enabled): New.
4422 (DEFAULT_FDPIC): New.
4423 (frv_flags): Use DEFAULT_FDPIC.
4424 (frv_pic_flag): Likewise.
4425 (OPTION_NOPIC): New.
4426 (md_longopts): Add -mnopic.
4427 (md_parse_option): Handle it.
4428 (md_show_usage): Add -mfdpic and -mnopic.
4432 * config/tc-m68k.c: Add find_cf_chip to print list of valid
4433 chips for invalid coldfire instructions, rename selectors
4434 for ColdFire sub-variants, add 521x,5249,547x,548x and aliases,
4435 add current_chip to track which chip is referred to(including save/restore),
4436 use current_chip to select control registers, not current_arch.
4437 (md_show_usage): Add new chips.
4438 * doc/c-m68k.texi: Document new command line switches.
4442 * tc-s390.h (md_do_align, HANDLE_ALIGN): Remove.
4443 (NOP_OPCODE): Define.
4444 (s390_align_code): Remove prototype.
4445 * tc-s390.c (s390_align_code): Remove.
4449 * config/tc-ia64.c (make_unw_section_name): Removed.
4450 (start_unwind_section): New function.
4451 (generate_unwind_image): Take const segT instead of const
4453 (dot_handlerdata): Adjusted.
4454 (dot_endp): Likewise.
4458 * config/obj-elf.c (obj_elf_change_section): Allow the
4459 ".note.GNU-stack" section has SHF_EXECINSTR.
4463 * config/obj-elf.c (get_section): Return bfd_boolean.
4464 (obj_elf_change_section): Call bfd_get_section_by_name_if
4465 instead of bfd_map_over_sections.
4469 * config/obj-elf.c (get_section): New function.
4470 (obj_elf_change_section): Support multiple sections with same
4475 * config/tc-arm.c (create_register_alias): Fix typo checking for
4476 case sensitive register aliases.
4477 (co_proc_number): Use error message string in all_reg_maps[]
4479 (cp_reg_required_here): Likewise.
4480 (fp_reg_required_here): Likewise.
4484 * dwarf2dbg.c (dwarf2_finish): Add SEC_DEBUGGING to section flags.
4488 * config/tc-mips.c (HAVE_32BIT_ADDRESSES, append_insn, macro_build)
4489 (load_address, macro, mips_ip, md_parse_option)
4490 (mips_force_relocation, mips_validate_fix, md_apply_fix3)
4491 (s_change_sec, pic_need_relax, tc_gen_reloc): Remove all
4492 embedded-PIC handling, and update comments.
4493 (SWITCH_TABLE): Remove.
4494 * config/tc-mips.h (DIFF_EXPR_OK): Delete.
4495 (enum mips_pic_level): Remove EMBEDDED_PIC.
4496 (EXTERN_FORCE_RELOC): Remove embedded-PIC handling.
4497 (TC_FORCE_RELOCATION): Update comment.
4498 * ecoff.c (ecoff_build_lineno): Add comment about some code that
4499 might be safe to remove now that MIPS embedded-PIC is gone.
4503 * config/obj-som.c (obj_som_init_stab_section): Add new arguments in
4504 call to obj_set_subsection_attributes.
4505 (obj_som_init_stab_section): Likewise.
4506 * config/tc-hppa.c (default_subspace_dict): Add comdat field.
4507 (pa_def_subspaces): Provide comdat default.
4508 (pa_subspace): Handle new "comdat" parameter. Set SEC_LINK_ONCE and
4509 not SEC_IS_COMMON if section is comdat, common or dup_common. Update
4510 calls to create_new_subspace and update_subspace to pass comdat flag.
4511 (create_new_subspace, update_subspace): Add new comdat argument. Use
4512 it in calls to obj_set_subsection_attributes.
4513 * doc/c-hppa.texi (.subspa, .nsubspa): Document new comdat parameter
4514 and use of comdat, common and dup_comm parameters.
4518 * config/obj-elf.c (obj_elf_change_section): Check if the old
4519 group name is NULL before comparison.
