1 2004-10-27 Tomer Levi <Tomer.Levi@nsc.com>
3 * config/tc-crx.c: Remove global variable 'post_inc_mode'.
4 (get_flags): New function.
5 (get_number_of_bits): Edit comments, update numeric values to supported
7 (process_label_constant): Don't support the colon format (SYMBOL:[s|m|l]).
8 (set_cons_rparams): Support argument type 'arg_rbase'.
9 (get_operandtype): Bug fix in 'rbase' operand type parsing.
10 (handle_LoadStor): Bug fix, first handle post-increment mode.
11 (getreg_image): Remove redundant code, update according to latest CRX spec.
12 (print_constant): Bug fix relate to 3-word instructions.
13 (assemble_insn): Bug fix, when matching instructions, verify also
14 instruction type (not only mnemonic).
15 Add various error checking.
16 (preprocess_reglist): Support HI/LO and user registers.
18 2004-10-25 David Mosberger-Tang <davidm@hpl.hp.com>
20 * config/tc-ia64.c (fixup_unw_records): Don't let the "t" value
21 in an epilogue directive go negative.
23 2004-10-25 H.J. Lu <hongjiu.lu@intel.com>
26 * config/tc-ia64.c (emit_one_bundle): Decrement md.num_slots_in_use
27 after reporting template error during manual bundling. Reported
28 by Michael Dupont, michaelx.dupont@intel.com.
30 2004-10-25 Daniel Jacobowitz <dan@debian.org>
32 * Makefile.am: Run dep-am.
33 * aclocal.m4: Regenerate with automake 1.9.2.
34 * Makefile.in: Regenerate with automake 1.9.2.
35 * doc/Makefile.in: Likewise.
37 * config/tc-arm.c: Include "dw2gencfi.h".
38 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
40 * config/tc-arm.h (TARGET_USE_CFIPOP, DWARF2_DEFAULT_RETURN_COLUMN)
41 (DWARF2_CIE_DATA_ALIGNMENT, tc_regname_to_dw2regnum)
42 (tc_cfi_frame_initial_instructions): Define.
43 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
46 2004-10-21 Tomer Levi <Tomer.Levi@nsc.com>
48 * config/tc-crx.c (assemble_insn): Check unsigned immediate
50 Update coprocessor id to be unsigned immediate.
52 2004-10-18 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
54 * config/tc-i386.c (O_secrel): Delete.
55 (tc_pe_dwarf2_emit_offset): New function.
56 * config/tc-i386.h (O_secrel): Define as O_md1.
57 (TC_DWARF2_EMIT_OFFSET): Define.
59 2004-10-18 Nick Clifton <nickc@redhat.com>
61 * config/tc-xstormy16.c (xstormy16_cons_fix_new): Accept and
62 ignore @fptr() directives for 4-byte fixups.
64 2004-10-15 Alan Modra <amodra@bigpond.net.au>
66 * Makefile.am: Run "make dep-am"
67 * Makefile.in: Regenerate.
69 2004-10-14 Bob Wilson <bob.wilson@acm.org>
71 * doc/c-xtensa.texi (Xtensa Options, Absolute Literals Directive):
72 Remove comments about placement of literal pools.
73 (Literal Directive): Update description of literal placement.
74 (Literal Prefix Directive): Remove statement that this does not apply
75 to absolute-mode literals. Describe new section naming scheme.
77 2004-10-12 Bob Wilson <bob.wilson@acm.org>
79 * config/tc-xtensa.c (xtensa_frequency_pseudo): Use set_subseg_freq.
80 (is_entry_opcode, is_movi_opcode, is_the_loop_opcode, is_jx_opcode,
81 is_windowed_return_opcode): Delete.
82 (xtensa_frob_label): Use get_subseg_target_freq.
83 (md_assemble): Inline call to is_entry_opcode.
84 (xtensa_handle_align): Inline call to get_frag_is_literal.
85 (relaxation_requirements): Inline call to is_jx_opcode.
86 (emit_single_op): Inline call to is_movi_opcode.
87 (xg_assemble_vliw_tokens): Inline calls to get_frag_is_insn,
88 get_frag_is_no_transform, is_entry_opcode, and
89 set_frag_is_specific_opcode. Use get_subseg_total_freq.
90 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags,
91 xtensa_fix_close_loop_end_frags, relax_frag_immed, convert_frag_immed):
92 Inline calls to get_frag_is_no_transform.
93 (next_instrs_are_b_retw): Inline call to is_windowed_return_opcode.
94 (xtensa_fix_short_loop_frags): Inline calls to is_the_loop_opcode and
95 get_frag_is_no_transform.
96 (convert_frag_immed_finish_loop): Inline calls to get_expression_value
97 and set_frag_is_no_transform.
98 (get_expression_value): Delete.
99 (subseg_map struct): Rename cur_total_freq to total_freq. Rename
100 cur_target_freq to target_freq.
101 (get_subseg_info): Split out code to create a new map entry into ...
102 (add_subseg_info): ... this new function.
103 (get_last_insn_flags): Check if get_subseg_info succeeded.
104 (set_last_insn_flags): Call add_subseg_info if needed.
105 (get_subseg_total_freq, get_subseg_target_freq, set_subseg_freq): New.
106 (xtensa_reorder_segments): Compute last_sec while counting sections.
107 Remove call to get_last_sec.
108 (get_last_sec): Delete.
109 (cache_literal_section): Inline call to retrieve_literal_seg and its
110 callees, seg_present and add_seg_list.
111 (retrieve_literal_seg, seg_present, add_seg_list): Delete.
112 (get_frag_is_insn, get_frag_is_no_transform,
113 set_frag_is_specific_opcode, set_frag_is_no_transform): Delete.
114 * config/tc-xtensa.h (MAX_SLOTS): Reduce from 31 to 15.
116 2004-10-12 Bob Wilson <bob.wilson@acm.org>
118 * config/tc-xtensa.c: Use ISO C90 formatting.
119 * config/tc-xtensa.h: Likewise.
120 * config/xtensa-istack.h: Likewise.
121 * config/xtensa-relax.c: Likewise.
122 * config/xtensa-relax.h: Likewise.
124 2004-10-12 Paul Brook <paul@codesourcery.com>
126 * config/tc-arm.c (md_begin): Change EF_ARM_EABI_VER3 to
129 * doc/c-arm.texi: Document that we actually support -meabi=4, not
132 2004-10-08 Bob Wilson <bob.wilson@acm.org>
134 * doc/as.texinfo (VTableEntry, VTableInherit): Add "directive" to index
136 (Acknowledgements): Use "GAS" instead of AS variable.
138 2004-10-08 Daniel Jacobowitz <dan@debian.org>
140 * config/tc-i386.c: Include "elf/x86-64.h".
141 (i386_elf_section_type): New function.
142 * config/tc-i386.h (md_elf_section_type): Define.
143 (i386_elf_section_type): New prototype.
145 2004-10-08 Linus Nielsen Feltzing <linus@haxx.se>
147 * config/m68k-parse.h (enum m68k_register): New control register,
149 * config/tc-m68k.c: Correct control register set for MCF5249.
151 2004-10-07 Bob Wilson <bob.wilson@acm.org>
152 Sterling Augustine <sterling@tensilica.com>
154 * config/tc-xtensa.c (absolute_literals_supported): New global flag.
155 (UNREACHABLE_MAX_WIDTH): Define.
156 (XTENSA_FETCH_WIDTH): Delete.
157 (cur_vinsn, xtensa_fetch_width, xt_saved_debug_type, past_xtensa_end,
158 prefer_const16, prefer_l32r): New global variables.
159 (LIT4_SECTION_NAME): Define.
160 (lit4_state struct): Add lit4_seg_name and lit4_seg fields.
161 (XTENSA_PROP_*, GET_XTENSA_PROP_*, SET_XTENSA_PROP_*): Define.
162 (frag_flags struct): New.
163 (xtensa_block_info struct): Move from tc-xtensa.h. Add flags field.
164 (subseg_map struct): Add cur_total_freq and cur_target_freq fields.
165 (bitfield, bit_is_set, set_bit, clear_bit): Define.
166 (MAX_FORMATS): Define.
167 (op_placement_info struct, op_placement_table): New.
168 (O_pltrel, O_hi16, O_lo16): Define.
169 (directiveE enum): Rename directive_generics to directive_transform.
170 Delete directive_relax. Add directive_schedule,
171 directive_absolute_literals, and directive_last_directive.
172 (directive_info): Rename "generics" to "transform". Delete "relax".
173 Add "schedule" and "absolute-literals".
174 (directive_state): Adjust entries to match changes in directive_info.
175 (xtensa_relax_statesE, RELAX_IMMED_MAXSTEPS): Move to tc-xtensa.h.
176 (xtensa_const16_opcode, xtensa_movi_opcode, xtensa_movi_n_opcode,
177 xtensa_l32r_opcode, xtensa_nop_opcode, xtensa_rsr_lcount_opcode): New.
178 (xtensa_j_opcode, xtensa_rsr_opcode): Delete.
179 (align_only_targets, software_a0_b_retw_interlock,
180 software_avoid_b_j_loop_end, maybe_has_b_j_loop_end,
181 software_avoid_short_loop, software_avoid_close_loop_end,
182 software_avoid_all_short_loops, specific_opcode): Delete.
183 (warn_unaligned_branch_targets): New.
184 (workaround_a0_b_retw, workaround_b_j_loop_end, workaround_short_loop,
185 workaround_close_loop_end, workaround_all_short_loops): Default FALSE.
186 (option_[no_]link_relax, option_[no_]transform,
187 option_[no_]absolute_literals, option_warn_unaligned_targets,
188 option_prefer_l32r, option_prefer_const16, option_target_hardware):
190 (option_[no_]align_only_targets, option_literal_section_name,
191 option_text_section_name, option_data_section_name,
192 option_bss_section_name, option_eb, option_el): Delete.
193 (md_longopts): Add entries for: [no-]transform, [no-]absolute-literals,
194 warn-unaligned-targets, prefer-l32r, prefer-const16, [no-]link-relax,
195 and target-hardware. Delete entries for [no-]target-align-only,
196 literal-section-name, text-section-name, data-section-name, and
198 (md_parse_option): Handle new options and remove old ones. Accept but
199 ignore [no-]density options. Warn for [no-]generics and [no-]relax
200 and treat them as [no-]transform.
201 (md_show_usage): Add new options and remove old ones.
202 (xtensa_setup_hw_workarounds): New.
203 (md_pseudo_table): Change "word" entry to use xtensa_elf_cons. Add
204 "long", "short", "loc" and "frequency" entries.
205 (use_generics): Rename to ...
206 (use_transform): ... this function. Add past_xtensa_end check.
207 (use_longcalls): Add past_xtensa_end check.
208 (code_density_available, can_relax): Delete.
209 (do_align_targets): New.
210 (get_directive): Accept dashes in directive names. Warn about
211 [no-]generics and [no-]relax directives and treat them as
213 (xtensa_begin_directive): Call md_flush_pending_output only for some
214 directives. Check for directives inside instruction bundles. Warn
215 about deprecated ".begin literal" usage. Warn and ignore [no-]density
216 directives. Handle new directives. Check generating_literals flag
218 (xtensa_end_directive): Check for directives inside instruction
219 bundles. Warn and ignore [no-]density directives. Handle new
220 directives. Call xtensa_set_frag_assembly_state.
221 (xtensa_loc_directive_seen, xtensa_dwarf2_directive_loc,
222 xtensa_dwarf2_emit_insn): New.
223 (xtensa_literal_position): Call md_flush_pending_output. Do not check
224 use_literal_section flag.
225 (xtensa_literal_pseudo): Call md_flush_pending_output. Handle absolute
226 literals. Use xtensa_elf_cons to parse the expression.
227 (xtensa_literal_prefix): Do not check use_literal_section. Support
228 ".lit4" sections for absolute literals. Change prefix convention to
229 replace ".text" (or ".t" in a linkonce section). No need to call
231 (xtensa_frequency_pseudo, xtensa_elf_cons, xtensa_elf_suffix): New.
232 (expression_end): Handle closing braces and colons.
233 (PLT_SUFFIX, plt_suffix): Delete.
234 (expression_maybe_register): Use new xtensa-isa.h functions. Use
235 xtensa_elf_suffix instead of checking for plt suffix, and handle O_lo16
236 and O_hi16 expressions as well.
237 (tokenize_arguments): Handle closing braces and colons.
238 (parse_arguments): Use new xtensa-isa.h functions. Handle "invisible"
239 operands and paired register syntax.
240 (get_invisible_operands): New.
241 (xg_translate_sysreg_op): Handle new Xtensa LX RSR/WSR/XSR syntax. Use
242 new xtensa-isa.h functions.
243 (xtensa_translate_old_userreg_ops, xtensa_translate_zero_immed): New.
244 (xg_translate_idioms): Check if inside bundle. Use use_transform.
245 Handle new Xtensa LX RSR/WSR/XSR syntax. Remove code to widen density
246 instructions. Use xtensa_translate_zero_immed.
247 (operand_is_immed, operand_is_pcrel_label): Delete.
248 (get_relaxable_immed): Use new xtensa-isa.h functions.
249 (get_opcode_from_buf): Add slot parameter. Use new xtensa-isa.h
251 (xtensa_print_insn_table, print_vliw_insn): New.
252 (is_direct_call_opcode): Use new xtensa-isa.h functions.
