3 * config/tc-mips.c (load_address): Implement GP optimization
4 for 64bit address space non-PIC. Fix formatting.
5 (macro): Likewise. Simplify code.
6 (md_parse_option): Don't bail out if -G 0 is set for PIC code.
7 (mips_after_parse_args): Simplify code.
11 * expr.c (operand): Remove redundant code enclosed by #ifdef
12 RELAX_PAREN_GROUPING....#endif.
14 * config/tc-mn10200.c (tc_gen_reloc): Handle the case where the
15 reloc is the difference of two symbols defined in the same
18 * config/tc-iq2000.c (line_comment_chars): Include the # character
19 as otherwise this breaks #APP/#NO_APP processing.
23 * config/tc-arc.c(md_assemble): Remove dead code for handling
24 immediate indexing of ld and st .
29 * config/tc-mips.c (percent_op): Add %tlsgd, %tlsldm, %dtprel_hi,
30 %dtprel_lo, %tprel_hi, %tprel_lo, and %gottprel.
31 (parse_relocation): Check for a word break after a relocation
33 (md_apply_fix3): Handle TLS relocations, and mark thread-local
38 * config/tc-ppc.c (ppc_fix_adjustable <ELF>): Remove bogus checks.
42 * as.c (main): Use unlink_if_ordinary instead of unlink.
43 * messages.c (as_fatal): Likewise.
47 * config/tc-i386.c (build_modrm_byte): Add lock prefix for cr8...15
49 (parse_register): Allow cr8...15 in all modes.
53 * config/tc-i386.c (intel_e11): If not followed by T_PTR, treat T_BYTE
54 etc. like normal symbol references (T_ID).
58 * symbols.c (fb_label_name): Fix silly thinko in last change.
62 * expr.c (integer_constant): Remove TARGET_WORD_SIZE hack.
63 * config/tc-m68k.h (TARGET_WORD_SIZE): Delete.
65 * symbols.c (fb_label_name): Allow an augend of 2 for mmix.
70 * config/tc-arc.c (md_assemble): Initialize suffix for extension
75 * config/obj-coff.c (fixup_segment): Delete sy_mri_common assertion.
77 * as.h (assert): Warning fix.
78 * expr.c (expr): Correct assertion.
79 * read.c (s_comm_internal): Remove assertion.
80 * write.c (relax_segment): Enable vma assertion only for BFD_ASSEMBLER.
81 (fixup_segment): Remove assertion.
82 * config/tc-dlx.c (machine_ip): Remove untrue assertions.
83 (md_apply_fix3): Likewise.
84 * config/tc-i370.c (md_begin): Correct assertion.
85 (i370_macro): Warning fix for assertion.
89 * configure.in (AC_C_BIGENDIAN): Invoke.
90 * configure: Regenerate.
91 * write.c (write_object_file <!BFD_ASSEMBLER>): Don't use sizeof
92 host variable to set string header size.
93 * config/obj-aout.c (obj_header_append): Don't use host structs.
94 (obj_symbol_to_chars): Likewise.
95 (obj_emit_strings): Likewise. Use the passed in output pointer.
96 * config/obj-aout.h (H_GET_FILE_SIZE): Include H_GET_LINENO_SIZE.
97 * config/obj-bout.c (obj_emit_relocations): Use md_reloc_size,
98 not sizeof host struct.
99 (obj_header_append, obj_symbol_to_chars): Don't use host structs.
100 (obj_emit_strings): Likewise.
101 * config/obj-bout.h (EXEC_BYTES_SIZE): Define.
102 (N_TXTOFF, H_GET_FILE_SIZE, H_GET_HEADER_SIZE): Use instead of
104 (H_SET_SYMBOL_TABLE_SIZE): Hard code sym size rather than using
106 (host_number_to_chars): Define.
107 * config/obj-hp300.c (hp300_header_append): Don't use sizeof
108 host internal struct to set header sizes.
109 * config/tc-i960.c (md_number_to_field): Warning fix.
110 (md_ri_to_chars): Use host byte order.
111 (get_cdisp, md_apply_fix3): Warning fix.
112 * config/tc-m68k.c (md_assemble): Don't use sizeof host short.
116 * doc/c-arc.texi: Update documentation about ARC's extension
121 * config/tc-tic4x.c (tic4x_gen_to_words): Changed mail
126 * cgen.c: Warning fixes.
127 * config/tc-arc.c: Likewise.
128 * config/tc-arm.c: Likewise.
129 * config/tc-avr.c: Likewise.
130 * config/tc-d10v.c: Likewise.
131 * config/tc-d30v.c: Likewise.
132 * config/tc-frv.c: Likewise.
133 * config/tc-frv.h: Likewise.
134 * config/tc-h8300.c: Likewise.
135 * config/tc-h8500.c: Likewise.
136 * config/tc-i370.c: Likewise.
137 * config/tc-i960.c: Likewise.
138 * config/tc-ia64.c: Likewise.
139 * config/tc-ip2k.c: Likewise.
140 * config/tc-m68hc11.c: Likewise.
141 * config/tc-maxq.c: Likewise.
142 * config/tc-mcore.c: Likewise.
143 * config/tc-mips.c: Likewise.
144 * config/tc-msp430.c: Likewise.
145 * config/tc-pj.c: Likewise.
146 * config/tc-ppc.c: Likewise.
147 * config/tc-ppc.h: Likewise.
148 * config/tc-s390.c: Likewise.
149 * config/tc-sh.c: Likewise.
150 * config/tc-sh64.c: Likewise.
151 * config/tc-tic4x.c: Likewise.
152 * config/tc-tic80.c: Likewise.
153 * config/tc-v850.c: Likewise.
154 * config/tc-vax.c: Likewise.
155 * config/tc-w65.c: Likewise.
156 * config/tc-xstormy16.c: Likewise.
157 * config/tc-z8k.c: Likewise.
161 * read.c (read_a_source_file): Reinstate TC_EQUAL_IN_INSN test.
162 * doc/internals.texi (TC_EQUAL_IN_INSN): Reinstate.
166 * config/tc-mips.c (struct proc): Change isym to
167 func_sym. New member func_end_sym.
168 (s_mips_ent): Update.
169 (s_mips_end): Ditto. Add code to compute function size.
173 * read.c: Warning fixes.
174 * config/obj-elf.c: Likewise.
178 * config/tc-mips.c (append_insn): Call dwarf2_emit_insn() before
183 * config/obj-coff.c (obj_coff_section): Replace SEC_SHARED with
186 * config/tc-tic54x.c (tic54x_bss): Replace SEC_BLOCK with
188 (demand_empty_rest_of_line): Likewise.
189 (tic54x_sblock): Likewise.
190 (tic54x_clink): Replace with SEC_CLINK with SEC_TIC54X_CLINK.
194 * read.c (address_bytes): New function.
195 (TC_ADDRESS_BYTES): Default for BSD_ASSEMBLER to address_bytes.
196 (potable): Add "dc.a".
197 (cons_worker): Handle "dc.a".
198 * doc/internals.texi (TC_ADDRESS_BYTES): Document.
202 * input-file.c (input_file_open): Rearrange to avoid warning.
206 * config/tc-hppa.h (TC_EQUAL_IN_INSN): Delete.
207 * read.c (read_a_source_file): Remove TC_EQUAL_IN_INSN test.
208 * doc/internals.texi (TC_EQUAL_IN_INSN): Delete.
212 * config/tc-mips.c (macro_build_ldst_constoffset): Fail on $at
213 uses after .set noat.
214 (load_address): Likewise.
215 (macro): Likewise. Don't try to avoid $at use by sacrificing
216 the target register before it is stored, it won't work.
220 * config/tc-ia64.c (emit_one_bundle): Stop filling a bundle if we
221 see an instruction that specifies a template.
225 * config/tc-openrisc.c (openrisc_relax_frag): Delete unused function.
226 * config/tc-sparc.c (sparc_ip): Make op_exp static to silence warnings.
227 * config/tc-tic80.c (build_insn): Init insn[1] to silence warning.
231 * NEWS: Mention "-mhint.b=[ok|warning|error]".
233 * config/tc-ia64.c (md): Add hint_b.
234 (emit_one_bundle): Handle md.hint_b for "hint".
235 (md_parse_option): Accepted "-mhint.b=[ok|warning|error]".
236 (md_show_usage): Add "-mhint.b=[ok|warning|error]".
237 (ia64_init): Set md.hint_b to error.
238 (md_assemble): Handle md.hint_b for "hint.b".
240 * doc/as.texinfo: Add "-mhint.b=[ok|warning|error]".
241 * doc/c-ia64.texi: Likewise.
245 * tc.h (struct relax_type, relax_typeS): Move from here..
246 * as.h: ..to here. Make rlx_forward and rlx_backward an offsetT.
247 * ecoff.c (ecoff_new_file): Add appfile param.
248 * ecoff.h (ecoff_new_file): Likewise.
249 * itbl-lex.h: New file.
250 * itbl-lex.l: Include itbl-lex.h.
251 * itbl-parse.y: Likewise.
252 (insntbl_line, yyparse, yylex): Move to itbl-lex.h.
253 * read.c (s_app_file_string): Mark appfile possibly unused.
254 * subsegs.c (seg_not_empty_p): Make sec possibly unused.
255 * subsegs.h (struct seg_info_trash): Delete.
256 (seg_info): Use segment_info_type instead.
257 * config/obj-coff.c (struct filename_list): Make filename const char *.
258 * config/obj-ecoff.h (obj_app_file): Pass app to ecoff_new_file.
259 * config/obj-elf.c (elf_file_symbol): Similarly.
260 * config/tc-a29k.c (md_apply_fix3): Make val a valueT. Don't use
262 * config/tc-arc.c (md_operand): Warning fix.
263 * config/tc-arm.c (arm_parse_reloc): Only define when OBJ_ELF.
264 (md_begin): Rearrange #if defined OBJ_COFF || defined OBJ_ELF.
265 * config/tc-cris.h (TC_IMPLICIT_LCOMM_ALIGNMENT): Use do while.
266 * config/tc-frv.c (frv_force_relocation): Warning fix.
267 * config/tc-m68k.c (md_parse_option): Delete unused var.
268 * config/tc-mcore.c (mylog2): Rename from log2 throughout.
269 * config/tc-sparc.c: Likewise.
270 (s_common): Warning fix.
271 * config/tc-mips.c (append_insn): Use unsigned long long expressions.
272 * config/tc-mmix.c (PUSHJSTUB_MAX, PUSHJSTUB_MIN): Define from
274 * config/tc-s390.c (s390_insn): Delete test of unsigned >= 0.
275 * config/tc-sh.c (sh_cfi_frame_initial_instructions,
276 sh_regname_to_dw2regnum): Only define for OBJ_ELF.
277 * config/tc-tic4x.c (tic4x_insert_reg): Use ISLOWER.
278 (tic4x_do_align): Use TIC_NOP_OPCODE.
279 * config/tc-tic4x.h (TIC_NOP_OPCODE): Rename from NOP_OPCODE.
280 * config/tc-vax.c: Include netinet/in.h.
281 (tc_headers_hook): Formatting.
282 * config/tc-xstormy16.c (md_pcrel_from_section): Correct parens.
286 * config/tc-ia64.c (ia64_parse_name): Don't advance 'name' when
287 parsing inN, locN, outN. Set 'idx' to offset register number starts
288 at. Don't handle numbers with leading zeroes or beyond 95. Remove
293 * config/tc-mips.c (load_address): Fix formatting.
294 (macro): Don't use AT if .set noat is in effect. Fix formatting.
295 Catch macros which are unexpandable without AT. Remove duplicate
297 (macro2): Remove duplicate zeroing of used_at.
301 * dw2gencfi.c (output_cie, output_fde): Use DW_CFA_nop rather
307 * config/tc-mips.c (reloc_needs_lo_p): Handle
308 BFD_RELOC_MIPS16_HI16_S.
309 (fixup_has_matching_lo_p): Handle BFD_RELOC_MIPS16_LO16.
310 (append_insn): Add BFD_RELOC_MIPS16_GPREL, BFD_RELOC_MIPS16_HI16_S
311 and BFD_RELOC_MIPS16_LO16 to relocs to suppress overflow
313 (mips16_ip): Resolve BFD_RELOC_MIPS16_HI16_S,
314 BFD_RELOC_MIPS16_HI16 and BFD_RELOC_MIPS16_LO16 for constants.
315 Call my_getSmallExpression() to parse percent operators.
316 (percent_op_match, mips_percent_op): Separate definitions.
317 (mips16_percent_op): Define percent operators for the MIPS16 mode.
318 (parse_relocation): Handle the MIPS16 mode using
320 (md_apply_fix3): Handle BFD_RELOC_MIPS16_HI16,
321 BFD_RELOC_MIPS16_HI16_S and BFD_RELOC_MIPS16_LO16.
325 * config/tc-ia64.c (md_apply_fix3): Call ia64_gen_real_reloc_type
326 instead of explicitly dealing with the translation; exclude
327 relocations that are already pcrel, however.
331 * config/tc-ia64.c: Include limits.h (if available).
332 (gr_values[0]): Set path to INT_MAX.
333 (dot_reg_val): Don't allow changing value of r0. Limit range of
334 general registers at r127.
335 (specify_resource): Default resource index is -1. Don't set resource
336 index (in case IA64_RS_RSE) without setting the specific flag.
337 (note_register_values): Check operand is O_constant before tracking
338 input value of moves. Add tracking for dep.z with constant inputs.
339 (print_dependency): Resource index of specific resource may be zero.
340 (check_dependencies): Likewise.
344 * config/tc-ia64.c (parse_operands): New local variables reg1, reg2,
345 reg_class. Check operands and emit diagnostics for illegal use of
350 * config/tc-ia64.c (ia64_gen_real_reloc_type): Define and initialize
351 new variables type, suffix, and width. Handle
352 BFD_RELOC_IA64_DIR(32|64)[LM]SB in FUNC_LT_FPTR_RELATIVE case.
353 Handle BFD_RELOC_IA64_DIR64[LM]SB in FUNC_TP_RELATIVE case. Add
354 FUNC_DTP_MODULE case. Handle BFD_RELOC_IA64_DIR32[LM]SB in
355 FUNC_DTP_RELATIVE case. Return incoming relocation type if
356 BFD_RELOC_IA64_IPLT[LM]SB in FUNC_IPLT_RELOC case. Generate warning
357 if unable to translate relocation type, using the new variables.
361 * config/tc-ia64.h (ia64_symbol_chars): Declare.
362 (ty_symbol_chars): Define.
363 * config/tc-ia64.c (ia64_symbol_chars): Define.
367 * config/tc-ia64.c (ia64_parse_name): Only update next character if
368 input_line_pointer was advanced.
372 * config/tc-sh.c (md_apply_fix3): Add parentheses around &
377 * config/tc-ia64.c (md_parse_option): Handle -xnone and -xdebugn.
378 (md_show_usage): Add -xnone, -xdebugn, and -xdebugx. Relocate default
380 (ia64_init): Set md.detect_dv.
381 (ia64_start_line): New static variable warned. Warn only once when
382 encountering explicit stops in automatic mode.
383 * doc/c-ia64.texi: Describe -xnone, -xdebugn, and -xdebugx.
384 * NEWS: Mention new default mode.
388 * config/tc-ia64.c (dot_rot): Add comment that name strings should
389 be freed when wiping out previous state. Canonicalize names before
390 use. Free name string when detecting redefinition.
391 (dot_pred_rel): Call generic expression parser to process arguments.
392 Handle O_register case for individual predicates and O_subtract for
394 (ia64_parse_name): Canonicalize name before looking it up in dynamic
396 (ia64_canonicalize_symbol_name): Strip off all trailing # characters.
397 Warn if multiple found, issue error if resulting symbol name has zero
399 (dot_alias): Canonicalize name before use.
403 * config/tc-ia64.c (unwind_diagnostic): Return -1 for warning
405 (in_procedure): Return -1 for warning.
406 (in_prologue): Likewise.
411 * config/tc-ia64.c (dot_xdata): Undo the last change.
412 (dot_float_cons): Likewise.
413 (dot_xstringer): Likewise.
414 (dot_xdata_ua): Likewise.
415 (dot_float_cons_ua): Likewise.
419 * NEWS: Mention "-munwind-check=[warning|error]".
421 * config/tc-ia64.c (md): Add unwind_check.
422 (unwind_diagnostic): New.
423 (in_procedure): Call unwind_diagnostic when a directive isn't
425 (in_prologue): Call unwind_diagnostic when a directive isn't in
427 (in_body): Call unwind_diagnostic when a directive isn't in
429 (dot_endp): Set md.unwind_check to error before calling
430 in_procedure and restore it after. When the name is missing or
431 couldn't be found, use the one from the last .proc if
432 md.unwind_check isn't error. Warn if md.unwind_check is
434 (md_parse_option): Handle "-munwind-check=[warning|error]".
435 (md_show_usage): Add "-munwind-check=[warning|error]".
436 (ia64_init): Set md.unwind_check to warning.
438 * doc/as.texinfo: Add "-munwind-check=[none|warning|error]".
439 * doc/c-ia64.texi: Likewise.
443 * config/tc-ia64.h (LEX_AT): Include LEX_BEGIN_NAME.
445 (ia64_parse_name): New third parameter.
446 (md_parse_name): Pass third argument.
447 * config/tc-ia64.c (pseudo_func): Placeholders use NULL as name.
448 (md_operand): Handling of '@'-prefixed symbols moved from here...
449 (ia64_parse_name): ...to here.
453 * config/tc-ia64.c (md): Remove last_groups and group_idx.
454 (errata_nop_necessary_p): Remove declaraction and definition.
455 (emit_one_bundle): Don't call errata_nop_necessary_p. Don't
456 update md.group_idx. Don't reset md.last_groups.