4523 * config/tc-mips.h (mips_dwarf2_addr_size): Prototype.
4527 * config/tc-mips.c (s_mipsset): Set default CPU type for .set mipsN.
4531 * config/tc-mips.c (md_longopts): Remove -membedded-pic option.
4532 (OPTION_MEMBEDDED_PIC): Remove.
4533 (OPTION_TRAP, OPTION_BREAK, OPTION_EB, OPTION_EL)
4534 (OPTION_FP32, OPTION_GP32, OPTION_CONSTRUCT_FLOATS)
4535 (OPTION_NO_CONSTRUCT_FLOATS, OPTIONS_FP64, OPTION_GP64)
4536 (OPTION_RELAX_BRANCH, OPTION_NO_RELAX_BRANCH)
4537 (OPTION_ELF_BASE): Renumber.
4538 (md_parse_option): Remove OPTION_MEMBEDDED_PIC handling.
4539 (md_show_usage): Remove mention of -membedded-pic.
4540 * doc/as.texinfo: Remove mention of -membedded-pic.
4544 * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Remove.
4545 * config/tc-mips.c (RDATA_SECTION_NAME, mips_target_format): Remove
4547 (md_begin, mips_ip, md_parse_option, s_change_sec, s_option,
4548 s_abicalls, nopic_need_relax, tc_gen_reloc): Remove uses of
4549 USE_GLOBAL_POINTER_OPT.
4553 * config/tc-mips.c (macro): One more use of load_delay_nop.
4557 * config/tc-mips.c (load_delay_nop): New function.
4558 (load_address, macro): Use load_delay_nop() to build a nop
4559 which can be omitted with gpr_interlocks.
4563 * config/tc-mips.c (hilo_interlocks, gpr_interlocks,
4564 cop_interlocks): Remove superfluous CPU entries.
4568 * config/tc-arm.c (mav_parse_offset): Value must be multiple of 4.
4572 * NEWS: Added support for EMAC instructions and MAC/EMAC
4574 * config/m68k-parse.h: Add ACC[123], ACCEXT{01,23}, MAC/EMAC
4575 scale factor tokens, trailing_ampersand to mark mask addressing
4576 for MAC/EMAC instructions.
4577 * config/m68k-parse.y: Add options_ampersand clause, '<<',
4579 (yylex): Handle '>', '<', and '&' following '+'.
4580 * config/tc-m68k.c: Set mcfmac/mcfemac on appropriate ColdFire
4581 architectures in archs[].
4582 (m68k-ip): Add '4', 'e', 'g', 'i', cases to handle mask addressing
4583 for MAC/EMAC instructions, ACC[0123], ACCEXT{01,23}, and '<<'/'>>'
4585 (m68k_ip): Handle trailing '&' on MAC/EMAC insns.
4586 (install_operand): Fix 'n' case, Add 'F', 'f', 'G', 'H', 'I', ']'
4588 Add EMAC operands to init_table[].
4592 * config/tc-arm.h (enum mstate): Move here, add MAP_UNDEFINED
4594 (TC_SEGMENT_INFO_TYPE): Define to enum mstate.
4595 * config/tc-arm.c (enum mstate): Delete from here.
4596 (mapping_state): Remove the static mapstate variable and instead
4597 store the state in the segment. This allows a per-section mapping
4598 state. Handle and ignore MAP_UNDEFINED states.
4599 (arm_elf_change_section): Get the current mapping state from the
4601 (s_ltorg): Set the mapping state to MAP_DATA.
4602 (arm_cleanup): Use arm_elf_change_section to get the mapping state
4603 for each pool as it is emitted.
4607 * config/tc-arm.h: Formatting tidy ups.
4611 * NEWS: Note that MIPS -membedded-pic option is deprecated.
4615 * config/tc-i386.h [TE_PE] (TC_CONS_FIX_NEW): Define.
4616 * config/tc-i386.c (md_pseudo_table) [TE_PE]: Add "secrel32".
