3 * dw2gencfi.c (output_cie, output_fde): Use DW_CFA_nop rather
9 * config/tc-mips.c (reloc_needs_lo_p): Handle
10 BFD_RELOC_MIPS16_HI16_S.
11 (fixup_has_matching_lo_p): Handle BFD_RELOC_MIPS16_LO16.
12 (append_insn): Add BFD_RELOC_MIPS16_GPREL, BFD_RELOC_MIPS16_HI16_S
13 and BFD_RELOC_MIPS16_LO16 to relocs to suppress overflow
15 (mips16_ip): Resolve BFD_RELOC_MIPS16_HI16_S,
16 BFD_RELOC_MIPS16_HI16 and BFD_RELOC_MIPS16_LO16 for constants.
17 Call my_getSmallExpression() to parse percent operators.
18 (percent_op_match, mips_percent_op): Separate definitions.
19 (mips16_percent_op): Define percent operators for the MIPS16 mode.
20 (parse_relocation): Handle the MIPS16 mode using
22 (md_apply_fix3): Handle BFD_RELOC_MIPS16_HI16,
23 BFD_RELOC_MIPS16_HI16_S and BFD_RELOC_MIPS16_LO16.
27 * config/tc-ia64.c (md_apply_fix3): Call ia64_gen_real_reloc_type
28 instead of explicitly dealing with the translation; exclude
29 relocations that are already pcrel, however.
33 * config/tc-ia64.c: Include limits.h (if available).
34 (gr_values[0]): Set path to INT_MAX.
35 (dot_reg_val): Don't allow changing value of r0. Limit range of
36 general registers at r127.
37 (specify_resource): Default resource index is -1. Don't set resource
38 index (in case IA64_RS_RSE) without setting the specific flag.
39 (note_register_values): Check operand is O_constant before tracking
40 input value of moves. Add tracking for dep.z with constant inputs.
41 (print_dependency): Resource index of specific resource may be zero.
42 (check_dependencies): Likewise.
46 * config/tc-ia64.c (parse_operands): New local variables reg1, reg2,
47 reg_class. Check operands and emit diagnostics for illegal use of
52 * config/tc-ia64.c (ia64_gen_real_reloc_type): Define and initialize
53 new variables type, suffix, and width. Handle
54 BFD_RELOC_IA64_DIR(32|64)[LM]SB in FUNC_LT_FPTR_RELATIVE case.
55 Handle BFD_RELOC_IA64_DIR64[LM]SB in FUNC_TP_RELATIVE case. Add
56 FUNC_DTP_MODULE case. Handle BFD_RELOC_IA64_DIR32[LM]SB in
57 FUNC_DTP_RELATIVE case. Return incoming relocation type if
58 BFD_RELOC_IA64_IPLT[LM]SB in FUNC_IPLT_RELOC case. Generate warning
59 if unable to translate relocation type, using the new variables.
63 * config/tc-ia64.h (ia64_symbol_chars): Declare.
64 (ty_symbol_chars): Define.
65 * config/tc-ia64.c (ia64_symbol_chars): Define.
69 * config/tc-ia64.c (ia64_parse_name): Only update next character if
70 input_line_pointer was advanced.
74 * config/tc-sh.c (md_apply_fix3): Add parentheses around &
79 * config/tc-ia64.c (md_parse_option): Handle -xnone and -xdebugn.
80 (md_show_usage): Add -xnone, -xdebugn, and -xdebugx. Relocate default
82 (ia64_init): Set md.detect_dv.
83 (ia64_start_line): New static variable warned. Warn only once when
84 encountering explicit stops in automatic mode.
85 * doc/c-ia64.texi: Describe -xnone, -xdebugn, and -xdebugx.
86 * NEWS: Mention new default mode.
90 * config/tc-ia64.c (dot_rot): Add comment that name strings should
91 be freed when wiping out previous state. Canonicalize names before
92 use. Free name string when detecting redefinition.
93 (dot_pred_rel): Call generic expression parser to process arguments.
94 Handle O_register case for individual predicates and O_subtract for
96 (ia64_parse_name): Canonicalize name before looking it up in dynamic
98 (ia64_canonicalize_symbol_name): Strip off all trailing # characters.
99 Warn if multiple found, issue error if resulting symbol name has zero
101 (dot_alias): Canonicalize name before use.
105 * config/tc-ia64.c (unwind_diagnostic): Return -1 for warning
107 (in_procedure): Return -1 for warning.
108 (in_prologue): Likewise.
113 (dot_xdata): Undo the last change. Section name is used by
115 (dot_float_cons): Likewise.
116 (dot_xstringer): Likewise.
117 (dot_xdata_ua): Likewise.
118 (dot_float_cons_ua): Likewise.
122 * NEWS: Mention "-munwind-check=[warning|error]".
124 * config/tc-ia64.c (md): Add unwind_check.
125 (unwind_diagnostic): New.
126 (in_procedure): Call unwind_diagnostic when a directive isn't
128 (in_prologue): Call unwind_diagnostic when a directive isn't in
130 (in_body): Call unwind_diagnostic when a directive isn't in
132 (dot_endp): Set md.unwind_check to error before calling
133 in_procedure and restore it after. When the name is missing or
134 couldn't be found, use the one from the last .proc if
135 md.unwind_check isn't error. Warn if md.unwind_check is
137 (md_parse_option): Handle "-munwind-check=[warning|error]".
138 (md_show_usage): Add "-munwind-check=[warning|error]".
139 (ia64_init): Set md.unwind_check to warning.
141 * doc/as.texinfo: Add "-munwind-check=[none|warning|error]".
142 * doc/c-ia64.texi: Likewise.
146 * config/tc-ia64.h (LEX_AT): Include LEX_BEGIN_NAME.
148 (ia64_parse_name): New third parameter.
149 (md_parse_name): Pass third argument.
150 * config/tc-ia64.c (pseudo_func): Placeholders use NULL as name.
151 (md_operand): Handling of '@'-prefixed symbols moved from here...
152 (ia64_parse_name): ...to here.
156 * config/tc-ia64.c (md): Remove last_groups and group_idx.
157 (errata_nop_necessary_p): Remove declaraction and definition.
158 (emit_one_bundle): Don't call errata_nop_necessary_p. Don't
159 update md.group_idx. Don't reset md.last_groups.
163 * config/tc-ia64.c (parse_section_name): Handle non-quoted first
165 (dot_xdata): Free section name after use.
166 (dot_float_cons): Likewise.
167 (dot_xstringer): Likewise.
168 (dot_xdata_ua): Likewise.
169 (dot_float_cons_ua): Likewise.
170 (md_pseudo_table): Add xdata16 and xdata16.ua.
174 * doc/all.texi: Add IA64.
175 * doc/as.texinfo: Likewise.
177 * doc/c-ia64.texi: Fix typos.
181 * config/tc-arm.c (do_t_ldmstm): Change BFD_RELOC_NONE to
183 (do_t_push_pop): Likewise.
184 (md_assemble): Likewise.
185 (md_apply_fix3): Handle BFD_RELOC_NONE correctly, make
186 BFD_RELOC_UNUSED same as previous meaning of BFD_RELOC_NONE.
187 (create_unwind_entry): Output dependency on the required personality
189 testsuite/gas/arm/unwind.d: Alter expected output to include
190 dependency on __aeabi_unwind_cpp_pr[01].
194 * as.h (seg_not_empty_p): Return int, not bfd_boolean.
195 * subsegs.c (seg_not_empty_p): Likewise.
199 * config/tc-maxq.c (md_estimate_size_before_relax): Correct the
200 relative jump calculation.
201 <md_convert_frag) : Likewise.
202 <output_disp): Likewise.
206 * write.c (write_object_file): Recognize warning-symbol construct
207 and skip object- and target- handling for the second symbol.
211 * config/tc-ia64.c (dot_pred_rel): Update comment. Handle @-prefixed
212 designators along with quoted ones. Free copy of quoted designator
217 * config/atof-ieee.c, config/obj-coff.c, config/obj-elf.c,
218 config/obj-ieee.c, config/obj-som.c, config/obj-vms.c,
219 config/tc-a29k.c, config/tc-alpha.c, config/tc-arc.c,
220 config/tc-arm.c, config/tc-d30v.c, config/tc-dlx.c,
221 config/tc-fr30.c, config/tc-h8300.c, config/tc-h8500.c,
222 config/tc-i370.c, config/tc-i386.c, config/tc-i960.c,
223 config/tc-ia64.c, config/tc-m32r.c, config/tc-m32r.h,
224 config/tc-m68hc11.c, config/tc-m68hc11.h, config/tc-mips.c,
225 config/tc-mn10200.c, config/tc-msp430.c, config/tc-ns32k.c,
226 config/tc-openrisc.c, config/tc-or32.c, config/tc-pdp11.c,
227 config/tc-pj.c, config/tc-sparc.h, config/tc-tic54x.c,
228 config/tc-tic80.c, config/tc-v850.c, config/tc-w65.c,
229 config/tc-xtensa.c, config/tc-z8k.c, config/xtensa-relax.c: Remove
230 #if 0'd code throughout.
234 * as.c (parse_args): Bump copyright date reported by --version to
239 * configure.tgt: Remove obsolete targets m68k-lynxos, sparc-lynxos
244 * macro.c (buffer_and_nest): Allow 'from' being NULL; handle anything
245 that can end with .endr in that case. Make requiring/permitting
246 pseudo-ops without leading dot closer to the logic in read.c serving
248 (expand_irp): Don't pass a mnemonic to buffer_and_nest as it will be
253 * macro.c (do_formals): Adjust to no longer accept empty parameter
255 (define_macro): Adjust to no longer accept empty macro name, garbage
256 following the parameters, or macros that were previously defined.
257 * read.c (s_bad_end): Declare.
258 (potable): Add endm. Handler for endr and endm is s_bad_end.
259 (s_bad_end): Rename from s_bad_endr. Adjust to handle both .endm
261 * read.h (s_bad_endr): Remove.
265 * config/tc-ia64.c (parse_operands): Parse all specified operands,
266 immediately discarding (but counting) those exceeding the maximum
267 possible amount. Track whether output and input operand counts ever
268 matched, and use this to better indicate which of the operands/
269 operand types was wrong; specifically don't default to pointing to
274 * config/tc-ia64.c (unwind): Remove proc_end (now an automatic
275 variable in dot_endp). Add body and insn. Make prologue,
276 prologue_mask, body, and insn bitfields.
277 (fixup_unw_records): Remove spurious new-lines from end of diagnostic
279 (in_procedure, in_prologue, in_body): New.
280 (dot_fframe, dot_vframe, dot_vframesp, dot_vframepsp, dot_save,
281 dot_restore, dot_restorereg, dot_restorereg_p, dot_handlerdata,
282 dot_unwentry, dot_altrp, dot_savemem, dot_saveg, dot_savef, dot_saveb,
283 dot_savegf, dot_spill, dot_spillreg, dot_spillmem, dot_spillreg_p,
284 dot_spillmem_p, dot_label_state, dot_copy_state, dot_unwabi,
285 dot_personality): Use the appropriate one of the above.
286 (dot_proc): Clear unwind.proc_start; set to current location only if
287 none of the entry points were valid. Check for non-zero-length entry
288 point names. Check that entry points aren't defined, yet. Clear
289 unwind.prologue, unwind.body, and unwind.insn.
290 (dot_body): Call in_procedure. Check that first directive in procedure
291 had no insns emitted before. Set unwind.body.
292 (dot_prologue): Call in_procedure. Check that not already in prologue.
293 Check that first directive in procedure had no insns emitted before.
295 (dot_endp): Call in_procedure. Declare proc_end. Check for non-zero-
296 length entry point names. Check that entry points became defined.
297 (md_assemble): Set unwind.insn once unwind.proc_start is defined.
301 * config/tc-ia64.c (emit_one_bundle): Snapshot manual bundling state
302 before actually using it. Don't generate an error in manual bundling
303 mode when looking at an insn requiring slot 2 but not yet at slot 2.
304 Don't generate an error in manual bundling mode when looking at an
305 insn required to be last in its group but the required slot hasn't
306 been reached, yet. Allow conversion from MII to MI;I for bundle
307 consisting of only 2 insns with the stop between them. Suppress
308 various meaningless errors resulting from detecting earlier ones.
312 * config/tc-ia64.c (parse_operands): Also handle alloc without first
317 * config/tc-z8k.c (md_assemble): Improve error detection.
321 * config/tc-ia64.c (ia64_estimate_size_before_relax): Allocate space
322 for personality routine pointer only if there is one.
323 (ia64_convert_frag): Likewise.
324 (generate_unwind_image): Likewise.
328 * config/tc-z8k.c (INSERT): Remove, not used anywhere.
329 (md_apply_fix3): Make relative branches out of range an error
330 instead of a warning. Display correct line number for out of
331 range branches/calls/memory accesses.
335 * dwarf2dbg.c (dwarf2_finish): Correct logic for determining when
336 to emit .debug_line and other debug sections.
337 * as.h (seg_not_empty_p): Declare.
338 * subsegs.c (seg_not_empty_p): New predicate.
342 * configure: Regenerate to track ../gettext.m4 change.
346 * config/tc-ia64.c (emit_one_bundle): Change "?imbf??" to "?ibmfxx".
350 * config/tc-ia64.c (emit_one_bundle): Add late resolution of move
351 to/from application registers dynamic insns.
352 (md_assemble): Defer resolution of move to/from application registers
353 dynamic insns when they can be issued on either the I- or M-units.
357 * config/tc-frv.c (md_apply_fix3): Mark TLS symbols as such.
359 * config/tc-frv.c (frv_pic_ptr): Add tlsmoff support.
361 * cgen.c (gas_cgen_parse_operand): Handle
362 CGEN_PARSE_OPERAND_SYMBOLIC.
363 * config/tc-frv.c (md_cgen_lookup_reloc): Handle TLS relocations.
364 (frv_force_relocation): Likewise. Fix handling of PIC
366 (md_apply_fix3): Likewise.
370 * as.h: Remove #if 0'd code.
371 * atof-generic.c (atof_generic): Likewise.
372 * ecoff.c (ecoff_directive_frame): Likewise.
373 * frags.h (FRAG_APPEND_1_CHAR): Likewise.
374 * itbl-ops.c (itbl_add_reg): Likewise.
375 * listing.c (calc_hex): Likewise.
376 * read.c (MASK_CHAR): Likewise.
377 * subsegs.c (subsegs_print_statistics): Likewise.
378 * symbols.c (indent): Likewise.
379 * write.c (write_relocs): Likewise.
380 (write_object_file): Likewise.
381 (relax_frag): Likewise.
385 * as.c (std_longopts): Add an entry for "--a" in order to prevent
386 getopt_long_only() from considering -a as an abbreviation for
388 (parse_args): Fix the parsing of -a=<file>.
393 * read.c (s_incbin): Adjust default count for skip. Check validity
394 of count and skip rigorously.
398 * config/tc-mips.c (dummy_opcode): Add init for new struct member.
402 * read.c (convert_to_bignum): New function, split out from...
403 (emit_expr): ...here. Handle the case where X_add_number is
404 positive and the input value is negative.
405 (output_big_sleb128): Fix setting of continuation bit. Check whether
406 the final byte needs to be sign-extended. Fix size-shrinking loop.
407 (emit_leb128_expr): When generating a signed leb128, see whether the
408 sign of an O_constant's X_add_number matches the sign of the input
409 value. Use a bignum if not.
413 * tc-sh.c (md_begin,md_parse_option): Change arch_sh1_up to
414 arch_sh_up in order to match the external name and make the
415 testsuite's job easier.
420 * config/tc-i386.c (i386_scale): Disallow 0 scale.
424 * config/tc-iq2000.c (s_iq2000_set): Fix thinko parsing
425 ignored_arguments array.
429 * write.c (write_object_file): Disallow a symbol equated to
434 * tc-maxq.c: Replace constants 10 and 20 with bfd_mach_maxq10 and
436 (md_pseudo_table): Add new pseudo ops for maxq10 and maxq20.
437 (maxq_target): New function: Set the machine type.
441 * config/tc-arm.c (FPU_DEFAULT): Define for TE_VXWORKS.
442 (md_begin): Handle TE_VXWORKS for FP defaults.
443 (md_apply_fix3): Correct rela offsets.
