3 * config/tc-arm.c (arm_adjust_symtab): Fetch elf_sym's binding
9 * configure.tgt: Accept any C library to accompany a GNU Linux
10 implementation, not just the GNU C library.
11 * configure.in: Likewise.
12 * configure: Regenerate.
16 * config/tc-tic30.c: Convert to ISO C90 formatting.
17 * config/tc-tic30.h: Convert to ISO C90 formatting.
18 * config/tc-pdp11.c: Convert to ISO C90 formatting.
19 * config/atof-vax.c: Convert to ISO C90 formatting.
23 * config/tc-mips.c (mips_frob_file): Sort BFD_RELOC_MIPS16_LO16
24 relocations correctly as well.
25 (mips_fix_adjustable): Don't make BFD_RELOC_MIPS16_LO16
26 relocations in mergeable sections section-relative either.
30 * config/tc-xtensa.c (md_apply_fix3): Recognize XTENSA_PLT relocations.
34 * config/tc-sh.c (sh_elf_final_processing): Fix compile time
35 warning by providing a prototype for sh_symbian_find_elf_flags.
37 * cgen.c (gas_cgen_parse_operand): Fix typo introduced by
42 * configure.tgt: Handle setting of bfd_gas for fmt=multi targets
43 along with other formats that set bfd_gas. Remove unnecessary
44 setting of bfd_gas. Delete strongarm cases in generic_target
50 * config/tc-m68k.c (TRUNC, SEXT): Define.
51 (issbyte, isubyte, issword, isuword, isbyte, isword): Use the above.
52 (m68k_ip): Truncate or sign extend expressions as appropriate.
54 (md_apply_fix3): Use SEXT.
58 * Makefile.am (OBJ_FORMAT_CFILES): Prune config/obj-vms.c.
59 (OBJ_FORMAT_HFILES): Prune config/obj-vms.h.
60 (obj-vms.o): Delete rule.
62 * Makefile.in: Regenerate.
63 * aclocal.m4: Regenerate.
64 * doc/Makefile.in: Regenerate.
65 * po/POTFILES.in: Regenerate.
69 * config/tc-m68k.c (m68k_elf_final_processing): Set file specific
70 flag for coldfire v4e.
74 * config/tc-xtensa.c (xg_apply_tentative_value): Rename to
75 xg_apply_fix_value and return a value to indicate success.
76 (md_pcrel_from): Skip check of fx_done. Return 0 if not PC-relative.
77 (xtensa_force_relocation): Remove checks for VTABLE relocs.
78 (xtensa_validate_fix_sub): New.
79 (xtensa_fix_adjustable): Remove check for external or weak symbols.
80 (tc_gen_reloc): Move code to handle difference of symbols and code to
81 apply tentative fix values to ...
82 (md_apply_fix3): ...here. Enable standard overflow checks for simple
83 8, 16, and 32 bit relocations. Apply fixes for slot-specific
84 relocations when linkrelax flag is not set.
85 * config/tc-xtensa.h (xtensa_validate_fix_sub): Add prototype.
86 (TC_FORCE_RELOCATION_SUB_SAME, TC_VALIDATE_FIX_SUB): Define.
90 * config/tc-i386.c (i386_scale): Beautify error message.
91 (Intel syntax comments): Update.
92 (struct intel_parser_s): Add fields in_offset, in_bracket, and
94 (intel_e04_1, intel_e05_1, intel_e05_1, intel_e09_1, intel_e10_1):
96 (intel_bracket_expr): Declare.
97 (i386_intel_operand): Initialize new intel_parser fields. Wrap most
98 of the function body in a loop allowing to split an operand into two.
99 Replace calls to malloc and checks of it returning non-NULL with
100 calls to xmalloc/xstrdup.
101 (intel_expr): SHORT no longer handled here. Add comment indicating
102 comparison ops need implementation.
103 (intel_e04, intel_e04_1): Combine, replace recursion with loop.
104 Check right operand of - does not specify a register when parsing
105 the address of a memory reference.
106 (intel_e05, intel_e05_1): Combine, replace recursion with loop.
107 Check operands do not specify a register when parsing the address of
109 (intel_e06, intel_e06_1): Likewise.
110 (intel_e09, intel_e09_1): Combine, replace recursion with loop. Also
111 handle SHORT as well as unary + and -. Don't accept : except for
112 segment overrides or in direct far jump/call insns.
113 (intel_brack_expr): New.
114 (intel_e10, intel_e10_1): Combine, replace recursion with loop. Use
116 (intel_e11): Replace chain of if/else-if by switch, alloing fall-
117 through in certain cases. Use intel_brack_expr. Add new diagnostics.
118 Allow symbolic constants as register scale value.
119 (intel_get_token): Replace call to malloc and check of return value
120 with call to xmalloc. Change handling for FLAT to match MASM's.
121 (intel_putback_token): Don't try to back up/free current token if
126 * configure.tgt: Set emulation for arm-*-eabi*.
127 * config/tc-arm.c (meabi_flags): Check EABI_DEFAULT.
128 * config/te-armeabi.h: New file.
129 * config/te-armlinuxeabi.h (EABI_DEFAULT): Define.
130 * config/te-symbian.h: Include "te-armeabi.h".
134 * cgen.c (gas_cgen_parse_operand): Copy opinfo parameter into a
135 local variable in case it is clobbered by the setjmp.
140 * configure.in (werror): New switch: Add -Werror to the
141 compiler command line. Enabled by default. Disable via
143 * configure: Regenerate.
147 * config/obj-coff.h: Convert to ISO C90 formatting.
148 * config/obj-coff.c: Convert to ISO C90 formatting.
152 * config/tc-arm.c (do_mla): Rename to do_mlas, take second
153 is_mls parameter; do not diagnose Rm==Rd when is_mls.
154 (do_mla, do_mls, five_bit_unsigned_immediate, bfci_lsb_and_width)
155 (do_bfc, do_bfi, do_bfx, do_rbit, do_mov16, do_ldsttv4): New functions.
156 (insns): Add ARMv6T2 instructions:
157 bfc bfi mls movw movt rbit sbfx ubfx ldrht ldrsht ldrsbt strht.
158 (arm_archs): Add V6T2 variants.
162 * NEWS: Add cutoff for changes in 2.16 release.
166 * expr.c (operand): Merge handling of unary + into that for unary
171 * config/tc-mips.c: Include dw2gencfi.h.
172 (mips_cfi_frame_initial_instructions): New.
173 * config/tc-mips.h (TARGET_USE_CFIPOP): Define.
174 (tc_cfi_frame_initial_instructions): Ditto.
175 (DWARF2_DEFAULT_RETURN_COLUMN): Ditto.
176 (DWARF2_CIE_DATA_ALIGNMENT): Ditto.
177 * Makefile.am: Update dependencies.
178 * Makefile.in: Regenerate.
182 * po/es.po: Commit new Spanish translation.
186 * po/tr.po: Commit new Turkish translation.
190 * config/tc-arm.c (tinsns): Add ARMv6K instructions sev, wfe,
195 * config-gas.com: Mark vax-vms as obsolete.
196 * configure.in: Remove fmt=vms support.
197 * config.in: Regenerate.
198 * configure: Regenerate.
199 * config/obj-vms.h, config/obj-vms.c, config/vms-conf.h: Remove.
203 * config/tc-xtensa.c (finish_vinsn): Include the last instruction slot
204 when checking if xg_resolve_literals needs to be called.
205 * config/tc-xtensa.h: Fix spelling typo in a comment.
209 * config/tc-tic54x.h (tic54x_macro_info): Change parameter type.
210 * config/tc-tic54x.c (tic54x_macro_info): Likewise. Replace hand-
211 crafted structure declarations with the types from macro.h.
215 * config/tc-mips.c (s_cpsetup): Use '__gnu_local_gp' instead of '_gp'
216 for -mno-shared optimization.
220 * config/tc-mips.c (MAX_VR4130_NOPS, MAX_DELAY_NOPS): New macros.
221 (MAX_NOPS): Bump to 4.
222 (mips_fix_vr4130): New variable.
223 (nops_for_vr4130): New function.
224 (nops_for_insn): Use MAX_DELAY_NOPS rather than MAX_NOPS. Use
225 nops_for_vr4130 if working around VR4130 errata.
226 (OPTION_FIX_VR4130, OPTION_NO_FIX_VR4130): New macros.
227 (md_longopts): Add -mfix-vr4130 and -mno-fix-vr4130.
228 (md_parse_option): Handle them.
229 (md_show_usage): Print them.
230 * doc/c-mips.texi: Document -mfix-vr4130 and -mno-fix-vr4130.
234 * config/tc-mips.c (append_insn): Remove cop_interlocks test from
239 * config/tc-mips.h (mips_flush_pending_output): Delete.
240 (mips_emit_delays): Declare.
241 (md_flush_pending_output): Use mips_emit_delays.
242 * config/tc-mips.c (mips_no_prev_insn): Remove parameter; always forget
243 the previous instructions.
244 (md_begin, append_insn, md_parse_option): Update callers.
245 (mips_emit_delay): Remove parameter. Move INSNS != 0 code to
247 (mips_align, s_change_sec, s_cons, s_float_cons, s_gpword)
248 (s_gpdword): Update callers.
249 (start_noreorder, end_noreorder): New functions.
250 (macro, macro2, mips16_macro, s_mipsset): Use them instead of
251 manipulating mips_opts or prev_nop_frag directly.
252 (mips_flush_pending_output): Delete.
256 * config/tc-mips.c (mips_move_labels): New function, taken from...
257 (append_insn, mips_emit_delays): ...here.
261 * config/tc-mips.c (MAX_NOPS): New macro.
262 (history): Resize to 1 + MAX_NOPS.
263 (fix_vr4120_class): New enumeration.
264 (vr4120_conflicts): New variable.
265 (init_vr4120_conflicts): New function.
267 (insn_uses_reg): Constify first argument.
268 (classify_vr4120_insn, insns_between, nops_for_insn, nops_for_sequence)
269 (nops_for_insn_or_target): New functions.
270 (append_insn): Use the new nops_for_* functions instead of inline
271 delay checks. Generalize prev_nop_frag handling to handle an
272 arbitrary history length. Insert nops into the history buffer
273 once the number of nops in prev_nop_frag is fixed.
274 (emit_delays): Use nops_for_insn instead of inline delay checks.
278 * config/tc-mips.c (append_insn): Remove now-redundant nops != 0
279 check from branch delay code. Remove unnecessary check for branches.
283 * config/tc-mips.c (dummy_opcode): Delete.
284 (nop_insn, mips16_nop_insn): New variables.
285 (NOP_INSN): New macro.
286 (insn_length, create_insn, install_insn, move_insn, add_fixed_insn)
287 (add_relaxed_insn, insert_into_history, emit_nop): New functions.
288 (md_begin): Initialize nop_insn and mips16_nop_insn.
289 (append_insn): Use the new emit_nop function to add nops, recording
290 them in the history buffer. Use add_fixed_insn or add_relaxed_insn
291 to reserve room for the instruction and install_insn to install the
292 final form. Use insert_into_history to record the instruction in
293 the history buffer. Use move_insn to do delay slot filling.
294 (mips_emit_delays): Use add_fixed_insn instead of the emit_nop macro.
295 (macro_build, mips16_macro_build, macro_build_lui, mips_ip)
296 (mips16_ip): Use create_insn to initialize mips_cl_insns.
300 * config/tc-mips.c (INSERT_BITS, EXTRACT_BITS, INSERT_OPERAND)
301 (EXTRACT_OPERAND, MIPS16_INSERT_OPERAND, MIPS16_EXTRACT_OPERAND): New.
302 (insn_uses_reg, reg_needs_delay, append_insn, macro_build)
303 (mips16_macro_build, macro_build_lui, mips16_macro, mips_ip)
304 (mips16_ip): Use the new macros instead of explicit masks and shifts.
308 * config/tc-mips.c (mips_cl_insn): Replace the valid_p, delay_slot_p
309 and extended_p fields with a single fixed_p field.
310 (append_insn, mips_no_prev_insn): Adjust accordingly.
314 * config/tc-mips.c (mips_cl_insn): Replace reloc_type array with
315 a single mips16_absolute_jump_p bit.
316 (append_insn): Adjust accordingly.
320 * config/tc-mips.h (mips_cl_insn): Move definition to...
321 * config/tc-mips.c (mips_cl_insn): ...here. Add new fields:
322 frag, where, fixp, reloc_type, valid_p, noreorder_p, delay_slot_p
324 (history): New variable.
325 (prev_insn, prev_prev_insn, prev_insn_valid, prev_insn_frag)
326 (prev_insn_where, prev_insn_reloc_type, prev_insn_fixp)
327 (prev_insn_is_delay_slot, prev_insn_unreordered, prev_insn_extended)
328 (prev_prev_insn_unreordered): Delete.
329 (reg_needs_delay, append_insn, mips_no_prev_insn, mips_emit_delays)
330 (macro_start): Replace uses of prev_insn* with the equivalent history[]
335 * doc/Makefile.am: Update as.info dependencies.
336 * aclocal.m4, Makefile.in, doc/Makefile.in: Regenerated.
340 * doc/as.texinfo: Add sentence to indicate redefining a macro is an
341 error, and point to .purgem documentation if someone really needs
343 * NEWS: Mention macro redefinition is now an error.
347 * config/tc-ia64.c (emit_one_bundle): Track last slot user insn was
348 emitted to. Add more precise diagnostics for non-fitting insns based
349 on that. Eliminate now superfluous special casing of MLX. Clear out
350 slot information when dropping an insn.
354 * config/tc-ia64.c (parse_section_name): Rename to...
355 (cross_section): In addition to separating the name from the rest of
356 the arguments, also carry out the operation.
357 (dot_xdata): Use cross_section.
358 (dot_float_cons): Likewise.
359 (dot_xstringer): Likewise.
360 (dot_xdata_ua): Likewise.
361 (dot_float_cons_ua): Likewise. Pass float_cons, not stmt_float_cons.
365 * po/gas.pot: Regenerate.
369 * config/tc-mips.c (macro_build_lui): Use '__gnu_local_gp'
370 instead of '_gp' for -mno-shared optimization.
372 (s_abicalls): Document it in the comment.
373 (md_show_usage): Document the -mno-shared option.
377 * config/tc-mips.c (mips_set_options): Add sym32 field.
378 (mips_opts): Initialize it.
379 (HAVE_32BIT_ADDRESSES): Set to true if pointers are 32 bits wide.
380 (HAVE_64BIT_ADDRESSES): Redefine as !HAVE_32BIT_ADDRESSES.
381 (HAVE_32BIT_SYMBOLS, HAVE_64BIT_SYMBOLS): New macros.
382 (load_address): Use HAVE_64BIT_SYMBOLS instead of HAVE_64BIT_ADDRESSES
383 when deciding whether to use a symbolic %highest/%higher expansion.
384 (macro): Likewise. Remove o64/n32 linux hack. Always use
385 ADDRESS_ADD*_INSN for address addition in the expansion of "dla"
386 and "la". Handle constants separately from symbolic expressions in
387 the "ld_st:" case, using 64-bit arithmetic if HAVE_64BIT_ADDRESSES
388 and using load_register to load the high part of the address.
389 (OPTION_MSYM32, OPTION_NO_MSYM32): New macros.
390 (OPTION_ELF_BASE): Bump by 2.
391 (md_longopts): Add entries for -msym32 and -mno-sym32.
392 (md_parse_option): Handle them.
393 (usage): Document them.
394 (s_mipsset): Handle ".set sym32" and ".set nosym32".
395 (s_cpload, s_cpsetup): Use HAVE_64BIT_SYMBOLS instead of
396 HAVE_64BIT_ADDRESSES to detect 64-bit values of "_gp".
397 * doc/c-mips.texi: Document ".set sym32", ".set nosym32",
398 -msym32 and -mno-sym32.
402 * config/tc-mips.c (load_address): Implement GP optimization
403 for 64bit address space non-PIC. Fix formatting.
404 (macro): Likewise. Simplify code.
405 (md_parse_option): Don't bail out if -G 0 is set for PIC code.
406 (mips_after_parse_args): Simplify code.
410 * expr.c (operand): Remove redundant code enclosed by #ifdef
411 RELAX_PAREN_GROUPING....#endif.
413 * config/tc-mn10200.c (tc_gen_reloc): Handle the case where the
414 reloc is the difference of two symbols defined in the same
417 * config/tc-iq2000.c (line_comment_chars): Include the # character
418 as otherwise this breaks #APP/#NO_APP processing.
422 * config/tc-arc.c(md_assemble): Remove dead code for handling
423 immediate indexing of ld and st .
428 * config/tc-mips.c (percent_op): Add %tlsgd, %tlsldm, %dtprel_hi,
429 %dtprel_lo, %tprel_hi, %tprel_lo, and %gottprel.
430 (parse_relocation): Check for a word break after a relocation
432 (md_apply_fix3): Handle TLS relocations, and mark thread-local
437 * config/tc-ppc.c (ppc_fix_adjustable <ELF>): Remove bogus checks.
441 * as.c (main): Use unlink_if_ordinary instead of unlink.
442 * messages.c (as_fatal): Likewise.
446 * config/tc-i386.c (build_modrm_byte): Add lock prefix for cr8...15
448 (parse_register): Allow cr8...15 in all modes.
452 * config/tc-i386.c (intel_e11): If not followed by T_PTR, treat T_BYTE
453 etc. like normal symbol references (T_ID).
457 * symbols.c (fb_label_name): Fix silly thinko in last change.
461 * expr.c (integer_constant): Remove TARGET_WORD_SIZE hack.
462 * config/tc-m68k.h (TARGET_WORD_SIZE): Delete.
464 * symbols.c (fb_label_name): Allow an augend of 2 for mmix.
469 * config/tc-arc.c (md_assemble): Initialize suffix for extension
474 * config/obj-coff.c (fixup_segment): Delete sy_mri_common assertion.
476 * as.h (assert): Warning fix.
477 * expr.c (expr): Correct assertion.
478 * read.c (s_comm_internal): Remove assertion.
479 * write.c (relax_segment): Enable vma assertion only for BFD_ASSEMBLER.
480 (fixup_segment): Remove assertion.
481 * config/tc-dlx.c (machine_ip): Remove untrue assertions.
482 (md_apply_fix3): Likewise.
483 * config/tc-i370.c (md_begin): Correct assertion.
484 (i370_macro): Warning fix for assertion.
488 * configure.in (AC_C_BIGENDIAN): Invoke.
489 * configure: Regenerate.
490 * write.c (write_object_file <!BFD_ASSEMBLER>): Don't use sizeof
491 host variable to set string header size.
492 * config/obj-aout.c (obj_header_append): Don't use host structs.
493 (obj_symbol_to_chars): Likewise.
494 (obj_emit_strings): Likewise. Use the passed in output pointer.
495 * config/obj-aout.h (H_GET_FILE_SIZE): Include H_GET_LINENO_SIZE.
496 * config/obj-bout.c (obj_emit_relocations): Use md_reloc_size,
497 not sizeof host struct.
498 (obj_header_append, obj_symbol_to_chars): Don't use host structs.
499 (obj_emit_strings): Likewise.
500 * config/obj-bout.h (EXEC_BYTES_SIZE): Define.
501 (N_TXTOFF, H_GET_FILE_SIZE, H_GET_HEADER_SIZE): Use instead of
503 (H_SET_SYMBOL_TABLE_SIZE): Hard code sym size rather than using
505 (host_number_to_chars): Define.
506 * config/obj-hp300.c (hp300_header_append): Don't use sizeof
507 host internal struct to set header sizes.
508 * config/tc-i960.c (md_number_to_field): Warning fix.
509 (md_ri_to_chars): Use host byte order.
510 (get_cdisp, md_apply_fix3): Warning fix.