4617 [TE_PE] (O_secrel): Define.
4618 [TE_PE] (x86_pe_cons_fix_new): New.
4619 [TE_PE] (pe_directive_secrel): Likewise.
4620 (tc_gen_reloc) [TE_PE]: Support BFD_RELOC_32_SECREL.
4624 * config/tc-mips.c (mips_dwarf2_addr_size): Revert part
4625 of previous patch for fix in gcc.
4629 * config/tc-xtensa.c (xg_assembler_literal): Fix a typo.
4633 * read.c (do_align): Call md_flush_pending_output, if defined.
4637 * expr.c (operand): Correct checks for ++ and --.
4641 * config/tc-generic.c: Add some comments.
4645 * doc/c-mips.texi (-m{no-,}fix-vr4120): Renamed from
4646 -{no-}mfix-vr4122-bugs.
4647 * config/tc-mips.c (mips_fix_vr4120): Renamed from mips_fix_4122_bugs.
4648 (append_insn, mips_emit_delays): Update accordingly.
4649 (OPTION_FIX_VR4120, OPTION_NO_FIX_VR4120): Renamed from *VR4122.
4650 (md_longopts): Change -{no-,}mfix-vr4122-bugs to -m{no-,}fix-vr4120.
4651 (md_parse_option): Update after above changes.
4652 (md_show_usage): Add -mfix-vr4120.
4656 * doc/as.texinfo (Sub-Sections): Conditionalize COFF-specific use
4657 of .section directive; add a reference to the ELF .subsection
4662 * config/tc-m32r.c (md_assemble): Fixed infinite loop bug
4667 * Makefile.am: Remove mips from aout targets.
4668 * Makefile.in: Regenerate.
4669 * configure.in: Remove mips-dec-bsd* target.
4670 * configure: Regenerate.
4675 * config/tc-ppc.c (ppc_elf_suffix): Add valid32 and valid64 fields
4676 to struct map_bfd. Adjust MAP macro, and define MAP32, MAP64.
4677 Update "mapping". Restrict some @ modifiers to 32 bit.
4682 * config/tc-sh.c (dont_adjust_reloc_32): New variable.
4683 (sh_fix_adjustable): Avoid adjusting BFD_RELOC_32 when
4684 dont_adjust_reloc_32 is set.
4685 (md_longopts): Add option -renesas.
4686 (md_parse_option, md_show_usage): Likewise.
4687 * doc/c-sh.texi: Likewise.
4691 * config/tc-dlx.c (md_assemble): set fx_no_overflow flag for
4692 hi16 and lo16 fixS structs.
4693 (md_assemble): generate bit_fixS for RELOC_DLX_LO16 in
4694 exactly the same way as for RELOC_DLX_REL16.
4695 (machine_ip): properly respect LO flag in the_insn and
4696 output RELOC_DLX_LO16 rather than RELOC_DLX_16.
4697 (md_apply_fix3): apply RELOC_DLX_LO16.
4701 Remove long-obsolete MPW support.
4702 * mpw-config.in, mpw-make.sed, mac-as.r: Remove files.
4703 * configure.in: Remove mention of ppc-*-mpw* config.
4704 * configure.in: Likewise.
4708 * config/tc-arm.c (meabi_flags): Make its use conditional upon
4709 OBJ_ELF being defined.
4713 * config/obj-aout.c (obj_aout_type): Remove #ifdef BFD_ASSEMBLER code.
4717 * config/tc-arm.c (meabi_flags): New variable.
4718 (arm_parse_eabi): New function.
4719 (md_begin): Set flags for EABI v3.
4721 (arm_long_opts): Add meabi.
4722 * doc/as.texinfo <ARM>: Document -meabi.
4723 * doc/c-arm.texi: Ditto.
4727 * config/tc-xtensa.c (xtensa_post_relax_hook): Create literal
4728 tables even when use_literal_section flag is not set.
4732 * config/tc-sh.c: Remove trailing whitespace.