444 (elf32_arm_target_format): Add VxWorks targets.
448 * configure.tgt: Set em=vxworks for *-*-vxworks.
449 * config/te-vxworks.h: New File.
453 * config/tc-arm.c (arm_cpus): Correct arch field for arm1026ej-s.
457 * config/tc-msp430.c (md_apply_fix3): Fix offset calculation for
462 * config/tc-ia64.c (md): Add member "loc_directive_seen".
463 (dot_loc): New function.
464 (md_pseudo_table): Add entry to map .loc to dot_loc().
465 (emit_one_bundle): Only call dwarf2_gen_line_info() if we have
466 seen a .loc directive or we're generating DWARF2 debug info for
472 * read.c (s_comm_internal): Don't zero end of name until size
473 expression has been parsed.
477 * config/tc-avr.c (mcu_types): Move attiny{13,2313} from avr4 to avr2.
481 * config/tc-crx.c: Support 'bcop' relaxation (dealt as in 'cmp&branch'
486 * configure.tgt: New.
487 * configure.in: Move setting of cpu_type, fmt, etc., to
489 * Makefile.am (CONFIG_STATUS_DEPENDENCIES): Add
490 $(srcdir)/configure.tgt.
491 * configure, Makefile.in: Rebuild.
495 * config/tc-avr.c: Add support for the new R_AVR_LDI, R_AVR_6 and
496 R_AVR_6_ADIW relocs for the LDI, ADIW/SBIW and LDD/STD
498 (avr_offset_expression): New function to parse offsets for LDI
500 (avr_operand): Use it.
501 (md_apply_fix3): Generate the relocs.
505 * config/tc-sh64.c (shmedia_md_apply_fix3): Add missing
506 BFD_RELOC_SH_IMMS10BY8 relocation.
508 * config/tc-sh64.c (shmedia_build_Mytes): Emit an error message rather
509 than just ignoring bad code.
513 * config/tc-v850.c (handle_lo16): New function.
514 (v850_reloc_prefix): Use it to check lo().
515 (md_assemble, md_apply_fix3): Handle BFD_RELOC_V850_LO16_SPLIT_OFFSET.
519 * configure.in: Add s390x-ibm-tpf support.
520 * configure: Regenerate.
524 * config/obj-elf.c (obj_elf_change_section): Only set type and
525 attributes on new sections. Emit warning when type of re-declared
526 section doesn't match.
530 * dw2gencfi.c (dot.cfi.startproc): Clear cur_cfa_offset so
531 '.cfi_startproc simple' doesn't inherit the old value.
535 * dw2gencfi.c (output_cfi_insn): Adjust DW_CFA_def_cfa_sf generation
536 to emit a signed and factored offset. Adjust DW_CFA_def_cfa_offset_sf
537 generation to emit a factored offset.
541 * config/tc-mips.c (macro) [M_LA_AB]: Give an error for a offset
542 which is too large in the case of NO_PIC without 64-bit
545 * config/tc-mips.c (mips_in_shared): New static variable.
546 (macro_build_lui): Permit "_gp" if !mips_in_shared.
547 (md_longopts): Add -mshared and -mno-shared.
548 (md_parse_option): Handle OPTION_MSHARED and OPTION_MNO_SHARED.
549 (s_cpload): Implement !mips_in_shared case.
550 (s_cpsetup): Likewise.
551 * doc/c-mips.texi (MIPS Opts): Document -mno-shared.
552 * NEWS: Mention -mno-shared.
556 * config/tc-arm.c (s_arm_unwind_fnend): Use R_ARM_PREL31 relocation
561 * config/tc-mips.c (append_insn): If we emit a nop during a relax
562 sequence, increase the size of the sequence.
564 * config/tc-mips.c (mips_cpu_info_table): Change "9000" entry to
569 * read.c (s_align): Use an align_limit temporary to allay a GCC
570 signed/unsigned comparison warning.
574 * Makefile.am (TARG_ENV_HFILES): Add te-armlinuxeabi.h.
575 * configure.in: Use it for arm*-*-linux-gnueabi*.
576 * config/tc-arm.c: Allow emulation file to set FPU_DEFAULT.
577 * config/te-armlinuxeabi.h: New file.
578 * Makefile.in: Regenerated.
579 * aclocal.m4: Likewise.
580 * configure: Likewise.
581 * doc/Makefile.in: Regenerated.
585 * config/tc-xtensa.c (xtensa_switch_section_emit_state): Use subseg_set.
586 (xtensa_restore_emit_state): Likewise.
590 * read.c (ALIGN_LIMIT): Define, increasing limit for BFD_ASSEMBLER.
595 * Makefile.am: Change LOCALEDIR to $(datadir)/share.
596 * Makefile.in: Regenerate.
600 * config/tc-crx.c: Major code cleanup. Remove unused variables and
601 functions, give functions a meaningful name, add comments.
602 (check_range): New function - Replace operand size calculation
604 (assemble_insn): Update Algorithm, improve error issuing.
606 (process_label_constant): Bug fix regarding COP_BRANCH_INS relocation
611 * config/tc-m32r.c (md_pcrel_from_section): Fixed a pcrel relocte
612 miss between different section in the same module.
613 (tc_gen_reloc): Likewise.
617 * gas/config/tc-avr.c (mcu_types): Add support for atmega165,
618 atmega325, atmega3250, atmega645 and atmega6450.
622 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
623 permissible for the selected instruction suffix.
624 (process_suffix): For DefaultSize instructions, suppressing the
625 guessing of a 'q' suffix if the instruction doesn't support it is
626 pointless, because only an 'l' suffix can be guessed in this place.
630 * config/tc-iq2000.c: Remove support for IQ10 processor.
631 Convert to ISO C90 formatting.
632 * config/tc-iq2000.h: Likewise.
636 * config/tc-mn10300.c (md_relax_table): More fixes to the offsets
637 in this table. They should be correct now.
641 * config/tc-i386.h (CpuMMX2): Declare. Artificial classifier to
642 indicate the MMX extensions added by both SSE and 3DNow!A.
643 (Cpu3dnowA): Declare.
644 (CpuUnknownFlags): Update.
645 * config/tc-i386.c (cpu_sub_arch_name): Declare.
646 (cpu_arch): i586 and pentium do not imply MMX. i686 and pentiumpro do
647 neither imply SSE nor MMX. k6 implies MMX. k6_2 additionally implies
648 3DNow!. Athlon additionally implies 3DNow!A. Several new
649 entries (those starting with a dot are for sub-arch specification).
650 (set_cpu_arch): Handle sub-arch specifications.
651 (parse_insn): Distinguish between instructions not supported because
652 of insufficient CPU features and because of 64-bit mode.
653 * doc/c-i386.texi: Describe enhanced .arch directive.
657 * config/tc-xtensa.c (xg_add_opcode_fix): Set fx_no_overflow.
661 * dwarf2dbg.c (dwarf2_finish): Don't write a .debug_line section
662 without a corresponding .debug_info section.
666 * read.c (potable): Add "error" and "warning".
667 (s_errwarn): New function.
668 * read.h (s_errwarn): Declare.
669 * doc/as.texinfo (Error, Warning): Document .error and .warning.
673 * config/tc-tic54x.c (tic54x_adjust_symtab): Adjust call to
678 * config/tc-msp430.c (struct rcodes_s, MSP430_RLC, msp430_rcodes,
679 struct hcodes_s, msp430_hcodes): From include/opcode/msp430.h.
683 * config/obj-coff.c (c_dot_file_symbol): Add "app" param.
684 (coff_adjust_symtab): Adjust call.
685 (crawl_symbols): Likewise.
686 * config/obj-coff.h (c_dot_file_symbol): Add "app" param.
687 (obj_app_file): Adjust.
691 * configure.in: Enable bfd_assember for the MAXQ port.
692 * configure: Regenerate.
697 * config/tc-xtensa.c (finish_vinsn): Clear pending instruction if
699 (check_t1_t2_reads_and_writes): Check for both reads and writes to
700 interfaces that are related as determined by xtensa_interface_class_id.
704 * config/tc-mn10300.c (md_relax_table): Fix off by one negative
705 offsets for conditional branches.
709 * config/tc-xtensa.c (MAX_IMMED6): Change value to 65.
713 * config/tc-xtensa.c (update_next_frag_state): Always add a NOP if
714 relaxing at the end of a loop. Don't mark frags as UNREACHABLE or
716 (relax_frag_immed): Update call to update_next_frag_state.
720 * obj.h (struct format_ops <app_file>): Add int param.
721 * read.h (s_app_file_string): Likewise.
722 * read.c (s_app_file_string): Likewise.
723 (s_app_file): Adjust s_app_file_string call.
724 * config/tc-mips.c (s_mips_file): Likewise.
725 * config/obj-coff.h (obj_app_file): Add app param.
726 * config/obj-ecoff.h (obj_app_file): Likewise.
727 * config/obj-multi.h (obj_app_file): Likewise.
728 * config/obj-elf.h (elf_file_symbol): Likewise.
729 * config/obj-elf.c (elf_file_symbol): Only emit one file symbol
730 if called for # preprocessor lines.
735 * symbols.c (resolve_symbol_value): Convert weak symbols only
737 (symbol_equated_reloc_p): Don't equate weaks when relocating
738 only for Windows PECOFF.
742 * config/tc-crx.c (print_insn): Check and set insn_addr.
743 * config/tc-crx.h (md_frag_check): Define.
748 * configure.in: Add support for new target maxq-coff.
749 * configure: Regenerate.
750 * NEWS: Mention new support.
751 * config/tc-maxq.c: New file.
752 * config/tc-maxq.h: New file.
753 * config/obj-coff.h: Add support for maxq-coff.
757 * symbols.c (any_external_name): Define.
758 (resolve_symbol_value): Convert weak symbols.
759 (S_SET_EXTERNAL): Support any_external_name.
760 (S_SET_NAME): Qualify parameter const.
761 (symbol_equated_reloc_p): Equate to weaks when relocating.
762 * symbols.h (S_SET_NAME): Qualfiy parameter const.
763 * tc.h (any_external_name): Declare.
764 * config/obj-coff.c ("coff/pe.h"): Include for BFD
766 (weak_is_altname): Declare and define.
767 (weak_name2altname): Same.
768 (weak_altname2name): Same.
769 (weak_uniquify): Same.
770 (weak_altprefix): Define.
771 (obj_coff_weak): Change .weak syntax and handling.
772 (coff_frob_symbol): Fix PE weak symbol alternates.
773 * config/obj-coff.h (USE_UNIQUE): Define.
774 * config/tc-i386.c (md_apply_fix3): Assume weak symbols
775 are in another segment.
776 (tc_gen_reloc): Remove broken addend hack.
777 doc/as.texinfo: Update.
781 * config/tc-xtensa.c (total_frag_text_expansion): New.
782 (md_estimate_size_before_relax): Use it.
783 (find_address_of_next_align_frag): Likewise.
787 * config/tc-crx.c: Rename argument types.
788 (processing_arg_number): Rename to 'cur_arg_num'.
789 (get_number_of_bits): Rename to 'set_operand_size'.
790 (get_operandtype): Rename to 'parse_operand', totally rewrite.
791 (set_cons_rparams): Rename to 'set_operand', totally rewrite.
792 (set_indexmode_parameters): Remove function, integrate its code into
794 (set_operand_size): Get rid of 'Operand Number' function parameter -
795 use global variable 'cur_arg_num' instead.
796 Use a local 'argument' pointer to reference the current argument.
797 (parse_operand): Likewise.
798 (set_operand): Likewise.
799 (process_label_constant): Likewise.
803 * config/tc-xtensa.c: Remove XTENSA_SECTION_RENAME ifdefs.
804 (add_section_rename): Delete. Inlined into...
805 (build_section_rename): ...here. Use xstrdup instead of strdup.
806 (xtensa_section_rename): Drop "const" from argument and return types.
807 (md_show_usage): Indent to match show_usage().
808 * config/tc-xtensa.h: Remove XTENSA_SECTION_RENAME ifdefs.
809 (tc_canonicalize_section_name): Define.
810 (md_elf_section_rename): Remove unused macro.
811 * doc/as.texinfo (Overview): Document Xtensa --rename-section option.
812 * doc/c-xtensa.texi (Xtensa Options): Likewise.
813 (Frame Directive): Delete.
817 * configure.in: Remove arm-*-oabi and thumb-*-oabi.
818 * config/tc-arm.c (target_oabi): Delete.
819 (md_apply_fix3, elf32_arm_target_format): Remove target_oabi checks.
820 (arm_opts): Remove moabi.
821 * doc/as.texinfo (Overview): Remove documentation of -moabi.
822 * doc/c-arm.texi (ARM Options): Likewise.
823 * configure: Regenerated.
827 * configure.in (crisv32): Recognize. AC_DEFINE_UNQUOTED
828 DEFAULT_CRIS_ARCH. Handle crisv32-*-linux-gnu* like
829 cris-*-linux-gnu* and crisv32-*-* like cris-*-*.
830 * configure: Regenerate.
831 * config/tc-cris.c (enum cris_archs): New.
832 (cris_mach, cris_arch_from_string, s_cris_arch, get_sup_reg)
833 (cris_insn_ver_valid_for_arch): New functions.
834 (DEFAULT_CRIS_ARCH): New macro, default to cris_any_v0_v10.
835 (cris_arch): New variable.
836 (md_pseudo_table): New pseudo .arch.
837 (err_for_dangerous_mul_placement): Initialize according to
839 (STATE_COND_BRANCH): Renamed from STATE_CONDITIONAL_BRANCH.
841 (STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON)
842 (STATE_ABS_BRANCH_V32, STATE_LAPC, BRANCH_BF_V32, BRANCH_BB_V32)
843 (BRANCH_WF_V32, BRANCH_WB_V32): New.
844 (BRANCH_BF, BRANCH_BB, BRANCH_WF, BRANCH_WB): Don't undef after
845 use in md_cris_relax_table.
846 (md_cris_relax_table): Add entries for STATE_COND_BRANCH_V32,
847 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
848 Update and improve head comment.
849 (OPTION_PIC): Define in terms of previous option, OPTION_US.
850 (OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): Similar.
852 (md_longopts): New option --march=...
853 (cris_any_v0_v10_long_jump_size, crisv32_long_jump_size): New
855 (md_long_jump_size): Initialize in terms of DEFAULT_CRIS_ARCH.
856 (HANDLE_RELAXABLE): New macro.
857 (md_estimate_size_before_relax): Use HANDLE_RELAXABLE for common
858 cases. Check for weak symbols and assume not relaxable. Handle
859 STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON,
860 STATE_ABS_BRANCH_V32, STATE_LAPC. Use new variable symbolP, not
862 (md_convert_frag): Handle STATE_COND_BRANCH_V32,
863 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
864 (cris_create_short_jump): Adjust for CRISv32.
865 (cris_relax_frag): Handle new states.
866 (md_create_long_jump): Ditto. Emit error for common_v10_v32.
867 (md_begin): Define symbols "..asm.arch.cris.v32",
868 "..asm.arch.cris.v10", "..asm.arch.cris.common_v10_v32" and
869 "..asm.arch.cris.any_v0_v10". Use cris_insn_ver_valid_for_arch
870 when entering opcode table entry points.
871 (md_assemble): Adjust branch handling for CRISv32. Handle LAPC
872 relaxation. In fix_new_exp call for main insn, pass 1 for pcrel
873 parameter for 8, 16 and 32-bit pc-relative insns and LAPC.
874 (cris_process_instruction): Initialize out_insnp->insn_type to
875 CRIS_INSN_NONE, not CRIS_INSN_NORMAL.
876 <case ']', '[', 'A', 'd', 'Q', 'N', 'n', 'Y', 'U', 'u', 'T'>: New
878 <case 'm'>: Check that modified_char == '.'.
879 <invalid operands>: Consume the rest of the line.
880 When operands don't match, skip over subsequent insns with
881 non-matching version specifier but same mnemonic.
882 <immediate constant, case SIZE_SPEC_REG>: Immediate operands for
883 special registers in CRISv32 are always 32 bit long.
884 <immediate constant, case SIZE_FIELD_SIGNED, SIZE_FIELD_UNSIGNED>:
886 (get_gen_reg): Only recognize "PC" when followed by "+]" for v32
887 and compatible. Recognize "ACR" for v32, unless followed by "+".