460 * config/tc-ia64.c (parse_section_name): Handle non-quoted first
462 (dot_xdata): Free section name after use.
463 (dot_float_cons): Likewise.
464 (dot_xstringer): Likewise.
465 (dot_xdata_ua): Likewise.
466 (dot_float_cons_ua): Likewise.
467 (md_pseudo_table): Add xdata16 and xdata16.ua.
471 * doc/all.texi: Add IA64.
472 * doc/as.texinfo: Likewise.
474 * doc/c-ia64.texi: Fix typos.
478 * config/tc-arm.c (do_t_ldmstm): Change BFD_RELOC_NONE to
480 (do_t_push_pop): Likewise.
481 (md_assemble): Likewise.
482 (md_apply_fix3): Handle BFD_RELOC_NONE correctly, make
483 BFD_RELOC_UNUSED same as previous meaning of BFD_RELOC_NONE.
484 (create_unwind_entry): Output dependency on the required personality
486 testsuite/gas/arm/unwind.d: Alter expected output to include
487 dependency on __aeabi_unwind_cpp_pr[01].
491 * as.h (seg_not_empty_p): Return int, not bfd_boolean.
492 * subsegs.c (seg_not_empty_p): Likewise.
496 * config/tc-maxq.c (md_estimate_size_before_relax): Correct the
497 relative jump calculation.
498 <md_convert_frag) : Likewise.
499 <output_disp): Likewise.
503 * write.c (write_object_file): Recognize warning-symbol construct
504 and skip object- and target- handling for the second symbol.
508 * config/tc-ia64.c (dot_pred_rel): Update comment. Handle @-prefixed
509 designators along with quoted ones. Free copy of quoted designator
514 * config/atof-ieee.c, config/obj-coff.c, config/obj-elf.c,
515 config/obj-ieee.c, config/obj-som.c, config/obj-vms.c,
516 config/tc-a29k.c, config/tc-alpha.c, config/tc-arc.c,
517 config/tc-arm.c, config/tc-d30v.c, config/tc-dlx.c,
518 config/tc-fr30.c, config/tc-h8300.c, config/tc-h8500.c,
519 config/tc-i370.c, config/tc-i386.c, config/tc-i960.c,
520 config/tc-ia64.c, config/tc-m32r.c, config/tc-m32r.h,
521 config/tc-m68hc11.c, config/tc-m68hc11.h, config/tc-mips.c,
522 config/tc-mn10200.c, config/tc-msp430.c, config/tc-ns32k.c,
523 config/tc-openrisc.c, config/tc-or32.c, config/tc-pdp11.c,
524 config/tc-pj.c, config/tc-sparc.h, config/tc-tic54x.c,
525 config/tc-tic80.c, config/tc-v850.c, config/tc-w65.c,
526 config/tc-xtensa.c, config/tc-z8k.c, config/xtensa-relax.c: Remove
527 #if 0'd code throughout.
531 * as.c (parse_args): Bump copyright date reported by --version to
536 * configure.tgt: Remove obsolete targets m68k-lynxos, sparc-lynxos
541 * macro.c (buffer_and_nest): Allow 'from' being NULL; handle anything
542 that can end with .endr in that case. Make requiring/permitting
543 pseudo-ops without leading dot closer to the logic in read.c serving
545 (expand_irp): Don't pass a mnemonic to buffer_and_nest as it will be
550 * macro.c (do_formals): Adjust to no longer accept empty parameter
552 (define_macro): Adjust to no longer accept empty macro name, garbage
553 following the parameters, or macros that were previously defined.
554 * read.c (s_bad_end): Declare.
555 (potable): Add endm. Handler for endr and endm is s_bad_end.
556 (s_bad_end): Rename from s_bad_endr. Adjust to handle both .endm
558 * read.h (s_bad_endr): Remove.
562 * config/tc-ia64.c (parse_operands): Parse all specified operands,
563 immediately discarding (but counting) those exceeding the maximum
564 possible amount. Track whether output and input operand counts ever
565 matched, and use this to better indicate which of the operands/
566 operand types was wrong; specifically don't default to pointing to
571 * config/tc-ia64.c (unwind): Remove proc_end (now an automatic
572 variable in dot_endp). Add body and insn. Make prologue,
573 prologue_mask, body, and insn bitfields.
574 (fixup_unw_records): Remove spurious new-lines from end of diagnostic
576 (in_procedure, in_prologue, in_body): New.
577 (dot_fframe, dot_vframe, dot_vframesp, dot_vframepsp, dot_save,
578 dot_restore, dot_restorereg, dot_restorereg_p, dot_handlerdata,
579 dot_unwentry, dot_altrp, dot_savemem, dot_saveg, dot_savef, dot_saveb,
580 dot_savegf, dot_spill, dot_spillreg, dot_spillmem, dot_spillreg_p,
581 dot_spillmem_p, dot_label_state, dot_copy_state, dot_unwabi,
582 dot_personality): Use the appropriate one of the above.
583 (dot_proc): Clear unwind.proc_start; set to current location only if
584 none of the entry points were valid. Check for non-zero-length entry
585 point names. Check that entry points aren't defined, yet. Clear
586 unwind.prologue, unwind.body, and unwind.insn.
587 (dot_body): Call in_procedure. Check that first directive in procedure
588 had no insns emitted before. Set unwind.body.
589 (dot_prologue): Call in_procedure. Check that not already in prologue.
590 Check that first directive in procedure had no insns emitted before.
592 (dot_endp): Call in_procedure. Declare proc_end. Check for non-zero-
593 length entry point names. Check that entry points became defined.
594 (md_assemble): Set unwind.insn once unwind.proc_start is defined.
598 * config/tc-ia64.c (emit_one_bundle): Snapshot manual bundling state
599 before actually using it. Don't generate an error in manual bundling
600 mode when looking at an insn requiring slot 2 but not yet at slot 2.
601 Don't generate an error in manual bundling mode when looking at an
602 insn required to be last in its group but the required slot hasn't
603 been reached, yet. Allow conversion from MII to MI;I for bundle
604 consisting of only 2 insns with the stop between them. Suppress
605 various meaningless errors resulting from detecting earlier ones.
609 * config/tc-ia64.c (parse_operands): Also handle alloc without first
614 * config/tc-z8k.c (md_assemble): Improve error detection.
618 * config/tc-ia64.c (ia64_estimate_size_before_relax): Allocate space
619 for personality routine pointer only if there is one.
620 (ia64_convert_frag): Likewise.
621 (generate_unwind_image): Likewise.
625 * config/tc-z8k.c (INSERT): Remove, not used anywhere.
626 (md_apply_fix3): Make relative branches out of range an error
627 instead of a warning. Display correct line number for out of
628 range branches/calls/memory accesses.
632 * dwarf2dbg.c (dwarf2_finish): Correct logic for determining when
633 to emit .debug_line and other debug sections.
634 * as.h (seg_not_empty_p): Declare.
635 * subsegs.c (seg_not_empty_p): New predicate.
639 * configure: Regenerate to track ../gettext.m4 change.
643 * config/tc-ia64.c (emit_one_bundle): Change "?imbf??" to "?ibmfxx".
647 * config/tc-ia64.c (emit_one_bundle): Add late resolution of move
648 to/from application registers dynamic insns.
649 (md_assemble): Defer resolution of move to/from application registers
650 dynamic insns when they can be issued on either the I- or M-units.
654 * config/tc-frv.c (md_apply_fix3): Mark TLS symbols as such.
656 * config/tc-frv.c (frv_pic_ptr): Add tlsmoff support.
658 * cgen.c (gas_cgen_parse_operand): Handle
659 CGEN_PARSE_OPERAND_SYMBOLIC.
660 * config/tc-frv.c (md_cgen_lookup_reloc): Handle TLS relocations.
661 (frv_force_relocation): Likewise. Fix handling of PIC
663 (md_apply_fix3): Likewise.
667 * as.h: Remove #if 0'd code.
668 * atof-generic.c (atof_generic): Likewise.
669 * ecoff.c (ecoff_directive_frame): Likewise.
670 * frags.h (FRAG_APPEND_1_CHAR): Likewise.
671 * itbl-ops.c (itbl_add_reg): Likewise.
672 * listing.c (calc_hex): Likewise.
673 * read.c (MASK_CHAR): Likewise.
674 * subsegs.c (subsegs_print_statistics): Likewise.
675 * symbols.c (indent): Likewise.
676 * write.c (write_relocs): Likewise.
677 (write_object_file): Likewise.
678 (relax_frag): Likewise.
682 * as.c (std_longopts): Add an entry for "--a" in order to prevent
683 getopt_long_only() from considering -a as an abbreviation for
685 (parse_args): Fix the parsing of -a=<file>.
690 * read.c (s_incbin): Adjust default count for skip. Check validity
691 of count and skip rigorously.
695 * config/tc-mips.c (dummy_opcode): Add init for new struct member.
699 * read.c (convert_to_bignum): New function, split out from...
700 (emit_expr): ...here. Handle the case where X_add_number is
701 positive and the input value is negative.
702 (output_big_sleb128): Fix setting of continuation bit. Check whether
703 the final byte needs to be sign-extended. Fix size-shrinking loop.
704 (emit_leb128_expr): When generating a signed leb128, see whether the
705 sign of an O_constant's X_add_number matches the sign of the input
706 value. Use a bignum if not.
710 * tc-sh.c (md_begin,md_parse_option): Change arch_sh1_up to
711 arch_sh_up in order to match the external name and make the
712 testsuite's job easier.
717 * config/tc-i386.c (i386_scale): Disallow 0 scale.
721 * config/tc-iq2000.c (s_iq2000_set): Fix thinko parsing
722 ignored_arguments array.
726 * write.c (write_object_file): Disallow a symbol equated to
731 * tc-maxq.c: Replace constants 10 and 20 with bfd_mach_maxq10 and
733 (md_pseudo_table): Add new pseudo ops for maxq10 and maxq20.
734 (maxq_target): New function: Set the machine type.
738 * config/tc-arm.c (FPU_DEFAULT): Define for TE_VXWORKS.
739 (md_begin): Handle TE_VXWORKS for FP defaults.
740 (md_apply_fix3): Correct rela offsets.
741 (elf32_arm_target_format): Add VxWorks targets.
745 * configure.tgt: Set em=vxworks for *-*-vxworks.
746 * config/te-vxworks.h: New File.
750 * config/tc-arm.c (arm_cpus): Correct arch field for arm1026ej-s.
754 * config/tc-msp430.c (md_apply_fix3): Fix offset calculation for
759 * config/tc-ia64.c (md): Add member "loc_directive_seen".
760 (dot_loc): New function.
761 (md_pseudo_table): Add entry to map .loc to dot_loc().
762 (emit_one_bundle): Only call dwarf2_gen_line_info() if we have
763 seen a .loc directive or we're generating DWARF2 debug info for
769 * read.c (s_comm_internal): Don't zero end of name until size
770 expression has been parsed.
774 * config/tc-avr.c (mcu_types): Move attiny{13,2313} from avr4 to avr2.
778 * config/tc-crx.c: Support 'bcop' relaxation (dealt as in 'cmp&branch'
783 * configure.tgt: New.
784 * configure.in: Move setting of cpu_type, fmt, etc., to
786 * Makefile.am (CONFIG_STATUS_DEPENDENCIES): Add
787 $(srcdir)/configure.tgt.
788 * configure, Makefile.in: Rebuild.
792 * config/tc-avr.c: Add support for the new R_AVR_LDI, R_AVR_6 and
793 R_AVR_6_ADIW relocs for the LDI, ADIW/SBIW and LDD/STD
795 (avr_offset_expression): New function to parse offsets for LDI
797 (avr_operand): Use it.
798 (md_apply_fix3): Generate the relocs.
802 * config/tc-sh64.c (shmedia_md_apply_fix3): Add missing
803 BFD_RELOC_SH_IMMS10BY8 relocation.
805 * config/tc-sh64.c (shmedia_build_Mytes): Emit an error message rather
806 than just ignoring bad code.
810 * config/tc-v850.c (handle_lo16): New function.
811 (v850_reloc_prefix): Use it to check lo().
812 (md_assemble, md_apply_fix3): Handle BFD_RELOC_V850_LO16_SPLIT_OFFSET.
816 * configure.in: Add s390x-ibm-tpf support.
817 * configure: Regenerate.
821 * config/obj-elf.c (obj_elf_change_section): Only set type and
822 attributes on new sections. Emit warning when type of re-declared
823 section doesn't match.
827 * dw2gencfi.c (dot.cfi.startproc): Clear cur_cfa_offset so
828 '.cfi_startproc simple' doesn't inherit the old value.
832 * dw2gencfi.c (output_cfi_insn): Adjust DW_CFA_def_cfa_sf generation
833 to emit a signed and factored offset. Adjust DW_CFA_def_cfa_offset_sf
834 generation to emit a factored offset.
838 * config/tc-mips.c (macro) [M_LA_AB]: Give an error for a offset
839 which is too large in the case of NO_PIC without 64-bit
842 * config/tc-mips.c (mips_in_shared): New static variable.
843 (macro_build_lui): Permit "_gp" if !mips_in_shared.
844 (md_longopts): Add -mshared and -mno-shared.
845 (md_parse_option): Handle OPTION_MSHARED and OPTION_MNO_SHARED.
846 (s_cpload): Implement !mips_in_shared case.
847 (s_cpsetup): Likewise.
848 * doc/c-mips.texi (MIPS Opts): Document -mno-shared.
849 * NEWS: Mention -mno-shared.
853 * config/tc-arm.c (s_arm_unwind_fnend): Use R_ARM_PREL31 relocation
858 * config/tc-mips.c (append_insn): If we emit a nop during a relax
859 sequence, increase the size of the sequence.
861 * config/tc-mips.c (mips_cpu_info_table): Change "9000" entry to
866 * read.c (s_align): Use an align_limit temporary to allay a GCC
867 signed/unsigned comparison warning.
871 * Makefile.am (TARG_ENV_HFILES): Add te-armlinuxeabi.h.
872 * configure.in: Use it for arm*-*-linux-gnueabi*.
873 * config/tc-arm.c: Allow emulation file to set FPU_DEFAULT.
874 * config/te-armlinuxeabi.h: New file.
875 * Makefile.in: Regenerated.
876 * aclocal.m4: Likewise.
877 * configure: Likewise.
878 * doc/Makefile.in: Regenerated.
882 * config/tc-xtensa.c (xtensa_switch_section_emit_state): Use subseg_set.
883 (xtensa_restore_emit_state): Likewise.
887 * read.c (ALIGN_LIMIT): Define, increasing limit for BFD_ASSEMBLER.
892 * Makefile.am: Change LOCALEDIR to $(datadir)/share.
893 * Makefile.in: Regenerate.
897 * config/tc-crx.c: Major code cleanup. Remove unused variables and
898 functions, give functions a meaningful name, add comments.
899 (check_range): New function - Replace operand size calculation
901 (assemble_insn): Update Algorithm, improve error issuing.
903 (process_label_constant): Bug fix regarding COP_BRANCH_INS relocation
908 * config/tc-m32r.c (md_pcrel_from_section): Fixed a pcrel relocte
909 miss between different section in the same module.
910 (tc_gen_reloc): Likewise.
914 * gas/config/tc-avr.c (mcu_types): Add support for atmega165,
915 atmega325, atmega3250, atmega645 and atmega6450.
919 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
920 permissible for the selected instruction suffix.
921 (process_suffix): For DefaultSize instructions, suppressing the
922 guessing of a 'q' suffix if the instruction doesn't support it is
923 pointless, because only an 'l' suffix can be guessed in this place.
927 * config/tc-iq2000.c: Remove support for IQ10 processor.
928 Convert to ISO C90 formatting.
929 * config/tc-iq2000.h: Likewise.
933 * config/tc-mn10300.c (md_relax_table): More fixes to the offsets
934 in this table. They should be correct now.
938 * config/tc-i386.h (CpuMMX2): Declare. Artificial classifier to
939 indicate the MMX extensions added by both SSE and 3DNow!A.
940 (Cpu3dnowA): Declare.
941 (CpuUnknownFlags): Update.
942 * config/tc-i386.c (cpu_sub_arch_name): Declare.
943 (cpu_arch): i586 and pentium do not imply MMX. i686 and pentiumpro do
944 neither imply SSE nor MMX. k6 implies MMX. k6_2 additionally implies
945 3DNow!. Athlon additionally implies 3DNow!A. Several new
946 entries (those starting with a dot are for sub-arch specification).
947 (set_cpu_arch): Handle sub-arch specifications.
948 (parse_insn): Distinguish between instructions not supported because
949 of insufficient CPU features and because of 64-bit mode.
950 * doc/c-i386.texi: Describe enhanced .arch directive.
954 * config/tc-xtensa.c (xg_add_opcode_fix): Set fx_no_overflow.
958 * dwarf2dbg.c (dwarf2_finish): Don't write a .debug_line section
959 without a corresponding .debug_info section.
963 * read.c (potable): Add "error" and "warning".
964 (s_errwarn): New function.
965 * read.h (s_errwarn): Declare.
966 * doc/as.texinfo (Error, Warning): Document .error and .warning.
970 * config/tc-tic54x.c (tic54x_adjust_symtab): Adjust call to
975 * config/tc-msp430.c (struct rcodes_s, MSP430_RLC, msp430_rcodes,
976 struct hcodes_s, msp430_hcodes): From include/opcode/msp430.h.
980 * config/obj-coff.c (c_dot_file_symbol): Add "app" param.
981 (coff_adjust_symtab): Adjust call.
982 (crawl_symbols): Likewise.
983 * config/obj-coff.h (c_dot_file_symbol): Add "app" param.
984 (obj_app_file): Adjust.
988 * configure.in: Enable bfd_assember for the MAXQ port.