511 * config/tc-m68k.c (md_assemble): Don't use sizeof host short.
515 * doc/c-arc.texi: Update documentation about ARC's extension
520 * config/tc-tic4x.c (tic4x_gen_to_words): Changed mail
525 * cgen.c: Warning fixes.
526 * config/tc-arc.c: Likewise.
527 * config/tc-arm.c: Likewise.
528 * config/tc-avr.c: Likewise.
529 * config/tc-d10v.c: Likewise.
530 * config/tc-d30v.c: Likewise.
531 * config/tc-frv.c: Likewise.
532 * config/tc-frv.h: Likewise.
533 * config/tc-h8300.c: Likewise.
534 * config/tc-h8500.c: Likewise.
535 * config/tc-i370.c: Likewise.
536 * config/tc-i960.c: Likewise.
537 * config/tc-ia64.c: Likewise.
538 * config/tc-ip2k.c: Likewise.
539 * config/tc-m68hc11.c: Likewise.
540 * config/tc-maxq.c: Likewise.
541 * config/tc-mcore.c: Likewise.
542 * config/tc-mips.c: Likewise.
543 * config/tc-msp430.c: Likewise.
544 * config/tc-pj.c: Likewise.
545 * config/tc-ppc.c: Likewise.
546 * config/tc-ppc.h: Likewise.
547 * config/tc-s390.c: Likewise.
548 * config/tc-sh.c: Likewise.
549 * config/tc-sh64.c: Likewise.
550 * config/tc-tic4x.c: Likewise.
551 * config/tc-tic80.c: Likewise.
552 * config/tc-v850.c: Likewise.
553 * config/tc-vax.c: Likewise.
554 * config/tc-w65.c: Likewise.
555 * config/tc-xstormy16.c: Likewise.
556 * config/tc-z8k.c: Likewise.
560 * read.c (read_a_source_file): Reinstate TC_EQUAL_IN_INSN test.
561 * doc/internals.texi (TC_EQUAL_IN_INSN): Reinstate.
565 * config/tc-mips.c (struct proc): Change isym to
566 func_sym. New member func_end_sym.
567 (s_mips_ent): Update.
568 (s_mips_end): Ditto. Add code to compute function size.
572 * read.c: Warning fixes.
573 * config/obj-elf.c: Likewise.
577 * config/tc-mips.c (append_insn): Call dwarf2_emit_insn() before
582 * config/obj-coff.c (obj_coff_section): Replace SEC_SHARED with
585 * config/tc-tic54x.c (tic54x_bss): Replace SEC_BLOCK with
587 (demand_empty_rest_of_line): Likewise.
588 (tic54x_sblock): Likewise.
589 (tic54x_clink): Replace with SEC_CLINK with SEC_TIC54X_CLINK.
593 * read.c (address_bytes): New function.
594 (TC_ADDRESS_BYTES): Default for BSD_ASSEMBLER to address_bytes.
595 (potable): Add "dc.a".
596 (cons_worker): Handle "dc.a".
597 * doc/internals.texi (TC_ADDRESS_BYTES): Document.
601 * input-file.c (input_file_open): Rearrange to avoid warning.
605 * config/tc-hppa.h (TC_EQUAL_IN_INSN): Delete.
606 * read.c (read_a_source_file): Remove TC_EQUAL_IN_INSN test.
607 * doc/internals.texi (TC_EQUAL_IN_INSN): Delete.
611 * config/tc-mips.c (macro_build_ldst_constoffset): Fail on $at
612 uses after .set noat.
613 (load_address): Likewise.
614 (macro): Likewise. Don't try to avoid $at use by sacrificing
615 the target register before it is stored, it won't work.
619 * config/tc-ia64.c (emit_one_bundle): Stop filling a bundle if we
620 see an instruction that specifies a template.
624 * config/tc-openrisc.c (openrisc_relax_frag): Delete unused function.
625 * config/tc-sparc.c (sparc_ip): Make op_exp static to silence warnings.
626 * config/tc-tic80.c (build_insn): Init insn[1] to silence warning.
630 * NEWS: Mention "-mhint.b=[ok|warning|error]".
632 * config/tc-ia64.c (md): Add hint_b.
633 (emit_one_bundle): Handle md.hint_b for "hint".
634 (md_parse_option): Accepted "-mhint.b=[ok|warning|error]".
635 (md_show_usage): Add "-mhint.b=[ok|warning|error]".
636 (ia64_init): Set md.hint_b to error.
637 (md_assemble): Handle md.hint_b for "hint.b".
639 * doc/as.texinfo: Add "-mhint.b=[ok|warning|error]".
640 * doc/c-ia64.texi: Likewise.
644 * tc.h (struct relax_type, relax_typeS): Move from here..
645 * as.h: ..to here. Make rlx_forward and rlx_backward an offsetT.
646 * ecoff.c (ecoff_new_file): Add appfile param.
647 * ecoff.h (ecoff_new_file): Likewise.
648 * itbl-lex.h: New file.
649 * itbl-lex.l: Include itbl-lex.h.
650 * itbl-parse.y: Likewise.
651 (insntbl_line, yyparse, yylex): Move to itbl-lex.h.
652 * read.c (s_app_file_string): Mark appfile possibly unused.
653 * subsegs.c (seg_not_empty_p): Make sec possibly unused.
654 * subsegs.h (struct seg_info_trash): Delete.
655 (seg_info): Use segment_info_type instead.
656 * config/obj-coff.c (struct filename_list): Make filename const char *.
657 * config/obj-ecoff.h (obj_app_file): Pass app to ecoff_new_file.
658 * config/obj-elf.c (elf_file_symbol): Similarly.
659 * config/tc-a29k.c (md_apply_fix3): Make val a valueT. Don't use
661 * config/tc-arc.c (md_operand): Warning fix.
662 * config/tc-arm.c (arm_parse_reloc): Only define when OBJ_ELF.
663 (md_begin): Rearrange #if defined OBJ_COFF || defined OBJ_ELF.
664 * config/tc-cris.h (TC_IMPLICIT_LCOMM_ALIGNMENT): Use do while.
665 * config/tc-frv.c (frv_force_relocation): Warning fix.
666 * config/tc-m68k.c (md_parse_option): Delete unused var.
667 * config/tc-mcore.c (mylog2): Rename from log2 throughout.
668 * config/tc-sparc.c: Likewise.
669 (s_common): Warning fix.
670 * config/tc-mips.c (append_insn): Use unsigned long long expressions.
671 * config/tc-mmix.c (PUSHJSTUB_MAX, PUSHJSTUB_MIN): Define from
673 * config/tc-s390.c (s390_insn): Delete test of unsigned >= 0.
674 * config/tc-sh.c (sh_cfi_frame_initial_instructions,
675 sh_regname_to_dw2regnum): Only define for OBJ_ELF.
676 * config/tc-tic4x.c (tic4x_insert_reg): Use ISLOWER.
677 (tic4x_do_align): Use TIC_NOP_OPCODE.
678 * config/tc-tic4x.h (TIC_NOP_OPCODE): Rename from NOP_OPCODE.
679 * config/tc-vax.c: Include netinet/in.h.
680 (tc_headers_hook): Formatting.
681 * config/tc-xstormy16.c (md_pcrel_from_section): Correct parens.
685 * config/tc-ia64.c (ia64_parse_name): Don't advance 'name' when
686 parsing inN, locN, outN. Set 'idx' to offset register number starts
687 at. Don't handle numbers with leading zeroes or beyond 95. Remove
692 * config/tc-mips.c (load_address): Fix formatting.
693 (macro): Don't use AT if .set noat is in effect. Fix formatting.
694 Catch macros which are unexpandable without AT. Remove duplicate
696 (macro2): Remove duplicate zeroing of used_at.
700 * dw2gencfi.c (output_cie, output_fde): Use DW_CFA_nop rather
706 * config/tc-mips.c (reloc_needs_lo_p): Handle
707 BFD_RELOC_MIPS16_HI16_S.
708 (fixup_has_matching_lo_p): Handle BFD_RELOC_MIPS16_LO16.
709 (append_insn): Add BFD_RELOC_MIPS16_GPREL, BFD_RELOC_MIPS16_HI16_S
710 and BFD_RELOC_MIPS16_LO16 to relocs to suppress overflow
712 (mips16_ip): Resolve BFD_RELOC_MIPS16_HI16_S,
713 BFD_RELOC_MIPS16_HI16 and BFD_RELOC_MIPS16_LO16 for constants.
714 Call my_getSmallExpression() to parse percent operators.
715 (percent_op_match, mips_percent_op): Separate definitions.
716 (mips16_percent_op): Define percent operators for the MIPS16 mode.
717 (parse_relocation): Handle the MIPS16 mode using
719 (md_apply_fix3): Handle BFD_RELOC_MIPS16_HI16,
720 BFD_RELOC_MIPS16_HI16_S and BFD_RELOC_MIPS16_LO16.
724 * config/tc-ia64.c (md_apply_fix3): Call ia64_gen_real_reloc_type
725 instead of explicitly dealing with the translation; exclude
726 relocations that are already pcrel, however.
730 * config/tc-ia64.c: Include limits.h (if available).
731 (gr_values[0]): Set path to INT_MAX.
732 (dot_reg_val): Don't allow changing value of r0. Limit range of
733 general registers at r127.
734 (specify_resource): Default resource index is -1. Don't set resource
735 index (in case IA64_RS_RSE) without setting the specific flag.
736 (note_register_values): Check operand is O_constant before tracking
737 input value of moves. Add tracking for dep.z with constant inputs.
738 (print_dependency): Resource index of specific resource may be zero.
739 (check_dependencies): Likewise.
743 * config/tc-ia64.c (parse_operands): New local variables reg1, reg2,
744 reg_class. Check operands and emit diagnostics for illegal use of
749 * config/tc-ia64.c (ia64_gen_real_reloc_type): Define and initialize
750 new variables type, suffix, and width. Handle
751 BFD_RELOC_IA64_DIR(32|64)[LM]SB in FUNC_LT_FPTR_RELATIVE case.
752 Handle BFD_RELOC_IA64_DIR64[LM]SB in FUNC_TP_RELATIVE case. Add
753 FUNC_DTP_MODULE case. Handle BFD_RELOC_IA64_DIR32[LM]SB in
754 FUNC_DTP_RELATIVE case. Return incoming relocation type if
755 BFD_RELOC_IA64_IPLT[LM]SB in FUNC_IPLT_RELOC case. Generate warning
756 if unable to translate relocation type, using the new variables.
760 * config/tc-ia64.h (ia64_symbol_chars): Declare.
761 (ty_symbol_chars): Define.
762 * config/tc-ia64.c (ia64_symbol_chars): Define.
766 * config/tc-ia64.c (ia64_parse_name): Only update next character if
767 input_line_pointer was advanced.
771 * config/tc-sh.c (md_apply_fix3): Add parentheses around &
776 * config/tc-ia64.c (md_parse_option): Handle -xnone and -xdebugn.
777 (md_show_usage): Add -xnone, -xdebugn, and -xdebugx. Relocate default
779 (ia64_init): Set md.detect_dv.
780 (ia64_start_line): New static variable warned. Warn only once when
781 encountering explicit stops in automatic mode.
782 * doc/c-ia64.texi: Describe -xnone, -xdebugn, and -xdebugx.
783 * NEWS: Mention new default mode.
787 * config/tc-ia64.c (dot_rot): Add comment that name strings should
788 be freed when wiping out previous state. Canonicalize names before
789 use. Free name string when detecting redefinition.
790 (dot_pred_rel): Call generic expression parser to process arguments.
791 Handle O_register case for individual predicates and O_subtract for
793 (ia64_parse_name): Canonicalize name before looking it up in dynamic
795 (ia64_canonicalize_symbol_name): Strip off all trailing # characters.
796 Warn if multiple found, issue error if resulting symbol name has zero
798 (dot_alias): Canonicalize name before use.
802 * config/tc-ia64.c (unwind_diagnostic): Return -1 for warning
804 (in_procedure): Return -1 for warning.
805 (in_prologue): Likewise.
810 * config/tc-ia64.c (dot_xdata): Undo the last change.
811 (dot_float_cons): Likewise.
812 (dot_xstringer): Likewise.
813 (dot_xdata_ua): Likewise.
814 (dot_float_cons_ua): Likewise.
818 * NEWS: Mention "-munwind-check=[warning|error]".
820 * config/tc-ia64.c (md): Add unwind_check.
821 (unwind_diagnostic): New.
822 (in_procedure): Call unwind_diagnostic when a directive isn't
824 (in_prologue): Call unwind_diagnostic when a directive isn't in
826 (in_body): Call unwind_diagnostic when a directive isn't in
828 (dot_endp): Set md.unwind_check to error before calling
829 in_procedure and restore it after. When the name is missing or
830 couldn't be found, use the one from the last .proc if
831 md.unwind_check isn't error. Warn if md.unwind_check is
833 (md_parse_option): Handle "-munwind-check=[warning|error]".
834 (md_show_usage): Add "-munwind-check=[warning|error]".
835 (ia64_init): Set md.unwind_check to warning.
837 * doc/as.texinfo: Add "-munwind-check=[none|warning|error]".
838 * doc/c-ia64.texi: Likewise.
842 * config/tc-ia64.h (LEX_AT): Include LEX_BEGIN_NAME.
844 (ia64_parse_name): New third parameter.
845 (md_parse_name): Pass third argument.
846 * config/tc-ia64.c (pseudo_func): Placeholders use NULL as name.
847 (md_operand): Handling of '@'-prefixed symbols moved from here...
848 (ia64_parse_name): ...to here.
852 * config/tc-ia64.c (md): Remove last_groups and group_idx.
853 (errata_nop_necessary_p): Remove declaraction and definition.
854 (emit_one_bundle): Don't call errata_nop_necessary_p. Don't
855 update md.group_idx. Don't reset md.last_groups.
859 * config/tc-ia64.c (parse_section_name): Handle non-quoted first
861 (dot_xdata): Free section name after use.
862 (dot_float_cons): Likewise.
863 (dot_xstringer): Likewise.
864 (dot_xdata_ua): Likewise.
865 (dot_float_cons_ua): Likewise.
866 (md_pseudo_table): Add xdata16 and xdata16.ua.
870 * doc/all.texi: Add IA64.
871 * doc/as.texinfo: Likewise.
873 * doc/c-ia64.texi: Fix typos.
877 * config/tc-arm.c (do_t_ldmstm): Change BFD_RELOC_NONE to
879 (do_t_push_pop): Likewise.
880 (md_assemble): Likewise.
881 (md_apply_fix3): Handle BFD_RELOC_NONE correctly, make
882 BFD_RELOC_UNUSED same as previous meaning of BFD_RELOC_NONE.
883 (create_unwind_entry): Output dependency on the required personality
885 testsuite/gas/arm/unwind.d: Alter expected output to include
886 dependency on __aeabi_unwind_cpp_pr[01].
890 * as.h (seg_not_empty_p): Return int, not bfd_boolean.
891 * subsegs.c (seg_not_empty_p): Likewise.
895 * config/tc-maxq.c (md_estimate_size_before_relax): Correct the
896 relative jump calculation.
897 <md_convert_frag) : Likewise.
898 <output_disp): Likewise.
902 * write.c (write_object_file): Recognize warning-symbol construct
903 and skip object- and target- handling for the second symbol.
907 * config/tc-ia64.c (dot_pred_rel): Update comment. Handle @-prefixed
908 designators along with quoted ones. Free copy of quoted designator
913 * config/atof-ieee.c, config/obj-coff.c, config/obj-elf.c,
914 config/obj-ieee.c, config/obj-som.c, config/obj-vms.c,
915 config/tc-a29k.c, config/tc-alpha.c, config/tc-arc.c,
916 config/tc-arm.c, config/tc-d30v.c, config/tc-dlx.c,
917 config/tc-fr30.c, config/tc-h8300.c, config/tc-h8500.c,
918 config/tc-i370.c, config/tc-i386.c, config/tc-i960.c,
919 config/tc-ia64.c, config/tc-m32r.c, config/tc-m32r.h,
920 config/tc-m68hc11.c, config/tc-m68hc11.h, config/tc-mips.c,
921 config/tc-mn10200.c, config/tc-msp430.c, config/tc-ns32k.c,
922 config/tc-openrisc.c, config/tc-or32.c, config/tc-pdp11.c,
923 config/tc-pj.c, config/tc-sparc.h, config/tc-tic54x.c,
924 config/tc-tic80.c, config/tc-v850.c, config/tc-w65.c,
925 config/tc-xtensa.c, config/tc-z8k.c, config/xtensa-relax.c: Remove
926 #if 0'd code throughout.
930 * as.c (parse_args): Bump copyright date reported by --version to
935 * configure.tgt: Remove obsolete targets m68k-lynxos, sparc-lynxos
940 * macro.c (buffer_and_nest): Allow 'from' being NULL; handle anything
941 that can end with .endr in that case. Make requiring/permitting
942 pseudo-ops without leading dot closer to the logic in read.c serving
944 (expand_irp): Don't pass a mnemonic to buffer_and_nest as it will be
949 * macro.c (do_formals): Adjust to no longer accept empty parameter
951 (define_macro): Adjust to no longer accept empty macro name, garbage
952 following the parameters, or macros that were previously defined.
953 * read.c (s_bad_end): Declare.
954 (potable): Add endm. Handler for endr and endm is s_bad_end.
955 (s_bad_end): Rename from s_bad_endr. Adjust to handle both .endm
957 * read.h (s_bad_endr): Remove.
961 * config/tc-ia64.c (parse_operands): Parse all specified operands,
962 immediately discarding (but counting) those exceeding the maximum
963 possible amount. Track whether output and input operand counts ever
964 matched, and use this to better indicate which of the operands/
965 operand types was wrong; specifically don't default to pointing to
970 * config/tc-ia64.c (unwind): Remove proc_end (now an automatic
971 variable in dot_endp). Add body and insn. Make prologue,
972 prologue_mask, body, and insn bitfields.
973 (fixup_unw_records): Remove spurious new-lines from end of diagnostic
975 (in_procedure, in_prologue, in_body): New.
976 (dot_fframe, dot_vframe, dot_vframesp, dot_vframepsp, dot_save,
977 dot_restore, dot_restorereg, dot_restorereg_p, dot_handlerdata,
978 dot_unwentry, dot_altrp, dot_savemem, dot_saveg, dot_savef, dot_saveb,
979 dot_savegf, dot_spill, dot_spillreg, dot_spillmem, dot_spillreg_p,
980 dot_spillmem_p, dot_label_state, dot_copy_state, dot_unwabi,
981 dot_personality): Use the appropriate one of the above.
982 (dot_proc): Clear unwind.proc_start; set to current location only if
983 none of the entry points were valid. Check for non-zero-length entry
984 point names. Check that entry points aren't defined, yet. Clear
985 unwind.prologue, unwind.body, and unwind.insn.
986 (dot_body): Call in_procedure. Check that first directive in procedure
987 had no insns emitted before. Set unwind.body.
988 (dot_prologue): Call in_procedure. Check that not already in prologue.
989 Check that first directive in procedure had no insns emitted before.
991 (dot_endp): Call in_procedure. Declare proc_end. Check for non-zero-
992 length entry point names. Check that entry points became defined.
993 (md_assemble): Set unwind.insn once unwind.proc_start is defined.
997 * config/tc-ia64.c (emit_one_bundle): Snapshot manual bundling state
998 before actually using it. Don't generate an error in manual bundling
999 mode when looking at an insn requiring slot 2 but not yet at slot 2.
1000 Don't generate an error in manual bundling mode when looking at an
1001 insn required to be last in its group but the required slot hasn't
1002 been reached, yet. Allow conversion from MII to MI;I for bundle
1003 consisting of only 2 insns with the stop between them. Suppress
1004 various meaningless errors resulting from detecting earlier ones.