4736 * doc/c-cris.texi (CRIS-Opts): Document --no-mul-bug-abort,
4737 --mul-bug-abort and the default behavior.
4738 * config/tc-cris.c (cris_insn_kind): New member CRIS_INSN_MUL.
4739 (err_for_dangerous_mul_placement): New variable.
4740 (STATE_MUL, OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): New
4742 (md_cris_relax_table): Have placeholder for STATE_MUL.
4743 (md_longopts): New options --mul-bug-abort and --no-mul-bug-abort.
4744 (cris_relax_frag) <case ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: New
4746 (md_estimate_size_before_relax) <case ENCODE_RELAX (STATE_MUL,
4747 STATE_BYTE)>: Ditto.
4748 (md_convert_frag) <ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: Check
4749 alignment and position of this frag, emit error message if
4751 (md_assemble): For a multiply insn and when checking it,
4752 transform the current frag into a special frag for that purpose.
4753 (md_parse_option) <case OPTION_MULBUG_ABORT_OFF, case
4754 OPTION_MULBUG_ABORT_ON>: Handle new options.
4758 * config/tc-xtensa.c (mark_literal_frags): New function.
4759 (xtensa_move_literals): Call mark_literal_frags for all literal
4760 segments, including init and fini literal segments.
4761 (xtensa_post_relax_hook): Swap use of xt_insn_sec and xt_literal_sec.
4765 * tc-hppa.c (cons_fix_new_hppa): Check for PC relative base type.
4766 (pa_comm): Set BSF_OBJECT in symbol flags.
4770 * Makefile.am: Run "make dep-am".
4771 * Makefile.in: Regenerate.
4772 * doc/Makefile.in: Regenerate.
4773 * config.in: Regenerate.
4774 * po/gas.pot: Regenerate.
4778 * read.c (read_a_source_file): Use demand_empty_rest_of_line.
4779 (demand_empty_rest_of_line): Issue an error here.
4780 (ignore_rest_of_line): Silently skip to end.
4781 (demand_copy_string): Issue an error, not warning.
4783 * config/obj-elf.c (obj_elf_section_name): Likewise.
4784 (obj_elf_section): Likewise.
4785 * config/tc-arc.c (arc_extoper): Remove bogus NULL checks.
4786 (arc_extinst): Likewise.
4787 * config/tc-ia64.c (dot_saveb): Use demand_empty_rest_of_line.
4788 (dot_spill): Likewise.
4789 (dot_unwabi): Likewise.
4790 (dot_prologue): Likewise.
4792 * expr.c (operand): Reject ++ and --.
4793 (operator): Likewise.
4797 * config/tc-sh.c: Include dw2gencfi.h.
4798 (sh_cfi_frame_initial_instructions): New function.
4799 (sh_regname_to_dw2regnum): Likewise.
4800 * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Move to the end of
4802 (TARGET_USE_CFIPOP): Define.
4803 (tc_cfi_frame_initial_instructions): Likewise.
4804 (tc_regname_to_dw2regnum): Likewise.
4805 (DWARF2_DEFAULT_RETURN_COLUMN, DWARF2_CIE_DATA_ALIGNMENT): Likewise.
4806 * Makefile.am: Update dependencies.
4807 * Makefile.in: Regenerate.
4811 * configure.in: Switch sh-*-rtems* to ELF. Add sh-*-rtemscoff*.
4812 * configure: Regenerate.
4816 * read.c (s_leb128): Call md_flush_pending_output.
4820 * config/tc-i386.c (output_insn): Handle PadLock instructions.
4821 * config/tc-i386.h (CpuPadLock): New define.
4822 (CpuUnknownFlags): Added CpuPadLock.
4826 * doc/c-hppa.texi (HPPA Directives): Fix typo.
4830 * dw2gencfi.c (output_cie): Align length to 4 byte boundary.
4831 (cfi_finish): Likewise for fde.
4835 * config/tc-ia64.c (md_assemble): Properly handle NULL
4837 (ia64_handle_align): Don't abort if failed to add a stop bit.
4841 * Makefile.in: Regenerated.