989 * configure: Regenerate.
994 * config/tc-xtensa.c (finish_vinsn): Clear pending instruction if
996 (check_t1_t2_reads_and_writes): Check for both reads and writes to
997 interfaces that are related as determined by xtensa_interface_class_id.
1001 * config/tc-mn10300.c (md_relax_table): Fix off by one negative
1002 offsets for conditional branches.
1006 * config/tc-xtensa.c (MAX_IMMED6): Change value to 65.
1010 * config/tc-xtensa.c (update_next_frag_state): Always add a NOP if
1011 relaxing at the end of a loop. Don't mark frags as UNREACHABLE or
1013 (relax_frag_immed): Update call to update_next_frag_state.
1017 * obj.h (struct format_ops <app_file>): Add int param.
1018 * read.h (s_app_file_string): Likewise.
1019 * read.c (s_app_file_string): Likewise.
1020 (s_app_file): Adjust s_app_file_string call.
1021 * config/tc-mips.c (s_mips_file): Likewise.
1022 * config/obj-coff.h (obj_app_file): Add app param.
1023 * config/obj-ecoff.h (obj_app_file): Likewise.
1024 * config/obj-multi.h (obj_app_file): Likewise.
1025 * config/obj-elf.h (elf_file_symbol): Likewise.
1026 * config/obj-elf.c (elf_file_symbol): Only emit one file symbol
1027 if called for # preprocessor lines.
1032 * symbols.c (resolve_symbol_value): Convert weak symbols only
1034 (symbol_equated_reloc_p): Don't equate weaks when relocating
1035 only for Windows PECOFF.
1039 * config/tc-crx.c (print_insn): Check and set insn_addr.
1040 * config/tc-crx.h (md_frag_check): Define.
1045 * configure.in: Add support for new target maxq-coff.
1046 * configure: Regenerate.
1047 * NEWS: Mention new support.
1048 * config/tc-maxq.c: New file.
1049 * config/tc-maxq.h: New file.
1050 * config/obj-coff.h: Add support for maxq-coff.
1054 * symbols.c (any_external_name): Define.
1055 (resolve_symbol_value): Convert weak symbols.
1056 (S_SET_EXTERNAL): Support any_external_name.
1057 (S_SET_NAME): Qualify parameter const.
1058 (symbol_equated_reloc_p): Equate to weaks when relocating.
1059 * symbols.h (S_SET_NAME): Qualfiy parameter const.
1060 * tc.h (any_external_name): Declare.
1061 * config/obj-coff.c ("coff/pe.h"): Include for BFD
1063 (weak_is_altname): Declare and define.
1064 (weak_name2altname): Same.
1065 (weak_altname2name): Same.
1066 (weak_uniquify): Same.
1067 (weak_altprefix): Define.
1068 (obj_coff_weak): Change .weak syntax and handling.
1069 (coff_frob_symbol): Fix PE weak symbol alternates.
1070 * config/obj-coff.h (USE_UNIQUE): Define.
1071 * config/tc-i386.c (md_apply_fix3): Assume weak symbols
1072 are in another segment.
1073 (tc_gen_reloc): Remove broken addend hack.
1074 doc/as.texinfo: Update.
1078 * config/tc-xtensa.c (total_frag_text_expansion): New.
1079 (md_estimate_size_before_relax): Use it.
1080 (find_address_of_next_align_frag): Likewise.
1084 * config/tc-crx.c: Rename argument types.
1085 (processing_arg_number): Rename to 'cur_arg_num'.
1086 (get_number_of_bits): Rename to 'set_operand_size'.
1087 (get_operandtype): Rename to 'parse_operand', totally rewrite.
1088 (set_cons_rparams): Rename to 'set_operand', totally rewrite.
1089 (set_indexmode_parameters): Remove function, integrate its code into
1091 (set_operand_size): Get rid of 'Operand Number' function parameter -
1092 use global variable 'cur_arg_num' instead.
1093 Use a local 'argument' pointer to reference the current argument.
1094 (parse_operand): Likewise.
1095 (set_operand): Likewise.
1096 (process_label_constant): Likewise.
1100 * config/tc-xtensa.c: Remove XTENSA_SECTION_RENAME ifdefs.
1101 (add_section_rename): Delete. Inlined into...
1102 (build_section_rename): ...here. Use xstrdup instead of strdup.
1103 (xtensa_section_rename): Drop "const" from argument and return types.
1104 (md_show_usage): Indent to match show_usage().
1105 * config/tc-xtensa.h: Remove XTENSA_SECTION_RENAME ifdefs.
1106 (tc_canonicalize_section_name): Define.
1107 (md_elf_section_rename): Remove unused macro.
1108 * doc/as.texinfo (Overview): Document Xtensa --rename-section option.
1109 * doc/c-xtensa.texi (Xtensa Options): Likewise.
1110 (Frame Directive): Delete.
1114 * configure.in: Remove arm-*-oabi and thumb-*-oabi.
1115 * config/tc-arm.c (target_oabi): Delete.
1116 (md_apply_fix3, elf32_arm_target_format): Remove target_oabi checks.
1117 (arm_opts): Remove moabi.
1118 * doc/as.texinfo (Overview): Remove documentation of -moabi.
1119 * doc/c-arm.texi (ARM Options): Likewise.
1120 * configure: Regenerated.
1124 * configure.in (crisv32): Recognize. AC_DEFINE_UNQUOTED
1125 DEFAULT_CRIS_ARCH. Handle crisv32-*-linux-gnu* like
1126 cris-*-linux-gnu* and crisv32-*-* like cris-*-*.
1127 * configure: Regenerate.
1128 * config/tc-cris.c (enum cris_archs): New.
1129 (cris_mach, cris_arch_from_string, s_cris_arch, get_sup_reg)
1130 (cris_insn_ver_valid_for_arch): New functions.
1131 (DEFAULT_CRIS_ARCH): New macro, default to cris_any_v0_v10.
1132 (cris_arch): New variable.
1133 (md_pseudo_table): New pseudo .arch.
1134 (err_for_dangerous_mul_placement): Initialize according to
1136 (STATE_COND_BRANCH): Renamed from STATE_CONDITIONAL_BRANCH.
1138 (STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON)
1139 (STATE_ABS_BRANCH_V32, STATE_LAPC, BRANCH_BF_V32, BRANCH_BB_V32)
1140 (BRANCH_WF_V32, BRANCH_WB_V32): New.
1141 (BRANCH_BF, BRANCH_BB, BRANCH_WF, BRANCH_WB): Don't undef after
1142 use in md_cris_relax_table.
1143 (md_cris_relax_table): Add entries for STATE_COND_BRANCH_V32,
1144 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
1145 Update and improve head comment.
1146 (OPTION_PIC): Define in terms of previous option, OPTION_US.
1147 (OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): Similar.
1149 (md_longopts): New option --march=...
1150 (cris_any_v0_v10_long_jump_size, crisv32_long_jump_size): New
1152 (md_long_jump_size): Initialize in terms of DEFAULT_CRIS_ARCH.
1153 (HANDLE_RELAXABLE): New macro.
1154 (md_estimate_size_before_relax): Use HANDLE_RELAXABLE for common
1155 cases. Check for weak symbols and assume not relaxable. Handle
1156 STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON,
1157 STATE_ABS_BRANCH_V32, STATE_LAPC. Use new variable symbolP, not
1159 (md_convert_frag): Handle STATE_COND_BRANCH_V32,
1160 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
1161 (cris_create_short_jump): Adjust for CRISv32.
1162 (cris_relax_frag): Handle new states.
1163 (md_create_long_jump): Ditto. Emit error for common_v10_v32.
1164 (md_begin): Define symbols "..asm.arch.cris.v32",
1165 "..asm.arch.cris.v10", "..asm.arch.cris.common_v10_v32" and
1166 "..asm.arch.cris.any_v0_v10". Use cris_insn_ver_valid_for_arch
1167 when entering opcode table entry points.
1168 (md_assemble): Adjust branch handling for CRISv32. Handle LAPC
1169 relaxation. In fix_new_exp call for main insn, pass 1 for pcrel
1170 parameter for 8, 16 and 32-bit pc-relative insns and LAPC.
1171 (cris_process_instruction): Initialize out_insnp->insn_type to
1172 CRIS_INSN_NONE, not CRIS_INSN_NORMAL.
1173 <case ']', '[', 'A', 'd', 'Q', 'N', 'n', 'Y', 'U', 'u', 'T'>: New
1175 <case 'm'>: Check that modified_char == '.'.
1176 <invalid operands>: Consume the rest of the line.
1177 When operands don't match, skip over subsequent insns with
1178 non-matching version specifier but same mnemonic.
1179 <immediate constant, case SIZE_SPEC_REG>: Immediate operands for
1180 special registers in CRISv32 are always 32 bit long.
1181 <immediate constant, case SIZE_FIELD_SIGNED, SIZE_FIELD_UNSIGNED>:
1183 (get_gen_reg): Only recognize "PC" when followed by "+]" for v32
1184 and compatible. Recognize "ACR" for v32, unless followed by "+".
1185 (get_spec_reg): Consider cris_arch when looking up register.
1186 (get_autoinc_prefix_or_indir_op): Don't recognize assignment for
1188 (get_3op_or_dip_prefix_op): Check for ']' after seeing '[rN+'.
1189 (cris_get_expression): Restore input_line_pointer if failing "early".
1190 (get_flags): Consider cris_arch and recognize flags accordingly.
1191 (branch_disp): Adjust for CRISv32.
1192 (gen_cond_branch_32): Similar. Emit error for common_v10_v32.
1193 (cris_number_to_imm): Use as_bad_where, not as_bad. Remove
1194 related FIXME. Don't insist on BFD_RELOC_32_PCREL fixup to be
1195 resolved. Don't enter zeros in object file for
1197 <case BFD_RELOC_CRIS_LAPCQ_OFFSET, BFD_RELOC_CRIS_SIGNED_16>
1198 <case BFD_RELOC_CRIS_SIGNED_8>: New case.
1199 (md_parse_option): Break out "return 1".
1200 <OPTION_ARCH> New case.
1201 (tc_gen_reloc): <case BFD_RELOC_CRIS_LAPCQ_OFFSET>
1202 <case BFD_RELOC_CRIS_SIGNED_16, BFD_RELOC_CRIS_SIGNED_8>
1203 <case BFD_RELOC_CRIS_UNSIGNED_8, BFD_RELOC_CRIS_UNSIGNED_16>
1204 <case BFD_RELOC_32_PCREL>: New cases.
1205 Addends for non-zero fx_pcrel are too in fx_offset.
1206 (md_show_usage): Show --march=<arch>.
1207 (md_apply_fix3): Adjust val for BFD_RELOC_CRIS_LAPCQ_OFFSET.
1208 (md_pcrel_from): BFD_RELOC_CRIS_LAPCQ_OFFSET is PC-relative too.
1209 (s_syntax) <struct syntaxes>: Properly constify member operand.
1210 * config/tc-cris.h (TARGET_MACH): Define.
1211 (cris_mach): Declare.
1212 * doc/as.texinfo (Overview) <CRIS>: Add --march=...
1213 * doc/c-cris.texi (CRIS-Symbols): New node for built-in symbols.
1214 (CRIS-Opts): Document --march=...
1215 (CRIS-Pseudos): Document .arch.
1219 * config/tc-i386.c (set_intel_syntax): Allow % in symbol names when
1220 intel syntax and no register prefix, allow $ in symbol names when
1222 (set_16bit_gcc_code_flag): Replace literal 'l' by LONG_MNEM_SUFFIX.
1223 (intel_float_operand): Add fourth return value indicating math control
1224 operations. Make classification more precise.
1225 (md_assemble): Complain if memory operand of mov[sz]x has no size
1227 (parse_insn): Translate word operands to floating point instructions
1228 operating on integers as well as control instructions to short ones
1229 as expected by AT&T syntax. Translate 'd' suffix to short one only for
1230 floating point instructions operating on non-integer operands.
1231 (match_template): Remove fldcw special case. Adjust q-suffix handling
1232 to permit it on fild/fistp/fisttp in AT&T mode.
1233 (process_suffix): Don't guess DefaultSize insns' suffix from
1234 stackop_size for certain floating point control instructions. Guess
1235 suffix for branch and [ls][gi]dt based on flag_code. Split error
1236 messages for Intel and AT&T syntax, and make the condition more strict
1237 for the former. Adjust suppressing of generation of operand size
1239 (intel parser): Allow the full set of MASM operators. Add FWORD, TBYTE,
1240 OWORD, and XMMWORD operand size specifiers (TBYTE replaces XWORD). Add
1241 more error checking.
1242 * config/tc-i386.h (BYTE_PTR WORD_PTR DWORD_PTR QWORD_PTR XWORD_PTR
1243 SHORT OFFSET_FLAT FLAT NONE_FOUND): Remove unused defines.
1247 * symbols.c (colon) [!WORKING_DOT_WORD]: Don't declare
1248 md_short_jump_size, md_long_jump_size.
1249 * write.c [!WORKING_DOT_WORD]: Ditto.
1250 * tc.h [!WORKING_DOT_WORD]: Declare them here. Drop const
1252 * config/tc-cris.h (md_short_jump_size, md_long_jump_size): Don't
1254 * config/tc-cris.c (md_short_jump_size, md_long_jump_size): Drop
1255 const qualifier in these definitions.
1256 * config/tc-i370.c, config/tc-m68k.c, config/tc-pdp11.c,
1257 config/tc-s390.c, config/tc-tahoe.c, config/tc-vax.c: Ditto.
1261 * dwarf2dbg.c (dwarf2_finish): Check for the existence of a file
1262 table before deciding to produce a .debug_line section to match up
1263 with a user provided .debug_info section.
1267 * config/tc-crx.c (getreg_image): Bug fix, a return value was
1268 mistakenly omitted from CRX_C_REGTYPE and CRX_CS_REGTYPE cases.
1272 * config/tc-crx.c: Remove global variable 'post_inc_mode'.
1273 (get_flags): New function.
1274 (get_number_of_bits): Edit comments, update numeric values to
1276 (process_label_constant): Don't support the colon
1277 format (SYMBOL:[s|m|l]).
1278 (set_cons_rparams): Support argument type 'arg_rbase'.
1279 (get_operandtype): Bug fix in 'rbase' operand type parsing.
1280 (handle_LoadStor): Bug fix, first handle post-increment mode.
1281 (getreg_image): Remove redundant code, update according to latest
1283 (print_constant): Bug fix relate to 3-word instructions.
1284 (assemble_insn): Bug fix, when matching instructions, verify also
1285 instruction type (not only mnemonic).
1286 Add various error checking.
1287 (preprocess_reglist): Support HI/LO and user registers.
1291 * config/tc-ia64.c (fixup_unw_records): Don't let the "t" value
1292 in an epilogue directive go negative.
1297 * config/tc-ia64.c (emit_one_bundle): Decrement md.num_slots_in_use
1298 after reporting template error during manual bundling. Reported
1303 * Makefile.am: Run dep-am.
1304 * aclocal.m4: Regenerate with automake 1.9.2.
1305 * Makefile.in: Regenerate with automake 1.9.2.
1306 * doc/Makefile.in: Likewise.
1308 * config/tc-arm.c: Include "dw2gencfi.h".
1309 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
1311 * config/tc-arm.h (TARGET_USE_CFIPOP, DWARF2_DEFAULT_RETURN_COLUMN)
1312 (DWARF2_CIE_DATA_ALIGNMENT, tc_regname_to_dw2regnum)
1313 (tc_cfi_frame_initial_instructions): Define.
1314 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
1319 * config/tc-crx.c (assemble_insn): Check unsigned immediate
1321 Update coprocessor id to be unsigned immediate.
1325 * config/tc-i386.c (O_secrel): Delete.
1326 (tc_pe_dwarf2_emit_offset): New function.
1327 * config/tc-i386.h (O_secrel): Define as O_md1.
1328 (TC_DWARF2_EMIT_OFFSET): Define.
1332 * config/tc-xstormy16.c (xstormy16_cons_fix_new): Accept and
1333 ignore @fptr() directives for 4-byte fixups.
1337 * Makefile.am: Run "make dep-am"
1338 * Makefile.in: Regenerate.
1342 * doc/c-xtensa.texi (Xtensa Options, Absolute Literals Directive):
1343 Remove comments about placement of literal pools.
1344 (Literal Directive): Update description of literal placement.
1345 (Literal Prefix Directive): Remove statement that this does not apply
1346 to absolute-mode literals. Describe new section naming scheme.
1350 * config/tc-xtensa.c (xtensa_frequency_pseudo): Use set_subseg_freq.
1351 (is_entry_opcode, is_movi_opcode, is_the_loop_opcode, is_jx_opcode,
1352 is_windowed_return_opcode): Delete.
1353 (xtensa_frob_label): Use get_subseg_target_freq.
1354 (md_assemble): Inline call to is_entry_opcode.
1355 (xtensa_handle_align): Inline call to get_frag_is_literal.
1356 (relaxation_requirements): Inline call to is_jx_opcode.
1357 (emit_single_op): Inline call to is_movi_opcode.
1358 (xg_assemble_vliw_tokens): Inline calls to get_frag_is_insn,
1359 get_frag_is_no_transform, is_entry_opcode, and
1360 set_frag_is_specific_opcode. Use get_subseg_total_freq.
1361 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags,
1362 xtensa_fix_close_loop_end_frags, relax_frag_immed, convert_frag_immed):
1363 Inline calls to get_frag_is_no_transform.
1364 (next_instrs_are_b_retw): Inline call to is_windowed_return_opcode.
1365 (xtensa_fix_short_loop_frags): Inline calls to is_the_loop_opcode and
1366 get_frag_is_no_transform.