253 (is_call_opcode, is_loop_opcode, is_conditional_branch_opcode,
254 is_branch_or_jump_opcode): Delete.
255 (is_movi_opcode, decode_reloc, encode_reloc, encode_alt_reloc): New.
256 (opnum_to_reloc, reloc_to_opnum): Delete.
257 (xtensa_insnbuf_set_operand, xtensa_insnbuf_get_operand): Use new
258 xtensa-isa.h functions. Operate on one slot of an instruction.
259 (xtensa_insnbuf_set_immediate_field, is_negatable_branch,
260 xg_get_insn_size): Delete.
261 (xg_get_build_instr_size): Use xg_get_single_size.
262 (xg_is_narrow_insn, xg_is_single_relaxable_insn): Update calls to
263 xg_build_widen_table. Use xg_get_single_size.
264 (xg_get_max_narrow_insn_size): Delete.
265 (xg_get_max_insn_widen_size, xg_get_max_insn_widen_literal_size,
266 xg_is_relaxable_insn): Update calls to xg_build_widen_table. Use
268 (xg_build_to_insn): Record the loc field. Handle OP_OPERAND_HI16U and
269 OP_OPERAND_LOW16U. Check xg_valid_literal_expression.
270 (xg_expand_to_stack, xg_expand_narrow): Update calls to
271 xg_build_widen_table. Use xg_get_single_size.
272 (xg_immeds_fit): Use new xtensa-isa.h functions. Update call to
274 (xg_symbolic_immeds_fit): Likewise. Also handle O_lo16 and O_hi16, and
275 treat weak symbols conservatively.
276 (xg_check_operand): Use new xtensa-isa.h functions.
277 (is_dnrange): Delete.
278 (xg_assembly_relax): Inline previous calls to tinsn_copy.
279 (xg_finish_frag): Specify separate relax states for the frag and slot0.
280 (is_branch_jmp_to_next, xg_add_branch_and_loop_targets): Use new
281 xtensa-isa.h functions.
282 (xg_instruction_matches_option_term, xg_instruction_matches_or_options,
283 xg_instruction_matches_options): New.
284 (xg_instruction_matches_rule): Handle O_register expressions. Call
285 xg_instruction_matches_options.
286 (transition_rule_cmp): New.
287 (xg_instruction_match): Update call to xg_build_simplify_table.
288 (xg_build_token_insn): Record loc fields.
289 (xg_simplify_insn): Check is_specific_opcode field and
290 density_supported flag.
291 (xg_expand_assembly_insn): Skip checking code_density_available. Use
292 new xtensa-isa.h functions. Call use_transform instead of can_relax.
293 (xg_assemble_literal): Add error handling for O_big. Call
294 record_alignment. Handle O_pltrel.
295 (xg_valid_literal_expression): New.
296 (xg_assemble_literal_space): Add slot parameter. Remove call to
297 set_expr_symbol_offset. Add call to record_alignment. Update call to
299 (xg_emit_insn): Delete.
300 (xg_emit_insn_to_buf): Add format parameter. Update calls to
301 xg_add_opcode_fix and xtensa_insnbuf_to_chars.
302 (xg_add_opcode_fix): Change opcode parameter to tinsn and add format
303 and slot parameters. Handle new "alternate" relocations for absolute
304 literals and CONST16 instructions. Check for bad uses of O_lo16 and
305 O_hi16. Use new xtensa-isa.h functions.
306 (xg_assemble_tokens): Delete.
307 (is_register_writer): Use new xtensa-isa.h functions.
308 (is_bad_loopend_opcode): Check for xtensa_rsr_lcount_opcode instead of
309 old-style RSR from LCOUNT.
310 (next_frag_opcode): Delete.
311 (next_frag_opcode_is_loop, next_frag_format_size, frag_format_size,
312 update_next_frag_state): New.
313 (update_next_frag_nop_state): Delete.
314 (next_frag_pre_opcode_bytes): Use next_frag_opcode_is_loop.
315 (xtensa_mark_literal_pool_location): Check use_literal_section flag and
316 the state of the absolute-literals directive. Add calls to
317 record_alignment and xtensa_set_frag_assembly_state. Call
318 xtensa_switch_to_non_abs_literal_fragment instead of
319 xtensa_switch_to_literal_fragment.
321 (assemble_nop): Use build_nop. Update call to xtensa_insnbuf_to_chars.
322 (get_expanded_loop_offset): Change check for undefined opcode to an
324 (xtensa_set_frag_assembly_state, relaxable_section,
325 xtensa_find_unmarked_state_frags, xtensa_find_unaligned_branch_targets,
326 xtensa_find_unaligned_loops, xg_apply_tentative_value): New.
327 (md_begin): Update call to xtensa_isa_init. Initialize linkrelax to 1.
328 Set lit4_seg_name. Call xg_init_vinsn. Initialize new global opcodes.
329 Call init_op_placement_info_table and xtensa_set_frag_assembly_state.
330 (xtensa_init_fix_data): New.
331 (xtensa_frob_label): Reset label symbol to the current frag. Check
332 do_align_targets and generating_literals flag. Propagate frequency
333 info to new alignment frag. Call xtensa_set_frag_assembly_state.
334 (xtensa_unrecognized_line): New.
335 (xtensa_flush_pending_output): Check if inside a bundle. Add a call
336 to xtensa_set_frag_assembly_state.
337 (error_reset_cur_vinsn): New.
338 (md_assemble): Remove check for literal frag. Remove call to
339 istack_init. Call use_transform instead of use_generics. Parse
340 explicit instruction format specifiers. Move code for
341 a0_b_retw_interlock workaround to xg_assemble_vliw_tokens. Call
342 error_reset_cur_vinsn on errors. Add call to get_invisible_operands.
343 Add dwarf2_where call. Remote automatic alignment for ENTRY
344 instructions. Move call to xtensa_clear_insn_labels to the end.
345 Rearrange to handle bundles.
346 (xtensa_cons_fix_new): Delete.
347 (xtensa_handle_align): New.
348 (xtensa_frag_init): Call xtensa_set_frag_assembly_state. Remove
349 assignment to is_no_density field.
350 (md_pcrel_from): Use new xtensa-isa.h functions. Use decode_reloc
351 instead of reloc_to_opnum. Handle "alternate" relocations.
352 (xtensa_force_relocation, xtensa_check_inside_bundle,
353 xtensa_elf_section_change_hook): New.
354 (xtensa_symbol_new_hook): Delete.
355 (xtensa_fix_adjustable): Check for difference of symbols with an
356 offset. Check for external and weak symbols.
357 (md_apply_fix3): Remove cases for XTENSA_OP{0,1,2} relocs.
358 (md_estimate_size_before_relax): Return expansion for the first slot.
359 (tc_gen_reloc): Handle difference of symbols by producing
360 XTENSA_DIFF{8,16,32} relocs and by writing the value of the difference
361 into the output. Handle new XTENSA_SLOT*_OP relocs by storing the
362 tentative values into the output when linkrelax is set.
363 (XTENSA_PROP_SEC_NAME): Define.
364 (xtensa_post_relax_hook): Call xtensa_find_unmarked_state_frags.
365 Create literal tables only if using literal sections. Create new
366 property tables instead of old instruction tables. Check for unaligned
367 branch targets and loops.
368 (finish_vinsn, find_vinsn_conflicts, check_t1_t2_reads_and_writes,
369 new_resource_table, clear_resource_table, resize_resource_table,
370 resources_available, reserve_resources, release_resources,
371 opcode_funcUnit_use_unit, opcode_funcUnit_use_stage,
372 resources_conflict, xg_find_narrowest_format, relaxation_requirements,
373 bundle_single_op, emit_single_op, xg_assemble_vliw_tokens): New.
374 (xtensa_end): Call xtensa_flush_pending_output. Set past_xtensa_end
375 flag. Update checks for workaround options. Call
376 xtensa_mark_narrow_branches and xtensa_mark_zcl_first_insns.
377 (xtensa_cleanup_align_frags): Add special case for branch targets.
378 Check for and mark unreachable frags.
379 (xtensa_fix_target_frags): Remove use of align_only_targets flag.
380 Use RELAX_LOOP_END_BYTES in special case for negatable branch at the
381 end of a zero-overhead loop body.
382 (frag_can_negate_branch): Handle instructions with multiple slots.
383 Use new xtensa-isa.h functions
384 (xtensa_mark_narrow_branches, is_narrow_branch_guaranteed_in_range,
385 xtensa_mark_zcl_first_insns): New.
386 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags): Error if
387 transformations are disabled.
388 (next_instrs_are_b_retw): Use new xtensa-isa.h functions. Handle
389 multislot instructions.
390 (xtensa_fix_close_loop_end_frags, xtensa_fix_short_loop_frags):
391 Likewise. Also error if transformations are disabled.
392 (unrelaxed_frag_max_size): New.
393 (unrelaxed_frag_min_insn_count, unrelax_frag_has_b_j): Use new
394 xtensa-isa.h functions.
395 (xtensa_sanity_check, is_empty_loop, is_local_forward_loop): Use
396 xtensa_opcode_is_loop instead of is_loop_opcode.
397 (get_text_align_power): Replace as_fatal with assertion.
398 (get_text_align_fill_size): Iterate instead of using modulus when
400 (get_noop_aligned_address): Assert that this is for a machine-dependent
401 RELAX_ALIGN_NEXT_OPCODE frag. Use next_frag_opcode_is_loop,
402 xg_get_single_size, and frag_format_size.
403 (get_widen_aligned_address): Rename to ...
404 (get_aligned_diff): ... this function. Add max_diff parameter.
405 Remove handling of rs_align/rs_align_code frags. Use
406 next_frag_format_size, get_text_align_power, get_text_align_fill_size,
407 next_frag_opcode_is_loop, and xg_get_single_size. Compute max_diff
408 and pass it back to caller.
409 (xtensa_relax_frag): Use relax_frag_loop_align. Add code for new
410 RELAX_SLOTS, RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN,
411 RELAX_FILL_NOP, and RELAX_UNREACHABLE frag types. Check relax_seen.
412 (relax_frag_text_align): Rename to ...
413 (relax_frag_loop_align): ... this function. Assume loops can only be
414 in the first slot of an instruction.
415 (relax_frag_add_nop): Use assemble_nop instead of constructing an OR
416 instruction. Remove call to frag_wane.
417 (relax_frag_narrow): Rename to ...
418 (relax_frag_for_align): ... this function. Extend to handle
419 RELAX_FILL_NOP and RELAX_UNREACHABLE, as well as RELAX_SLOTS with
420 RELAX_NARROW for the first slot.
421 (find_address_of_next_align_frag, bytes_to_stretch): New.
422 (future_alignment_required): Use find_address_of_next_align_frag and
423 bytes_to_stretch. Look ahead to subsequent frags to make smarter
425 (relax_frag_immed): Add format, slot, and estimate_only parameters.
426 Check if transformations are enabled for b_j_loop_end workaround.
427 Use new xtensa-isa.h functions and handle multislot instructions.
428 Update call to xg_assembly_relax.
429 (md_convert_frag): Handle new RELAX_SLOTS, RELAX_UNREACHABLE,
430 RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN, and RELAX_FILL_NOP
432 (convert_frag_narrow): Add segP, format and slot parameters. Call
433 convert_frag_immed for branch instructions. Adjust calls to
434 tinsn_from_chars, tinsn_immed_from_frag, and xg_emit_insn_to_buf. Use
435 xg_get_single_size and xg_get_single_format.
436 (convert_frag_fill_nop): New.
437 (convert_frag_immed): Add format and slot parameters. Handle multislot
438 instructions and use new xtensa-isa.h functions. Update calls to
439 tinsn_immed_from_frag and xg_assembly_relax. Check if transformations
440 enabled for b_j_loop_end workaround. Use build_nop instead of
441 assemble_nop. Check is_specific_opcode flag. Check for unreachable
442 frags. Use xg_get_single_size. Handle O_pltrel.
443 (fix_new_exp_in_seg): Remove check for old plt flag.
444 (convert_frag_immed_finish_loop): Update calls to tinsn_from_chars and
445 xtensa_insnbuf_to_chars. Call tinsn_immed_from_frag. Change check
446 for loop opcode to an assertion. Mark all frags up to the end of the
447 loop as not transformable.
448 (get_last_insn_flags, set_last_insn_flags): Use get_subseg_info.
449 (get_subseg_info): New.
450 (xtensa_move_literals): Call xtensa_set_frag_assembly_state. Add null
452 (xtensa_switch_to_literal_fragment): Rewrite to handle absolute
453 literals and use xtensa_switch_to_non_abs_literal_fragment otherwise.
454 (xtensa_switch_to_non_abs_literal_fragment): New.
455 (cache_literal_section): Add is_code parameter and pass it through to
456 retrieve_literal_seg.
457 (retrieve_literal_seg): Add is_code parameter and use it to set the
458 flags on the literal section. Handle case where head parameter is 0.
459 (get_frag_is_no_transform, set_frag_is_specific_opcode,
460 set_frag_is_no_transform): New.
461 (xtensa_create_property_segments): Add end_property_function parameter
462 and pass it through to add_xt_block_frags. Call bfd_get_section_flags
463 and skip SEC_DEBUGGING and !SEC_ALLOC sections.
464 (xtensa_create_xproperty_segments, section_has_xproperty): New.
465 (add_xt_block_frags): Add end_property_function parameter and call it
466 if it is non-zero. Call xtensa_frag_flags_init.
