1 2004-11-05 Sterling Augustine <sterling@tensilica.com>
3 * config/tc-xtensa.c (total_frag_text_expansion): New.
4 (md_estimate_size_before_relax): Use it.
5 (find_address_of_next_align_frag): Likewise.
7 2004-11-05 Tomer Levi <Tomer.Levi@nsc.com>
9 * config/tc-crx.c: Rename argument types.
10 (processing_arg_number): Rename to 'cur_arg_num'.
11 (get_number_of_bits): Rename to 'set_operand_size'.
12 (get_operandtype): Rename to 'parse_operand', totally rewrite.
13 (set_cons_rparams): Rename to 'set_operand', totally rewrite.
14 (set_indexmode_parameters): Remove function, integrate its code into
16 (set_operand_size): Get rid of 'Operand Number' function parameter -
17 use global variable 'cur_arg_num' instead.
18 Use a local 'argument' pointer to reference the current argument.
19 (parse_operand): Likewise.
20 (set_operand): Likewise.
21 (process_label_constant): Likewise.
23 2004-11-04 Bob Wilson <bob.wilson@acm.org>
25 * config/tc-xtensa.c: Remove XTENSA_SECTION_RENAME ifdefs.
26 (add_section_rename): Delete. Inlined into...
27 (build_section_rename): ...here. Use xstrdup instead of strdup.
28 (xtensa_section_rename): Drop "const" from argument and return types.
29 (md_show_usage): Indent to match show_usage().
30 * config/tc-xtensa.h: Remove XTENSA_SECTION_RENAME ifdefs.
31 (tc_canonicalize_section_name): Define.
32 (md_elf_section_rename): Remove unused macro.
33 * doc/as.texinfo (Overview): Document Xtensa --rename-section option.
34 * doc/c-xtensa.texi (Xtensa Options): Likewise.
35 (Frame Directive): Delete.
37 2004-11-04 Daniel Jacobowitz <dan@debian.org>
39 * configure.in: Remove arm-*-oabi and thumb-*-oabi.
40 * config/tc-arm.c (target_oabi): Delete.
41 (md_apply_fix3, elf32_arm_target_format): Remove target_oabi checks.
42 (arm_opts): Remove moabi.
43 * doc/as.texinfo (Overview): Remove documentation of -moabi.
44 * doc/c-arm.texi (ARM Options): Likewise.
45 * configure: Regenerated.
47 2004-11-04 Hans-Peter Nilsson <hp@axis.com>
49 * configure.in (crisv32): Recognize. AC_DEFINE_UNQUOTED
50 DEFAULT_CRIS_ARCH. Handle crisv32-*-linux-gnu* like
51 cris-*-linux-gnu* and crisv32-*-* like cris-*-*.
52 * configure: Regenerate.
53 * config/tc-cris.c (enum cris_archs): New.
54 (cris_mach, cris_arch_from_string, s_cris_arch, get_sup_reg)
55 (cris_insn_ver_valid_for_arch): New functions.
56 (DEFAULT_CRIS_ARCH): New macro, default to cris_any_v0_v10.
57 (cris_arch): New variable.
58 (md_pseudo_table): New pseudo .arch.
59 (err_for_dangerous_mul_placement): Initialize according to
61 (STATE_COND_BRANCH): Renamed from STATE_CONDITIONAL_BRANCH.
63 (STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON)
64 (STATE_ABS_BRANCH_V32, STATE_LAPC, BRANCH_BF_V32, BRANCH_BB_V32)
65 (BRANCH_WF_V32, BRANCH_WB_V32): New.
66 (BRANCH_BF, BRANCH_BB, BRANCH_WF, BRANCH_WB): Don't undef after
67 use in md_cris_relax_table.
68 (md_cris_relax_table): Add entries for STATE_COND_BRANCH_V32,
69 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
70 Update and improve head comment.
71 (OPTION_PIC): Define in terms of previous option, OPTION_US.
72 (OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): Similar.
74 (md_longopts): New option --march=...
75 (cris_any_v0_v10_long_jump_size, crisv32_long_jump_size): New
77 (md_long_jump_size): Initialize in terms of DEFAULT_CRIS_ARCH.
78 (HANDLE_RELAXABLE): New macro.