1367 (convert_frag_immed_finish_loop): Inline calls to get_expression_value
1368 and set_frag_is_no_transform.
1369 (get_expression_value): Delete.
1370 (subseg_map struct): Rename cur_total_freq to total_freq. Rename
1371 cur_target_freq to target_freq.
1372 (get_subseg_info): Split out code to create a new map entry into ...
1373 (add_subseg_info): ... this new function.
1374 (get_last_insn_flags): Check if get_subseg_info succeeded.
1375 (set_last_insn_flags): Call add_subseg_info if needed.
1376 (get_subseg_total_freq, get_subseg_target_freq, set_subseg_freq): New.
1377 (xtensa_reorder_segments): Compute last_sec while counting sections.
1378 Remove call to get_last_sec.
1379 (get_last_sec): Delete.
1380 (cache_literal_section): Inline call to retrieve_literal_seg and its
1381 callees, seg_present and add_seg_list.
1382 (retrieve_literal_seg, seg_present, add_seg_list): Delete.
1383 (get_frag_is_insn, get_frag_is_no_transform,
1384 set_frag_is_specific_opcode, set_frag_is_no_transform): Delete.
1385 * config/tc-xtensa.h (MAX_SLOTS): Reduce from 31 to 15.
1389 * config/tc-xtensa.c: Use ISO C90 formatting.
1390 * config/tc-xtensa.h: Likewise.
1391 * config/xtensa-istack.h: Likewise.
1392 * config/xtensa-relax.c: Likewise.
1393 * config/xtensa-relax.h: Likewise.
1397 * config/tc-arm.c (md_begin): Change EF_ARM_EABI_VER3 to
1400 * doc/c-arm.texi: Document that we actually support -meabi=4, not
1405 * doc/as.texinfo (VTableEntry, VTableInherit): Add "directive" to index
1407 (Acknowledgements): Use "GAS" instead of AS variable.
1411 * config/tc-i386.c: Include "elf/x86-64.h".
1412 (i386_elf_section_type): New function.
1413 * config/tc-i386.h (md_elf_section_type): Define.
1414 (i386_elf_section_type): New prototype.
1418 * config/m68k-parse.h (enum m68k_register): New control register,
1420 * config/tc-m68k.c: Correct control register set for MCF5249.
1425 * config/tc-xtensa.c (absolute_literals_supported): New global flag.
1426 (UNREACHABLE_MAX_WIDTH): Define.
1427 (XTENSA_FETCH_WIDTH): Delete.
1428 (cur_vinsn, xtensa_fetch_width, xt_saved_debug_type, past_xtensa_end,
1429 prefer_const16, prefer_l32r): New global variables.
1430 (LIT4_SECTION_NAME): Define.
1431 (lit4_state struct): Add lit4_seg_name and lit4_seg fields.
1432 (XTENSA_PROP_*, GET_XTENSA_PROP_*, SET_XTENSA_PROP_*): Define.
1433 (frag_flags struct): New.
1434 (xtensa_block_info struct): Move from tc-xtensa.h. Add flags field.
1435 (subseg_map struct): Add cur_total_freq and cur_target_freq fields.
1436 (bitfield, bit_is_set, set_bit, clear_bit): Define.
1437 (MAX_FORMATS): Define.
1438 (op_placement_info struct, op_placement_table): New.
1439 (O_pltrel, O_hi16, O_lo16): Define.
1440 (directiveE enum): Rename directive_generics to directive_transform.
1441 Delete directive_relax. Add directive_schedule,
1442 directive_absolute_literals, and directive_last_directive.
1443 (directive_info): Rename "generics" to "transform". Delete "relax".
1444 Add "schedule" and "absolute-literals".
1445 (directive_state): Adjust entries to match changes in directive_info.
1446 (xtensa_relax_statesE, RELAX_IMMED_MAXSTEPS): Move to tc-xtensa.h.
1447 (xtensa_const16_opcode, xtensa_movi_opcode, xtensa_movi_n_opcode,
1448 xtensa_l32r_opcode, xtensa_nop_opcode, xtensa_rsr_lcount_opcode): New.
1449 (xtensa_j_opcode, xtensa_rsr_opcode): Delete.
1450 (align_only_targets, software_a0_b_retw_interlock,
1451 software_avoid_b_j_loop_end, maybe_has_b_j_loop_end,
1452 software_avoid_short_loop, software_avoid_close_loop_end,
1453 software_avoid_all_short_loops, specific_opcode): Delete.
1454 (warn_unaligned_branch_targets): New.
1455 (workaround_a0_b_retw, workaround_b_j_loop_end, workaround_short_loop,
1456 workaround_close_loop_end, workaround_all_short_loops): Default FALSE.
1457 (option_[no_]link_relax, option_[no_]transform,
1458 option_[no_]absolute_literals, option_warn_unaligned_targets,
1459 option_prefer_l32r, option_prefer_const16, option_target_hardware):
1461 (option_[no_]align_only_targets, option_literal_section_name,
1462 option_text_section_name, option_data_section_name,
1463 option_bss_section_name, option_eb, option_el): Delete.
1464 (md_longopts): Add entries for: [no-]transform, [no-]absolute-literals,
1465 warn-unaligned-targets, prefer-l32r, prefer-const16, [no-]link-relax,
1466 and target-hardware. Delete entries for [no-]target-align-only,
1467 literal-section-name, text-section-name, data-section-name, and
1469 (md_parse_option): Handle new options and remove old ones. Accept but
1470 ignore [no-]density options. Warn for [no-]generics and [no-]relax
1471 and treat them as [no-]transform.
1472 (md_show_usage): Add new options and remove old ones.
1473 (xtensa_setup_hw_workarounds): New.
1474 (md_pseudo_table): Change "word" entry to use xtensa_elf_cons. Add
1475 "long", "short", "loc" and "frequency" entries.
1476 (use_generics): Rename to ...
1477 (use_transform): ... this function. Add past_xtensa_end check.
1478 (use_longcalls): Add past_xtensa_end check.
1479 (code_density_available, can_relax): Delete.
1480 (do_align_targets): New.
1481 (get_directive): Accept dashes in directive names. Warn about
1482 [no-]generics and [no-]relax directives and treat them as
1484 (xtensa_begin_directive): Call md_flush_pending_output only for some
1485 directives. Check for directives inside instruction bundles. Warn
1486 about deprecated ".begin literal" usage. Warn and ignore [no-]density
1487 directives. Handle new directives. Check generating_literals flag
1489 (xtensa_end_directive): Check for directives inside instruction
1490 bundles. Warn and ignore [no-]density directives. Handle new
1491 directives. Call xtensa_set_frag_assembly_state.
1492 (xtensa_loc_directive_seen, xtensa_dwarf2_directive_loc,
1493 xtensa_dwarf2_emit_insn): New.
1494 (xtensa_literal_position): Call md_flush_pending_output. Do not check
1495 use_literal_section flag.
1496 (xtensa_literal_pseudo): Call md_flush_pending_output. Handle absolute
1497 literals. Use xtensa_elf_cons to parse the expression.
1498 (xtensa_literal_prefix): Do not check use_literal_section. Support
1499 ".lit4" sections for absolute literals. Change prefix convention to
1500 replace ".text" (or ".t" in a linkonce section). No need to call
1502 (xtensa_frequency_pseudo, xtensa_elf_cons, xtensa_elf_suffix): New.
1503 (expression_end): Handle closing braces and colons.
1504 (PLT_SUFFIX, plt_suffix): Delete.
1505 (expression_maybe_register): Use new xtensa-isa.h functions. Use
1506 xtensa_elf_suffix instead of checking for plt suffix, and handle O_lo16
1507 and O_hi16 expressions as well.
1508 (tokenize_arguments): Handle closing braces and colons.
1509 (parse_arguments): Use new xtensa-isa.h functions. Handle "invisible"
1510 operands and paired register syntax.
1511 (get_invisible_operands): New.
1512 (xg_translate_sysreg_op): Handle new Xtensa LX RSR/WSR/XSR syntax. Use
1513 new xtensa-isa.h functions.
1514 (xtensa_translate_old_userreg_ops, xtensa_translate_zero_immed): New.
1515 (xg_translate_idioms): Check if inside bundle. Use use_transform.
1516 Handle new Xtensa LX RSR/WSR/XSR syntax. Remove code to widen density
1517 instructions. Use xtensa_translate_zero_immed.
1518 (operand_is_immed, operand_is_pcrel_label): Delete.
1519 (get_relaxable_immed): Use new xtensa-isa.h functions.
1520 (get_opcode_from_buf): Add slot parameter. Use new xtensa-isa.h
1522 (xtensa_print_insn_table, print_vliw_insn): New.
1523 (is_direct_call_opcode): Use new xtensa-isa.h functions.
1524 (is_call_opcode, is_loop_opcode, is_conditional_branch_opcode,
1525 is_branch_or_jump_opcode): Delete.
1526 (is_movi_opcode, decode_reloc, encode_reloc, encode_alt_reloc): New.
1527 (opnum_to_reloc, reloc_to_opnum): Delete.
1528 (xtensa_insnbuf_set_operand, xtensa_insnbuf_get_operand): Use new
1529 xtensa-isa.h functions. Operate on one slot of an instruction.
1530 (xtensa_insnbuf_set_immediate_field, is_negatable_branch,
1531 xg_get_insn_size): Delete.
1532 (xg_get_build_instr_size): Use xg_get_single_size.
1533 (xg_is_narrow_insn, xg_is_single_relaxable_insn): Update calls to
1534 xg_build_widen_table. Use xg_get_single_size.
1535 (xg_get_max_narrow_insn_size): Delete.
1536 (xg_get_max_insn_widen_size, xg_get_max_insn_widen_literal_size,
1537 xg_is_relaxable_insn): Update calls to xg_build_widen_table. Use
1539 (xg_build_to_insn): Record the loc field. Handle OP_OPERAND_HI16U and
1540 OP_OPERAND_LOW16U. Check xg_valid_literal_expression.
1541 (xg_expand_to_stack, xg_expand_narrow): Update calls to
1542 xg_build_widen_table. Use xg_get_single_size.
1543 (xg_immeds_fit): Use new xtensa-isa.h functions. Update call to
1545 (xg_symbolic_immeds_fit): Likewise. Also handle O_lo16 and O_hi16, and
1546 treat weak symbols conservatively.
1547 (xg_check_operand): Use new xtensa-isa.h functions.
1548 (is_dnrange): Delete.
1549 (xg_assembly_relax): Inline previous calls to tinsn_copy.
1550 (xg_finish_frag): Specify separate relax states for the frag and slot0.
1551 (is_branch_jmp_to_next, xg_add_branch_and_loop_targets): Use new
1552 xtensa-isa.h functions.
1553 (xg_instruction_matches_option_term, xg_instruction_matches_or_options,
1554 xg_instruction_matches_options): New.
1555 (xg_instruction_matches_rule): Handle O_register expressions. Call
1556 xg_instruction_matches_options.
1557 (transition_rule_cmp): New.
1558 (xg_instruction_match): Update call to xg_build_simplify_table.
1559 (xg_build_token_insn): Record loc fields.
1560 (xg_simplify_insn): Check is_specific_opcode field and
1561 density_supported flag.
1562 (xg_expand_assembly_insn): Skip checking code_density_available. Use
1563 new xtensa-isa.h functions. Call use_transform instead of can_relax.
1564 (xg_assemble_literal): Add error handling for O_big. Call
1565 record_alignment. Handle O_pltrel.
1566 (xg_valid_literal_expression): New.
1567 (xg_assemble_literal_space): Add slot parameter. Remove call to
1568 set_expr_symbol_offset. Add call to record_alignment. Update call to
1570 (xg_emit_insn): Delete.
1571 (xg_emit_insn_to_buf): Add format parameter. Update calls to
1572 xg_add_opcode_fix and xtensa_insnbuf_to_chars.
1573 (xg_add_opcode_fix): Change opcode parameter to tinsn and add format
1574 and slot parameters. Handle new "alternate" relocations for absolute
1575 literals and CONST16 instructions. Check for bad uses of O_lo16 and
1576 O_hi16. Use new xtensa-isa.h functions.
1577 (xg_assemble_tokens): Delete.
1578 (is_register_writer): Use new xtensa-isa.h functions.
1579 (is_bad_loopend_opcode): Check for xtensa_rsr_lcount_opcode instead of
1580 old-style RSR from LCOUNT.
1581 (next_frag_opcode): Delete.
1582 (next_frag_opcode_is_loop, next_frag_format_size, frag_format_size,
1583 update_next_frag_state): New.
1584 (update_next_frag_nop_state): Delete.
1585 (next_frag_pre_opcode_bytes): Use next_frag_opcode_is_loop.
1586 (xtensa_mark_literal_pool_location): Check use_literal_section flag and
1587 the state of the absolute-literals directive. Add calls to
1588 record_alignment and xtensa_set_frag_assembly_state. Call
1589 xtensa_switch_to_non_abs_literal_fragment instead of
1590 xtensa_switch_to_literal_fragment.
1592 (assemble_nop): Use build_nop. Update call to xtensa_insnbuf_to_chars.
1593 (get_expanded_loop_offset): Change check for undefined opcode to an
1595 (xtensa_set_frag_assembly_state, relaxable_section,
1596 xtensa_find_unmarked_state_frags, xtensa_find_unaligned_branch_targets,
1597 xtensa_find_unaligned_loops, xg_apply_tentative_value): New.
1598 (md_begin): Update call to xtensa_isa_init. Initialize linkrelax to 1.
1599 Set lit4_seg_name. Call xg_init_vinsn. Initialize new global opcodes.
1600 Call init_op_placement_info_table and xtensa_set_frag_assembly_state.
1601 (xtensa_init_fix_data): New.
1602 (xtensa_frob_label): Reset label symbol to the current frag. Check
1603 do_align_targets and generating_literals flag. Propagate frequency
1604 info to new alignment frag. Call xtensa_set_frag_assembly_state.
1605 (xtensa_unrecognized_line): New.
1606 (xtensa_flush_pending_output): Check if inside a bundle. Add a call
1607 to xtensa_set_frag_assembly_state.
1608 (error_reset_cur_vinsn): New.
1609 (md_assemble): Remove check for literal frag. Remove call to
1610 istack_init. Call use_transform instead of use_generics. Parse
1611 explicit instruction format specifiers. Move code for
1612 a0_b_retw_interlock workaround to xg_assemble_vliw_tokens. Call
1613 error_reset_cur_vinsn on errors. Add call to get_invisible_operands.
1614 Add dwarf2_where call. Remote automatic alignment for ENTRY
1615 instructions. Move call to xtensa_clear_insn_labels to the end.
1616 Rearrange to handle bundles.
1617 (xtensa_cons_fix_new): Delete.
1618 (xtensa_handle_align): New.
1619 (xtensa_frag_init): Call xtensa_set_frag_assembly_state. Remove
1620 assignment to is_no_density field.
1621 (md_pcrel_from): Use new xtensa-isa.h functions. Use decode_reloc
1622 instead of reloc_to_opnum. Handle "alternate" relocations.
1623 (xtensa_force_relocation, xtensa_check_inside_bundle,
1624 xtensa_elf_section_change_hook): New.
1625 (xtensa_symbol_new_hook): Delete.
1626 (xtensa_fix_adjustable): Check for difference of symbols with an
1627 offset. Check for external and weak symbols.
1628 (md_apply_fix3): Remove cases for XTENSA_OP{0,1,2} relocs.
1629 (md_estimate_size_before_relax): Return expansion for the first slot.
1630 (tc_gen_reloc): Handle difference of symbols by producing
1631 XTENSA_DIFF{8,16,32} relocs and by writing the value of the difference
1632 into the output. Handle new XTENSA_SLOT*_OP relocs by storing the
1633 tentative values into the output when linkrelax is set.
1634 (XTENSA_PROP_SEC_NAME): Define.
1635 (xtensa_post_relax_hook): Call xtensa_find_unmarked_state_frags.
1636 Create literal tables only if using literal sections. Create new
1637 property tables instead of old instruction tables. Check for unaligned
1638 branch targets and loops.
1639 (finish_vinsn, find_vinsn_conflicts, check_t1_t2_reads_and_writes,
1640 new_resource_table, clear_resource_table, resize_resource_table,
1641 resources_available, reserve_resources, release_resources,
1642 opcode_funcUnit_use_unit, opcode_funcUnit_use_stage,
1643 resources_conflict, xg_find_narrowest_format, relaxation_requirements,
1644 bundle_single_op, emit_single_op, xg_assemble_vliw_tokens): New.
1645 (xtensa_end): Call xtensa_flush_pending_output. Set past_xtensa_end
1646 flag. Update checks for workaround options. Call
1647 xtensa_mark_narrow_branches and xtensa_mark_zcl_first_insns.
1648 (xtensa_cleanup_align_frags): Add special case for branch targets.
1649 Check for and mark unreachable frags.
1650 (xtensa_fix_target_frags): Remove use of align_only_targets flag.
1651 Use RELAX_LOOP_END_BYTES in special case for negatable branch at the
1652 end of a zero-overhead loop body.
1653 (frag_can_negate_branch): Handle instructions with multiple slots.
1654 Use new xtensa-isa.h functions
1655 (xtensa_mark_narrow_branches, is_narrow_branch_guaranteed_in_range,
1656 xtensa_mark_zcl_first_insns): New.
1657 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags): Error if
1658 transformations are disabled.
1659 (next_instrs_are_b_retw): Use new xtensa-isa.h functions. Handle
1660 multislot instructions.
1661 (xtensa_fix_close_loop_end_frags, xtensa_fix_short_loop_frags):
1662 Likewise. Also error if transformations are disabled.
1663 (unrelaxed_frag_max_size): New.
1664 (unrelaxed_frag_min_insn_count, unrelax_frag_has_b_j): Use new
1665 xtensa-isa.h functions.