1008 * config/tc-ia64.c (parse_operands): Also handle alloc without first
1013 * config/tc-z8k.c (md_assemble): Improve error detection.
1017 * config/tc-ia64.c (ia64_estimate_size_before_relax): Allocate space
1018 for personality routine pointer only if there is one.
1019 (ia64_convert_frag): Likewise.
1020 (generate_unwind_image): Likewise.
1024 * config/tc-z8k.c (INSERT): Remove, not used anywhere.
1025 (md_apply_fix3): Make relative branches out of range an error
1026 instead of a warning. Display correct line number for out of
1027 range branches/calls/memory accesses.
1031 * dwarf2dbg.c (dwarf2_finish): Correct logic for determining when
1032 to emit .debug_line and other debug sections.
1033 * as.h (seg_not_empty_p): Declare.
1034 * subsegs.c (seg_not_empty_p): New predicate.
1038 * configure: Regenerate to track ../gettext.m4 change.
1042 * config/tc-ia64.c (emit_one_bundle): Change "?imbf??" to "?ibmfxx".
1046 * config/tc-ia64.c (emit_one_bundle): Add late resolution of move
1047 to/from application registers dynamic insns.
1048 (md_assemble): Defer resolution of move to/from application registers
1049 dynamic insns when they can be issued on either the I- or M-units.
1053 * config/tc-frv.c (md_apply_fix3): Mark TLS symbols as such.
1055 * config/tc-frv.c (frv_pic_ptr): Add tlsmoff support.
1057 * cgen.c (gas_cgen_parse_operand): Handle
1058 CGEN_PARSE_OPERAND_SYMBOLIC.
1059 * config/tc-frv.c (md_cgen_lookup_reloc): Handle TLS relocations.
1060 (frv_force_relocation): Likewise. Fix handling of PIC
1062 (md_apply_fix3): Likewise.
1066 * as.h: Remove #if 0'd code.
1067 * atof-generic.c (atof_generic): Likewise.
1068 * ecoff.c (ecoff_directive_frame): Likewise.
1069 * frags.h (FRAG_APPEND_1_CHAR): Likewise.
1070 * itbl-ops.c (itbl_add_reg): Likewise.
1071 * listing.c (calc_hex): Likewise.
1072 * read.c (MASK_CHAR): Likewise.
1073 * subsegs.c (subsegs_print_statistics): Likewise.
1074 * symbols.c (indent): Likewise.
1075 * write.c (write_relocs): Likewise.
1076 (write_object_file): Likewise.
1077 (relax_frag): Likewise.
1081 * as.c (std_longopts): Add an entry for "--a" in order to prevent
1082 getopt_long_only() from considering -a as an abbreviation for
1084 (parse_args): Fix the parsing of -a=<file>.
1089 * read.c (s_incbin): Adjust default count for skip. Check validity
1090 of count and skip rigorously.
1094 * config/tc-mips.c (dummy_opcode): Add init for new struct member.
1098 * read.c (convert_to_bignum): New function, split out from...
1099 (emit_expr): ...here. Handle the case where X_add_number is
1100 positive and the input value is negative.
1101 (output_big_sleb128): Fix setting of continuation bit. Check whether
1102 the final byte needs to be sign-extended. Fix size-shrinking loop.
1103 (emit_leb128_expr): When generating a signed leb128, see whether the
1104 sign of an O_constant's X_add_number matches the sign of the input
1105 value. Use a bignum if not.
1109 * tc-sh.c (md_begin,md_parse_option): Change arch_sh1_up to
1110 arch_sh_up in order to match the external name and make the
1111 testsuite's job easier.
1116 * config/tc-i386.c (i386_scale): Disallow 0 scale.
1120 * config/tc-iq2000.c (s_iq2000_set): Fix thinko parsing
1121 ignored_arguments array.
1125 * write.c (write_object_file): Disallow a symbol equated to
1130 * tc-maxq.c: Replace constants 10 and 20 with bfd_mach_maxq10 and
1132 (md_pseudo_table): Add new pseudo ops for maxq10 and maxq20.
1133 (maxq_target): New function: Set the machine type.
1137 * config/tc-arm.c (FPU_DEFAULT): Define for TE_VXWORKS.
1138 (md_begin): Handle TE_VXWORKS for FP defaults.
1139 (md_apply_fix3): Correct rela offsets.
1140 (elf32_arm_target_format): Add VxWorks targets.
1144 * configure.tgt: Set em=vxworks for *-*-vxworks.
1145 * config/te-vxworks.h: New File.
1149 * config/tc-arm.c (arm_cpus): Correct arch field for arm1026ej-s.
1153 * config/tc-msp430.c (md_apply_fix3): Fix offset calculation for
1158 * config/tc-ia64.c (md): Add member "loc_directive_seen".
1159 (dot_loc): New function.
1160 (md_pseudo_table): Add entry to map .loc to dot_loc().
1161 (emit_one_bundle): Only call dwarf2_gen_line_info() if we have
1162 seen a .loc directive or we're generating DWARF2 debug info for
1168 * read.c (s_comm_internal): Don't zero end of name until size
1169 expression has been parsed.
1173 * config/tc-avr.c (mcu_types): Move attiny{13,2313} from avr4 to avr2.
1177 * config/tc-crx.c: Support 'bcop' relaxation (dealt as in 'cmp&branch'
1182 * configure.tgt: New.
1183 * configure.in: Move setting of cpu_type, fmt, etc., to
1185 * Makefile.am (CONFIG_STATUS_DEPENDENCIES): Add
1186 $(srcdir)/configure.tgt.
1187 * configure, Makefile.in: Rebuild.
1191 * config/tc-avr.c: Add support for the new R_AVR_LDI, R_AVR_6 and
1192 R_AVR_6_ADIW relocs for the LDI, ADIW/SBIW and LDD/STD
1194 (avr_offset_expression): New function to parse offsets for LDI
1196 (avr_operand): Use it.
1197 (md_apply_fix3): Generate the relocs.
1201 * config/tc-sh64.c (shmedia_md_apply_fix3): Add missing
1202 BFD_RELOC_SH_IMMS10BY8 relocation.
1204 * config/tc-sh64.c (shmedia_build_Mytes): Emit an error message rather
1205 than just ignoring bad code.
1209 * config/tc-v850.c (handle_lo16): New function.
1210 (v850_reloc_prefix): Use it to check lo().
1211 (md_assemble, md_apply_fix3): Handle BFD_RELOC_V850_LO16_SPLIT_OFFSET.
1215 * configure.in: Add s390x-ibm-tpf support.
1216 * configure: Regenerate.
1220 * config/obj-elf.c (obj_elf_change_section): Only set type and
1221 attributes on new sections. Emit warning when type of re-declared
1222 section doesn't match.
1226 * dw2gencfi.c (dot.cfi.startproc): Clear cur_cfa_offset so
1227 '.cfi_startproc simple' doesn't inherit the old value.
1231 * dw2gencfi.c (output_cfi_insn): Adjust DW_CFA_def_cfa_sf generation
1232 to emit a signed and factored offset. Adjust DW_CFA_def_cfa_offset_sf
1233 generation to emit a factored offset.
1237 * config/tc-mips.c (macro) [M_LA_AB]: Give an error for a offset
1238 which is too large in the case of NO_PIC without 64-bit
1241 * config/tc-mips.c (mips_in_shared): New static variable.
1242 (macro_build_lui): Permit "_gp" if !mips_in_shared.
1243 (md_longopts): Add -mshared and -mno-shared.
1244 (md_parse_option): Handle OPTION_MSHARED and OPTION_MNO_SHARED.
1245 (s_cpload): Implement !mips_in_shared case.
1246 (s_cpsetup): Likewise.
1247 * doc/c-mips.texi (MIPS Opts): Document -mno-shared.
1248 * NEWS: Mention -mno-shared.
1252 * config/tc-arm.c (s_arm_unwind_fnend): Use R_ARM_PREL31 relocation
1257 * config/tc-mips.c (append_insn): If we emit a nop during a relax
1258 sequence, increase the size of the sequence.
1260 * config/tc-mips.c (mips_cpu_info_table): Change "9000" entry to
1265 * read.c (s_align): Use an align_limit temporary to allay a GCC
1266 signed/unsigned comparison warning.
1270 * Makefile.am (TARG_ENV_HFILES): Add te-armlinuxeabi.h.
1271 * configure.in: Use it for arm*-*-linux-gnueabi*.
1272 * config/tc-arm.c: Allow emulation file to set FPU_DEFAULT.
1273 * config/te-armlinuxeabi.h: New file.
1274 * Makefile.in: Regenerated.
1275 * aclocal.m4: Likewise.
1276 * configure: Likewise.
1277 * doc/Makefile.in: Regenerated.
1281 * config/tc-xtensa.c (xtensa_switch_section_emit_state): Use subseg_set.
1282 (xtensa_restore_emit_state): Likewise.
1286 * read.c (ALIGN_LIMIT): Define, increasing limit for BFD_ASSEMBLER.
1291 * Makefile.am: Change LOCALEDIR to $(datadir)/share.
1292 * Makefile.in: Regenerate.
1296 * config/tc-crx.c: Major code cleanup. Remove unused variables and
1297 functions, give functions a meaningful name, add comments.
1298 (check_range): New function - Replace operand size calculation
1299 with range checking.
1300 (assemble_insn): Update Algorithm, improve error issuing.
1302 (process_label_constant): Bug fix regarding COP_BRANCH_INS relocation
1307 * config/tc-m32r.c (md_pcrel_from_section): Fixed a pcrel relocte
1308 miss between different section in the same module.
1309 (tc_gen_reloc): Likewise.
1313 * gas/config/tc-avr.c (mcu_types): Add support for atmega165,
1314 atmega325, atmega3250, atmega645 and atmega6450.
1318 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
1319 permissible for the selected instruction suffix.
1320 (process_suffix): For DefaultSize instructions, suppressing the
1321 guessing of a 'q' suffix if the instruction doesn't support it is
1322 pointless, because only an 'l' suffix can be guessed in this place.
1326 * config/tc-iq2000.c: Remove support for IQ10 processor.
1327 Convert to ISO C90 formatting.
1328 * config/tc-iq2000.h: Likewise.
1332 * config/tc-mn10300.c (md_relax_table): More fixes to the offsets
1333 in this table. They should be correct now.
1337 * config/tc-i386.h (CpuMMX2): Declare. Artificial classifier to
1338 indicate the MMX extensions added by both SSE and 3DNow!A.
1339 (Cpu3dnowA): Declare.
1340 (CpuUnknownFlags): Update.
1341 * config/tc-i386.c (cpu_sub_arch_name): Declare.
1342 (cpu_arch): i586 and pentium do not imply MMX. i686 and pentiumpro do
1343 neither imply SSE nor MMX. k6 implies MMX. k6_2 additionally implies
1344 3DNow!. Athlon additionally implies 3DNow!A. Several new
1345 entries (those starting with a dot are for sub-arch specification).
1346 (set_cpu_arch): Handle sub-arch specifications.
1347 (parse_insn): Distinguish between instructions not supported because
1348 of insufficient CPU features and because of 64-bit mode.
1349 * doc/c-i386.texi: Describe enhanced .arch directive.
1353 * config/tc-xtensa.c (xg_add_opcode_fix): Set fx_no_overflow.
1357 * dwarf2dbg.c (dwarf2_finish): Don't write a .debug_line section
1358 without a corresponding .debug_info section.
1362 * read.c (potable): Add "error" and "warning".
1363 (s_errwarn): New function.
1364 * read.h (s_errwarn): Declare.
1365 * doc/as.texinfo (Error, Warning): Document .error and .warning.
1369 * config/tc-tic54x.c (tic54x_adjust_symtab): Adjust call to
1374 * config/tc-msp430.c (struct rcodes_s, MSP430_RLC, msp430_rcodes,
1375 struct hcodes_s, msp430_hcodes): From include/opcode/msp430.h.
1379 * config/obj-coff.c (c_dot_file_symbol): Add "app" param.
1380 (coff_adjust_symtab): Adjust call.
1381 (crawl_symbols): Likewise.
1382 * config/obj-coff.h (c_dot_file_symbol): Add "app" param.
1383 (obj_app_file): Adjust.
1387 * configure.in: Enable bfd_assember for the MAXQ port.
1388 * configure: Regenerate.
1393 * config/tc-xtensa.c (finish_vinsn): Clear pending instruction if
1394 there is a conflict.
1395 (check_t1_t2_reads_and_writes): Check for both reads and writes to
1396 interfaces that are related as determined by xtensa_interface_class_id.
1400 * config/tc-mn10300.c (md_relax_table): Fix off by one negative
1401 offsets for conditional branches.
1405 * config/tc-xtensa.c (MAX_IMMED6): Change value to 65.
1409 * config/tc-xtensa.c (update_next_frag_state): Always add a NOP if
1410 relaxing at the end of a loop. Don't mark frags as UNREACHABLE or
1412 (relax_frag_immed): Update call to update_next_frag_state.
1416 * obj.h (struct format_ops <app_file>): Add int param.
1417 * read.h (s_app_file_string): Likewise.
1418 * read.c (s_app_file_string): Likewise.
1419 (s_app_file): Adjust s_app_file_string call.
1420 * config/tc-mips.c (s_mips_file): Likewise.
1421 * config/obj-coff.h (obj_app_file): Add app param.
1422 * config/obj-ecoff.h (obj_app_file): Likewise.
1423 * config/obj-multi.h (obj_app_file): Likewise.
1424 * config/obj-elf.h (elf_file_symbol): Likewise.
1425 * config/obj-elf.c (elf_file_symbol): Only emit one file symbol
1426 if called for # preprocessor lines.
1431 * symbols.c (resolve_symbol_value): Convert weak symbols only
1433 (symbol_equated_reloc_p): Don't equate weaks when relocating
1434 only for Windows PECOFF.
1438 * config/tc-crx.c (print_insn): Check and set insn_addr.
1439 * config/tc-crx.h (md_frag_check): Define.
1444 * configure.in: Add support for new target maxq-coff.
1445 * configure: Regenerate.
1446 * NEWS: Mention new support.
1447 * config/tc-maxq.c: New file.
1448 * config/tc-maxq.h: New file.
1449 * config/obj-coff.h: Add support for maxq-coff.
1453 * symbols.c (any_external_name): Define.
1454 (resolve_symbol_value): Convert weak symbols.
1455 (S_SET_EXTERNAL): Support any_external_name.
1456 (S_SET_NAME): Qualify parameter const.
1457 (symbol_equated_reloc_p): Equate to weaks when relocating.
1458 * symbols.h (S_SET_NAME): Qualfiy parameter const.
1459 * tc.h (any_external_name): Declare.
1460 * config/obj-coff.c ("coff/pe.h"): Include for BFD
1462 (weak_is_altname): Declare and define.
1463 (weak_name2altname): Same.
1464 (weak_altname2name): Same.
1465 (weak_uniquify): Same.
1466 (weak_altprefix): Define.
1467 (obj_coff_weak): Change .weak syntax and handling.
1468 (coff_frob_symbol): Fix PE weak symbol alternates.
1469 * config/obj-coff.h (USE_UNIQUE): Define.
1470 * config/tc-i386.c (md_apply_fix3): Assume weak symbols
1471 are in another segment.
1472 (tc_gen_reloc): Remove broken addend hack.
1473 doc/as.texinfo: Update.
1477 * config/tc-xtensa.c (total_frag_text_expansion): New.
1478 (md_estimate_size_before_relax): Use it.
1479 (find_address_of_next_align_frag): Likewise.
1483 * config/tc-crx.c: Rename argument types.
1484 (processing_arg_number): Rename to 'cur_arg_num'.
1485 (get_number_of_bits): Rename to 'set_operand_size'.
1486 (get_operandtype): Rename to 'parse_operand', totally rewrite.
1487 (set_cons_rparams): Rename to 'set_operand', totally rewrite.
1488 (set_indexmode_parameters): Remove function, integrate its code into
1490 (set_operand_size): Get rid of 'Operand Number' function parameter -
1491 use global variable 'cur_arg_num' instead.
1492 Use a local 'argument' pointer to reference the current argument.
1493 (parse_operand): Likewise.
1494 (set_operand): Likewise.
1495 (process_label_constant): Likewise.
1499 * config/tc-xtensa.c: Remove XTENSA_SECTION_RENAME ifdefs.
1500 (add_section_rename): Delete. Inlined into...
1501 (build_section_rename): ...here. Use xstrdup instead of strdup.
1502 (xtensa_section_rename): Drop "const" from argument and return types.
1503 (md_show_usage): Indent to match show_usage().
1504 * config/tc-xtensa.h: Remove XTENSA_SECTION_RENAME ifdefs.
1505 (tc_canonicalize_section_name): Define.
1506 (md_elf_section_rename): Remove unused macro.
1507 * doc/as.texinfo (Overview): Document Xtensa --rename-section option.
1508 * doc/c-xtensa.texi (Xtensa Options): Likewise.
1509 (Frame Directive): Delete.
1513 * configure.in: Remove arm-*-oabi and thumb-*-oabi.
1514 * config/tc-arm.c (target_oabi): Delete.
1515 (md_apply_fix3, elf32_arm_target_format): Remove target_oabi checks.
1516 (arm_opts): Remove moabi.
1517 * doc/as.texinfo (Overview): Remove documentation of -moabi.
1518 * doc/c-arm.texi (ARM Options): Likewise.
1519 * configure: Regenerated.
1523 * configure.in (crisv32): Recognize. AC_DEFINE_UNQUOTED
1524 DEFAULT_CRIS_ARCH. Handle crisv32-*-linux-gnu* like
1525 cris-*-linux-gnu* and crisv32-*-* like cris-*-*.
1526 * configure: Regenerate.
1527 * config/tc-cris.c (enum cris_archs): New.
1528 (cris_mach, cris_arch_from_string, s_cris_arch, get_sup_reg)
1529 (cris_insn_ver_valid_for_arch): New functions.
1530 (DEFAULT_CRIS_ARCH): New macro, default to cris_any_v0_v10.
1531 (cris_arch): New variable.
1532 (md_pseudo_table): New pseudo .arch.
1533 (err_for_dangerous_mul_placement): Initialize according to
1535 (STATE_COND_BRANCH): Renamed from STATE_CONDITIONAL_BRANCH.
1537 (STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON)
1538 (STATE_ABS_BRANCH_V32, STATE_LAPC, BRANCH_BF_V32, BRANCH_BB_V32)
1539 (BRANCH_WF_V32, BRANCH_WB_V32): New.
1540 (BRANCH_BF, BRANCH_BB, BRANCH_WF, BRANCH_WB): Don't undef after
1541 use in md_cris_relax_table.
1542 (md_cris_relax_table): Add entries for STATE_COND_BRANCH_V32,
1543 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
1544 Update and improve head comment.
1545 (OPTION_PIC): Define in terms of previous option, OPTION_US.
1546 (OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): Similar.
1548 (md_longopts): New option --march=...
1549 (cris_any_v0_v10_long_jump_size, crisv32_long_jump_size): New
1551 (md_long_jump_size): Initialize in terms of DEFAULT_CRIS_ARCH.
1552 (HANDLE_RELAXABLE): New macro.
1553 (md_estimate_size_before_relax): Use HANDLE_RELAXABLE for common
1554 cases. Check for weak symbols and assume not relaxable. Handle
1555 STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON,
1556 STATE_ABS_BRANCH_V32, STATE_LAPC. Use new variable symbolP, not
1558 (md_convert_frag): Handle STATE_COND_BRANCH_V32,
1559 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
1560 (cris_create_short_jump): Adjust for CRISv32.
1561 (cris_relax_frag): Handle new states.
1562 (md_create_long_jump): Ditto. Emit error for common_v10_v32.