467 (xtensa_frag_flags_is_empty, xtensa_frag_flags_init,
468 get_frag_property_flags, frag_flags_to_number,
469 xtensa_frag_flags_combinable, xt_block_aligned_size,
470 xtensa_xt_block_combine, add_xt_prop_frags,
471 init_op_placement_info_table, opcode_fits_format_slot,
472 xg_get_single_size, xg_get_single_format): New.
473 (istack_push): Inline call to tinsn_copy.
474 (tinsn_copy): Delete.
475 (tinsn_has_invalid_symbolic_operands): Handle O_hi16 and O_lo16 and
476 CONST16 opcodes. Handle O_big, O_illegal, and O_absent.
477 (tinsn_has_complex_operands): Handle O_hi16 and O_lo16.
478 (tinsn_to_insnbuf): Use xg_get_single_format and new xtensa-isa.h
479 functions. Handle invisible operands.
480 (tinsn_to_slotbuf): New.
481 (tinsn_check_arguments): Use new xtensa-isa.h functions.
482 (tinsn_from_chars): Add slot parameter. Rewrite using xg_init_vinsn,
483 vinsn_from_chars, and xg_free_vinsn.
484 (tinsn_from_insnbuf): New.
485 (tinsn_immed_from_frag): Add slot parameter and handle multislot
486 instructions. Handle symbol differences.
487 (get_num_stack_text_bytes): Use xg_get_single_size.
488 (xg_init_vinsn, xg_clear_vinsn, vinsn_has_specific_opcodes,
489 xg_free_vinsn, vinsn_to_insnbuf, vinsn_from_chars, expr_is_register,
490 get_expr_register, set_expr_symbol_offset_diff): New.
491 * config/tc-xtensa.h (MAX_SLOTS): Define.
492 (xtensa_relax_statesE): Move from tc-xtensa.c. Add
493 RELAX_CHECK_ALIGN_NEXT_OPCODE, RELAX_MAYBE_DESIRE_ALIGN, RELAX_SLOTS,
494 RELAX_FILL_NOP, RELAX_UNREACHABLE, RELAX_MAYBE_UNREACHABLE, and
496 (RELAX_IMMED_MAXSTEPS): Move from tc-xtensa.c.
497 (xtensa_frag_type struct): Add is_assembly_state_set,
498 use_absolute_literals, relax_seen, is_unreachable, is_specific_opcode,
499 is_align, is_text_align, alignment, and is_first_loop_insn fields.
500 Replace is_generics and is_relax fields by is_no_transform field.
501 Delete is_text and is_longcalls fields. Change text_expansion and
502 literal_expansion to arrays of MAX_SLOTS entries. Add arrays of
503 per-slot information: literal_frags, slot_subtypes, slot_symbols,
504 slot_sub_symbols, and slot_offsets. Add fr_prev field.
505 (xtensa_fix_data struct): New.
506 (xtensa_symfield_type struct): Delete plt field.
507 (xtensa_block_info struct): Move definition to tc-xtensa.h. Add
508 forward declaration here.
509 (xt_section_type enum): Delete xt_insn_sec. Add xt_prop_sec.
510 (XTENSA_SECTION_RENAME): Undefine.
511 (TC_FIX_TYPE, TC_INIT_FIX_DATA, TC_FORCE_RELOCATION, NO_PSEUDO_DOT,
512 tc_unrecognized_line, md_do_align, md_elf_section_change_hook,
513 HANDLE_ALIGN, TC_LINKRELAX_FIXUP, SUB_SEGMENT_ALIGN): Define.
514 (TC_CONS_FIX_NEW, tc_symbol_new_hook): Delete.
515 (unit_num_copies_func, opcode_num_units_func,
516 opcode_funcUnit_use_unit_func, opcode_funcUnit_use_stage_func): New.
517 (resource_table struct): New.
518 * config/xtensa-istack.h (MAX_INSN_ARGS): Increase from 6 to 10.
519 (TInsn struct): Add keep_wide, loc, fixup, record_fix, subtype,
520 literal_space, symbol, sub_symbol, offset, and literal_frag fields.
521 (tinsn_copy): Delete prototype.
522 (vliw_insn struct): New.
523 * config/xtensa-relax.c (insn_pattern_struct): Add options field.
524 (widen_spec_list): Add option conditions for density and boolean
525 instructions. Add expansions using CONST16 and conditions for using
526 CONST16 vs. L32R. Use new Xtensa LX RSR/WSR syntax. Add entries for
528 (simplify_spec_list): Add option conditions for density instructions.
529 Add entry for NOP instruction.
530 (append_transition): Add cmp function pointer parameter and use it to
531 insert the new entry in order.
532 (operand_function_LOW16U, operand_function_HI16U): New.
533 (xg_has_userdef_op_fn, xg_apply_userdef_op_fn): Handle
534 OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
535 (enter_opname, split_string): Use xstrdup instead of strdup.
536 (init_insn_pattern): Initialize new options field.
537 (clear_req_or_option_list, clear_req_option_list,
538 clone_req_or_option_list, clone_req_option_list, parse_option_cond):
540 (parse_insn_pattern): Parse option conditions.
541 (transition_applies): New.
542 (build_transition): Use new xtensa-isa.h functions. Fix incorrectly
543 swapped last arguments in calls to append_constant_value_condition.
544 Call clone_req_option_list. Add warning about invalid opcode.
545 Handle LOW16U and HI16U function names.
546 (build_transition_table): Add cmp parameter and use it in calls to
547 append_transition. Use new xtensa-isa.h functions. Check
548 transition_applies before adding entries.
549 (xg_build_widen_table, xg_build_simplify_table): Add cmp parameter and
550 pass it through to build_transition_table.
551 * config/xtensa-relax.h (ReqOrOptionList, ReqOrOption, ReqOptionList,
552 ReqOption, transition_cmp_fn): New types.
553 (OpType enum): Add OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
554 (transition_rule struct): Add options field.
555 * doc/as.texinfo (Overview): Update Xtensa options.
556 * doc/c-xtensa.texi (Xtensa Options): Delete --[no-]density,
557 --[no-]relax, and --[no-]generics options. Update descriptions of
558 --text-section-literals and --[no-]longcalls. Add
559 --[no-]absolute-literals and --[no-]transform.
560 (Xtensa Syntax): Add description of syntax for FLIX instructions.
561 Remove use of "generic" and "specific" terminology for opcodes.
562 (Xtensa Registers): Generalize the syntax description to include
563 user-defined register files.
564 (Xtensa Automatic Alignment): Update.
565 (Xtensa Branch Relaxation): Mention limitation of unconditional jumps.
566 (Xtensa Call Relaxation): Linker can now remove most of the overhead.
567 (Xtensa Directives): Remove confusing rules about precedence.
568 (Density Directive, Relax Directive): Delete.
569 (Schedule Directive): New.
570 (Generics Directive): Rename to ...
571 (Transform Directive): ... this node.
572 (Literal Directive): Update for absolute literals. Missing
573 literal_position directive is now an error.
574 (Literal Position Directive): Update for absolute literals.
575 (Freeregs Directive): Delete.
576 (Absolute Literals Directive): New.
577 (Frame Directive): Minor editing.
578 * Makefile.am (DEPTC_xtensa_elf, DEPOBJ_xtensa_elf, DEP_xtensa_elf):
580 * Makefile.in: Regenerate.
582 2004-10-07 Richard Sandiford <rsandifo@redhat.com>
584 * config/tc-mips.c (append_insn): Use fix_new rather than fix_new_exp
585 to build the second and third fixups for a composite relocation.
586 (macro_read_relocs): New function.
587 (macro_build): Use it.
588 (s_cpsetup): Pass all three composite relocation codes to macro_build.
589 Simplify fragging code accordingly.
590 (s_gpdword): Use fix_new rather than fix_new_exp for the second part
591 of the composite relocation. Set fx_tcbit in both fixups.
593 2004-10-07 Richard Sandiford <rsandifo@redhat.com>
595 * config/tc-mips.c (append_insn): Set fx_tcbit for composite relocs.
596 (md_apply_fix3): Don't treat composite relocs as done.
598 2004-10-07 Jan Beulich <jbeulich@novell.com>
600 * macro.c (macro_expand_body): When ELF, use .LL rather than LL as
601 prefix for symbol names generated from the LOCAL macro directive.
603 * dw2gencfi.c (select_cie_for_fde): When separating CIE out from
604 FDE, treat a DW_CFA_remember_state as we do a DW_CFA_advance_loc.
606 2004-10-07 Tomer Levi <Tomer.Levi@nsc.com>
608 * config/tc-crx.c (preprocess_reglist): Handle Co-processor
610 (md_assemble): Add error checking for Co-Processor instructions.
611 (get_cinv_parameters): Add 'b' option to invalidate the
614 2004-10-05 Paul Brook <paul@codesourcery.com>
616 * config/tc-arm.c (unwind): New variable.
617 (vfp_sp_encode_reg): New function.
618 (vfp_sp_reg_required_here): Use it.
619 (vfp_sp_reg_list, vfp_dp_reg_list): Remove.
620 (vfp_parse_reg_list): New function.
621 (s_arm_unwind_fnstart, s_arm_unwind_fnend, s_arm_unwind_cantunwind,
622 s_arm_unwind_personality, s_arm_unwind_personalityindex,
623 s_arm_unwind_handlerdata, s_arm_unwind_save, s_arm_unwind_movsp,
624 s_arm_unwind_pad, s_arm_unwind_setfp, s_arm_unwind_raw): New
626 (md_pseudo_table): Add them.
627 (do_vfp_reg2_from_sp2): Use vfp_parse_reg_list and vfp_sp_encode_reg.
628 (do_vfp_sp2_from_reg2, vfp_sp_ldstm, vfp_dp_ldstm): Ditto.
629 (set_section, add_unwind_adjustsp, flush_pending_unwind,
630 finish_unwind_opcodes, start_unwind_section, create_unwind_entry,
631 require_hashconst, add_unwind_opcode): New functions.
632 * doc/tc-arm.text: Document unwinding opcodes.
633 * NEWS: Mention the new feature.
635 2004-10-04 Eric Christopher <echristo@redhat.com>
637 * config/tc-mips.c (md_apply_fix3): Remove erroneous assert.
639 2004-10-01 H.J. Lu <hongjiu.lu@intel.com>
641 * config/tc-ppc.c (md_apply_fix3): Call S_SET_THREAD_LOCAL for
643 * config/tc-s390.c (md_apply_fix3): Likewise.
644 * config/tc-sparc.c (md_apply_fix3): Likewise.
646 2004-10-01 Paul Brook <paul@codesourcery.com>
648 * config/tc-arm.c (arm_elf_section_type): New function.
649 (arm_elf_change_section): Set section link for exidx sections.
650 * config/tc-arm.h (arm_elf_section_type): Add prototype.
651 (md_elf_section_type): Define.
653 2004-10-01 Bill Farmer <Bill@the-farmers.freeserve.co.uk>
655 * config/tc-pdp11.c (md_apply_fix3): Change to sign of the SOB
656 instruction's offset.
658 2004-10-01 Adam Nemet <anemet@lnxw.com>
660 * (TARGET_FORMAT): Remove LynxOS COFF definition.
662 2004-10-01 Ravi Ramaseshan <ravi.ramaseshan@codito.com>
664 * config/tc-arc.c (tc_gen_reloc): Don't assume fixP->fx_addsy is an
665 asymbol *, instead use symbol_get_bfdsym.
667 2004-09-30 Linus Nielsen Feltzing <linus@haxx.se>
669 * config/tc-m68k.c (select_control_regs): Add mcf5249.
671 2004-09-30 Paul Brook <paul@codesourcery.com>
673 * config/tc-arm.c (do_smi, do_nop): New functions.
674 (insns): Add ARMv6ZK instructions.
675 (md_apply_fix3): Handle BFD_RELOC_ARM_SMI.
676 (tc_gen_reloc): Ditto.
677 (arm_cpus): Add mpcore and arm1176.
678 (arm_archs): Add armv6{k,z,zk}.
679 * doc/c-arm.texi: Document new cores and architectures.
681 2004-09-30 Nick Clifton <nickc@redhat.com>
683 * config/tc-arm.c: Use ISO C90 formatting.
685 2004-09-30 Vladimir Ivanov <vladitx@nucleusys.com>
687 * config/tc-arm.c (mav_reg_required_here): Allow REG_TYPE_CN
688 as alternative when REG_TYPE_MVF, REG_TYPE_MVD, REG_TYPE_MVFX or
689 REG_TYPE_MVDX is expected.
691 2004-09-29 Marc Bevand <m.bevand@gmail.com>
693 * doc/c-i386.texi (i386-Mnemonics): Fix typo.
695 2004-09-21 James E Wilson <wilson@specifixinc.com>
697 * config/tc-ia64.c (ENCODED_PSP_OFFSET): New.
698 (output_rp_psprel, output_pfs_psprel, output_preds_psprel,
699 output_spill_base, output_unat_psprel, output_lc_psprel,
700 output_fpsr_psprel, output_priunat_psprel, output_bsp_psprel,
701 output_bsprestore_psprel, output_rnat_psprel, output_spill_psprel,
702 output_spill_psprel_p): Use it.
704 2004-09-20 Tomer Levi <Tomer.Levi@nsc.com>
706 * config/tc-crx.c (handle_LoadStor): New function.
