1 2004-11-10 Alan Modra <amodra@bigpond.net.au>
3 * obj.h (struct format_ops <app_file>): Add int param.
4 * read.h (s_app_file_string): Likewise.
5 * read.c (s_app_file_string): Likewise.
6 (s_app_file): Adjust s_app_file_string call.
7 * config/tc-mips.c (s_mips_file): Likewise.
8 * config/obj-coff.h (obj_app_file): Add app param.
9 * config/obj-ecoff.h (obj_app_file): Likewise.
10 * config/obj-multi.h (obj_app_file): Likewise.
11 * config/obj-elf.h (elf_file_symbol): Likewise.
12 * config/obj-elf.c (elf_file_symbol): Only emit one file symbol
13 if called for # preprocessor lines.
15 2004-11-08 H.J. Lu <hongjiu.lu@intel.com>
18 * symbols.c (resolve_symbol_value): Convert weak symbols only
20 (symbol_equated_reloc_p): Don't equate weaks when relocating
21 only for Windows PECOFF.
23 2004-11-08 Tomer Levi <Tomer.Levi@nsc.com>
25 * config/tc-crx.c (print_insn): Check and set insn_addr.
26 * config/tc-crx.h (md_frag_check): Define.
28 2004-11-08 Inderpreet Singh <inderpreetb@nioda.hcltech.com>
29 Vineet Sharma <vineets@noida.hcltech.com>
31 * configure.in: Add support for new target maxq-coff.
32 * configure: Regenerate.
33 * NEWS: Mention new support.
34 * config/tc-maxq.c: New file.
35 * config/tc-maxq.h: New file.
36 * config/obj-coff.h: Add support for maxq-coff.
38 2004-11-08 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
40 * symbols.c (any_external_name): Define.
41 (resolve_symbol_value): Convert weak symbols.
42 (S_SET_EXTERNAL): Support any_external_name.
43 (S_SET_NAME): Qualify parameter const.
44 (symbol_equated_reloc_p): Equate to weaks when relocating.
45 * symbols.h (S_SET_NAME): Qualfiy parameter const.
46 * tc.h (any_external_name): Declare.
47 * config/obj-coff.c ("coff/pe.h"): Include for BFD
49 (weak_is_altname): Declare and define.
50 (weak_name2altname): Same.
51 (weak_altname2name): Same.
52 (weak_uniquify): Same.
53 (weak_altprefix): Define.
54 (obj_coff_weak): Change .weak syntax and handling.
55 (coff_frob_symbol): Fix PE weak symbol alternates.
56 * config/obj-coff.h (USE_UNIQUE): Define.
57 * config/tc-i386.c (md_apply_fix3): Assume weak symbols
58 are in another segment.
59 (tc_gen_reloc): Remove broken addend hack.
60 doc/as.texinfo: Update.
62 2004-11-05 Sterling Augustine <sterling@tensilica.com>
64 * config/tc-xtensa.c (total_frag_text_expansion): New.
65 (md_estimate_size_before_relax): Use it.
66 (find_address_of_next_align_frag): Likewise.
68 2004-11-05 Tomer Levi <Tomer.Levi@nsc.com>
70 * config/tc-crx.c: Rename argument types.
71 (processing_arg_number): Rename to 'cur_arg_num'.
72 (get_number_of_bits): Rename to 'set_operand_size'.
73 (get_operandtype): Rename to 'parse_operand', totally rewrite.
74 (set_cons_rparams): Rename to 'set_operand', totally rewrite.
75 (set_indexmode_parameters): Remove function, integrate its code into
77 (set_operand_size): Get rid of 'Operand Number' function parameter -
78 use global variable 'cur_arg_num' instead.
79 Use a local 'argument' pointer to reference the current argument.
80 (parse_operand): Likewise.
81 (set_operand): Likewise.
82 (process_label_constant): Likewise.
84 2004-11-04 Bob Wilson <bob.wilson@acm.org>
86 * config/tc-xtensa.c: Remove XTENSA_SECTION_RENAME ifdefs.
87 (add_section_rename): Delete. Inlined into...
88 (build_section_rename): ...here. Use xstrdup instead of strdup.
89 (xtensa_section_rename): Drop "const" from argument and return types.
90 (md_show_usage): Indent to match show_usage().
91 * config/tc-xtensa.h: Remove XTENSA_SECTION_RENAME ifdefs.
92 (tc_canonicalize_section_name): Define.
93 (md_elf_section_rename): Remove unused macro.
94 * doc/as.texinfo (Overview): Document Xtensa --rename-section option.
95 * doc/c-xtensa.texi (Xtensa Options): Likewise.
96 (Frame Directive): Delete.
98 2004-11-04 Daniel Jacobowitz <dan@debian.org>
100 * configure.in: Remove arm-*-oabi and thumb-*-oabi.
101 * config/tc-arm.c (target_oabi): Delete.
102 (md_apply_fix3, elf32_arm_target_format): Remove target_oabi checks.
103 (arm_opts): Remove moabi.
104 * doc/as.texinfo (Overview): Remove documentation of -moabi.
105 * doc/c-arm.texi (ARM Options): Likewise.
106 * configure: Regenerated.
108 2004-11-04 Hans-Peter Nilsson <hp@axis.com>
110 * configure.in (crisv32): Recognize. AC_DEFINE_UNQUOTED
111 DEFAULT_CRIS_ARCH. Handle crisv32-*-linux-gnu* like
112 cris-*-linux-gnu* and crisv32-*-* like cris-*-*.
113 * configure: Regenerate.
114 * config/tc-cris.c (enum cris_archs): New.
115 (cris_mach, cris_arch_from_string, s_cris_arch, get_sup_reg)
116 (cris_insn_ver_valid_for_arch): New functions.
117 (DEFAULT_CRIS_ARCH): New macro, default to cris_any_v0_v10.
118 (cris_arch): New variable.
119 (md_pseudo_table): New pseudo .arch.
120 (err_for_dangerous_mul_placement): Initialize according to
122 (STATE_COND_BRANCH): Renamed from STATE_CONDITIONAL_BRANCH.
124 (STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON)
125 (STATE_ABS_BRANCH_V32, STATE_LAPC, BRANCH_BF_V32, BRANCH_BB_V32)
126 (BRANCH_WF_V32, BRANCH_WB_V32): New.
127 (BRANCH_BF, BRANCH_BB, BRANCH_WF, BRANCH_WB): Don't undef after
128 use in md_cris_relax_table.
129 (md_cris_relax_table): Add entries for STATE_COND_BRANCH_V32,
130 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
131 Update and improve head comment.
132 (OPTION_PIC): Define in terms of previous option, OPTION_US.
133 (OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): Similar.
135 (md_longopts): New option --march=...
136 (cris_any_v0_v10_long_jump_size, crisv32_long_jump_size): New
138 (md_long_jump_size): Initialize in terms of DEFAULT_CRIS_ARCH.
139 (HANDLE_RELAXABLE): New macro.
140 (md_estimate_size_before_relax): Use HANDLE_RELAXABLE for common
141 cases. Check for weak symbols and assume not relaxable. Handle
142 STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON,
143 STATE_ABS_BRANCH_V32, STATE_LAPC. Use new variable symbolP, not
145 (md_convert_frag): Handle STATE_COND_BRANCH_V32,
146 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
147 (cris_create_short_jump): Adjust for CRISv32.
148 (cris_relax_frag): Handle new states.
149 (md_create_long_jump): Ditto. Emit error for common_v10_v32.
150 (md_begin): Define symbols "..asm.arch.cris.v32",
151 "..asm.arch.cris.v10", "..asm.arch.cris.common_v10_v32" and
152 "..asm.arch.cris.any_v0_v10". Use cris_insn_ver_valid_for_arch
153 when entering opcode table entry points.
154 (md_assemble): Adjust branch handling for CRISv32. Handle LAPC
155 relaxation. In fix_new_exp call for main insn, pass 1 for pcrel
156 parameter for 8, 16 and 32-bit pc-relative insns and LAPC.
157 (cris_process_instruction): Initialize out_insnp->insn_type to
158 CRIS_INSN_NONE, not CRIS_INSN_NORMAL.
159 <case ']', '[', 'A', 'd', 'Q', 'N', 'n', 'Y', 'U', 'u', 'T'>: New
161 <case 'm'>: Check that modified_char == '.'.
162 <invalid operands>: Consume the rest of the line.
163 When operands don't match, skip over subsequent insns with
164 non-matching version specifier but same mnemonic.
165 <immediate constant, case SIZE_SPEC_REG>: Immediate operands for
166 special registers in CRISv32 are always 32 bit long.
167 <immediate constant, case SIZE_FIELD_SIGNED, SIZE_FIELD_UNSIGNED>:
169 (get_gen_reg): Only recognize "PC" when followed by "+]" for v32
170 and compatible. Recognize "ACR" for v32, unless followed by "+".
171 (get_spec_reg): Consider cris_arch when looking up register.
172 (get_autoinc_prefix_or_indir_op): Don't recognize assignment for
174 (get_3op_or_dip_prefix_op): Check for ']' after seeing '[rN+'.
175 (cris_get_expression): Restore input_line_pointer if failing "early".
176 (get_flags): Consider cris_arch and recognize flags accordingly.
177 (branch_disp): Adjust for CRISv32.
178 (gen_cond_branch_32): Similar. Emit error for common_v10_v32.
179 (cris_number_to_imm): Use as_bad_where, not as_bad. Remove
180 related FIXME. Don't insist on BFD_RELOC_32_PCREL fixup to be
181 resolved. Don't enter zeros in object file for
183 <case BFD_RELOC_CRIS_LAPCQ_OFFSET, BFD_RELOC_CRIS_SIGNED_16>
184 <case BFD_RELOC_CRIS_SIGNED_8>: New case.
185 (md_parse_option): Break out "return 1".
186 <OPTION_ARCH> New case.
187 (tc_gen_reloc): <case BFD_RELOC_CRIS_LAPCQ_OFFSET>
188 <case BFD_RELOC_CRIS_SIGNED_16, BFD_RELOC_CRIS_SIGNED_8>
189 <case BFD_RELOC_CRIS_UNSIGNED_8, BFD_RELOC_CRIS_UNSIGNED_16>
190 <case BFD_RELOC_32_PCREL>: New cases.
191 Addends for non-zero fx_pcrel are too in fx_offset.
192 (md_show_usage): Show --march=<arch>.
193 (md_apply_fix3): Adjust val for BFD_RELOC_CRIS_LAPCQ_OFFSET.
194 (md_pcrel_from): BFD_RELOC_CRIS_LAPCQ_OFFSET is PC-relative too.
195 (s_syntax) <struct syntaxes>: Properly constify member operand.
196 * config/tc-cris.h (TARGET_MACH): Define.
197 (cris_mach): Declare.
198 * doc/as.texinfo (Overview) <CRIS>: Add --march=...
199 * doc/c-cris.texi (CRIS-Symbols): New node for built-in symbols.
200 (CRIS-Opts): Document --march=...
201 (CRIS-Pseudos): Document .arch.
203 2004-11-04 Jan Beulich <jbeulich@novell.com>
205 * config/tc-i386.c (set_intel_syntax): Allow % in symbol names when
206 intel syntax and no register prefix, allow $ in symbol names when
208 (set_16bit_gcc_code_flag): Replace literal 'l' by LONG_MNEM_SUFFIX.
209 (intel_float_operand): Add fourth return value indicating math control
210 operations. Make classification more precise.
211 (md_assemble): Complain if memory operand of mov[sz]x has no size
213 (parse_insn): Translate word operands to floating point instructions
214 operating on integers as well as control instructions to short ones
215 as expected by AT&T syntax. Translate 'd' suffix to short one only for
216 floating point instructions operating on non-integer operands.
217 (match_template): Remove fldcw special case. Adjust q-suffix handling
218 to permit it on fild/fistp/fisttp in AT&T mode.
219 (process_suffix): Don't guess DefaultSize insns' suffix from
220 stackop_size for certain floating point control instructions. Guess
221 suffix for branch and [ls][gi]dt based on flag_code. Split error
222 messages for Intel and AT&T syntax, and make the condition more strict
223 for the former. Adjust suppressing of generation of operand size
225 (intel parser): Allow the full set of MASM operators. Add FWORD, TBYTE,
226 OWORD, and XMMWORD operand size specifiers (TBYTE replaces XWORD). Add
228 * config/tc-i386.h (BYTE_PTR WORD_PTR DWORD_PTR QWORD_PTR XWORD_PTR
229 SHORT OFFSET_FLAT FLAT NONE_FOUND): Remove unused defines.
231 2004-11-03 Hans-Peter Nilsson <hp@axis.com>
233 * symbols.c (colon) [!WORKING_DOT_WORD]: Don't declare
234 md_short_jump_size, md_long_jump_size.
235 * write.c [!WORKING_DOT_WORD]: Ditto.
236 * tc.h [!WORKING_DOT_WORD]: Declare them here. Drop const
238 * config/tc-cris.h (md_short_jump_size, md_long_jump_size): Don't
240 * config/tc-cris.c (md_short_jump_size, md_long_jump_size): Drop
241 const qualifier in these definitions.
242 * config/tc-i370.c, config/tc-m68k.c, config/tc-pdp11.c,
243 config/tc-s390.c, config/tc-tahoe.c, config/tc-vax.c: Ditto.
245 2004-11-02 Nick Clifton <nickc@redhat.com>
247 * dwarf2dbg.c (dwarf2_finish): Check for the existence of a file
248 table before deciding to produce a .debug_line section to match up
249 with a user provided .debug_info section.
251 2004-10-28 Tomer Levi <Tomer.Levi@nsc.com>
253 * config/tc-crx.c (getreg_image): Bug fix, a return value was
254 mistakenly omitted from CRX_C_REGTYPE and CRX_CS_REGTYPE cases.
256 2004-10-27 Tomer Levi <Tomer.Levi@nsc.com>
258 * config/tc-crx.c: Remove global variable 'post_inc_mode'.
259 (get_flags): New function.
260 (get_number_of_bits): Edit comments, update numeric values to
262 (process_label_constant): Don't support the colon
263 format (SYMBOL:[s|m|l]).
264 (set_cons_rparams): Support argument type 'arg_rbase'.
265 (get_operandtype): Bug fix in 'rbase' operand type parsing.
266 (handle_LoadStor): Bug fix, first handle post-increment mode.
267 (getreg_image): Remove redundant code, update according to latest
269 (print_constant): Bug fix relate to 3-word instructions.
270 (assemble_insn): Bug fix, when matching instructions, verify also
271 instruction type (not only mnemonic).
272 Add various error checking.
273 (preprocess_reglist): Support HI/LO and user registers.
275 2004-10-25 David Mosberger-Tang <davidm@hpl.hp.com>
277 * config/tc-ia64.c (fixup_unw_records): Don't let the "t" value
278 in an epilogue directive go negative.
280 2004-10-25 H.J. Lu <hongjiu.lu@intel.com>
283 * config/tc-ia64.c (emit_one_bundle): Decrement md.num_slots_in_use
284 after reporting template error during manual bundling. Reported
285 by Michael Dupont, michaelx.dupont@intel.com.