888 (get_spec_reg): Consider cris_arch when looking up register.
889 (get_autoinc_prefix_or_indir_op): Don't recognize assignment for
891 (get_3op_or_dip_prefix_op): Check for ']' after seeing '[rN+'.
892 (cris_get_expression): Restore input_line_pointer if failing "early".
893 (get_flags): Consider cris_arch and recognize flags accordingly.
894 (branch_disp): Adjust for CRISv32.
895 (gen_cond_branch_32): Similar. Emit error for common_v10_v32.
896 (cris_number_to_imm): Use as_bad_where, not as_bad. Remove
897 related FIXME. Don't insist on BFD_RELOC_32_PCREL fixup to be
898 resolved. Don't enter zeros in object file for
900 <case BFD_RELOC_CRIS_LAPCQ_OFFSET, BFD_RELOC_CRIS_SIGNED_16>
901 <case BFD_RELOC_CRIS_SIGNED_8>: New case.
902 (md_parse_option): Break out "return 1".
903 <OPTION_ARCH> New case.
904 (tc_gen_reloc): <case BFD_RELOC_CRIS_LAPCQ_OFFSET>
905 <case BFD_RELOC_CRIS_SIGNED_16, BFD_RELOC_CRIS_SIGNED_8>
906 <case BFD_RELOC_CRIS_UNSIGNED_8, BFD_RELOC_CRIS_UNSIGNED_16>
907 <case BFD_RELOC_32_PCREL>: New cases.
908 Addends for non-zero fx_pcrel are too in fx_offset.
909 (md_show_usage): Show --march=<arch>.
910 (md_apply_fix3): Adjust val for BFD_RELOC_CRIS_LAPCQ_OFFSET.
911 (md_pcrel_from): BFD_RELOC_CRIS_LAPCQ_OFFSET is PC-relative too.
912 (s_syntax) <struct syntaxes>: Properly constify member operand.
913 * config/tc-cris.h (TARGET_MACH): Define.
914 (cris_mach): Declare.
915 * doc/as.texinfo (Overview) <CRIS>: Add --march=...
916 * doc/c-cris.texi (CRIS-Symbols): New node for built-in symbols.
917 (CRIS-Opts): Document --march=...
918 (CRIS-Pseudos): Document .arch.
922 * config/tc-i386.c (set_intel_syntax): Allow % in symbol names when
923 intel syntax and no register prefix, allow $ in symbol names when
925 (set_16bit_gcc_code_flag): Replace literal 'l' by LONG_MNEM_SUFFIX.
926 (intel_float_operand): Add fourth return value indicating math control
927 operations. Make classification more precise.
928 (md_assemble): Complain if memory operand of mov[sz]x has no size
930 (parse_insn): Translate word operands to floating point instructions
931 operating on integers as well as control instructions to short ones
932 as expected by AT&T syntax. Translate 'd' suffix to short one only for
933 floating point instructions operating on non-integer operands.
934 (match_template): Remove fldcw special case. Adjust q-suffix handling
935 to permit it on fild/fistp/fisttp in AT&T mode.
936 (process_suffix): Don't guess DefaultSize insns' suffix from
937 stackop_size for certain floating point control instructions. Guess
938 suffix for branch and [ls][gi]dt based on flag_code. Split error
939 messages for Intel and AT&T syntax, and make the condition more strict
940 for the former. Adjust suppressing of generation of operand size
942 (intel parser): Allow the full set of MASM operators. Add FWORD, TBYTE,
943 OWORD, and XMMWORD operand size specifiers (TBYTE replaces XWORD). Add
945 * config/tc-i386.h (BYTE_PTR WORD_PTR DWORD_PTR QWORD_PTR XWORD_PTR
946 SHORT OFFSET_FLAT FLAT NONE_FOUND): Remove unused defines.
950 * symbols.c (colon) [!WORKING_DOT_WORD]: Don't declare
951 md_short_jump_size, md_long_jump_size.
952 * write.c [!WORKING_DOT_WORD]: Ditto.
953 * tc.h [!WORKING_DOT_WORD]: Declare them here. Drop const
955 * config/tc-cris.h (md_short_jump_size, md_long_jump_size): Don't
957 * config/tc-cris.c (md_short_jump_size, md_long_jump_size): Drop
958 const qualifier in these definitions.
959 * config/tc-i370.c, config/tc-m68k.c, config/tc-pdp11.c,
960 config/tc-s390.c, config/tc-tahoe.c, config/tc-vax.c: Ditto.
964 * dwarf2dbg.c (dwarf2_finish): Check for the existence of a file
965 table before deciding to produce a .debug_line section to match up
966 with a user provided .debug_info section.
970 * config/tc-crx.c (getreg_image): Bug fix, a return value was
971 mistakenly omitted from CRX_C_REGTYPE and CRX_CS_REGTYPE cases.
975 * config/tc-crx.c: Remove global variable 'post_inc_mode'.
976 (get_flags): New function.
977 (get_number_of_bits): Edit comments, update numeric values to
979 (process_label_constant): Don't support the colon
980 format (SYMBOL:[s|m|l]).
981 (set_cons_rparams): Support argument type 'arg_rbase'.
982 (get_operandtype): Bug fix in 'rbase' operand type parsing.
983 (handle_LoadStor): Bug fix, first handle post-increment mode.
984 (getreg_image): Remove redundant code, update according to latest
986 (print_constant): Bug fix relate to 3-word instructions.
987 (assemble_insn): Bug fix, when matching instructions, verify also
988 instruction type (not only mnemonic).
989 Add various error checking.
990 (preprocess_reglist): Support HI/LO and user registers.
994 * config/tc-ia64.c (fixup_unw_records): Don't let the "t" value
995 in an epilogue directive go negative.
1000 * config/tc-ia64.c (emit_one_bundle): Decrement md.num_slots_in_use
1001 after reporting template error during manual bundling. Reported
1006 * Makefile.am: Run dep-am.
1007 * aclocal.m4: Regenerate with automake 1.9.2.
1008 * Makefile.in: Regenerate with automake 1.9.2.
1009 * doc/Makefile.in: Likewise.
1011 * config/tc-arm.c: Include "dw2gencfi.h".
1012 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
1014 * config/tc-arm.h (TARGET_USE_CFIPOP, DWARF2_DEFAULT_RETURN_COLUMN)
1015 (DWARF2_CIE_DATA_ALIGNMENT, tc_regname_to_dw2regnum)
1016 (tc_cfi_frame_initial_instructions): Define.
1017 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
1022 * config/tc-crx.c (assemble_insn): Check unsigned immediate
1024 Update coprocessor id to be unsigned immediate.
1028 * config/tc-i386.c (O_secrel): Delete.
1029 (tc_pe_dwarf2_emit_offset): New function.
1030 * config/tc-i386.h (O_secrel): Define as O_md1.
1031 (TC_DWARF2_EMIT_OFFSET): Define.
1035 * config/tc-xstormy16.c (xstormy16_cons_fix_new): Accept and
1036 ignore @fptr() directives for 4-byte fixups.
1040 * Makefile.am: Run "make dep-am"
1041 * Makefile.in: Regenerate.
1045 * doc/c-xtensa.texi (Xtensa Options, Absolute Literals Directive):
1046 Remove comments about placement of literal pools.
1047 (Literal Directive): Update description of literal placement.
1048 (Literal Prefix Directive): Remove statement that this does not apply
1049 to absolute-mode literals. Describe new section naming scheme.
1053 * config/tc-xtensa.c (xtensa_frequency_pseudo): Use set_subseg_freq.
1054 (is_entry_opcode, is_movi_opcode, is_the_loop_opcode, is_jx_opcode,
1055 is_windowed_return_opcode): Delete.
1056 (xtensa_frob_label): Use get_subseg_target_freq.
1057 (md_assemble): Inline call to is_entry_opcode.
1058 (xtensa_handle_align): Inline call to get_frag_is_literal.
1059 (relaxation_requirements): Inline call to is_jx_opcode.
1060 (emit_single_op): Inline call to is_movi_opcode.
1061 (xg_assemble_vliw_tokens): Inline calls to get_frag_is_insn,
1062 get_frag_is_no_transform, is_entry_opcode, and
1063 set_frag_is_specific_opcode. Use get_subseg_total_freq.
1064 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags,
1065 xtensa_fix_close_loop_end_frags, relax_frag_immed, convert_frag_immed):
1066 Inline calls to get_frag_is_no_transform.
1067 (next_instrs_are_b_retw): Inline call to is_windowed_return_opcode.
1068 (xtensa_fix_short_loop_frags): Inline calls to is_the_loop_opcode and
1069 get_frag_is_no_transform.
1070 (convert_frag_immed_finish_loop): Inline calls to get_expression_value
1071 and set_frag_is_no_transform.
1072 (get_expression_value): Delete.
1073 (subseg_map struct): Rename cur_total_freq to total_freq. Rename
1074 cur_target_freq to target_freq.
1075 (get_subseg_info): Split out code to create a new map entry into ...
1076 (add_subseg_info): ... this new function.
1077 (get_last_insn_flags): Check if get_subseg_info succeeded.
1078 (set_last_insn_flags): Call add_subseg_info if needed.
1079 (get_subseg_total_freq, get_subseg_target_freq, set_subseg_freq): New.
1080 (xtensa_reorder_segments): Compute last_sec while counting sections.
1081 Remove call to get_last_sec.
1082 (get_last_sec): Delete.
1083 (cache_literal_section): Inline call to retrieve_literal_seg and its
1084 callees, seg_present and add_seg_list.
1085 (retrieve_literal_seg, seg_present, add_seg_list): Delete.
1086 (get_frag_is_insn, get_frag_is_no_transform,
1087 set_frag_is_specific_opcode, set_frag_is_no_transform): Delete.
1088 * config/tc-xtensa.h (MAX_SLOTS): Reduce from 31 to 15.
1092 * config/tc-xtensa.c: Use ISO C90 formatting.
1093 * config/tc-xtensa.h: Likewise.
1094 * config/xtensa-istack.h: Likewise.
1095 * config/xtensa-relax.c: Likewise.
1096 * config/xtensa-relax.h: Likewise.
1100 * config/tc-arm.c (md_begin): Change EF_ARM_EABI_VER3 to
1103 * doc/c-arm.texi: Document that we actually support -meabi=4, not
1108 * doc/as.texinfo (VTableEntry, VTableInherit): Add "directive" to index
1110 (Acknowledgements): Use "GAS" instead of AS variable.
1114 * config/tc-i386.c: Include "elf/x86-64.h".
1115 (i386_elf_section_type): New function.
1116 * config/tc-i386.h (md_elf_section_type): Define.
1117 (i386_elf_section_type): New prototype.
1121 * config/m68k-parse.h (enum m68k_register): New control register,
1123 * config/tc-m68k.c: Correct control register set for MCF5249.
1128 * config/tc-xtensa.c (absolute_literals_supported): New global flag.
1129 (UNREACHABLE_MAX_WIDTH): Define.
1130 (XTENSA_FETCH_WIDTH): Delete.
1131 (cur_vinsn, xtensa_fetch_width, xt_saved_debug_type, past_xtensa_end,
1132 prefer_const16, prefer_l32r): New global variables.
1133 (LIT4_SECTION_NAME): Define.
1134 (lit4_state struct): Add lit4_seg_name and lit4_seg fields.
1135 (XTENSA_PROP_*, GET_XTENSA_PROP_*, SET_XTENSA_PROP_*): Define.
1136 (frag_flags struct): New.
1137 (xtensa_block_info struct): Move from tc-xtensa.h. Add flags field.
1138 (subseg_map struct): Add cur_total_freq and cur_target_freq fields.
1139 (bitfield, bit_is_set, set_bit, clear_bit): Define.
1140 (MAX_FORMATS): Define.
1141 (op_placement_info struct, op_placement_table): New.
1142 (O_pltrel, O_hi16, O_lo16): Define.
1143 (directiveE enum): Rename directive_generics to directive_transform.
1144 Delete directive_relax. Add directive_schedule,
1145 directive_absolute_literals, and directive_last_directive.
1146 (directive_info): Rename "generics" to "transform". Delete "relax".
1147 Add "schedule" and "absolute-literals".
1148 (directive_state): Adjust entries to match changes in directive_info.
1149 (xtensa_relax_statesE, RELAX_IMMED_MAXSTEPS): Move to tc-xtensa.h.
1150 (xtensa_const16_opcode, xtensa_movi_opcode, xtensa_movi_n_opcode,
1151 xtensa_l32r_opcode, xtensa_nop_opcode, xtensa_rsr_lcount_opcode): New.
1152 (xtensa_j_opcode, xtensa_rsr_opcode): Delete.
1153 (align_only_targets, software_a0_b_retw_interlock,
1154 software_avoid_b_j_loop_end, maybe_has_b_j_loop_end,
1155 software_avoid_short_loop, software_avoid_close_loop_end,
1156 software_avoid_all_short_loops, specific_opcode): Delete.
1157 (warn_unaligned_branch_targets): New.
1158 (workaround_a0_b_retw, workaround_b_j_loop_end, workaround_short_loop,
1159 workaround_close_loop_end, workaround_all_short_loops): Default FALSE.
1160 (option_[no_]link_relax, option_[no_]transform,
1161 option_[no_]absolute_literals, option_warn_unaligned_targets,
1162 option_prefer_l32r, option_prefer_const16, option_target_hardware):
1164 (option_[no_]align_only_targets, option_literal_section_name,
1165 option_text_section_name, option_data_section_name,
1166 option_bss_section_name, option_eb, option_el): Delete.
1167 (md_longopts): Add entries for: [no-]transform, [no-]absolute-literals,
1168 warn-unaligned-targets, prefer-l32r, prefer-const16, [no-]link-relax,
1169 and target-hardware. Delete entries for [no-]target-align-only,
1170 literal-section-name, text-section-name, data-section-name, and
1172 (md_parse_option): Handle new options and remove old ones. Accept but
1173 ignore [no-]density options. Warn for [no-]generics and [no-]relax
1174 and treat them as [no-]transform.
1175 (md_show_usage): Add new options and remove old ones.
1176 (xtensa_setup_hw_workarounds): New.
1177 (md_pseudo_table): Change "word" entry to use xtensa_elf_cons. Add
1178 "long", "short", "loc" and "frequency" entries.
1179 (use_generics): Rename to ...
1180 (use_transform): ... this function. Add past_xtensa_end check.
1181 (use_longcalls): Add past_xtensa_end check.
1182 (code_density_available, can_relax): Delete.
1183 (do_align_targets): New.
1184 (get_directive): Accept dashes in directive names. Warn about
1185 [no-]generics and [no-]relax directives and treat them as
1187 (xtensa_begin_directive): Call md_flush_pending_output only for some
1188 directives. Check for directives inside instruction bundles. Warn
1189 about deprecated ".begin literal" usage. Warn and ignore [no-]density
1190 directives. Handle new directives. Check generating_literals flag
1192 (xtensa_end_directive): Check for directives inside instruction
1193 bundles. Warn and ignore [no-]density directives. Handle new
1194 directives. Call xtensa_set_frag_assembly_state.
1195 (xtensa_loc_directive_seen, xtensa_dwarf2_directive_loc,
1196 xtensa_dwarf2_emit_insn): New.
1197 (xtensa_literal_position): Call md_flush_pending_output. Do not check
1198 use_literal_section flag.
1199 (xtensa_literal_pseudo): Call md_flush_pending_output. Handle absolute
1200 literals. Use xtensa_elf_cons to parse the expression.
1201 (xtensa_literal_prefix): Do not check use_literal_section. Support
1202 ".lit4" sections for absolute literals. Change prefix convention to
1203 replace ".text" (or ".t" in a linkonce section). No need to call
1205 (xtensa_frequency_pseudo, xtensa_elf_cons, xtensa_elf_suffix): New.
1206 (expression_end): Handle closing braces and colons.
1207 (PLT_SUFFIX, plt_suffix): Delete.
1208 (expression_maybe_register): Use new xtensa-isa.h functions. Use
1209 xtensa_elf_suffix instead of checking for plt suffix, and handle O_lo16
1210 and O_hi16 expressions as well.
1211 (tokenize_arguments): Handle closing braces and colons.