1666 (xtensa_sanity_check, is_empty_loop, is_local_forward_loop): Use
1667 xtensa_opcode_is_loop instead of is_loop_opcode.
1668 (get_text_align_power): Replace as_fatal with assertion.
1669 (get_text_align_fill_size): Iterate instead of using modulus when
1671 (get_noop_aligned_address): Assert that this is for a machine-dependent
1672 RELAX_ALIGN_NEXT_OPCODE frag. Use next_frag_opcode_is_loop,
1673 xg_get_single_size, and frag_format_size.
1674 (get_widen_aligned_address): Rename to ...
1675 (get_aligned_diff): ... this function. Add max_diff parameter.
1676 Remove handling of rs_align/rs_align_code frags. Use
1677 next_frag_format_size, get_text_align_power, get_text_align_fill_size,
1678 next_frag_opcode_is_loop, and xg_get_single_size. Compute max_diff
1679 and pass it back to caller.
1680 (xtensa_relax_frag): Use relax_frag_loop_align. Add code for new
1681 RELAX_SLOTS, RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN,
1682 RELAX_FILL_NOP, and RELAX_UNREACHABLE frag types. Check relax_seen.
1683 (relax_frag_text_align): Rename to ...
1684 (relax_frag_loop_align): ... this function. Assume loops can only be
1685 in the first slot of an instruction.
1686 (relax_frag_add_nop): Use assemble_nop instead of constructing an OR
1687 instruction. Remove call to frag_wane.
1688 (relax_frag_narrow): Rename to ...
1689 (relax_frag_for_align): ... this function. Extend to handle
1690 RELAX_FILL_NOP and RELAX_UNREACHABLE, as well as RELAX_SLOTS with
1691 RELAX_NARROW for the first slot.
1692 (find_address_of_next_align_frag, bytes_to_stretch): New.
1693 (future_alignment_required): Use find_address_of_next_align_frag and
1694 bytes_to_stretch. Look ahead to subsequent frags to make smarter
1695 alignment decisions.
1696 (relax_frag_immed): Add format, slot, and estimate_only parameters.
1697 Check if transformations are enabled for b_j_loop_end workaround.
1698 Use new xtensa-isa.h functions and handle multislot instructions.
1699 Update call to xg_assembly_relax.
1700 (md_convert_frag): Handle new RELAX_SLOTS, RELAX_UNREACHABLE,
1701 RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN, and RELAX_FILL_NOP
1703 (convert_frag_narrow): Add segP, format and slot parameters. Call
1704 convert_frag_immed for branch instructions. Adjust calls to
1705 tinsn_from_chars, tinsn_immed_from_frag, and xg_emit_insn_to_buf. Use
1706 xg_get_single_size and xg_get_single_format.
1707 (convert_frag_fill_nop): New.
1708 (convert_frag_immed): Add format and slot parameters. Handle multislot
1709 instructions and use new xtensa-isa.h functions. Update calls to
1710 tinsn_immed_from_frag and xg_assembly_relax. Check if transformations
1711 enabled for b_j_loop_end workaround. Use build_nop instead of
1712 assemble_nop. Check is_specific_opcode flag. Check for unreachable
1713 frags. Use xg_get_single_size. Handle O_pltrel.
1714 (fix_new_exp_in_seg): Remove check for old plt flag.
1715 (convert_frag_immed_finish_loop): Update calls to tinsn_from_chars and
1716 xtensa_insnbuf_to_chars. Call tinsn_immed_from_frag. Change check
1717 for loop opcode to an assertion. Mark all frags up to the end of the
1718 loop as not transformable.
1719 (get_last_insn_flags, set_last_insn_flags): Use get_subseg_info.
1720 (get_subseg_info): New.
1721 (xtensa_move_literals): Call xtensa_set_frag_assembly_state. Add null
1723 (xtensa_switch_to_literal_fragment): Rewrite to handle absolute
1724 literals and use xtensa_switch_to_non_abs_literal_fragment otherwise.
1725 (xtensa_switch_to_non_abs_literal_fragment): New.
1726 (cache_literal_section): Add is_code parameter and pass it through to
1727 retrieve_literal_seg.
1728 (retrieve_literal_seg): Add is_code parameter and use it to set the
1729 flags on the literal section. Handle case where head parameter is 0.
1730 (get_frag_is_no_transform, set_frag_is_specific_opcode,
1731 set_frag_is_no_transform): New.
1732 (xtensa_create_property_segments): Add end_property_function parameter
1733 and pass it through to add_xt_block_frags. Call bfd_get_section_flags
1734 and skip SEC_DEBUGGING and !SEC_ALLOC sections.
1735 (xtensa_create_xproperty_segments, section_has_xproperty): New.
1736 (add_xt_block_frags): Add end_property_function parameter and call it
1737 if it is non-zero. Call xtensa_frag_flags_init.
1738 (xtensa_frag_flags_is_empty, xtensa_frag_flags_init,
1739 get_frag_property_flags, frag_flags_to_number,
1740 xtensa_frag_flags_combinable, xt_block_aligned_size,
1741 xtensa_xt_block_combine, add_xt_prop_frags,
1742 init_op_placement_info_table, opcode_fits_format_slot,
1743 xg_get_single_size, xg_get_single_format): New.
1744 (istack_push): Inline call to tinsn_copy.
1745 (tinsn_copy): Delete.
1746 (tinsn_has_invalid_symbolic_operands): Handle O_hi16 and O_lo16 and
1747 CONST16 opcodes. Handle O_big, O_illegal, and O_absent.
1748 (tinsn_has_complex_operands): Handle O_hi16 and O_lo16.
1749 (tinsn_to_insnbuf): Use xg_get_single_format and new xtensa-isa.h
1750 functions. Handle invisible operands.
1751 (tinsn_to_slotbuf): New.
1752 (tinsn_check_arguments): Use new xtensa-isa.h functions.
1753 (tinsn_from_chars): Add slot parameter. Rewrite using xg_init_vinsn,
1754 vinsn_from_chars, and xg_free_vinsn.
1755 (tinsn_from_insnbuf): New.
1756 (tinsn_immed_from_frag): Add slot parameter and handle multislot
1757 instructions. Handle symbol differences.
1758 (get_num_stack_text_bytes): Use xg_get_single_size.
1759 (xg_init_vinsn, xg_clear_vinsn, vinsn_has_specific_opcodes,
1760 xg_free_vinsn, vinsn_to_insnbuf, vinsn_from_chars, expr_is_register,
1761 get_expr_register, set_expr_symbol_offset_diff): New.
1762 * config/tc-xtensa.h (MAX_SLOTS): Define.
1763 (xtensa_relax_statesE): Move from tc-xtensa.c. Add
1764 RELAX_CHECK_ALIGN_NEXT_OPCODE, RELAX_MAYBE_DESIRE_ALIGN, RELAX_SLOTS,
1765 RELAX_FILL_NOP, RELAX_UNREACHABLE, RELAX_MAYBE_UNREACHABLE, and
1767 (RELAX_IMMED_MAXSTEPS): Move from tc-xtensa.c.
1768 (xtensa_frag_type struct): Add is_assembly_state_set,
1769 use_absolute_literals, relax_seen, is_unreachable, is_specific_opcode,
1770 is_align, is_text_align, alignment, and is_first_loop_insn fields.
1771 Replace is_generics and is_relax fields by is_no_transform field.
1772 Delete is_text and is_longcalls fields. Change text_expansion and
1773 literal_expansion to arrays of MAX_SLOTS entries. Add arrays of
1774 per-slot information: literal_frags, slot_subtypes, slot_symbols,
1775 slot_sub_symbols, and slot_offsets. Add fr_prev field.
1776 (xtensa_fix_data struct): New.
1777 (xtensa_symfield_type struct): Delete plt field.
1778 (xtensa_block_info struct): Move definition to tc-xtensa.h. Add
1779 forward declaration here.
1780 (xt_section_type enum): Delete xt_insn_sec. Add xt_prop_sec.
1781 (XTENSA_SECTION_RENAME): Undefine.
1782 (TC_FIX_TYPE, TC_INIT_FIX_DATA, TC_FORCE_RELOCATION, NO_PSEUDO_DOT,
1783 tc_unrecognized_line, md_do_align, md_elf_section_change_hook,
1784 HANDLE_ALIGN, TC_LINKRELAX_FIXUP, SUB_SEGMENT_ALIGN): Define.
1785 (TC_CONS_FIX_NEW, tc_symbol_new_hook): Delete.
1786 (unit_num_copies_func, opcode_num_units_func,
1787 opcode_funcUnit_use_unit_func, opcode_funcUnit_use_stage_func): New.
1788 (resource_table struct): New.
1789 * config/xtensa-istack.h (MAX_INSN_ARGS): Increase from 6 to 10.
1790 (TInsn struct): Add keep_wide, loc, fixup, record_fix, subtype,
1791 literal_space, symbol, sub_symbol, offset, and literal_frag fields.
1792 (tinsn_copy): Delete prototype.
1793 (vliw_insn struct): New.
1794 * config/xtensa-relax.c (insn_pattern_struct): Add options field.
1795 (widen_spec_list): Add option conditions for density and boolean
1796 instructions. Add expansions using CONST16 and conditions for using
1797 CONST16 vs. L32R. Use new Xtensa LX RSR/WSR syntax. Add entries for
1799 (simplify_spec_list): Add option conditions for density instructions.
1800 Add entry for NOP instruction.
1801 (append_transition): Add cmp function pointer parameter and use it to
1802 insert the new entry in order.
1803 (operand_function_LOW16U, operand_function_HI16U): New.
1804 (xg_has_userdef_op_fn, xg_apply_userdef_op_fn): Handle
1805 OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
1806 (enter_opname, split_string): Use xstrdup instead of strdup.
1807 (init_insn_pattern): Initialize new options field.
1808 (clear_req_or_option_list, clear_req_option_list,
1809 clone_req_or_option_list, clone_req_option_list, parse_option_cond):
1811 (parse_insn_pattern): Parse option conditions.
1812 (transition_applies): New.
1813 (build_transition): Use new xtensa-isa.h functions. Fix incorrectly
1814 swapped last arguments in calls to append_constant_value_condition.
1815 Call clone_req_option_list. Add warning about invalid opcode.
1816 Handle LOW16U and HI16U function names.
1817 (build_transition_table): Add cmp parameter and use it in calls to
1818 append_transition. Use new xtensa-isa.h functions. Check
1819 transition_applies before adding entries.
1820 (xg_build_widen_table, xg_build_simplify_table): Add cmp parameter and
1821 pass it through to build_transition_table.
1822 * config/xtensa-relax.h (ReqOrOptionList, ReqOrOption, ReqOptionList,
1823 ReqOption, transition_cmp_fn): New types.
1824 (OpType enum): Add OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
1825 (transition_rule struct): Add options field.
1826 * doc/as.texinfo (Overview): Update Xtensa options.
1827 * doc/c-xtensa.texi (Xtensa Options): Delete --[no-]density,
1828 --[no-]relax, and --[no-]generics options. Update descriptions of
1829 --text-section-literals and --[no-]longcalls. Add
1830 --[no-]absolute-literals and --[no-]transform.
1831 (Xtensa Syntax): Add description of syntax for FLIX instructions.
1832 Remove use of "generic" and "specific" terminology for opcodes.
1833 (Xtensa Registers): Generalize the syntax description to include
1834 user-defined register files.
1835 (Xtensa Automatic Alignment): Update.
1836 (Xtensa Branch Relaxation): Mention limitation of unconditional jumps.
1837 (Xtensa Call Relaxation): Linker can now remove most of the overhead.
1838 (Xtensa Directives): Remove confusing rules about precedence.
1839 (Density Directive, Relax Directive): Delete.
1840 (Schedule Directive): New.
1841 (Generics Directive): Rename to ...
1842 (Transform Directive): ... this node.
1843 (Literal Directive): Update for absolute literals. Missing
1844 literal_position directive is now an error.
1845 (Literal Position Directive): Update for absolute literals.
1846 (Freeregs Directive): Delete.
1847 (Absolute Literals Directive): New.
1848 (Frame Directive): Minor editing.
1849 * Makefile.am (DEPTC_xtensa_elf, DEPOBJ_xtensa_elf, DEP_xtensa_elf):
1850 Update dependencies.
1851 * Makefile.in: Regenerate.
1855 * config/tc-mips.c (append_insn): Use fix_new rather than fix_new_exp
1856 to build the second and third fixups for a composite relocation.
1857 (macro_read_relocs): New function.
1858 (macro_build): Use it.
1859 (s_cpsetup): Pass all three composite relocation codes to macro_build.
1860 Simplify fragging code accordingly.
1861 (s_gpdword): Use fix_new rather than fix_new_exp for the second part
1862 of the composite relocation. Set fx_tcbit in both fixups.
1866 * config/tc-mips.c (append_insn): Set fx_tcbit for composite relocs.
1867 (md_apply_fix3): Don't treat composite relocs as done.
1871 * macro.c (macro_expand_body): When ELF, use .LL rather than LL as
1872 prefix for symbol names generated from the LOCAL macro directive.
1874 * dw2gencfi.c (select_cie_for_fde): When separating CIE out from
1875 FDE, treat a DW_CFA_remember_state as we do a DW_CFA_advance_loc.
1879 * config/tc-crx.c (preprocess_reglist): Handle Co-processor
1881 (md_assemble): Add error checking for Co-Processor instructions.
1882 (get_cinv_parameters): Add 'b' option to invalidate the
1883 branch-target cache.
1887 * config/tc-arm.c (unwind): New variable.
1888 (vfp_sp_encode_reg): New function.
1889 (vfp_sp_reg_required_here): Use it.
1890 (vfp_sp_reg_list, vfp_dp_reg_list): Remove.
1891 (vfp_parse_reg_list): New function.
1892 (s_arm_unwind_fnstart, s_arm_unwind_fnend, s_arm_unwind_cantunwind,
1893 s_arm_unwind_personality, s_arm_unwind_personalityindex,
1894 s_arm_unwind_handlerdata, s_arm_unwind_save, s_arm_unwind_movsp,
1895 s_arm_unwind_pad, s_arm_unwind_setfp, s_arm_unwind_raw): New
1897 (md_pseudo_table): Add them.
1898 (do_vfp_reg2_from_sp2): Use vfp_parse_reg_list and vfp_sp_encode_reg.
1899 (do_vfp_sp2_from_reg2, vfp_sp_ldstm, vfp_dp_ldstm): Ditto.
1900 (set_section, add_unwind_adjustsp, flush_pending_unwind,
1901 finish_unwind_opcodes, start_unwind_section, create_unwind_entry,
1902 require_hashconst, add_unwind_opcode): New functions.
1903 * doc/c-arm.texi: Document unwinding opcodes.
1904 * NEWS: Mention the new feature.
1908 * config/tc-mips.c (md_apply_fix3): Remove erroneous assert.
1912 * config/tc-ppc.c (md_apply_fix3): Call S_SET_THREAD_LOCAL for
1914 * config/tc-s390.c (md_apply_fix3): Likewise.
1915 * config/tc-sparc.c (md_apply_fix3): Likewise.
1919 * config/tc-arm.c (arm_elf_section_type): New function.
1920 (arm_elf_change_section): Set section link for exidx sections.
1921 * config/tc-arm.h (arm_elf_section_type): Add prototype.
1922 (md_elf_section_type): Define.
1926 * config/tc-pdp11.c (md_apply_fix3): Change to sign of the SOB
1927 instruction's offset.
1931 * (TARGET_FORMAT): Remove LynxOS COFF definition.
1935 * config/tc-arc.c (tc_gen_reloc): Don't assume fixP->fx_addsy is an
1936 asymbol *, instead use symbol_get_bfdsym.
1940 * config/tc-m68k.c (select_control_regs): Add mcf5249.
1944 * config/tc-arm.c (do_smi, do_nop): New functions.
1945 (insns): Add ARMv6ZK instructions.
1946 (md_apply_fix3): Handle BFD_RELOC_ARM_SMI.
1947 (tc_gen_reloc): Ditto.
1948 (arm_cpus): Add mpcore and arm1176.
1949 (arm_archs): Add armv6{k,z,zk}.
1950 * doc/c-arm.texi: Document new cores and architectures.
1954 * config/tc-arm.c: Use ISO C90 formatting.
1958 * config/tc-arm.c (mav_reg_required_here): Allow REG_TYPE_CN
1959 as alternative when REG_TYPE_MVF, REG_TYPE_MVD, REG_TYPE_MVFX or
1960 REG_TYPE_MVDX is expected.
1964 * doc/c-i386.texi (i386-Mnemonics): Fix typo.
1968 * config/tc-ia64.c (ENCODED_PSP_OFFSET): New.
1969 (output_rp_psprel, output_pfs_psprel, output_preds_psprel,
1970 output_spill_base, output_unat_psprel, output_lc_psprel,
1971 output_fpsr_psprel, output_priunat_psprel, output_bsp_psprel,
1972 output_bsprestore_psprel, output_rnat_psprel, output_spill_psprel,
1973 output_spill_psprel_p): Use it.
1977 * config/tc-crx.c (handle_LoadStor): New function.
1978 Handle load/stor unique instructions before parsing.
1982 * config/tc-arm.c (s_arm_rel31): New funciton.
1983 (md_pseudo_table): Add .rel31.
1984 (md_apply_fix3): Handle BFD_RELOC_ARM_TARGET2,
1985 BFD_RELOC_32_PCREL and BFD_RELOC_ARM_PREL31.
1986 (tc_gen_reloc): Handle BFD_RELOC_ARM_PREL31 and BFD_RELOC_ARM_TARGET2.
1987 (arm_fix_adjustable): Return 0 for BFD_RELOC_ARM_TARGET2.
1988 (arm_parse_reloc): Add (target2).