707 Handle load/stor unique instructions before parsing.
709 2004-09-17 Paul Brook <paul@codesourcery.com>
711 * config/tc-arm.c (s_arm_rel31): New funciton.
712 (md_pseudo_table): Add .rel31.
713 (md_apply_fix3): Handle BFD_RELOC_ARM_TARGET2,
714 BFD_RELOC_32_PCREL and BFD_RELOC_ARM_PREL31.
715 (tc_gen_reloc): Handle BFD_RELOC_ARM_PREL31 and BFD_RELOC_ARM_TARGET2.
716 (arm_fix_adjustable): Return 0 for BFD_RELOC_ARM_TARGET2.
717 (arm_parse_reloc): Add (target2).
719 2004-09-17 Alan Modra <amodra@bigpond.net.au>
721 * Makefile.am: Run "make dep-am".
722 * Makefile.in: Regenerate.
723 * aclocal.m4: Regenerate.
724 * configure: Regenerate.
725 * doc/Makefile.in: Regenerate.
726 * po/POTFILES.in: Regenerate.
727 * po/gas.pot: Regenerate.
729 2004-09-14 Hideki IWAMOTO <h-iwamoto@kit.hi-ho.ne.jp>
731 * config/tc-mmix.c [!LLONG_MIN]: Correct #elsif to #elif.
734 2004-09-13 Paul Brook <paul@codesourcery.com>
736 * config/tc-arm.c: Rename RELABS to TARGET1.
738 2004-09-13 Alan Modra <amodra@bigpond.net.au>
740 * messages.c (as_internal_value_out_of_range): Cast values passed
741 to as_bad_where or as_warn_where to proper type.
743 2004-09-11 Theodore A. Roth <troth@openavr.org>
745 * config/tc-avr.c: Add support for
746 atmega48, atmega88, atmega168, attiny13, attiny2313, at90can128.
748 2004-09-09 Alan Modra <amodra@bigpond.net.au>
750 * dw2gencfi.c (select_cie_for_fde): When separating CIE out
751 from FDE, treat a CFI_escape as we do a DW_CFA_advance_loc.
753 2004-09-08 Paul Brook <paul@codesourcery.com>
755 * config/obj-elf.c (obj_elf_section_type): Handle init_array,
756 fini_array and preinit_array section types.
757 * config/tc-ia64.c (ia64_elf_section_type): Remove init_array
759 * doc/as.texinfo: Document extra section types.
761 2004-09-02 Mark Mitchell <mark@codesourcery.com>
763 * Makefile.am (TARG_ENV_HFILES): Add te-symbian.h.
764 * Makefile.in: Regenerated.
765 * configure.in: Set em for arm*-*-symbianelf*.
766 * configure: Regenerated.
767 * config/tc-arm.c (elf32_arm_target_format): Use Symbian target
768 vectors when appropriate.
769 * config/te-symbian.h: New file.
771 2004-09-03 Tomer Levi <Tomer.Levi@nsc.com>
773 * config/tc-crx.c (gettrap): Exception vector can be case
775 (process_label_constant): Fix a 32-bit displacement bug in branch
777 (get_operandtype) : Bug fix, wrong operand was used.
778 (process_label_constant): Initialize relocation type to
781 2004-09-01 Richard Earnshaw < reanrsha@arm.com>
783 * tc-arm.c (arm_cpus, arm_fpus): Allow <cpu>-s as well as <cpu>s
784 for synthesizable cores.
786 * doc/c-arm.texi (ARM Options): Document canonical names of CPUs.
788 2004-08-25 Dmitry Diky <diwil@spec.ru>
790 * config/tc-msp430.c: Clean-up the code.
791 (md_relax_table): New relax table.
792 (mcu_types): Sort MCU types.
793 (md_pseudo_table): Add .profiler pseudo handler.
794 (pow2value): New function.
795 (msp430_profiler): New function.
796 (msp430_operands): Add new insns handlers.
797 (msp430_srcoperand): Add register operand handler, allow complex
799 (md_estimate_size_before_relax): Rewritten.
800 (md_convert_frag): Rewritten.
801 (msp430_relax_frag): New function.
802 * config/tc-msp430.h (md_relax_frag): define macro
803 * doc/c-msp430.texi: Update information.
805 2004-08-24 Nick Clifton <nickc@redhat.com>
807 * as.c (std_shortopts): Allow -g to take an optional argument.
808 (parse_args): Pass any switch starting with -g on to the backend
811 2004-08-18 Mark Mitchell <mark@codesourcery.com>
813 * configure.in (arm*-*-symbianelf*): New target.
814 (arm*-*-eabi*): Likewise.
815 * configure: Regenerated.
817 2004-08-18 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
818 * config/tc-mips.c (append_insn): Handle delay slots in branch likely
821 2004-08-18 Jakub Jelinek <jakub@redhat.com>
823 * config/tc-ia64.c (start_unwind_section): Add linkonce_empty
824 argument, don't do anything if current section is not
825 .gnu.linkonce.t.* and linkonce_empty is set.
826 (generate_unwind_image, dot_endp): Adjust callers, call
827 start_unwind_section (*, 1) if nothing will be put into the
830 2004-08-17 Nick Clifton <nickc@redhat.com>
832 * as.c (MD_DEBUG_FORMAT_SELECTOR): Provide default definition.
833 (show_usage): Add -g.
834 (std_longopts): Add --gen-debug. Alpha sort the table.
835 (parse_args): Print an error message if a switch is not handled.
836 Handle the -g switch, calling md_debug_format_selector() if
838 * NEWS: Mention new feature.
839 * doc/as.texinfo: Document new switch.
840 * doc/internals.texi: Document behaviour of md_parse_option.
842 * config/tc-arm.c (md_parse_option): Do not issue an error message
843 if the switch is not recognised.
844 * config/tc-m68k.c (md_parse_option): Likewise.
845 * config/tc-pdp11.c (md_parse_option): Likewise.
846 * config/tc-v850.c (md_parse_option): Likewise.
848 * as.h: Fix up formatting.
851 2004-08-16 Nick Clifton <nickc@redhat.com>
853 * macro.c (macro_set_alternate): Use ISO C90 formatting.
855 * configure.in: Sort architecture based tables alphabetically.
856 * configure: Regenerate.
858 2004-08-16 Alan Modra <amodra@bigpond.net.au>
860 * config/tc-ppc.c (tc_ppc_regname_to_dw2regnum <regnames>): Replace
861 { "cc", 68 }, with { "cr", 70 }.
863 2004-08-13 Jan Beulich <jbeulich@novell.com>
864 Nick Clifton <nickc@redhat.com>
866 * as.c: Add and handle new --alternate command line option.
867 * macro.c (macro_set_alternate): New.
868 * macro.h (macro_set_alternate): Declare.
869 * read.c: Add and handle new .altmacro and .noaltmacro directives.
870 * doc/as.texinfo: Document new command line option and pseudo-ops
871 as well as insert documentation originating from gasp about
872 alternate macro syntax.
873 * NEWS: Mention new command line option and pseudo-ops.
875 2004-08-10 Mark Mitchell <mark@codesourcery.com>
877 * expr.c (operand): Handle the "~", "-", and "!" operators applied
880 2004-08-06 Paul Brook <paul@codesourcery.com>
882 * config/tc-arm.c (md_apply_fix3, tc_gen_reloc, arm_parse_reloc):
883 Handle new relocations.
884 * include/elf/arm.h (elf_arm_reloc_type): Add new EABI relocations.
886 2004-08-05 Bob Wilson <bob.wilson@acm.org>
888 * write.c (relax_segment): Use was_address instead of address when
889 setting fr_fix field for align frag due to backwards .org.
891 2004-07-29 Alexandre Oliva <aoliva@redhat.com>
893 Introduce SH2a support.
894 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
895 * config/tc-sh.c (get_specific): Change arch_sh2a_up to
897 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
898 * config/tc-sh.c (md_parse_option): Add sh2a-nofpu ISA handling.
899 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
900 * config/tc-sh.c (sh_elf_final_processing): Move sh2a recognition
901 to end of conditional expression.
902 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
903 * config/tc-sh.c: Add sh2a-nofpu support.
904 2003-12-29 DJ Delorie <dj@redhat.com>
905 * tc-sh.c: Add sh2a support.
906 (parse_reg): Add tbr.
907 (parse_at): Support @@(disp,tbr).
908 (get_specific): Support sh2a opcodes.
909 (insert4): New, for 4 byte relocs.
910 (build_Mytes): Support sh2a opcodes.
911 (md_apply_fix3_Mytes): Support sh2a opcodes.
912 2003-12-02 Michael Snyder <msnyder@redhat.com>
913 * config/tc-sh.c (md_parse_option): Handle sh2a.
914 (sh_elf_final_processing): Ditto.
916 2004-07-27 Jason Thorpe <thorpej@wasabisystems.com>
918 * config/tc-hppa.h (TARGET_FORMAT): Set to "elf32-hppa-netbsd"
921 2004-07-27 Alan Modra <amodra@bigpond.net.au>
923 * config/tc-ppc.c (ppc_frob_file_before_adjust): Warn if .toc too big.
924 (ppc_arch): Expand comment.
926 2004-07-27 Tomer Levi <Tomer.Levi@nsc.com>
928 * config/tc-crx.c: Support evaluating the difference between two
930 * config/tc-crx.h: Likewise.
932 2004-07-26 H.J. Lu <hongjiu.lu@intel.com>
934 * config/tc-ia64.c (start_unwind_section): Set the linked-to
936 (ia64_elf_section_change_hook): Set the linked-to section for
939 2004-07-26 Dmitry Diky <diwil@spec.ru>
941 * config/tc-msp430.c: Add new subtargets: msp430x1610,
942 msp430x1611, msp430x1612, msp430x415, msp430x417, msp430xG437,
943 msp430xG438, msp430xG439.
945 2004-07-25 Daniel Jacobowitz <dan@debian.org>
947 * doc/as.texinfo (Section, PushSection): Correct documentation
950 2004-07-21 Jan Beulich <jbeulich@novell.com>
952 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
953 permissible for the selected instruction suffix.
954 (match_template): Don't permit 64-bit general purpose operands in
956 (finalize_imm): Permit 64-bit immediates.
957 (build_modrm_byte): Don't treat 32-bit addressing in 64-bit mode
958 specially except for the width of the used base and/or index
959 registers. For 32-bit displacements, use sign-extended
960 relocations only when using 64-bit addressing.
961 Force zero displacement on rip-relative addressing when there is
962 no other displacement.
963 (i386_index_check): Don't treat 32-bit addressing in 64-bit mode
964 specially except for the width of the used base and/or index
966 (parse_register): Disallow Reg64 registers in 32-bit mode.
968 * config/tc-i386.c: For DefaultSize instructions, don't guess a 'q'
969 suffix if the instruction doesn't support it.
971 2004-07-20 Maciej W. Rozycki <macro@linux-mips.org>
973 * config/tc-mips.c (append_insn): Handle constant expressions with
974 no associated relocation.
975 (mips_ip): Cancel the expression after use for the Q format
977 (parse_relocation): Return no relocation for unsupported
979 (my_getSmallExpression): Return no relocation if no relocation
982 2004-07-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
984 * config/obj-som.c (adjust_stab_sections): Add prototype.
985 (obj_som_compiler, obj_som_version, obj_som_copyright,
986 adjust_stab_sections): Add ATTRIBUTE_UNUSED to unused arguments.
987 * config/tc-hppa.c (update_subspace): Likewise.
988 (is_defined_subspace): Amplify comment.
989 * config/obj-som.h (som_frob_file): Add prototype.
991 2004-07-19 Christopher Faylor <cgf@timesys.com>
992 H.J. Lu <hongjiu.lu@intel.com>
994 * subsegs.c (section_symbol): Don't create a new segment when
995 existing segment is undefined.
997 2004-07-16 Richard Earnshaw <rearnsha@arm.com>
999 * config/tc-arm.c: Include include/opcode/arm.h.
1000 (ARM_EXT_*, ARM_ARCH_*, ARM_ANY, ARM_ALL, COPROC_ANY): Delete.
1001 (FPU_FPA_EXT_* FPU_VFP_EXT_*, FPU_ANY, FPU_NONE, FPU_MAVERICK): Delete.
1002 (FPU_ARCH_*): Delete.
1003 * Makefile.am: Update dependencies.
1004 * Makefile.in: Regenerate.
1006 2004-07-15 Nick Clifton <nickc@redhat.com>
1008 * configure.in: Accept armbe as a big-endian arm configuration.
1009 * configure: Regenerate.
1011 2004-07-13 Thomas Nystrom <thn@saeab.se>
1013 * config/tc-i386.c (T_SHIFTOP): New constant.
1014 (intel_e05_1): Handle '&', '|' and T_SHIFTOP.
1015 (intel_el1): Handle '~'.
1016 (intel_get_token): Handle '<>', '&', '|' and '~'.
1018 2004-07-13 Nick Clifton <nickc@redhat.com>
1020 (md_assemble): Remove spurious newline from end of as_bad error
1022 (intel_e05_1): Likewise.
1023 (intel_e11): Likewise.
1024 (intel_match_token): Likewise.
1026 2004-07-11 Andreas Schwab <schwab@suse.de>
1028 * config/tc-m68k.c: Convert to C90. Remove redundant
1029 declarations. Indentation fixup.
1030 [M68KCOFF]: Include "obj-coff.h" instead of declaring
1031 obj_coff_section ourselves.