287 2004-10-25 Daniel Jacobowitz <dan@debian.org>
289 * Makefile.am: Run dep-am.
290 * aclocal.m4: Regenerate with automake 1.9.2.
291 * Makefile.in: Regenerate with automake 1.9.2.
292 * doc/Makefile.in: Likewise.
294 * config/tc-arm.c: Include "dw2gencfi.h".
295 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
297 * config/tc-arm.h (TARGET_USE_CFIPOP, DWARF2_DEFAULT_RETURN_COLUMN)
298 (DWARF2_CIE_DATA_ALIGNMENT, tc_regname_to_dw2regnum)
299 (tc_cfi_frame_initial_instructions): Define.
300 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
303 2004-10-21 Tomer Levi <Tomer.Levi@nsc.com>
305 * config/tc-crx.c (assemble_insn): Check unsigned immediate
307 Update coprocessor id to be unsigned immediate.
309 2004-10-18 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
311 * config/tc-i386.c (O_secrel): Delete.
312 (tc_pe_dwarf2_emit_offset): New function.
313 * config/tc-i386.h (O_secrel): Define as O_md1.
314 (TC_DWARF2_EMIT_OFFSET): Define.
316 2004-10-18 Nick Clifton <nickc@redhat.com>
318 * config/tc-xstormy16.c (xstormy16_cons_fix_new): Accept and
319 ignore @fptr() directives for 4-byte fixups.
321 2004-10-15 Alan Modra <amodra@bigpond.net.au>
323 * Makefile.am: Run "make dep-am"
324 * Makefile.in: Regenerate.
326 2004-10-14 Bob Wilson <bob.wilson@acm.org>
328 * doc/c-xtensa.texi (Xtensa Options, Absolute Literals Directive):
329 Remove comments about placement of literal pools.
330 (Literal Directive): Update description of literal placement.
331 (Literal Prefix Directive): Remove statement that this does not apply
332 to absolute-mode literals. Describe new section naming scheme.
334 2004-10-12 Bob Wilson <bob.wilson@acm.org>
336 * config/tc-xtensa.c (xtensa_frequency_pseudo): Use set_subseg_freq.
337 (is_entry_opcode, is_movi_opcode, is_the_loop_opcode, is_jx_opcode,
338 is_windowed_return_opcode): Delete.
339 (xtensa_frob_label): Use get_subseg_target_freq.
340 (md_assemble): Inline call to is_entry_opcode.
341 (xtensa_handle_align): Inline call to get_frag_is_literal.
342 (relaxation_requirements): Inline call to is_jx_opcode.
343 (emit_single_op): Inline call to is_movi_opcode.
344 (xg_assemble_vliw_tokens): Inline calls to get_frag_is_insn,
345 get_frag_is_no_transform, is_entry_opcode, and
346 set_frag_is_specific_opcode. Use get_subseg_total_freq.
347 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags,
348 xtensa_fix_close_loop_end_frags, relax_frag_immed, convert_frag_immed):
349 Inline calls to get_frag_is_no_transform.
350 (next_instrs_are_b_retw): Inline call to is_windowed_return_opcode.
351 (xtensa_fix_short_loop_frags): Inline calls to is_the_loop_opcode and
352 get_frag_is_no_transform.
353 (convert_frag_immed_finish_loop): Inline calls to get_expression_value
354 and set_frag_is_no_transform.
355 (get_expression_value): Delete.
356 (subseg_map struct): Rename cur_total_freq to total_freq. Rename
357 cur_target_freq to target_freq.
358 (get_subseg_info): Split out code to create a new map entry into ...
359 (add_subseg_info): ... this new function.
360 (get_last_insn_flags): Check if get_subseg_info succeeded.
361 (set_last_insn_flags): Call add_subseg_info if needed.
362 (get_subseg_total_freq, get_subseg_target_freq, set_subseg_freq): New.
363 (xtensa_reorder_segments): Compute last_sec while counting sections.
364 Remove call to get_last_sec.
365 (get_last_sec): Delete.
366 (cache_literal_section): Inline call to retrieve_literal_seg and its
367 callees, seg_present and add_seg_list.
368 (retrieve_literal_seg, seg_present, add_seg_list): Delete.
369 (get_frag_is_insn, get_frag_is_no_transform,
370 set_frag_is_specific_opcode, set_frag_is_no_transform): Delete.
371 * config/tc-xtensa.h (MAX_SLOTS): Reduce from 31 to 15.
373 2004-10-12 Bob Wilson <bob.wilson@acm.org>
375 * config/tc-xtensa.c: Use ISO C90 formatting.
376 * config/tc-xtensa.h: Likewise.
377 * config/xtensa-istack.h: Likewise.
378 * config/xtensa-relax.c: Likewise.
379 * config/xtensa-relax.h: Likewise.
381 2004-10-12 Paul Brook <paul@codesourcery.com>
383 * config/tc-arm.c (md_begin): Change EF_ARM_EABI_VER3 to
386 * doc/c-arm.texi: Document that we actually support -meabi=4, not
389 2004-10-08 Bob Wilson <bob.wilson@acm.org>
391 * doc/as.texinfo (VTableEntry, VTableInherit): Add "directive" to index
393 (Acknowledgements): Use "GAS" instead of AS variable.
395 2004-10-08 Daniel Jacobowitz <dan@debian.org>
397 * config/tc-i386.c: Include "elf/x86-64.h".
398 (i386_elf_section_type): New function.
399 * config/tc-i386.h (md_elf_section_type): Define.
400 (i386_elf_section_type): New prototype.
402 2004-10-08 Linus Nielsen Feltzing <linus@haxx.se>
404 * config/m68k-parse.h (enum m68k_register): New control register,
406 * config/tc-m68k.c: Correct control register set for MCF5249.
408 2004-10-07 Bob Wilson <bob.wilson@acm.org>
409 Sterling Augustine <sterling@tensilica.com>
411 * config/tc-xtensa.c (absolute_literals_supported): New global flag.
412 (UNREACHABLE_MAX_WIDTH): Define.
413 (XTENSA_FETCH_WIDTH): Delete.
414 (cur_vinsn, xtensa_fetch_width, xt_saved_debug_type, past_xtensa_end,
415 prefer_const16, prefer_l32r): New global variables.
416 (LIT4_SECTION_NAME): Define.
417 (lit4_state struct): Add lit4_seg_name and lit4_seg fields.
418 (XTENSA_PROP_*, GET_XTENSA_PROP_*, SET_XTENSA_PROP_*): Define.
419 (frag_flags struct): New.
420 (xtensa_block_info struct): Move from tc-xtensa.h. Add flags field.
421 (subseg_map struct): Add cur_total_freq and cur_target_freq fields.
422 (bitfield, bit_is_set, set_bit, clear_bit): Define.
423 (MAX_FORMATS): Define.
424 (op_placement_info struct, op_placement_table): New.
425 (O_pltrel, O_hi16, O_lo16): Define.
426 (directiveE enum): Rename directive_generics to directive_transform.
427 Delete directive_relax. Add directive_schedule,
428 directive_absolute_literals, and directive_last_directive.
429 (directive_info): Rename "generics" to "transform". Delete "relax".
430 Add "schedule" and "absolute-literals".
431 (directive_state): Adjust entries to match changes in directive_info.
432 (xtensa_relax_statesE, RELAX_IMMED_MAXSTEPS): Move to tc-xtensa.h.
433 (xtensa_const16_opcode, xtensa_movi_opcode, xtensa_movi_n_opcode,
434 xtensa_l32r_opcode, xtensa_nop_opcode, xtensa_rsr_lcount_opcode): New.
435 (xtensa_j_opcode, xtensa_rsr_opcode): Delete.
436 (align_only_targets, software_a0_b_retw_interlock,
437 software_avoid_b_j_loop_end, maybe_has_b_j_loop_end,
438 software_avoid_short_loop, software_avoid_close_loop_end,
439 software_avoid_all_short_loops, specific_opcode): Delete.
440 (warn_unaligned_branch_targets): New.
441 (workaround_a0_b_retw, workaround_b_j_loop_end, workaround_short_loop,
442 workaround_close_loop_end, workaround_all_short_loops): Default FALSE.
443 (option_[no_]link_relax, option_[no_]transform,
444 option_[no_]absolute_literals, option_warn_unaligned_targets,
445 option_prefer_l32r, option_prefer_const16, option_target_hardware):
447 (option_[no_]align_only_targets, option_literal_section_name,
448 option_text_section_name, option_data_section_name,
449 option_bss_section_name, option_eb, option_el): Delete.
450 (md_longopts): Add entries for: [no-]transform, [no-]absolute-literals,
451 warn-unaligned-targets, prefer-l32r, prefer-const16, [no-]link-relax,
452 and target-hardware. Delete entries for [no-]target-align-only,
453 literal-section-name, text-section-name, data-section-name, and
455 (md_parse_option): Handle new options and remove old ones. Accept but
456 ignore [no-]density options. Warn for [no-]generics and [no-]relax
457 and treat them as [no-]transform.
458 (md_show_usage): Add new options and remove old ones.
459 (xtensa_setup_hw_workarounds): New.
460 (md_pseudo_table): Change "word" entry to use xtensa_elf_cons. Add
461 "long", "short", "loc" and "frequency" entries.
462 (use_generics): Rename to ...
463 (use_transform): ... this function. Add past_xtensa_end check.
464 (use_longcalls): Add past_xtensa_end check.
465 (code_density_available, can_relax): Delete.
466 (do_align_targets): New.
467 (get_directive): Accept dashes in directive names. Warn about
468 [no-]generics and [no-]relax directives and treat them as
470 (xtensa_begin_directive): Call md_flush_pending_output only for some
471 directives. Check for directives inside instruction bundles. Warn
472 about deprecated ".begin literal" usage. Warn and ignore [no-]density
473 directives. Handle new directives. Check generating_literals flag
475 (xtensa_end_directive): Check for directives inside instruction
476 bundles. Warn and ignore [no-]density directives. Handle new
477 directives. Call xtensa_set_frag_assembly_state.
478 (xtensa_loc_directive_seen, xtensa_dwarf2_directive_loc,
479 xtensa_dwarf2_emit_insn): New.
480 (xtensa_literal_position): Call md_flush_pending_output. Do not check
481 use_literal_section flag.
482 (xtensa_literal_pseudo): Call md_flush_pending_output. Handle absolute
483 literals. Use xtensa_elf_cons to parse the expression.
484 (xtensa_literal_prefix): Do not check use_literal_section. Support
485 ".lit4" sections for absolute literals. Change prefix convention to
486 replace ".text" (or ".t" in a linkonce section). No need to call
488 (xtensa_frequency_pseudo, xtensa_elf_cons, xtensa_elf_suffix): New.
489 (expression_end): Handle closing braces and colons.
490 (PLT_SUFFIX, plt_suffix): Delete.
491 (expression_maybe_register): Use new xtensa-isa.h functions. Use
492 xtensa_elf_suffix instead of checking for plt suffix, and handle O_lo16
493 and O_hi16 expressions as well.
494 (tokenize_arguments): Handle closing braces and colons.
495 (parse_arguments): Use new xtensa-isa.h functions. Handle "invisible"
496 operands and paired register syntax.
497 (get_invisible_operands): New.
498 (xg_translate_sysreg_op): Handle new Xtensa LX RSR/WSR/XSR syntax. Use
499 new xtensa-isa.h functions.
500 (xtensa_translate_old_userreg_ops, xtensa_translate_zero_immed): New.
501 (xg_translate_idioms): Check if inside bundle. Use use_transform.
502 Handle new Xtensa LX RSR/WSR/XSR syntax. Remove code to widen density
503 instructions. Use xtensa_translate_zero_immed.
504 (operand_is_immed, operand_is_pcrel_label): Delete.
505 (get_relaxable_immed): Use new xtensa-isa.h functions.
506 (get_opcode_from_buf): Add slot parameter. Use new xtensa-isa.h
508 (xtensa_print_insn_table, print_vliw_insn): New.
509 (is_direct_call_opcode): Use new xtensa-isa.h functions.
510 (is_call_opcode, is_loop_opcode, is_conditional_branch_opcode,
511 is_branch_or_jump_opcode): Delete.
512 (is_movi_opcode, decode_reloc, encode_reloc, encode_alt_reloc): New.
513 (opnum_to_reloc, reloc_to_opnum): Delete.
514 (xtensa_insnbuf_set_operand, xtensa_insnbuf_get_operand): Use new
515 xtensa-isa.h functions. Operate on one slot of an instruction.
516 (xtensa_insnbuf_set_immediate_field, is_negatable_branch,
517 xg_get_insn_size): Delete.
518 (xg_get_build_instr_size): Use xg_get_single_size.
519 (xg_is_narrow_insn, xg_is_single_relaxable_insn): Update calls to
520 xg_build_widen_table. Use xg_get_single_size.
521 (xg_get_max_narrow_insn_size): Delete.
522 (xg_get_max_insn_widen_size, xg_get_max_insn_widen_literal_size,
523 xg_is_relaxable_insn): Update calls to xg_build_widen_table. Use
525 (xg_build_to_insn): Record the loc field. Handle OP_OPERAND_HI16U and
526 OP_OPERAND_LOW16U. Check xg_valid_literal_expression.
527 (xg_expand_to_stack, xg_expand_narrow): Update calls to
528 xg_build_widen_table. Use xg_get_single_size.
529 (xg_immeds_fit): Use new xtensa-isa.h functions. Update call to
531 (xg_symbolic_immeds_fit): Likewise. Also handle O_lo16 and O_hi16, and
532 treat weak symbols conservatively.
533 (xg_check_operand): Use new xtensa-isa.h functions.
534 (is_dnrange): Delete.
535 (xg_assembly_relax): Inline previous calls to tinsn_copy.
536 (xg_finish_frag): Specify separate relax states for the frag and slot0.
537 (is_branch_jmp_to_next, xg_add_branch_and_loop_targets): Use new
538 xtensa-isa.h functions.
539 (xg_instruction_matches_option_term, xg_instruction_matches_or_options,
540 xg_instruction_matches_options): New.
541 (xg_instruction_matches_rule): Handle O_register expressions. Call
542 xg_instruction_matches_options.
543 (transition_rule_cmp): New.
544 (xg_instruction_match): Update call to xg_build_simplify_table.
545 (xg_build_token_insn): Record loc fields.
546 (xg_simplify_insn): Check is_specific_opcode field and
547 density_supported flag.
548 (xg_expand_assembly_insn): Skip checking code_density_available. Use
549 new xtensa-isa.h functions. Call use_transform instead of can_relax.
550 (xg_assemble_literal): Add error handling for O_big. Call
551 record_alignment. Handle O_pltrel.
552 (xg_valid_literal_expression): New.
553 (xg_assemble_literal_space): Add slot parameter. Remove call to
554 set_expr_symbol_offset. Add call to record_alignment. Update call to
556 (xg_emit_insn): Delete.
557 (xg_emit_insn_to_buf): Add format parameter. Update calls to
558 xg_add_opcode_fix and xtensa_insnbuf_to_chars.