1992 * Makefile.am: Run "make dep-am".
1993 * Makefile.in: Regenerate.
1994 * aclocal.m4: Regenerate.
1995 * configure: Regenerate.
1996 * doc/Makefile.in: Regenerate.
1997 * po/POTFILES.in: Regenerate.
1998 * po/gas.pot: Regenerate.
2002 * config/tc-mmix.c [!LLONG_MIN]: Correct #elsif to #elif.
2003 [!LLONG_MAX]: Ditto.
2007 * config/tc-arm.c: Rename RELABS to TARGET1.
2011 * messages.c (as_internal_value_out_of_range): Cast values passed
2012 to as_bad_where or as_warn_where to proper type.
2016 * config/tc-avr.c: Add support for
2017 atmega48, atmega88, atmega168, attiny13, attiny2313, at90can128.
2021 * dw2gencfi.c (select_cie_for_fde): When separating CIE out
2022 from FDE, treat a CFI_escape as we do a DW_CFA_advance_loc.
2026 * config/obj-elf.c (obj_elf_section_type): Handle init_array,
2027 fini_array and preinit_array section types.
2028 * config/tc-ia64.c (ia64_elf_section_type): Remove init_array
2030 * doc/as.texinfo: Document extra section types.
2034 * Makefile.am (TARG_ENV_HFILES): Add te-symbian.h.
2035 * Makefile.in: Regenerated.
2036 * configure.in: Set em for arm*-*-symbianelf*.
2037 * configure: Regenerated.
2038 * config/tc-arm.c (elf32_arm_target_format): Use Symbian target
2039 vectors when appropriate.
2040 * config/te-symbian.h: New file.
2044 * config/tc-crx.c (gettrap): Exception vector can be case
2046 (process_label_constant): Fix a 32-bit displacement bug in branch
2048 (get_operandtype) : Bug fix, wrong operand was used.
2049 (process_label_constant): Initialize relocation type to
2054 * tc-arm.c (arm_cpus, arm_fpus): Allow <cpu>-s as well as <cpu>s
2055 for synthesizable cores.
2057 * doc/c-arm.texi (ARM Options): Document canonical names of CPUs.
2061 * config/tc-msp430.c: Clean-up the code.
2062 (md_relax_table): New relax table.
2063 (mcu_types): Sort MCU types.
2064 (md_pseudo_table): Add .profiler pseudo handler.
2065 (pow2value): New function.
2066 (msp430_profiler): New function.
2067 (msp430_operands): Add new insns handlers.
2068 (msp430_srcoperand): Add register operand handler, allow complex
2070 (md_estimate_size_before_relax): Rewritten.
2071 (md_convert_frag): Rewritten.
2072 (msp430_relax_frag): New function.
2073 * config/tc-msp430.h (md_relax_frag): define macro
2074 * doc/c-msp430.texi: Update information.
2078 * as.c (std_shortopts): Allow -g to take an optional argument.
2079 (parse_args): Pass any switch starting with -g on to the backend
2084 * configure.in (arm*-*-symbianelf*): New target.
2085 (arm*-*-eabi*): Likewise.
2086 * configure: Regenerated.
2089 * config/tc-mips.c (append_insn): Handle delay slots in branch likely
2094 * config/tc-ia64.c (start_unwind_section): Add linkonce_empty
2095 argument, don't do anything if current section is not
2096 .gnu.linkonce.t.* and linkonce_empty is set.
2097 (generate_unwind_image, dot_endp): Adjust callers, call
2098 start_unwind_section (*, 1) if nothing will be put into the
2103 * as.c (MD_DEBUG_FORMAT_SELECTOR): Provide default definition.
2104 (show_usage): Add -g.
2105 (std_longopts): Add --gen-debug. Alpha sort the table.
2106 (parse_args): Print an error message if a switch is not handled.
2107 Handle the -g switch, calling md_debug_format_selector() if
2109 * NEWS: Mention new feature.
2110 * doc/as.texinfo: Document new switch.
2111 * doc/internals.texi: Document behaviour of md_parse_option.
2113 * config/tc-arm.c (md_parse_option): Do not issue an error message
2114 if the switch is not recognised.
2115 * config/tc-m68k.c (md_parse_option): Likewise.
2116 * config/tc-pdp11.c (md_parse_option): Likewise.
2117 * config/tc-v850.c (md_parse_option): Likewise.
2119 * as.h: Fix up formatting.
2124 * macro.c (macro_set_alternate): Use ISO C90 formatting.
2126 * configure.in: Sort architecture based tables alphabetically.
2127 * configure: Regenerate.
2131 * config/tc-ppc.c (tc_ppc_regname_to_dw2regnum <regnames>): Replace
2132 { "cc", 68 }, with { "cr", 70 }.
2137 * as.c: Add and handle new --alternate command line option.
2138 * macro.c (macro_set_alternate): New.
2139 * macro.h (macro_set_alternate): Declare.
2140 * read.c: Add and handle new .altmacro and .noaltmacro directives.
2141 * doc/as.texinfo: Document new command line option and pseudo-ops
2142 as well as insert documentation originating from gasp about
2143 alternate macro syntax.
2144 * NEWS: Mention new command line option and pseudo-ops.
2148 * expr.c (operand): Handle the "~", "-", and "!" operators applied
2153 * config/tc-arm.c (md_apply_fix3, tc_gen_reloc, arm_parse_reloc):
2154 Handle new relocations.
2155 * include/elf/arm.h (elf_arm_reloc_type): Add new EABI relocations.
2159 * write.c (relax_segment): Use was_address instead of address when
2160 setting fr_fix field for align frag due to backwards .org.
2164 Introduce SH2a support.
2166 * config/tc-sh.c (get_specific): Change arch_sh2a_up to
2169 * config/tc-sh.c (md_parse_option): Add sh2a-nofpu ISA handling.
2171 * config/tc-sh.c (sh_elf_final_processing): Move sh2a recognition
2172 to end of conditional expression.
2174 * config/tc-sh.c: Add sh2a-nofpu support.
2176 * tc-sh.c: Add sh2a support.
2177 (parse_reg): Add tbr.
2178 (parse_at): Support @@(disp,tbr).
2179 (get_specific): Support sh2a opcodes.
2180 (insert4): New, for 4 byte relocs.
2181 (build_Mytes): Support sh2a opcodes.
2182 (md_apply_fix3_Mytes): Support sh2a opcodes.
2184 * config/tc-sh.c (md_parse_option): Handle sh2a.
2185 (sh_elf_final_processing): Ditto.
2189 * config/tc-hppa.h (TARGET_FORMAT): Set to "elf32-hppa-netbsd"
2194 * config/tc-ppc.c (ppc_frob_file_before_adjust): Warn if .toc too big.
2195 (ppc_arch): Expand comment.
2199 * config/tc-crx.c: Support evaluating the difference between two
2201 * config/tc-crx.h: Likewise.
2205 * config/tc-ia64.c (start_unwind_section): Set the linked-to
2207 (ia64_elf_section_change_hook): Set the linked-to section for
2212 * config/tc-msp430.c: Add new subtargets: msp430x1610,
2213 msp430x1611, msp430x1612, msp430x415, msp430x417, msp430xG437,
2214 msp430xG438, msp430xG439.
2218 * doc/as.texinfo (Section, PushSection): Correct documentation
2223 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
2224 permissible for the selected instruction suffix.
2225 (match_template): Don't permit 64-bit general purpose operands in
2227 (finalize_imm): Permit 64-bit immediates.
2228 (build_modrm_byte): Don't treat 32-bit addressing in 64-bit mode
2229 specially except for the width of the used base and/or index
2230 registers. For 32-bit displacements, use sign-extended
2231 relocations only when using 64-bit addressing.
2232 Force zero displacement on rip-relative addressing when there is
2233 no other displacement.
2234 (i386_index_check): Don't treat 32-bit addressing in 64-bit mode
2235 specially except for the width of the used base and/or index
2237 (parse_register): Disallow Reg64 registers in 32-bit mode.
2239 * config/tc-i386.c: For DefaultSize instructions, don't guess a 'q'
2240 suffix if the instruction doesn't support it.
2244 * config/tc-mips.c (append_insn): Handle constant expressions with
2245 no associated relocation.
2246 (mips_ip): Cancel the expression after use for the Q format
2248 (parse_relocation): Return no relocation for unsupported
2250 (my_getSmallExpression): Return no relocation if no relocation
2255 * config/obj-som.c (adjust_stab_sections): Add prototype.
2256 (obj_som_compiler, obj_som_version, obj_som_copyright,
2257 adjust_stab_sections): Add ATTRIBUTE_UNUSED to unused arguments.
2258 * config/tc-hppa.c (update_subspace): Likewise.
2259 (is_defined_subspace): Amplify comment.
2260 * config/obj-som.h (som_frob_file): Add prototype.
2265 * subsegs.c (section_symbol): Don't create a new segment when
2266 existing segment is undefined.
2270 * config/tc-arm.c: Include include/opcode/arm.h.
2271 (ARM_EXT_*, ARM_ARCH_*, ARM_ANY, ARM_ALL, COPROC_ANY): Delete.
2272 (FPU_FPA_EXT_* FPU_VFP_EXT_*, FPU_ANY, FPU_NONE, FPU_MAVERICK): Delete.
2273 (FPU_ARCH_*): Delete.
2274 * Makefile.am: Update dependencies.
2275 * Makefile.in: Regenerate.
2279 * configure.in: Accept armbe as a big-endian arm configuration.
2280 * configure: Regenerate.
2284 * config/tc-i386.c (T_SHIFTOP): New constant.
2285 (intel_e05_1): Handle '&', '|' and T_SHIFTOP.
2286 (intel_el1): Handle '~'.
2287 (intel_get_token): Handle '<>', '&', '|' and '~'.
2291 (md_assemble): Remove spurious newline from end of as_bad error
2293 (intel_e05_1): Likewise.
2294 (intel_e11): Likewise.
2295 (intel_match_token): Likewise.
2299 * config/tc-m68k.c: Convert to C90. Remove redundant
2300 declarations. Indentation fixup.
2301 [M68KCOFF]: Include "obj-coff.h" instead of declaring
2302 obj_coff_section ourselves.
2306 * config/tc-ia64.c (default_big_endian): New.
2307 (dot_byteorder, md_begin): Use it.
2308 (md_parse_option): Set it.
2312 * configure.in: Change sh-sybmian-elf to sh-*-symbianelf.
2313 * configure: Regenerate.
2314 * NEWS: Change sh-sybmian-elf to sh-*-symbianelf.
2315 * config/tc-sh.c (sh_elf_final_processing): Use renamed version of
2316 sh_find_elf_flags if necessary.
2320 * config/tc-mips.c (mips_fix_adjustable): If the full addend is
2321 going to be split into more than one in-place addend, return 0
2322 for relocations against mergeable sections. Associate comments
2327 * Makefile.am (CPU_TYPES): Add crx.
2328 (TARGET_CPU_CFILES): Add config/tc-crx.c.
2329 (TARGET_CPU_HFILES): Add config/tc-crx.h.
2330 (DEPTC_crx_elf): New target.
2331 (DEPOBJ_crx_elf): Likewise.
2332 (DEP_crx_elf): Likewise.
2333 * Makefile.in: Regenerate.
2334 * configure.in: Add crx* target.
2335 * configure: Regenerate.
2336 * config/tc-crx.c: New file.
2337 * config/tc-crx.h: New file.
2338 * NEWS: Mention new target.
2342 * config.in: Undefine TARGET_SYMBIAN by default.
2344 * configure: Regenerate. Add sh-symbian-elf target. If
2345 selected define TARGET_SYMBIAN.
2346 * config/tc-sh.h (TARGET_FORMAT): Select a Symbian target
2347 format if TARGET_SYMBIAN has been defined.
2349 * output-file.c (output_file_create): Report the target format
2350 chosen when bfd_openw reports that it is invalid.
2352 * config/obj-coff.c (coff_pseudo_table): Only define the weak
2353 pseudo for BFD based assemblers.
2358 * config/tc-sh.c (md_assemble): Change isspace to ISSPACE.
2359 (md_parse_option): Remove redundant -isa testing.
2360 Make bfd_arch variable const.
2361 (md_show_usage): Make bfd_arch variable const.
2365 * config/tc-ia64.c (emit_one_bundle): Check and set insn_addr.
2366 * config/tc-ia64.h (md_frag_check): Define.
2370 * config/obj-coff.c (obj_coff_weak): New .weak syntax for PE weak
2372 * doc/as.texinfo (Weak): Document PE weak symbols.
2376 * config/tc-mips.c (HAVE_IN_PLACE_ADDENDS): New macro.
2377 (reloc_needs_lo_p): Only return true if HAVE_IN_PLACE_ADDENDS.
2378 (mips_frob_file): Rework so that only a single pass through the
2379 relocs is needed. Allow %lo()s to have higher offsets than their
2380 corresponding %hi()s or %got()s.
2384 * config/tc-arm.c (md_apply_fix3:BFD_RELOC_ARM_IMMEDIATE): Do not
2385 allow values which have come from undefined symbols.
2386 Always consider this fixup to have been processed as a reloc
2387 cannot be generated for it.
2391 * frags.h (struct frag): Add has_code and insn_addr fields.
2392 * write.c (cvt_frag_to_fill): Invoke md_frag_check.
2393 * config/tc-ppc.c (md_assemble): Check and set insn_addr.
2394 * config/tc-ppc.h (md_frag_check): Define.
2398 * doc/Makefile.am (info): Rename goal to...
2399 (info-local): ... this, to preserve implicit dependencies.
2400 * doc/Makefile.in: Regenerate with automake 1.8.5.
2404 * config/tc-m32r.c (md_convert_frag): Changed for @PLT.
2405 (m32r_cgen_record_fixup_exp): Changed for @GOTOFF, @GOT.
2406 (m32r_fix_adjustable): Changed for @GOTOFF, @GOT, @PLT.
2407 (tc_gen_reloc): Likewise.
2408 (m32r_end_of_match): Add for @GOTOFF, @GOT, @PLT.
2409 (m32r_parse_name): Likewise.
2410 (m32r_cgen_parse_fix_exp): Likewise.
2411 * config/tc-m32r.h (md_parse_name): Define for @GOTOFF, @GOT, @PLT.
2412 (O_PIC_reloc): Likewise.
2413 (TC_CGEN_PARSE_FIX_EXP): Likewise..
2414 * cgen.c (gas_cgen_parse_operand): Add TC_CGEN_PARSE_FIX_EXP
2415 for @GOTOFF, @GOT, @PLT.
2419 * gas/symbols.c: While discarding ordinary local absolute symbols
2420 when --strip-local-absolute is in effect, retain file symbols.
2424 * config/tc-m68k.c (mri_chip): Replace current_chip, not augment.
2425 (md_parse_option): Likewise.
2429 * config/tc-i386.c: Deal with LEX_QM the same way as with LEX_AT.
2430 * config/te-netware.h: New file.
2431 * config/te-ppcnw.h: Delete: Obsolete.
2432 * configure.in: Eliminate ill NetWare targets. Make generic
2433 NetWare target use proper emulation.
2434 * Makefile.am: Eliminate reference to obsolete te-ppcnw.h, add
2435 reference to new te-netware.h.
2436 * configure: Regenerate.
2437 * Makefile.in: Regenerate.
2441 * config/tc-s390.c (s390_insn): Avoid incorrect signed/unsigned
2442 comparison in .insn pseudo operation.
2446 * config/obj-coff.c (coff_adjust_section_syms): Use
2447 bfd_get_section_size instead of bfd_get_section_size_before_reloc.
2448 (coff_frob_section): Likewise.
2449 * config/tc-mips.c (md_apply_fix3): Likewise.
2450 * config/obj-elf.c (elf_frob_file): Use bfd_set_section_size.
2451 (elf_frob_file_after_relocs): Likewise.
2455 * config/tc-hppa.c (log2): Rename to exact_log2.
2456 (pa_next_subseg): Delete unused function.
2457 (create_new_space): Mark unused arguments with ATTRIBUTE_UNUSED.
2458 (create_new_subspace): Likewise.
2461 * config/tc-hppa.c (hppa_fix_adjustable): Allow reduction of fake
2462 labels. Fix warning.
2466 * config/tc-mn10300.h (tc_fix_adjustable): Define.
2467 * config/tc-mn10300.c (mn10300_fix_adjustable): Don't adjust debug
2468 or non-merged symbols.
2472 * config/tc-ia64.c (remove_marked_resource): Save, clear and
2473 restore the old slot when inserting srlz.i/srlz.d.
2477 * Makefile.am: Regenerate dependecies.
2478 * Makefile.in: Regenerate.
2479 * config/tc-sh.c (valid_arch): Make unsigned.
2480 (preset_target_arch): Likewise.
2481 (md_begin): Use new architecture flags system.
2482 (get_specific): Likewise.
2483 (assemble_ppi): Likewise.
2484 (md_assemble): Likewise. Also fix error check for bad opcodes.
2485 (md_parse_option): Likewise. Also generate -isa values according
2486 to the table in bfd/cpu-sh.c instead of just constants. Also
2487 allow <arch>-up ISA variants.
2488 (sh_elf_final_processing): Replace if-else chain with a call to
2489 sh_find_elf_flags().
2490 * testsuite/gas/sh/arch: New directory.
2491 * testsuite/gas/sh/arch/arch.exp: New test script.
2492 * testsuite/gas/sh/arch/arch_expected.txt: New file.
2493 * testsuite/gas/sh/arch/sh.s: New file.
2494 * testsuite/gas/sh/arch/sh2.s: New file.
2495 * testsuite/gas/sh/arch/sh-dsp.s: New file.
2496 * testsuite/gas/sh/arch/sh2e.s: New file.
2497 * testsuite/gas/sh/arch/sh3-nommu.s: New file.