1033 2004-07-09 James E Wilson <wilson@specifixinc.com>
1035 * config/tc-ia64.c (default_big_endian): New.
1036 (dot_byteorder, md_begin): Use it.
1037 (md_parse_option): Set it.
1039 2004-07-09 Nick Clifton <nickc@redhat.com>
1041 * configure.in: Change sh-sybmian-elf to sh-*-symbianelf.
1042 * configure: Regenerate.
1043 * NEWS: Change sh-sybmian-elf to sh-*-symbianelf.
1044 * config/tc-sh.c (sh_elf_final_processing): Use renamed version of
1045 sh_find_elf_flags if necessary.
1047 2004-07-08 Richard Sandiford <rsandifo@redhat.com>
1049 * config/tc-mips.c (mips_fix_adjustable): If the full addend is
1050 going to be split into more than one in-place addend, return 0
1051 for relocations against mergeable sections. Associate comments
1054 2004-07-07 Tomer Levi <Tomer.Levi@nsc.com>
1056 * Makefile.am (CPU_TYPES): Add crx.
1057 (TARGET_CPU_CFILES): Add config/tc-crx.c.
1058 (TARGET_CPU_HFILES): Add config/tc-crx.h.
1059 (DEPTC_crx_elf): New target.
1060 (DEPOBJ_crx_elf): Likewise.
1061 (DEP_crx_elf): Likewise.
1062 * Makefile.in: Regenerate.
1063 * configure.in: Add crx* target.
1064 * configure: Regenerate.
1065 * config/tc-crx.c: New file.
1066 * config/tc-crx.h: New file.
1067 * NEWS: Mention new target.
1069 2004-07-06 Nick Clifton <nickc@redhat.com>
1071 * config.in: Undefine TARGET_SYMBIAN by default.
1073 * configure: Regenerate. Add sh-symbian-elf target. If
1074 selected define TARGET_SYMBIAN.
1075 * config/tc-sh.h (TARGET_FORMAT): Select a Symbian target
1076 format if TARGET_SYMBIAN has been defined.
1078 * output-file.c (output_file_create): Report the target format
1079 chosen when bfd_openw reports that it is invalid.
1081 * config/obj-coff.c (coff_pseudo_table): Only define the weak
1082 pseudo for BFD based assemblers.
1084 2004-07-05 Andrew Stubbs <andrew.stubbs@superh.com>
1087 * config/tc-sh.c (md_assemble): Change isspace to ISSPACE.
1088 (md_parse_option): Remove redundant -isa testing.
1089 Make bfd_arch variable const.
1090 (md_show_usage): Make bfd_arch variable const.
1092 2004-07-03 James E Wilson <wilson@specifixinc.com>
1094 * config/tc-ia64.c (emit_one_bundle): Check and set insn_addr.
1095 * config/tc-ia64.h (md_frag_check): Define.
1097 2004-07-03 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
1099 * config/obj-coff.c (obj_coff_weak): New .weak syntax for PE weak
1101 * doc/as.texinfo (Weak): Document PE weak symbols.
1103 2004-07-03 Richard Sandiford <rsandifo@redhat.com>
1105 * config/tc-mips.c (HAVE_IN_PLACE_ADDENDS): New macro.
1106 (reloc_needs_lo_p): Only return true if HAVE_IN_PLACE_ADDENDS.
1107 (mips_frob_file): Rework so that only a single pass through the
1108 relocs is needed. Allow %lo()s to have higher offsets than their
1109 corresponding %hi()s or %got()s.
1111 2004-07-02 Nick Clifton <nickc@redhat.com>
1113 * config/tc-arm.c (md_apply_fix3:BFD_RELOC_ARM_IMMEDIATE): Do not
1114 allow values which have come from undefined symbols.
1115 Always consider this fixup to have been processed as a reloc
1116 cannot be generated for it.
1118 2004-07-02 Alan Modra <amodra@bigpond.net.au>
1120 * frags.h (struct frag): Add has_code and insn_addr fields.
1121 * write.c (cvt_frag_to_fill): Invoke md_frag_check.
1122 * config/tc-ppc.c (md_assemble): Check and set insn_addr.
1123 * config/tc-ppc.h (md_frag_check): Define.
1125 2004-06-28 Maciej W. Rozycki <macro@linux-mips.org>
1127 * doc/Makefile.am (info): Rename goal to...
1128 (info-local): ... this, to preserve implicit dependencies.
1129 * doc/Makefile.in: Regenerate with automake 1.8.5.
1131 2004-06-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1133 * config/tc-m32r.c (md_convert_frag): Changed for @PLT.
1134 (m32r_cgen_record_fixup_exp): Changed for @GOTOFF, @GOT.
1135 (m32r_fix_adjustable): Changed for @GOTOFF, @GOT, @PLT.
1136 (tc_gen_reloc): Likewise.
1137 (m32r_end_of_match): Add for @GOTOFF, @GOT, @PLT.
1138 (m32r_parse_name): Likewise.
1139 (m32r_cgen_parse_fix_exp): Likewise.
1140 * config/tc-m32r.h (md_parse_name): Define for @GOTOFF, @GOT, @PLT.
1141 (O_PIC_reloc): Likewise.
1142 (TC_CGEN_PARSE_FIX_EXP): Likewise..
1143 * cgen.c (gas_cgen_parse_operand): Add TC_CGEN_PARSE_FIX_EXP
1144 for @GOTOFF, @GOT, @PLT.
1146 2004-06-21 Jan Beulich <jbeulich@novell.com>
1148 * gas/symbols.c: While discarding ordinary local absolute symbols
1149 when --strip-local-absolute is in effect, retain file symbols.
1151 2004-06-20 Andreas Schwab <schwab@suse.de>
1153 * config/tc-m68k.c (mri_chip): Replace current_chip, not augment.
1154 (md_parse_option): Likewise.
1156 2004-06-17 Jan Beulich <jbeulich@novell.com>
1158 * config/tc-i386.c: Deal with LEX_QM the same way as with LEX_AT.
1159 * config/te-netware.h: New file.
1160 * config/te-ppcnw.h: Delete: Obsolete.
1161 * configure.in: Eliminate ill NetWare targets. Make generic
1162 NetWare target use proper emulation.
1163 * Makefile.am: Eliminate reference to obsolete te-ppcnw.h, add
1164 reference to new te-netware.h.
1165 * configure: Regenerate.
1166 * Makefile.in: Regenerate.
1168 2004-06-15 Martin Schwidefsky <schwidefsky@de.ibm.com>
1170 * config/tc-s390.c (s390_insn): Avoid incorrect signed/unsigned
1171 comparison in .insn pseudo operation.
1173 2004-06-15 Alan Modra <amodra@bigpond.net.au>
1175 * config/obj-coff.c (coff_adjust_section_syms): Use
1176 bfd_get_section_size instead of bfd_get_section_size_before_reloc.
1177 (coff_frob_section): Likewise.
1178 * config/tc-mips.c (md_apply_fix3): Likewise.
1179 * config/obj-elf.c (elf_frob_file): Use bfd_set_section_size.
1180 (elf_frob_file_after_relocs): Likewise.
1182 2004-06-10 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1184 * config/tc-hppa.c (log2): Rename to exact_log2.
1185 (pa_next_subseg): Delete unused function.
1186 (create_new_space): Mark unused arguments with ATTRIBUTE_UNUSED.
1187 (create_new_subspace): Likewise.
1190 * config/tc-hppa.c (hppa_fix_adjustable): Allow reduction of fake
1191 labels. Fix warning.
1193 2004-05-28 DJ Delorie <dj@redhat.com>
1195 * config/tc-mn10300.h (tc_fix_adjustable): Define.
1196 * config/tc-mn10300.c (mn10300_fix_adjustable): Don't adjust debug
1197 or non-merged symbols.
1199 2004-05-28 H.J. Lu <hongjiu.lu@intel.com>
1201 * config/tc-ia64.c (remove_marked_resource): Save, clear and
1202 restore the old slot when inserting srlz.i/srlz.d.
1204 2004-05-28 Andrew Stubbs <andrew.stubbs@superh.com>
1206 * Makefile.am: Regenerate dependecies.
1207 * Makefile.in: Regenerate.
1208 * config/tc-sh.c (valid_arch): Make unsigned.
1209 (preset_target_arch): Likewise.
1210 (md_begin): Use new architecture flags system.
1211 (get_specific): Likewise.
1212 (assemble_ppi): Likewise.
1213 (md_assemble): Likewise. Also fix error check for bad opcodes.
1214 (md_parse_option): Likewise. Also generate -isa values according
1215 to the table in bfd/cpu-sh.c instead of just constants. Also
1216 allow <arch>-up ISA variants.
1217 (sh_elf_final_processing): Replace if-else chain with a call to
1218 sh_find_elf_flags().
1219 * testsuite/gas/sh/arch: New directory.
1220 * testsuite/gas/sh/arch/arch.exp: New test script.
1221 * testsuite/gas/sh/arch/arch_expected.txt: New file.
1222 * testsuite/gas/sh/arch/sh.s: New file.
1223 * testsuite/gas/sh/arch/sh2.s: New file.
1224 * testsuite/gas/sh/arch/sh-dsp.s: New file.
1225 * testsuite/gas/sh/arch/sh2e.s: New file.
1226 * testsuite/gas/sh/arch/sh3-nommu.s: New file.
1227 * testsuite/gas/sh/arch/sh3.s: New file.
1228 * testsuite/gas/sh/arch/sh3-dsp.s: New file.
1229 * testsuite/gas/sh/arch/sh3e.s: New file.
1230 * testsuite/gas/sh/arch/sh4-nommu-nofpu.s: New file.
1231 * testsuite/gas/sh/arch/sh4-nofpu.s: New file.
1232 * testsuite/gas/sh/arch/sh4.s: New file.
1233 * testsuite/gas/sh/arch/sh4a-nofpu.s: New file.
1234 * testsuite/gas/sh/arch/sh4al-dsp.s: New file.
1235 * testsuite/gas/sh/arch/sh4a.s: New file.
1237 2004-05-28 Peter Barada <peter@the-baradas.com>
1239 * config/gc-m68k.c(m68k_ip): Convert mode 5 addressing
1240 with zero offset into mode 2 addressing to save a word.
1242 2004-05-27 H.J. Lu <hongjiu.lu@intel.com>
1244 * config/tc-ia64.c (ar_is_in_integer_unit): Removed.
1245 (ar_is_only_in_integer_unit): New.
1246 (ar_is_only_in_memory_unit): New.
1247 (generate_unwind_image): Silence gcc on 32bit host.
1248 (md_assemble): Use ar_is_only_in_integer_unit instead of
1249 ar_is_in_integer_unit. Check AR access.
1251 2004-05-27 Peter Barada <peter@the-baradas.com>
1253 * config/tc-m68k.c (md_begin): Sort the opcode table into
1255 (m68k_compare_opcode): New function to do the sorting.
1257 2004-05-24 Peter Barada <peter@the-baradas.com>
1259 * config/m68k-parse.y(operand): Allow for MAC/EMAC mask
1260 addressing on MIT style operands.
1261 * config/m68k-parse.y(yylex): Allow '-&' for predecrement
1263 * config/tc-m68k.c(install_operand): Comment 'G' and 'H' type
1266 2004-05-23 Alan Modra <amodra@bigpond.net.au>
1268 * expr.c (operand, operator): Don't reject '++' and '--'.
1270 2004-05-20 Richard Sandiford <rsandifo@redhat.com>
1272 * config/tc-mips.c (append_insn): Use ISA-encoded addresses in MIPS16
1275 2004-05-17 Adam Nemet <anemet@lnxw.com>
1277 * configure.in: Add ppc-*-lynxos*. Update i386-*-lynxos* to ELF.
1278 * configure: Regenerate.
1280 2004-05-13 Paul Brook <paul@codesourcery.com>
1282 * dw2gencfi.c (output_cie): Handle dwarf3 format CIE entries.
1284 2004-05-13 Joel Sherrill <joel@oarcorp.com>
1286 * configure.in (or32-*-rtems*): Switch to elf.
1287 * configure: Regenerate.
1289 2004-05-13 Nick Clifton <nickc@redhat.com>
1291 * po/fr.po: Updated French translation.
1293 2004-05-11 Nick Clifton <nickc@redhat.com>
1295 * doc/as.texinfo (Section): Document G and T flags to .section
1296 directive. Document the extra arguments that the G flag
1297 requires. Document the #tls flag.
1299 2004-05-11 H.J. Lu <hongjiu.lu@intel.com>
1301 * subsegs.c (section_symbol): Create a new section symbol if
1302 the existing one doesn't match.
1303 * symbols.c (symbol_set_bfdsym): Don't reset BFD section symbol.
1305 2004-05-07 Richard Sandiford <rsandifo@redhat.com>
1307 * config/tc-mips.c (append_insn, mips_emit_delays): Extend -mfix-vr4120
1308 to cope with VR4181A errata MD(1) and MD(4).
1310 2004-05-07 Brian Ford <ford@vss.fsi.com>
1312 * NEWS: Mention .secrel32 for pe[i]-i386.
1314 2004-05-07 Alexandre Oliva <aoliva@redhat.com>
1316 * config/tc-frv.h (MAX_MEM_FOR_RS_ALIGN_CODE): New.
1317 (HANDLE_ALIGN): New.