559 (xg_add_opcode_fix): Change opcode parameter to tinsn and add format
560 and slot parameters. Handle new "alternate" relocations for absolute
561 literals and CONST16 instructions. Check for bad uses of O_lo16 and
562 O_hi16. Use new xtensa-isa.h functions.
563 (xg_assemble_tokens): Delete.
564 (is_register_writer): Use new xtensa-isa.h functions.
565 (is_bad_loopend_opcode): Check for xtensa_rsr_lcount_opcode instead of
566 old-style RSR from LCOUNT.
567 (next_frag_opcode): Delete.
568 (next_frag_opcode_is_loop, next_frag_format_size, frag_format_size,
569 update_next_frag_state): New.
570 (update_next_frag_nop_state): Delete.
571 (next_frag_pre_opcode_bytes): Use next_frag_opcode_is_loop.
572 (xtensa_mark_literal_pool_location): Check use_literal_section flag and
573 the state of the absolute-literals directive. Add calls to
574 record_alignment and xtensa_set_frag_assembly_state. Call
575 xtensa_switch_to_non_abs_literal_fragment instead of
576 xtensa_switch_to_literal_fragment.
578 (assemble_nop): Use build_nop. Update call to xtensa_insnbuf_to_chars.
579 (get_expanded_loop_offset): Change check for undefined opcode to an
581 (xtensa_set_frag_assembly_state, relaxable_section,
582 xtensa_find_unmarked_state_frags, xtensa_find_unaligned_branch_targets,
583 xtensa_find_unaligned_loops, xg_apply_tentative_value): New.
584 (md_begin): Update call to xtensa_isa_init. Initialize linkrelax to 1.
585 Set lit4_seg_name. Call xg_init_vinsn. Initialize new global opcodes.
586 Call init_op_placement_info_table and xtensa_set_frag_assembly_state.
587 (xtensa_init_fix_data): New.
588 (xtensa_frob_label): Reset label symbol to the current frag. Check
589 do_align_targets and generating_literals flag. Propagate frequency
590 info to new alignment frag. Call xtensa_set_frag_assembly_state.
591 (xtensa_unrecognized_line): New.
592 (xtensa_flush_pending_output): Check if inside a bundle. Add a call
593 to xtensa_set_frag_assembly_state.
594 (error_reset_cur_vinsn): New.
595 (md_assemble): Remove check for literal frag. Remove call to
596 istack_init. Call use_transform instead of use_generics. Parse
597 explicit instruction format specifiers. Move code for
598 a0_b_retw_interlock workaround to xg_assemble_vliw_tokens. Call
599 error_reset_cur_vinsn on errors. Add call to get_invisible_operands.
600 Add dwarf2_where call. Remote automatic alignment for ENTRY
601 instructions. Move call to xtensa_clear_insn_labels to the end.
602 Rearrange to handle bundles.
603 (xtensa_cons_fix_new): Delete.
604 (xtensa_handle_align): New.
605 (xtensa_frag_init): Call xtensa_set_frag_assembly_state. Remove
606 assignment to is_no_density field.
607 (md_pcrel_from): Use new xtensa-isa.h functions. Use decode_reloc
608 instead of reloc_to_opnum. Handle "alternate" relocations.
609 (xtensa_force_relocation, xtensa_check_inside_bundle,
610 xtensa_elf_section_change_hook): New.
611 (xtensa_symbol_new_hook): Delete.
612 (xtensa_fix_adjustable): Check for difference of symbols with an
613 offset. Check for external and weak symbols.
614 (md_apply_fix3): Remove cases for XTENSA_OP{0,1,2} relocs.
615 (md_estimate_size_before_relax): Return expansion for the first slot.
616 (tc_gen_reloc): Handle difference of symbols by producing
617 XTENSA_DIFF{8,16,32} relocs and by writing the value of the difference
618 into the output. Handle new XTENSA_SLOT*_OP relocs by storing the
619 tentative values into the output when linkrelax is set.
620 (XTENSA_PROP_SEC_NAME): Define.
621 (xtensa_post_relax_hook): Call xtensa_find_unmarked_state_frags.
622 Create literal tables only if using literal sections. Create new
623 property tables instead of old instruction tables. Check for unaligned
624 branch targets and loops.
625 (finish_vinsn, find_vinsn_conflicts, check_t1_t2_reads_and_writes,
626 new_resource_table, clear_resource_table, resize_resource_table,
627 resources_available, reserve_resources, release_resources,
628 opcode_funcUnit_use_unit, opcode_funcUnit_use_stage,
629 resources_conflict, xg_find_narrowest_format, relaxation_requirements,
630 bundle_single_op, emit_single_op, xg_assemble_vliw_tokens): New.
631 (xtensa_end): Call xtensa_flush_pending_output. Set past_xtensa_end
632 flag. Update checks for workaround options. Call
633 xtensa_mark_narrow_branches and xtensa_mark_zcl_first_insns.
634 (xtensa_cleanup_align_frags): Add special case for branch targets.
635 Check for and mark unreachable frags.
636 (xtensa_fix_target_frags): Remove use of align_only_targets flag.
637 Use RELAX_LOOP_END_BYTES in special case for negatable branch at the
638 end of a zero-overhead loop body.
639 (frag_can_negate_branch): Handle instructions with multiple slots.
640 Use new xtensa-isa.h functions
641 (xtensa_mark_narrow_branches, is_narrow_branch_guaranteed_in_range,
642 xtensa_mark_zcl_first_insns): New.
643 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags): Error if
644 transformations are disabled.
645 (next_instrs_are_b_retw): Use new xtensa-isa.h functions. Handle
646 multislot instructions.
647 (xtensa_fix_close_loop_end_frags, xtensa_fix_short_loop_frags):
648 Likewise. Also error if transformations are disabled.
649 (unrelaxed_frag_max_size): New.
650 (unrelaxed_frag_min_insn_count, unrelax_frag_has_b_j): Use new
651 xtensa-isa.h functions.
652 (xtensa_sanity_check, is_empty_loop, is_local_forward_loop): Use
653 xtensa_opcode_is_loop instead of is_loop_opcode.
654 (get_text_align_power): Replace as_fatal with assertion.
655 (get_text_align_fill_size): Iterate instead of using modulus when
657 (get_noop_aligned_address): Assert that this is for a machine-dependent
658 RELAX_ALIGN_NEXT_OPCODE frag. Use next_frag_opcode_is_loop,
659 xg_get_single_size, and frag_format_size.
660 (get_widen_aligned_address): Rename to ...
661 (get_aligned_diff): ... this function. Add max_diff parameter.
662 Remove handling of rs_align/rs_align_code frags. Use
663 next_frag_format_size, get_text_align_power, get_text_align_fill_size,
664 next_frag_opcode_is_loop, and xg_get_single_size. Compute max_diff
665 and pass it back to caller.
666 (xtensa_relax_frag): Use relax_frag_loop_align. Add code for new
667 RELAX_SLOTS, RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN,
668 RELAX_FILL_NOP, and RELAX_UNREACHABLE frag types. Check relax_seen.
669 (relax_frag_text_align): Rename to ...
670 (relax_frag_loop_align): ... this function. Assume loops can only be
671 in the first slot of an instruction.
672 (relax_frag_add_nop): Use assemble_nop instead of constructing an OR
673 instruction. Remove call to frag_wane.
674 (relax_frag_narrow): Rename to ...
675 (relax_frag_for_align): ... this function. Extend to handle
676 RELAX_FILL_NOP and RELAX_UNREACHABLE, as well as RELAX_SLOTS with
677 RELAX_NARROW for the first slot.
678 (find_address_of_next_align_frag, bytes_to_stretch): New.
679 (future_alignment_required): Use find_address_of_next_align_frag and
680 bytes_to_stretch. Look ahead to subsequent frags to make smarter
682 (relax_frag_immed): Add format, slot, and estimate_only parameters.
683 Check if transformations are enabled for b_j_loop_end workaround.
684 Use new xtensa-isa.h functions and handle multislot instructions.
685 Update call to xg_assembly_relax.
686 (md_convert_frag): Handle new RELAX_SLOTS, RELAX_UNREACHABLE,
687 RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN, and RELAX_FILL_NOP
689 (convert_frag_narrow): Add segP, format and slot parameters. Call
690 convert_frag_immed for branch instructions. Adjust calls to
691 tinsn_from_chars, tinsn_immed_from_frag, and xg_emit_insn_to_buf. Use
692 xg_get_single_size and xg_get_single_format.
693 (convert_frag_fill_nop): New.
694 (convert_frag_immed): Add format and slot parameters. Handle multislot
695 instructions and use new xtensa-isa.h functions. Update calls to
696 tinsn_immed_from_frag and xg_assembly_relax. Check if transformations
697 enabled for b_j_loop_end workaround. Use build_nop instead of
698 assemble_nop. Check is_specific_opcode flag. Check for unreachable
699 frags. Use xg_get_single_size. Handle O_pltrel.
700 (fix_new_exp_in_seg): Remove check for old plt flag.
701 (convert_frag_immed_finish_loop): Update calls to tinsn_from_chars and
702 xtensa_insnbuf_to_chars. Call tinsn_immed_from_frag. Change check
703 for loop opcode to an assertion. Mark all frags up to the end of the
704 loop as not transformable.
705 (get_last_insn_flags, set_last_insn_flags): Use get_subseg_info.
706 (get_subseg_info): New.
707 (xtensa_move_literals): Call xtensa_set_frag_assembly_state. Add null
709 (xtensa_switch_to_literal_fragment): Rewrite to handle absolute
710 literals and use xtensa_switch_to_non_abs_literal_fragment otherwise.
711 (xtensa_switch_to_non_abs_literal_fragment): New.
712 (cache_literal_section): Add is_code parameter and pass it through to
713 retrieve_literal_seg.
714 (retrieve_literal_seg): Add is_code parameter and use it to set the
715 flags on the literal section. Handle case where head parameter is 0.
716 (get_frag_is_no_transform, set_frag_is_specific_opcode,
717 set_frag_is_no_transform): New.
718 (xtensa_create_property_segments): Add end_property_function parameter
719 and pass it through to add_xt_block_frags. Call bfd_get_section_flags
720 and skip SEC_DEBUGGING and !SEC_ALLOC sections.
721 (xtensa_create_xproperty_segments, section_has_xproperty): New.
722 (add_xt_block_frags): Add end_property_function parameter and call it
723 if it is non-zero. Call xtensa_frag_flags_init.
724 (xtensa_frag_flags_is_empty, xtensa_frag_flags_init,
725 get_frag_property_flags, frag_flags_to_number,
726 xtensa_frag_flags_combinable, xt_block_aligned_size,
727 xtensa_xt_block_combine, add_xt_prop_frags,
728 init_op_placement_info_table, opcode_fits_format_slot,
729 xg_get_single_size, xg_get_single_format): New.
730 (istack_push): Inline call to tinsn_copy.
731 (tinsn_copy): Delete.
732 (tinsn_has_invalid_symbolic_operands): Handle O_hi16 and O_lo16 and
733 CONST16 opcodes. Handle O_big, O_illegal, and O_absent.
734 (tinsn_has_complex_operands): Handle O_hi16 and O_lo16.
735 (tinsn_to_insnbuf): Use xg_get_single_format and new xtensa-isa.h
736 functions. Handle invisible operands.
737 (tinsn_to_slotbuf): New.
738 (tinsn_check_arguments): Use new xtensa-isa.h functions.
739 (tinsn_from_chars): Add slot parameter. Rewrite using xg_init_vinsn,
740 vinsn_from_chars, and xg_free_vinsn.
741 (tinsn_from_insnbuf): New.
742 (tinsn_immed_from_frag): Add slot parameter and handle multislot
743 instructions. Handle symbol differences.
744 (get_num_stack_text_bytes): Use xg_get_single_size.
745 (xg_init_vinsn, xg_clear_vinsn, vinsn_has_specific_opcodes,
746 xg_free_vinsn, vinsn_to_insnbuf, vinsn_from_chars, expr_is_register,
747 get_expr_register, set_expr_symbol_offset_diff): New.
748 * config/tc-xtensa.h (MAX_SLOTS): Define.
749 (xtensa_relax_statesE): Move from tc-xtensa.c. Add
750 RELAX_CHECK_ALIGN_NEXT_OPCODE, RELAX_MAYBE_DESIRE_ALIGN, RELAX_SLOTS,
751 RELAX_FILL_NOP, RELAX_UNREACHABLE, RELAX_MAYBE_UNREACHABLE, and
753 (RELAX_IMMED_MAXSTEPS): Move from tc-xtensa.c.
754 (xtensa_frag_type struct): Add is_assembly_state_set,
755 use_absolute_literals, relax_seen, is_unreachable, is_specific_opcode,
756 is_align, is_text_align, alignment, and is_first_loop_insn fields.
757 Replace is_generics and is_relax fields by is_no_transform field.
758 Delete is_text and is_longcalls fields. Change text_expansion and
759 literal_expansion to arrays of MAX_SLOTS entries. Add arrays of
760 per-slot information: literal_frags, slot_subtypes, slot_symbols,
761 slot_sub_symbols, and slot_offsets. Add fr_prev field.
762 (xtensa_fix_data struct): New.
763 (xtensa_symfield_type struct): Delete plt field.
764 (xtensa_block_info struct): Move definition to tc-xtensa.h. Add
765 forward declaration here.
766 (xt_section_type enum): Delete xt_insn_sec. Add xt_prop_sec.
767 (XTENSA_SECTION_RENAME): Undefine.
768 (TC_FIX_TYPE, TC_INIT_FIX_DATA, TC_FORCE_RELOCATION, NO_PSEUDO_DOT,
769 tc_unrecognized_line, md_do_align, md_elf_section_change_hook,
770 HANDLE_ALIGN, TC_LINKRELAX_FIXUP, SUB_SEGMENT_ALIGN): Define.
771 (TC_CONS_FIX_NEW, tc_symbol_new_hook): Delete.
772 (unit_num_copies_func, opcode_num_units_func,
773 opcode_funcUnit_use_unit_func, opcode_funcUnit_use_stage_func): New.
774 (resource_table struct): New.
775 * config/xtensa-istack.h (MAX_INSN_ARGS): Increase from 6 to 10.
776 (TInsn struct): Add keep_wide, loc, fixup, record_fix, subtype,
777 literal_space, symbol, sub_symbol, offset, and literal_frag fields.
778 (tinsn_copy): Delete prototype.
779 (vliw_insn struct): New.
780 * config/xtensa-relax.c (insn_pattern_struct): Add options field.
781 (widen_spec_list): Add option conditions for density and boolean
782 instructions. Add expansions using CONST16 and conditions for using
783 CONST16 vs. L32R. Use new Xtensa LX RSR/WSR syntax. Add entries for
785 (simplify_spec_list): Add option conditions for density instructions.
786 Add entry for NOP instruction.
787 (append_transition): Add cmp function pointer parameter and use it to
788 insert the new entry in order.
789 (operand_function_LOW16U, operand_function_HI16U): New.
790 (xg_has_userdef_op_fn, xg_apply_userdef_op_fn): Handle
791 OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
792 (enter_opname, split_string): Use xstrdup instead of strdup.
793 (init_insn_pattern): Initialize new options field.