1563 (md_begin): Define symbols "..asm.arch.cris.v32",
1564 "..asm.arch.cris.v10", "..asm.arch.cris.common_v10_v32" and
1565 "..asm.arch.cris.any_v0_v10". Use cris_insn_ver_valid_for_arch
1566 when entering opcode table entry points.
1567 (md_assemble): Adjust branch handling for CRISv32. Handle LAPC
1568 relaxation. In fix_new_exp call for main insn, pass 1 for pcrel
1569 parameter for 8, 16 and 32-bit pc-relative insns and LAPC.
1570 (cris_process_instruction): Initialize out_insnp->insn_type to
1571 CRIS_INSN_NONE, not CRIS_INSN_NORMAL.
1572 <case ']', '[', 'A', 'd', 'Q', 'N', 'n', 'Y', 'U', 'u', 'T'>: New
1574 <case 'm'>: Check that modified_char == '.'.
1575 <invalid operands>: Consume the rest of the line.
1576 When operands don't match, skip over subsequent insns with
1577 non-matching version specifier but same mnemonic.
1578 <immediate constant, case SIZE_SPEC_REG>: Immediate operands for
1579 special registers in CRISv32 are always 32 bit long.
1580 <immediate constant, case SIZE_FIELD_SIGNED, SIZE_FIELD_UNSIGNED>:
1582 (get_gen_reg): Only recognize "PC" when followed by "+]" for v32
1583 and compatible. Recognize "ACR" for v32, unless followed by "+".
1584 (get_spec_reg): Consider cris_arch when looking up register.
1585 (get_autoinc_prefix_or_indir_op): Don't recognize assignment for
1587 (get_3op_or_dip_prefix_op): Check for ']' after seeing '[rN+'.
1588 (cris_get_expression): Restore input_line_pointer if failing "early".
1589 (get_flags): Consider cris_arch and recognize flags accordingly.
1590 (branch_disp): Adjust for CRISv32.
1591 (gen_cond_branch_32): Similar. Emit error for common_v10_v32.
1592 (cris_number_to_imm): Use as_bad_where, not as_bad. Remove
1593 related FIXME. Don't insist on BFD_RELOC_32_PCREL fixup to be
1594 resolved. Don't enter zeros in object file for
1596 <case BFD_RELOC_CRIS_LAPCQ_OFFSET, BFD_RELOC_CRIS_SIGNED_16>
1597 <case BFD_RELOC_CRIS_SIGNED_8>: New case.
1598 (md_parse_option): Break out "return 1".
1599 <OPTION_ARCH> New case.
1600 (tc_gen_reloc): <case BFD_RELOC_CRIS_LAPCQ_OFFSET>
1601 <case BFD_RELOC_CRIS_SIGNED_16, BFD_RELOC_CRIS_SIGNED_8>
1602 <case BFD_RELOC_CRIS_UNSIGNED_8, BFD_RELOC_CRIS_UNSIGNED_16>
1603 <case BFD_RELOC_32_PCREL>: New cases.
1604 Addends for non-zero fx_pcrel are too in fx_offset.
1605 (md_show_usage): Show --march=<arch>.
1606 (md_apply_fix3): Adjust val for BFD_RELOC_CRIS_LAPCQ_OFFSET.
1607 (md_pcrel_from): BFD_RELOC_CRIS_LAPCQ_OFFSET is PC-relative too.
1608 (s_syntax) <struct syntaxes>: Properly constify member operand.
1609 * config/tc-cris.h (TARGET_MACH): Define.
1610 (cris_mach): Declare.
1611 * doc/as.texinfo (Overview) <CRIS>: Add --march=...
1612 * doc/c-cris.texi (CRIS-Symbols): New node for built-in symbols.
1613 (CRIS-Opts): Document --march=...
1614 (CRIS-Pseudos): Document .arch.
1618 * config/tc-i386.c (set_intel_syntax): Allow % in symbol names when
1619 intel syntax and no register prefix, allow $ in symbol names when
1621 (set_16bit_gcc_code_flag): Replace literal 'l' by LONG_MNEM_SUFFIX.
1622 (intel_float_operand): Add fourth return value indicating math control
1623 operations. Make classification more precise.
1624 (md_assemble): Complain if memory operand of mov[sz]x has no size
1626 (parse_insn): Translate word operands to floating point instructions
1627 operating on integers as well as control instructions to short ones
1628 as expected by AT&T syntax. Translate 'd' suffix to short one only for
1629 floating point instructions operating on non-integer operands.
1630 (match_template): Remove fldcw special case. Adjust q-suffix handling
1631 to permit it on fild/fistp/fisttp in AT&T mode.
1632 (process_suffix): Don't guess DefaultSize insns' suffix from
1633 stackop_size for certain floating point control instructions. Guess
1634 suffix for branch and [ls][gi]dt based on flag_code. Split error
1635 messages for Intel and AT&T syntax, and make the condition more strict
1636 for the former. Adjust suppressing of generation of operand size
1638 (intel parser): Allow the full set of MASM operators. Add FWORD, TBYTE,
1639 OWORD, and XMMWORD operand size specifiers (TBYTE replaces XWORD). Add
1640 more error checking.
1641 * config/tc-i386.h (BYTE_PTR WORD_PTR DWORD_PTR QWORD_PTR XWORD_PTR
1642 SHORT OFFSET_FLAT FLAT NONE_FOUND): Remove unused defines.
1646 * symbols.c (colon) [!WORKING_DOT_WORD]: Don't declare
1647 md_short_jump_size, md_long_jump_size.
1648 * write.c [!WORKING_DOT_WORD]: Ditto.
1649 * tc.h [!WORKING_DOT_WORD]: Declare them here. Drop const
1651 * config/tc-cris.h (md_short_jump_size, md_long_jump_size): Don't
1653 * config/tc-cris.c (md_short_jump_size, md_long_jump_size): Drop
1654 const qualifier in these definitions.
1655 * config/tc-i370.c, config/tc-m68k.c, config/tc-pdp11.c,
1656 config/tc-s390.c, config/tc-tahoe.c, config/tc-vax.c: Ditto.
1660 * dwarf2dbg.c (dwarf2_finish): Check for the existence of a file
1661 table before deciding to produce a .debug_line section to match up
1662 with a user provided .debug_info section.
1666 * config/tc-crx.c (getreg_image): Bug fix, a return value was
1667 mistakenly omitted from CRX_C_REGTYPE and CRX_CS_REGTYPE cases.
1671 * config/tc-crx.c: Remove global variable 'post_inc_mode'.
1672 (get_flags): New function.
1673 (get_number_of_bits): Edit comments, update numeric values to
1675 (process_label_constant): Don't support the colon
1676 format (SYMBOL:[s|m|l]).
1677 (set_cons_rparams): Support argument type 'arg_rbase'.
1678 (get_operandtype): Bug fix in 'rbase' operand type parsing.
1679 (handle_LoadStor): Bug fix, first handle post-increment mode.
1680 (getreg_image): Remove redundant code, update according to latest
1682 (print_constant): Bug fix relate to 3-word instructions.
1683 (assemble_insn): Bug fix, when matching instructions, verify also
1684 instruction type (not only mnemonic).
1685 Add various error checking.
1686 (preprocess_reglist): Support HI/LO and user registers.
1690 * config/tc-ia64.c (fixup_unw_records): Don't let the "t" value
1691 in an epilogue directive go negative.
1696 * config/tc-ia64.c (emit_one_bundle): Decrement md.num_slots_in_use
1697 after reporting template error during manual bundling. Reported
1702 * Makefile.am: Run dep-am.
1703 * aclocal.m4: Regenerate with automake 1.9.2.
1704 * Makefile.in: Regenerate with automake 1.9.2.
1705 * doc/Makefile.in: Likewise.
1707 * config/tc-arm.c: Include "dw2gencfi.h".
1708 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
1710 * config/tc-arm.h (TARGET_USE_CFIPOP, DWARF2_DEFAULT_RETURN_COLUMN)
1711 (DWARF2_CIE_DATA_ALIGNMENT, tc_regname_to_dw2regnum)
1712 (tc_cfi_frame_initial_instructions): Define.
1713 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
1718 * config/tc-crx.c (assemble_insn): Check unsigned immediate
1720 Update coprocessor id to be unsigned immediate.
1724 * config/tc-i386.c (O_secrel): Delete.
1725 (tc_pe_dwarf2_emit_offset): New function.
1726 * config/tc-i386.h (O_secrel): Define as O_md1.
1727 (TC_DWARF2_EMIT_OFFSET): Define.
1731 * config/tc-xstormy16.c (xstormy16_cons_fix_new): Accept and
1732 ignore @fptr() directives for 4-byte fixups.
1736 * Makefile.am: Run "make dep-am"
1737 * Makefile.in: Regenerate.
1741 * doc/c-xtensa.texi (Xtensa Options, Absolute Literals Directive):
1742 Remove comments about placement of literal pools.
1743 (Literal Directive): Update description of literal placement.
1744 (Literal Prefix Directive): Remove statement that this does not apply
1745 to absolute-mode literals. Describe new section naming scheme.
1749 * config/tc-xtensa.c (xtensa_frequency_pseudo): Use set_subseg_freq.
1750 (is_entry_opcode, is_movi_opcode, is_the_loop_opcode, is_jx_opcode,
1751 is_windowed_return_opcode): Delete.
1752 (xtensa_frob_label): Use get_subseg_target_freq.
1753 (md_assemble): Inline call to is_entry_opcode.
1754 (xtensa_handle_align): Inline call to get_frag_is_literal.
1755 (relaxation_requirements): Inline call to is_jx_opcode.
1756 (emit_single_op): Inline call to is_movi_opcode.
1757 (xg_assemble_vliw_tokens): Inline calls to get_frag_is_insn,
1758 get_frag_is_no_transform, is_entry_opcode, and
1759 set_frag_is_specific_opcode. Use get_subseg_total_freq.
1760 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags,
1761 xtensa_fix_close_loop_end_frags, relax_frag_immed, convert_frag_immed):
1762 Inline calls to get_frag_is_no_transform.
1763 (next_instrs_are_b_retw): Inline call to is_windowed_return_opcode.
1764 (xtensa_fix_short_loop_frags): Inline calls to is_the_loop_opcode and
1765 get_frag_is_no_transform.
1766 (convert_frag_immed_finish_loop): Inline calls to get_expression_value
1767 and set_frag_is_no_transform.
1768 (get_expression_value): Delete.
1769 (subseg_map struct): Rename cur_total_freq to total_freq. Rename
1770 cur_target_freq to target_freq.
1771 (get_subseg_info): Split out code to create a new map entry into ...
1772 (add_subseg_info): ... this new function.
1773 (get_last_insn_flags): Check if get_subseg_info succeeded.
1774 (set_last_insn_flags): Call add_subseg_info if needed.
1775 (get_subseg_total_freq, get_subseg_target_freq, set_subseg_freq): New.
1776 (xtensa_reorder_segments): Compute last_sec while counting sections.
1777 Remove call to get_last_sec.
1778 (get_last_sec): Delete.
1779 (cache_literal_section): Inline call to retrieve_literal_seg and its
1780 callees, seg_present and add_seg_list.
1781 (retrieve_literal_seg, seg_present, add_seg_list): Delete.
1782 (get_frag_is_insn, get_frag_is_no_transform,
1783 set_frag_is_specific_opcode, set_frag_is_no_transform): Delete.
1784 * config/tc-xtensa.h (MAX_SLOTS): Reduce from 31 to 15.
1788 * config/tc-xtensa.c: Use ISO C90 formatting.
1789 * config/tc-xtensa.h: Likewise.
1790 * config/xtensa-istack.h: Likewise.
1791 * config/xtensa-relax.c: Likewise.
1792 * config/xtensa-relax.h: Likewise.
1796 * config/tc-arm.c (md_begin): Change EF_ARM_EABI_VER3 to
1799 * doc/c-arm.texi: Document that we actually support -meabi=4, not
1804 * doc/as.texinfo (VTableEntry, VTableInherit): Add "directive" to index
1806 (Acknowledgements): Use "GAS" instead of AS variable.
1810 * config/tc-i386.c: Include "elf/x86-64.h".
1811 (i386_elf_section_type): New function.
1812 * config/tc-i386.h (md_elf_section_type): Define.
1813 (i386_elf_section_type): New prototype.
1817 * config/m68k-parse.h (enum m68k_register): New control register,
1819 * config/tc-m68k.c: Correct control register set for MCF5249.
1824 * config/tc-xtensa.c (absolute_literals_supported): New global flag.
1825 (UNREACHABLE_MAX_WIDTH): Define.
1826 (XTENSA_FETCH_WIDTH): Delete.
1827 (cur_vinsn, xtensa_fetch_width, xt_saved_debug_type, past_xtensa_end,
1828 prefer_const16, prefer_l32r): New global variables.
1829 (LIT4_SECTION_NAME): Define.
1830 (lit4_state struct): Add lit4_seg_name and lit4_seg fields.
1831 (XTENSA_PROP_*, GET_XTENSA_PROP_*, SET_XTENSA_PROP_*): Define.
1832 (frag_flags struct): New.
1833 (xtensa_block_info struct): Move from tc-xtensa.h. Add flags field.
1834 (subseg_map struct): Add cur_total_freq and cur_target_freq fields.
1835 (bitfield, bit_is_set, set_bit, clear_bit): Define.
1836 (MAX_FORMATS): Define.
1837 (op_placement_info struct, op_placement_table): New.
1838 (O_pltrel, O_hi16, O_lo16): Define.
1839 (directiveE enum): Rename directive_generics to directive_transform.
1840 Delete directive_relax. Add directive_schedule,
1841 directive_absolute_literals, and directive_last_directive.
1842 (directive_info): Rename "generics" to "transform". Delete "relax".
1843 Add "schedule" and "absolute-literals".
1844 (directive_state): Adjust entries to match changes in directive_info.
1845 (xtensa_relax_statesE, RELAX_IMMED_MAXSTEPS): Move to tc-xtensa.h.
1846 (xtensa_const16_opcode, xtensa_movi_opcode, xtensa_movi_n_opcode,
1847 xtensa_l32r_opcode, xtensa_nop_opcode, xtensa_rsr_lcount_opcode): New.
1848 (xtensa_j_opcode, xtensa_rsr_opcode): Delete.
1849 (align_only_targets, software_a0_b_retw_interlock,
1850 software_avoid_b_j_loop_end, maybe_has_b_j_loop_end,
1851 software_avoid_short_loop, software_avoid_close_loop_end,
1852 software_avoid_all_short_loops, specific_opcode): Delete.
1853 (warn_unaligned_branch_targets): New.
1854 (workaround_a0_b_retw, workaround_b_j_loop_end, workaround_short_loop,
1855 workaround_close_loop_end, workaround_all_short_loops): Default FALSE.
1856 (option_[no_]link_relax, option_[no_]transform,
1857 option_[no_]absolute_literals, option_warn_unaligned_targets,
1858 option_prefer_l32r, option_prefer_const16, option_target_hardware):
1860 (option_[no_]align_only_targets, option_literal_section_name,
1861 option_text_section_name, option_data_section_name,
1862 option_bss_section_name, option_eb, option_el): Delete.
1863 (md_longopts): Add entries for: [no-]transform, [no-]absolute-literals,
1864 warn-unaligned-targets, prefer-l32r, prefer-const16, [no-]link-relax,
1865 and target-hardware. Delete entries for [no-]target-align-only,
1866 literal-section-name, text-section-name, data-section-name, and
1868 (md_parse_option): Handle new options and remove old ones. Accept but
1869 ignore [no-]density options. Warn for [no-]generics and [no-]relax
1870 and treat them as [no-]transform.
1871 (md_show_usage): Add new options and remove old ones.
1872 (xtensa_setup_hw_workarounds): New.
1873 (md_pseudo_table): Change "word" entry to use xtensa_elf_cons. Add
1874 "long", "short", "loc" and "frequency" entries.
1875 (use_generics): Rename to ...
1876 (use_transform): ... this function. Add past_xtensa_end check.
1877 (use_longcalls): Add past_xtensa_end check.
1878 (code_density_available, can_relax): Delete.
1879 (do_align_targets): New.
1880 (get_directive): Accept dashes in directive names. Warn about
1881 [no-]generics and [no-]relax directives and treat them as
1883 (xtensa_begin_directive): Call md_flush_pending_output only for some
1884 directives. Check for directives inside instruction bundles. Warn
1885 about deprecated ".begin literal" usage. Warn and ignore [no-]density
1886 directives. Handle new directives. Check generating_literals flag
1888 (xtensa_end_directive): Check for directives inside instruction
1889 bundles. Warn and ignore [no-]density directives. Handle new
1890 directives. Call xtensa_set_frag_assembly_state.
1891 (xtensa_loc_directive_seen, xtensa_dwarf2_directive_loc,
1892 xtensa_dwarf2_emit_insn): New.
1893 (xtensa_literal_position): Call md_flush_pending_output. Do not check
1894 use_literal_section flag.
1895 (xtensa_literal_pseudo): Call md_flush_pending_output. Handle absolute
1896 literals. Use xtensa_elf_cons to parse the expression.
1897 (xtensa_literal_prefix): Do not check use_literal_section. Support
1898 ".lit4" sections for absolute literals. Change prefix convention to
1899 replace ".text" (or ".t" in a linkonce section). No need to call
1901 (xtensa_frequency_pseudo, xtensa_elf_cons, xtensa_elf_suffix): New.
1902 (expression_end): Handle closing braces and colons.
1903 (PLT_SUFFIX, plt_suffix): Delete.
1904 (expression_maybe_register): Use new xtensa-isa.h functions. Use
1905 xtensa_elf_suffix instead of checking for plt suffix, and handle O_lo16
1906 and O_hi16 expressions as well.
1907 (tokenize_arguments): Handle closing braces and colons.
1908 (parse_arguments): Use new xtensa-isa.h functions. Handle "invisible"
1909 operands and paired register syntax.
1910 (get_invisible_operands): New.
1911 (xg_translate_sysreg_op): Handle new Xtensa LX RSR/WSR/XSR syntax. Use
1912 new xtensa-isa.h functions.
1913 (xtensa_translate_old_userreg_ops, xtensa_translate_zero_immed): New.
1914 (xg_translate_idioms): Check if inside bundle. Use use_transform.
1915 Handle new Xtensa LX RSR/WSR/XSR syntax. Remove code to widen density
1916 instructions. Use xtensa_translate_zero_immed.
1917 (operand_is_immed, operand_is_pcrel_label): Delete.
1918 (get_relaxable_immed): Use new xtensa-isa.h functions.
1919 (get_opcode_from_buf): Add slot parameter. Use new xtensa-isa.h
1921 (xtensa_print_insn_table, print_vliw_insn): New.
1922 (is_direct_call_opcode): Use new xtensa-isa.h functions.
1923 (is_call_opcode, is_loop_opcode, is_conditional_branch_opcode,
1924 is_branch_or_jump_opcode): Delete.
1925 (is_movi_opcode, decode_reloc, encode_reloc, encode_alt_reloc): New.
1926 (opnum_to_reloc, reloc_to_opnum): Delete.
1927 (xtensa_insnbuf_set_operand, xtensa_insnbuf_get_operand): Use new
1928 xtensa-isa.h functions. Operate on one slot of an instruction.
1929 (xtensa_insnbuf_set_immediate_field, is_negatable_branch,
1930 xg_get_insn_size): Delete.
1931 (xg_get_build_instr_size): Use xg_get_single_size.
1932 (xg_is_narrow_insn, xg_is_single_relaxable_insn): Update calls to
1933 xg_build_widen_table. Use xg_get_single_size.
1934 (xg_get_max_narrow_insn_size): Delete.