1212 (parse_arguments): Use new xtensa-isa.h functions. Handle "invisible"
1213 operands and paired register syntax.
1214 (get_invisible_operands): New.
1215 (xg_translate_sysreg_op): Handle new Xtensa LX RSR/WSR/XSR syntax. Use
1216 new xtensa-isa.h functions.
1217 (xtensa_translate_old_userreg_ops, xtensa_translate_zero_immed): New.
1218 (xg_translate_idioms): Check if inside bundle. Use use_transform.
1219 Handle new Xtensa LX RSR/WSR/XSR syntax. Remove code to widen density
1220 instructions. Use xtensa_translate_zero_immed.
1221 (operand_is_immed, operand_is_pcrel_label): Delete.
1222 (get_relaxable_immed): Use new xtensa-isa.h functions.
1223 (get_opcode_from_buf): Add slot parameter. Use new xtensa-isa.h
1225 (xtensa_print_insn_table, print_vliw_insn): New.
1226 (is_direct_call_opcode): Use new xtensa-isa.h functions.
1227 (is_call_opcode, is_loop_opcode, is_conditional_branch_opcode,
1228 is_branch_or_jump_opcode): Delete.
1229 (is_movi_opcode, decode_reloc, encode_reloc, encode_alt_reloc): New.
1230 (opnum_to_reloc, reloc_to_opnum): Delete.
1231 (xtensa_insnbuf_set_operand, xtensa_insnbuf_get_operand): Use new
1232 xtensa-isa.h functions. Operate on one slot of an instruction.
1233 (xtensa_insnbuf_set_immediate_field, is_negatable_branch,
1234 xg_get_insn_size): Delete.
1235 (xg_get_build_instr_size): Use xg_get_single_size.
1236 (xg_is_narrow_insn, xg_is_single_relaxable_insn): Update calls to
1237 xg_build_widen_table. Use xg_get_single_size.
1238 (xg_get_max_narrow_insn_size): Delete.
1239 (xg_get_max_insn_widen_size, xg_get_max_insn_widen_literal_size,
1240 xg_is_relaxable_insn): Update calls to xg_build_widen_table. Use
1242 (xg_build_to_insn): Record the loc field. Handle OP_OPERAND_HI16U and
1243 OP_OPERAND_LOW16U. Check xg_valid_literal_expression.
1244 (xg_expand_to_stack, xg_expand_narrow): Update calls to
1245 xg_build_widen_table. Use xg_get_single_size.
1246 (xg_immeds_fit): Use new xtensa-isa.h functions. Update call to
1248 (xg_symbolic_immeds_fit): Likewise. Also handle O_lo16 and O_hi16, and
1249 treat weak symbols conservatively.
1250 (xg_check_operand): Use new xtensa-isa.h functions.
1251 (is_dnrange): Delete.
1252 (xg_assembly_relax): Inline previous calls to tinsn_copy.
1253 (xg_finish_frag): Specify separate relax states for the frag and slot0.
1254 (is_branch_jmp_to_next, xg_add_branch_and_loop_targets): Use new
1255 xtensa-isa.h functions.
1256 (xg_instruction_matches_option_term, xg_instruction_matches_or_options,
1257 xg_instruction_matches_options): New.
1258 (xg_instruction_matches_rule): Handle O_register expressions. Call
1259 xg_instruction_matches_options.
1260 (transition_rule_cmp): New.
1261 (xg_instruction_match): Update call to xg_build_simplify_table.
1262 (xg_build_token_insn): Record loc fields.
1263 (xg_simplify_insn): Check is_specific_opcode field and
1264 density_supported flag.
1265 (xg_expand_assembly_insn): Skip checking code_density_available. Use
1266 new xtensa-isa.h functions. Call use_transform instead of can_relax.
1267 (xg_assemble_literal): Add error handling for O_big. Call
1268 record_alignment. Handle O_pltrel.
1269 (xg_valid_literal_expression): New.
1270 (xg_assemble_literal_space): Add slot parameter. Remove call to
1271 set_expr_symbol_offset. Add call to record_alignment. Update call to
1273 (xg_emit_insn): Delete.
1274 (xg_emit_insn_to_buf): Add format parameter. Update calls to
1275 xg_add_opcode_fix and xtensa_insnbuf_to_chars.
1276 (xg_add_opcode_fix): Change opcode parameter to tinsn and add format
1277 and slot parameters. Handle new "alternate" relocations for absolute
1278 literals and CONST16 instructions. Check for bad uses of O_lo16 and
1279 O_hi16. Use new xtensa-isa.h functions.
1280 (xg_assemble_tokens): Delete.
1281 (is_register_writer): Use new xtensa-isa.h functions.
1282 (is_bad_loopend_opcode): Check for xtensa_rsr_lcount_opcode instead of
1283 old-style RSR from LCOUNT.
1284 (next_frag_opcode): Delete.
1285 (next_frag_opcode_is_loop, next_frag_format_size, frag_format_size,
1286 update_next_frag_state): New.
1287 (update_next_frag_nop_state): Delete.
1288 (next_frag_pre_opcode_bytes): Use next_frag_opcode_is_loop.
1289 (xtensa_mark_literal_pool_location): Check use_literal_section flag and
1290 the state of the absolute-literals directive. Add calls to
1291 record_alignment and xtensa_set_frag_assembly_state. Call
1292 xtensa_switch_to_non_abs_literal_fragment instead of
1293 xtensa_switch_to_literal_fragment.
1295 (assemble_nop): Use build_nop. Update call to xtensa_insnbuf_to_chars.
1296 (get_expanded_loop_offset): Change check for undefined opcode to an
1298 (xtensa_set_frag_assembly_state, relaxable_section,
1299 xtensa_find_unmarked_state_frags, xtensa_find_unaligned_branch_targets,
1300 xtensa_find_unaligned_loops, xg_apply_tentative_value): New.
1301 (md_begin): Update call to xtensa_isa_init. Initialize linkrelax to 1.
1302 Set lit4_seg_name. Call xg_init_vinsn. Initialize new global opcodes.
1303 Call init_op_placement_info_table and xtensa_set_frag_assembly_state.
1304 (xtensa_init_fix_data): New.
1305 (xtensa_frob_label): Reset label symbol to the current frag. Check
1306 do_align_targets and generating_literals flag. Propagate frequency
1307 info to new alignment frag. Call xtensa_set_frag_assembly_state.
1308 (xtensa_unrecognized_line): New.
1309 (xtensa_flush_pending_output): Check if inside a bundle. Add a call
1310 to xtensa_set_frag_assembly_state.
1311 (error_reset_cur_vinsn): New.
1312 (md_assemble): Remove check for literal frag. Remove call to
1313 istack_init. Call use_transform instead of use_generics. Parse
1314 explicit instruction format specifiers. Move code for
1315 a0_b_retw_interlock workaround to xg_assemble_vliw_tokens. Call
1316 error_reset_cur_vinsn on errors. Add call to get_invisible_operands.
1317 Add dwarf2_where call. Remote automatic alignment for ENTRY
1318 instructions. Move call to xtensa_clear_insn_labels to the end.
1319 Rearrange to handle bundles.
1320 (xtensa_cons_fix_new): Delete.
1321 (xtensa_handle_align): New.
1322 (xtensa_frag_init): Call xtensa_set_frag_assembly_state. Remove
1323 assignment to is_no_density field.
1324 (md_pcrel_from): Use new xtensa-isa.h functions. Use decode_reloc
1325 instead of reloc_to_opnum. Handle "alternate" relocations.
1326 (xtensa_force_relocation, xtensa_check_inside_bundle,
1327 xtensa_elf_section_change_hook): New.
1328 (xtensa_symbol_new_hook): Delete.
1329 (xtensa_fix_adjustable): Check for difference of symbols with an
1330 offset. Check for external and weak symbols.
1331 (md_apply_fix3): Remove cases for XTENSA_OP{0,1,2} relocs.
1332 (md_estimate_size_before_relax): Return expansion for the first slot.
1333 (tc_gen_reloc): Handle difference of symbols by producing
1334 XTENSA_DIFF{8,16,32} relocs and by writing the value of the difference
1335 into the output. Handle new XTENSA_SLOT*_OP relocs by storing the
1336 tentative values into the output when linkrelax is set.
1337 (XTENSA_PROP_SEC_NAME): Define.
1338 (xtensa_post_relax_hook): Call xtensa_find_unmarked_state_frags.
1339 Create literal tables only if using literal sections. Create new
1340 property tables instead of old instruction tables. Check for unaligned
1341 branch targets and loops.
1342 (finish_vinsn, find_vinsn_conflicts, check_t1_t2_reads_and_writes,
1343 new_resource_table, clear_resource_table, resize_resource_table,
1344 resources_available, reserve_resources, release_resources,
1345 opcode_funcUnit_use_unit, opcode_funcUnit_use_stage,
1346 resources_conflict, xg_find_narrowest_format, relaxation_requirements,
1347 bundle_single_op, emit_single_op, xg_assemble_vliw_tokens): New.
1348 (xtensa_end): Call xtensa_flush_pending_output. Set past_xtensa_end
1349 flag. Update checks for workaround options. Call
1350 xtensa_mark_narrow_branches and xtensa_mark_zcl_first_insns.
1351 (xtensa_cleanup_align_frags): Add special case for branch targets.
1352 Check for and mark unreachable frags.
1353 (xtensa_fix_target_frags): Remove use of align_only_targets flag.
1354 Use RELAX_LOOP_END_BYTES in special case for negatable branch at the
1355 end of a zero-overhead loop body.
1356 (frag_can_negate_branch): Handle instructions with multiple slots.
1357 Use new xtensa-isa.h functions
1358 (xtensa_mark_narrow_branches, is_narrow_branch_guaranteed_in_range,
1359 xtensa_mark_zcl_first_insns): New.
1360 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags): Error if
1361 transformations are disabled.
1362 (next_instrs_are_b_retw): Use new xtensa-isa.h functions. Handle
1363 multislot instructions.
1364 (xtensa_fix_close_loop_end_frags, xtensa_fix_short_loop_frags):
1365 Likewise. Also error if transformations are disabled.
1366 (unrelaxed_frag_max_size): New.
1367 (unrelaxed_frag_min_insn_count, unrelax_frag_has_b_j): Use new
1368 xtensa-isa.h functions.
1369 (xtensa_sanity_check, is_empty_loop, is_local_forward_loop): Use
1370 xtensa_opcode_is_loop instead of is_loop_opcode.
1371 (get_text_align_power): Replace as_fatal with assertion.
1372 (get_text_align_fill_size): Iterate instead of using modulus when
1374 (get_noop_aligned_address): Assert that this is for a machine-dependent
1375 RELAX_ALIGN_NEXT_OPCODE frag. Use next_frag_opcode_is_loop,
1376 xg_get_single_size, and frag_format_size.
1377 (get_widen_aligned_address): Rename to ...
1378 (get_aligned_diff): ... this function. Add max_diff parameter.
1379 Remove handling of rs_align/rs_align_code frags. Use
1380 next_frag_format_size, get_text_align_power, get_text_align_fill_size,
1381 next_frag_opcode_is_loop, and xg_get_single_size. Compute max_diff
1382 and pass it back to caller.
1383 (xtensa_relax_frag): Use relax_frag_loop_align. Add code for new
1384 RELAX_SLOTS, RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN,
1385 RELAX_FILL_NOP, and RELAX_UNREACHABLE frag types. Check relax_seen.
1386 (relax_frag_text_align): Rename to ...
1387 (relax_frag_loop_align): ... this function. Assume loops can only be
1388 in the first slot of an instruction.
1389 (relax_frag_add_nop): Use assemble_nop instead of constructing an OR
1390 instruction. Remove call to frag_wane.
1391 (relax_frag_narrow): Rename to ...
1392 (relax_frag_for_align): ... this function. Extend to handle
1393 RELAX_FILL_NOP and RELAX_UNREACHABLE, as well as RELAX_SLOTS with
1394 RELAX_NARROW for the first slot.
1395 (find_address_of_next_align_frag, bytes_to_stretch): New.
1396 (future_alignment_required): Use find_address_of_next_align_frag and
1397 bytes_to_stretch. Look ahead to subsequent frags to make smarter
1398 alignment decisions.
1399 (relax_frag_immed): Add format, slot, and estimate_only parameters.
1400 Check if transformations are enabled for b_j_loop_end workaround.
1401 Use new xtensa-isa.h functions and handle multislot instructions.
1402 Update call to xg_assembly_relax.
1403 (md_convert_frag): Handle new RELAX_SLOTS, RELAX_UNREACHABLE,
1404 RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN, and RELAX_FILL_NOP
1406 (convert_frag_narrow): Add segP, format and slot parameters. Call
1407 convert_frag_immed for branch instructions. Adjust calls to
1408 tinsn_from_chars, tinsn_immed_from_frag, and xg_emit_insn_to_buf. Use
1409 xg_get_single_size and xg_get_single_format.
1410 (convert_frag_fill_nop): New.
1411 (convert_frag_immed): Add format and slot parameters. Handle multislot
1412 instructions and use new xtensa-isa.h functions. Update calls to
1413 tinsn_immed_from_frag and xg_assembly_relax. Check if transformations
1414 enabled for b_j_loop_end workaround. Use build_nop instead of
1415 assemble_nop. Check is_specific_opcode flag. Check for unreachable
1416 frags. Use xg_get_single_size. Handle O_pltrel.
1417 (fix_new_exp_in_seg): Remove check for old plt flag.
1418 (convert_frag_immed_finish_loop): Update calls to tinsn_from_chars and
1419 xtensa_insnbuf_to_chars. Call tinsn_immed_from_frag. Change check
1420 for loop opcode to an assertion. Mark all frags up to the end of the
1421 loop as not transformable.
1422 (get_last_insn_flags, set_last_insn_flags): Use get_subseg_info.
1423 (get_subseg_info): New.
1424 (xtensa_move_literals): Call xtensa_set_frag_assembly_state. Add null
1426 (xtensa_switch_to_literal_fragment): Rewrite to handle absolute
1427 literals and use xtensa_switch_to_non_abs_literal_fragment otherwise.
1428 (xtensa_switch_to_non_abs_literal_fragment): New.
1429 (cache_literal_section): Add is_code parameter and pass it through to
1430 retrieve_literal_seg.
1431 (retrieve_literal_seg): Add is_code parameter and use it to set the
1432 flags on the literal section. Handle case where head parameter is 0.
1433 (get_frag_is_no_transform, set_frag_is_specific_opcode,
1434 set_frag_is_no_transform): New.
1435 (xtensa_create_property_segments): Add end_property_function parameter
1436 and pass it through to add_xt_block_frags. Call bfd_get_section_flags
1437 and skip SEC_DEBUGGING and !SEC_ALLOC sections.
1438 (xtensa_create_xproperty_segments, section_has_xproperty): New.
1439 (add_xt_block_frags): Add end_property_function parameter and call it
1440 if it is non-zero. Call xtensa_frag_flags_init.
1441 (xtensa_frag_flags_is_empty, xtensa_frag_flags_init,
1442 get_frag_property_flags, frag_flags_to_number,
1443 xtensa_frag_flags_combinable, xt_block_aligned_size,
1444 xtensa_xt_block_combine, add_xt_prop_frags,
1445 init_op_placement_info_table, opcode_fits_format_slot,
1446 xg_get_single_size, xg_get_single_format): New.
1447 (istack_push): Inline call to tinsn_copy.
1448 (tinsn_copy): Delete.
1449 (tinsn_has_invalid_symbolic_operands): Handle O_hi16 and O_lo16 and
1450 CONST16 opcodes. Handle O_big, O_illegal, and O_absent.
1451 (tinsn_has_complex_operands): Handle O_hi16 and O_lo16.
1452 (tinsn_to_insnbuf): Use xg_get_single_format and new xtensa-isa.h
1453 functions. Handle invisible operands.
1454 (tinsn_to_slotbuf): New.
1455 (tinsn_check_arguments): Use new xtensa-isa.h functions.
1456 (tinsn_from_chars): Add slot parameter. Rewrite using xg_init_vinsn,
1457 vinsn_from_chars, and xg_free_vinsn.
1458 (tinsn_from_insnbuf): New.
1459 (tinsn_immed_from_frag): Add slot parameter and handle multislot
1460 instructions. Handle symbol differences.