794 (clear_req_or_option_list, clear_req_option_list,
795 clone_req_or_option_list, clone_req_option_list, parse_option_cond):
797 (parse_insn_pattern): Parse option conditions.
798 (transition_applies): New.
799 (build_transition): Use new xtensa-isa.h functions. Fix incorrectly
800 swapped last arguments in calls to append_constant_value_condition.
801 Call clone_req_option_list. Add warning about invalid opcode.
802 Handle LOW16U and HI16U function names.
803 (build_transition_table): Add cmp parameter and use it in calls to
804 append_transition. Use new xtensa-isa.h functions. Check
805 transition_applies before adding entries.
806 (xg_build_widen_table, xg_build_simplify_table): Add cmp parameter and
807 pass it through to build_transition_table.
808 * config/xtensa-relax.h (ReqOrOptionList, ReqOrOption, ReqOptionList,
809 ReqOption, transition_cmp_fn): New types.
810 (OpType enum): Add OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
811 (transition_rule struct): Add options field.
812 * doc/as.texinfo (Overview): Update Xtensa options.
813 * doc/c-xtensa.texi (Xtensa Options): Delete --[no-]density,
814 --[no-]relax, and --[no-]generics options. Update descriptions of
815 --text-section-literals and --[no-]longcalls. Add
816 --[no-]absolute-literals and --[no-]transform.
817 (Xtensa Syntax): Add description of syntax for FLIX instructions.
818 Remove use of "generic" and "specific" terminology for opcodes.
819 (Xtensa Registers): Generalize the syntax description to include
820 user-defined register files.
821 (Xtensa Automatic Alignment): Update.
822 (Xtensa Branch Relaxation): Mention limitation of unconditional jumps.
823 (Xtensa Call Relaxation): Linker can now remove most of the overhead.
824 (Xtensa Directives): Remove confusing rules about precedence.
825 (Density Directive, Relax Directive): Delete.
826 (Schedule Directive): New.
827 (Generics Directive): Rename to ...
828 (Transform Directive): ... this node.
829 (Literal Directive): Update for absolute literals. Missing
830 literal_position directive is now an error.
831 (Literal Position Directive): Update for absolute literals.
832 (Freeregs Directive): Delete.
833 (Absolute Literals Directive): New.
834 (Frame Directive): Minor editing.
835 * Makefile.am (DEPTC_xtensa_elf, DEPOBJ_xtensa_elf, DEP_xtensa_elf):
837 * Makefile.in: Regenerate.
839 2004-10-07 Richard Sandiford <rsandifo@redhat.com>
841 * config/tc-mips.c (append_insn): Use fix_new rather than fix_new_exp
842 to build the second and third fixups for a composite relocation.
843 (macro_read_relocs): New function.
844 (macro_build): Use it.
845 (s_cpsetup): Pass all three composite relocation codes to macro_build.
846 Simplify fragging code accordingly.
847 (s_gpdword): Use fix_new rather than fix_new_exp for the second part
848 of the composite relocation. Set fx_tcbit in both fixups.
850 2004-10-07 Richard Sandiford <rsandifo@redhat.com>
852 * config/tc-mips.c (append_insn): Set fx_tcbit for composite relocs.
853 (md_apply_fix3): Don't treat composite relocs as done.
855 2004-10-07 Jan Beulich <jbeulich@novell.com>
857 * macro.c (macro_expand_body): When ELF, use .LL rather than LL as
858 prefix for symbol names generated from the LOCAL macro directive.
860 * dw2gencfi.c (select_cie_for_fde): When separating CIE out from
861 FDE, treat a DW_CFA_remember_state as we do a DW_CFA_advance_loc.
863 2004-10-07 Tomer Levi <Tomer.Levi@nsc.com>
865 * config/tc-crx.c (preprocess_reglist): Handle Co-processor
867 (md_assemble): Add error checking for Co-Processor instructions.
868 (get_cinv_parameters): Add 'b' option to invalidate the
871 2004-10-05 Paul Brook <paul@codesourcery.com>
873 * config/tc-arm.c (unwind): New variable.
874 (vfp_sp_encode_reg): New function.
875 (vfp_sp_reg_required_here): Use it.
876 (vfp_sp_reg_list, vfp_dp_reg_list): Remove.
877 (vfp_parse_reg_list): New function.
878 (s_arm_unwind_fnstart, s_arm_unwind_fnend, s_arm_unwind_cantunwind,
879 s_arm_unwind_personality, s_arm_unwind_personalityindex,
880 s_arm_unwind_handlerdata, s_arm_unwind_save, s_arm_unwind_movsp,
881 s_arm_unwind_pad, s_arm_unwind_setfp, s_arm_unwind_raw): New
883 (md_pseudo_table): Add them.
884 (do_vfp_reg2_from_sp2): Use vfp_parse_reg_list and vfp_sp_encode_reg.
885 (do_vfp_sp2_from_reg2, vfp_sp_ldstm, vfp_dp_ldstm): Ditto.
886 (set_section, add_unwind_adjustsp, flush_pending_unwind,
887 finish_unwind_opcodes, start_unwind_section, create_unwind_entry,
888 require_hashconst, add_unwind_opcode): New functions.
889 * doc/tc-arm.text: Document unwinding opcodes.
890 * NEWS: Mention the new feature.
892 2004-10-04 Eric Christopher <echristo@redhat.com>
894 * config/tc-mips.c (md_apply_fix3): Remove erroneous assert.
896 2004-10-01 H.J. Lu <hongjiu.lu@intel.com>
898 * config/tc-ppc.c (md_apply_fix3): Call S_SET_THREAD_LOCAL for
900 * config/tc-s390.c (md_apply_fix3): Likewise.
901 * config/tc-sparc.c (md_apply_fix3): Likewise.
903 2004-10-01 Paul Brook <paul@codesourcery.com>
905 * config/tc-arm.c (arm_elf_section_type): New function.
906 (arm_elf_change_section): Set section link for exidx sections.
907 * config/tc-arm.h (arm_elf_section_type): Add prototype.
908 (md_elf_section_type): Define.
910 2004-10-01 Bill Farmer <Bill@the-farmers.freeserve.co.uk>
912 * config/tc-pdp11.c (md_apply_fix3): Change to sign of the SOB
913 instruction's offset.
915 2004-10-01 Adam Nemet <anemet@lnxw.com>
917 * (TARGET_FORMAT): Remove LynxOS COFF definition.
919 2004-10-01 Ravi Ramaseshan <ravi.ramaseshan@codito.com>
921 * config/tc-arc.c (tc_gen_reloc): Don't assume fixP->fx_addsy is an
922 asymbol *, instead use symbol_get_bfdsym.
924 2004-09-30 Linus Nielsen Feltzing <linus@haxx.se>
926 * config/tc-m68k.c (select_control_regs): Add mcf5249.
928 2004-09-30 Paul Brook <paul@codesourcery.com>
930 * config/tc-arm.c (do_smi, do_nop): New functions.
931 (insns): Add ARMv6ZK instructions.
932 (md_apply_fix3): Handle BFD_RELOC_ARM_SMI.
933 (tc_gen_reloc): Ditto.
934 (arm_cpus): Add mpcore and arm1176.
935 (arm_archs): Add armv6{k,z,zk}.
936 * doc/c-arm.texi: Document new cores and architectures.
938 2004-09-30 Nick Clifton <nickc@redhat.com>
940 * config/tc-arm.c: Use ISO C90 formatting.
942 2004-09-30 Vladimir Ivanov <vladitx@nucleusys.com>
944 * config/tc-arm.c (mav_reg_required_here): Allow REG_TYPE_CN
945 as alternative when REG_TYPE_MVF, REG_TYPE_MVD, REG_TYPE_MVFX or
946 REG_TYPE_MVDX is expected.
948 2004-09-29 Marc Bevand <m.bevand@gmail.com>
950 * doc/c-i386.texi (i386-Mnemonics): Fix typo.
952 2004-09-21 James E Wilson <wilson@specifixinc.com>
954 * config/tc-ia64.c (ENCODED_PSP_OFFSET): New.
955 (output_rp_psprel, output_pfs_psprel, output_preds_psprel,
956 output_spill_base, output_unat_psprel, output_lc_psprel,
957 output_fpsr_psprel, output_priunat_psprel, output_bsp_psprel,
958 output_bsprestore_psprel, output_rnat_psprel, output_spill_psprel,
959 output_spill_psprel_p): Use it.
961 2004-09-20 Tomer Levi <Tomer.Levi@nsc.com>
963 * config/tc-crx.c (handle_LoadStor): New function.
964 Handle load/stor unique instructions before parsing.
966 2004-09-17 Paul Brook <paul@codesourcery.com>
968 * config/tc-arm.c (s_arm_rel31): New funciton.
969 (md_pseudo_table): Add .rel31.
970 (md_apply_fix3): Handle BFD_RELOC_ARM_TARGET2,
971 BFD_RELOC_32_PCREL and BFD_RELOC_ARM_PREL31.
972 (tc_gen_reloc): Handle BFD_RELOC_ARM_PREL31 and BFD_RELOC_ARM_TARGET2.
973 (arm_fix_adjustable): Return 0 for BFD_RELOC_ARM_TARGET2.
974 (arm_parse_reloc): Add (target2).
976 2004-09-17 Alan Modra <amodra@bigpond.net.au>
978 * Makefile.am: Run "make dep-am".
979 * Makefile.in: Regenerate.
980 * aclocal.m4: Regenerate.
981 * configure: Regenerate.
982 * doc/Makefile.in: Regenerate.
983 * po/POTFILES.in: Regenerate.
984 * po/gas.pot: Regenerate.
986 2004-09-14 Hideki IWAMOTO <h-iwamoto@kit.hi-ho.ne.jp>
988 * config/tc-mmix.c [!LLONG_MIN]: Correct #elsif to #elif.
991 2004-09-13 Paul Brook <paul@codesourcery.com>
993 * config/tc-arm.c: Rename RELABS to TARGET1.
995 2004-09-13 Alan Modra <amodra@bigpond.net.au>
997 * messages.c (as_internal_value_out_of_range): Cast values passed
998 to as_bad_where or as_warn_where to proper type.
1000 2004-09-11 Theodore A. Roth <troth@openavr.org>
1002 * config/tc-avr.c: Add support for
1003 atmega48, atmega88, atmega168, attiny13, attiny2313, at90can128.
1005 2004-09-09 Alan Modra <amodra@bigpond.net.au>
1007 * dw2gencfi.c (select_cie_for_fde): When separating CIE out
1008 from FDE, treat a CFI_escape as we do a DW_CFA_advance_loc.
1010 2004-09-08 Paul Brook <paul@codesourcery.com>
1012 * config/obj-elf.c (obj_elf_section_type): Handle init_array,
1013 fini_array and preinit_array section types.
1014 * config/tc-ia64.c (ia64_elf_section_type): Remove init_array
1016 * doc/as.texinfo: Document extra section types.
1018 2004-09-02 Mark Mitchell <mark@codesourcery.com>
1020 * Makefile.am (TARG_ENV_HFILES): Add te-symbian.h.
1021 * Makefile.in: Regenerated.
1022 * configure.in: Set em for arm*-*-symbianelf*.
1023 * configure: Regenerated.
1024 * config/tc-arm.c (elf32_arm_target_format): Use Symbian target
1025 vectors when appropriate.
1026 * config/te-symbian.h: New file.
1028 2004-09-03 Tomer Levi <Tomer.Levi@nsc.com>
1030 * config/tc-crx.c (gettrap): Exception vector can be case
1032 (process_label_constant): Fix a 32-bit displacement bug in branch
1034 (get_operandtype) : Bug fix, wrong operand was used.
1035 (process_label_constant): Initialize relocation type to
1038 2004-09-01 Richard Earnshaw < reanrsha@arm.com>
1040 * tc-arm.c (arm_cpus, arm_fpus): Allow <cpu>-s as well as <cpu>s
1041 for synthesizable cores.
1043 * doc/c-arm.texi (ARM Options): Document canonical names of CPUs.
1045 2004-08-25 Dmitry Diky <diwil@spec.ru>
1047 * config/tc-msp430.c: Clean-up the code.
1048 (md_relax_table): New relax table.
1049 (mcu_types): Sort MCU types.
1050 (md_pseudo_table): Add .profiler pseudo handler.
1051 (pow2value): New function.
1052 (msp430_profiler): New function.
1053 (msp430_operands): Add new insns handlers.
1054 (msp430_srcoperand): Add register operand handler, allow complex
1056 (md_estimate_size_before_relax): Rewritten.
1057 (md_convert_frag): Rewritten.
1058 (msp430_relax_frag): New function.
1059 * config/tc-msp430.h (md_relax_frag): define macro
1060 * doc/c-msp430.texi: Update information.
1062 2004-08-24 Nick Clifton <nickc@redhat.com>
1064 * as.c (std_shortopts): Allow -g to take an optional argument.
1065 (parse_args): Pass any switch starting with -g on to the backend
1068 2004-08-18 Mark Mitchell <mark@codesourcery.com>
1070 * configure.in (arm*-*-symbianelf*): New target.
1071 (arm*-*-eabi*): Likewise.
1072 * configure: Regenerated.
1074 2004-08-18 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1075 * config/tc-mips.c (append_insn): Handle delay slots in branch likely
1078 2004-08-18 Jakub Jelinek <jakub@redhat.com>
1080 * config/tc-ia64.c (start_unwind_section): Add linkonce_empty
1081 argument, don't do anything if current section is not
1082 .gnu.linkonce.t.* and linkonce_empty is set.
1083 (generate_unwind_image, dot_endp): Adjust callers, call
1084 start_unwind_section (*, 1) if nothing will be put into the
1087 2004-08-17 Nick Clifton <nickc@redhat.com>
1089 * as.c (MD_DEBUG_FORMAT_SELECTOR): Provide default definition.
1090 (show_usage): Add -g.
1091 (std_longopts): Add --gen-debug. Alpha sort the table.
1092 (parse_args): Print an error message if a switch is not handled.
1093 Handle the -g switch, calling md_debug_format_selector() if
1095 * NEWS: Mention new feature.
1096 * doc/as.texinfo: Document new switch.
1097 * doc/internals.texi: Document behaviour of md_parse_option.
1099 * config/tc-arm.c (md_parse_option): Do not issue an error message
1100 if the switch is not recognised.
1101 * config/tc-m68k.c (md_parse_option): Likewise.
1102 * config/tc-pdp11.c (md_parse_option): Likewise.
1103 * config/tc-v850.c (md_parse_option): Likewise.
1105 * as.h: Fix up formatting.
1108 2004-08-16 Nick Clifton <nickc@redhat.com>
1110 * macro.c (macro_set_alternate): Use ISO C90 formatting.
1112 * configure.in: Sort architecture based tables alphabetically.
1113 * configure: Regenerate.
1115 2004-08-16 Alan Modra <amodra@bigpond.net.au>
1117 * config/tc-ppc.c (tc_ppc_regname_to_dw2regnum <regnames>): Replace
1118 { "cc", 68 }, with { "cr", 70 }.
1120 2004-08-13 Jan Beulich <jbeulich@novell.com>
1121 Nick Clifton <nickc@redhat.com>
1123 * as.c: Add and handle new --alternate command line option.