79 (md_estimate_size_before_relax): Use HANDLE_RELAXABLE for common
80 cases. Check for weak symbols and assume not relaxable. Handle
81 STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON,
82 STATE_ABS_BRANCH_V32, STATE_LAPC. Use new variable symbolP, not
84 (md_convert_frag): Handle STATE_COND_BRANCH_V32,
85 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
86 (cris_create_short_jump): Adjust for CRISv32.
87 (cris_relax_frag): Handle new states.
88 (md_create_long_jump): Ditto. Emit error for common_v10_v32.
89 (md_begin): Define symbols "..asm.arch.cris.v32",
90 "..asm.arch.cris.v10", "..asm.arch.cris.common_v10_v32" and
91 "..asm.arch.cris.any_v0_v10". Use cris_insn_ver_valid_for_arch
92 when entering opcode table entry points.
93 (md_assemble): Adjust branch handling for CRISv32. Handle LAPC
94 relaxation. In fix_new_exp call for main insn, pass 1 for pcrel
95 parameter for 8, 16 and 32-bit pc-relative insns and LAPC.
96 (cris_process_instruction): Initialize out_insnp->insn_type to
97 CRIS_INSN_NONE, not CRIS_INSN_NORMAL.
98 <case ']', '[', 'A', 'd', 'Q', 'N', 'n', 'Y', 'U', 'u', 'T'>: New
100 <case 'm'>: Check that modified_char == '.'.
101 <invalid operands>: Consume the rest of the line.
102 When operands don't match, skip over subsequent insns with
103 non-matching version specifier but same mnemonic.
104 <immediate constant, case SIZE_SPEC_REG>: Immediate operands for
105 special registers in CRISv32 are always 32 bit long.
106 <immediate constant, case SIZE_FIELD_SIGNED, SIZE_FIELD_UNSIGNED>:
108 (get_gen_reg): Only recognize "PC" when followed by "+]" for v32
109 and compatible. Recognize "ACR" for v32, unless followed by "+".
110 (get_spec_reg): Consider cris_arch when looking up register.
111 (get_autoinc_prefix_or_indir_op): Don't recognize assignment for
113 (get_3op_or_dip_prefix_op): Check for ']' after seeing '[rN+'.
114 (cris_get_expression): Restore input_line_pointer if failing "early".
115 (get_flags): Consider cris_arch and recognize flags accordingly.
116 (branch_disp): Adjust for CRISv32.
117 (gen_cond_branch_32): Similar. Emit error for common_v10_v32.
118 (cris_number_to_imm): Use as_bad_where, not as_bad. Remove
119 related FIXME. Don't insist on BFD_RELOC_32_PCREL fixup to be
120 resolved. Don't enter zeros in object file for
122 <case BFD_RELOC_CRIS_LAPCQ_OFFSET, BFD_RELOC_CRIS_SIGNED_16>
123 <case BFD_RELOC_CRIS_SIGNED_8>: New case.
124 (md_parse_option): Break out "return 1".
125 <OPTION_ARCH> New case.
126 (tc_gen_reloc): <case BFD_RELOC_CRIS_LAPCQ_OFFSET>
127 <case BFD_RELOC_CRIS_SIGNED_16, BFD_RELOC_CRIS_SIGNED_8>
128 <case BFD_RELOC_CRIS_UNSIGNED_8, BFD_RELOC_CRIS_UNSIGNED_16>
129 <case BFD_RELOC_32_PCREL>: New cases.
130 Addends for non-zero fx_pcrel are too in fx_offset.
131 (md_show_usage): Show --march=<arch>.
132 (md_apply_fix3): Adjust val for BFD_RELOC_CRIS_LAPCQ_OFFSET.
133 (md_pcrel_from): BFD_RELOC_CRIS_LAPCQ_OFFSET is PC-relative too.
134 (s_syntax) <struct syntaxes>: Properly constify member operand.
135 * config/tc-cris.h (TARGET_MACH): Define.
136 (cris_mach): Declare.
137 * doc/as.texinfo (Overview) <CRIS>: Add --march=...
138 * doc/c-cris.texi (CRIS-Symbols): New node for built-in symbols.
139 (CRIS-Opts): Document --march=...
140 (CRIS-Pseudos): Document .arch.
142 2004-11-04 Jan Beulich <jbeulich@novell.com>
144 * config/tc-i386.c (set_intel_syntax): Allow % in symbol names when
145 intel syntax and no register prefix, allow $ in symbol names when
147 (set_16bit_gcc_code_flag): Replace literal 'l' by LONG_MNEM_SUFFIX.
148 (intel_float_operand): Add fourth return value indicating math control
149 operations. Make classification more precise.