1319 2004-05-06 Daniel Jacobowitz <dan@debian.org>
1321 * Makefile.am (DIST_SUBDIRS): Define.
1322 * aclocal.m4: Regenerate with automake 1.8.4.
1323 * Makefile.in: Likewise.
1324 * doc/Makefile.in: Likewise.
1326 2004-05-06 David Mosberger-Tang <davidm@hpl.hp.com>
1328 * config/tc-ia64.c (dot_serialize): Declare.
1329 (dot_serialize): New function.
1330 (md_pseudo_table): Add ".serialize.data" and
1331 ".serialize.instruction" directives.
1333 2004-05-06 Nick Clifton <nickc@redhat.com>
1335 * messages (as_internal_value_out_of_range): Print a message about
1336 a value being out of range. Be consistent about whether the
1337 values are printed in decimal or hexadecimal.
1338 (as_warn_value_out_of_range): Generate a warning message about an
1340 (as_bad_value_out_of_range): Generate an error message about an
1342 * as.h: Prototype the new functions.
1343 * config/tc-alpha.c (insert_operand): Use new function.
1344 * config/tc-arc.c (arc_insert_operand): Likewise.
1345 * config/tc-mn10200.c (mn10200_insert_operand): Likewise.
1346 * config/tc-mn10300.c (mn10300_insert_operand): Likewise.
1347 * config/tc-ppc.c (ppc_insert_operand): Likewise.
1348 * config/tc-s390.c (s390_insert_operand): Likewise.
1349 * config/tc-v850.c (v850_insert_operand): Likewise.
1351 2004-05-05 Alexandre Oliva <aoliva@redhat.com>
1353 * configure.in: Set em=linux for frv-*-*linux*.
1354 * configure: Rebuilt.
1355 * config/tc-frv.h (TARGET_FORMAT): Use elf32-frvfdpic if...
1356 (frv_md_fdpic_enabled): New.
1357 * config/tc-frv.c (frv_md_fdpic_enabled): New.
1358 (DEFAULT_FDPIC): New.
1359 (frv_flags): Use DEFAULT_FDPIC.
1360 (frv_pic_flag): Likewise.
1361 (OPTION_NOPIC): New.
1362 (md_longopts): Add -mnopic.
1363 (md_parse_option): Handle it.
1364 (md_show_usage): Add -mfdpic and -mnopic.
1366 2004-05-05 Peter Barada <peter@the-baradas.com>
1368 * config/tc-m68k.c: Add find_cf_chip to print list of valid
1369 chips for invalid coldfire instructions, rename selectors
1370 for ColdFire sub-variants, add 521x,5249,547x,548x and aliases,
1371 add current_chip to track which chip is referred to(including save/restore),
1372 use current_chip to select control registers, not current_arch.
1373 (md_show_usage): Add new chips.
1374 * doc/c-m68k.texi: Document new command line switches.
1376 2004-05-05 Jakub Jelinek <jakub@redhat.com>
1378 * tc-s390.h (md_do_align, HANDLE_ALIGN): Remove.
1379 (NOP_OPCODE): Define.
1380 (s390_align_code): Remove prototype.
1381 * tc-s390.c (s390_align_code): Remove.
1383 2004-05-04 H.J. Lu <hongjiu.lu@intel.com>
1385 * config/tc-ia64.c (make_unw_section_name): Removed.
1386 (start_unwind_section): New function.
1387 (generate_unwind_image): Take const segT instead of const
1389 (dot_handlerdata): Adjusted.
1390 (dot_endp): Likewise.
1392 2004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1394 * config/obj-elf.c (obj_elf_change_section): Allow the
1395 ".note.GNU-stack" section has SHF_EXECINSTR.
1397 2004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1399 * config/obj-elf.c (get_section): Return bfd_boolean.
1400 (obj_elf_change_section): Call bfd_get_section_by_name_if
1401 instead of bfd_map_over_sections.
1403 2004-04-30 H.J. Lu <hongjiu.lu@intel.com>
1405 * config/obj-elf.c (get_section): New function.
1406 (obj_elf_change_section): Support multiple sections with same
1409 2004-04-30 Nick Clifton <nickc@redhat.com>
1411 * config/tc-arm.c (create_register_alias): Fix typo checking for
1412 case sensitive register aliases.
1413 (co_proc_number): Use error message string in all_reg_maps[]
1415 (cp_reg_required_here): Likewise.
1416 (fp_reg_required_here): Likewise.
1418 2004-04-29 Brian Ford <ford@vss.fsi.com>
1420 * dwarf2dbg.c (dwarf2_finish): Add SEC_DEBUGGING to section flags.
1422 2004-04-28 Chris Demetriou <cgd@broadcom.com>
1424 * config/tc-mips.c (HAVE_32BIT_ADDRESSES, append_insn, macro_build)
1425 (load_address, macro, mips_ip, md_parse_option)
1426 (mips_force_relocation, mips_validate_fix, md_apply_fix3)
1427 (s_change_sec, pic_need_relax, tc_gen_reloc): Remove all
1428 embedded-PIC handling, and update comments.
1429 (SWITCH_TABLE): Remove.
1430 * config/tc-mips.h (DIFF_EXPR_OK): Delete.
1431 (enum mips_pic_level): Remove EMBEDDED_PIC.
1432 (EXTERN_FORCE_RELOC): Remove embedded-PIC handling.
1433 (TC_FORCE_RELOCATION): Update comment.
1434 * ecoff.c (ecoff_build_lineno): Add comment about some code that
1435 might be safe to remove now that MIPS embedded-PIC is gone.
1437 2004-04-28 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1439 * config/obj-som.c (obj_som_init_stab_section): Add new arguments in
1440 call to obj_set_subsection_attributes.
1441 (obj_som_init_stab_section): Likewise.
1442 * config/tc-hppa.c (default_subspace_dict): Add comdat field.
1443 (pa_def_subspaces): Provide comdat default.
1444 (pa_subspace): Handle new "comdat" parameter. Set SEC_LINK_ONCE and
1445 not SEC_IS_COMMON if section is comdat, common or dup_common. Update
1446 calls to create_new_subspace and update_subspace to pass comdat flag.
1447 (create_new_subspace, update_subspace): Add new comdat argument. Use
1448 it in calls to obj_set_subsection_attributes.
1449 * doc/c-hppa.texi (.subspa, .nsubspa): Document new comdat parameter
1450 and use of comdat, common and dup_comm parameters.
1452 2004-04-26 H.J. Lu <hongjiu.lu@intel.com>
1454 * config/obj-elf.c (obj_elf_change_section): Check if the old
1455 group name is NULL before comparison.
1457 2004-04-23 Chris Demetriou <cgd@broadcom.com>
1459 * config/tc-mips.h (mips_dwarf2_addr_size): Prototype.
1461 2004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1463 * config/tc-mips.c (s_mipsset): Set default CPU type for .set mipsN.
1465 2004-04-23 Chris Demetriou <cgd@broadcom.com>
1467 * config/tc-mips.c (md_longopts): Remove -membedded-pic option.
1468 (OPTION_MEMBEDDED_PIC): Remove.
1469 (OPTION_TRAP, OPTION_BREAK, OPTION_EB, OPTION_EL)
1470 (OPTION_FP32, OPTION_GP32, OPTION_CONSTRUCT_FLOATS)
1471 (OPTION_NO_CONSTRUCT_FLOATS, OPTIONS_FP64, OPTION_GP64)
1472 (OPTION_RELAX_BRANCH, OPTION_NO_RELAX_BRANCH)
1473 (OPTION_ELF_BASE): Renumber.
1474 (md_parse_option): Remove OPTION_MEMBEDDED_PIC handling.
1475 (md_show_usage): Remove mention of -membedded-pic.
1476 * doc/as.texinfo: Remove mention of -membedded-pic.
1478 2004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1480 * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Remove.
1481 * config/tc-mips.c (RDATA_SECTION_NAME, mips_target_format): Remove
1483 (md_begin, mips_ip, md_parse_option, s_change_sec, s_option,
1484 s_abicalls, nopic_need_relax, tc_gen_reloc): Remove uses of
1485 USE_GLOBAL_POINTER_OPT.
1487 2004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1489 * config/tc-mips.c (macro): One more use of load_delay_nop.
1491 2004-04-22 Atsushi Nemoto <anemo@mba.ocn.ne.jp>
1493 * config/tc-mips.c (load_delay_nop): New function.
1494 (load_address, macro): Use load_delay_nop() to build a nop
1495 which can be omitted with gpr_interlocks.
1497 2004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1499 * config/tc-mips.c (hilo_interlocks, gpr_interlocks,
1500 cop_interlocks): Remove superfluous CPU entries.
1502 2004-04-22 Paul Brook <paul@codesourcery.com>
1504 * config/tc-arm.c (mav_parse_offset): Value must be multiple of 4.
1506 2004-04-22 Peter Barada <peter@the-baradas.com>
1508 * NEWS: Added support for EMAC instructions and MAC/EMAC
1510 * config/m68k-parse.h: Add ACC[123], ACCEXT{01,23}, MAC/EMAC
1511 scale factor tokens, trailing_ampersand to mark mask addressing
1512 for MAC/EMAC instructions.
1513 * config/m68k-parse.y: Add options_ampersand clause, '<<',
1515 (yylex): Handle '>', '<', and '&' following '+'.
1516 * config/tc-m68k.c: Set mcfmac/mcfemac on appropriate ColdFire
1517 architectures in archs[].
1518 (m68k-ip): Add '4', 'e', 'g', 'i', cases to handle mask addressing
1519 for MAC/EMAC instructions, ACC[0123], ACCEXT{01,23}, and '<<'/'>>'
1521 (m68k_ip): Handle trailing '&' on MAC/EMAC insns.
1522 (install_operand): Fix 'n' case, Add 'F', 'f', 'G', 'H', 'I', ']'
1524 Add EMAC operands to init_table[].
1526 2004-04-22 Bruno De Bus <bdebus@elis.ugent.be>
1528 * config/tc-arm.h (enum mstate): Move here, add MAP_UNDEFINED
1530 (TC_SEGMENT_INFO_TYPE): Define to enum mstate.
1531 * config/tc-arm.c (enum mstate): Delete from here.
1532 (mapping_state): Remove the static mapstate variable and instead
1533 store the state in the segment. This allows a per-section mapping
1534 state. Handle and ignore MAP_UNDEFINED states.
1535 (arm_elf_change_section): Get the current mapping state from the
1537 (s_ltorg): Set the mapping state to MAP_DATA.
1538 (arm_cleanup): Use arm_elf_change_section to get the mapping state
1539 for each pool as it is emitted.
1541 2004-04-22 Nick Clifton <nickc@redhat.com>
1543 * config/tc-arm.h: Formatting tidy ups.
1545 2004-04-20 Chris Demetriou <cgd@broadcom.com>
1547 * NEWS: Note that MIPS -membedded-pic option is deprecated.
1549 2004-04-20 DJ Delorie <dj@redhat.com>
1551 * config/tc-i386.h [TE_PE] (TC_CONS_FIX_NEW): Define.
1552 * config/tc-i386.c (md_pseudo_table) [TE_PE]: Add "secrel32".
1553 [TE_PE] (O_secrel): Define.
1554 [TE_PE] (x86_pe_cons_fix_new): New.
1555 [TE_PE] (pe_directive_secrel): Likewise.
1556 (tc_gen_reloc) [TE_PE]: Support BFD_RELOC_32_SECREL.
1558 2004-04-19 Eric Christopher <echristo@redhat.com>
1560 * config/tc-mips.c (mips_dwarf2_addr_size): Revert part
1561 of previous patch for fix in gcc.
1563 2004-04-19 Jakub Jelinek <jakub@redhat.com>
1565 * config/tc-xtensa.c (xg_assembler_literal): Fix a typo.
1567 2004-04-19 Nathan Sidwell <nathan@codesourcery.com>
1569 * read.c (do_align): Call md_flush_pending_output, if defined.
1571 2004-04-16 Alan Modra <amodra@bigpond.net.au>
1573 * expr.c (operand): Correct checks for ++ and --.
1575 2004-04-14 H.J. Lu <hongjiu.lu@intel.com>
1577 * config/tc-generic.c: Add some comments.
1579 2004-04-14 Richard Sandiford <rsandifo@redhat.com>
1581 * doc/c-mips.texi (-m{no-,}fix-vr4120): Renamed from
1582 -{no-}mfix-vr4122-bugs.
1583 * config/tc-mips.c (mips_fix_vr4120): Renamed from mips_fix_4122_bugs.
1584 (append_insn, mips_emit_delays): Update accordingly.
1585 (OPTION_FIX_VR4120, OPTION_NO_FIX_VR4120): Renamed from *VR4122.
1586 (md_longopts): Change -{no-,}mfix-vr4122-bugs to -m{no-,}fix-vr4120.
1587 (md_parse_option): Update after above changes.
1588 (md_show_usage): Add -mfix-vr4120.
1590 2004-04-13 Bob Wilson <bob.wilson@acm.org>
1592 * doc/as.texinfo (Sub-Sections): Conditionalize COFF-specific use
1593 of .section directive; add a reference to the ELF .subsection
1596 2004-04-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1598 * config/tc-m32r.c (md_assemble): Fixed infinite loop bug
1601 2004-04-11 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1603 * Makefile.am: Remove mips from aout targets.
1604 * Makefile.in: Regenerate.
1605 * configure.in: Remove mips-dec-bsd* target.
1606 * configure: Regenerate.