1124 * macro.c (macro_set_alternate): New.
1125 * macro.h (macro_set_alternate): Declare.
1126 * read.c: Add and handle new .altmacro and .noaltmacro directives.
1127 * doc/as.texinfo: Document new command line option and pseudo-ops
1128 as well as insert documentation originating from gasp about
1129 alternate macro syntax.
1130 * NEWS: Mention new command line option and pseudo-ops.
1132 2004-08-10 Mark Mitchell <mark@codesourcery.com>
1134 * expr.c (operand): Handle the "~", "-", and "!" operators applied
1137 2004-08-06 Paul Brook <paul@codesourcery.com>
1139 * config/tc-arm.c (md_apply_fix3, tc_gen_reloc, arm_parse_reloc):
1140 Handle new relocations.
1141 * include/elf/arm.h (elf_arm_reloc_type): Add new EABI relocations.
1143 2004-08-05 Bob Wilson <bob.wilson@acm.org>
1145 * write.c (relax_segment): Use was_address instead of address when
1146 setting fr_fix field for align frag due to backwards .org.
1148 2004-07-29 Alexandre Oliva <aoliva@redhat.com>
1150 Introduce SH2a support.
1151 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
1152 * config/tc-sh.c (get_specific): Change arch_sh2a_up to
1154 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
1155 * config/tc-sh.c (md_parse_option): Add sh2a-nofpu ISA handling.
1156 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
1157 * config/tc-sh.c (sh_elf_final_processing): Move sh2a recognition
1158 to end of conditional expression.
1159 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
1160 * config/tc-sh.c: Add sh2a-nofpu support.
1161 2003-12-29 DJ Delorie <dj@redhat.com>
1162 * tc-sh.c: Add sh2a support.
1163 (parse_reg): Add tbr.
1164 (parse_at): Support @@(disp,tbr).
1165 (get_specific): Support sh2a opcodes.
1166 (insert4): New, for 4 byte relocs.
1167 (build_Mytes): Support sh2a opcodes.
1168 (md_apply_fix3_Mytes): Support sh2a opcodes.
1169 2003-12-02 Michael Snyder <msnyder@redhat.com>
1170 * config/tc-sh.c (md_parse_option): Handle sh2a.
1171 (sh_elf_final_processing): Ditto.
1173 2004-07-27 Jason Thorpe <thorpej@wasabisystems.com>
1175 * config/tc-hppa.h (TARGET_FORMAT): Set to "elf32-hppa-netbsd"
1178 2004-07-27 Alan Modra <amodra@bigpond.net.au>
1180 * config/tc-ppc.c (ppc_frob_file_before_adjust): Warn if .toc too big.
1181 (ppc_arch): Expand comment.
1183 2004-07-27 Tomer Levi <Tomer.Levi@nsc.com>
1185 * config/tc-crx.c: Support evaluating the difference between two
1187 * config/tc-crx.h: Likewise.
1189 2004-07-26 H.J. Lu <hongjiu.lu@intel.com>
1191 * config/tc-ia64.c (start_unwind_section): Set the linked-to
1193 (ia64_elf_section_change_hook): Set the linked-to section for
1196 2004-07-26 Dmitry Diky <diwil@spec.ru>
1198 * config/tc-msp430.c: Add new subtargets: msp430x1610,
1199 msp430x1611, msp430x1612, msp430x415, msp430x417, msp430xG437,
1200 msp430xG438, msp430xG439.
1202 2004-07-25 Daniel Jacobowitz <dan@debian.org>
1204 * doc/as.texinfo (Section, PushSection): Correct documentation
1207 2004-07-21 Jan Beulich <jbeulich@novell.com>
1209 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
1210 permissible for the selected instruction suffix.
1211 (match_template): Don't permit 64-bit general purpose operands in
1213 (finalize_imm): Permit 64-bit immediates.
1214 (build_modrm_byte): Don't treat 32-bit addressing in 64-bit mode
1215 specially except for the width of the used base and/or index
1216 registers. For 32-bit displacements, use sign-extended
1217 relocations only when using 64-bit addressing.
1218 Force zero displacement on rip-relative addressing when there is
1219 no other displacement.
1220 (i386_index_check): Don't treat 32-bit addressing in 64-bit mode
1221 specially except for the width of the used base and/or index
1223 (parse_register): Disallow Reg64 registers in 32-bit mode.
1225 * config/tc-i386.c: For DefaultSize instructions, don't guess a 'q'
1226 suffix if the instruction doesn't support it.
1228 2004-07-20 Maciej W. Rozycki <macro@linux-mips.org>
1230 * config/tc-mips.c (append_insn): Handle constant expressions with
1231 no associated relocation.
1232 (mips_ip): Cancel the expression after use for the Q format
1234 (parse_relocation): Return no relocation for unsupported
1236 (my_getSmallExpression): Return no relocation if no relocation
1239 2004-07-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1241 * config/obj-som.c (adjust_stab_sections): Add prototype.
1242 (obj_som_compiler, obj_som_version, obj_som_copyright,
1243 adjust_stab_sections): Add ATTRIBUTE_UNUSED to unused arguments.
1244 * config/tc-hppa.c (update_subspace): Likewise.
1245 (is_defined_subspace): Amplify comment.
1246 * config/obj-som.h (som_frob_file): Add prototype.
1248 2004-07-19 Christopher Faylor <cgf@timesys.com>
1249 H.J. Lu <hongjiu.lu@intel.com>
1251 * subsegs.c (section_symbol): Don't create a new segment when
1252 existing segment is undefined.
1254 2004-07-16 Richard Earnshaw <rearnsha@arm.com>
1256 * config/tc-arm.c: Include include/opcode/arm.h.
1257 (ARM_EXT_*, ARM_ARCH_*, ARM_ANY, ARM_ALL, COPROC_ANY): Delete.
1258 (FPU_FPA_EXT_* FPU_VFP_EXT_*, FPU_ANY, FPU_NONE, FPU_MAVERICK): Delete.
1259 (FPU_ARCH_*): Delete.
1260 * Makefile.am: Update dependencies.
1261 * Makefile.in: Regenerate.
1263 2004-07-15 Nick Clifton <nickc@redhat.com>
1265 * configure.in: Accept armbe as a big-endian arm configuration.
1266 * configure: Regenerate.
1268 2004-07-13 Thomas Nystrom <thn@saeab.se>
1270 * config/tc-i386.c (T_SHIFTOP): New constant.
1271 (intel_e05_1): Handle '&', '|' and T_SHIFTOP.
1272 (intel_el1): Handle '~'.
1273 (intel_get_token): Handle '<>', '&', '|' and '~'.
1275 2004-07-13 Nick Clifton <nickc@redhat.com>
1277 (md_assemble): Remove spurious newline from end of as_bad error
1279 (intel_e05_1): Likewise.
1280 (intel_e11): Likewise.
1281 (intel_match_token): Likewise.
1283 2004-07-11 Andreas Schwab <schwab@suse.de>
1285 * config/tc-m68k.c: Convert to C90. Remove redundant
1286 declarations. Indentation fixup.
1287 [M68KCOFF]: Include "obj-coff.h" instead of declaring
1288 obj_coff_section ourselves.
1290 2004-07-09 James E Wilson <wilson@specifixinc.com>
1292 * config/tc-ia64.c (default_big_endian): New.
1293 (dot_byteorder, md_begin): Use it.
1294 (md_parse_option): Set it.
1296 2004-07-09 Nick Clifton <nickc@redhat.com>
1298 * configure.in: Change sh-sybmian-elf to sh-*-symbianelf.
1299 * configure: Regenerate.
1300 * NEWS: Change sh-sybmian-elf to sh-*-symbianelf.
1301 * config/tc-sh.c (sh_elf_final_processing): Use renamed version of
1302 sh_find_elf_flags if necessary.
1304 2004-07-08 Richard Sandiford <rsandifo@redhat.com>
1306 * config/tc-mips.c (mips_fix_adjustable): If the full addend is
1307 going to be split into more than one in-place addend, return 0
1308 for relocations against mergeable sections. Associate comments
1311 2004-07-07 Tomer Levi <Tomer.Levi@nsc.com>
1313 * Makefile.am (CPU_TYPES): Add crx.
1314 (TARGET_CPU_CFILES): Add config/tc-crx.c.
1315 (TARGET_CPU_HFILES): Add config/tc-crx.h.
1316 (DEPTC_crx_elf): New target.
1317 (DEPOBJ_crx_elf): Likewise.
1318 (DEP_crx_elf): Likewise.
1319 * Makefile.in: Regenerate.
1320 * configure.in: Add crx* target.
1321 * configure: Regenerate.
1322 * config/tc-crx.c: New file.
1323 * config/tc-crx.h: New file.
1324 * NEWS: Mention new target.
1326 2004-07-06 Nick Clifton <nickc@redhat.com>
1328 * config.in: Undefine TARGET_SYMBIAN by default.
1330 * configure: Regenerate. Add sh-symbian-elf target. If
1331 selected define TARGET_SYMBIAN.
1332 * config/tc-sh.h (TARGET_FORMAT): Select a Symbian target
1333 format if TARGET_SYMBIAN has been defined.
1335 * output-file.c (output_file_create): Report the target format
1336 chosen when bfd_openw reports that it is invalid.
1338 * config/obj-coff.c (coff_pseudo_table): Only define the weak
1339 pseudo for BFD based assemblers.
1341 2004-07-05 Andrew Stubbs <andrew.stubbs@superh.com>
1344 * config/tc-sh.c (md_assemble): Change isspace to ISSPACE.
1345 (md_parse_option): Remove redundant -isa testing.
1346 Make bfd_arch variable const.
1347 (md_show_usage): Make bfd_arch variable const.
1349 2004-07-03 James E Wilson <wilson@specifixinc.com>
1351 * config/tc-ia64.c (emit_one_bundle): Check and set insn_addr.
1352 * config/tc-ia64.h (md_frag_check): Define.
1354 2004-07-03 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
1356 * config/obj-coff.c (obj_coff_weak): New .weak syntax for PE weak
1358 * doc/as.texinfo (Weak): Document PE weak symbols.
1360 2004-07-03 Richard Sandiford <rsandifo@redhat.com>
1362 * config/tc-mips.c (HAVE_IN_PLACE_ADDENDS): New macro.
1363 (reloc_needs_lo_p): Only return true if HAVE_IN_PLACE_ADDENDS.
1364 (mips_frob_file): Rework so that only a single pass through the
1365 relocs is needed. Allow %lo()s to have higher offsets than their
1366 corresponding %hi()s or %got()s.
1368 2004-07-02 Nick Clifton <nickc@redhat.com>
1370 * config/tc-arm.c (md_apply_fix3:BFD_RELOC_ARM_IMMEDIATE): Do not
1371 allow values which have come from undefined symbols.
1372 Always consider this fixup to have been processed as a reloc
1373 cannot be generated for it.
1375 2004-07-02 Alan Modra <amodra@bigpond.net.au>
1377 * frags.h (struct frag): Add has_code and insn_addr fields.
1378 * write.c (cvt_frag_to_fill): Invoke md_frag_check.
1379 * config/tc-ppc.c (md_assemble): Check and set insn_addr.
1380 * config/tc-ppc.h (md_frag_check): Define.
1382 2004-06-28 Maciej W. Rozycki <macro@linux-mips.org>
1384 * doc/Makefile.am (info): Rename goal to...
1385 (info-local): ... this, to preserve implicit dependencies.
1386 * doc/Makefile.in: Regenerate with automake 1.8.5.
1388 2004-06-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1390 * config/tc-m32r.c (md_convert_frag): Changed for @PLT.
1391 (m32r_cgen_record_fixup_exp): Changed for @GOTOFF, @GOT.
1392 (m32r_fix_adjustable): Changed for @GOTOFF, @GOT, @PLT.
1393 (tc_gen_reloc): Likewise.
1394 (m32r_end_of_match): Add for @GOTOFF, @GOT, @PLT.
1395 (m32r_parse_name): Likewise.
1396 (m32r_cgen_parse_fix_exp): Likewise.
1397 * config/tc-m32r.h (md_parse_name): Define for @GOTOFF, @GOT, @PLT.
1398 (O_PIC_reloc): Likewise.
1399 (TC_CGEN_PARSE_FIX_EXP): Likewise..
1400 * cgen.c (gas_cgen_parse_operand): Add TC_CGEN_PARSE_FIX_EXP
1401 for @GOTOFF, @GOT, @PLT.
1403 2004-06-21 Jan Beulich <jbeulich@novell.com>
1405 * gas/symbols.c: While discarding ordinary local absolute symbols
1406 when --strip-local-absolute is in effect, retain file symbols.
1408 2004-06-20 Andreas Schwab <schwab@suse.de>
1410 * config/tc-m68k.c (mri_chip): Replace current_chip, not augment.
1411 (md_parse_option): Likewise.
1413 2004-06-17 Jan Beulich <jbeulich@novell.com>
1415 * config/tc-i386.c: Deal with LEX_QM the same way as with LEX_AT.
1416 * config/te-netware.h: New file.
1417 * config/te-ppcnw.h: Delete: Obsolete.
1418 * configure.in: Eliminate ill NetWare targets. Make generic
1419 NetWare target use proper emulation.
1420 * Makefile.am: Eliminate reference to obsolete te-ppcnw.h, add
1421 reference to new te-netware.h.
1422 * configure: Regenerate.
1423 * Makefile.in: Regenerate.
1425 2004-06-15 Martin Schwidefsky <schwidefsky@de.ibm.com>
1427 * config/tc-s390.c (s390_insn): Avoid incorrect signed/unsigned
1428 comparison in .insn pseudo operation.
1430 2004-06-15 Alan Modra <amodra@bigpond.net.au>
1432 * config/obj-coff.c (coff_adjust_section_syms): Use
1433 bfd_get_section_size instead of bfd_get_section_size_before_reloc.
1434 (coff_frob_section): Likewise.
1435 * config/tc-mips.c (md_apply_fix3): Likewise.
1436 * config/obj-elf.c (elf_frob_file): Use bfd_set_section_size.
1437 (elf_frob_file_after_relocs): Likewise.
1439 2004-06-10 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1441 * config/tc-hppa.c (log2): Rename to exact_log2.
1442 (pa_next_subseg): Delete unused function.
1443 (create_new_space): Mark unused arguments with ATTRIBUTE_UNUSED.
1444 (create_new_subspace): Likewise.
1447 * config/tc-hppa.c (hppa_fix_adjustable): Allow reduction of fake
1448 labels. Fix warning.
1450 2004-05-28 DJ Delorie <dj@redhat.com>
1452 * config/tc-mn10300.h (tc_fix_adjustable): Define.
1453 * config/tc-mn10300.c (mn10300_fix_adjustable): Don't adjust debug
1454 or non-merged symbols.
1456 2004-05-28 H.J. Lu <hongjiu.lu@intel.com>
1458 * config/tc-ia64.c (remove_marked_resource): Save, clear and
1459 restore the old slot when inserting srlz.i/srlz.d.
1461 2004-05-28 Andrew Stubbs <andrew.stubbs@superh.com>
1463 * Makefile.am: Regenerate dependecies.