1461 (get_num_stack_text_bytes): Use xg_get_single_size.
1462 (xg_init_vinsn, xg_clear_vinsn, vinsn_has_specific_opcodes,
1463 xg_free_vinsn, vinsn_to_insnbuf, vinsn_from_chars, expr_is_register,
1464 get_expr_register, set_expr_symbol_offset_diff): New.
1465 * config/tc-xtensa.h (MAX_SLOTS): Define.
1466 (xtensa_relax_statesE): Move from tc-xtensa.c. Add
1467 RELAX_CHECK_ALIGN_NEXT_OPCODE, RELAX_MAYBE_DESIRE_ALIGN, RELAX_SLOTS,
1468 RELAX_FILL_NOP, RELAX_UNREACHABLE, RELAX_MAYBE_UNREACHABLE, and
1470 (RELAX_IMMED_MAXSTEPS): Move from tc-xtensa.c.
1471 (xtensa_frag_type struct): Add is_assembly_state_set,
1472 use_absolute_literals, relax_seen, is_unreachable, is_specific_opcode,
1473 is_align, is_text_align, alignment, and is_first_loop_insn fields.
1474 Replace is_generics and is_relax fields by is_no_transform field.
1475 Delete is_text and is_longcalls fields. Change text_expansion and
1476 literal_expansion to arrays of MAX_SLOTS entries. Add arrays of
1477 per-slot information: literal_frags, slot_subtypes, slot_symbols,
1478 slot_sub_symbols, and slot_offsets. Add fr_prev field.
1479 (xtensa_fix_data struct): New.
1480 (xtensa_symfield_type struct): Delete plt field.
1481 (xtensa_block_info struct): Move definition to tc-xtensa.h. Add
1482 forward declaration here.
1483 (xt_section_type enum): Delete xt_insn_sec. Add xt_prop_sec.
1484 (XTENSA_SECTION_RENAME): Undefine.
1485 (TC_FIX_TYPE, TC_INIT_FIX_DATA, TC_FORCE_RELOCATION, NO_PSEUDO_DOT,
1486 tc_unrecognized_line, md_do_align, md_elf_section_change_hook,
1487 HANDLE_ALIGN, TC_LINKRELAX_FIXUP, SUB_SEGMENT_ALIGN): Define.
1488 (TC_CONS_FIX_NEW, tc_symbol_new_hook): Delete.
1489 (unit_num_copies_func, opcode_num_units_func,
1490 opcode_funcUnit_use_unit_func, opcode_funcUnit_use_stage_func): New.
1491 (resource_table struct): New.
1492 * config/xtensa-istack.h (MAX_INSN_ARGS): Increase from 6 to 10.
1493 (TInsn struct): Add keep_wide, loc, fixup, record_fix, subtype,
1494 literal_space, symbol, sub_symbol, offset, and literal_frag fields.
1495 (tinsn_copy): Delete prototype.
1496 (vliw_insn struct): New.
1497 * config/xtensa-relax.c (insn_pattern_struct): Add options field.
1498 (widen_spec_list): Add option conditions for density and boolean
1499 instructions. Add expansions using CONST16 and conditions for using
1500 CONST16 vs. L32R. Use new Xtensa LX RSR/WSR syntax. Add entries for
1502 (simplify_spec_list): Add option conditions for density instructions.
1503 Add entry for NOP instruction.
1504 (append_transition): Add cmp function pointer parameter and use it to
1505 insert the new entry in order.
1506 (operand_function_LOW16U, operand_function_HI16U): New.
1507 (xg_has_userdef_op_fn, xg_apply_userdef_op_fn): Handle
1508 OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
1509 (enter_opname, split_string): Use xstrdup instead of strdup.
1510 (init_insn_pattern): Initialize new options field.
1511 (clear_req_or_option_list, clear_req_option_list,
1512 clone_req_or_option_list, clone_req_option_list, parse_option_cond):
1514 (parse_insn_pattern): Parse option conditions.
1515 (transition_applies): New.
1516 (build_transition): Use new xtensa-isa.h functions. Fix incorrectly
1517 swapped last arguments in calls to append_constant_value_condition.
1518 Call clone_req_option_list. Add warning about invalid opcode.
1519 Handle LOW16U and HI16U function names.
1520 (build_transition_table): Add cmp parameter and use it in calls to
1521 append_transition. Use new xtensa-isa.h functions. Check
1522 transition_applies before adding entries.
1523 (xg_build_widen_table, xg_build_simplify_table): Add cmp parameter and
1524 pass it through to build_transition_table.
1525 * config/xtensa-relax.h (ReqOrOptionList, ReqOrOption, ReqOptionList,
1526 ReqOption, transition_cmp_fn): New types.
1527 (OpType enum): Add OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
1528 (transition_rule struct): Add options field.
1529 * doc/as.texinfo (Overview): Update Xtensa options.
1530 * doc/c-xtensa.texi (Xtensa Options): Delete --[no-]density,
1531 --[no-]relax, and --[no-]generics options. Update descriptions of
1532 --text-section-literals and --[no-]longcalls. Add
1533 --[no-]absolute-literals and --[no-]transform.
1534 (Xtensa Syntax): Add description of syntax for FLIX instructions.
1535 Remove use of "generic" and "specific" terminology for opcodes.
1536 (Xtensa Registers): Generalize the syntax description to include
1537 user-defined register files.
1538 (Xtensa Automatic Alignment): Update.
1539 (Xtensa Branch Relaxation): Mention limitation of unconditional jumps.
1540 (Xtensa Call Relaxation): Linker can now remove most of the overhead.
1541 (Xtensa Directives): Remove confusing rules about precedence.
1542 (Density Directive, Relax Directive): Delete.
1543 (Schedule Directive): New.
1544 (Generics Directive): Rename to ...
1545 (Transform Directive): ... this node.
1546 (Literal Directive): Update for absolute literals. Missing
1547 literal_position directive is now an error.
1548 (Literal Position Directive): Update for absolute literals.
1549 (Freeregs Directive): Delete.
1550 (Absolute Literals Directive): New.
1551 (Frame Directive): Minor editing.
1552 * Makefile.am (DEPTC_xtensa_elf, DEPOBJ_xtensa_elf, DEP_xtensa_elf):
1553 Update dependencies.
1554 * Makefile.in: Regenerate.
1558 * config/tc-mips.c (append_insn): Use fix_new rather than fix_new_exp
1559 to build the second and third fixups for a composite relocation.
1560 (macro_read_relocs): New function.
1561 (macro_build): Use it.
1562 (s_cpsetup): Pass all three composite relocation codes to macro_build.
1563 Simplify fragging code accordingly.
1564 (s_gpdword): Use fix_new rather than fix_new_exp for the second part
1565 of the composite relocation. Set fx_tcbit in both fixups.
1569 * config/tc-mips.c (append_insn): Set fx_tcbit for composite relocs.
1570 (md_apply_fix3): Don't treat composite relocs as done.
1574 * macro.c (macro_expand_body): When ELF, use .LL rather than LL as
1575 prefix for symbol names generated from the LOCAL macro directive.
1577 * dw2gencfi.c (select_cie_for_fde): When separating CIE out from
1578 FDE, treat a DW_CFA_remember_state as we do a DW_CFA_advance_loc.
1582 * config/tc-crx.c (preprocess_reglist): Handle Co-processor
1584 (md_assemble): Add error checking for Co-Processor instructions.
1585 (get_cinv_parameters): Add 'b' option to invalidate the
1586 branch-target cache.
1590 * config/tc-arm.c (unwind): New variable.
1591 (vfp_sp_encode_reg): New function.
1592 (vfp_sp_reg_required_here): Use it.
1593 (vfp_sp_reg_list, vfp_dp_reg_list): Remove.
1594 (vfp_parse_reg_list): New function.
1595 (s_arm_unwind_fnstart, s_arm_unwind_fnend, s_arm_unwind_cantunwind,
1596 s_arm_unwind_personality, s_arm_unwind_personalityindex,
1597 s_arm_unwind_handlerdata, s_arm_unwind_save, s_arm_unwind_movsp,
1598 s_arm_unwind_pad, s_arm_unwind_setfp, s_arm_unwind_raw): New
1600 (md_pseudo_table): Add them.
1601 (do_vfp_reg2_from_sp2): Use vfp_parse_reg_list and vfp_sp_encode_reg.
1602 (do_vfp_sp2_from_reg2, vfp_sp_ldstm, vfp_dp_ldstm): Ditto.
1603 (set_section, add_unwind_adjustsp, flush_pending_unwind,
1604 finish_unwind_opcodes, start_unwind_section, create_unwind_entry,
1605 require_hashconst, add_unwind_opcode): New functions.
1606 * doc/tc-arm.text: Document unwinding opcodes.
1607 * NEWS: Mention the new feature.
1611 * config/tc-mips.c (md_apply_fix3): Remove erroneous assert.
1615 * config/tc-ppc.c (md_apply_fix3): Call S_SET_THREAD_LOCAL for
1617 * config/tc-s390.c (md_apply_fix3): Likewise.
1618 * config/tc-sparc.c (md_apply_fix3): Likewise.
1622 * config/tc-arm.c (arm_elf_section_type): New function.
1623 (arm_elf_change_section): Set section link for exidx sections.
1624 * config/tc-arm.h (arm_elf_section_type): Add prototype.
1625 (md_elf_section_type): Define.
1629 * config/tc-pdp11.c (md_apply_fix3): Change to sign of the SOB
1630 instruction's offset.
1634 * (TARGET_FORMAT): Remove LynxOS COFF definition.
1638 * config/tc-arc.c (tc_gen_reloc): Don't assume fixP->fx_addsy is an
1639 asymbol *, instead use symbol_get_bfdsym.
1643 * config/tc-m68k.c (select_control_regs): Add mcf5249.
1647 * config/tc-arm.c (do_smi, do_nop): New functions.
1648 (insns): Add ARMv6ZK instructions.
1649 (md_apply_fix3): Handle BFD_RELOC_ARM_SMI.
1650 (tc_gen_reloc): Ditto.
1651 (arm_cpus): Add mpcore and arm1176.
1652 (arm_archs): Add armv6{k,z,zk}.
1653 * doc/c-arm.texi: Document new cores and architectures.
1657 * config/tc-arm.c: Use ISO C90 formatting.
1661 * config/tc-arm.c (mav_reg_required_here): Allow REG_TYPE_CN
1662 as alternative when REG_TYPE_MVF, REG_TYPE_MVD, REG_TYPE_MVFX or
1663 REG_TYPE_MVDX is expected.
1667 * doc/c-i386.texi (i386-Mnemonics): Fix typo.
1671 * config/tc-ia64.c (ENCODED_PSP_OFFSET): New.
1672 (output_rp_psprel, output_pfs_psprel, output_preds_psprel,
1673 output_spill_base, output_unat_psprel, output_lc_psprel,
1674 output_fpsr_psprel, output_priunat_psprel, output_bsp_psprel,
1675 output_bsprestore_psprel, output_rnat_psprel, output_spill_psprel,
1676 output_spill_psprel_p): Use it.
1680 * config/tc-crx.c (handle_LoadStor): New function.
1681 Handle load/stor unique instructions before parsing.
1685 * config/tc-arm.c (s_arm_rel31): New funciton.
1686 (md_pseudo_table): Add .rel31.
1687 (md_apply_fix3): Handle BFD_RELOC_ARM_TARGET2,
1688 BFD_RELOC_32_PCREL and BFD_RELOC_ARM_PREL31.
1689 (tc_gen_reloc): Handle BFD_RELOC_ARM_PREL31 and BFD_RELOC_ARM_TARGET2.
1690 (arm_fix_adjustable): Return 0 for BFD_RELOC_ARM_TARGET2.
1691 (arm_parse_reloc): Add (target2).
1695 * Makefile.am: Run "make dep-am".
1696 * Makefile.in: Regenerate.
1697 * aclocal.m4: Regenerate.
1698 * configure: Regenerate.
1699 * doc/Makefile.in: Regenerate.
1700 * po/POTFILES.in: Regenerate.
1701 * po/gas.pot: Regenerate.
1705 * config/tc-mmix.c [!LLONG_MIN]: Correct #elsif to #elif.
1706 [!LLONG_MAX]: Ditto.
1710 * config/tc-arm.c: Rename RELABS to TARGET1.
1714 * messages.c (as_internal_value_out_of_range): Cast values passed
1715 to as_bad_where or as_warn_where to proper type.
1719 * config/tc-avr.c: Add support for
1720 atmega48, atmega88, atmega168, attiny13, attiny2313, at90can128.
1724 * dw2gencfi.c (select_cie_for_fde): When separating CIE out
1725 from FDE, treat a CFI_escape as we do a DW_CFA_advance_loc.
1729 * config/obj-elf.c (obj_elf_section_type): Handle init_array,
1730 fini_array and preinit_array section types.
1731 * config/tc-ia64.c (ia64_elf_section_type): Remove init_array
1733 * doc/as.texinfo: Document extra section types.
1737 * Makefile.am (TARG_ENV_HFILES): Add te-symbian.h.
1738 * Makefile.in: Regenerated.
1739 * configure.in: Set em for arm*-*-symbianelf*.
1740 * configure: Regenerated.
1741 * config/tc-arm.c (elf32_arm_target_format): Use Symbian target
1742 vectors when appropriate.
1743 * config/te-symbian.h: New file.
1747 * config/tc-crx.c (gettrap): Exception vector can be case
1749 (process_label_constant): Fix a 32-bit displacement bug in branch
1751 (get_operandtype) : Bug fix, wrong operand was used.
1752 (process_label_constant): Initialize relocation type to
1757 * tc-arm.c (arm_cpus, arm_fpus): Allow <cpu>-s as well as <cpu>s
1758 for synthesizable cores.
1760 * doc/c-arm.texi (ARM Options): Document canonical names of CPUs.
1764 * config/tc-msp430.c: Clean-up the code.
1765 (md_relax_table): New relax table.
1766 (mcu_types): Sort MCU types.
1767 (md_pseudo_table): Add .profiler pseudo handler.
1768 (pow2value): New function.
1769 (msp430_profiler): New function.
1770 (msp430_operands): Add new insns handlers.
1771 (msp430_srcoperand): Add register operand handler, allow complex
1773 (md_estimate_size_before_relax): Rewritten.
1774 (md_convert_frag): Rewritten.
1775 (msp430_relax_frag): New function.
1776 * config/tc-msp430.h (md_relax_frag): define macro
1777 * doc/c-msp430.texi: Update information.
1781 * as.c (std_shortopts): Allow -g to take an optional argument.
1782 (parse_args): Pass any switch starting with -g on to the backend
1787 * configure.in (arm*-*-symbianelf*): New target.
1788 (arm*-*-eabi*): Likewise.
1789 * configure: Regenerated.
1792 * config/tc-mips.c (append_insn): Handle delay slots in branch likely
1797 * config/tc-ia64.c (start_unwind_section): Add linkonce_empty
1798 argument, don't do anything if current section is not
1799 .gnu.linkonce.t.* and linkonce_empty is set.
1800 (generate_unwind_image, dot_endp): Adjust callers, call
1801 start_unwind_section (*, 1) if nothing will be put into the
1806 * as.c (MD_DEBUG_FORMAT_SELECTOR): Provide default definition.
1807 (show_usage): Add -g.
1808 (std_longopts): Add --gen-debug. Alpha sort the table.
1809 (parse_args): Print an error message if a switch is not handled.
1810 Handle the -g switch, calling md_debug_format_selector() if
1812 * NEWS: Mention new feature.
1813 * doc/as.texinfo: Document new switch.
1814 * doc/internals.texi: Document behaviour of md_parse_option.
1816 * config/tc-arm.c (md_parse_option): Do not issue an error message
1817 if the switch is not recognised.
1818 * config/tc-m68k.c (md_parse_option): Likewise.
1819 * config/tc-pdp11.c (md_parse_option): Likewise.
1820 * config/tc-v850.c (md_parse_option): Likewise.
1822 * as.h: Fix up formatting.
1827 * macro.c (macro_set_alternate): Use ISO C90 formatting.
1829 * configure.in: Sort architecture based tables alphabetically.
1830 * configure: Regenerate.
1834 * config/tc-ppc.c (tc_ppc_regname_to_dw2regnum <regnames>): Replace
1835 { "cc", 68 }, with { "cr", 70 }.