1608 2004-04-07 Alan Modra <amodra@bigpond.net.au>
1611 * config/tc-ppc.c (ppc_elf_suffix): Add valid32 and valid64 fields
1612 to struct map_bfd. Adjust MAP macro, and define MAP32, MAP64.
1613 Update "mapping". Restrict some @ modifiers to 32 bit.
1615 2004-04-01 Asgari Jinia <asgarij@kpitcummins.com>
1616 Dhananjay Deshpande <dhananjayd@kpitcummins.com>
1618 * config/tc-sh.c (dont_adjust_reloc_32): New variable.
1619 (sh_fix_adjustable): Avoid adjusting BFD_RELOC_32 when
1620 dont_adjust_reloc_32 is set.
1621 (md_longopts): Add option -renesas.
1622 (md_parse_option, md_show_usage): Likewise.
1623 * doc/c-sh.texi: Likewise.
1625 2004-04-01 Dave Korn <dk@artimi.com>
1627 * config/tc-dlx.c (md_assemble): set fx_no_overflow flag for
1628 hi16 and lo16 fixS structs.
1629 (md_assemble): generate bit_fixS for RELOC_DLX_LO16 in
1630 exactly the same way as for RELOC_DLX_REL16.
1631 (machine_ip): properly respect LO flag in the_insn and
1632 output RELOC_DLX_LO16 rather than RELOC_DLX_16.
1633 (md_apply_fix3): apply RELOC_DLX_LO16.
1635 2004-03-30 Stan Shebs <shebs@apple.com>
1637 Remove long-obsolete MPW support.
1638 * mpw-config.in, mpw-make.sed, mac-as.r: Remove files.
1639 * configure.in: Remove mention of ppc-*-mpw* config.
1640 * configure.in: Likewise.
1642 2004-03-30 Nick Clifton <nickc@redhat.com>
1644 * config/tc-arm.c (meabi_flags): Make its use conditional upon
1645 OBJ_ELF being defined.
1647 2004-03-27 Alan Modra <amodra@bigpond.net.au>
1649 * config/obj-aout.c (obj_aout_type): Remove #ifdef BFD_ASSEMBLER code.
1651 2004-03-23 Paul Brook <paul@codesourcery.com>
1653 * config/tc-arm.c (meabi_flags): New variable.
1654 (arm_parse_eabi): New function.
1655 (md_begin): Set flags for EABI v3.
1657 (arm_long_opts): Add meabi.
1658 * doc/as.texinf <ARM>: Document -meabi.
1659 * doc/c-arm.texi: Ditto.
1661 2004-03-22 Bob Wilson <bob.wilson@acm.org>
1663 * config/tc-xtensa.c (xtensa_post_relax_hook): Create literal
1664 tables even when use_literal_section flag is not set.
1666 2004-03-22 Alan Modra <amodra@bigpond.net.au>
1668 * config/tc-sh.c: Remove trailing whitespace.
1670 2004-03-22 Hans-Peter Nilsson <hp@axis.com>
1672 * doc/c-cris.texi (CRIS-Opts): Document --no-mul-bug-abort,
1673 --mul-bug-abort and the default behavior.
1674 * config/tc-cris.c (cris_insn_kind): New member CRIS_INSN_MUL.
1675 (err_for_dangerous_mul_placement): New variable.
1676 (STATE_MUL, OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): New
1678 (md_cris_relax_table): Have placeholder for STATE_MUL.
1679 (md_longopts): New options --mul-bug-abort and --no-mul-bug-abort.
1680 (cris_relax_frag) <case ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: New
1682 (md_estimate_size_before_relax) <case ENCODE_RELAX (STATE_MUL,
1683 STATE_BYTE)>: Ditto.
1684 (md_convert_frag) <ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: Check
1685 alignment and position of this frag, emit error message if
1687 (md_assemble): For a multiply insn and when checking it,
1688 transform the current frag into a special frag for that purpose.
1689 (md_parse_option) <case OPTION_MULBUG_ABORT_OFF, case
1690 OPTION_MULBUG_ABORT_ON>: Handle new options.
1692 2004-03-19 Bob Wilson <bob.wilson@acm.org>
1694 * config/tc-xtensa.c (mark_literal_frags): New function.
1695 (xtensa_move_literals): Call mark_literal_frags for all literal
1696 segments, including init and fini literal segments.
1697 (xtensa_post_relax_hook): Swap use of xt_insn_sec and xt_literal_sec.
1699 2004-03-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1701 * tc-hppa.c (cons_fix_new_hppa): Check for PC relative base type.
1702 (pa_comm): Set BSF_OBJECT in symbol flags.
1704 2004-03-19 Alan Modra <amodra@bigpond.net.au>
1706 * Makefile.am: Run "make dep-am".
1707 * Makefile.in: Regenerate.
1708 * doc/Makefile.in: Regenerate.
1709 * config.in: Regenerate.
1710 * po/gas.pot: Regenerate.
1712 2004-03-18 Nathan Sidwell <nathan@codesourcery.com>
1714 * read.c (read_a_source_file): Use demand_empty_rest_of_line.
1715 (demand_empty_rest_of_line): Issue an error here.
1716 (ignore_rest_of_line): Silently skip to end.
1717 (demand_copy_string): Issue an error, not warning.
1719 * config/obj-elf.c (obj_elf_section_name): Likewise.
1720 (obj_elf_section): Likewise.
1721 * config/tc-arc.c (arc_extoper): Remove bogus NULL checks.
1722 (arc_extinst): Likewise.
1723 * config/tc-ia64.c (dot_saveb): Use demand_empty_rest_of_line.
1724 (dot_spill): Likewise.
1725 (dot_unwabi): Likewise.
1726 (dot_prologue): Likewise.
1728 * expr.c (operand): Reject ++ and --.
1729 (operator): Likewise.
1731 2004-03-17 Kaz Kojima <kkojima@rr.iij4u.or.jp>
1733 * config/tc-sh.c: Include dw2gencfi.h.
1734 (sh_cfi_frame_initial_instructions): New function.
1735 (sh_regname_to_dw2regnum): Likewise.
1736 * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Move to the end of
1738 (TARGET_USE_CFIPOP): Define.
1739 (tc_cfi_frame_initial_instructions): Likewise.
1740 (tc_regname_to_dw2regnum): Likewise.
1741 (DWARF2_DEFAULT_RETURN_COLUMN, DWARF2_CIE_DATA_ALIGNMENT): Likewise.
1742 * Makefile.am: Update dependencies.
1743 * Makefile.in: Regenerate.
1745 2004-03-17 Ralf Corsepius <corsepiu@faw.uni-ulm.de>
1747 * configure.in: Switch sh-*-rtems* to ELF. Add sh-*-rtemscoff*.
1748 * configure: Regenerate.
1750 2004-03-12 Bob Wilson <bob.wilson@acm.org>
1752 * read.c (s_leb128): Call md_flush_pending_output.
1754 2004-03-12 Michal Ludvig <mludvig@suse.cz>
1756 * config/tc-i386.c (output_insn): Handle PadLock instructions.
1757 * config/tc-i386.h (CpuPadLock): New define.
1758 (CpuUnknownFlags): Added CpuPadLock.
1760 2004-03-07 Andreas Schwab <schwab@suse.de>
1762 * doc/c-hppa.texi (HPPA Directives): Fix typo.
1764 2004-03-07 Richard Henderson <rth@redhat.com>
1766 * dw2gencfi.c (output_cie): Align length to 4 byte boundary.
1767 (cfi_finish): Likewise for fde.
1769 2004-03-05 H.J. Lu <hongjiu.lu@intel.com>
1771 * config/tc-ia64.c (md_assemble): Properly handle NULL
1773 (ia64_handle_align): Don't abort if failed to add a stop bit.
1775 2004-03-04 H.J. Lu <hongjiu.lu@intel.com>
1777 * Makefile.in: Regenerated.
1778 * aclocal.m4: Likewise.
1779 * configure: Likewise.
1780 * doc/Makefile.in: Likewise.
1782 2004-03-03 H.J. Lu <hongjiu.lu@intel.com>
1784 * config/tc-ia64.c (dot_align): New.
1785 (ia64_do_align): Make it static.
1786 (md_pseudo_table): Use "dot_align" for "align".
1787 (ia64_md_do_align): Don't set align_frag here.
1788 (ia64_handle_align): Add a stop bit to the previous bundle if
1791 * config/tc-ia64.h (ia64_do_align): Removed.
1793 2003-03-03 Andrew Stubbs <andrew.stubbs@superh.com>
1795 * config/tc-sh.c (md_parse_option): Add -isa=sh4-nofpu and
1796 -isa=sh4-nommu-nofpu options. Adjust help messages accordingly.
1797 (sh_elf_final_processing): Output BFD type sh4_nofpu if that is
1798 the most general type or the user specifically requested it.
1799 (md_assemble): Add a new error message for when an instruction
1800 is understood, but is not allowed due to an -isa option.
1802 2004-03-02 H.J. Lu <hongjiu.lu@intel.com>
1804 * config/tc-ia64.c (align_frag): New.
1805 (md_assemble): Set the tc_frag_data field in align_frag for
1806 IA64_OPCODE_FIRST instructions.
1807 (ia64_md_do_align): Set align_frag.
1808 (ia64_handle_align): Add a stop bit if needed.
1810 * config/tc-ia64.h (TC_FRAG_TYPE): New.
1811 (TC_FRAG_INIT): New.
1813 2004-03-01 Richard Sandiford <rsandifo@redhat.com>
1815 * config/tc-frv.c (fr400_audio): New variable.
1816 (md_parse_option, md_show_usage): Add -mcpu=fr405 and -mcpu=fr450.
1817 (md_parse_option): Set fr400_audio for -mcpu=fr400 and -mcpu=fr405.
1818 (target_implements_insn_p): New function.
1819 (md_assemble): Report an error if the processor doesn't implement
1822 2004-02-27 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1824 * config/tc-m32r.c (md_longopts): Added -no-bitinst option.
1825 (md_parse_option): Ditto.
1826 (OPTION_NO_SPECIAL_M32R): Added.
1827 (md_show_usage): Document it.
1828 (enable_speial_m32r): Changed a default value from 0 to 1.
1829 * doc/c-m32r.texi: Document the -no-bitinst option.
1831 2004-02-27 Nick Clifton <nickc@redhat.com>
1833 * config/tc-sh.c (get_operand): Revert previous delta.
1834 (tc_gen_reloc): Check for an unknown reloc type before processing
1837 2004-02-27 Hannes Reinecke <hare@suse.de>
1839 * config/tc-s390.c (s390_insn): Correct range check for opcode in
1840 .insn pseudo operation.
1842 2004-02-27 Anil Paranjpe <anilp1@kpitcummins.com>
1844 * config/tc-sh.c (get_operand): In case of #Imm, check has been
1845 added for wrong syntax.
1847 2004-02-26 Eric Christopher <echristo@redhat.com>
1849 * config/tc-mips.c (mips_dwarf2_addr_size): New.
1850 * config/tc-mips.h (DWARF2_ADDR_SIZE): Use.
1852 2004-02-26 Andrew Stubbs <andrew.stubbs@superh.com>
1854 * config/tc-sh.c (build_Mytes): Add REG_N_D and REG_N_B01
1855 nibble types to assembler.
1857 2004-02-25 Fred Fish <fnf@redhat.com>
1859 * config/tc-iq2000.c: Add missing \n\ in multiline string literal.
1861 2004-02-20 James E Wilson <wilson@specifixinc.com>
1863 * config/tc-ia64.c (slot_index): New arg before_relax. Use instead of
1865 (fixup_unw_records): New arg before_relax. Pass to slot_index.
1866 (ia64_estimate_size_before_relax): New.
1867 (ia64_convert_frag): Pass 0 to fixup_unw_records. Add comment.
1868 (generate_unwind_image): Pass 1 to fixup_unw_records.
1869 * config/tc-ia64.h (ia64_estimate_size_before_relax): Declare.
1870 (md_estimate_size_before_relax): Call ia64_estimate_size_before_relax.
1872 2004-02-19 Jakub Jelinek <jakub@redhat.com>
1874 * stabs.c (generate_asm_file): Avoid warning about use of
1875 uninitialized variable.
1877 2004-02-18 David Mosberger <davidm@hpl.hp.com>
1879 * config/tc-ia64.c (ia64_flush_insns): In addition to prologue,
1880 body, and endp, allow unwind records which do not have a "t"
1881 (time/instruction) field.
1883 2004-02-17 Petko Manolov <petkan@nucleusys.com>
1885 * config/tc-arm.c (do_mav_dspsc_1): Correct offset of CRn.
1886 (do_mav_dspsc_2): Likewise.
1887 Fix accumulator registers move opcodes.
1889 2004-02-13 Hannes Reinecke <hare@suse.de>
1890 Jakub Jelinek <jakub@redhat.com>
1892 * dwarf2dbg.c (get_filenum): Do not read beyond allocated memory.
1894 2004-02-10 Steve Ellcey <sje@cup.hp.com>
1896 * config/tc-ia64.h (ia64_frob_symbol): New declaration.
1897 (tc_frob_symbol): New macro definition.
1898 * config/tc-ia64.c (ia64_frob_symbol): New routine.
1900 2004-02-09 Daniel Jacobowitz <drow@mvista.com>
1902 * config/tc-arm.c (md_begin): Mark .note.gnu.arm.ident as
1905 2004-02-09 Nathan Sidwell <nathan@codesourcery.com>
1907 * read.h (IGNORE_OPCODE_CASE): Do not define. Replace with ...