150 (md_assemble): Complain if memory operand of mov[sz]x has no size
152 (parse_insn): Translate word operands to floating point instructions
153 operating on integers as well as control instructions to short ones
154 as expected by AT&T syntax. Translate 'd' suffix to short one only for
155 floating point instructions operating on non-integer operands.
156 (match_template): Remove fldcw special case. Adjust q-suffix handling
157 to permit it on fild/fistp/fisttp in AT&T mode.
158 (process_suffix): Don't guess DefaultSize insns' suffix from
159 stackop_size for certain floating point control instructions. Guess
160 suffix for branch and [ls][gi]dt based on flag_code. Split error
161 messages for Intel and AT&T syntax, and make the condition more strict
162 for the former. Adjust suppressing of generation of operand size
164 (intel parser): Allow the full set of MASM operators. Add FWORD, TBYTE,
165 OWORD, and XMMWORD operand size specifiers (TBYTE replaces XWORD). Add
167 * config/tc-i386.h (BYTE_PTR WORD_PTR DWORD_PTR QWORD_PTR XWORD_PTR
168 SHORT OFFSET_FLAT FLAT NONE_FOUND): Remove unused defines.
170 2004-11-03 Hans-Peter Nilsson <hp@axis.com>
172 * symbols.c (colon) [!WORKING_DOT_WORD]: Don't declare
173 md_short_jump_size, md_long_jump_size.
174 * write.c [!WORKING_DOT_WORD]: Ditto.
175 * tc.h [!WORKING_DOT_WORD]: Declare them here. Drop const
177 * config/tc-cris.h (md_short_jump_size, md_long_jump_size): Don't
179 * config/tc-cris.c (md_short_jump_size, md_long_jump_size): Drop
180 const qualifier in these definitions.
181 * config/tc-i370.c, config/tc-m68k.c, config/tc-pdp11.c,
182 config/tc-s390.c, config/tc-tahoe.c, config/tc-vax.c: Ditto.
184 2004-11-02 Nick Clifton <nickc@redhat.com>
186 * dwarf2dbg.c (dwarf2_finish): Check for the existence of a file
187 table before deciding to produce a .debug_line section to match up
188 with a user provided .debug_info section.
190 2004-10-28 Tomer Levi <Tomer.Levi@nsc.com>
192 * config/tc-crx.c (getreg_image): Bug fix, a return value was
193 mistakenly omitted from CRX_C_REGTYPE and CRX_CS_REGTYPE cases.
195 2004-10-27 Tomer Levi <Tomer.Levi@nsc.com>
197 * config/tc-crx.c: Remove global variable 'post_inc_mode'.
198 (get_flags): New function.
199 (get_number_of_bits): Edit comments, update numeric values to supported
201 (process_label_constant): Don't support the colon format (SYMBOL:[s|m|l]).
202 (set_cons_rparams): Support argument type 'arg_rbase'.
203 (get_operandtype): Bug fix in 'rbase' operand type parsing.
204 (handle_LoadStor): Bug fix, first handle post-increment mode.
205 (getreg_image): Remove redundant code, update according to latest CRX spec.
206 (print_constant): Bug fix relate to 3-word instructions.
207 (assemble_insn): Bug fix, when matching instructions, verify also
208 instruction type (not only mnemonic).
209 Add various error checking.
210 (preprocess_reglist): Support HI/LO and user registers.
212 2004-10-25 David Mosberger-Tang <davidm@hpl.hp.com>
214 * config/tc-ia64.c (fixup_unw_records): Don't let the "t" value
215 in an epilogue directive go negative.
217 2004-10-25 H.J. Lu <hongjiu.lu@intel.com>
220 * config/tc-ia64.c (emit_one_bundle): Decrement md.num_slots_in_use
221 after reporting template error during manual bundling. Reported
222 by Michael Dupont, michaelx.dupont@intel.com.
224 2004-10-25 Daniel Jacobowitz <dan@debian.org>
226 * Makefile.am: Run dep-am.
227 * aclocal.m4: Regenerate with automake 1.9.2.
228 * Makefile.in: Regenerate with automake 1.9.2.
229 * doc/Makefile.in: Likewise.
231 * config/tc-arm.c: Include "dw2gencfi.h".
232 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
234 * config/tc-arm.h (TARGET_USE_CFIPOP, DWARF2_DEFAULT_RETURN_COLUMN)
235 (DWARF2_CIE_DATA_ALIGNMENT, tc_regname_to_dw2regnum)
236 (tc_cfi_frame_initial_instructions): Define.