1840 * as.c: Add and handle new --alternate command line option.
1841 * macro.c (macro_set_alternate): New.
1842 * macro.h (macro_set_alternate): Declare.
1843 * read.c: Add and handle new .altmacro and .noaltmacro directives.
1844 * doc/as.texinfo: Document new command line option and pseudo-ops
1845 as well as insert documentation originating from gasp about
1846 alternate macro syntax.
1847 * NEWS: Mention new command line option and pseudo-ops.
1851 * expr.c (operand): Handle the "~", "-", and "!" operators applied
1856 * config/tc-arm.c (md_apply_fix3, tc_gen_reloc, arm_parse_reloc):
1857 Handle new relocations.
1858 * include/elf/arm.h (elf_arm_reloc_type): Add new EABI relocations.
1862 * write.c (relax_segment): Use was_address instead of address when
1863 setting fr_fix field for align frag due to backwards .org.
1867 Introduce SH2a support.
1869 * config/tc-sh.c (get_specific): Change arch_sh2a_up to
1872 * config/tc-sh.c (md_parse_option): Add sh2a-nofpu ISA handling.
1874 * config/tc-sh.c (sh_elf_final_processing): Move sh2a recognition
1875 to end of conditional expression.
1877 * config/tc-sh.c: Add sh2a-nofpu support.
1879 * tc-sh.c: Add sh2a support.
1880 (parse_reg): Add tbr.
1881 (parse_at): Support @@(disp,tbr).
1882 (get_specific): Support sh2a opcodes.
1883 (insert4): New, for 4 byte relocs.
1884 (build_Mytes): Support sh2a opcodes.
1885 (md_apply_fix3_Mytes): Support sh2a opcodes.
1887 * config/tc-sh.c (md_parse_option): Handle sh2a.
1888 (sh_elf_final_processing): Ditto.
1892 * config/tc-hppa.h (TARGET_FORMAT): Set to "elf32-hppa-netbsd"
1897 * config/tc-ppc.c (ppc_frob_file_before_adjust): Warn if .toc too big.
1898 (ppc_arch): Expand comment.
1902 * config/tc-crx.c: Support evaluating the difference between two
1904 * config/tc-crx.h: Likewise.
1908 * config/tc-ia64.c (start_unwind_section): Set the linked-to
1910 (ia64_elf_section_change_hook): Set the linked-to section for
1915 * config/tc-msp430.c: Add new subtargets: msp430x1610,
1916 msp430x1611, msp430x1612, msp430x415, msp430x417, msp430xG437,
1917 msp430xG438, msp430xG439.
1921 * doc/as.texinfo (Section, PushSection): Correct documentation
1926 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
1927 permissible for the selected instruction suffix.
1928 (match_template): Don't permit 64-bit general purpose operands in
1930 (finalize_imm): Permit 64-bit immediates.
1931 (build_modrm_byte): Don't treat 32-bit addressing in 64-bit mode
1932 specially except for the width of the used base and/or index
1933 registers. For 32-bit displacements, use sign-extended
1934 relocations only when using 64-bit addressing.
1935 Force zero displacement on rip-relative addressing when there is
1936 no other displacement.
1937 (i386_index_check): Don't treat 32-bit addressing in 64-bit mode
1938 specially except for the width of the used base and/or index
1940 (parse_register): Disallow Reg64 registers in 32-bit mode.
1942 * config/tc-i386.c: For DefaultSize instructions, don't guess a 'q'
1943 suffix if the instruction doesn't support it.
1947 * config/tc-mips.c (append_insn): Handle constant expressions with
1948 no associated relocation.
1949 (mips_ip): Cancel the expression after use for the Q format
1951 (parse_relocation): Return no relocation for unsupported
1953 (my_getSmallExpression): Return no relocation if no relocation
1958 * config/obj-som.c (adjust_stab_sections): Add prototype.
1959 (obj_som_compiler, obj_som_version, obj_som_copyright,
1960 adjust_stab_sections): Add ATTRIBUTE_UNUSED to unused arguments.
1961 * config/tc-hppa.c (update_subspace): Likewise.
1962 (is_defined_subspace): Amplify comment.
1963 * config/obj-som.h (som_frob_file): Add prototype.
1968 * subsegs.c (section_symbol): Don't create a new segment when
1969 existing segment is undefined.
1973 * config/tc-arm.c: Include include/opcode/arm.h.
1974 (ARM_EXT_*, ARM_ARCH_*, ARM_ANY, ARM_ALL, COPROC_ANY): Delete.
1975 (FPU_FPA_EXT_* FPU_VFP_EXT_*, FPU_ANY, FPU_NONE, FPU_MAVERICK): Delete.
1976 (FPU_ARCH_*): Delete.
1977 * Makefile.am: Update dependencies.
1978 * Makefile.in: Regenerate.
1982 * configure.in: Accept armbe as a big-endian arm configuration.
1983 * configure: Regenerate.
1987 * config/tc-i386.c (T_SHIFTOP): New constant.
1988 (intel_e05_1): Handle '&', '|' and T_SHIFTOP.
1989 (intel_el1): Handle '~'.
1990 (intel_get_token): Handle '<>', '&', '|' and '~'.
1994 (md_assemble): Remove spurious newline from end of as_bad error
1996 (intel_e05_1): Likewise.
1997 (intel_e11): Likewise.
1998 (intel_match_token): Likewise.
2002 * config/tc-m68k.c: Convert to C90. Remove redundant
2003 declarations. Indentation fixup.
2004 [M68KCOFF]: Include "obj-coff.h" instead of declaring
2005 obj_coff_section ourselves.
2009 * config/tc-ia64.c (default_big_endian): New.
2010 (dot_byteorder, md_begin): Use it.
2011 (md_parse_option): Set it.
2015 * configure.in: Change sh-sybmian-elf to sh-*-symbianelf.
2016 * configure: Regenerate.
2017 * NEWS: Change sh-sybmian-elf to sh-*-symbianelf.
2018 * config/tc-sh.c (sh_elf_final_processing): Use renamed version of
2019 sh_find_elf_flags if necessary.
2023 * config/tc-mips.c (mips_fix_adjustable): If the full addend is
2024 going to be split into more than one in-place addend, return 0
2025 for relocations against mergeable sections. Associate comments
2030 * Makefile.am (CPU_TYPES): Add crx.
2031 (TARGET_CPU_CFILES): Add config/tc-crx.c.
2032 (TARGET_CPU_HFILES): Add config/tc-crx.h.
2033 (DEPTC_crx_elf): New target.
2034 (DEPOBJ_crx_elf): Likewise.
2035 (DEP_crx_elf): Likewise.
2036 * Makefile.in: Regenerate.
2037 * configure.in: Add crx* target.
2038 * configure: Regenerate.
2039 * config/tc-crx.c: New file.
2040 * config/tc-crx.h: New file.
2041 * NEWS: Mention new target.
2045 * config.in: Undefine TARGET_SYMBIAN by default.
2047 * configure: Regenerate. Add sh-symbian-elf target. If
2048 selected define TARGET_SYMBIAN.
2049 * config/tc-sh.h (TARGET_FORMAT): Select a Symbian target
2050 format if TARGET_SYMBIAN has been defined.
2052 * output-file.c (output_file_create): Report the target format
2053 chosen when bfd_openw reports that it is invalid.
2055 * config/obj-coff.c (coff_pseudo_table): Only define the weak
2056 pseudo for BFD based assemblers.
2061 * config/tc-sh.c (md_assemble): Change isspace to ISSPACE.
2062 (md_parse_option): Remove redundant -isa testing.
2063 Make bfd_arch variable const.
2064 (md_show_usage): Make bfd_arch variable const.
2068 * config/tc-ia64.c (emit_one_bundle): Check and set insn_addr.
2069 * config/tc-ia64.h (md_frag_check): Define.
2073 * config/obj-coff.c (obj_coff_weak): New .weak syntax for PE weak
2075 * doc/as.texinfo (Weak): Document PE weak symbols.
2079 * config/tc-mips.c (HAVE_IN_PLACE_ADDENDS): New macro.
2080 (reloc_needs_lo_p): Only return true if HAVE_IN_PLACE_ADDENDS.
2081 (mips_frob_file): Rework so that only a single pass through the
2082 relocs is needed. Allow %lo()s to have higher offsets than their
2083 corresponding %hi()s or %got()s.
2087 * config/tc-arm.c (md_apply_fix3:BFD_RELOC_ARM_IMMEDIATE): Do not
2088 allow values which have come from undefined symbols.
2089 Always consider this fixup to have been processed as a reloc
2090 cannot be generated for it.
2094 * frags.h (struct frag): Add has_code and insn_addr fields.
2095 * write.c (cvt_frag_to_fill): Invoke md_frag_check.
2096 * config/tc-ppc.c (md_assemble): Check and set insn_addr.
2097 * config/tc-ppc.h (md_frag_check): Define.
2101 * doc/Makefile.am (info): Rename goal to...
2102 (info-local): ... this, to preserve implicit dependencies.
2103 * doc/Makefile.in: Regenerate with automake 1.8.5.
2107 * config/tc-m32r.c (md_convert_frag): Changed for @PLT.
2108 (m32r_cgen_record_fixup_exp): Changed for @GOTOFF, @GOT.
2109 (m32r_fix_adjustable): Changed for @GOTOFF, @GOT, @PLT.
2110 (tc_gen_reloc): Likewise.
2111 (m32r_end_of_match): Add for @GOTOFF, @GOT, @PLT.
2112 (m32r_parse_name): Likewise.
2113 (m32r_cgen_parse_fix_exp): Likewise.
2114 * config/tc-m32r.h (md_parse_name): Define for @GOTOFF, @GOT, @PLT.
2115 (O_PIC_reloc): Likewise.
2116 (TC_CGEN_PARSE_FIX_EXP): Likewise..
2117 * cgen.c (gas_cgen_parse_operand): Add TC_CGEN_PARSE_FIX_EXP
2118 for @GOTOFF, @GOT, @PLT.
2122 * gas/symbols.c: While discarding ordinary local absolute symbols
2123 when --strip-local-absolute is in effect, retain file symbols.
2127 * config/tc-m68k.c (mri_chip): Replace current_chip, not augment.
2128 (md_parse_option): Likewise.
2132 * config/tc-i386.c: Deal with LEX_QM the same way as with LEX_AT.
2133 * config/te-netware.h: New file.
2134 * config/te-ppcnw.h: Delete: Obsolete.
2135 * configure.in: Eliminate ill NetWare targets. Make generic
2136 NetWare target use proper emulation.
2137 * Makefile.am: Eliminate reference to obsolete te-ppcnw.h, add
2138 reference to new te-netware.h.
2139 * configure: Regenerate.
2140 * Makefile.in: Regenerate.
2144 * config/tc-s390.c (s390_insn): Avoid incorrect signed/unsigned
2145 comparison in .insn pseudo operation.
2149 * config/obj-coff.c (coff_adjust_section_syms): Use
2150 bfd_get_section_size instead of bfd_get_section_size_before_reloc.
2151 (coff_frob_section): Likewise.
2152 * config/tc-mips.c (md_apply_fix3): Likewise.
2153 * config/obj-elf.c (elf_frob_file): Use bfd_set_section_size.
2154 (elf_frob_file_after_relocs): Likewise.
2158 * config/tc-hppa.c (log2): Rename to exact_log2.
2159 (pa_next_subseg): Delete unused function.
2160 (create_new_space): Mark unused arguments with ATTRIBUTE_UNUSED.
2161 (create_new_subspace): Likewise.
2164 * config/tc-hppa.c (hppa_fix_adjustable): Allow reduction of fake
2165 labels. Fix warning.
2169 * config/tc-mn10300.h (tc_fix_adjustable): Define.
2170 * config/tc-mn10300.c (mn10300_fix_adjustable): Don't adjust debug
2171 or non-merged symbols.
2175 * config/tc-ia64.c (remove_marked_resource): Save, clear and
2176 restore the old slot when inserting srlz.i/srlz.d.
2180 * Makefile.am: Regenerate dependecies.
2181 * Makefile.in: Regenerate.
2182 * config/tc-sh.c (valid_arch): Make unsigned.
2183 (preset_target_arch): Likewise.
2184 (md_begin): Use new architecture flags system.
2185 (get_specific): Likewise.
2186 (assemble_ppi): Likewise.
2187 (md_assemble): Likewise. Also fix error check for bad opcodes.
2188 (md_parse_option): Likewise. Also generate -isa values according
2189 to the table in bfd/cpu-sh.c instead of just constants. Also
2190 allow <arch>-up ISA variants.
2191 (sh_elf_final_processing): Replace if-else chain with a call to
2192 sh_find_elf_flags().
2193 * testsuite/gas/sh/arch: New directory.
2194 * testsuite/gas/sh/arch/arch.exp: New test script.
2195 * testsuite/gas/sh/arch/arch_expected.txt: New file.
2196 * testsuite/gas/sh/arch/sh.s: New file.
2197 * testsuite/gas/sh/arch/sh2.s: New file.
2198 * testsuite/gas/sh/arch/sh-dsp.s: New file.
2199 * testsuite/gas/sh/arch/sh2e.s: New file.
2200 * testsuite/gas/sh/arch/sh3-nommu.s: New file.
2201 * testsuite/gas/sh/arch/sh3.s: New file.
2202 * testsuite/gas/sh/arch/sh3-dsp.s: New file.
2203 * testsuite/gas/sh/arch/sh3e.s: New file.
2204 * testsuite/gas/sh/arch/sh4-nommu-nofpu.s: New file.
2205 * testsuite/gas/sh/arch/sh4-nofpu.s: New file.
2206 * testsuite/gas/sh/arch/sh4.s: New file.
2207 * testsuite/gas/sh/arch/sh4a-nofpu.s: New file.
2208 * testsuite/gas/sh/arch/sh4al-dsp.s: New file.
2209 * testsuite/gas/sh/arch/sh4a.s: New file.
2213 * config/gc-m68k.c(m68k_ip): Convert mode 5 addressing
2214 with zero offset into mode 2 addressing to save a word.
2218 * config/tc-ia64.c (ar_is_in_integer_unit): Removed.
2219 (ar_is_only_in_integer_unit): New.
2220 (ar_is_only_in_memory_unit): New.
2221 (generate_unwind_image): Silence gcc on 32bit host.
2222 (md_assemble): Use ar_is_only_in_integer_unit instead of
2223 ar_is_in_integer_unit. Check AR access.
2227 * config/tc-m68k.c (md_begin): Sort the opcode table into
2229 (m68k_compare_opcode): New function to do the sorting.
2233 * config/m68k-parse.y(operand): Allow for MAC/EMAC mask
2234 addressing on MIT style operands.
2235 * config/m68k-parse.y(yylex): Allow '-&' for predecrement
2237 * config/tc-m68k.c(install_operand): Comment 'G' and 'H' type
2242 * expr.c (operand, operator): Don't reject '++' and '--'.
2246 * config/tc-mips.c (append_insn): Use ISA-encoded addresses in MIPS16
2251 * configure.in: Add ppc-*-lynxos*. Update i386-*-lynxos* to ELF.
2252 * configure: Regenerate.
2256 * dw2gencfi.c (output_cie): Handle dwarf3 format CIE entries.
2260 * configure.in (or32-*-rtems*): Switch to elf.
2261 * configure: Regenerate.
2265 * po/fr.po: Updated French translation.
2269 * doc/as.texinfo (Section): Document G and T flags to .section
2270 directive. Document the extra arguments that the G flag
2271 requires. Document the #tls flag.
2275 * subsegs.c (section_symbol): Create a new section symbol if
2276 the existing one doesn't match.
2277 * symbols.c (symbol_set_bfdsym): Don't reset BFD section symbol.
2281 * config/tc-mips.c (append_insn, mips_emit_delays): Extend -mfix-vr4120
2282 to cope with VR4181A errata MD(1) and MD(4).
2286 * NEWS: Mention .secrel32 for pe[i]-i386.
2290 * config/tc-frv.h (MAX_MEM_FOR_RS_ALIGN_CODE): New.
2291 (HANDLE_ALIGN): New.
2295 * Makefile.am (DIST_SUBDIRS): Define.
2296 * aclocal.m4: Regenerate with automake 1.8.4.