4842 * aclocal.m4: Likewise.
4843 * configure: Likewise.
4844 * doc/Makefile.in: Likewise.
4848 * config/tc-ia64.c (dot_align): New.
4849 (ia64_do_align): Make it static.
4850 (md_pseudo_table): Use "dot_align" for "align".
4851 (ia64_md_do_align): Don't set align_frag here.
4852 (ia64_handle_align): Add a stop bit to the previous bundle if
4855 * config/tc-ia64.h (ia64_do_align): Removed.
4859 * config/tc-sh.c (md_parse_option): Add -isa=sh4-nofpu and
4860 -isa=sh4-nommu-nofpu options. Adjust help messages accordingly.
4861 (sh_elf_final_processing): Output BFD type sh4_nofpu if that is
4862 the most general type or the user specifically requested it.
4863 (md_assemble): Add a new error message for when an instruction
4864 is understood, but is not allowed due to an -isa option.
4868 * config/tc-ia64.c (align_frag): New.
4869 (md_assemble): Set the tc_frag_data field in align_frag for
4870 IA64_OPCODE_FIRST instructions.
4871 (ia64_md_do_align): Set align_frag.
4872 (ia64_handle_align): Add a stop bit if needed.
4874 * config/tc-ia64.h (TC_FRAG_TYPE): New.
4875 (TC_FRAG_INIT): New.
4879 * config/tc-frv.c (fr400_audio): New variable.
4880 (md_parse_option, md_show_usage): Add -mcpu=fr405 and -mcpu=fr450.
4881 (md_parse_option): Set fr400_audio for -mcpu=fr400 and -mcpu=fr405.
4882 (target_implements_insn_p): New function.
4883 (md_assemble): Report an error if the processor doesn't implement
4888 * config/tc-m32r.c (md_longopts): Added -no-bitinst option.
4889 (md_parse_option): Ditto.
4890 (OPTION_NO_SPECIAL_M32R): Added.
4891 (md_show_usage): Document it.
4892 (enable_speial_m32r): Changed a default value from 0 to 1.
4893 * doc/c-m32r.texi: Document the -no-bitinst option.
4897 * config/tc-sh.c (get_operand): Revert previous delta.
4898 (tc_gen_reloc): Check for an unknown reloc type before processing
4903 * config/tc-s390.c (s390_insn): Correct range check for opcode in
4904 .insn pseudo operation.
4908 * config/tc-sh.c (get_operand): In case of #Imm, check has been
4909 added for wrong syntax.
4913 * config/tc-mips.c (mips_dwarf2_addr_size): New.
4914 * config/tc-mips.h (DWARF2_ADDR_SIZE): Use.
4918 * config/tc-sh.c (build_Mytes): Add REG_N_D and REG_N_B01
4919 nibble types to assembler.
4923 * config/tc-iq2000.c: Add missing \n\ in multiline string literal.
4927 * config/tc-ia64.c (slot_index): New arg before_relax. Use instead of
4929 (fixup_unw_records): New arg before_relax. Pass to slot_index.
4930 (ia64_estimate_size_before_relax): New.
4931 (ia64_convert_frag): Pass 0 to fixup_unw_records. Add comment.
4932 (generate_unwind_image): Pass 1 to fixup_unw_records.
4933 * config/tc-ia64.h (ia64_estimate_size_before_relax): Declare.
4934 (md_estimate_size_before_relax): Call ia64_estimate_size_before_relax.
4938 * stabs.c (generate_asm_file): Avoid warning about use of
4939 uninitialized variable.
4943 * config/tc-ia64.c (ia64_flush_insns): In addition to prologue,
4944 body, and endp, allow unwind records which do not have a "t"
4945 (time/instruction) field.
4949 * config/tc-arm.c (do_mav_dspsc_1): Correct offset of CRn.
4950 (do_mav_dspsc_2): Likewise.
4951 Fix accumulator registers move opcodes.
4956 * dwarf2dbg.c (get_filenum): Do not read beyond allocated memory.