2498 * testsuite/gas/sh/arch/sh3.s: New file.
2499 * testsuite/gas/sh/arch/sh3-dsp.s: New file.
2500 * testsuite/gas/sh/arch/sh3e.s: New file.
2501 * testsuite/gas/sh/arch/sh4-nommu-nofpu.s: New file.
2502 * testsuite/gas/sh/arch/sh4-nofpu.s: New file.
2503 * testsuite/gas/sh/arch/sh4.s: New file.
2504 * testsuite/gas/sh/arch/sh4a-nofpu.s: New file.
2505 * testsuite/gas/sh/arch/sh4al-dsp.s: New file.
2506 * testsuite/gas/sh/arch/sh4a.s: New file.
2510 * config/gc-m68k.c(m68k_ip): Convert mode 5 addressing
2511 with zero offset into mode 2 addressing to save a word.
2515 * config/tc-ia64.c (ar_is_in_integer_unit): Removed.
2516 (ar_is_only_in_integer_unit): New.
2517 (ar_is_only_in_memory_unit): New.
2518 (generate_unwind_image): Silence gcc on 32bit host.
2519 (md_assemble): Use ar_is_only_in_integer_unit instead of
2520 ar_is_in_integer_unit. Check AR access.
2524 * config/tc-m68k.c (md_begin): Sort the opcode table into
2526 (m68k_compare_opcode): New function to do the sorting.
2530 * config/m68k-parse.y(operand): Allow for MAC/EMAC mask
2531 addressing on MIT style operands.
2532 * config/m68k-parse.y(yylex): Allow '-&' for predecrement
2534 * config/tc-m68k.c(install_operand): Comment 'G' and 'H' type
2539 * expr.c (operand, operator): Don't reject '++' and '--'.
2543 * config/tc-mips.c (append_insn): Use ISA-encoded addresses in MIPS16
2548 * configure.in: Add ppc-*-lynxos*. Update i386-*-lynxos* to ELF.
2549 * configure: Regenerate.
2553 * dw2gencfi.c (output_cie): Handle dwarf3 format CIE entries.
2557 * configure.in (or32-*-rtems*): Switch to elf.
2558 * configure: Regenerate.
2562 * po/fr.po: Updated French translation.
2566 * doc/as.texinfo (Section): Document G and T flags to .section
2567 directive. Document the extra arguments that the G flag
2568 requires. Document the #tls flag.
2572 * subsegs.c (section_symbol): Create a new section symbol if
2573 the existing one doesn't match.
2574 * symbols.c (symbol_set_bfdsym): Don't reset BFD section symbol.
2578 * config/tc-mips.c (append_insn, mips_emit_delays): Extend -mfix-vr4120
2579 to cope with VR4181A errata MD(1) and MD(4).
2583 * NEWS: Mention .secrel32 for pe[i]-i386.
2587 * config/tc-frv.h (MAX_MEM_FOR_RS_ALIGN_CODE): New.
2588 (HANDLE_ALIGN): New.
2592 * Makefile.am (DIST_SUBDIRS): Define.
2593 * aclocal.m4: Regenerate with automake 1.8.4.
2594 * Makefile.in: Likewise.
2595 * doc/Makefile.in: Likewise.
2599 * config/tc-ia64.c (dot_serialize): Declare.
2600 (dot_serialize): New function.
2601 (md_pseudo_table): Add ".serialize.data" and
2602 ".serialize.instruction" directives.
2606 * messages (as_internal_value_out_of_range): Print a message about
2607 a value being out of range. Be consistent about whether the
2608 values are printed in decimal or hexadecimal.
2609 (as_warn_value_out_of_range): Generate a warning message about an
2611 (as_bad_value_out_of_range): Generate an error message about an
2613 * as.h: Prototype the new functions.
2614 * config/tc-alpha.c (insert_operand): Use new function.
2615 * config/tc-arc.c (arc_insert_operand): Likewise.
2616 * config/tc-mn10200.c (mn10200_insert_operand): Likewise.
2617 * config/tc-mn10300.c (mn10300_insert_operand): Likewise.
2618 * config/tc-ppc.c (ppc_insert_operand): Likewise.
2619 * config/tc-s390.c (s390_insert_operand): Likewise.
2620 * config/tc-v850.c (v850_insert_operand): Likewise.
2624 * configure.in: Set em=linux for frv-*-*linux*.
2625 * configure: Rebuilt.
2626 * config/tc-frv.h (TARGET_FORMAT): Use elf32-frvfdpic if...
2627 (frv_md_fdpic_enabled): New.
2628 * config/tc-frv.c (frv_md_fdpic_enabled): New.
2629 (DEFAULT_FDPIC): New.
2630 (frv_flags): Use DEFAULT_FDPIC.
2631 (frv_pic_flag): Likewise.
2632 (OPTION_NOPIC): New.
2633 (md_longopts): Add -mnopic.
2634 (md_parse_option): Handle it.
2635 (md_show_usage): Add -mfdpic and -mnopic.
2639 * config/tc-m68k.c: Add find_cf_chip to print list of valid
2640 chips for invalid coldfire instructions, rename selectors
2641 for ColdFire sub-variants, add 521x,5249,547x,548x and aliases,
2642 add current_chip to track which chip is referred to(including save/restore),
2643 use current_chip to select control registers, not current_arch.
2644 (md_show_usage): Add new chips.
2645 * doc/c-m68k.texi: Document new command line switches.
2649 * tc-s390.h (md_do_align, HANDLE_ALIGN): Remove.
2650 (NOP_OPCODE): Define.
2651 (s390_align_code): Remove prototype.
2652 * tc-s390.c (s390_align_code): Remove.
2656 * config/tc-ia64.c (make_unw_section_name): Removed.
2657 (start_unwind_section): New function.
2658 (generate_unwind_image): Take const segT instead of const
2660 (dot_handlerdata): Adjusted.
2661 (dot_endp): Likewise.
2665 * config/obj-elf.c (obj_elf_change_section): Allow the
2666 ".note.GNU-stack" section has SHF_EXECINSTR.
2670 * config/obj-elf.c (get_section): Return bfd_boolean.
2671 (obj_elf_change_section): Call bfd_get_section_by_name_if
2672 instead of bfd_map_over_sections.
2676 * config/obj-elf.c (get_section): New function.
2677 (obj_elf_change_section): Support multiple sections with same
2682 * config/tc-arm.c (create_register_alias): Fix typo checking for
2683 case sensitive register aliases.
2684 (co_proc_number): Use error message string in all_reg_maps[]
2686 (cp_reg_required_here): Likewise.
2687 (fp_reg_required_here): Likewise.
2691 * dwarf2dbg.c (dwarf2_finish): Add SEC_DEBUGGING to section flags.
2695 * config/tc-mips.c (HAVE_32BIT_ADDRESSES, append_insn, macro_build)
2696 (load_address, macro, mips_ip, md_parse_option)
2697 (mips_force_relocation, mips_validate_fix, md_apply_fix3)
2698 (s_change_sec, pic_need_relax, tc_gen_reloc): Remove all
2699 embedded-PIC handling, and update comments.
2700 (SWITCH_TABLE): Remove.
2701 * config/tc-mips.h (DIFF_EXPR_OK): Delete.
2702 (enum mips_pic_level): Remove EMBEDDED_PIC.
2703 (EXTERN_FORCE_RELOC): Remove embedded-PIC handling.
2704 (TC_FORCE_RELOCATION): Update comment.
2705 * ecoff.c (ecoff_build_lineno): Add comment about some code that
2706 might be safe to remove now that MIPS embedded-PIC is gone.
2710 * config/obj-som.c (obj_som_init_stab_section): Add new arguments in
2711 call to obj_set_subsection_attributes.
2712 (obj_som_init_stab_section): Likewise.
2713 * config/tc-hppa.c (default_subspace_dict): Add comdat field.
2714 (pa_def_subspaces): Provide comdat default.
2715 (pa_subspace): Handle new "comdat" parameter. Set SEC_LINK_ONCE and
2716 not SEC_IS_COMMON if section is comdat, common or dup_common. Update
2717 calls to create_new_subspace and update_subspace to pass comdat flag.
2718 (create_new_subspace, update_subspace): Add new comdat argument. Use
2719 it in calls to obj_set_subsection_attributes.
2720 * doc/c-hppa.texi (.subspa, .nsubspa): Document new comdat parameter
2721 and use of comdat, common and dup_comm parameters.
2725 * config/obj-elf.c (obj_elf_change_section): Check if the old
2726 group name is NULL before comparison.
2730 * config/tc-mips.h (mips_dwarf2_addr_size): Prototype.
2734 * config/tc-mips.c (s_mipsset): Set default CPU type for .set mipsN.
2738 * config/tc-mips.c (md_longopts): Remove -membedded-pic option.
2739 (OPTION_MEMBEDDED_PIC): Remove.
2740 (OPTION_TRAP, OPTION_BREAK, OPTION_EB, OPTION_EL)
2741 (OPTION_FP32, OPTION_GP32, OPTION_CONSTRUCT_FLOATS)
2742 (OPTION_NO_CONSTRUCT_FLOATS, OPTIONS_FP64, OPTION_GP64)
2743 (OPTION_RELAX_BRANCH, OPTION_NO_RELAX_BRANCH)
2744 (OPTION_ELF_BASE): Renumber.
2745 (md_parse_option): Remove OPTION_MEMBEDDED_PIC handling.
2746 (md_show_usage): Remove mention of -membedded-pic.
2747 * doc/as.texinfo: Remove mention of -membedded-pic.
2751 * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Remove.
2752 * config/tc-mips.c (RDATA_SECTION_NAME, mips_target_format): Remove
2754 (md_begin, mips_ip, md_parse_option, s_change_sec, s_option,
2755 s_abicalls, nopic_need_relax, tc_gen_reloc): Remove uses of
2756 USE_GLOBAL_POINTER_OPT.
2760 * config/tc-mips.c (macro): One more use of load_delay_nop.
2764 * config/tc-mips.c (load_delay_nop): New function.
2765 (load_address, macro): Use load_delay_nop() to build a nop
2766 which can be omitted with gpr_interlocks.
2770 * config/tc-mips.c (hilo_interlocks, gpr_interlocks,
2771 cop_interlocks): Remove superfluous CPU entries.
2775 * config/tc-arm.c (mav_parse_offset): Value must be multiple of 4.
2779 * NEWS: Added support for EMAC instructions and MAC/EMAC
2781 * config/m68k-parse.h: Add ACC[123], ACCEXT{01,23}, MAC/EMAC
2782 scale factor tokens, trailing_ampersand to mark mask addressing
2783 for MAC/EMAC instructions.
2784 * config/m68k-parse.y: Add options_ampersand clause, '<<',
2786 (yylex): Handle '>', '<', and '&' following '+'.
2787 * config/tc-m68k.c: Set mcfmac/mcfemac on appropriate ColdFire
2788 architectures in archs[].
2789 (m68k-ip): Add '4', 'e', 'g', 'i', cases to handle mask addressing
2790 for MAC/EMAC instructions, ACC[0123], ACCEXT{01,23}, and '<<'/'>>'
2792 (m68k_ip): Handle trailing '&' on MAC/EMAC insns.
2793 (install_operand): Fix 'n' case, Add 'F', 'f', 'G', 'H', 'I', ']'
2795 Add EMAC operands to init_table[].
2799 * config/tc-arm.h (enum mstate): Move here, add MAP_UNDEFINED
2801 (TC_SEGMENT_INFO_TYPE): Define to enum mstate.
2802 * config/tc-arm.c (enum mstate): Delete from here.
2803 (mapping_state): Remove the static mapstate variable and instead
2804 store the state in the segment. This allows a per-section mapping
2805 state. Handle and ignore MAP_UNDEFINED states.
2806 (arm_elf_change_section): Get the current mapping state from the
2808 (s_ltorg): Set the mapping state to MAP_DATA.
2809 (arm_cleanup): Use arm_elf_change_section to get the mapping state
2810 for each pool as it is emitted.
2814 * config/tc-arm.h: Formatting tidy ups.
2818 * NEWS: Note that MIPS -membedded-pic option is deprecated.
2822 * config/tc-i386.h [TE_PE] (TC_CONS_FIX_NEW): Define.
2823 * config/tc-i386.c (md_pseudo_table) [TE_PE]: Add "secrel32".
2824 [TE_PE] (O_secrel): Define.
2825 [TE_PE] (x86_pe_cons_fix_new): New.
2826 [TE_PE] (pe_directive_secrel): Likewise.
2827 (tc_gen_reloc) [TE_PE]: Support BFD_RELOC_32_SECREL.
2831 * config/tc-mips.c (mips_dwarf2_addr_size): Revert part
2832 of previous patch for fix in gcc.
2836 * config/tc-xtensa.c (xg_assembler_literal): Fix a typo.
2840 * read.c (do_align): Call md_flush_pending_output, if defined.
2844 * expr.c (operand): Correct checks for ++ and --.
2848 * config/tc-generic.c: Add some comments.
2852 * doc/c-mips.texi (-m{no-,}fix-vr4120): Renamed from
2853 -{no-}mfix-vr4122-bugs.
2854 * config/tc-mips.c (mips_fix_vr4120): Renamed from mips_fix_4122_bugs.
2855 (append_insn, mips_emit_delays): Update accordingly.
2856 (OPTION_FIX_VR4120, OPTION_NO_FIX_VR4120): Renamed from *VR4122.
2857 (md_longopts): Change -{no-,}mfix-vr4122-bugs to -m{no-,}fix-vr4120.
2858 (md_parse_option): Update after above changes.
2859 (md_show_usage): Add -mfix-vr4120.
2863 * doc/as.texinfo (Sub-Sections): Conditionalize COFF-specific use
2864 of .section directive; add a reference to the ELF .subsection
2869 * config/tc-m32r.c (md_assemble): Fixed infinite loop bug
2874 * Makefile.am: Remove mips from aout targets.
2875 * Makefile.in: Regenerate.
2876 * configure.in: Remove mips-dec-bsd* target.
2877 * configure: Regenerate.
2882 * config/tc-ppc.c (ppc_elf_suffix): Add valid32 and valid64 fields
2883 to struct map_bfd. Adjust MAP macro, and define MAP32, MAP64.
2884 Update "mapping". Restrict some @ modifiers to 32 bit.
2889 * config/tc-sh.c (dont_adjust_reloc_32): New variable.
2890 (sh_fix_adjustable): Avoid adjusting BFD_RELOC_32 when
2891 dont_adjust_reloc_32 is set.
2892 (md_longopts): Add option -renesas.
2893 (md_parse_option, md_show_usage): Likewise.
2894 * doc/c-sh.texi: Likewise.
2898 * config/tc-dlx.c (md_assemble): set fx_no_overflow flag for
2899 hi16 and lo16 fixS structs.
2900 (md_assemble): generate bit_fixS for RELOC_DLX_LO16 in
2901 exactly the same way as for RELOC_DLX_REL16.
2902 (machine_ip): properly respect LO flag in the_insn and
2903 output RELOC_DLX_LO16 rather than RELOC_DLX_16.
2904 (md_apply_fix3): apply RELOC_DLX_LO16.
2908 Remove long-obsolete MPW support.
2909 * mpw-config.in, mpw-make.sed, mac-as.r: Remove files.
2910 * configure.in: Remove mention of ppc-*-mpw* config.
2911 * configure.in: Likewise.
2915 * config/tc-arm.c (meabi_flags): Make its use conditional upon
2916 OBJ_ELF being defined.
2920 * config/obj-aout.c (obj_aout_type): Remove #ifdef BFD_ASSEMBLER code.
2924 * config/tc-arm.c (meabi_flags): New variable.
2925 (arm_parse_eabi): New function.
2926 (md_begin): Set flags for EABI v3.
2928 (arm_long_opts): Add meabi.
2929 * doc/as.texinfo <ARM>: Document -meabi.
2930 * doc/c-arm.texi: Ditto.
2934 * config/tc-xtensa.c (xtensa_post_relax_hook): Create literal
2935 tables even when use_literal_section flag is not set.
2939 * config/tc-sh.c: Remove trailing whitespace.
2943 * doc/c-cris.texi (CRIS-Opts): Document --no-mul-bug-abort,
2944 --mul-bug-abort and the default behavior.
2945 * config/tc-cris.c (cris_insn_kind): New member CRIS_INSN_MUL.
2946 (err_for_dangerous_mul_placement): New variable.
2947 (STATE_MUL, OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): New
2949 (md_cris_relax_table): Have placeholder for STATE_MUL.
2950 (md_longopts): New options --mul-bug-abort and --no-mul-bug-abort.
2951 (cris_relax_frag) <case ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: New
2953 (md_estimate_size_before_relax) <case ENCODE_RELAX (STATE_MUL,
2954 STATE_BYTE)>: Ditto.
2955 (md_convert_frag) <ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: Check
2956 alignment and position of this frag, emit error message if
2958 (md_assemble): For a multiply insn and when checking it,
2959 transform the current frag into a special frag for that purpose.
2960 (md_parse_option) <case OPTION_MULBUG_ABORT_OFF, case
2961 OPTION_MULBUG_ABORT_ON>: Handle new options.
2965 * config/tc-xtensa.c (mark_literal_frags): New function.
2966 (xtensa_move_literals): Call mark_literal_frags for all literal
2967 segments, including init and fini literal segments.
2968 (xtensa_post_relax_hook): Swap use of xt_insn_sec and xt_literal_sec.
2972 * tc-hppa.c (cons_fix_new_hppa): Check for PC relative base type.
2973 (pa_comm): Set BSF_OBJECT in symbol flags.
2977 * Makefile.am: Run "make dep-am".
2978 * Makefile.in: Regenerate.
2979 * doc/Makefile.in: Regenerate.
2980 * config.in: Regenerate.