237 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
240 2004-10-21 Tomer Levi <Tomer.Levi@nsc.com>
242 * config/tc-crx.c (assemble_insn): Check unsigned immediate
244 Update coprocessor id to be unsigned immediate.
246 2004-10-18 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
248 * config/tc-i386.c (O_secrel): Delete.
249 (tc_pe_dwarf2_emit_offset): New function.
250 * config/tc-i386.h (O_secrel): Define as O_md1.
251 (TC_DWARF2_EMIT_OFFSET): Define.
253 2004-10-18 Nick Clifton <nickc@redhat.com>
255 * config/tc-xstormy16.c (xstormy16_cons_fix_new): Accept and
256 ignore @fptr() directives for 4-byte fixups.
258 2004-10-15 Alan Modra <amodra@bigpond.net.au>
260 * Makefile.am: Run "make dep-am"
261 * Makefile.in: Regenerate.
263 2004-10-14 Bob Wilson <bob.wilson@acm.org>
265 * doc/c-xtensa.texi (Xtensa Options, Absolute Literals Directive):
266 Remove comments about placement of literal pools.
267 (Literal Directive): Update description of literal placement.
268 (Literal Prefix Directive): Remove statement that this does not apply
269 to absolute-mode literals. Describe new section naming scheme.
271 2004-10-12 Bob Wilson <bob.wilson@acm.org>
273 * config/tc-xtensa.c (xtensa_frequency_pseudo): Use set_subseg_freq.
274 (is_entry_opcode, is_movi_opcode, is_the_loop_opcode, is_jx_opcode,
275 is_windowed_return_opcode): Delete.
276 (xtensa_frob_label): Use get_subseg_target_freq.
277 (md_assemble): Inline call to is_entry_opcode.
278 (xtensa_handle_align): Inline call to get_frag_is_literal.
279 (relaxation_requirements): Inline call to is_jx_opcode.
280 (emit_single_op): Inline call to is_movi_opcode.
281 (xg_assemble_vliw_tokens): Inline calls to get_frag_is_insn,
282 get_frag_is_no_transform, is_entry_opcode, and
283 set_frag_is_specific_opcode. Use get_subseg_total_freq.
284 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags,
285 xtensa_fix_close_loop_end_frags, relax_frag_immed, convert_frag_immed):
286 Inline calls to get_frag_is_no_transform.
287 (next_instrs_are_b_retw): Inline call to is_windowed_return_opcode.
288 (xtensa_fix_short_loop_frags): Inline calls to is_the_loop_opcode and
289 get_frag_is_no_transform.
290 (convert_frag_immed_finish_loop): Inline calls to get_expression_value
291 and set_frag_is_no_transform.
292 (get_expression_value): Delete.
293 (subseg_map struct): Rename cur_total_freq to total_freq. Rename
294 cur_target_freq to target_freq.
295 (get_subseg_info): Split out code to create a new map entry into ...
296 (add_subseg_info): ... this new function.
297 (get_last_insn_flags): Check if get_subseg_info succeeded.
298 (set_last_insn_flags): Call add_subseg_info if needed.
299 (get_subseg_total_freq, get_subseg_target_freq, set_subseg_freq): New.
300 (xtensa_reorder_segments): Compute last_sec while counting sections.
301 Remove call to get_last_sec.
302 (get_last_sec): Delete.
303 (cache_literal_section): Inline call to retrieve_literal_seg and its
304 callees, seg_present and add_seg_list.
305 (retrieve_literal_seg, seg_present, add_seg_list): Delete.
306 (get_frag_is_insn, get_frag_is_no_transform,
307 set_frag_is_specific_opcode, set_frag_is_no_transform): Delete.
308 * config/tc-xtensa.h (MAX_SLOTS): Reduce from 31 to 15.
310 2004-10-12 Bob Wilson <bob.wilson@acm.org>
312 * config/tc-xtensa.c: Use ISO C90 formatting.
313 * config/tc-xtensa.h: Likewise.
314 * config/xtensa-istack.h: Likewise.
315 * config/xtensa-relax.c: Likewise.
316 * config/xtensa-relax.h: Likewise.