4960 * config/tc-ia64.h (ia64_frob_symbol): New declaration.
4961 (tc_frob_symbol): New macro definition.
4962 * config/tc-ia64.c (ia64_frob_symbol): New routine.
4966 * config/tc-arm.c (md_begin): Mark .note.gnu.arm.ident as
4971 * read.h (IGNORE_OPCODE_CASE): Do not define. Replace with ...
4972 (TC_CASE_SENSITIVE): ... this.
4973 * read.c: Replace IGNORE_OPCODE_CASE with TC_CASE_SENSITIVE.
4974 * doc/internals.texi (TC_CASE_SENSITIVE): Document.
4978 * config/tc-ia64.c (dot_endp): Delete call to output_endp.
4979 (generate_unwind_image): Re-add it here.
4983 * dwarf2dbg.c (DWARF2_ADDR_SIZE): Remove trailing ';'
4984 * read.h (SKIP_WHITESPACE): Turn into an expression.
4985 * read.c (read_a_source_file): A pseudo is removed by having a
4990 * config/tc-ia64.c (output_endp): New.
4991 (count_bits): Delete.
4992 (ia64_flush_insns, process_one_record, optimize_unw_records): Handle
4993 endp unwind records.
4994 (fixup_unw_records): Handle endp unwind records. Delete code for
4995 shortening prologue regions not followed by a body record.
4996 (dot_endp): Call add_unwind_entry to emit endp unwind record.
4997 * config/tc-ia64.h (unw_record_type): Add endp.
5001 * config/tc-ia64.c (ia64_convert_frag): Call md_number_to_chars to
5002 fill padding bytes with zeroes.
5003 (emit_one_bundle): New locals last_ptr, end_ptr. Rewrite code that
5004 sets unwind_record slot_number and slot_frag fields.
5008 * config/tc-mips.c (add_got_offset_hilo): New function.
5009 (macro): Use load_register() and add_got_offset_hilo() to load
5010 constants instead of hardcoding code sequences throughout.
5014 * config/tc-ia64.c (emit_one_bundle): Add proper indentation.
5018 * config/tc-m68k.h (EXTERN_FORCE_RELOC): Handle m68k-uclinux specially,
5020 * config/tc-m68k.c (RELAXABLE_SYMBOL): Use EXTERN_FORCE_RELOC instead
5021 of hard-coded test for TARGET_OS=elf.
5025 * config/tc-mips.c (hilo_interlocks): Change definition
5026 so that MIPS32, MIPS64 and later ISAs are included, along with
5027 the already-included machines. Update comments.
5031 * config/tc-arm.c (tc_gen_reloc): Improve error message for
5032 undefined local labels.
5036 * config/tc-mips.c (load_address, macro): Update comments about
5037 NewABI GP relaxation.
5041 * config/tc-mips.c (macro_build): Remove place and counter arguments.
5042 (mips_build_lui, macro_build_ldst_constoffset): Likewise.
5043 (mips16_macro_build, macro_build_jalr): Remove counter argument.
5044 (set_at, load_register, load_address, move_register): Likewise.
5045 (load_got_offset, add_got_offset): Likewise.
5046 Update all calls and tidy accordingly.
5050 * config/tc-mips.c (RELAX_ENCODE): Remove WARN argument.
5051 (RELAX_FIRST, RELAX_SECOND): Turn into 8-bit quantities.
5052 (RELAX_USE_SECOND): Bump to 0x10000.
5053 (RELAX_SECOND_LONGER, RELAX_NOMACRO, RELAX_DELAY_SLOT): New flags.
5054 (mips_macro_warning): New variable.
5055 (md_assemble): Wrap macro expansion in macro_start() and macro_end().
5056 (s_cpload, s_cpsetup, s_cprestore, s_cpreturn): Likewise.
5057 (relax_close_frag): Set mips_macro_warning.first_frag. Adjust use
5059 (append_insn): Update mips_macro_warning.sizes.
5060 (macro_start, macro_warning, macro_end): New functions.