1464 * Makefile.in: Regenerate.
1465 * config/tc-sh.c (valid_arch): Make unsigned.
1466 (preset_target_arch): Likewise.
1467 (md_begin): Use new architecture flags system.
1468 (get_specific): Likewise.
1469 (assemble_ppi): Likewise.
1470 (md_assemble): Likewise. Also fix error check for bad opcodes.
1471 (md_parse_option): Likewise. Also generate -isa values according
1472 to the table in bfd/cpu-sh.c instead of just constants. Also
1473 allow <arch>-up ISA variants.
1474 (sh_elf_final_processing): Replace if-else chain with a call to
1475 sh_find_elf_flags().
1476 * testsuite/gas/sh/arch: New directory.
1477 * testsuite/gas/sh/arch/arch.exp: New test script.
1478 * testsuite/gas/sh/arch/arch_expected.txt: New file.
1479 * testsuite/gas/sh/arch/sh.s: New file.
1480 * testsuite/gas/sh/arch/sh2.s: New file.
1481 * testsuite/gas/sh/arch/sh-dsp.s: New file.
1482 * testsuite/gas/sh/arch/sh2e.s: New file.
1483 * testsuite/gas/sh/arch/sh3-nommu.s: New file.
1484 * testsuite/gas/sh/arch/sh3.s: New file.
1485 * testsuite/gas/sh/arch/sh3-dsp.s: New file.
1486 * testsuite/gas/sh/arch/sh3e.s: New file.
1487 * testsuite/gas/sh/arch/sh4-nommu-nofpu.s: New file.
1488 * testsuite/gas/sh/arch/sh4-nofpu.s: New file.
1489 * testsuite/gas/sh/arch/sh4.s: New file.
1490 * testsuite/gas/sh/arch/sh4a-nofpu.s: New file.
1491 * testsuite/gas/sh/arch/sh4al-dsp.s: New file.
1492 * testsuite/gas/sh/arch/sh4a.s: New file.
1494 2004-05-28 Peter Barada <peter@the-baradas.com>
1496 * config/gc-m68k.c(m68k_ip): Convert mode 5 addressing
1497 with zero offset into mode 2 addressing to save a word.
1499 2004-05-27 H.J. Lu <hongjiu.lu@intel.com>
1501 * config/tc-ia64.c (ar_is_in_integer_unit): Removed.
1502 (ar_is_only_in_integer_unit): New.
1503 (ar_is_only_in_memory_unit): New.
1504 (generate_unwind_image): Silence gcc on 32bit host.
1505 (md_assemble): Use ar_is_only_in_integer_unit instead of
1506 ar_is_in_integer_unit. Check AR access.
1508 2004-05-27 Peter Barada <peter@the-baradas.com>
1510 * config/tc-m68k.c (md_begin): Sort the opcode table into
1512 (m68k_compare_opcode): New function to do the sorting.
1514 2004-05-24 Peter Barada <peter@the-baradas.com>
1516 * config/m68k-parse.y(operand): Allow for MAC/EMAC mask
1517 addressing on MIT style operands.
1518 * config/m68k-parse.y(yylex): Allow '-&' for predecrement
1520 * config/tc-m68k.c(install_operand): Comment 'G' and 'H' type
1523 2004-05-23 Alan Modra <amodra@bigpond.net.au>
1525 * expr.c (operand, operator): Don't reject '++' and '--'.
1527 2004-05-20 Richard Sandiford <rsandifo@redhat.com>
1529 * config/tc-mips.c (append_insn): Use ISA-encoded addresses in MIPS16
1532 2004-05-17 Adam Nemet <anemet@lnxw.com>
1534 * configure.in: Add ppc-*-lynxos*. Update i386-*-lynxos* to ELF.
1535 * configure: Regenerate.
1537 2004-05-13 Paul Brook <paul@codesourcery.com>
1539 * dw2gencfi.c (output_cie): Handle dwarf3 format CIE entries.
1541 2004-05-13 Joel Sherrill <joel@oarcorp.com>
1543 * configure.in (or32-*-rtems*): Switch to elf.
1544 * configure: Regenerate.
1546 2004-05-13 Nick Clifton <nickc@redhat.com>
1548 * po/fr.po: Updated French translation.
1550 2004-05-11 Nick Clifton <nickc@redhat.com>
1552 * doc/as.texinfo (Section): Document G and T flags to .section
1553 directive. Document the extra arguments that the G flag
1554 requires. Document the #tls flag.
1556 2004-05-11 H.J. Lu <hongjiu.lu@intel.com>
1558 * subsegs.c (section_symbol): Create a new section symbol if
1559 the existing one doesn't match.
1560 * symbols.c (symbol_set_bfdsym): Don't reset BFD section symbol.
1562 2004-05-07 Richard Sandiford <rsandifo@redhat.com>
1564 * config/tc-mips.c (append_insn, mips_emit_delays): Extend -mfix-vr4120
1565 to cope with VR4181A errata MD(1) and MD(4).
1567 2004-05-07 Brian Ford <ford@vss.fsi.com>
1569 * NEWS: Mention .secrel32 for pe[i]-i386.
1571 2004-05-07 Alexandre Oliva <aoliva@redhat.com>
1573 * config/tc-frv.h (MAX_MEM_FOR_RS_ALIGN_CODE): New.
1574 (HANDLE_ALIGN): New.
1576 2004-05-06 Daniel Jacobowitz <dan@debian.org>
1578 * Makefile.am (DIST_SUBDIRS): Define.
1579 * aclocal.m4: Regenerate with automake 1.8.4.
1580 * Makefile.in: Likewise.
1581 * doc/Makefile.in: Likewise.
1583 2004-05-06 David Mosberger-Tang <davidm@hpl.hp.com>
1585 * config/tc-ia64.c (dot_serialize): Declare.
1586 (dot_serialize): New function.
1587 (md_pseudo_table): Add ".serialize.data" and
1588 ".serialize.instruction" directives.
1590 2004-05-06 Nick Clifton <nickc@redhat.com>
1592 * messages (as_internal_value_out_of_range): Print a message about
1593 a value being out of range. Be consistent about whether the
1594 values are printed in decimal or hexadecimal.
1595 (as_warn_value_out_of_range): Generate a warning message about an
1597 (as_bad_value_out_of_range): Generate an error message about an
1599 * as.h: Prototype the new functions.
1600 * config/tc-alpha.c (insert_operand): Use new function.
1601 * config/tc-arc.c (arc_insert_operand): Likewise.
1602 * config/tc-mn10200.c (mn10200_insert_operand): Likewise.
1603 * config/tc-mn10300.c (mn10300_insert_operand): Likewise.
1604 * config/tc-ppc.c (ppc_insert_operand): Likewise.
1605 * config/tc-s390.c (s390_insert_operand): Likewise.
1606 * config/tc-v850.c (v850_insert_operand): Likewise.
1608 2004-05-05 Alexandre Oliva <aoliva@redhat.com>
1610 * configure.in: Set em=linux for frv-*-*linux*.
1611 * configure: Rebuilt.
1612 * config/tc-frv.h (TARGET_FORMAT): Use elf32-frvfdpic if...
1613 (frv_md_fdpic_enabled): New.
1614 * config/tc-frv.c (frv_md_fdpic_enabled): New.
1615 (DEFAULT_FDPIC): New.
1616 (frv_flags): Use DEFAULT_FDPIC.
1617 (frv_pic_flag): Likewise.
1618 (OPTION_NOPIC): New.
1619 (md_longopts): Add -mnopic.
1620 (md_parse_option): Handle it.
1621 (md_show_usage): Add -mfdpic and -mnopic.
1623 2004-05-05 Peter Barada <peter@the-baradas.com>
1625 * config/tc-m68k.c: Add find_cf_chip to print list of valid
1626 chips for invalid coldfire instructions, rename selectors
1627 for ColdFire sub-variants, add 521x,5249,547x,548x and aliases,
1628 add current_chip to track which chip is referred to(including save/restore),
1629 use current_chip to select control registers, not current_arch.
1630 (md_show_usage): Add new chips.
1631 * doc/c-m68k.texi: Document new command line switches.
1633 2004-05-05 Jakub Jelinek <jakub@redhat.com>
1635 * tc-s390.h (md_do_align, HANDLE_ALIGN): Remove.
1636 (NOP_OPCODE): Define.
1637 (s390_align_code): Remove prototype.
1638 * tc-s390.c (s390_align_code): Remove.
1640 2004-05-04 H.J. Lu <hongjiu.lu@intel.com>
1642 * config/tc-ia64.c (make_unw_section_name): Removed.
1643 (start_unwind_section): New function.
1644 (generate_unwind_image): Take const segT instead of const
1646 (dot_handlerdata): Adjusted.
1647 (dot_endp): Likewise.
1649 2004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1651 * config/obj-elf.c (obj_elf_change_section): Allow the
1652 ".note.GNU-stack" section has SHF_EXECINSTR.
1654 2004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1656 * config/obj-elf.c (get_section): Return bfd_boolean.
1657 (obj_elf_change_section): Call bfd_get_section_by_name_if
1658 instead of bfd_map_over_sections.
1660 2004-04-30 H.J. Lu <hongjiu.lu@intel.com>
1662 * config/obj-elf.c (get_section): New function.
1663 (obj_elf_change_section): Support multiple sections with same
1666 2004-04-30 Nick Clifton <nickc@redhat.com>
1668 * config/tc-arm.c (create_register_alias): Fix typo checking for
1669 case sensitive register aliases.
1670 (co_proc_number): Use error message string in all_reg_maps[]
1672 (cp_reg_required_here): Likewise.
1673 (fp_reg_required_here): Likewise.
1675 2004-04-29 Brian Ford <ford@vss.fsi.com>
1677 * dwarf2dbg.c (dwarf2_finish): Add SEC_DEBUGGING to section flags.
1679 2004-04-28 Chris Demetriou <cgd@broadcom.com>
1681 * config/tc-mips.c (HAVE_32BIT_ADDRESSES, append_insn, macro_build)
1682 (load_address, macro, mips_ip, md_parse_option)
1683 (mips_force_relocation, mips_validate_fix, md_apply_fix3)
1684 (s_change_sec, pic_need_relax, tc_gen_reloc): Remove all
1685 embedded-PIC handling, and update comments.
1686 (SWITCH_TABLE): Remove.
1687 * config/tc-mips.h (DIFF_EXPR_OK): Delete.
1688 (enum mips_pic_level): Remove EMBEDDED_PIC.
1689 (EXTERN_FORCE_RELOC): Remove embedded-PIC handling.
1690 (TC_FORCE_RELOCATION): Update comment.
1691 * ecoff.c (ecoff_build_lineno): Add comment about some code that
1692 might be safe to remove now that MIPS embedded-PIC is gone.
1694 2004-04-28 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1696 * config/obj-som.c (obj_som_init_stab_section): Add new arguments in
1697 call to obj_set_subsection_attributes.
1698 (obj_som_init_stab_section): Likewise.
1699 * config/tc-hppa.c (default_subspace_dict): Add comdat field.
1700 (pa_def_subspaces): Provide comdat default.
1701 (pa_subspace): Handle new "comdat" parameter. Set SEC_LINK_ONCE and
1702 not SEC_IS_COMMON if section is comdat, common or dup_common. Update
1703 calls to create_new_subspace and update_subspace to pass comdat flag.
1704 (create_new_subspace, update_subspace): Add new comdat argument. Use
1705 it in calls to obj_set_subsection_attributes.
1706 * doc/c-hppa.texi (.subspa, .nsubspa): Document new comdat parameter
1707 and use of comdat, common and dup_comm parameters.
1709 2004-04-26 H.J. Lu <hongjiu.lu@intel.com>
1711 * config/obj-elf.c (obj_elf_change_section): Check if the old
1712 group name is NULL before comparison.
1714 2004-04-23 Chris Demetriou <cgd@broadcom.com>
1716 * config/tc-mips.h (mips_dwarf2_addr_size): Prototype.
1718 2004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1720 * config/tc-mips.c (s_mipsset): Set default CPU type for .set mipsN.
1722 2004-04-23 Chris Demetriou <cgd@broadcom.com>
1724 * config/tc-mips.c (md_longopts): Remove -membedded-pic option.
1725 (OPTION_MEMBEDDED_PIC): Remove.
1726 (OPTION_TRAP, OPTION_BREAK, OPTION_EB, OPTION_EL)
1727 (OPTION_FP32, OPTION_GP32, OPTION_CONSTRUCT_FLOATS)
1728 (OPTION_NO_CONSTRUCT_FLOATS, OPTIONS_FP64, OPTION_GP64)
1729 (OPTION_RELAX_BRANCH, OPTION_NO_RELAX_BRANCH)
1730 (OPTION_ELF_BASE): Renumber.
1731 (md_parse_option): Remove OPTION_MEMBEDDED_PIC handling.
1732 (md_show_usage): Remove mention of -membedded-pic.
1733 * doc/as.texinfo: Remove mention of -membedded-pic.
1735 2004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1737 * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Remove.
1738 * config/tc-mips.c (RDATA_SECTION_NAME, mips_target_format): Remove
1740 (md_begin, mips_ip, md_parse_option, s_change_sec, s_option,
1741 s_abicalls, nopic_need_relax, tc_gen_reloc): Remove uses of
1742 USE_GLOBAL_POINTER_OPT.
1744 2004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1746 * config/tc-mips.c (macro): One more use of load_delay_nop.
1748 2004-04-22 Atsushi Nemoto <anemo@mba.ocn.ne.jp>
1750 * config/tc-mips.c (load_delay_nop): New function.
1751 (load_address, macro): Use load_delay_nop() to build a nop
1752 which can be omitted with gpr_interlocks.
1754 2004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1756 * config/tc-mips.c (hilo_interlocks, gpr_interlocks,
1757 cop_interlocks): Remove superfluous CPU entries.
1759 2004-04-22 Paul Brook <paul@codesourcery.com>
1761 * config/tc-arm.c (mav_parse_offset): Value must be multiple of 4.
1763 2004-04-22 Peter Barada <peter@the-baradas.com>
1765 * NEWS: Added support for EMAC instructions and MAC/EMAC
1767 * config/m68k-parse.h: Add ACC[123], ACCEXT{01,23}, MAC/EMAC
1768 scale factor tokens, trailing_ampersand to mark mask addressing
1769 for MAC/EMAC instructions.
1770 * config/m68k-parse.y: Add options_ampersand clause, '<<',
1772 (yylex): Handle '>', '<', and '&' following '+'.
1773 * config/tc-m68k.c: Set mcfmac/mcfemac on appropriate ColdFire
1774 architectures in archs[].
1775 (m68k-ip): Add '4', 'e', 'g', 'i', cases to handle mask addressing
1776 for MAC/EMAC instructions, ACC[0123], ACCEXT{01,23}, and '<<'/'>>'
1778 (m68k_ip): Handle trailing '&' on MAC/EMAC insns.
1779 (install_operand): Fix 'n' case, Add 'F', 'f', 'G', 'H', 'I', ']'
1781 Add EMAC operands to init_table[].
1783 2004-04-22 Bruno De Bus <bdebus@elis.ugent.be>
1785 * config/tc-arm.h (enum mstate): Move here, add MAP_UNDEFINED
1787 (TC_SEGMENT_INFO_TYPE): Define to enum mstate.