2981 * po/gas.pot: Regenerate.
2985 * read.c (read_a_source_file): Use demand_empty_rest_of_line.
2986 (demand_empty_rest_of_line): Issue an error here.
2987 (ignore_rest_of_line): Silently skip to end.
2988 (demand_copy_string): Issue an error, not warning.
2990 * config/obj-elf.c (obj_elf_section_name): Likewise.
2991 (obj_elf_section): Likewise.
2992 * config/tc-arc.c (arc_extoper): Remove bogus NULL checks.
2993 (arc_extinst): Likewise.
2994 * config/tc-ia64.c (dot_saveb): Use demand_empty_rest_of_line.
2995 (dot_spill): Likewise.
2996 (dot_unwabi): Likewise.
2997 (dot_prologue): Likewise.
2999 * expr.c (operand): Reject ++ and --.
3000 (operator): Likewise.
3004 * config/tc-sh.c: Include dw2gencfi.h.
3005 (sh_cfi_frame_initial_instructions): New function.
3006 (sh_regname_to_dw2regnum): Likewise.
3007 * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Move to the end of
3009 (TARGET_USE_CFIPOP): Define.
3010 (tc_cfi_frame_initial_instructions): Likewise.
3011 (tc_regname_to_dw2regnum): Likewise.
3012 (DWARF2_DEFAULT_RETURN_COLUMN, DWARF2_CIE_DATA_ALIGNMENT): Likewise.
3013 * Makefile.am: Update dependencies.
3014 * Makefile.in: Regenerate.
3018 * configure.in: Switch sh-*-rtems* to ELF. Add sh-*-rtemscoff*.
3019 * configure: Regenerate.
3023 * read.c (s_leb128): Call md_flush_pending_output.
3027 * config/tc-i386.c (output_insn): Handle PadLock instructions.
3028 * config/tc-i386.h (CpuPadLock): New define.
3029 (CpuUnknownFlags): Added CpuPadLock.
3033 * doc/c-hppa.texi (HPPA Directives): Fix typo.
3037 * dw2gencfi.c (output_cie): Align length to 4 byte boundary.
3038 (cfi_finish): Likewise for fde.
3042 * config/tc-ia64.c (md_assemble): Properly handle NULL
3044 (ia64_handle_align): Don't abort if failed to add a stop bit.
3048 * Makefile.in: Regenerated.
3049 * aclocal.m4: Likewise.
3050 * configure: Likewise.
3051 * doc/Makefile.in: Likewise.
3055 * config/tc-ia64.c (dot_align): New.
3056 (ia64_do_align): Make it static.
3057 (md_pseudo_table): Use "dot_align" for "align".
3058 (ia64_md_do_align): Don't set align_frag here.
3059 (ia64_handle_align): Add a stop bit to the previous bundle if
3062 * config/tc-ia64.h (ia64_do_align): Removed.
3066 * config/tc-sh.c (md_parse_option): Add -isa=sh4-nofpu and
3067 -isa=sh4-nommu-nofpu options. Adjust help messages accordingly.
3068 (sh_elf_final_processing): Output BFD type sh4_nofpu if that is
3069 the most general type or the user specifically requested it.
3070 (md_assemble): Add a new error message for when an instruction
3071 is understood, but is not allowed due to an -isa option.
3075 * config/tc-ia64.c (align_frag): New.
3076 (md_assemble): Set the tc_frag_data field in align_frag for
3077 IA64_OPCODE_FIRST instructions.
3078 (ia64_md_do_align): Set align_frag.
3079 (ia64_handle_align): Add a stop bit if needed.
3081 * config/tc-ia64.h (TC_FRAG_TYPE): New.
3082 (TC_FRAG_INIT): New.
3086 * config/tc-frv.c (fr400_audio): New variable.
3087 (md_parse_option, md_show_usage): Add -mcpu=fr405 and -mcpu=fr450.
3088 (md_parse_option): Set fr400_audio for -mcpu=fr400 and -mcpu=fr405.
3089 (target_implements_insn_p): New function.
3090 (md_assemble): Report an error if the processor doesn't implement
3095 * config/tc-m32r.c (md_longopts): Added -no-bitinst option.
3096 (md_parse_option): Ditto.
3097 (OPTION_NO_SPECIAL_M32R): Added.
3098 (md_show_usage): Document it.
3099 (enable_speial_m32r): Changed a default value from 0 to 1.
3100 * doc/c-m32r.texi: Document the -no-bitinst option.
3104 * config/tc-sh.c (get_operand): Revert previous delta.
3105 (tc_gen_reloc): Check for an unknown reloc type before processing
3110 * config/tc-s390.c (s390_insn): Correct range check for opcode in
3111 .insn pseudo operation.
3115 * config/tc-sh.c (get_operand): In case of #Imm, check has been
3116 added for wrong syntax.
3120 * config/tc-mips.c (mips_dwarf2_addr_size): New.
3121 * config/tc-mips.h (DWARF2_ADDR_SIZE): Use.
3125 * config/tc-sh.c (build_Mytes): Add REG_N_D and REG_N_B01
3126 nibble types to assembler.
3130 * config/tc-iq2000.c: Add missing \n\ in multiline string literal.
3134 * config/tc-ia64.c (slot_index): New arg before_relax. Use instead of
3136 (fixup_unw_records): New arg before_relax. Pass to slot_index.
3137 (ia64_estimate_size_before_relax): New.
3138 (ia64_convert_frag): Pass 0 to fixup_unw_records. Add comment.
3139 (generate_unwind_image): Pass 1 to fixup_unw_records.
3140 * config/tc-ia64.h (ia64_estimate_size_before_relax): Declare.
3141 (md_estimate_size_before_relax): Call ia64_estimate_size_before_relax.
3145 * stabs.c (generate_asm_file): Avoid warning about use of
3146 uninitialized variable.
3150 * config/tc-ia64.c (ia64_flush_insns): In addition to prologue,
3151 body, and endp, allow unwind records which do not have a "t"
3152 (time/instruction) field.
3156 * config/tc-arm.c (do_mav_dspsc_1): Correct offset of CRn.
3157 (do_mav_dspsc_2): Likewise.
3158 Fix accumulator registers move opcodes.
3163 * dwarf2dbg.c (get_filenum): Do not read beyond allocated memory.
3167 * config/tc-ia64.h (ia64_frob_symbol): New declaration.
3168 (tc_frob_symbol): New macro definition.
3169 * config/tc-ia64.c (ia64_frob_symbol): New routine.
3173 * config/tc-arm.c (md_begin): Mark .note.gnu.arm.ident as
3178 * read.h (IGNORE_OPCODE_CASE): Do not define. Replace with ...
3179 (TC_CASE_SENSITIVE): ... this.
3180 * read.c: Replace IGNORE_OPCODE_CASE with TC_CASE_SENSITIVE.
3181 * doc/internals.texi (TC_CASE_SENSITIVE): Document.
3185 * config/tc-ia64.c (dot_endp): Delete call to output_endp.
3186 (generate_unwind_image): Re-add it here.
3190 * dwarf2dbg.c (DWARF2_ADDR_SIZE): Remove trailing ';'
3191 * read.h (SKIP_WHITESPACE): Turn into an expression.
3192 * read.c (read_a_source_file): A pseudo is removed by having a
3197 * config/tc-ia64.c (output_endp): New.
3198 (count_bits): Delete.
3199 (ia64_flush_insns, process_one_record, optimize_unw_records): Handle
3200 endp unwind records.
3201 (fixup_unw_records): Handle endp unwind records. Delete code for
3202 shortening prologue regions not followed by a body record.
3203 (dot_endp): Call add_unwind_entry to emit endp unwind record.
3204 * config/tc-ia64.h (unw_record_type): Add endp.
3208 * config/tc-ia64.c (ia64_convert_frag): Call md_number_to_chars to
3209 fill padding bytes with zeroes.
3210 (emit_one_bundle): New locals last_ptr, end_ptr. Rewrite code that
3211 sets unwind_record slot_number and slot_frag fields.
3215 * config/tc-mips.c (add_got_offset_hilo): New function.
3216 (macro): Use load_register() and add_got_offset_hilo() to load
3217 constants instead of hardcoding code sequences throughout.
3221 * config/tc-ia64.c (emit_one_bundle): Add proper indentation.
3225 * config/tc-m68k.h (EXTERN_FORCE_RELOC): Handle m68k-uclinux specially,
3227 * config/tc-m68k.c (RELAXABLE_SYMBOL): Use EXTERN_FORCE_RELOC instead
3228 of hard-coded test for TARGET_OS=elf.
3232 * config/tc-mips.c (hilo_interlocks): Change definition
3233 so that MIPS32, MIPS64 and later ISAs are included, along with
3234 the already-included machines. Update comments.
3238 * config/tc-arm.c (tc_gen_reloc): Improve error message for
3239 undefined local labels.
3243 * config/tc-mips.c (load_address, macro): Update comments about
3244 NewABI GP relaxation.
3248 * config/tc-mips.c (macro_build): Remove place and counter arguments.
3249 (mips_build_lui, macro_build_ldst_constoffset): Likewise.
3250 (mips16_macro_build, macro_build_jalr): Remove counter argument.
3251 (set_at, load_register, load_address, move_register): Likewise.
3252 (load_got_offset, add_got_offset): Likewise.
3253 Update all calls and tidy accordingly.
3257 * config/tc-mips.c (RELAX_ENCODE): Remove WARN argument.
3258 (RELAX_FIRST, RELAX_SECOND): Turn into 8-bit quantities.
3259 (RELAX_USE_SECOND): Bump to 0x10000.
3260 (RELAX_SECOND_LONGER, RELAX_NOMACRO, RELAX_DELAY_SLOT): New flags.
3261 (mips_macro_warning): New variable.
3262 (md_assemble): Wrap macro expansion in macro_start() and macro_end().
3263 (s_cpload, s_cpsetup, s_cprestore, s_cpreturn): Likewise.
3264 (relax_close_frag): Set mips_macro_warning.first_frag. Adjust use
3266 (append_insn): Update mips_macro_warning.sizes.
3267 (macro_start, macro_warning, macro_end): New functions.
3268 (macro_build): Don't emit warnings here.
3269 (macro_build_lui, md_estimate_size_before_relax): ...or here.
3270 (md_convert_frag): Check for cases where one macro alternative
3271 needs a warning and the other doesn't. Emit a warning if the
3272 longer sequence was chosen.
3276 * config/tc-mips.h (tc_frag_data_type, TC_FRAG_TYPE): Remove.
3277 * config/tc-mips.c (RELAX_ENCODE): Take three arguments: the size of
3278 the first sequence, the size of the second sequence, and a flag
3279 that says whether we should warn.
3280 (RELAX_OLD, RELAX_NEW, RELAX_RELOC[123]): Delete.
3281 (RELAX_FIRST, RELAX_SECOND): New.
3282 (mips_relax): New variable.
3283 (relax_close_frag, relax_start, relax_switch, relax_end): New fns.
3284 (append_insn): Remove "place" argument. Use mips_relax.sequence
3285 rather than "place" to check whether we're expanding the second
3286 alternative of a relaxable macro. Remove redundant check for
3287 branch relaxation. If generating a normal insn, and there
3288 is not enough room in the current frag, call relax_close_frag()
3289 to close it. Update mips_relax.sizes[]. Emit fixups for the
3290 second version of a relaxable macro. Record the first relaxable
3291 fixup in mips_relax. Remove tc_gen_reloc workaround.
3292 (macro_build): Remove all uses of "place". Use mips_relax.sequence
3293 in the same way as in append_insn.
3294 (mips16_macro_build): Remove "place" argument.
3295 (macro_build_lui): As for macro_build. Don't drop the add_symbol
3296 when generating the second version of a relaxable macro.
3297 (load_got_offset, add_got_offset): New functions.
3298 (load_address, macro): Use new relaxation machinery. Remove
3299 tc_gen_reloc workarounds.
3300 (md_estimate_size_before_relax): Set RELAX_USE_SECOND if the second
3301 version of a relaxable macro is needed. Return -RELAX_SECOND if the
3302 first version is needed.
3303 (tc_gen_reloc): Remove relaxation handling.
3304 (md_convert_frag): Go through the fixups for a relaxable macro and
3305 mark those that belong to the unneeded alternative as done. If the
3306 second alternative is needed, adjust the fixup addresses to account
3307 for the deleted first alternative.
3311 * frags.h (frag_room): Declare.
3312 * frags.c (frag_room): New function.
3313 * doc/internals.texi: Document it.
3317 * config/tc-mips.c (append_insn): Don't do r3900 interlock
3318 optimization for -mtune=r3900, as this will break on other CPUs.
3322 * config/tc-m68hc11.c (build_indexed_byte): movb and movw cannot
3323 be relaxed, use fixup.
3324 (md_apply_fix3): Use 5 bit reloc from movb and movw fixup.
3328 * config/tc-sparc.c (sparc_ip): Disallow %f32-%f63 for single
3333 * config/tc-mips.c (append_insn): Properly detect variant frags
3334 that preclude swapping of relaxed branches. Correctly swap
3335 instructions between frags when dealing with relaxed branches.
3339 * acinclude.m4: Quote names of macros to be defined by AC_DEFUN
3341 * aclocal.m4: Regenerate.
3342 * configure: Regenerate.
3346 * config/tc-h8300.c (build_bytes): Apply relaxation to bit
3351 * config/tc-mips.c (macro_build_jalr): When adding an R_MIPS_JALR
3352 reloc, reserve space for the delay slot as well as the jalr itself.
3356 * config/tc-arm.c (do_vfp_reg2_from_sp2): Rename from do_vfp_sp_reg2.
3357 (do_vfp_sp2_from_reg2): New function.
3359 (do_vfp_dp_from_reg2): Check return values properly.
3363 * config/tc-mips.c (warn_nops): Remove static variable.
3364 (macro): Remove test of warn_nops.
3365 (md_shortops): Remove 'n'.
3366 (md_parse_option): Remove 'n' case.
3367 (md_show_usage): Remove -n.
3368 * doc/as.texinfo (Overview): Remove MIPS -n option.
3369 * doc/c-mips.texi (MIPS Opts): Remove mention -n.
3370 * NEWS: Mention removal of MIPS -n option.
3372 * config/tc-mips.c (ISA_HAS_COPROC_DELAYS): Remove.
3373 (cop_interlocks): Check ISA level.
3374 (cop_mem_interlocks): Define.
3375 (reg_needs_delay): Check cop_interlocks rather than
3376 ISA_HAS_COPROC_DELAYS.
3377 (append_insn): Likewise. Use cop_mem_interlocks rather than
3378 directly checking mips_opts.isa.
3379 (mips_emit_delays): Likewise.
3383 * config/tc-ia64.c (unwind): Move next_slot_number and
3384 next_slot_frag to ...
3385 (unw_rec_list): Here.
3386 (free_list_records): Removed.
3387 (output_unw_records): Likewise.
3388 (generate_unwind_image): Make it void.
3389 (alloc_record): Initialize next_slot_number and next_slot_frag.
3390 (slot_index): Take .org, .space and .align into account.
3391 (fixup_unw_records): Don't set slot_number to 0. Use
3392 list->next_slot_number and list->next_slot_frag instead of
3393 unwind.next_slot_number and unwind.next_slot_frag.
3394 (ia64_convert_frag): New.
3395 (generate_unwind_image): Generate a rs_machine_dependent frag
3397 (emit_one_bundle): Use list->next_slot_number and
3398 list->next_slot_frag instead of unwind.next_slot_number and
3399 unwind.next_slot_frag.
3401 * config/tc-ia64.h (md_convert_frag): Defined as
3403 (md_estimate_size_before_relax): Defined as (f)->fr_var.
3408 * config/tc-frv.h (md_apply_fix3): Don't define.
3409 * config/tc-frv.c (md_apply_fix3): New. Shift/truncate %hi/%lo
3411 * config/tc-frv.h (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
3413 * config/tc-frv.c (line_separator_chars): Add `!'.
3415 * config/tc-frv.c (md_assemble): Clear insn upfront.
3417 * config/tc-frv.c (OPTION_FDPIC): New macro.
3418 (md_longopts): Add mfdpic.
3419 (md_parse_option): Handle it.
3421 * config/tc-frv.c (md_cgen_lookup_reloc) <FRV_OPERAND_D12,
3422 FRV_OPERAND_S12>: Use reloc type encoded in fix-up.
3423 (frv_pic_ptr): Parse funcdesc.
3427 * doc/as.texinfo: Let texi2pod parse asconfig.texi and
3428 gasver.texi. Remove duplicate symbol definitions for texi2pod.
3432 * Makefile.am (Makefile): Move the dependency on
3433 $(BFDDIR)/configure.in to...
3434 (CONFIG_STATUS_DEPENDENCIES): ... here.
3435 (AUTOMAKE_OPTIONS): Require automake 1.8.
3436 * Makefile.in: Regenerate.
3437 * doc/Makefile.am (BASEDIR, BFDDIR): Define.
3438 (CONFIG_STATUS_DEPENDENCIES): Add a dependency on
3439 $(BFDDIR)/configure.in here as well.
3440 * doc/Makefile.in: Regenerate.
3444 * Makefile.am (install, install-info, RECURSIVE_TARGETS): Remove.
3445 * Makefile.in: Regenerate.
3446 * aclocal.m4: Regenerate.
3447 * doc/Makefile.am (install, install-info): Remove.
3448 (install-data-local): A new hook for install-info.
3449 (AUTOMAKE_OPTIONS): Require automake 1.8.
3450 * doc/Makefile.in: Regenerate.
3454 * doc/c-sh.texi: Update description about floating point behavior
3459 * configure.in: Add m68k-uClinux target.
3460 * configure: Regenerate.
3462 For older changes see ChangeLog-0203
3468 version-control: never