318 2004-10-12 Paul Brook <paul@codesourcery.com>
320 * config/tc-arm.c (md_begin): Change EF_ARM_EABI_VER3 to
323 * doc/c-arm.texi: Document that we actually support -meabi=4, not
326 2004-10-08 Bob Wilson <bob.wilson@acm.org>
328 * doc/as.texinfo (VTableEntry, VTableInherit): Add "directive" to index
330 (Acknowledgements): Use "GAS" instead of AS variable.
332 2004-10-08 Daniel Jacobowitz <dan@debian.org>
334 * config/tc-i386.c: Include "elf/x86-64.h".
335 (i386_elf_section_type): New function.
336 * config/tc-i386.h (md_elf_section_type): Define.
337 (i386_elf_section_type): New prototype.
339 2004-10-08 Linus Nielsen Feltzing <linus@haxx.se>
341 * config/m68k-parse.h (enum m68k_register): New control register,
343 * config/tc-m68k.c: Correct control register set for MCF5249.
345 2004-10-07 Bob Wilson <bob.wilson@acm.org>
346 Sterling Augustine <sterling@tensilica.com>
348 * config/tc-xtensa.c (absolute_literals_supported): New global flag.
349 (UNREACHABLE_MAX_WIDTH): Define.
350 (XTENSA_FETCH_WIDTH): Delete.
351 (cur_vinsn, xtensa_fetch_width, xt_saved_debug_type, past_xtensa_end,
352 prefer_const16, prefer_l32r): New global variables.
353 (LIT4_SECTION_NAME): Define.
354 (lit4_state struct): Add lit4_seg_name and lit4_seg fields.
355 (XTENSA_PROP_*, GET_XTENSA_PROP_*, SET_XTENSA_PROP_*): Define.
356 (frag_flags struct): New.
357 (xtensa_block_info struct): Move from tc-xtensa.h. Add flags field.
358 (subseg_map struct): Add cur_total_freq and cur_target_freq fields.
359 (bitfield, bit_is_set, set_bit, clear_bit): Define.
360 (MAX_FORMATS): Define.
361 (op_placement_info struct, op_placement_table): New.
362 (O_pltrel, O_hi16, O_lo16): Define.
363 (directiveE enum): Rename directive_generics to directive_transform.
364 Delete directive_relax. Add directive_schedule,
365 directive_absolute_literals, and directive_last_directive.
366 (directive_info): Rename "generics" to "transform". Delete "relax".
367 Add "schedule" and "absolute-literals".
368 (directive_state): Adjust entries to match changes in directive_info.
369 (xtensa_relax_statesE, RELAX_IMMED_MAXSTEPS): Move to tc-xtensa.h.
370 (xtensa_const16_opcode, xtensa_movi_opcode, xtensa_movi_n_opcode,
371 xtensa_l32r_opcode, xtensa_nop_opcode, xtensa_rsr_lcount_opcode): New.
372 (xtensa_j_opcode, xtensa_rsr_opcode): Delete.
373 (align_only_targets, software_a0_b_retw_interlock,
374 software_avoid_b_j_loop_end, maybe_has_b_j_loop_end,
375 software_avoid_short_loop, software_avoid_close_loop_end,
376 software_avoid_all_short_loops, specific_opcode): Delete.
377 (warn_unaligned_branch_targets): New.
378 (workaround_a0_b_retw, workaround_b_j_loop_end, workaround_short_loop,
379 workaround_close_loop_end, workaround_all_short_loops): Default FALSE.
380 (option_[no_]link_relax, option_[no_]transform,
381 option_[no_]absolute_literals, option_warn_unaligned_targets,
382 option_prefer_l32r, option_prefer_const16, option_target_hardware):
384 (option_[no_]align_only_targets, option_literal_section_name,
385 option_text_section_name, option_data_section_name,
386 option_bss_section_name, option_eb, option_el): Delete.
387 (md_longopts): Add entries for: [no-]transform, [no-]absolute-literals,
388 warn-unaligned-targets, prefer-l32r, prefer-const16, [no-]link-relax,
389 and target-hardware. Delete entries for [no-]target-align-only,
390 literal-section-name, text-section-name, data-section-name, and
392 (md_parse_option): Handle new options and remove old ones. Accept but
393 ignore [no-]density options. Warn for [no-]generics and [no-]relax
394 and treat them as [no-]transform.
395 (md_show_usage): Add new options and remove old ones.
396 (xtensa_setup_hw_workarounds): New.
397 (md_pseudo_table): Change "word" entry to use xtensa_elf_cons. Add
398 "long", "short", "loc" and "frequency" entries.