1935 (xg_get_max_insn_widen_size, xg_get_max_insn_widen_literal_size,
1936 xg_is_relaxable_insn): Update calls to xg_build_widen_table. Use
1938 (xg_build_to_insn): Record the loc field. Handle OP_OPERAND_HI16U and
1939 OP_OPERAND_LOW16U. Check xg_valid_literal_expression.
1940 (xg_expand_to_stack, xg_expand_narrow): Update calls to
1941 xg_build_widen_table. Use xg_get_single_size.
1942 (xg_immeds_fit): Use new xtensa-isa.h functions. Update call to
1944 (xg_symbolic_immeds_fit): Likewise. Also handle O_lo16 and O_hi16, and
1945 treat weak symbols conservatively.
1946 (xg_check_operand): Use new xtensa-isa.h functions.
1947 (is_dnrange): Delete.
1948 (xg_assembly_relax): Inline previous calls to tinsn_copy.
1949 (xg_finish_frag): Specify separate relax states for the frag and slot0.
1950 (is_branch_jmp_to_next, xg_add_branch_and_loop_targets): Use new
1951 xtensa-isa.h functions.
1952 (xg_instruction_matches_option_term, xg_instruction_matches_or_options,
1953 xg_instruction_matches_options): New.
1954 (xg_instruction_matches_rule): Handle O_register expressions. Call
1955 xg_instruction_matches_options.
1956 (transition_rule_cmp): New.
1957 (xg_instruction_match): Update call to xg_build_simplify_table.
1958 (xg_build_token_insn): Record loc fields.
1959 (xg_simplify_insn): Check is_specific_opcode field and
1960 density_supported flag.
1961 (xg_expand_assembly_insn): Skip checking code_density_available. Use
1962 new xtensa-isa.h functions. Call use_transform instead of can_relax.
1963 (xg_assemble_literal): Add error handling for O_big. Call
1964 record_alignment. Handle O_pltrel.
1965 (xg_valid_literal_expression): New.
1966 (xg_assemble_literal_space): Add slot parameter. Remove call to
1967 set_expr_symbol_offset. Add call to record_alignment. Update call to
1969 (xg_emit_insn): Delete.
1970 (xg_emit_insn_to_buf): Add format parameter. Update calls to
1971 xg_add_opcode_fix and xtensa_insnbuf_to_chars.
1972 (xg_add_opcode_fix): Change opcode parameter to tinsn and add format
1973 and slot parameters. Handle new "alternate" relocations for absolute
1974 literals and CONST16 instructions. Check for bad uses of O_lo16 and
1975 O_hi16. Use new xtensa-isa.h functions.
1976 (xg_assemble_tokens): Delete.
1977 (is_register_writer): Use new xtensa-isa.h functions.
1978 (is_bad_loopend_opcode): Check for xtensa_rsr_lcount_opcode instead of
1979 old-style RSR from LCOUNT.
1980 (next_frag_opcode): Delete.
1981 (next_frag_opcode_is_loop, next_frag_format_size, frag_format_size,
1982 update_next_frag_state): New.
1983 (update_next_frag_nop_state): Delete.
1984 (next_frag_pre_opcode_bytes): Use next_frag_opcode_is_loop.
1985 (xtensa_mark_literal_pool_location): Check use_literal_section flag and
1986 the state of the absolute-literals directive. Add calls to
1987 record_alignment and xtensa_set_frag_assembly_state. Call
1988 xtensa_switch_to_non_abs_literal_fragment instead of
1989 xtensa_switch_to_literal_fragment.
1991 (assemble_nop): Use build_nop. Update call to xtensa_insnbuf_to_chars.
1992 (get_expanded_loop_offset): Change check for undefined opcode to an
1994 (xtensa_set_frag_assembly_state, relaxable_section,
1995 xtensa_find_unmarked_state_frags, xtensa_find_unaligned_branch_targets,
1996 xtensa_find_unaligned_loops, xg_apply_tentative_value): New.
1997 (md_begin): Update call to xtensa_isa_init. Initialize linkrelax to 1.
1998 Set lit4_seg_name. Call xg_init_vinsn. Initialize new global opcodes.
1999 Call init_op_placement_info_table and xtensa_set_frag_assembly_state.
2000 (xtensa_init_fix_data): New.
2001 (xtensa_frob_label): Reset label symbol to the current frag. Check
2002 do_align_targets and generating_literals flag. Propagate frequency
2003 info to new alignment frag. Call xtensa_set_frag_assembly_state.
2004 (xtensa_unrecognized_line): New.
2005 (xtensa_flush_pending_output): Check if inside a bundle. Add a call
2006 to xtensa_set_frag_assembly_state.
2007 (error_reset_cur_vinsn): New.
2008 (md_assemble): Remove check for literal frag. Remove call to
2009 istack_init. Call use_transform instead of use_generics. Parse
2010 explicit instruction format specifiers. Move code for
2011 a0_b_retw_interlock workaround to xg_assemble_vliw_tokens. Call
2012 error_reset_cur_vinsn on errors. Add call to get_invisible_operands.
2013 Add dwarf2_where call. Remote automatic alignment for ENTRY
2014 instructions. Move call to xtensa_clear_insn_labels to the end.
2015 Rearrange to handle bundles.
2016 (xtensa_cons_fix_new): Delete.
2017 (xtensa_handle_align): New.
2018 (xtensa_frag_init): Call xtensa_set_frag_assembly_state. Remove
2019 assignment to is_no_density field.
2020 (md_pcrel_from): Use new xtensa-isa.h functions. Use decode_reloc
2021 instead of reloc_to_opnum. Handle "alternate" relocations.
2022 (xtensa_force_relocation, xtensa_check_inside_bundle,
2023 xtensa_elf_section_change_hook): New.
2024 (xtensa_symbol_new_hook): Delete.
2025 (xtensa_fix_adjustable): Check for difference of symbols with an
2026 offset. Check for external and weak symbols.
2027 (md_apply_fix3): Remove cases for XTENSA_OP{0,1,2} relocs.
2028 (md_estimate_size_before_relax): Return expansion for the first slot.
2029 (tc_gen_reloc): Handle difference of symbols by producing
2030 XTENSA_DIFF{8,16,32} relocs and by writing the value of the difference
2031 into the output. Handle new XTENSA_SLOT*_OP relocs by storing the
2032 tentative values into the output when linkrelax is set.
2033 (XTENSA_PROP_SEC_NAME): Define.
2034 (xtensa_post_relax_hook): Call xtensa_find_unmarked_state_frags.
2035 Create literal tables only if using literal sections. Create new
2036 property tables instead of old instruction tables. Check for unaligned
2037 branch targets and loops.
2038 (finish_vinsn, find_vinsn_conflicts, check_t1_t2_reads_and_writes,
2039 new_resource_table, clear_resource_table, resize_resource_table,
2040 resources_available, reserve_resources, release_resources,
2041 opcode_funcUnit_use_unit, opcode_funcUnit_use_stage,
2042 resources_conflict, xg_find_narrowest_format, relaxation_requirements,
2043 bundle_single_op, emit_single_op, xg_assemble_vliw_tokens): New.
2044 (xtensa_end): Call xtensa_flush_pending_output. Set past_xtensa_end
2045 flag. Update checks for workaround options. Call
2046 xtensa_mark_narrow_branches and xtensa_mark_zcl_first_insns.
2047 (xtensa_cleanup_align_frags): Add special case for branch targets.
2048 Check for and mark unreachable frags.
2049 (xtensa_fix_target_frags): Remove use of align_only_targets flag.
2050 Use RELAX_LOOP_END_BYTES in special case for negatable branch at the
2051 end of a zero-overhead loop body.
2052 (frag_can_negate_branch): Handle instructions with multiple slots.
2053 Use new xtensa-isa.h functions
2054 (xtensa_mark_narrow_branches, is_narrow_branch_guaranteed_in_range,
2055 xtensa_mark_zcl_first_insns): New.
2056 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags): Error if
2057 transformations are disabled.
2058 (next_instrs_are_b_retw): Use new xtensa-isa.h functions. Handle
2059 multislot instructions.
2060 (xtensa_fix_close_loop_end_frags, xtensa_fix_short_loop_frags):
2061 Likewise. Also error if transformations are disabled.
2062 (unrelaxed_frag_max_size): New.
2063 (unrelaxed_frag_min_insn_count, unrelax_frag_has_b_j): Use new
2064 xtensa-isa.h functions.
2065 (xtensa_sanity_check, is_empty_loop, is_local_forward_loop): Use
2066 xtensa_opcode_is_loop instead of is_loop_opcode.
2067 (get_text_align_power): Replace as_fatal with assertion.
2068 (get_text_align_fill_size): Iterate instead of using modulus when
2070 (get_noop_aligned_address): Assert that this is for a machine-dependent
2071 RELAX_ALIGN_NEXT_OPCODE frag. Use next_frag_opcode_is_loop,
2072 xg_get_single_size, and frag_format_size.
2073 (get_widen_aligned_address): Rename to ...
2074 (get_aligned_diff): ... this function. Add max_diff parameter.
2075 Remove handling of rs_align/rs_align_code frags. Use
2076 next_frag_format_size, get_text_align_power, get_text_align_fill_size,
2077 next_frag_opcode_is_loop, and xg_get_single_size. Compute max_diff
2078 and pass it back to caller.
2079 (xtensa_relax_frag): Use relax_frag_loop_align. Add code for new
2080 RELAX_SLOTS, RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN,
2081 RELAX_FILL_NOP, and RELAX_UNREACHABLE frag types. Check relax_seen.
2082 (relax_frag_text_align): Rename to ...
2083 (relax_frag_loop_align): ... this function. Assume loops can only be
2084 in the first slot of an instruction.
2085 (relax_frag_add_nop): Use assemble_nop instead of constructing an OR
2086 instruction. Remove call to frag_wane.
2087 (relax_frag_narrow): Rename to ...
2088 (relax_frag_for_align): ... this function. Extend to handle
2089 RELAX_FILL_NOP and RELAX_UNREACHABLE, as well as RELAX_SLOTS with
2090 RELAX_NARROW for the first slot.
2091 (find_address_of_next_align_frag, bytes_to_stretch): New.
2092 (future_alignment_required): Use find_address_of_next_align_frag and
2093 bytes_to_stretch. Look ahead to subsequent frags to make smarter
2094 alignment decisions.
2095 (relax_frag_immed): Add format, slot, and estimate_only parameters.
2096 Check if transformations are enabled for b_j_loop_end workaround.
2097 Use new xtensa-isa.h functions and handle multislot instructions.
2098 Update call to xg_assembly_relax.
2099 (md_convert_frag): Handle new RELAX_SLOTS, RELAX_UNREACHABLE,
2100 RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN, and RELAX_FILL_NOP
2102 (convert_frag_narrow): Add segP, format and slot parameters. Call
2103 convert_frag_immed for branch instructions. Adjust calls to
2104 tinsn_from_chars, tinsn_immed_from_frag, and xg_emit_insn_to_buf. Use
2105 xg_get_single_size and xg_get_single_format.
2106 (convert_frag_fill_nop): New.
2107 (convert_frag_immed): Add format and slot parameters. Handle multislot
2108 instructions and use new xtensa-isa.h functions. Update calls to
2109 tinsn_immed_from_frag and xg_assembly_relax. Check if transformations
2110 enabled for b_j_loop_end workaround. Use build_nop instead of
2111 assemble_nop. Check is_specific_opcode flag. Check for unreachable
2112 frags. Use xg_get_single_size. Handle O_pltrel.
2113 (fix_new_exp_in_seg): Remove check for old plt flag.
2114 (convert_frag_immed_finish_loop): Update calls to tinsn_from_chars and
2115 xtensa_insnbuf_to_chars. Call tinsn_immed_from_frag. Change check
2116 for loop opcode to an assertion. Mark all frags up to the end of the
2117 loop as not transformable.
2118 (get_last_insn_flags, set_last_insn_flags): Use get_subseg_info.
2119 (get_subseg_info): New.
2120 (xtensa_move_literals): Call xtensa_set_frag_assembly_state. Add null
2122 (xtensa_switch_to_literal_fragment): Rewrite to handle absolute
2123 literals and use xtensa_switch_to_non_abs_literal_fragment otherwise.
2124 (xtensa_switch_to_non_abs_literal_fragment): New.
2125 (cache_literal_section): Add is_code parameter and pass it through to
2126 retrieve_literal_seg.
2127 (retrieve_literal_seg): Add is_code parameter and use it to set the
2128 flags on the literal section. Handle case where head parameter is 0.
2129 (get_frag_is_no_transform, set_frag_is_specific_opcode,
2130 set_frag_is_no_transform): New.
2131 (xtensa_create_property_segments): Add end_property_function parameter
2132 and pass it through to add_xt_block_frags. Call bfd_get_section_flags
2133 and skip SEC_DEBUGGING and !SEC_ALLOC sections.
2134 (xtensa_create_xproperty_segments, section_has_xproperty): New.
2135 (add_xt_block_frags): Add end_property_function parameter and call it
2136 if it is non-zero. Call xtensa_frag_flags_init.
2137 (xtensa_frag_flags_is_empty, xtensa_frag_flags_init,
2138 get_frag_property_flags, frag_flags_to_number,
2139 xtensa_frag_flags_combinable, xt_block_aligned_size,
2140 xtensa_xt_block_combine, add_xt_prop_frags,
2141 init_op_placement_info_table, opcode_fits_format_slot,
2142 xg_get_single_size, xg_get_single_format): New.
2143 (istack_push): Inline call to tinsn_copy.
2144 (tinsn_copy): Delete.
2145 (tinsn_has_invalid_symbolic_operands): Handle O_hi16 and O_lo16 and
2146 CONST16 opcodes. Handle O_big, O_illegal, and O_absent.
2147 (tinsn_has_complex_operands): Handle O_hi16 and O_lo16.
2148 (tinsn_to_insnbuf): Use xg_get_single_format and new xtensa-isa.h
2149 functions. Handle invisible operands.
2150 (tinsn_to_slotbuf): New.
2151 (tinsn_check_arguments): Use new xtensa-isa.h functions.
2152 (tinsn_from_chars): Add slot parameter. Rewrite using xg_init_vinsn,
2153 vinsn_from_chars, and xg_free_vinsn.
2154 (tinsn_from_insnbuf): New.
2155 (tinsn_immed_from_frag): Add slot parameter and handle multislot
2156 instructions. Handle symbol differences.
2157 (get_num_stack_text_bytes): Use xg_get_single_size.
2158 (xg_init_vinsn, xg_clear_vinsn, vinsn_has_specific_opcodes,
2159 xg_free_vinsn, vinsn_to_insnbuf, vinsn_from_chars, expr_is_register,
2160 get_expr_register, set_expr_symbol_offset_diff): New.
2161 * config/tc-xtensa.h (MAX_SLOTS): Define.
2162 (xtensa_relax_statesE): Move from tc-xtensa.c. Add
2163 RELAX_CHECK_ALIGN_NEXT_OPCODE, RELAX_MAYBE_DESIRE_ALIGN, RELAX_SLOTS,
2164 RELAX_FILL_NOP, RELAX_UNREACHABLE, RELAX_MAYBE_UNREACHABLE, and
2166 (RELAX_IMMED_MAXSTEPS): Move from tc-xtensa.c.
2167 (xtensa_frag_type struct): Add is_assembly_state_set,
2168 use_absolute_literals, relax_seen, is_unreachable, is_specific_opcode,
2169 is_align, is_text_align, alignment, and is_first_loop_insn fields.
2170 Replace is_generics and is_relax fields by is_no_transform field.
2171 Delete is_text and is_longcalls fields. Change text_expansion and
2172 literal_expansion to arrays of MAX_SLOTS entries. Add arrays of
2173 per-slot information: literal_frags, slot_subtypes, slot_symbols,
2174 slot_sub_symbols, and slot_offsets. Add fr_prev field.
2175 (xtensa_fix_data struct): New.
2176 (xtensa_symfield_type struct): Delete plt field.
2177 (xtensa_block_info struct): Move definition to tc-xtensa.h. Add
2178 forward declaration here.
2179 (xt_section_type enum): Delete xt_insn_sec. Add xt_prop_sec.
2180 (XTENSA_SECTION_RENAME): Undefine.
2181 (TC_FIX_TYPE, TC_INIT_FIX_DATA, TC_FORCE_RELOCATION, NO_PSEUDO_DOT,
2182 tc_unrecognized_line, md_do_align, md_elf_section_change_hook,
2183 HANDLE_ALIGN, TC_LINKRELAX_FIXUP, SUB_SEGMENT_ALIGN): Define.
2184 (TC_CONS_FIX_NEW, tc_symbol_new_hook): Delete.
2185 (unit_num_copies_func, opcode_num_units_func,
2186 opcode_funcUnit_use_unit_func, opcode_funcUnit_use_stage_func): New.
2187 (resource_table struct): New.
2188 * config/xtensa-istack.h (MAX_INSN_ARGS): Increase from 6 to 10.
2189 (TInsn struct): Add keep_wide, loc, fixup, record_fix, subtype,
2190 literal_space, symbol, sub_symbol, offset, and literal_frag fields.
2191 (tinsn_copy): Delete prototype.
2192 (vliw_insn struct): New.
2193 * config/xtensa-relax.c (insn_pattern_struct): Add options field.
2194 (widen_spec_list): Add option conditions for density and boolean
2195 instructions. Add expansions using CONST16 and conditions for using
2196 CONST16 vs. L32R. Use new Xtensa LX RSR/WSR syntax. Add entries for
2198 (simplify_spec_list): Add option conditions for density instructions.
2199 Add entry for NOP instruction.
2200 (append_transition): Add cmp function pointer parameter and use it to
2201 insert the new entry in order.
2202 (operand_function_LOW16U, operand_function_HI16U): New.
2203 (xg_has_userdef_op_fn, xg_apply_userdef_op_fn): Handle
2204 OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
2205 (enter_opname, split_string): Use xstrdup instead of strdup.
2206 (init_insn_pattern): Initialize new options field.
2207 (clear_req_or_option_list, clear_req_option_list,
2208 clone_req_or_option_list, clone_req_option_list, parse_option_cond):
2210 (parse_insn_pattern): Parse option conditions.
2211 (transition_applies): New.
2212 (build_transition): Use new xtensa-isa.h functions. Fix incorrectly
2213 swapped last arguments in calls to append_constant_value_condition.
2214 Call clone_req_option_list. Add warning about invalid opcode.
2215 Handle LOW16U and HI16U function names.
2216 (build_transition_table): Add cmp parameter and use it in calls to
2217 append_transition. Use new xtensa-isa.h functions. Check
2218 transition_applies before adding entries.
2219 (xg_build_widen_table, xg_build_simplify_table): Add cmp parameter and
2220 pass it through to build_transition_table.
2221 * config/xtensa-relax.h (ReqOrOptionList, ReqOrOption, ReqOptionList,
2222 ReqOption, transition_cmp_fn): New types.
2223 (OpType enum): Add OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
2224 (transition_rule struct): Add options field.
2225 * doc/as.texinfo (Overview): Update Xtensa options.
2226 * doc/c-xtensa.texi (Xtensa Options): Delete --[no-]density,
2227 --[no-]relax, and --[no-]generics options. Update descriptions of
2228 --text-section-literals and --[no-]longcalls. Add
2229 --[no-]absolute-literals and --[no-]transform.
2230 (Xtensa Syntax): Add description of syntax for FLIX instructions.
2231 Remove use of "generic" and "specific" terminology for opcodes.
2232 (Xtensa Registers): Generalize the syntax description to include
2233 user-defined register files.
2234 (Xtensa Automatic Alignment): Update.
2235 (Xtensa Branch Relaxation): Mention limitation of unconditional jumps.
2236 (Xtensa Call Relaxation): Linker can now remove most of the overhead.
2237 (Xtensa Directives): Remove confusing rules about precedence.
2238 (Density Directive, Relax Directive): Delete.