5061 (macro_build): Don't emit warnings here.
5062 (macro_build_lui, md_estimate_size_before_relax): ...or here.
5063 (md_convert_frag): Check for cases where one macro alternative
5064 needs a warning and the other doesn't. Emit a warning if the
5065 longer sequence was chosen.
5069 * config/tc-mips.h (tc_frag_data_type, TC_FRAG_TYPE): Remove.
5070 * config/tc-mips.c (RELAX_ENCODE): Take three arguments: the size of
5071 the first sequence, the size of the second sequence, and a flag
5072 that says whether we should warn.
5073 (RELAX_OLD, RELAX_NEW, RELAX_RELOC[123]): Delete.
5074 (RELAX_FIRST, RELAX_SECOND): New.
5075 (mips_relax): New variable.
5076 (relax_close_frag, relax_start, relax_switch, relax_end): New fns.
5077 (append_insn): Remove "place" argument. Use mips_relax.sequence
5078 rather than "place" to check whether we're expanding the second
5079 alternative of a relaxable macro. Remove redundant check for
5080 branch relaxation. If generating a normal insn, and there
5081 is not enough room in the current frag, call relax_close_frag()
5082 to close it. Update mips_relax.sizes[]. Emit fixups for the
5083 second version of a relaxable macro. Record the first relaxable
5084 fixup in mips_relax. Remove tc_gen_reloc workaround.
5085 (macro_build): Remove all uses of "place". Use mips_relax.sequence
5086 in the same way as in append_insn.
5087 (mips16_macro_build): Remove "place" argument.
5088 (macro_build_lui): As for macro_build. Don't drop the add_symbol
5089 when generating the second version of a relaxable macro.
5090 (load_got_offset, add_got_offset): New functions.
5091 (load_address, macro): Use new relaxation machinery. Remove
5092 tc_gen_reloc workarounds.
5093 (md_estimate_size_before_relax): Set RELAX_USE_SECOND if the second
5094 version of a relaxable macro is needed. Return -RELAX_SECOND if the
5095 first version is needed.
5096 (tc_gen_reloc): Remove relaxation handling.
5097 (md_convert_frag): Go through the fixups for a relaxable macro and
5098 mark those that belong to the unneeded alternative as done. If the
5099 second alternative is needed, adjust the fixup addresses to account
5100 for the deleted first alternative.
5104 * frags.h (frag_room): Declare.
5105 * frags.c (frag_room): New function.
5106 * doc/internals.texi: Document it.
5110 * config/tc-mips.c (append_insn): Don't do r3900 interlock
5111 optimization for -mtune=r3900, as this will break on other CPUs.
5115 * config/tc-m68hc11.c (build_indexed_byte): movb and movw cannot
5116 be relaxed, use fixup.
5117 (md_apply_fix3): Use 5 bit reloc from movb and movw fixup.
5121 * config/tc-sparc.c (sparc_ip): Disallow %f32-%f63 for single
5126 * config/tc-mips.c (append_insn): Properly detect variant frags
5127 that preclude swapping of relaxed branches. Correctly swap
5128 instructions between frags when dealing with relaxed branches.
5132 * acinclude.m4: Quote names of macros to be defined by AC_DEFUN
5134 * aclocal.m4: Regenerate.
5135 * configure: Regenerate.
5139 * config/tc-h8300.c (build_bytes): Apply relaxation to bit
5144 * config/tc-mips.c (macro_build_jalr): When adding an R_MIPS_JALR
5145 reloc, reserve space for the delay slot as well as the jalr itself.
5149 * config/tc-arm.c (do_vfp_reg2_from_sp2): Rename from do_vfp_sp_reg2.
5150 (do_vfp_sp2_from_reg2): New function.
5152 (do_vfp_dp_from_reg2): Check return values properly.
5156 * config/tc-mips.c (warn_nops): Remove static variable.
5157 (macro): Remove test of warn_nops.
5158 (md_shortops): Remove 'n'.
5159 (md_parse_option): Remove 'n' case.