2297 * Makefile.in: Likewise.
2298 * doc/Makefile.in: Likewise.
2302 * config/tc-ia64.c (dot_serialize): Declare.
2303 (dot_serialize): New function.
2304 (md_pseudo_table): Add ".serialize.data" and
2305 ".serialize.instruction" directives.
2309 * messages (as_internal_value_out_of_range): Print a message about
2310 a value being out of range. Be consistent about whether the
2311 values are printed in decimal or hexadecimal.
2312 (as_warn_value_out_of_range): Generate a warning message about an
2314 (as_bad_value_out_of_range): Generate an error message about an
2316 * as.h: Prototype the new functions.
2317 * config/tc-alpha.c (insert_operand): Use new function.
2318 * config/tc-arc.c (arc_insert_operand): Likewise.
2319 * config/tc-mn10200.c (mn10200_insert_operand): Likewise.
2320 * config/tc-mn10300.c (mn10300_insert_operand): Likewise.
2321 * config/tc-ppc.c (ppc_insert_operand): Likewise.
2322 * config/tc-s390.c (s390_insert_operand): Likewise.
2323 * config/tc-v850.c (v850_insert_operand): Likewise.
2327 * configure.in: Set em=linux for frv-*-*linux*.
2328 * configure: Rebuilt.
2329 * config/tc-frv.h (TARGET_FORMAT): Use elf32-frvfdpic if...
2330 (frv_md_fdpic_enabled): New.
2331 * config/tc-frv.c (frv_md_fdpic_enabled): New.
2332 (DEFAULT_FDPIC): New.
2333 (frv_flags): Use DEFAULT_FDPIC.
2334 (frv_pic_flag): Likewise.
2335 (OPTION_NOPIC): New.
2336 (md_longopts): Add -mnopic.
2337 (md_parse_option): Handle it.
2338 (md_show_usage): Add -mfdpic and -mnopic.
2342 * config/tc-m68k.c: Add find_cf_chip to print list of valid
2343 chips for invalid coldfire instructions, rename selectors
2344 for ColdFire sub-variants, add 521x,5249,547x,548x and aliases,
2345 add current_chip to track which chip is referred to(including save/restore),
2346 use current_chip to select control registers, not current_arch.
2347 (md_show_usage): Add new chips.
2348 * doc/c-m68k.texi: Document new command line switches.
2352 * tc-s390.h (md_do_align, HANDLE_ALIGN): Remove.
2353 (NOP_OPCODE): Define.
2354 (s390_align_code): Remove prototype.
2355 * tc-s390.c (s390_align_code): Remove.
2359 * config/tc-ia64.c (make_unw_section_name): Removed.
2360 (start_unwind_section): New function.
2361 (generate_unwind_image): Take const segT instead of const
2363 (dot_handlerdata): Adjusted.
2364 (dot_endp): Likewise.
2368 * config/obj-elf.c (obj_elf_change_section): Allow the
2369 ".note.GNU-stack" section has SHF_EXECINSTR.
2373 * config/obj-elf.c (get_section): Return bfd_boolean.
2374 (obj_elf_change_section): Call bfd_get_section_by_name_if
2375 instead of bfd_map_over_sections.
2379 * config/obj-elf.c (get_section): New function.
2380 (obj_elf_change_section): Support multiple sections with same
2385 * config/tc-arm.c (create_register_alias): Fix typo checking for
2386 case sensitive register aliases.
2387 (co_proc_number): Use error message string in all_reg_maps[]
2389 (cp_reg_required_here): Likewise.
2390 (fp_reg_required_here): Likewise.
2394 * dwarf2dbg.c (dwarf2_finish): Add SEC_DEBUGGING to section flags.
2398 * config/tc-mips.c (HAVE_32BIT_ADDRESSES, append_insn, macro_build)
2399 (load_address, macro, mips_ip, md_parse_option)
2400 (mips_force_relocation, mips_validate_fix, md_apply_fix3)
2401 (s_change_sec, pic_need_relax, tc_gen_reloc): Remove all
2402 embedded-PIC handling, and update comments.
2403 (SWITCH_TABLE): Remove.
2404 * config/tc-mips.h (DIFF_EXPR_OK): Delete.
2405 (enum mips_pic_level): Remove EMBEDDED_PIC.
2406 (EXTERN_FORCE_RELOC): Remove embedded-PIC handling.
2407 (TC_FORCE_RELOCATION): Update comment.
2408 * ecoff.c (ecoff_build_lineno): Add comment about some code that
2409 might be safe to remove now that MIPS embedded-PIC is gone.
2413 * config/obj-som.c (obj_som_init_stab_section): Add new arguments in
2414 call to obj_set_subsection_attributes.
2415 (obj_som_init_stab_section): Likewise.
2416 * config/tc-hppa.c (default_subspace_dict): Add comdat field.
2417 (pa_def_subspaces): Provide comdat default.
2418 (pa_subspace): Handle new "comdat" parameter. Set SEC_LINK_ONCE and
2419 not SEC_IS_COMMON if section is comdat, common or dup_common. Update
2420 calls to create_new_subspace and update_subspace to pass comdat flag.
2421 (create_new_subspace, update_subspace): Add new comdat argument. Use
2422 it in calls to obj_set_subsection_attributes.
2423 * doc/c-hppa.texi (.subspa, .nsubspa): Document new comdat parameter
2424 and use of comdat, common and dup_comm parameters.
2428 * config/obj-elf.c (obj_elf_change_section): Check if the old
2429 group name is NULL before comparison.
2433 * config/tc-mips.h (mips_dwarf2_addr_size): Prototype.
2437 * config/tc-mips.c (s_mipsset): Set default CPU type for .set mipsN.
2441 * config/tc-mips.c (md_longopts): Remove -membedded-pic option.
2442 (OPTION_MEMBEDDED_PIC): Remove.
2443 (OPTION_TRAP, OPTION_BREAK, OPTION_EB, OPTION_EL)
2444 (OPTION_FP32, OPTION_GP32, OPTION_CONSTRUCT_FLOATS)
2445 (OPTION_NO_CONSTRUCT_FLOATS, OPTIONS_FP64, OPTION_GP64)
2446 (OPTION_RELAX_BRANCH, OPTION_NO_RELAX_BRANCH)
2447 (OPTION_ELF_BASE): Renumber.
2448 (md_parse_option): Remove OPTION_MEMBEDDED_PIC handling.
2449 (md_show_usage): Remove mention of -membedded-pic.
2450 * doc/as.texinfo: Remove mention of -membedded-pic.
2454 * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Remove.
2455 * config/tc-mips.c (RDATA_SECTION_NAME, mips_target_format): Remove
2457 (md_begin, mips_ip, md_parse_option, s_change_sec, s_option,
2458 s_abicalls, nopic_need_relax, tc_gen_reloc): Remove uses of
2459 USE_GLOBAL_POINTER_OPT.
2463 * config/tc-mips.c (macro): One more use of load_delay_nop.
2467 * config/tc-mips.c (load_delay_nop): New function.
2468 (load_address, macro): Use load_delay_nop() to build a nop
2469 which can be omitted with gpr_interlocks.
2473 * config/tc-mips.c (hilo_interlocks, gpr_interlocks,
2474 cop_interlocks): Remove superfluous CPU entries.
2478 * config/tc-arm.c (mav_parse_offset): Value must be multiple of 4.
2482 * NEWS: Added support for EMAC instructions and MAC/EMAC
2484 * config/m68k-parse.h: Add ACC[123], ACCEXT{01,23}, MAC/EMAC
2485 scale factor tokens, trailing_ampersand to mark mask addressing
2486 for MAC/EMAC instructions.
2487 * config/m68k-parse.y: Add options_ampersand clause, '<<',
2489 (yylex): Handle '>', '<', and '&' following '+'.
2490 * config/tc-m68k.c: Set mcfmac/mcfemac on appropriate ColdFire
2491 architectures in archs[].
2492 (m68k-ip): Add '4', 'e', 'g', 'i', cases to handle mask addressing
2493 for MAC/EMAC instructions, ACC[0123], ACCEXT{01,23}, and '<<'/'>>'
2495 (m68k_ip): Handle trailing '&' on MAC/EMAC insns.
2496 (install_operand): Fix 'n' case, Add 'F', 'f', 'G', 'H', 'I', ']'
2498 Add EMAC operands to init_table[].
2502 * config/tc-arm.h (enum mstate): Move here, add MAP_UNDEFINED
2504 (TC_SEGMENT_INFO_TYPE): Define to enum mstate.
2505 * config/tc-arm.c (enum mstate): Delete from here.
2506 (mapping_state): Remove the static mapstate variable and instead
2507 store the state in the segment. This allows a per-section mapping
2508 state. Handle and ignore MAP_UNDEFINED states.
2509 (arm_elf_change_section): Get the current mapping state from the
2511 (s_ltorg): Set the mapping state to MAP_DATA.
2512 (arm_cleanup): Use arm_elf_change_section to get the mapping state
2513 for each pool as it is emitted.
2517 * config/tc-arm.h: Formatting tidy ups.
2521 * NEWS: Note that MIPS -membedded-pic option is deprecated.
2525 * config/tc-i386.h [TE_PE] (TC_CONS_FIX_NEW): Define.
2526 * config/tc-i386.c (md_pseudo_table) [TE_PE]: Add "secrel32".
2527 [TE_PE] (O_secrel): Define.
2528 [TE_PE] (x86_pe_cons_fix_new): New.
2529 [TE_PE] (pe_directive_secrel): Likewise.
2530 (tc_gen_reloc) [TE_PE]: Support BFD_RELOC_32_SECREL.
2534 * config/tc-mips.c (mips_dwarf2_addr_size): Revert part
2535 of previous patch for fix in gcc.
2539 * config/tc-xtensa.c (xg_assembler_literal): Fix a typo.
2543 * read.c (do_align): Call md_flush_pending_output, if defined.
2547 * expr.c (operand): Correct checks for ++ and --.
2551 * config/tc-generic.c: Add some comments.
2555 * doc/c-mips.texi (-m{no-,}fix-vr4120): Renamed from
2556 -{no-}mfix-vr4122-bugs.
2557 * config/tc-mips.c (mips_fix_vr4120): Renamed from mips_fix_4122_bugs.
2558 (append_insn, mips_emit_delays): Update accordingly.
2559 (OPTION_FIX_VR4120, OPTION_NO_FIX_VR4120): Renamed from *VR4122.
2560 (md_longopts): Change -{no-,}mfix-vr4122-bugs to -m{no-,}fix-vr4120.
2561 (md_parse_option): Update after above changes.
2562 (md_show_usage): Add -mfix-vr4120.
2566 * doc/as.texinfo (Sub-Sections): Conditionalize COFF-specific use
2567 of .section directive; add a reference to the ELF .subsection
2572 * config/tc-m32r.c (md_assemble): Fixed infinite loop bug
2577 * Makefile.am: Remove mips from aout targets.
2578 * Makefile.in: Regenerate.
2579 * configure.in: Remove mips-dec-bsd* target.
2580 * configure: Regenerate.
2585 * config/tc-ppc.c (ppc_elf_suffix): Add valid32 and valid64 fields
2586 to struct map_bfd. Adjust MAP macro, and define MAP32, MAP64.
2587 Update "mapping". Restrict some @ modifiers to 32 bit.
2592 * config/tc-sh.c (dont_adjust_reloc_32): New variable.
2593 (sh_fix_adjustable): Avoid adjusting BFD_RELOC_32 when
2594 dont_adjust_reloc_32 is set.
2595 (md_longopts): Add option -renesas.
2596 (md_parse_option, md_show_usage): Likewise.
2597 * doc/c-sh.texi: Likewise.
2601 * config/tc-dlx.c (md_assemble): set fx_no_overflow flag for
2602 hi16 and lo16 fixS structs.
2603 (md_assemble): generate bit_fixS for RELOC_DLX_LO16 in
2604 exactly the same way as for RELOC_DLX_REL16.
2605 (machine_ip): properly respect LO flag in the_insn and
2606 output RELOC_DLX_LO16 rather than RELOC_DLX_16.
2607 (md_apply_fix3): apply RELOC_DLX_LO16.
2611 Remove long-obsolete MPW support.
2612 * mpw-config.in, mpw-make.sed, mac-as.r: Remove files.
2613 * configure.in: Remove mention of ppc-*-mpw* config.
2614 * configure.in: Likewise.
2618 * config/tc-arm.c (meabi_flags): Make its use conditional upon
2619 OBJ_ELF being defined.
2623 * config/obj-aout.c (obj_aout_type): Remove #ifdef BFD_ASSEMBLER code.
2627 * config/tc-arm.c (meabi_flags): New variable.
2628 (arm_parse_eabi): New function.
2629 (md_begin): Set flags for EABI v3.
2631 (arm_long_opts): Add meabi.
2632 * doc/as.texinf <ARM>: Document -meabi.
2633 * doc/c-arm.texi: Ditto.
2637 * config/tc-xtensa.c (xtensa_post_relax_hook): Create literal
2638 tables even when use_literal_section flag is not set.
2642 * config/tc-sh.c: Remove trailing whitespace.
2646 * doc/c-cris.texi (CRIS-Opts): Document --no-mul-bug-abort,
2647 --mul-bug-abort and the default behavior.
2648 * config/tc-cris.c (cris_insn_kind): New member CRIS_INSN_MUL.
2649 (err_for_dangerous_mul_placement): New variable.
2650 (STATE_MUL, OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): New
2652 (md_cris_relax_table): Have placeholder for STATE_MUL.
2653 (md_longopts): New options --mul-bug-abort and --no-mul-bug-abort.
2654 (cris_relax_frag) <case ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: New
2656 (md_estimate_size_before_relax) <case ENCODE_RELAX (STATE_MUL,
2657 STATE_BYTE)>: Ditto.
2658 (md_convert_frag) <ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: Check
2659 alignment and position of this frag, emit error message if
2661 (md_assemble): For a multiply insn and when checking it,
2662 transform the current frag into a special frag for that purpose.
2663 (md_parse_option) <case OPTION_MULBUG_ABORT_OFF, case
2664 OPTION_MULBUG_ABORT_ON>: Handle new options.
2668 * config/tc-xtensa.c (mark_literal_frags): New function.
2669 (xtensa_move_literals): Call mark_literal_frags for all literal
2670 segments, including init and fini literal segments.
2671 (xtensa_post_relax_hook): Swap use of xt_insn_sec and xt_literal_sec.
2675 * tc-hppa.c (cons_fix_new_hppa): Check for PC relative base type.
2676 (pa_comm): Set BSF_OBJECT in symbol flags.
2680 * Makefile.am: Run "make dep-am".
2681 * Makefile.in: Regenerate.
2682 * doc/Makefile.in: Regenerate.
2683 * config.in: Regenerate.
2684 * po/gas.pot: Regenerate.
2688 * read.c (read_a_source_file): Use demand_empty_rest_of_line.
2689 (demand_empty_rest_of_line): Issue an error here.
2690 (ignore_rest_of_line): Silently skip to end.
2691 (demand_copy_string): Issue an error, not warning.
2693 * config/obj-elf.c (obj_elf_section_name): Likewise.
2694 (obj_elf_section): Likewise.
2695 * config/tc-arc.c (arc_extoper): Remove bogus NULL checks.
2696 (arc_extinst): Likewise.
2697 * config/tc-ia64.c (dot_saveb): Use demand_empty_rest_of_line.
2698 (dot_spill): Likewise.
2699 (dot_unwabi): Likewise.
2700 (dot_prologue): Likewise.
2702 * expr.c (operand): Reject ++ and --.
2703 (operator): Likewise.
2707 * config/tc-sh.c: Include dw2gencfi.h.
2708 (sh_cfi_frame_initial_instructions): New function.
2709 (sh_regname_to_dw2regnum): Likewise.
2710 * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Move to the end of
2712 (TARGET_USE_CFIPOP): Define.
2713 (tc_cfi_frame_initial_instructions): Likewise.
2714 (tc_regname_to_dw2regnum): Likewise.
2715 (DWARF2_DEFAULT_RETURN_COLUMN, DWARF2_CIE_DATA_ALIGNMENT): Likewise.
2716 * Makefile.am: Update dependencies.
2717 * Makefile.in: Regenerate.
2721 * configure.in: Switch sh-*-rtems* to ELF. Add sh-*-rtemscoff*.