2239 (Schedule Directive): New.
2240 (Generics Directive): Rename to ...
2241 (Transform Directive): ... this node.
2242 (Literal Directive): Update for absolute literals. Missing
2243 literal_position directive is now an error.
2244 (Literal Position Directive): Update for absolute literals.
2245 (Freeregs Directive): Delete.
2246 (Absolute Literals Directive): New.
2247 (Frame Directive): Minor editing.
2248 * Makefile.am (DEPTC_xtensa_elf, DEPOBJ_xtensa_elf, DEP_xtensa_elf):
2249 Update dependencies.
2250 * Makefile.in: Regenerate.
2254 * config/tc-mips.c (append_insn): Use fix_new rather than fix_new_exp
2255 to build the second and third fixups for a composite relocation.
2256 (macro_read_relocs): New function.
2257 (macro_build): Use it.
2258 (s_cpsetup): Pass all three composite relocation codes to macro_build.
2259 Simplify fragging code accordingly.
2260 (s_gpdword): Use fix_new rather than fix_new_exp for the second part
2261 of the composite relocation. Set fx_tcbit in both fixups.
2265 * config/tc-mips.c (append_insn): Set fx_tcbit for composite relocs.
2266 (md_apply_fix3): Don't treat composite relocs as done.
2270 * macro.c (macro_expand_body): When ELF, use .LL rather than LL as
2271 prefix for symbol names generated from the LOCAL macro directive.
2273 * dw2gencfi.c (select_cie_for_fde): When separating CIE out from
2274 FDE, treat a DW_CFA_remember_state as we do a DW_CFA_advance_loc.
2278 * config/tc-crx.c (preprocess_reglist): Handle Co-processor
2280 (md_assemble): Add error checking for Co-Processor instructions.
2281 (get_cinv_parameters): Add 'b' option to invalidate the
2282 branch-target cache.
2286 * config/tc-arm.c (unwind): New variable.
2287 (vfp_sp_encode_reg): New function.
2288 (vfp_sp_reg_required_here): Use it.
2289 (vfp_sp_reg_list, vfp_dp_reg_list): Remove.
2290 (vfp_parse_reg_list): New function.
2291 (s_arm_unwind_fnstart, s_arm_unwind_fnend, s_arm_unwind_cantunwind,
2292 s_arm_unwind_personality, s_arm_unwind_personalityindex,
2293 s_arm_unwind_handlerdata, s_arm_unwind_save, s_arm_unwind_movsp,
2294 s_arm_unwind_pad, s_arm_unwind_setfp, s_arm_unwind_raw): New
2296 (md_pseudo_table): Add them.
2297 (do_vfp_reg2_from_sp2): Use vfp_parse_reg_list and vfp_sp_encode_reg.
2298 (do_vfp_sp2_from_reg2, vfp_sp_ldstm, vfp_dp_ldstm): Ditto.
2299 (set_section, add_unwind_adjustsp, flush_pending_unwind,
2300 finish_unwind_opcodes, start_unwind_section, create_unwind_entry,
2301 require_hashconst, add_unwind_opcode): New functions.
2302 * doc/c-arm.texi: Document unwinding opcodes.
2303 * NEWS: Mention the new feature.
2307 * config/tc-mips.c (md_apply_fix3): Remove erroneous assert.
2311 * config/tc-ppc.c (md_apply_fix3): Call S_SET_THREAD_LOCAL for
2313 * config/tc-s390.c (md_apply_fix3): Likewise.
2314 * config/tc-sparc.c (md_apply_fix3): Likewise.
2318 * config/tc-arm.c (arm_elf_section_type): New function.
2319 (arm_elf_change_section): Set section link for exidx sections.
2320 * config/tc-arm.h (arm_elf_section_type): Add prototype.
2321 (md_elf_section_type): Define.
2325 * config/tc-pdp11.c (md_apply_fix3): Change to sign of the SOB
2326 instruction's offset.
2330 * (TARGET_FORMAT): Remove LynxOS COFF definition.
2334 * config/tc-arc.c (tc_gen_reloc): Don't assume fixP->fx_addsy is an
2335 asymbol *, instead use symbol_get_bfdsym.
2339 * config/tc-m68k.c (select_control_regs): Add mcf5249.
2343 * config/tc-arm.c (do_smi, do_nop): New functions.
2344 (insns): Add ARMv6ZK instructions.
2345 (md_apply_fix3): Handle BFD_RELOC_ARM_SMI.
2346 (tc_gen_reloc): Ditto.
2347 (arm_cpus): Add mpcore and arm1176.
2348 (arm_archs): Add armv6{k,z,zk}.
2349 * doc/c-arm.texi: Document new cores and architectures.
2353 * config/tc-arm.c: Use ISO C90 formatting.
2357 * config/tc-arm.c (mav_reg_required_here): Allow REG_TYPE_CN
2358 as alternative when REG_TYPE_MVF, REG_TYPE_MVD, REG_TYPE_MVFX or
2359 REG_TYPE_MVDX is expected.
2363 * doc/c-i386.texi (i386-Mnemonics): Fix typo.
2367 * config/tc-ia64.c (ENCODED_PSP_OFFSET): New.
2368 (output_rp_psprel, output_pfs_psprel, output_preds_psprel,
2369 output_spill_base, output_unat_psprel, output_lc_psprel,
2370 output_fpsr_psprel, output_priunat_psprel, output_bsp_psprel,
2371 output_bsprestore_psprel, output_rnat_psprel, output_spill_psprel,
2372 output_spill_psprel_p): Use it.
2376 * config/tc-crx.c (handle_LoadStor): New function.
2377 Handle load/stor unique instructions before parsing.
2381 * config/tc-arm.c (s_arm_rel31): New funciton.
2382 (md_pseudo_table): Add .rel31.
2383 (md_apply_fix3): Handle BFD_RELOC_ARM_TARGET2,
2384 BFD_RELOC_32_PCREL and BFD_RELOC_ARM_PREL31.
2385 (tc_gen_reloc): Handle BFD_RELOC_ARM_PREL31 and BFD_RELOC_ARM_TARGET2.
2386 (arm_fix_adjustable): Return 0 for BFD_RELOC_ARM_TARGET2.
2387 (arm_parse_reloc): Add (target2).
2391 * Makefile.am: Run "make dep-am".
2392 * Makefile.in: Regenerate.
2393 * aclocal.m4: Regenerate.
2394 * configure: Regenerate.
2395 * doc/Makefile.in: Regenerate.
2396 * po/POTFILES.in: Regenerate.
2397 * po/gas.pot: Regenerate.
2401 * config/tc-mmix.c [!LLONG_MIN]: Correct #elsif to #elif.
2402 [!LLONG_MAX]: Ditto.
2406 * config/tc-arm.c: Rename RELABS to TARGET1.
2410 * messages.c (as_internal_value_out_of_range): Cast values passed
2411 to as_bad_where or as_warn_where to proper type.
2415 * config/tc-avr.c: Add support for
2416 atmega48, atmega88, atmega168, attiny13, attiny2313, at90can128.
2420 * dw2gencfi.c (select_cie_for_fde): When separating CIE out
2421 from FDE, treat a CFI_escape as we do a DW_CFA_advance_loc.
2425 * config/obj-elf.c (obj_elf_section_type): Handle init_array,
2426 fini_array and preinit_array section types.
2427 * config/tc-ia64.c (ia64_elf_section_type): Remove init_array
2429 * doc/as.texinfo: Document extra section types.
2433 * Makefile.am (TARG_ENV_HFILES): Add te-symbian.h.
2434 * Makefile.in: Regenerated.
2435 * configure.in: Set em for arm*-*-symbianelf*.
2436 * configure: Regenerated.
2437 * config/tc-arm.c (elf32_arm_target_format): Use Symbian target
2438 vectors when appropriate.
2439 * config/te-symbian.h: New file.
2443 * config/tc-crx.c (gettrap): Exception vector can be case
2445 (process_label_constant): Fix a 32-bit displacement bug in branch
2447 (get_operandtype) : Bug fix, wrong operand was used.
2448 (process_label_constant): Initialize relocation type to
2453 * tc-arm.c (arm_cpus, arm_fpus): Allow <cpu>-s as well as <cpu>s
2454 for synthesizable cores.
2456 * doc/c-arm.texi (ARM Options): Document canonical names of CPUs.
2460 * config/tc-msp430.c: Clean-up the code.
2461 (md_relax_table): New relax table.
2462 (mcu_types): Sort MCU types.
2463 (md_pseudo_table): Add .profiler pseudo handler.
2464 (pow2value): New function.
2465 (msp430_profiler): New function.
2466 (msp430_operands): Add new insns handlers.
2467 (msp430_srcoperand): Add register operand handler, allow complex
2469 (md_estimate_size_before_relax): Rewritten.
2470 (md_convert_frag): Rewritten.
2471 (msp430_relax_frag): New function.
2472 * config/tc-msp430.h (md_relax_frag): define macro
2473 * doc/c-msp430.texi: Update information.
2477 * as.c (std_shortopts): Allow -g to take an optional argument.
2478 (parse_args): Pass any switch starting with -g on to the backend
2483 * configure.in (arm*-*-symbianelf*): New target.
2484 (arm*-*-eabi*): Likewise.
2485 * configure: Regenerated.
2488 * config/tc-mips.c (append_insn): Handle delay slots in branch likely
2493 * config/tc-ia64.c (start_unwind_section): Add linkonce_empty
2494 argument, don't do anything if current section is not
2495 .gnu.linkonce.t.* and linkonce_empty is set.
2496 (generate_unwind_image, dot_endp): Adjust callers, call
2497 start_unwind_section (*, 1) if nothing will be put into the
2502 * as.c (MD_DEBUG_FORMAT_SELECTOR): Provide default definition.
2503 (show_usage): Add -g.
2504 (std_longopts): Add --gen-debug. Alpha sort the table.
2505 (parse_args): Print an error message if a switch is not handled.
2506 Handle the -g switch, calling md_debug_format_selector() if
2508 * NEWS: Mention new feature.
2509 * doc/as.texinfo: Document new switch.
2510 * doc/internals.texi: Document behaviour of md_parse_option.
2512 * config/tc-arm.c (md_parse_option): Do not issue an error message
2513 if the switch is not recognised.
2514 * config/tc-m68k.c (md_parse_option): Likewise.
2515 * config/tc-pdp11.c (md_parse_option): Likewise.
2516 * config/tc-v850.c (md_parse_option): Likewise.
2518 * as.h: Fix up formatting.
2523 * macro.c (macro_set_alternate): Use ISO C90 formatting.
2525 * configure.in: Sort architecture based tables alphabetically.
2526 * configure: Regenerate.
2530 * config/tc-ppc.c (tc_ppc_regname_to_dw2regnum <regnames>): Replace
2531 { "cc", 68 }, with { "cr", 70 }.
2536 * as.c: Add and handle new --alternate command line option.
2537 * macro.c (macro_set_alternate): New.
2538 * macro.h (macro_set_alternate): Declare.
2539 * read.c: Add and handle new .altmacro and .noaltmacro directives.
2540 * doc/as.texinfo: Document new command line option and pseudo-ops
2541 as well as insert documentation originating from gasp about
2542 alternate macro syntax.
2543 * NEWS: Mention new command line option and pseudo-ops.
2547 * expr.c (operand): Handle the "~", "-", and "!" operators applied
2552 * config/tc-arm.c (md_apply_fix3, tc_gen_reloc, arm_parse_reloc):
2553 Handle new relocations.
2554 * include/elf/arm.h (elf_arm_reloc_type): Add new EABI relocations.
2558 * write.c (relax_segment): Use was_address instead of address when
2559 setting fr_fix field for align frag due to backwards .org.
2563 Introduce SH2a support.
2565 * config/tc-sh.c (get_specific): Change arch_sh2a_up to
2568 * config/tc-sh.c (md_parse_option): Add sh2a-nofpu ISA handling.
2570 * config/tc-sh.c (sh_elf_final_processing): Move sh2a recognition
2571 to end of conditional expression.
2573 * config/tc-sh.c: Add sh2a-nofpu support.
2575 * tc-sh.c: Add sh2a support.
2576 (parse_reg): Add tbr.
2577 (parse_at): Support @@(disp,tbr).
2578 (get_specific): Support sh2a opcodes.
2579 (insert4): New, for 4 byte relocs.
2580 (build_Mytes): Support sh2a opcodes.
2581 (md_apply_fix3_Mytes): Support sh2a opcodes.
2583 * config/tc-sh.c (md_parse_option): Handle sh2a.
2584 (sh_elf_final_processing): Ditto.
2588 * config/tc-hppa.h (TARGET_FORMAT): Set to "elf32-hppa-netbsd"
2593 * config/tc-ppc.c (ppc_frob_file_before_adjust): Warn if .toc too big.
2594 (ppc_arch): Expand comment.
2598 * config/tc-crx.c: Support evaluating the difference between two
2600 * config/tc-crx.h: Likewise.
2604 * config/tc-ia64.c (start_unwind_section): Set the linked-to
2606 (ia64_elf_section_change_hook): Set the linked-to section for
2611 * config/tc-msp430.c: Add new subtargets: msp430x1610,
2612 msp430x1611, msp430x1612, msp430x415, msp430x417, msp430xG437,
2613 msp430xG438, msp430xG439.
2617 * doc/as.texinfo (Section, PushSection): Correct documentation
2622 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
2623 permissible for the selected instruction suffix.
2624 (match_template): Don't permit 64-bit general purpose operands in
2626 (finalize_imm): Permit 64-bit immediates.
2627 (build_modrm_byte): Don't treat 32-bit addressing in 64-bit mode
2628 specially except for the width of the used base and/or index
2629 registers. For 32-bit displacements, use sign-extended
2630 relocations only when using 64-bit addressing.
2631 Force zero displacement on rip-relative addressing when there is
2632 no other displacement.
2633 (i386_index_check): Don't treat 32-bit addressing in 64-bit mode
2634 specially except for the width of the used base and/or index
2636 (parse_register): Disallow Reg64 registers in 32-bit mode.
2638 * config/tc-i386.c: For DefaultSize instructions, don't guess a 'q'
2639 suffix if the instruction doesn't support it.
2643 * config/tc-mips.c (append_insn): Handle constant expressions with
2644 no associated relocation.
2645 (mips_ip): Cancel the expression after use for the Q format
2647 (parse_relocation): Return no relocation for unsupported
2649 (my_getSmallExpression): Return no relocation if no relocation
2654 * config/obj-som.c (adjust_stab_sections): Add prototype.
2655 (obj_som_compiler, obj_som_version, obj_som_copyright,
2656 adjust_stab_sections): Add ATTRIBUTE_UNUSED to unused arguments.
2657 * config/tc-hppa.c (update_subspace): Likewise.
2658 (is_defined_subspace): Amplify comment.
2659 * config/obj-som.h (som_frob_file): Add prototype.
2664 * subsegs.c (section_symbol): Don't create a new segment when
2665 existing segment is undefined.
2669 * config/tc-arm.c: Include include/opcode/arm.h.
2670 (ARM_EXT_*, ARM_ARCH_*, ARM_ANY, ARM_ALL, COPROC_ANY): Delete.
2671 (FPU_FPA_EXT_* FPU_VFP_EXT_*, FPU_ANY, FPU_NONE, FPU_MAVERICK): Delete.
2672 (FPU_ARCH_*): Delete.
2673 * Makefile.am: Update dependencies.
2674 * Makefile.in: Regenerate.
2678 * configure.in: Accept armbe as a big-endian arm configuration.
2679 * configure: Regenerate.
2683 * config/tc-i386.c (T_SHIFTOP): New constant.
2684 (intel_e05_1): Handle '&', '|' and T_SHIFTOP.
2685 (intel_el1): Handle '~'.
2686 (intel_get_token): Handle '<>', '&', '|' and '~'.
2690 (md_assemble): Remove spurious newline from end of as_bad error
2692 (intel_e05_1): Likewise.
2693 (intel_e11): Likewise.
2694 (intel_match_token): Likewise.
2698 * config/tc-m68k.c: Convert to C90. Remove redundant
2699 declarations. Indentation fixup.
2700 [M68KCOFF]: Include "obj-coff.h" instead of declaring
2701 obj_coff_section ourselves.
2705 * config/tc-ia64.c (default_big_endian): New.
2706 (dot_byteorder, md_begin): Use it.
2707 (md_parse_option): Set it.
2711 * configure.in: Change sh-sybmian-elf to sh-*-symbianelf.
2712 * configure: Regenerate.
2713 * NEWS: Change sh-sybmian-elf to sh-*-symbianelf.
2714 * config/tc-sh.c (sh_elf_final_processing): Use renamed version of
2715 sh_find_elf_flags if necessary.
2719 * config/tc-mips.c (mips_fix_adjustable): If the full addend is
2720 going to be split into more than one in-place addend, return 0
2721 for relocations against mergeable sections. Associate comments
2726 * Makefile.am (CPU_TYPES): Add crx.
2727 (TARGET_CPU_CFILES): Add config/tc-crx.c.
2728 (TARGET_CPU_HFILES): Add config/tc-crx.h.
2729 (DEPTC_crx_elf): New target.
2730 (DEPOBJ_crx_elf): Likewise.
2731 (DEP_crx_elf): Likewise.
2732 * Makefile.in: Regenerate.
2733 * configure.in: Add crx* target.
2734 * configure: Regenerate.
2735 * config/tc-crx.c: New file.
2736 * config/tc-crx.h: New file.
2737 * NEWS: Mention new target.
2741 * config.in: Undefine TARGET_SYMBIAN by default.
2743 * configure: Regenerate. Add sh-symbian-elf target. If
2744 selected define TARGET_SYMBIAN.
2745 * config/tc-sh.h (TARGET_FORMAT): Select a Symbian target
2746 format if TARGET_SYMBIAN has been defined.
2748 * output-file.c (output_file_create): Report the target format
2749 chosen when bfd_openw reports that it is invalid.
2751 * config/obj-coff.c (coff_pseudo_table): Only define the weak
2752 pseudo for BFD based assemblers.
2757 * config/tc-sh.c (md_assemble): Change isspace to ISSPACE.
2758 (md_parse_option): Remove redundant -isa testing.
2759 Make bfd_arch variable const.
2760 (md_show_usage): Make bfd_arch variable const.
2764 * config/tc-ia64.c (emit_one_bundle): Check and set insn_addr.
2765 * config/tc-ia64.h (md_frag_check): Define.
2769 * config/obj-coff.c (obj_coff_weak): New .weak syntax for PE weak
2771 * doc/as.texinfo (Weak): Document PE weak symbols.
2775 * config/tc-mips.c (HAVE_IN_PLACE_ADDENDS): New macro.
2776 (reloc_needs_lo_p): Only return true if HAVE_IN_PLACE_ADDENDS.
2777 (mips_frob_file): Rework so that only a single pass through the
2778 relocs is needed. Allow %lo()s to have higher offsets than their
2779 corresponding %hi()s or %got()s.
2783 * config/tc-arm.c (md_apply_fix3:BFD_RELOC_ARM_IMMEDIATE): Do not
2784 allow values which have come from undefined symbols.
2785 Always consider this fixup to have been processed as a reloc
2786 cannot be generated for it.
2790 * frags.h (struct frag): Add has_code and insn_addr fields.
2791 * write.c (cvt_frag_to_fill): Invoke md_frag_check.
2792 * config/tc-ppc.c (md_assemble): Check and set insn_addr.
2793 * config/tc-ppc.h (md_frag_check): Define.
2797 * doc/Makefile.am (info): Rename goal to...
2798 (info-local): ... this, to preserve implicit dependencies.
2799 * doc/Makefile.in: Regenerate with automake 1.8.5.
2803 * config/tc-m32r.c (md_convert_frag): Changed for @PLT.