399 (use_generics): Rename to ...
400 (use_transform): ... this function. Add past_xtensa_end check.
401 (use_longcalls): Add past_xtensa_end check.
402 (code_density_available, can_relax): Delete.
403 (do_align_targets): New.
404 (get_directive): Accept dashes in directive names. Warn about
405 [no-]generics and [no-]relax directives and treat them as
407 (xtensa_begin_directive): Call md_flush_pending_output only for some
408 directives. Check for directives inside instruction bundles. Warn
409 about deprecated ".begin literal" usage. Warn and ignore [no-]density
410 directives. Handle new directives. Check generating_literals flag
412 (xtensa_end_directive): Check for directives inside instruction
413 bundles. Warn and ignore [no-]density directives. Handle new
414 directives. Call xtensa_set_frag_assembly_state.
415 (xtensa_loc_directive_seen, xtensa_dwarf2_directive_loc,
416 xtensa_dwarf2_emit_insn): New.
417 (xtensa_literal_position): Call md_flush_pending_output. Do not check
418 use_literal_section flag.
419 (xtensa_literal_pseudo): Call md_flush_pending_output. Handle absolute
420 literals. Use xtensa_elf_cons to parse the expression.
421 (xtensa_literal_prefix): Do not check use_literal_section. Support
422 ".lit4" sections for absolute literals. Change prefix convention to
423 replace ".text" (or ".t" in a linkonce section). No need to call
425 (xtensa_frequency_pseudo, xtensa_elf_cons, xtensa_elf_suffix): New.
426 (expression_end): Handle closing braces and colons.
427 (PLT_SUFFIX, plt_suffix): Delete.
428 (expression_maybe_register): Use new xtensa-isa.h functions. Use
429 xtensa_elf_suffix instead of checking for plt suffix, and handle O_lo16
430 and O_hi16 expressions as well.
431 (tokenize_arguments): Handle closing braces and colons.
432 (parse_arguments): Use new xtensa-isa.h functions. Handle "invisible"
433 operands and paired register syntax.
434 (get_invisible_operands): New.
435 (xg_translate_sysreg_op): Handle new Xtensa LX RSR/WSR/XSR syntax. Use
436 new xtensa-isa.h functions.
437 (xtensa_translate_old_userreg_ops, xtensa_translate_zero_immed): New.
438 (xg_translate_idioms): Check if inside bundle. Use use_transform.
439 Handle new Xtensa LX RSR/WSR/XSR syntax. Remove code to widen density
440 instructions. Use xtensa_translate_zero_immed.
441 (operand_is_immed, operand_is_pcrel_label): Delete.
442 (get_relaxable_immed): Use new xtensa-isa.h functions.
443 (get_opcode_from_buf): Add slot parameter. Use new xtensa-isa.h
445 (xtensa_print_insn_table, print_vliw_insn): New.
446 (is_direct_call_opcode): Use new xtensa-isa.h functions.
447 (is_call_opcode, is_loop_opcode, is_conditional_branch_opcode,
448 is_branch_or_jump_opcode): Delete.
449 (is_movi_opcode, decode_reloc, encode_reloc, encode_alt_reloc): New.
450 (opnum_to_reloc, reloc_to_opnum): Delete.
451 (xtensa_insnbuf_set_operand, xtensa_insnbuf_get_operand): Use new
452 xtensa-isa.h functions. Operate on one slot of an instruction.
453 (xtensa_insnbuf_set_immediate_field, is_negatable_branch,
454 xg_get_insn_size): Delete.
455 (xg_get_build_instr_size): Use xg_get_single_size.
456 (xg_is_narrow_insn, xg_is_single_relaxable_insn): Update calls to
457 xg_build_widen_table. Use xg_get_single_size.
458 (xg_get_max_narrow_insn_size): Delete.
459 (xg_get_max_insn_widen_size, xg_get_max_insn_widen_literal_size,
460 xg_is_relaxable_insn): Update calls to xg_build_widen_table. Use
462 (xg_build_to_insn): Record the loc field. Handle OP_OPERAND_HI16U and
463 OP_OPERAND_LOW16U. Check xg_valid_literal_expression.
464 (xg_expand_to_stack, xg_expand_narrow): Update calls to
465 xg_build_widen_table. Use xg_get_single_size.
466 (xg_immeds_fit): Use new xtensa-isa.h functions. Update call to
468 (xg_symbolic_immeds_fit): Likewise. Also handle O_lo16 and O_hi16, and
469 treat weak symbols conservatively.