2722 * configure: Regenerate.
2726 * read.c (s_leb128): Call md_flush_pending_output.
2730 * config/tc-i386.c (output_insn): Handle PadLock instructions.
2731 * config/tc-i386.h (CpuPadLock): New define.
2732 (CpuUnknownFlags): Added CpuPadLock.
2736 * doc/c-hppa.texi (HPPA Directives): Fix typo.
2740 * dw2gencfi.c (output_cie): Align length to 4 byte boundary.
2741 (cfi_finish): Likewise for fde.
2745 * config/tc-ia64.c (md_assemble): Properly handle NULL
2747 (ia64_handle_align): Don't abort if failed to add a stop bit.
2751 * Makefile.in: Regenerated.
2752 * aclocal.m4: Likewise.
2753 * configure: Likewise.
2754 * doc/Makefile.in: Likewise.
2758 * config/tc-ia64.c (dot_align): New.
2759 (ia64_do_align): Make it static.
2760 (md_pseudo_table): Use "dot_align" for "align".
2761 (ia64_md_do_align): Don't set align_frag here.
2762 (ia64_handle_align): Add a stop bit to the previous bundle if
2765 * config/tc-ia64.h (ia64_do_align): Removed.
2769 * config/tc-sh.c (md_parse_option): Add -isa=sh4-nofpu and
2770 -isa=sh4-nommu-nofpu options. Adjust help messages accordingly.
2771 (sh_elf_final_processing): Output BFD type sh4_nofpu if that is
2772 the most general type or the user specifically requested it.
2773 (md_assemble): Add a new error message for when an instruction
2774 is understood, but is not allowed due to an -isa option.
2778 * config/tc-ia64.c (align_frag): New.
2779 (md_assemble): Set the tc_frag_data field in align_frag for
2780 IA64_OPCODE_FIRST instructions.
2781 (ia64_md_do_align): Set align_frag.
2782 (ia64_handle_align): Add a stop bit if needed.
2784 * config/tc-ia64.h (TC_FRAG_TYPE): New.
2785 (TC_FRAG_INIT): New.
2789 * config/tc-frv.c (fr400_audio): New variable.
2790 (md_parse_option, md_show_usage): Add -mcpu=fr405 and -mcpu=fr450.
2791 (md_parse_option): Set fr400_audio for -mcpu=fr400 and -mcpu=fr405.
2792 (target_implements_insn_p): New function.
2793 (md_assemble): Report an error if the processor doesn't implement
2798 * config/tc-m32r.c (md_longopts): Added -no-bitinst option.
2799 (md_parse_option): Ditto.
2800 (OPTION_NO_SPECIAL_M32R): Added.
2801 (md_show_usage): Document it.
2802 (enable_speial_m32r): Changed a default value from 0 to 1.
2803 * doc/c-m32r.texi: Document the -no-bitinst option.
2807 * config/tc-sh.c (get_operand): Revert previous delta.
2808 (tc_gen_reloc): Check for an unknown reloc type before processing
2813 * config/tc-s390.c (s390_insn): Correct range check for opcode in
2814 .insn pseudo operation.
2818 * config/tc-sh.c (get_operand): In case of #Imm, check has been
2819 added for wrong syntax.
2823 * config/tc-mips.c (mips_dwarf2_addr_size): New.
2824 * config/tc-mips.h (DWARF2_ADDR_SIZE): Use.
2828 * config/tc-sh.c (build_Mytes): Add REG_N_D and REG_N_B01
2829 nibble types to assembler.
2833 * config/tc-iq2000.c: Add missing \n\ in multiline string literal.
2837 * config/tc-ia64.c (slot_index): New arg before_relax. Use instead of
2839 (fixup_unw_records): New arg before_relax. Pass to slot_index.
2840 (ia64_estimate_size_before_relax): New.
2841 (ia64_convert_frag): Pass 0 to fixup_unw_records. Add comment.
2842 (generate_unwind_image): Pass 1 to fixup_unw_records.
2843 * config/tc-ia64.h (ia64_estimate_size_before_relax): Declare.
2844 (md_estimate_size_before_relax): Call ia64_estimate_size_before_relax.
2848 * stabs.c (generate_asm_file): Avoid warning about use of
2849 uninitialized variable.
2853 * config/tc-ia64.c (ia64_flush_insns): In addition to prologue,
2854 body, and endp, allow unwind records which do not have a "t"
2855 (time/instruction) field.
2859 * config/tc-arm.c (do_mav_dspsc_1): Correct offset of CRn.
2860 (do_mav_dspsc_2): Likewise.
2861 Fix accumulator registers move opcodes.
2866 * dwarf2dbg.c (get_filenum): Do not read beyond allocated memory.
2870 * config/tc-ia64.h (ia64_frob_symbol): New declaration.
2871 (tc_frob_symbol): New macro definition.
2872 * config/tc-ia64.c (ia64_frob_symbol): New routine.
2876 * config/tc-arm.c (md_begin): Mark .note.gnu.arm.ident as
2881 * read.h (IGNORE_OPCODE_CASE): Do not define. Replace with ...
2882 (TC_CASE_SENSITIVE): ... this.
2883 * read.c: Replace IGNORE_OPCODE_CASE with TC_CASE_SENSITIVE.
2884 * doc/internals.texi (TC_CASE_SENSITIVE): Document.
2888 * config/tc-ia64.c (dot_endp): Delete call to output_endp.
2889 (generate_unwind_image): Re-add it here.
2893 * dwarf2dbg.c (DWARF2_ADDR_SIZE): Remove trailing ';'
2894 * read.h (SKIP_WHITESPACE): Turn into an expression.
2895 * read.c (read_a_source_file): A pseudo is removed by having a
2900 * config/tc-ia64.c (output_endp): New.
2901 (count_bits): Delete.
2902 (ia64_flush_insns, process_one_record, optimize_unw_records): Handle
2903 endp unwind records.
2904 (fixup_unw_records): Handle endp unwind records. Delete code for
2905 shortening prologue regions not followed by a body record.
2906 (dot_endp): Call add_unwind_entry to emit endp unwind record.
2907 * config/tc-ia64.h (unw_record_type): Add endp.
2911 * config/tc-ia64.c (ia64_convert_frag): Call md_number_to_chars to
2912 fill padding bytes with zeroes.
2913 (emit_one_bundle): New locals last_ptr, end_ptr. Rewrite code that
2914 sets unwind_record slot_number and slot_frag fields.
2918 * config/tc-mips.c (add_got_offset_hilo): New function.
2919 (macro): Use load_register() and add_got_offset_hilo() to load
2920 constants instead of hardcoding code sequences throughout.
2924 * config/tc-ia64.c (emit_one_bundle): Add proper indentation.
2928 * config/tc-m68k.h (EXTERN_FORCE_RELOC): Handle m68k-uclinux specially,
2930 * config/tc-m68k.c (RELAXABLE_SYMBOL): Use EXTERN_FORCE_RELOC instead
2931 of hard-coded test for TARGET_OS=elf.
2935 * config/tc-mips.c (hilo_interlocks): Change definition
2936 so that MIPS32, MIPS64 and later ISAs are included, along with
2937 the already-included machines. Update comments.
2941 * config/tc-arm.c (tc_gen_reloc): Improve error message for
2942 undefined local labels.
2946 * config/tc-mips.c (load_address, macro): Update comments about
2947 NewABI GP relaxation.
2951 * config/tc-mips.c (macro_build): Remove place and counter arguments.
2952 (mips_build_lui, macro_build_ldst_constoffset): Likewise.
2953 (mips16_macro_build, macro_build_jalr): Remove counter argument.
2954 (set_at, load_register, load_address, move_register): Likewise.
2955 (load_got_offset, add_got_offset): Likewise.
2956 Update all calls and tidy accordingly.
2960 * config/tc-mips.c (RELAX_ENCODE): Remove WARN argument.
2961 (RELAX_FIRST, RELAX_SECOND): Turn into 8-bit quantities.
2962 (RELAX_USE_SECOND): Bump to 0x10000.
2963 (RELAX_SECOND_LONGER, RELAX_NOMACRO, RELAX_DELAY_SLOT): New flags.
2964 (mips_macro_warning): New variable.
2965 (md_assemble): Wrap macro expansion in macro_start() and macro_end().
2966 (s_cpload, s_cpsetup, s_cprestore, s_cpreturn): Likewise.
2967 (relax_close_frag): Set mips_macro_warning.first_frag. Adjust use
2969 (append_insn): Update mips_macro_warning.sizes.
2970 (macro_start, macro_warning, macro_end): New functions.
2971 (macro_build): Don't emit warnings here.
2972 (macro_build_lui, md_estimate_size_before_relax): ...or here.
2973 (md_convert_frag): Check for cases where one macro alternative
2974 needs a warning and the other doesn't. Emit a warning if the
2975 longer sequence was chosen.
2979 * config/tc-mips.h (tc_frag_data_type, TC_FRAG_TYPE): Remove.
2980 * config/tc-mips.c (RELAX_ENCODE): Take three arguments: the size of
2981 the first sequence, the size of the second sequence, and a flag
2982 that says whether we should warn.
2983 (RELAX_OLD, RELAX_NEW, RELAX_RELOC[123]): Delete.
2984 (RELAX_FIRST, RELAX_SECOND): New.
2985 (mips_relax): New variable.
2986 (relax_close_frag, relax_start, relax_switch, relax_end): New fns.
2987 (append_insn): Remove "place" argument. Use mips_relax.sequence
2988 rather than "place" to check whether we're expanding the second
2989 alternative of a relaxable macro. Remove redundant check for
2990 branch relaxation. If generating a normal insn, and there
2991 is not enough room in the current frag, call relax_close_frag()
2992 to close it. Update mips_relax.sizes[]. Emit fixups for the
2993 second version of a relaxable macro. Record the first relaxable
2994 fixup in mips_relax. Remove tc_gen_reloc workaround.
2995 (macro_build): Remove all uses of "place". Use mips_relax.sequence
2996 in the same way as in append_insn.
2997 (mips16_macro_build): Remove "place" argument.
2998 (macro_build_lui): As for macro_build. Don't drop the add_symbol
2999 when generating the second version of a relaxable macro.
3000 (load_got_offset, add_got_offset): New functions.
3001 (load_address, macro): Use new relaxation machinery. Remove
3002 tc_gen_reloc workarounds.
3003 (md_estimate_size_before_relax): Set RELAX_USE_SECOND if the second
3004 version of a relaxable macro is needed. Return -RELAX_SECOND if the
3005 first version is needed.
3006 (tc_gen_reloc): Remove relaxation handling.
3007 (md_convert_frag): Go through the fixups for a relaxable macro and
3008 mark those that belong to the unneeded alternative as done. If the
3009 second alternative is needed, adjust the fixup addresses to account
3010 for the deleted first alternative.
3014 * frags.h (frag_room): Declare.
3015 * frags.c (frag_room): New function.
3016 * doc/internals.texi: Document it.
3020 * config/tc-mips.c (append_insn): Don't do r3900 interlock
3021 optimization for -mtune=r3900, as this will break on other CPUs.
3025 * config/tc-m68hc11.c (build_indexed_byte): movb and movw cannot
3026 be relaxed, use fixup.
3027 (md_apply_fix3): Use 5 bit reloc from movb and movw fixup.
3031 * config/tc-sparc.c (sparc_ip): Disallow %f32-%f63 for single
3036 * config/tc-mips.c (append_insn): Properly detect variant frags
3037 that preclude swapping of relaxed branches. Correctly swap
3038 instructions between frags when dealing with relaxed branches.
3042 * acinclude.m4: Quote names of macros to be defined by AC_DEFUN
3044 * aclocal.m4: Regenerate.
3045 * configure: Regenerate.
3049 * config/tc-h8300.c (build_bytes): Apply relaxation to bit
3054 * config/tc-mips.c (macro_build_jalr): When adding an R_MIPS_JALR
3055 reloc, reserve space for the delay slot as well as the jalr itself.
3059 * config/tc-arm.c (do_vfp_reg2_from_sp2): Rename from do_vfp_sp_reg2.
3060 (do_vfp_sp2_from_reg2): New function.
3062 (do_vfp_dp_from_reg2): Check return values properly.
3066 * config/tc-mips.c (warn_nops): Remove static variable.
3067 (macro): Remove test of warn_nops.
3068 (md_shortops): Remove 'n'.
3069 (md_parse_option): Remove 'n' case.
3070 (md_show_usage): Remove -n.
3071 * doc/as.texinfo (Overview): Remove MIPS -n option.
3072 * doc/c-mips.texi (MIPS Opts): Remove mention -n.
3073 * NEWS: Mention removal of MIPS -n option.
3075 * config/tc-mips.c (ISA_HAS_COPROC_DELAYS): Remove.
3076 (cop_interlocks): Check ISA level.
3077 (cop_mem_interlocks): Define.
3078 (reg_needs_delay): Check cop_interlocks rather than
3079 ISA_HAS_COPROC_DELAYS.
3080 (append_insn): Likewise. Use cop_mem_interlocks rather than
3081 directly checking mips_opts.isa.
3082 (mips_emit_delays): Likewise.
3086 * config/tc-ia64.c (unwind): Move next_slot_number and
3087 next_slot_frag to ...
3088 (unw_rec_list): Here.
3089 (free_list_records): Removed.
3090 (output_unw_records): Likewise.
3091 (generate_unwind_image): Make it void.
3092 (alloc_record): Initialize next_slot_number and next_slot_frag.
3093 (slot_index): Take .org, .space and .align into account.
3094 (fixup_unw_records): Don't set slot_number to 0. Use
3095 list->next_slot_number and list->next_slot_frag instead of
3096 unwind.next_slot_number and unwind.next_slot_frag.
3097 (ia64_convert_frag): New.
3098 (generate_unwind_image): Generate a rs_machine_dependent frag
3100 (emit_one_bundle): Use list->next_slot_number and
3101 list->next_slot_frag instead of unwind.next_slot_number and
3102 unwind.next_slot_frag.
3104 * config/tc-ia64.h (md_convert_frag): Defined as
3106 (md_estimate_size_before_relax): Defined as (f)->fr_var.
3111 * config/tc-frv.h (md_apply_fix3): Don't define.
3112 * config/tc-frv.c (md_apply_fix3): New. Shift/truncate %hi/%lo
3114 * config/tc-frv.h (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
3116 * config/tc-frv.c (line_separator_chars): Add `!'.
3118 * config/tc-frv.c (md_assemble): Clear insn upfront.
3120 * config/tc-frv.c (OPTION_FDPIC): New macro.
3121 (md_longopts): Add mfdpic.
3122 (md_parse_option): Handle it.
3124 * config/tc-frv.c (md_cgen_lookup_reloc) <FRV_OPERAND_D12,
3125 FRV_OPERAND_S12>: Use reloc type encoded in fix-up.
3126 (frv_pic_ptr): Parse funcdesc.
3130 * doc/as.texinfo: Let texi2pod parse asconfig.texi and
3131 gasver.texi. Remove duplicate symbol definitions for texi2pod.
3135 * Makefile.am (Makefile): Move the dependency on
3136 $(BFDDIR)/configure.in to...
3137 (CONFIG_STATUS_DEPENDENCIES): ... here.
3138 (AUTOMAKE_OPTIONS): Require automake 1.8.
3139 * Makefile.in: Regenerate.
3140 * doc/Makefile.am (BASEDIR, BFDDIR): Define.
3141 (CONFIG_STATUS_DEPENDENCIES): Add a dependency on
3142 $(BFDDIR)/configure.in here as well.
3143 * doc/Makefile.in: Regenerate.
3147 * Makefile.am (install, install-info, RECURSIVE_TARGETS): Remove.
3148 * Makefile.in: Regenerate.
3149 * aclocal.m4: Regenerate.
3150 * doc/Makefile.am (install, install-info): Remove.
3151 (install-data-local): A new hook for install-info.
3152 (AUTOMAKE_OPTIONS): Require automake 1.8.
3153 * doc/Makefile.in: Regenerate.
3157 * doc/c-sh.texi: Update description about floating point behavior
3162 * configure.in: Add m68k-uClinux target.
3163 * configure: Regenerate.
3165 For older changes see ChangeLog-0203
3171 version-control: never
projects / binutils.git / blob