2804 (m32r_cgen_record_fixup_exp): Changed for @GOTOFF, @GOT.
2805 (m32r_fix_adjustable): Changed for @GOTOFF, @GOT, @PLT.
2806 (tc_gen_reloc): Likewise.
2807 (m32r_end_of_match): Add for @GOTOFF, @GOT, @PLT.
2808 (m32r_parse_name): Likewise.
2809 (m32r_cgen_parse_fix_exp): Likewise.
2810 * config/tc-m32r.h (md_parse_name): Define for @GOTOFF, @GOT, @PLT.
2811 (O_PIC_reloc): Likewise.
2812 (TC_CGEN_PARSE_FIX_EXP): Likewise..
2813 * cgen.c (gas_cgen_parse_operand): Add TC_CGEN_PARSE_FIX_EXP
2814 for @GOTOFF, @GOT, @PLT.
2818 * gas/symbols.c: While discarding ordinary local absolute symbols
2819 when --strip-local-absolute is in effect, retain file symbols.
2823 * config/tc-m68k.c (mri_chip): Replace current_chip, not augment.
2824 (md_parse_option): Likewise.
2828 * config/tc-i386.c: Deal with LEX_QM the same way as with LEX_AT.
2829 * config/te-netware.h: New file.
2830 * config/te-ppcnw.h: Delete: Obsolete.
2831 * configure.in: Eliminate ill NetWare targets. Make generic
2832 NetWare target use proper emulation.
2833 * Makefile.am: Eliminate reference to obsolete te-ppcnw.h, add
2834 reference to new te-netware.h.
2835 * configure: Regenerate.
2836 * Makefile.in: Regenerate.
2840 * config/tc-s390.c (s390_insn): Avoid incorrect signed/unsigned
2841 comparison in .insn pseudo operation.
2845 * config/obj-coff.c (coff_adjust_section_syms): Use
2846 bfd_get_section_size instead of bfd_get_section_size_before_reloc.
2847 (coff_frob_section): Likewise.
2848 * config/tc-mips.c (md_apply_fix3): Likewise.
2849 * config/obj-elf.c (elf_frob_file): Use bfd_set_section_size.
2850 (elf_frob_file_after_relocs): Likewise.
2854 * config/tc-hppa.c (log2): Rename to exact_log2.
2855 (pa_next_subseg): Delete unused function.
2856 (create_new_space): Mark unused arguments with ATTRIBUTE_UNUSED.
2857 (create_new_subspace): Likewise.
2860 * config/tc-hppa.c (hppa_fix_adjustable): Allow reduction of fake
2861 labels. Fix warning.
2865 * config/tc-mn10300.h (tc_fix_adjustable): Define.
2866 * config/tc-mn10300.c (mn10300_fix_adjustable): Don't adjust debug
2867 or non-merged symbols.
2871 * config/tc-ia64.c (remove_marked_resource): Save, clear and
2872 restore the old slot when inserting srlz.i/srlz.d.
2876 * Makefile.am: Regenerate dependecies.
2877 * Makefile.in: Regenerate.
2878 * config/tc-sh.c (valid_arch): Make unsigned.
2879 (preset_target_arch): Likewise.
2880 (md_begin): Use new architecture flags system.
2881 (get_specific): Likewise.
2882 (assemble_ppi): Likewise.
2883 (md_assemble): Likewise. Also fix error check for bad opcodes.
2884 (md_parse_option): Likewise. Also generate -isa values according
2885 to the table in bfd/cpu-sh.c instead of just constants. Also
2886 allow <arch>-up ISA variants.
2887 (sh_elf_final_processing): Replace if-else chain with a call to
2888 sh_find_elf_flags().
2889 * testsuite/gas/sh/arch: New directory.
2890 * testsuite/gas/sh/arch/arch.exp: New test script.
2891 * testsuite/gas/sh/arch/arch_expected.txt: New file.
2892 * testsuite/gas/sh/arch/sh.s: New file.
2893 * testsuite/gas/sh/arch/sh2.s: New file.
2894 * testsuite/gas/sh/arch/sh-dsp.s: New file.
2895 * testsuite/gas/sh/arch/sh2e.s: New file.
2896 * testsuite/gas/sh/arch/sh3-nommu.s: New file.
2897 * testsuite/gas/sh/arch/sh3.s: New file.
2898 * testsuite/gas/sh/arch/sh3-dsp.s: New file.
2899 * testsuite/gas/sh/arch/sh3e.s: New file.
2900 * testsuite/gas/sh/arch/sh4-nommu-nofpu.s: New file.
2901 * testsuite/gas/sh/arch/sh4-nofpu.s: New file.
2902 * testsuite/gas/sh/arch/sh4.s: New file.
2903 * testsuite/gas/sh/arch/sh4a-nofpu.s: New file.
2904 * testsuite/gas/sh/arch/sh4al-dsp.s: New file.
2905 * testsuite/gas/sh/arch/sh4a.s: New file.
2909 * config/gc-m68k.c(m68k_ip): Convert mode 5 addressing
2910 with zero offset into mode 2 addressing to save a word.
2914 * config/tc-ia64.c (ar_is_in_integer_unit): Removed.
2915 (ar_is_only_in_integer_unit): New.
2916 (ar_is_only_in_memory_unit): New.
2917 (generate_unwind_image): Silence gcc on 32bit host.
2918 (md_assemble): Use ar_is_only_in_integer_unit instead of
2919 ar_is_in_integer_unit. Check AR access.
2923 * config/tc-m68k.c (md_begin): Sort the opcode table into
2925 (m68k_compare_opcode): New function to do the sorting.
2929 * config/m68k-parse.y(operand): Allow for MAC/EMAC mask
2930 addressing on MIT style operands.
2931 * config/m68k-parse.y(yylex): Allow '-&' for predecrement
2933 * config/tc-m68k.c(install_operand): Comment 'G' and 'H' type
2938 * expr.c (operand, operator): Don't reject '++' and '--'.
2942 * config/tc-mips.c (append_insn): Use ISA-encoded addresses in MIPS16
2947 * configure.in: Add ppc-*-lynxos*. Update i386-*-lynxos* to ELF.
2948 * configure: Regenerate.
2952 * dw2gencfi.c (output_cie): Handle dwarf3 format CIE entries.
2956 * configure.in (or32-*-rtems*): Switch to elf.
2957 * configure: Regenerate.
2961 * po/fr.po: Updated French translation.
2965 * doc/as.texinfo (Section): Document G and T flags to .section
2966 directive. Document the extra arguments that the G flag
2967 requires. Document the #tls flag.
2971 * subsegs.c (section_symbol): Create a new section symbol if
2972 the existing one doesn't match.
2973 * symbols.c (symbol_set_bfdsym): Don't reset BFD section symbol.
2977 * config/tc-mips.c (append_insn, mips_emit_delays): Extend -mfix-vr4120
2978 to cope with VR4181A errata MD(1) and MD(4).
2982 * NEWS: Mention .secrel32 for pe[i]-i386.
2986 * config/tc-frv.h (MAX_MEM_FOR_RS_ALIGN_CODE): New.
2987 (HANDLE_ALIGN): New.
2991 * Makefile.am (DIST_SUBDIRS): Define.
2992 * aclocal.m4: Regenerate with automake 1.8.4.
2993 * Makefile.in: Likewise.
2994 * doc/Makefile.in: Likewise.
2998 * config/tc-ia64.c (dot_serialize): Declare.
2999 (dot_serialize): New function.
3000 (md_pseudo_table): Add ".serialize.data" and
3001 ".serialize.instruction" directives.
3005 * messages (as_internal_value_out_of_range): Print a message about
3006 a value being out of range. Be consistent about whether the
3007 values are printed in decimal or hexadecimal.
3008 (as_warn_value_out_of_range): Generate a warning message about an
3010 (as_bad_value_out_of_range): Generate an error message about an
3012 * as.h: Prototype the new functions.
3013 * config/tc-alpha.c (insert_operand): Use new function.
3014 * config/tc-arc.c (arc_insert_operand): Likewise.
3015 * config/tc-mn10200.c (mn10200_insert_operand): Likewise.
3016 * config/tc-mn10300.c (mn10300_insert_operand): Likewise.
3017 * config/tc-ppc.c (ppc_insert_operand): Likewise.
3018 * config/tc-s390.c (s390_insert_operand): Likewise.
3019 * config/tc-v850.c (v850_insert_operand): Likewise.
3023 * configure.in: Set em=linux for frv-*-*linux*.
3024 * configure: Rebuilt.
3025 * config/tc-frv.h (TARGET_FORMAT): Use elf32-frvfdpic if...
3026 (frv_md_fdpic_enabled): New.
3027 * config/tc-frv.c (frv_md_fdpic_enabled): New.
3028 (DEFAULT_FDPIC): New.
3029 (frv_flags): Use DEFAULT_FDPIC.
3030 (frv_pic_flag): Likewise.
3031 (OPTION_NOPIC): New.
3032 (md_longopts): Add -mnopic.
3033 (md_parse_option): Handle it.
3034 (md_show_usage): Add -mfdpic and -mnopic.
3038 * config/tc-m68k.c: Add find_cf_chip to print list of valid
3039 chips for invalid coldfire instructions, rename selectors
3040 for ColdFire sub-variants, add 521x,5249,547x,548x and aliases,
3041 add current_chip to track which chip is referred to(including save/restore),
3042 use current_chip to select control registers, not current_arch.
3043 (md_show_usage): Add new chips.
3044 * doc/c-m68k.texi: Document new command line switches.
3048 * tc-s390.h (md_do_align, HANDLE_ALIGN): Remove.
3049 (NOP_OPCODE): Define.
3050 (s390_align_code): Remove prototype.
3051 * tc-s390.c (s390_align_code): Remove.
3055 * config/tc-ia64.c (make_unw_section_name): Removed.
3056 (start_unwind_section): New function.
3057 (generate_unwind_image): Take const segT instead of const
3059 (dot_handlerdata): Adjusted.
3060 (dot_endp): Likewise.
3064 * config/obj-elf.c (obj_elf_change_section): Allow the
3065 ".note.GNU-stack" section has SHF_EXECINSTR.
3069 * config/obj-elf.c (get_section): Return bfd_boolean.
3070 (obj_elf_change_section): Call bfd_get_section_by_name_if
3071 instead of bfd_map_over_sections.
3075 * config/obj-elf.c (get_section): New function.
3076 (obj_elf_change_section): Support multiple sections with same
3081 * config/tc-arm.c (create_register_alias): Fix typo checking for
3082 case sensitive register aliases.
3083 (co_proc_number): Use error message string in all_reg_maps[]
3085 (cp_reg_required_here): Likewise.
3086 (fp_reg_required_here): Likewise.
3090 * dwarf2dbg.c (dwarf2_finish): Add SEC_DEBUGGING to section flags.
3094 * config/tc-mips.c (HAVE_32BIT_ADDRESSES, append_insn, macro_build)
3095 (load_address, macro, mips_ip, md_parse_option)
3096 (mips_force_relocation, mips_validate_fix, md_apply_fix3)
3097 (s_change_sec, pic_need_relax, tc_gen_reloc): Remove all
3098 embedded-PIC handling, and update comments.
3099 (SWITCH_TABLE): Remove.
3100 * config/tc-mips.h (DIFF_EXPR_OK): Delete.
3101 (enum mips_pic_level): Remove EMBEDDED_PIC.
3102 (EXTERN_FORCE_RELOC): Remove embedded-PIC handling.
3103 (TC_FORCE_RELOCATION): Update comment.
3104 * ecoff.c (ecoff_build_lineno): Add comment about some code that
3105 might be safe to remove now that MIPS embedded-PIC is gone.
3109 * config/obj-som.c (obj_som_init_stab_section): Add new arguments in
3110 call to obj_set_subsection_attributes.
3111 (obj_som_init_stab_section): Likewise.
3112 * config/tc-hppa.c (default_subspace_dict): Add comdat field.
3113 (pa_def_subspaces): Provide comdat default.
3114 (pa_subspace): Handle new "comdat" parameter. Set SEC_LINK_ONCE and
3115 not SEC_IS_COMMON if section is comdat, common or dup_common. Update
3116 calls to create_new_subspace and update_subspace to pass comdat flag.
3117 (create_new_subspace, update_subspace): Add new comdat argument. Use
3118 it in calls to obj_set_subsection_attributes.
3119 * doc/c-hppa.texi (.subspa, .nsubspa): Document new comdat parameter
3120 and use of comdat, common and dup_comm parameters.
3124 * config/obj-elf.c (obj_elf_change_section): Check if the old
3125 group name is NULL before comparison.
3129 * config/tc-mips.h (mips_dwarf2_addr_size): Prototype.
3133 * config/tc-mips.c (s_mipsset): Set default CPU type for .set mipsN.
3137 * config/tc-mips.c (md_longopts): Remove -membedded-pic option.
3138 (OPTION_MEMBEDDED_PIC): Remove.
3139 (OPTION_TRAP, OPTION_BREAK, OPTION_EB, OPTION_EL)
3140 (OPTION_FP32, OPTION_GP32, OPTION_CONSTRUCT_FLOATS)
3141 (OPTION_NO_CONSTRUCT_FLOATS, OPTIONS_FP64, OPTION_GP64)
3142 (OPTION_RELAX_BRANCH, OPTION_NO_RELAX_BRANCH)
3143 (OPTION_ELF_BASE): Renumber.
3144 (md_parse_option): Remove OPTION_MEMBEDDED_PIC handling.
3145 (md_show_usage): Remove mention of -membedded-pic.
3146 * doc/as.texinfo: Remove mention of -membedded-pic.
3150 * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Remove.
3151 * config/tc-mips.c (RDATA_SECTION_NAME, mips_target_format): Remove
3153 (md_begin, mips_ip, md_parse_option, s_change_sec, s_option,
3154 s_abicalls, nopic_need_relax, tc_gen_reloc): Remove uses of
3155 USE_GLOBAL_POINTER_OPT.
3159 * config/tc-mips.c (macro): One more use of load_delay_nop.
3163 * config/tc-mips.c (load_delay_nop): New function.
3164 (load_address, macro): Use load_delay_nop() to build a nop
3165 which can be omitted with gpr_interlocks.
3169 * config/tc-mips.c (hilo_interlocks, gpr_interlocks,
3170 cop_interlocks): Remove superfluous CPU entries.
3174 * config/tc-arm.c (mav_parse_offset): Value must be multiple of 4.
3178 * NEWS: Added support for EMAC instructions and MAC/EMAC
3180 * config/m68k-parse.h: Add ACC[123], ACCEXT{01,23}, MAC/EMAC
3181 scale factor tokens, trailing_ampersand to mark mask addressing
3182 for MAC/EMAC instructions.
3183 * config/m68k-parse.y: Add options_ampersand clause, '<<',
3185 (yylex): Handle '>', '<', and '&' following '+'.
3186 * config/tc-m68k.c: Set mcfmac/mcfemac on appropriate ColdFire
3187 architectures in archs[].
3188 (m68k-ip): Add '4', 'e', 'g', 'i', cases to handle mask addressing
3189 for MAC/EMAC instructions, ACC[0123], ACCEXT{01,23}, and '<<'/'>>'
3191 (m68k_ip): Handle trailing '&' on MAC/EMAC insns.
3192 (install_operand): Fix 'n' case, Add 'F', 'f', 'G', 'H', 'I', ']'
3194 Add EMAC operands to init_table[].
3198 * config/tc-arm.h (enum mstate): Move here, add MAP_UNDEFINED
3200 (TC_SEGMENT_INFO_TYPE): Define to enum mstate.
3201 * config/tc-arm.c (enum mstate): Delete from here.
3202 (mapping_state): Remove the static mapstate variable and instead
3203 store the state in the segment. This allows a per-section mapping
3204 state. Handle and ignore MAP_UNDEFINED states.
3205 (arm_elf_change_section): Get the current mapping state from the
3207 (s_ltorg): Set the mapping state to MAP_DATA.
3208 (arm_cleanup): Use arm_elf_change_section to get the mapping state
3209 for each pool as it is emitted.
3213 * config/tc-arm.h: Formatting tidy ups.
3217 * NEWS: Note that MIPS -membedded-pic option is deprecated.
3221 * config/tc-i386.h [TE_PE] (TC_CONS_FIX_NEW): Define.
3222 * config/tc-i386.c (md_pseudo_table) [TE_PE]: Add "secrel32".
3223 [TE_PE] (O_secrel): Define.
3224 [TE_PE] (x86_pe_cons_fix_new): New.
3225 [TE_PE] (pe_directive_secrel): Likewise.
3226 (tc_gen_reloc) [TE_PE]: Support BFD_RELOC_32_SECREL.
3230 * config/tc-mips.c (mips_dwarf2_addr_size): Revert part
3231 of previous patch for fix in gcc.
3235 * config/tc-xtensa.c (xg_assembler_literal): Fix a typo.
3239 * read.c (do_align): Call md_flush_pending_output, if defined.
3243 * expr.c (operand): Correct checks for ++ and --.
3247 * config/tc-generic.c: Add some comments.
3251 * doc/c-mips.texi (-m{no-,}fix-vr4120): Renamed from
3252 -{no-}mfix-vr4122-bugs.
3253 * config/tc-mips.c (mips_fix_vr4120): Renamed from mips_fix_4122_bugs.
3254 (append_insn, mips_emit_delays): Update accordingly.
3255 (OPTION_FIX_VR4120, OPTION_NO_FIX_VR4120): Renamed from *VR4122.
3256 (md_longopts): Change -{no-,}mfix-vr4122-bugs to -m{no-,}fix-vr4120.
3257 (md_parse_option): Update after above changes.
3258 (md_show_usage): Add -mfix-vr4120.
3262 * doc/as.texinfo (Sub-Sections): Conditionalize COFF-specific use
3263 of .section directive; add a reference to the ELF .subsection
3268 * config/tc-m32r.c (md_assemble): Fixed infinite loop bug
3273 * Makefile.am: Remove mips from aout targets.
3274 * Makefile.in: Regenerate.
3275 * configure.in: Remove mips-dec-bsd* target.
3276 * configure: Regenerate.
3281 * config/tc-ppc.c (ppc_elf_suffix): Add valid32 and valid64 fields
3282 to struct map_bfd. Adjust MAP macro, and define MAP32, MAP64.
3283 Update "mapping". Restrict some @ modifiers to 32 bit.
3288 * config/tc-sh.c (dont_adjust_reloc_32): New variable.
3289 (sh_fix_adjustable): Avoid adjusting BFD_RELOC_32 when
3290 dont_adjust_reloc_32 is set.
3291 (md_longopts): Add option -renesas.
3292 (md_parse_option, md_show_usage): Likewise.
3293 * doc/c-sh.texi: Likewise.
3297 * config/tc-dlx.c (md_assemble): set fx_no_overflow flag for
3298 hi16 and lo16 fixS structs.
3299 (md_assemble): generate bit_fixS for RELOC_DLX_LO16 in
3300 exactly the same way as for RELOC_DLX_REL16.
3301 (machine_ip): properly respect LO flag in the_insn and
3302 output RELOC_DLX_LO16 rather than RELOC_DLX_16.
3303 (md_apply_fix3): apply RELOC_DLX_LO16.
3307 Remove long-obsolete MPW support.
3308 * mpw-config.in, mpw-make.sed, mac-as.r: Remove files.
3309 * configure.in: Remove mention of ppc-*-mpw* config.
3310 * configure.in: Likewise.
3314 * config/tc-arm.c (meabi_flags): Make its use conditional upon
3315 OBJ_ELF being defined.
3319 * config/obj-aout.c (obj_aout_type): Remove #ifdef BFD_ASSEMBLER code.
3323 * config/tc-arm.c (meabi_flags): New variable.
3324 (arm_parse_eabi): New function.
3325 (md_begin): Set flags for EABI v3.