470 (xg_check_operand): Use new xtensa-isa.h functions.
471 (is_dnrange): Delete.
472 (xg_assembly_relax): Inline previous calls to tinsn_copy.
473 (xg_finish_frag): Specify separate relax states for the frag and slot0.
474 (is_branch_jmp_to_next, xg_add_branch_and_loop_targets): Use new
475 xtensa-isa.h functions.
476 (xg_instruction_matches_option_term, xg_instruction_matches_or_options,
477 xg_instruction_matches_options): New.
478 (xg_instruction_matches_rule): Handle O_register expressions. Call
479 xg_instruction_matches_options.
480 (transition_rule_cmp): New.
481 (xg_instruction_match): Update call to xg_build_simplify_table.
482 (xg_build_token_insn): Record loc fields.
483 (xg_simplify_insn): Check is_specific_opcode field and
484 density_supported flag.
485 (xg_expand_assembly_insn): Skip checking code_density_available. Use
486 new xtensa-isa.h functions. Call use_transform instead of can_relax.
487 (xg_assemble_literal): Add error handling for O_big. Call
488 record_alignment. Handle O_pltrel.
489 (xg_valid_literal_expression): New.
490 (xg_assemble_literal_space): Add slot parameter. Remove call to
491 set_expr_symbol_offset. Add call to record_alignment. Update call to
493 (xg_emit_insn): Delete.
494 (xg_emit_insn_to_buf): Add format parameter. Update calls to
495 xg_add_opcode_fix and xtensa_insnbuf_to_chars.
496 (xg_add_opcode_fix): Change opcode parameter to tinsn and add format
497 and slot parameters. Handle new "alternate" relocations for absolute
498 literals and CONST16 instructions. Check for bad uses of O_lo16 and
499 O_hi16. Use new xtensa-isa.h functions.
500 (xg_assemble_tokens): Delete.
501 (is_register_writer): Use new xtensa-isa.h functions.
502 (is_bad_loopend_opcode): Check for xtensa_rsr_lcount_opcode instead of
503 old-style RSR from LCOUNT.
504 (next_frag_opcode): Delete.
505 (next_frag_opcode_is_loop, next_frag_format_size, frag_format_size,
506 update_next_frag_state): New.
507 (update_next_frag_nop_state): Delete.
508 (next_frag_pre_opcode_bytes): Use next_frag_opcode_is_loop.
509 (xtensa_mark_literal_pool_location): Check use_literal_section flag and
510 the state of the absolute-literals directive. Add calls to
511 record_alignment and xtensa_set_frag_assembly_state. Call
512 xtensa_switch_to_non_abs_literal_fragment instead of
513 xtensa_switch_to_literal_fragment.
515 (assemble_nop): Use build_nop. Update call to xtensa_insnbuf_to_chars.
516 (get_expanded_loop_offset): Change check for undefined opcode to an
518 (xtensa_set_frag_assembly_state, relaxable_section,
519 xtensa_find_unmarked_state_frags, xtensa_find_unaligned_branch_targets,
520 xtensa_find_unaligned_loops, xg_apply_tentative_value): New.
521 (md_begin): Update call to xtensa_isa_init. Initialize linkrelax to 1.
522 Set lit4_seg_name. Call xg_init_vinsn. Initialize new global opcodes.
523 Call init_op_placement_info_table and xtensa_set_frag_assembly_state.
524 (xtensa_init_fix_data): New.
525 (xtensa_frob_label): Reset label symbol to the current frag. Check
526 do_align_targets and generating_literals flag. Propagate frequency
527 info to new alignment frag. Call xtensa_set_frag_assembly_state.
528 (xtensa_unrecognized_line): New.
529 (xtensa_flush_pending_output): Check if inside a bundle. Add a call
530 to xtensa_set_frag_assembly_state.
531 (error_reset_cur_vinsn): New.
532 (md_assemble): Remove check for literal frag. Remove call to
533 istack_init. Call use_transform instead of use_generics. Parse
534 explicit instruction format specifiers. Move code for
535 a0_b_retw_interlock workaround to xg_assemble_vliw_tokens. Call
536 error_reset_cur_vinsn on errors. Add call to get_invisible_operands.
537 Add dwarf2_where call. Remote automatic alignment for ENTRY
538 instructions. Move call to xtensa_clear_insn_labels to the end.