5160 (md_show_usage): Remove -n.
5161 * doc/as.texinfo (Overview): Remove MIPS -n option.
5162 * doc/c-mips.texi (MIPS Opts): Remove mention -n.
5163 * NEWS: Mention removal of MIPS -n option.
5165 * config/tc-mips.c (ISA_HAS_COPROC_DELAYS): Remove.
5166 (cop_interlocks): Check ISA level.
5167 (cop_mem_interlocks): Define.
5168 (reg_needs_delay): Check cop_interlocks rather than
5169 ISA_HAS_COPROC_DELAYS.
5170 (append_insn): Likewise. Use cop_mem_interlocks rather than
5171 directly checking mips_opts.isa.
5172 (mips_emit_delays): Likewise.
5176 * config/tc-ia64.c (unwind): Move next_slot_number and
5177 next_slot_frag to ...
5178 (unw_rec_list): Here.
5179 (free_list_records): Removed.
5180 (output_unw_records): Likewise.
5181 (generate_unwind_image): Make it void.
5182 (alloc_record): Initialize next_slot_number and next_slot_frag.
5183 (slot_index): Take .org, .space and .align into account.
5184 (fixup_unw_records): Don't set slot_number to 0. Use
5185 list->next_slot_number and list->next_slot_frag instead of
5186 unwind.next_slot_number and unwind.next_slot_frag.
5187 (ia64_convert_frag): New.
5188 (generate_unwind_image): Generate a rs_machine_dependent frag
5190 (emit_one_bundle): Use list->next_slot_number and
5191 list->next_slot_frag instead of unwind.next_slot_number and
5192 unwind.next_slot_frag.
5194 * config/tc-ia64.h (md_convert_frag): Defined as
5196 (md_estimate_size_before_relax): Defined as (f)->fr_var.
5201 * config/tc-frv.h (md_apply_fix3): Don't define.
5202 * config/tc-frv.c (md_apply_fix3): New. Shift/truncate %hi/%lo
5204 * config/tc-frv.h (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
5206 * config/tc-frv.c (line_separator_chars): Add `!'.
5208 * config/tc-frv.c (md_assemble): Clear insn upfront.
5210 * config/tc-frv.c (OPTION_FDPIC): New macro.
5211 (md_longopts): Add mfdpic.
5212 (md_parse_option): Handle it.
5214 * config/tc-frv.c (md_cgen_lookup_reloc) <FRV_OPERAND_D12,
5215 FRV_OPERAND_S12>: Use reloc type encoded in fix-up.
5216 (frv_pic_ptr): Parse funcdesc.
5220 * doc/as.texinfo: Let texi2pod parse asconfig.texi and
5221 gasver.texi. Remove duplicate symbol definitions for texi2pod.
5225 * Makefile.am (Makefile): Move the dependency on
5226 $(BFDDIR)/configure.in to...
5227 (CONFIG_STATUS_DEPENDENCIES): ... here.
5228 (AUTOMAKE_OPTIONS): Require automake 1.8.
5229 * Makefile.in: Regenerate.
5230 * doc/Makefile.am (BASEDIR, BFDDIR): Define.
5231 (CONFIG_STATUS_DEPENDENCIES): Add a dependency on
5232 $(BFDDIR)/configure.in here as well.
5233 * doc/Makefile.in: Regenerate.
5237 * Makefile.am (install, install-info, RECURSIVE_TARGETS): Remove.
5238 * Makefile.in: Regenerate.
5239 * aclocal.m4: Regenerate.
5240 * doc/Makefile.am (install, install-info): Remove.
5241 (install-data-local): A new hook for install-info.
5242 (AUTOMAKE_OPTIONS): Require automake 1.8.
5243 * doc/Makefile.in: Regenerate.
5247 * doc/c-sh.texi: Update description about floating point behavior
5252 * configure.in: Add m68k-uClinux target.
5253 * configure: Regenerate.
5255 For older changes see ChangeLog-0203
5261 version-control: never
projects / binutils.git / blob