1788 * config/tc-arm.c (enum mstate): Delete from here.
1789 (mapping_state): Remove the static mapstate variable and instead
1790 store the state in the segment. This allows a per-section mapping
1791 state. Handle and ignore MAP_UNDEFINED states.
1792 (arm_elf_change_section): Get the current mapping state from the
1794 (s_ltorg): Set the mapping state to MAP_DATA.
1795 (arm_cleanup): Use arm_elf_change_section to get the mapping state
1796 for each pool as it is emitted.
1798 2004-04-22 Nick Clifton <nickc@redhat.com>
1800 * config/tc-arm.h: Formatting tidy ups.
1802 2004-04-20 Chris Demetriou <cgd@broadcom.com>
1804 * NEWS: Note that MIPS -membedded-pic option is deprecated.
1806 2004-04-20 DJ Delorie <dj@redhat.com>
1808 * config/tc-i386.h [TE_PE] (TC_CONS_FIX_NEW): Define.
1809 * config/tc-i386.c (md_pseudo_table) [TE_PE]: Add "secrel32".
1810 [TE_PE] (O_secrel): Define.
1811 [TE_PE] (x86_pe_cons_fix_new): New.
1812 [TE_PE] (pe_directive_secrel): Likewise.
1813 (tc_gen_reloc) [TE_PE]: Support BFD_RELOC_32_SECREL.
1815 2004-04-19 Eric Christopher <echristo@redhat.com>
1817 * config/tc-mips.c (mips_dwarf2_addr_size): Revert part
1818 of previous patch for fix in gcc.
1820 2004-04-19 Jakub Jelinek <jakub@redhat.com>
1822 * config/tc-xtensa.c (xg_assembler_literal): Fix a typo.
1824 2004-04-19 Nathan Sidwell <nathan@codesourcery.com>
1826 * read.c (do_align): Call md_flush_pending_output, if defined.
1828 2004-04-16 Alan Modra <amodra@bigpond.net.au>
1830 * expr.c (operand): Correct checks for ++ and --.
1832 2004-04-14 H.J. Lu <hongjiu.lu@intel.com>
1834 * config/tc-generic.c: Add some comments.
1836 2004-04-14 Richard Sandiford <rsandifo@redhat.com>
1838 * doc/c-mips.texi (-m{no-,}fix-vr4120): Renamed from
1839 -{no-}mfix-vr4122-bugs.
1840 * config/tc-mips.c (mips_fix_vr4120): Renamed from mips_fix_4122_bugs.
1841 (append_insn, mips_emit_delays): Update accordingly.
1842 (OPTION_FIX_VR4120, OPTION_NO_FIX_VR4120): Renamed from *VR4122.
1843 (md_longopts): Change -{no-,}mfix-vr4122-bugs to -m{no-,}fix-vr4120.
1844 (md_parse_option): Update after above changes.
1845 (md_show_usage): Add -mfix-vr4120.
1847 2004-04-13 Bob Wilson <bob.wilson@acm.org>
1849 * doc/as.texinfo (Sub-Sections): Conditionalize COFF-specific use
1850 of .section directive; add a reference to the ELF .subsection
1853 2004-04-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1855 * config/tc-m32r.c (md_assemble): Fixed infinite loop bug
1858 2004-04-11 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1860 * Makefile.am: Remove mips from aout targets.
1861 * Makefile.in: Regenerate.
1862 * configure.in: Remove mips-dec-bsd* target.
1863 * configure: Regenerate.
1865 2004-04-07 Alan Modra <amodra@bigpond.net.au>
1868 * config/tc-ppc.c (ppc_elf_suffix): Add valid32 and valid64 fields
1869 to struct map_bfd. Adjust MAP macro, and define MAP32, MAP64.
1870 Update "mapping". Restrict some @ modifiers to 32 bit.
1872 2004-04-01 Asgari Jinia <asgarij@kpitcummins.com>
1873 Dhananjay Deshpande <dhananjayd@kpitcummins.com>
1875 * config/tc-sh.c (dont_adjust_reloc_32): New variable.
1876 (sh_fix_adjustable): Avoid adjusting BFD_RELOC_32 when
1877 dont_adjust_reloc_32 is set.
1878 (md_longopts): Add option -renesas.
1879 (md_parse_option, md_show_usage): Likewise.
1880 * doc/c-sh.texi: Likewise.
1882 2004-04-01 Dave Korn <dk@artimi.com>
1884 * config/tc-dlx.c (md_assemble): set fx_no_overflow flag for
1885 hi16 and lo16 fixS structs.
1886 (md_assemble): generate bit_fixS for RELOC_DLX_LO16 in
1887 exactly the same way as for RELOC_DLX_REL16.
1888 (machine_ip): properly respect LO flag in the_insn and
1889 output RELOC_DLX_LO16 rather than RELOC_DLX_16.
1890 (md_apply_fix3): apply RELOC_DLX_LO16.
1892 2004-03-30 Stan Shebs <shebs@apple.com>
1894 Remove long-obsolete MPW support.
1895 * mpw-config.in, mpw-make.sed, mac-as.r: Remove files.
1896 * configure.in: Remove mention of ppc-*-mpw* config.
1897 * configure.in: Likewise.
1899 2004-03-30 Nick Clifton <nickc@redhat.com>
1901 * config/tc-arm.c (meabi_flags): Make its use conditional upon
1902 OBJ_ELF being defined.
1904 2004-03-27 Alan Modra <amodra@bigpond.net.au>
1906 * config/obj-aout.c (obj_aout_type): Remove #ifdef BFD_ASSEMBLER code.
1908 2004-03-23 Paul Brook <paul@codesourcery.com>
1910 * config/tc-arm.c (meabi_flags): New variable.
1911 (arm_parse_eabi): New function.
1912 (md_begin): Set flags for EABI v3.
1914 (arm_long_opts): Add meabi.
1915 * doc/as.texinf <ARM>: Document -meabi.
1916 * doc/c-arm.texi: Ditto.
1918 2004-03-22 Bob Wilson <bob.wilson@acm.org>
1920 * config/tc-xtensa.c (xtensa_post_relax_hook): Create literal
1921 tables even when use_literal_section flag is not set.
1923 2004-03-22 Alan Modra <amodra@bigpond.net.au>
1925 * config/tc-sh.c: Remove trailing whitespace.
1927 2004-03-22 Hans-Peter Nilsson <hp@axis.com>
1929 * doc/c-cris.texi (CRIS-Opts): Document --no-mul-bug-abort,
1930 --mul-bug-abort and the default behavior.
1931 * config/tc-cris.c (cris_insn_kind): New member CRIS_INSN_MUL.
1932 (err_for_dangerous_mul_placement): New variable.
1933 (STATE_MUL, OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): New
1935 (md_cris_relax_table): Have placeholder for STATE_MUL.
1936 (md_longopts): New options --mul-bug-abort and --no-mul-bug-abort.
1937 (cris_relax_frag) <case ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: New
1939 (md_estimate_size_before_relax) <case ENCODE_RELAX (STATE_MUL,
1940 STATE_BYTE)>: Ditto.
1941 (md_convert_frag) <ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: Check
1942 alignment and position of this frag, emit error message if
1944 (md_assemble): For a multiply insn and when checking it,
1945 transform the current frag into a special frag for that purpose.
1946 (md_parse_option) <case OPTION_MULBUG_ABORT_OFF, case
1947 OPTION_MULBUG_ABORT_ON>: Handle new options.
1949 2004-03-19 Bob Wilson <bob.wilson@acm.org>
1951 * config/tc-xtensa.c (mark_literal_frags): New function.
1952 (xtensa_move_literals): Call mark_literal_frags for all literal
1953 segments, including init and fini literal segments.
1954 (xtensa_post_relax_hook): Swap use of xt_insn_sec and xt_literal_sec.
1956 2004-03-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1958 * tc-hppa.c (cons_fix_new_hppa): Check for PC relative base type.
1959 (pa_comm): Set BSF_OBJECT in symbol flags.
1961 2004-03-19 Alan Modra <amodra@bigpond.net.au>
1963 * Makefile.am: Run "make dep-am".
1964 * Makefile.in: Regenerate.
1965 * doc/Makefile.in: Regenerate.
1966 * config.in: Regenerate.
1967 * po/gas.pot: Regenerate.
1969 2004-03-18 Nathan Sidwell <nathan@codesourcery.com>
1971 * read.c (read_a_source_file): Use demand_empty_rest_of_line.
1972 (demand_empty_rest_of_line): Issue an error here.
1973 (ignore_rest_of_line): Silently skip to end.
1974 (demand_copy_string): Issue an error, not warning.
1976 * config/obj-elf.c (obj_elf_section_name): Likewise.
1977 (obj_elf_section): Likewise.
1978 * config/tc-arc.c (arc_extoper): Remove bogus NULL checks.
1979 (arc_extinst): Likewise.
1980 * config/tc-ia64.c (dot_saveb): Use demand_empty_rest_of_line.
1981 (dot_spill): Likewise.
1982 (dot_unwabi): Likewise.
1983 (dot_prologue): Likewise.
1985 * expr.c (operand): Reject ++ and --.
1986 (operator): Likewise.
1988 2004-03-17 Kaz Kojima <kkojima@rr.iij4u.or.jp>
1990 * config/tc-sh.c: Include dw2gencfi.h.
1991 (sh_cfi_frame_initial_instructions): New function.
1992 (sh_regname_to_dw2regnum): Likewise.
1993 * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Move to the end of
1995 (TARGET_USE_CFIPOP): Define.
1996 (tc_cfi_frame_initial_instructions): Likewise.
1997 (tc_regname_to_dw2regnum): Likewise.
1998 (DWARF2_DEFAULT_RETURN_COLUMN, DWARF2_CIE_DATA_ALIGNMENT): Likewise.
1999 * Makefile.am: Update dependencies.
2000 * Makefile.in: Regenerate.
2002 2004-03-17 Ralf Corsepius <corsepiu@faw.uni-ulm.de>
2004 * configure.in: Switch sh-*-rtems* to ELF. Add sh-*-rtemscoff*.
2005 * configure: Regenerate.
2007 2004-03-12 Bob Wilson <bob.wilson@acm.org>
2009 * read.c (s_leb128): Call md_flush_pending_output.
2011 2004-03-12 Michal Ludvig <mludvig@suse.cz>
2013 * config/tc-i386.c (output_insn): Handle PadLock instructions.
2014 * config/tc-i386.h (CpuPadLock): New define.
2015 (CpuUnknownFlags): Added CpuPadLock.
2017 2004-03-07 Andreas Schwab <schwab@suse.de>
2019 * doc/c-hppa.texi (HPPA Directives): Fix typo.
2021 2004-03-07 Richard Henderson <rth@redhat.com>
2023 * dw2gencfi.c (output_cie): Align length to 4 byte boundary.
2024 (cfi_finish): Likewise for fde.
2026 2004-03-05 H.J. Lu <hongjiu.lu@intel.com>
2028 * config/tc-ia64.c (md_assemble): Properly handle NULL
2030 (ia64_handle_align): Don't abort if failed to add a stop bit.
2032 2004-03-04 H.J. Lu <hongjiu.lu@intel.com>
2034 * Makefile.in: Regenerated.
2035 * aclocal.m4: Likewise.
2036 * configure: Likewise.
2037 * doc/Makefile.in: Likewise.
2039 2004-03-03 H.J. Lu <hongjiu.lu@intel.com>
2041 * config/tc-ia64.c (dot_align): New.
2042 (ia64_do_align): Make it static.
2043 (md_pseudo_table): Use "dot_align" for "align".
2044 (ia64_md_do_align): Don't set align_frag here.
2045 (ia64_handle_align): Add a stop bit to the previous bundle if
2048 * config/tc-ia64.h (ia64_do_align): Removed.
2050 2003-03-03 Andrew Stubbs <andrew.stubbs@superh.com>
2052 * config/tc-sh.c (md_parse_option): Add -isa=sh4-nofpu and
2053 -isa=sh4-nommu-nofpu options. Adjust help messages accordingly.
2054 (sh_elf_final_processing): Output BFD type sh4_nofpu if that is
2055 the most general type or the user specifically requested it.
2056 (md_assemble): Add a new error message for when an instruction
2057 is understood, but is not allowed due to an -isa option.
2059 2004-03-02 H.J. Lu <hongjiu.lu@intel.com>
2061 * config/tc-ia64.c (align_frag): New.
2062 (md_assemble): Set the tc_frag_data field in align_frag for
2063 IA64_OPCODE_FIRST instructions.
2064 (ia64_md_do_align): Set align_frag.
2065 (ia64_handle_align): Add a stop bit if needed.
2067 * config/tc-ia64.h (TC_FRAG_TYPE): New.
2068 (TC_FRAG_INIT): New.
2070 2004-03-01 Richard Sandiford <rsandifo@redhat.com>
2072 * config/tc-frv.c (fr400_audio): New variable.
2073 (md_parse_option, md_show_usage): Add -mcpu=fr405 and -mcpu=fr450.
2074 (md_parse_option): Set fr400_audio for -mcpu=fr400 and -mcpu=fr405.
2075 (target_implements_insn_p): New function.
2076 (md_assemble): Report an error if the processor doesn't implement
2079 2004-02-27 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2081 * config/tc-m32r.c (md_longopts): Added -no-bitinst option.
2082 (md_parse_option): Ditto.
2083 (OPTION_NO_SPECIAL_M32R): Added.
2084 (md_show_usage): Document it.
2085 (enable_speial_m32r): Changed a default value from 0 to 1.
2086 * doc/c-m32r.texi: Document the -no-bitinst option.
2088 2004-02-27 Nick Clifton <nickc@redhat.com>
2090 * config/tc-sh.c (get_operand): Revert previous delta.
2091 (tc_gen_reloc): Check for an unknown reloc type before processing
2094 2004-02-27 Hannes Reinecke <hare@suse.de>
2096 * config/tc-s390.c (s390_insn): Correct range check for opcode in
2097 .insn pseudo operation.
2099 2004-02-27 Anil Paranjpe <anilp1@kpitcummins.com>
2101 * config/tc-sh.c (get_operand): In case of #Imm, check has been
2102 added for wrong syntax.
2104 2004-02-26 Eric Christopher <echristo@redhat.com>
2106 * config/tc-mips.c (mips_dwarf2_addr_size): New.
2107 * config/tc-mips.h (DWARF2_ADDR_SIZE): Use.
2109 2004-02-26 Andrew Stubbs <andrew.stubbs@superh.com>
2111 * config/tc-sh.c (build_Mytes): Add REG_N_D and REG_N_B01
2112 nibble types to assembler.
2114 2004-02-25 Fred Fish <fnf@redhat.com>
2116 * config/tc-iq2000.c: Add missing \n\ in multiline string literal.
2118 2004-02-20 James E Wilson <wilson@specifixinc.com>
2120 * config/tc-ia64.c (slot_index): New arg before_relax. Use instead of
2122 (fixup_unw_records): New arg before_relax. Pass to slot_index.