3327 (arm_long_opts): Add meabi.
3328 * doc/as.texinfo <ARM>: Document -meabi.
3329 * doc/c-arm.texi: Ditto.
3333 * config/tc-xtensa.c (xtensa_post_relax_hook): Create literal
3334 tables even when use_literal_section flag is not set.
3338 * config/tc-sh.c: Remove trailing whitespace.
3342 * doc/c-cris.texi (CRIS-Opts): Document --no-mul-bug-abort,
3343 --mul-bug-abort and the default behavior.
3344 * config/tc-cris.c (cris_insn_kind): New member CRIS_INSN_MUL.
3345 (err_for_dangerous_mul_placement): New variable.
3346 (STATE_MUL, OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): New
3348 (md_cris_relax_table): Have placeholder for STATE_MUL.
3349 (md_longopts): New options --mul-bug-abort and --no-mul-bug-abort.
3350 (cris_relax_frag) <case ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: New
3352 (md_estimate_size_before_relax) <case ENCODE_RELAX (STATE_MUL,
3353 STATE_BYTE)>: Ditto.
3354 (md_convert_frag) <ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: Check
3355 alignment and position of this frag, emit error message if
3357 (md_assemble): For a multiply insn and when checking it,
3358 transform the current frag into a special frag for that purpose.
3359 (md_parse_option) <case OPTION_MULBUG_ABORT_OFF, case
3360 OPTION_MULBUG_ABORT_ON>: Handle new options.
3364 * config/tc-xtensa.c (mark_literal_frags): New function.
3365 (xtensa_move_literals): Call mark_literal_frags for all literal
3366 segments, including init and fini literal segments.
3367 (xtensa_post_relax_hook): Swap use of xt_insn_sec and xt_literal_sec.
3371 * tc-hppa.c (cons_fix_new_hppa): Check for PC relative base type.
3372 (pa_comm): Set BSF_OBJECT in symbol flags.
3376 * Makefile.am: Run "make dep-am".
3377 * Makefile.in: Regenerate.
3378 * doc/Makefile.in: Regenerate.
3379 * config.in: Regenerate.
3380 * po/gas.pot: Regenerate.
3384 * read.c (read_a_source_file): Use demand_empty_rest_of_line.
3385 (demand_empty_rest_of_line): Issue an error here.
3386 (ignore_rest_of_line): Silently skip to end.
3387 (demand_copy_string): Issue an error, not warning.
3389 * config/obj-elf.c (obj_elf_section_name): Likewise.
3390 (obj_elf_section): Likewise.
3391 * config/tc-arc.c (arc_extoper): Remove bogus NULL checks.
3392 (arc_extinst): Likewise.
3393 * config/tc-ia64.c (dot_saveb): Use demand_empty_rest_of_line.
3394 (dot_spill): Likewise.
3395 (dot_unwabi): Likewise.
3396 (dot_prologue): Likewise.
3398 * expr.c (operand): Reject ++ and --.
3399 (operator): Likewise.
3403 * config/tc-sh.c: Include dw2gencfi.h.
3404 (sh_cfi_frame_initial_instructions): New function.
3405 (sh_regname_to_dw2regnum): Likewise.
3406 * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Move to the end of
3408 (TARGET_USE_CFIPOP): Define.
3409 (tc_cfi_frame_initial_instructions): Likewise.
3410 (tc_regname_to_dw2regnum): Likewise.
3411 (DWARF2_DEFAULT_RETURN_COLUMN, DWARF2_CIE_DATA_ALIGNMENT): Likewise.
3412 * Makefile.am: Update dependencies.
3413 * Makefile.in: Regenerate.
3417 * configure.in: Switch sh-*-rtems* to ELF. Add sh-*-rtemscoff*.
3418 * configure: Regenerate.
3422 * read.c (s_leb128): Call md_flush_pending_output.
3426 * config/tc-i386.c (output_insn): Handle PadLock instructions.
3427 * config/tc-i386.h (CpuPadLock): New define.
3428 (CpuUnknownFlags): Added CpuPadLock.
3432 * doc/c-hppa.texi (HPPA Directives): Fix typo.
3436 * dw2gencfi.c (output_cie): Align length to 4 byte boundary.
3437 (cfi_finish): Likewise for fde.
3441 * config/tc-ia64.c (md_assemble): Properly handle NULL
3443 (ia64_handle_align): Don't abort if failed to add a stop bit.
3447 * Makefile.in: Regenerated.
3448 * aclocal.m4: Likewise.
3449 * configure: Likewise.
3450 * doc/Makefile.in: Likewise.
3454 * config/tc-ia64.c (dot_align): New.
3455 (ia64_do_align): Make it static.
3456 (md_pseudo_table): Use "dot_align" for "align".
3457 (ia64_md_do_align): Don't set align_frag here.
3458 (ia64_handle_align): Add a stop bit to the previous bundle if
3461 * config/tc-ia64.h (ia64_do_align): Removed.
3465 * config/tc-sh.c (md_parse_option): Add -isa=sh4-nofpu and
3466 -isa=sh4-nommu-nofpu options. Adjust help messages accordingly.
3467 (sh_elf_final_processing): Output BFD type sh4_nofpu if that is
3468 the most general type or the user specifically requested it.
3469 (md_assemble): Add a new error message for when an instruction
3470 is understood, but is not allowed due to an -isa option.
3474 * config/tc-ia64.c (align_frag): New.
3475 (md_assemble): Set the tc_frag_data field in align_frag for
3476 IA64_OPCODE_FIRST instructions.
3477 (ia64_md_do_align): Set align_frag.
3478 (ia64_handle_align): Add a stop bit if needed.
3480 * config/tc-ia64.h (TC_FRAG_TYPE): New.
3481 (TC_FRAG_INIT): New.
3485 * config/tc-frv.c (fr400_audio): New variable.
3486 (md_parse_option, md_show_usage): Add -mcpu=fr405 and -mcpu=fr450.
3487 (md_parse_option): Set fr400_audio for -mcpu=fr400 and -mcpu=fr405.
3488 (target_implements_insn_p): New function.
3489 (md_assemble): Report an error if the processor doesn't implement
3494 * config/tc-m32r.c (md_longopts): Added -no-bitinst option.
3495 (md_parse_option): Ditto.
3496 (OPTION_NO_SPECIAL_M32R): Added.
3497 (md_show_usage): Document it.
3498 (enable_speial_m32r): Changed a default value from 0 to 1.
3499 * doc/c-m32r.texi: Document the -no-bitinst option.
3503 * config/tc-sh.c (get_operand): Revert previous delta.
3504 (tc_gen_reloc): Check for an unknown reloc type before processing
3509 * config/tc-s390.c (s390_insn): Correct range check for opcode in
3510 .insn pseudo operation.
3514 * config/tc-sh.c (get_operand): In case of #Imm, check has been
3515 added for wrong syntax.
3519 * config/tc-mips.c (mips_dwarf2_addr_size): New.
3520 * config/tc-mips.h (DWARF2_ADDR_SIZE): Use.
3524 * config/tc-sh.c (build_Mytes): Add REG_N_D and REG_N_B01
3525 nibble types to assembler.
3529 * config/tc-iq2000.c: Add missing \n\ in multiline string literal.
3533 * config/tc-ia64.c (slot_index): New arg before_relax. Use instead of
3535 (fixup_unw_records): New arg before_relax. Pass to slot_index.
3536 (ia64_estimate_size_before_relax): New.
3537 (ia64_convert_frag): Pass 0 to fixup_unw_records. Add comment.
3538 (generate_unwind_image): Pass 1 to fixup_unw_records.
3539 * config/tc-ia64.h (ia64_estimate_size_before_relax): Declare.
3540 (md_estimate_size_before_relax): Call ia64_estimate_size_before_relax.
3544 * stabs.c (generate_asm_file): Avoid warning about use of
3545 uninitialized variable.
3549 * config/tc-ia64.c (ia64_flush_insns): In addition to prologue,
3550 body, and endp, allow unwind records which do not have a "t"
3551 (time/instruction) field.
3555 * config/tc-arm.c (do_mav_dspsc_1): Correct offset of CRn.
3556 (do_mav_dspsc_2): Likewise.
3557 Fix accumulator registers move opcodes.
3562 * dwarf2dbg.c (get_filenum): Do not read beyond allocated memory.
3566 * config/tc-ia64.h (ia64_frob_symbol): New declaration.
3567 (tc_frob_symbol): New macro definition.
3568 * config/tc-ia64.c (ia64_frob_symbol): New routine.
3572 * config/tc-arm.c (md_begin): Mark .note.gnu.arm.ident as
3577 * read.h (IGNORE_OPCODE_CASE): Do not define. Replace with ...
3578 (TC_CASE_SENSITIVE): ... this.
3579 * read.c: Replace IGNORE_OPCODE_CASE with TC_CASE_SENSITIVE.
3580 * doc/internals.texi (TC_CASE_SENSITIVE): Document.
3584 * config/tc-ia64.c (dot_endp): Delete call to output_endp.
3585 (generate_unwind_image): Re-add it here.
3589 * dwarf2dbg.c (DWARF2_ADDR_SIZE): Remove trailing ';'
3590 * read.h (SKIP_WHITESPACE): Turn into an expression.
3591 * read.c (read_a_source_file): A pseudo is removed by having a
3596 * config/tc-ia64.c (output_endp): New.
3597 (count_bits): Delete.
3598 (ia64_flush_insns, process_one_record, optimize_unw_records): Handle
3599 endp unwind records.
3600 (fixup_unw_records): Handle endp unwind records. Delete code for
3601 shortening prologue regions not followed by a body record.
3602 (dot_endp): Call add_unwind_entry to emit endp unwind record.
3603 * config/tc-ia64.h (unw_record_type): Add endp.
3607 * config/tc-ia64.c (ia64_convert_frag): Call md_number_to_chars to
3608 fill padding bytes with zeroes.
3609 (emit_one_bundle): New locals last_ptr, end_ptr. Rewrite code that
3610 sets unwind_record slot_number and slot_frag fields.
3614 * config/tc-mips.c (add_got_offset_hilo): New function.
3615 (macro): Use load_register() and add_got_offset_hilo() to load
3616 constants instead of hardcoding code sequences throughout.
3620 * config/tc-ia64.c (emit_one_bundle): Add proper indentation.
3624 * config/tc-m68k.h (EXTERN_FORCE_RELOC): Handle m68k-uclinux specially,
3626 * config/tc-m68k.c (RELAXABLE_SYMBOL): Use EXTERN_FORCE_RELOC instead
3627 of hard-coded test for TARGET_OS=elf.
3631 * config/tc-mips.c (hilo_interlocks): Change definition
3632 so that MIPS32, MIPS64 and later ISAs are included, along with
3633 the already-included machines. Update comments.
3637 * config/tc-arm.c (tc_gen_reloc): Improve error message for
3638 undefined local labels.
3642 * config/tc-mips.c (load_address, macro): Update comments about
3643 NewABI GP relaxation.
3647 * config/tc-mips.c (macro_build): Remove place and counter arguments.
3648 (mips_build_lui, macro_build_ldst_constoffset): Likewise.
3649 (mips16_macro_build, macro_build_jalr): Remove counter argument.
3650 (set_at, load_register, load_address, move_register): Likewise.
3651 (load_got_offset, add_got_offset): Likewise.
3652 Update all calls and tidy accordingly.
3656 * config/tc-mips.c (RELAX_ENCODE): Remove WARN argument.
3657 (RELAX_FIRST, RELAX_SECOND): Turn into 8-bit quantities.
3658 (RELAX_USE_SECOND): Bump to 0x10000.
3659 (RELAX_SECOND_LONGER, RELAX_NOMACRO, RELAX_DELAY_SLOT): New flags.
3660 (mips_macro_warning): New variable.
3661 (md_assemble): Wrap macro expansion in macro_start() and macro_end().
3662 (s_cpload, s_cpsetup, s_cprestore, s_cpreturn): Likewise.
3663 (relax_close_frag): Set mips_macro_warning.first_frag. Adjust use
3665 (append_insn): Update mips_macro_warning.sizes.
3666 (macro_start, macro_warning, macro_end): New functions.
3667 (macro_build): Don't emit warnings here.
3668 (macro_build_lui, md_estimate_size_before_relax): ...or here.
3669 (md_convert_frag): Check for cases where one macro alternative
3670 needs a warning and the other doesn't. Emit a warning if the
3671 longer sequence was chosen.
3675 * config/tc-mips.h (tc_frag_data_type, TC_FRAG_TYPE): Remove.
3676 * config/tc-mips.c (RELAX_ENCODE): Take three arguments: the size of
3677 the first sequence, the size of the second sequence, and a flag
3678 that says whether we should warn.
3679 (RELAX_OLD, RELAX_NEW, RELAX_RELOC[123]): Delete.
3680 (RELAX_FIRST, RELAX_SECOND): New.
3681 (mips_relax): New variable.
3682 (relax_close_frag, relax_start, relax_switch, relax_end): New fns.
3683 (append_insn): Remove "place" argument. Use mips_relax.sequence
3684 rather than "place" to check whether we're expanding the second
3685 alternative of a relaxable macro. Remove redundant check for
3686 branch relaxation. If generating a normal insn, and there
3687 is not enough room in the current frag, call relax_close_frag()
3688 to close it. Update mips_relax.sizes[]. Emit fixups for the
3689 second version of a relaxable macro. Record the first relaxable
3690 fixup in mips_relax. Remove tc_gen_reloc workaround.
3691 (macro_build): Remove all uses of "place". Use mips_relax.sequence
3692 in the same way as in append_insn.
3693 (mips16_macro_build): Remove "place" argument.
3694 (macro_build_lui): As for macro_build. Don't drop the add_symbol
3695 when generating the second version of a relaxable macro.
3696 (load_got_offset, add_got_offset): New functions.
3697 (load_address, macro): Use new relaxation machinery. Remove
3698 tc_gen_reloc workarounds.
3699 (md_estimate_size_before_relax): Set RELAX_USE_SECOND if the second
3700 version of a relaxable macro is needed. Return -RELAX_SECOND if the
3701 first version is needed.
3702 (tc_gen_reloc): Remove relaxation handling.
3703 (md_convert_frag): Go through the fixups for a relaxable macro and
3704 mark those that belong to the unneeded alternative as done. If the
3705 second alternative is needed, adjust the fixup addresses to account
3706 for the deleted first alternative.
3710 * frags.h (frag_room): Declare.
3711 * frags.c (frag_room): New function.
3712 * doc/internals.texi: Document it.
3716 * config/tc-mips.c (append_insn): Don't do r3900 interlock
3717 optimization for -mtune=r3900, as this will break on other CPUs.
3721 * config/tc-m68hc11.c (build_indexed_byte): movb and movw cannot
3722 be relaxed, use fixup.
3723 (md_apply_fix3): Use 5 bit reloc from movb and movw fixup.
3727 * config/tc-sparc.c (sparc_ip): Disallow %f32-%f63 for single
3732 * config/tc-mips.c (append_insn): Properly detect variant frags
3733 that preclude swapping of relaxed branches. Correctly swap
3734 instructions between frags when dealing with relaxed branches.
3738 * acinclude.m4: Quote names of macros to be defined by AC_DEFUN
3740 * aclocal.m4: Regenerate.
3741 * configure: Regenerate.
3745 * config/tc-h8300.c (build_bytes): Apply relaxation to bit
3750 * config/tc-mips.c (macro_build_jalr): When adding an R_MIPS_JALR
3751 reloc, reserve space for the delay slot as well as the jalr itself.
3755 * config/tc-arm.c (do_vfp_reg2_from_sp2): Rename from do_vfp_sp_reg2.
3756 (do_vfp_sp2_from_reg2): New function.
3758 (do_vfp_dp_from_reg2): Check return values properly.
3762 * config/tc-mips.c (warn_nops): Remove static variable.
3763 (macro): Remove test of warn_nops.
3764 (md_shortops): Remove 'n'.
3765 (md_parse_option): Remove 'n' case.
3766 (md_show_usage): Remove -n.
3767 * doc/as.texinfo (Overview): Remove MIPS -n option.
3768 * doc/c-mips.texi (MIPS Opts): Remove mention -n.
3769 * NEWS: Mention removal of MIPS -n option.
3771 * config/tc-mips.c (ISA_HAS_COPROC_DELAYS): Remove.
3772 (cop_interlocks): Check ISA level.
3773 (cop_mem_interlocks): Define.
3774 (reg_needs_delay): Check cop_interlocks rather than
3775 ISA_HAS_COPROC_DELAYS.
3776 (append_insn): Likewise. Use cop_mem_interlocks rather than
3777 directly checking mips_opts.isa.
3778 (mips_emit_delays): Likewise.
3782 * config/tc-ia64.c (unwind): Move next_slot_number and
3783 next_slot_frag to ...
3784 (unw_rec_list): Here.
3785 (free_list_records): Removed.
3786 (output_unw_records): Likewise.
3787 (generate_unwind_image): Make it void.
3788 (alloc_record): Initialize next_slot_number and next_slot_frag.
3789 (slot_index): Take .org, .space and .align into account.
3790 (fixup_unw_records): Don't set slot_number to 0. Use
3791 list->next_slot_number and list->next_slot_frag instead of
3792 unwind.next_slot_number and unwind.next_slot_frag.
3793 (ia64_convert_frag): New.
3794 (generate_unwind_image): Generate a rs_machine_dependent frag
3796 (emit_one_bundle): Use list->next_slot_number and
3797 list->next_slot_frag instead of unwind.next_slot_number and
3798 unwind.next_slot_frag.
3800 * config/tc-ia64.h (md_convert_frag): Defined as
3802 (md_estimate_size_before_relax): Defined as (f)->fr_var.
3807 * config/tc-frv.h (md_apply_fix3): Don't define.
3808 * config/tc-frv.c (md_apply_fix3): New. Shift/truncate %hi/%lo
3810 * config/tc-frv.h (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
3812 * config/tc-frv.c (line_separator_chars): Add `!'.
3814 * config/tc-frv.c (md_assemble): Clear insn upfront.
3816 * config/tc-frv.c (OPTION_FDPIC): New macro.
3817 (md_longopts): Add mfdpic.
3818 (md_parse_option): Handle it.
3820 * config/tc-frv.c (md_cgen_lookup_reloc) <FRV_OPERAND_D12,
3821 FRV_OPERAND_S12>: Use reloc type encoded in fix-up.
3822 (frv_pic_ptr): Parse funcdesc.
3826 * doc/as.texinfo: Let texi2pod parse asconfig.texi and
3827 gasver.texi. Remove duplicate symbol definitions for texi2pod.
3831 * Makefile.am (Makefile): Move the dependency on
3832 $(BFDDIR)/configure.in to...
3833 (CONFIG_STATUS_DEPENDENCIES): ... here.
3834 (AUTOMAKE_OPTIONS): Require automake 1.8.
3835 * Makefile.in: Regenerate.
3836 * doc/Makefile.am (BASEDIR, BFDDIR): Define.
3837 (CONFIG_STATUS_DEPENDENCIES): Add a dependency on
3838 $(BFDDIR)/configure.in here as well.
3839 * doc/Makefile.in: Regenerate.
3843 * Makefile.am (install, install-info, RECURSIVE_TARGETS): Remove.
3844 * Makefile.in: Regenerate.
3845 * aclocal.m4: Regenerate.
3846 * doc/Makefile.am (install, install-info): Remove.
3847 (install-data-local): A new hook for install-info.
3848 (AUTOMAKE_OPTIONS): Require automake 1.8.
3849 * doc/Makefile.in: Regenerate.
3853 * doc/c-sh.texi: Update description about floating point behavior
3858 * configure.in: Add m68k-uClinux target.
3859 * configure: Regenerate.
3861 For older changes see ChangeLog-0203
3867 version-control: never