1908 (TC_CASE_SENSITIVE): ... this.
1909 * read.c: Replace IGNORE_OPCODE_CASE with TC_CASE_SENSITIVE.
1910 * doc/internals.texi (TC_CASE_SENSITIVE): Document.
1912 2004-02-06 James E Wilson <wilson@specifixinc.com>
1914 * config/tc-ia64.c (dot_endp): Delete call to output_endp.
1915 (generate_unwind_image): Re-add it here.
1917 2004-02-06 Nathan Sidwell <nathan@codesourcery.com>
1919 * dwarf2dbg.c (DWARF2_ADDR_SIZE): Remove trailing ';'
1920 * read.h (SKIP_WHITESPACE): Turn into an expression.
1921 * read.c (read_a_source_file): A pseudo is removed by having a
1924 2004-02-05 James E Wilson <wilson@specifixinc.com>
1926 * config/tc-ia64.c (output_endp): New.
1927 (count_bits): Delete.
1928 (ia64_flush_insns, process_one_record, optimize_unw_records): Handle
1929 endp unwind records.
1930 (fixup_unw_records): Handle endp unwind records. Delete code for
1931 shortening prologue regions not followed by a body record.
1932 (dot_endp): Call add_unwind_entry to emit endp unwind record.
1933 * config/tc-ia64.h (unw_record_type): Add endp.
1935 2004-02-03 James E Wilson <wilson@specifixinc.com>
1937 * config/tc-ia64.c (ia64_convert_frag): Call md_number_to_chars to
1938 fill padding bytes with zeroes.
1939 (emit_one_bundle): New locals last_ptr, end_ptr. Rewrite code that
1940 sets unwind_record slot_number and slot_frag fields.
1942 2004-02-02 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
1944 * config/tc-mips.c (add_got_offset_hilo): New function.
1945 (macro): Use load_register() and add_got_offset_hilo() to load
1946 constants instead of hardcoding code sequences throughout.
1948 2004-01-28 H.J. Lu <hongjiu.lu@intel.com>
1950 * config/tc-ia64.c (emit_one_bundle): Add proper indentation.
1952 2004-01-26 Bernardo Innocenti <bernie@develer.com>
1954 * config/tc-m68k.h (EXTERN_FORCE_RELOC): Handle m68k-uclinux specially,
1956 * config/tc-m68k.c (RELAXABLE_SYMBOL): Use EXTERN_FORCE_RELOC instead
1957 of hard-coded test for TARGET_OS=elf.
1959 2004-01-24 Chris Demetriou <cgd@broadcom.com>
1961 * config/tc-mips.c (hilo_interlocks): Change definition
1962 so that MIPS32, MIPS64 and later ISAs are included, along with
1963 the already-included machines. Update comments.
1965 2004-01-23 Daniel Jacobowitz <drow@mvista.com>
1967 * config/tc-arm.c (tc_gen_reloc): Improve error message for
1968 undefined local labels.
1970 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
1972 * config/tc-mips.c (load_address, macro): Update comments about
1973 NewABI GP relaxation.
1975 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
1977 * config/tc-mips.c (macro_build): Remove place and counter arguments.
1978 (mips_build_lui, macro_build_ldst_constoffset): Likewise.
1979 (mips16_macro_build, macro_build_jalr): Remove counter argument.
1980 (set_at, load_register, load_address, move_register): Likewise.
1981 (load_got_offset, add_got_offset): Likewise.
1982 Update all calls and tidy accordingly.
1984 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
1986 * config/tc-mips.c (RELAX_ENCODE): Remove WARN argument.
1987 (RELAX_FIRST, RELAX_SECOND): Turn into 8-bit quantities.
1988 (RELAX_USE_SECOND): Bump to 0x10000.
1989 (RELAX_SECOND_LONGER, RELAX_NOMACRO, RELAX_DELAY_SLOT): New flags.
1990 (mips_macro_warning): New variable.
1991 (md_assemble): Wrap macro expansion in macro_start() and macro_end().
1992 (s_cpload, s_cpsetup, s_cprestore, s_cpreturn): Likewise.
1993 (relax_close_frag): Set mips_macro_warning.first_frag. Adjust use
1995 (append_insn): Update mips_macro_warning.sizes.
1996 (macro_start, macro_warning, macro_end): New functions.
1997 (macro_build): Don't emit warnings here.
1998 (macro_build_lui, md_estimate_size_before_relax): ...or here.
1999 (md_convert_frag): Check for cases where one macro alternative
2000 needs a warning and the other doesn't. Emit a warning if the
2001 longer sequence was chosen.
2003 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2005 * config/tc-mips.h (tc_frag_data_type, TC_FRAG_TYPE): Remove.
2006 * config/tc-mips.c (RELAX_ENCODE): Take three arguments: the size of
2007 the first sequence, the size of the second sequence, and a flag
2008 that says whether we should warn.
2009 (RELAX_OLD, RELAX_NEW, RELAX_RELOC[123]): Delete.
2010 (RELAX_FIRST, RELAX_SECOND): New.
2011 (mips_relax): New variable.
2012 (relax_close_frag, relax_start, relax_switch, relax_end): New fns.
2013 (append_insn): Remove "place" argument. Use mips_relax.sequence
2014 rather than "place" to check whether we're expanding the second
2015 alternative of a relaxable macro. Remove redundant check for
2016 branch relaxation. If generating a normal insn, and there
2017 is not enough room in the current frag, call relax_close_frag()
2018 to close it. Update mips_relax.sizes[]. Emit fixups for the
2019 second version of a relaxable macro. Record the first relaxable
2020 fixup in mips_relax. Remove tc_gen_reloc workaround.
2021 (macro_build): Remove all uses of "place". Use mips_relax.sequence
2022 in the same way as in append_insn.
2023 (mips16_macro_build): Remove "place" argument.
2024 (macro_build_lui): As for macro_build. Don't drop the add_symbol
2025 when generating the second version of a relaxable macro.
2026 (load_got_offset, add_got_offset): New functions.
2027 (load_address, macro): Use new relaxation machinery. Remove
2028 tc_gen_reloc workarounds.
2029 (md_estimate_size_before_relax): Set RELAX_USE_SECOND if the second
2030 version of a relaxable macro is needed. Return -RELAX_SECOND if the
2031 first version is needed.
2032 (tc_gen_reloc): Remove relaxation handling.
2033 (md_convert_frag): Go through the fixups for a relaxable macro and
2034 mark those that belong to the unneeded alternative as done. If the
2035 second alternative is needed, adjust the fixup addresses to account
2036 for the deleted first alternative.
2038 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2040 * frags.h (frag_room): Declare.
2041 * frags.c (frag_room): New function.
2042 * doc/internals.texi: Document it.
2044 2004-01-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
2046 * config/tc-mips.c (append_insn): Don't do r3900 interlock
2047 optimization for -mtune=r3900, as this will break on other CPUs.
2049 2004-01-11 Tom Rix <tcrix@worldnet.att.net>
2051 * config/tc-m68hc11.c (build_indexed_byte): movb and movw cannot
2052 be relaxed, use fixup.
2053 (md_apply_fix3): Use 5 bit reloc from movb and movw fixup.
2055 2004-01-19 Jakub Jelinek <jakub@redhat.com>
2057 * config/tc-sparc.c (sparc_ip): Disallow %f32-%f63 for single
2060 2004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2062 * config/tc-mips.c (append_insn): Properly detect variant frags
2063 that preclude swapping of relaxed branches. Correctly swap
2064 instructions between frags when dealing with relaxed branches.
2066 2004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2068 * acinclude.m4: Quote names of macros to be defined by AC_DEFUN
2070 * aclocal.m4: Regenerate.
2071 * configure: Regenerate.
2073 2004-01-12 Anil Paranjpe <anilp1@KPITCummins.com>
2075 * config/tc-h8300.c (build_bytes): Apply relaxation to bit
2078 2004-01-12 Richard Sandiford <rsandifo@redhat.com>
2080 * config/tc-mips.c (macro_build_jalr): When adding an R_MIPS_JALR
2081 reloc, reserve space for the delay slot as well as the jalr itself.
2083 2004-01-09 Paul Brook <paul@codesourcery.com>
2085 * config/tc-arm.c (do_vfp_reg2_from_sp2): Rename from do_vfp_sp_reg2.
2086 (do_vfp_sp2_from_reg2): New function.
2088 (do_vfp_dp_from_reg2): Check return values properly.
2090 2004-01-08 Ian Lance Taylor <ian@wasabisystems.com>
2092 * config/tc-mips.c (warn_nops): Remove static variable.
2093 (macro): Remove test of warn_nops.
2094 (md_shortops): Remove 'n'.
2095 (md_parse_option): Remove 'n' case.
2096 (md_show_usage): Remove -n.
2097 * doc/as.texinfo (Overview): Remove MIPS -n option.
2098 * doc/c-mips.texi (MIPS Opts): Remove mention -n.
2099 * NEWS: Mention removal of MIPS -n option.
2101 * config/tc-mips.c (ISA_HAS_COPROC_DELAYS): Remove.
2102 (cop_interlocks): Check ISA level.
2103 (cop_mem_interlocks): Define.
2104 (reg_needs_delay): Check cop_interlocks rather than
2105 ISA_HAS_COPROC_DELAYS.
2106 (append_insn): Likewise. Use cop_mem_interlocks rather than
2107 directly checking mips_opts.isa.
2108 (mips_emit_delays): Likewise.
2110 2004-01-07 H.J. Lu <hongjiu.lu@intel.com>
2112 * config/tc-ia64.c (unwind): Move next_slot_number and
2113 next_slot_frag to ...
2114 (unw_rec_list): Here.
2115 (free_list_records): Removed.
2116 (output_unw_records): Likewise.
2117 (generate_unwind_image): Make it void.
2118 (alloc_record): Initialize next_slot_number and next_slot_frag.
2119 (slot_index): Take .org, .space and .align into account.
2120 (fixup_unw_records): Don't set slot_number to 0. Use
2121 list->next_slot_number and list->next_slot_frag instead of
2122 unwind.next_slot_number and unwind.next_slot_frag.
2123 (ia64_convert_frag): New.
2124 (generate_unwind_image): Generate a rs_machine_dependent frag
2126 (emit_one_bundle): Use list->next_slot_number and
2127 list->next_slot_frag instead of unwind.next_slot_number and
2128 unwind.next_slot_frag.
2130 * config/tc-ia64.h (md_convert_frag): Defined as
2132 (md_estimate_size_before_relax): Defined as (f)->fr_var.
2134 2004-01-06 Alexandre Oliva <aoliva@redhat.com>
2136 2003-12-19 Alexandre Oliva <aoliva@redhat.com>
2137 * config/tc-frv.h (md_apply_fix3): Don't define.
2138 * config/tc-frv.c (md_apply_fix3): New. Shift/truncate %hi/%lo
2140 * config/tc-frv.h (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
2141 2003-10-07 Alexandre Oliva <aoliva@redhat.com>
2142 * config/tc-frv.c (line_separator_chars): Add `!'.
2143 2003-09-19 Alexandre Oliva <aoliva@redhat.com>
2144 * config/tc-frv.c (md_assemble): Clear insn upfront.
2145 2003-09-18 Alexandre Oliva <aoliva@redhat.com>
2146 * config/tc-frv.c (OPTION_FDPIC): New macro.
2147 (md_longopts): Add mfdpic.
2148 (md_parse_option): Handle it.
2149 2003-08-04 Alexandre Oliva <aoliva@redhat.com>
2150 * config/tc-frv.c (md_cgen_lookup_reloc) <FRV_OPERAND_D12,
2151 FRV_OPERAND_S12>: Use reloc type encoded in fix-up.
2152 (frv_pic_ptr): Parse funcdesc.
2154 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2156 * doc/as.texinfo: Let texi2pod parse asconfig.texi and
2157 gasver.texi. Remove duplicate symbol definitions for texi2pod.
2159 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2161 * Makefile.am (Makefile): Move the dependency on
2162 $(BFDDIR)/configure.in to...
2163 (CONFIG_STATUS_DEPENDENCIES): ... here.
2164 (AUTOMAKE_OPTIONS): Require automake 1.8.
2165 * Makefile.in: Regenerate.
2166 * doc/Makefile.am (BASEDIR, BFDDIR): Define.
2167 (CONFIG_STATUS_DEPENDENCIES): Add a dependency on
2168 $(BFDDIR)/configure.in here as well.
2169 * doc/Makefile.in: Regenerate.
2171 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2173 * Makefile.am (install, install-info, RECURSIVE_TARGETS): Remove.
2174 * Makefile.in: Regenerate.
2175 * aclocal.m4: Regenerate.
2176 * doc/Makefile.am (install, install-info): Remove.
2177 (install-data-local): A new hook for install-info.
2178 (AUTOMAKE_OPTIONS): Require automake 1.8.
2179 * doc/Makefile.in: Regenerate.
2181 2004-01-02 Nutan Singh <nutan@kpitcummins.com>
2183 * doc/c-sh.texi: Update description about floating point behavior
2186 2004-01-02 Bernardo Innocenti <bernie@develer.com>
2188 * configure.in: Add m68k-uClinux target.
2189 * configure: Regenerate.
2191 For older changes see ChangeLog-0203
2197 version-control: never