2123 (ia64_estimate_size_before_relax): New.
2124 (ia64_convert_frag): Pass 0 to fixup_unw_records. Add comment.
2125 (generate_unwind_image): Pass 1 to fixup_unw_records.
2126 * config/tc-ia64.h (ia64_estimate_size_before_relax): Declare.
2127 (md_estimate_size_before_relax): Call ia64_estimate_size_before_relax.
2129 2004-02-19 Jakub Jelinek <jakub@redhat.com>
2131 * stabs.c (generate_asm_file): Avoid warning about use of
2132 uninitialized variable.
2134 2004-02-18 David Mosberger <davidm@hpl.hp.com>
2136 * config/tc-ia64.c (ia64_flush_insns): In addition to prologue,
2137 body, and endp, allow unwind records which do not have a "t"
2138 (time/instruction) field.
2140 2004-02-17 Petko Manolov <petkan@nucleusys.com>
2142 * config/tc-arm.c (do_mav_dspsc_1): Correct offset of CRn.
2143 (do_mav_dspsc_2): Likewise.
2144 Fix accumulator registers move opcodes.
2146 2004-02-13 Hannes Reinecke <hare@suse.de>
2147 Jakub Jelinek <jakub@redhat.com>
2149 * dwarf2dbg.c (get_filenum): Do not read beyond allocated memory.
2151 2004-02-10 Steve Ellcey <sje@cup.hp.com>
2153 * config/tc-ia64.h (ia64_frob_symbol): New declaration.
2154 (tc_frob_symbol): New macro definition.
2155 * config/tc-ia64.c (ia64_frob_symbol): New routine.
2157 2004-02-09 Daniel Jacobowitz <drow@mvista.com>
2159 * config/tc-arm.c (md_begin): Mark .note.gnu.arm.ident as
2162 2004-02-09 Nathan Sidwell <nathan@codesourcery.com>
2164 * read.h (IGNORE_OPCODE_CASE): Do not define. Replace with ...
2165 (TC_CASE_SENSITIVE): ... this.
2166 * read.c: Replace IGNORE_OPCODE_CASE with TC_CASE_SENSITIVE.
2167 * doc/internals.texi (TC_CASE_SENSITIVE): Document.
2169 2004-02-06 James E Wilson <wilson@specifixinc.com>
2171 * config/tc-ia64.c (dot_endp): Delete call to output_endp.
2172 (generate_unwind_image): Re-add it here.
2174 2004-02-06 Nathan Sidwell <nathan@codesourcery.com>
2176 * dwarf2dbg.c (DWARF2_ADDR_SIZE): Remove trailing ';'
2177 * read.h (SKIP_WHITESPACE): Turn into an expression.
2178 * read.c (read_a_source_file): A pseudo is removed by having a
2181 2004-02-05 James E Wilson <wilson@specifixinc.com>
2183 * config/tc-ia64.c (output_endp): New.
2184 (count_bits): Delete.
2185 (ia64_flush_insns, process_one_record, optimize_unw_records): Handle
2186 endp unwind records.
2187 (fixup_unw_records): Handle endp unwind records. Delete code for
2188 shortening prologue regions not followed by a body record.
2189 (dot_endp): Call add_unwind_entry to emit endp unwind record.
2190 * config/tc-ia64.h (unw_record_type): Add endp.
2192 2004-02-03 James E Wilson <wilson@specifixinc.com>
2194 * config/tc-ia64.c (ia64_convert_frag): Call md_number_to_chars to
2195 fill padding bytes with zeroes.
2196 (emit_one_bundle): New locals last_ptr, end_ptr. Rewrite code that
2197 sets unwind_record slot_number and slot_frag fields.
2199 2004-02-02 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2201 * config/tc-mips.c (add_got_offset_hilo): New function.
2202 (macro): Use load_register() and add_got_offset_hilo() to load
2203 constants instead of hardcoding code sequences throughout.
2205 2004-01-28 H.J. Lu <hongjiu.lu@intel.com>
2207 * config/tc-ia64.c (emit_one_bundle): Add proper indentation.
2209 2004-01-26 Bernardo Innocenti <bernie@develer.com>
2211 * config/tc-m68k.h (EXTERN_FORCE_RELOC): Handle m68k-uclinux specially,
2213 * config/tc-m68k.c (RELAXABLE_SYMBOL): Use EXTERN_FORCE_RELOC instead
2214 of hard-coded test for TARGET_OS=elf.
2216 2004-01-24 Chris Demetriou <cgd@broadcom.com>
2218 * config/tc-mips.c (hilo_interlocks): Change definition
2219 so that MIPS32, MIPS64 and later ISAs are included, along with
2220 the already-included machines. Update comments.
2222 2004-01-23 Daniel Jacobowitz <drow@mvista.com>
2224 * config/tc-arm.c (tc_gen_reloc): Improve error message for
2225 undefined local labels.
2227 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2229 * config/tc-mips.c (load_address, macro): Update comments about
2230 NewABI GP relaxation.
2232 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2234 * config/tc-mips.c (macro_build): Remove place and counter arguments.
2235 (mips_build_lui, macro_build_ldst_constoffset): Likewise.
2236 (mips16_macro_build, macro_build_jalr): Remove counter argument.
2237 (set_at, load_register, load_address, move_register): Likewise.
2238 (load_got_offset, add_got_offset): Likewise.
2239 Update all calls and tidy accordingly.
2241 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2243 * config/tc-mips.c (RELAX_ENCODE): Remove WARN argument.
2244 (RELAX_FIRST, RELAX_SECOND): Turn into 8-bit quantities.
2245 (RELAX_USE_SECOND): Bump to 0x10000.
2246 (RELAX_SECOND_LONGER, RELAX_NOMACRO, RELAX_DELAY_SLOT): New flags.
2247 (mips_macro_warning): New variable.
2248 (md_assemble): Wrap macro expansion in macro_start() and macro_end().
2249 (s_cpload, s_cpsetup, s_cprestore, s_cpreturn): Likewise.
2250 (relax_close_frag): Set mips_macro_warning.first_frag. Adjust use
2252 (append_insn): Update mips_macro_warning.sizes.
2253 (macro_start, macro_warning, macro_end): New functions.
2254 (macro_build): Don't emit warnings here.
2255 (macro_build_lui, md_estimate_size_before_relax): ...or here.
2256 (md_convert_frag): Check for cases where one macro alternative
2257 needs a warning and the other doesn't. Emit a warning if the
2258 longer sequence was chosen.
2260 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2262 * config/tc-mips.h (tc_frag_data_type, TC_FRAG_TYPE): Remove.
2263 * config/tc-mips.c (RELAX_ENCODE): Take three arguments: the size of
2264 the first sequence, the size of the second sequence, and a flag
2265 that says whether we should warn.
2266 (RELAX_OLD, RELAX_NEW, RELAX_RELOC[123]): Delete.
2267 (RELAX_FIRST, RELAX_SECOND): New.
2268 (mips_relax): New variable.
2269 (relax_close_frag, relax_start, relax_switch, relax_end): New fns.
2270 (append_insn): Remove "place" argument. Use mips_relax.sequence
2271 rather than "place" to check whether we're expanding the second
2272 alternative of a relaxable macro. Remove redundant check for
2273 branch relaxation. If generating a normal insn, and there
2274 is not enough room in the current frag, call relax_close_frag()
2275 to close it. Update mips_relax.sizes[]. Emit fixups for the
2276 second version of a relaxable macro. Record the first relaxable
2277 fixup in mips_relax. Remove tc_gen_reloc workaround.
2278 (macro_build): Remove all uses of "place". Use mips_relax.sequence
2279 in the same way as in append_insn.
2280 (mips16_macro_build): Remove "place" argument.
2281 (macro_build_lui): As for macro_build. Don't drop the add_symbol
2282 when generating the second version of a relaxable macro.
2283 (load_got_offset, add_got_offset): New functions.
2284 (load_address, macro): Use new relaxation machinery. Remove
2285 tc_gen_reloc workarounds.
2286 (md_estimate_size_before_relax): Set RELAX_USE_SECOND if the second
2287 version of a relaxable macro is needed. Return -RELAX_SECOND if the
2288 first version is needed.
2289 (tc_gen_reloc): Remove relaxation handling.
2290 (md_convert_frag): Go through the fixups for a relaxable macro and
2291 mark those that belong to the unneeded alternative as done. If the
2292 second alternative is needed, adjust the fixup addresses to account
2293 for the deleted first alternative.
2295 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2297 * frags.h (frag_room): Declare.
2298 * frags.c (frag_room): New function.
2299 * doc/internals.texi: Document it.
2301 2004-01-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
2303 * config/tc-mips.c (append_insn): Don't do r3900 interlock
2304 optimization for -mtune=r3900, as this will break on other CPUs.
2306 2004-01-11 Tom Rix <tcrix@worldnet.att.net>
2308 * config/tc-m68hc11.c (build_indexed_byte): movb and movw cannot
2309 be relaxed, use fixup.
2310 (md_apply_fix3): Use 5 bit reloc from movb and movw fixup.
2312 2004-01-19 Jakub Jelinek <jakub@redhat.com>
2314 * config/tc-sparc.c (sparc_ip): Disallow %f32-%f63 for single
2317 2004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2319 * config/tc-mips.c (append_insn): Properly detect variant frags
2320 that preclude swapping of relaxed branches. Correctly swap
2321 instructions between frags when dealing with relaxed branches.
2323 2004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2325 * acinclude.m4: Quote names of macros to be defined by AC_DEFUN
2327 * aclocal.m4: Regenerate.
2328 * configure: Regenerate.
2330 2004-01-12 Anil Paranjpe <anilp1@KPITCummins.com>
2332 * config/tc-h8300.c (build_bytes): Apply relaxation to bit
2335 2004-01-12 Richard Sandiford <rsandifo@redhat.com>
2337 * config/tc-mips.c (macro_build_jalr): When adding an R_MIPS_JALR
2338 reloc, reserve space for the delay slot as well as the jalr itself.
2340 2004-01-09 Paul Brook <paul@codesourcery.com>
2342 * config/tc-arm.c (do_vfp_reg2_from_sp2): Rename from do_vfp_sp_reg2.
2343 (do_vfp_sp2_from_reg2): New function.
2345 (do_vfp_dp_from_reg2): Check return values properly.
2347 2004-01-08 Ian Lance Taylor <ian@wasabisystems.com>
2349 * config/tc-mips.c (warn_nops): Remove static variable.
2350 (macro): Remove test of warn_nops.
2351 (md_shortops): Remove 'n'.
2352 (md_parse_option): Remove 'n' case.
2353 (md_show_usage): Remove -n.
2354 * doc/as.texinfo (Overview): Remove MIPS -n option.
2355 * doc/c-mips.texi (MIPS Opts): Remove mention -n.
2356 * NEWS: Mention removal of MIPS -n option.
2358 * config/tc-mips.c (ISA_HAS_COPROC_DELAYS): Remove.
2359 (cop_interlocks): Check ISA level.
2360 (cop_mem_interlocks): Define.
2361 (reg_needs_delay): Check cop_interlocks rather than
2362 ISA_HAS_COPROC_DELAYS.
2363 (append_insn): Likewise. Use cop_mem_interlocks rather than
2364 directly checking mips_opts.isa.
2365 (mips_emit_delays): Likewise.
2367 2004-01-07 H.J. Lu <hongjiu.lu@intel.com>
2369 * config/tc-ia64.c (unwind): Move next_slot_number and
2370 next_slot_frag to ...
2371 (unw_rec_list): Here.
2372 (free_list_records): Removed.
2373 (output_unw_records): Likewise.
2374 (generate_unwind_image): Make it void.
2375 (alloc_record): Initialize next_slot_number and next_slot_frag.
2376 (slot_index): Take .org, .space and .align into account.
2377 (fixup_unw_records): Don't set slot_number to 0. Use
2378 list->next_slot_number and list->next_slot_frag instead of
2379 unwind.next_slot_number and unwind.next_slot_frag.
2380 (ia64_convert_frag): New.
2381 (generate_unwind_image): Generate a rs_machine_dependent frag
2383 (emit_one_bundle): Use list->next_slot_number and
2384 list->next_slot_frag instead of unwind.next_slot_number and
2385 unwind.next_slot_frag.
2387 * config/tc-ia64.h (md_convert_frag): Defined as
2389 (md_estimate_size_before_relax): Defined as (f)->fr_var.
2391 2004-01-06 Alexandre Oliva <aoliva@redhat.com>
2393 2003-12-19 Alexandre Oliva <aoliva@redhat.com>
2394 * config/tc-frv.h (md_apply_fix3): Don't define.
2395 * config/tc-frv.c (md_apply_fix3): New. Shift/truncate %hi/%lo
2397 * config/tc-frv.h (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
2398 2003-10-07 Alexandre Oliva <aoliva@redhat.com>
2399 * config/tc-frv.c (line_separator_chars): Add `!'.
2400 2003-09-19 Alexandre Oliva <aoliva@redhat.com>
2401 * config/tc-frv.c (md_assemble): Clear insn upfront.
2402 2003-09-18 Alexandre Oliva <aoliva@redhat.com>
2403 * config/tc-frv.c (OPTION_FDPIC): New macro.
2404 (md_longopts): Add mfdpic.
2405 (md_parse_option): Handle it.
2406 2003-08-04 Alexandre Oliva <aoliva@redhat.com>
2407 * config/tc-frv.c (md_cgen_lookup_reloc) <FRV_OPERAND_D12,
2408 FRV_OPERAND_S12>: Use reloc type encoded in fix-up.
2409 (frv_pic_ptr): Parse funcdesc.
2411 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2413 * doc/as.texinfo: Let texi2pod parse asconfig.texi and
2414 gasver.texi. Remove duplicate symbol definitions for texi2pod.
2416 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2418 * Makefile.am (Makefile): Move the dependency on
2419 $(BFDDIR)/configure.in to...
2420 (CONFIG_STATUS_DEPENDENCIES): ... here.
2421 (AUTOMAKE_OPTIONS): Require automake 1.8.
2422 * Makefile.in: Regenerate.
2423 * doc/Makefile.am (BASEDIR, BFDDIR): Define.
2424 (CONFIG_STATUS_DEPENDENCIES): Add a dependency on
2425 $(BFDDIR)/configure.in here as well.
2426 * doc/Makefile.in: Regenerate.
2428 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2430 * Makefile.am (install, install-info, RECURSIVE_TARGETS): Remove.
2431 * Makefile.in: Regenerate.
2432 * aclocal.m4: Regenerate.
2433 * doc/Makefile.am (install, install-info): Remove.
2434 (install-data-local): A new hook for install-info.
2435 (AUTOMAKE_OPTIONS): Require automake 1.8.
2436 * doc/Makefile.in: Regenerate.
2438 2004-01-02 Nutan Singh <nutan@kpitcummins.com>
2440 * doc/c-sh.texi: Update description about floating point behavior
2443 2004-01-02 Bernardo Innocenti <bernie@develer.com>
2445 * configure.in: Add m68k-uClinux target.
2446 * configure: Regenerate.
2448 For older changes see ChangeLog-0203
2454 version-control: never