539 Rearrange to handle bundles.
540 (xtensa_cons_fix_new): Delete.
541 (xtensa_handle_align): New.
542 (xtensa_frag_init): Call xtensa_set_frag_assembly_state. Remove
543 assignment to is_no_density field.
544 (md_pcrel_from): Use new xtensa-isa.h functions. Use decode_reloc
545 instead of reloc_to_opnum. Handle "alternate" relocations.
546 (xtensa_force_relocation, xtensa_check_inside_bundle,
547 xtensa_elf_section_change_hook): New.
548 (xtensa_symbol_new_hook): Delete.
549 (xtensa_fix_adjustable): Check for difference of symbols with an
550 offset. Check for external and weak symbols.
551 (md_apply_fix3): Remove cases for XTENSA_OP{0,1,2} relocs.
552 (md_estimate_size_before_relax): Return expansion for the first slot.
553 (tc_gen_reloc): Handle difference of symbols by producing
554 XTENSA_DIFF{8,16,32} relocs and by writing the value of the difference
555 into the output. Handle new XTENSA_SLOT*_OP relocs by storing the
556 tentative values into the output when linkrelax is set.
557 (XTENSA_PROP_SEC_NAME): Define.
558 (xtensa_post_relax_hook): Call xtensa_find_unmarked_state_frags.
559 Create literal tables only if using literal sections. Create new
560 property tables instead of old instruction tables. Check for unaligned
561 branch targets and loops.
562 (finish_vinsn, find_vinsn_conflicts, check_t1_t2_reads_and_writes,
563 new_resource_table, clear_resource_table, resize_resource_table,
564 resources_available, reserve_resources, release_resources,
565 opcode_funcUnit_use_unit, opcode_funcUnit_use_stage,
566 resources_conflict, xg_find_narrowest_format, relaxation_requirements,
567 bundle_single_op, emit_single_op, xg_assemble_vliw_tokens): New.
568 (xtensa_end): Call xtensa_flush_pending_output. Set past_xtensa_end
569 flag. Update checks for workaround options. Call
570 xtensa_mark_narrow_branches and xtensa_mark_zcl_first_insns.
571 (xtensa_cleanup_align_frags): Add special case for branch targets.
572 Check for and mark unreachable frags.
573 (xtensa_fix_target_frags): Remove use of align_only_targets flag.
574 Use RELAX_LOOP_END_BYTES in special case for negatable branch at the
575 end of a zero-overhead loop body.
576 (frag_can_negate_branch): Handle instructions with multiple slots.
577 Use new xtensa-isa.h functions
578 (xtensa_mark_narrow_branches, is_narrow_branch_guaranteed_in_range,
579 xtensa_mark_zcl_first_insns): New.
580 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags): Error if
581 transformations are disabled.
582 (next_instrs_are_b_retw): Use new xtensa-isa.h functions. Handle
583 multislot instructions.
584 (xtensa_fix_close_loop_end_frags, xtensa_fix_short_loop_frags):
585 Likewise. Also error if transformations are disabled.
586 (unrelaxed_frag_max_size): New.
587 (unrelaxed_frag_min_insn_count, unrelax_frag_has_b_j): Use new
588 xtensa-isa.h functions.
589 (xtensa_sanity_check, is_empty_loop, is_local_forward_loop): Use
590 xtensa_opcode_is_loop instead of is_loop_opcode.
591 (get_text_align_power): Replace as_fatal with assertion.
592 (get_text_align_fill_size): Iterate instead of using modulus when
594 (get_noop_aligned_address): Assert that this is for a machine-dependent
595 RELAX_ALIGN_NEXT_OPCODE frag. Use next_frag_opcode_is_loop,
596 xg_get_single_size, and frag_format_size.
597 (get_widen_aligned_address): Rename to ...
598 (get_aligned_diff): ... this function. Add max_diff parameter.
599 Remove handling of rs_align/rs_align_code frags. Use
600 next_frag_format_size, get_text_align_power, get_text_align_fill_size,
601 next_frag_opcode_is_loop, and xg_get_single_size. Compute max_diff
602 and pass it back to caller.
603 (xtensa_relax_frag): Use relax_frag_loop_align. Add code for new
604 RELAX_SLOTS, RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN,
605 RELAX_FILL_NOP, and RELAX_UNREACHABLE frag types. Check relax_seen.
606 (relax_frag_text_align): Rename to ...
607 (relax_frag_loop_align): ... this function. Assume loops can only be
608 in the first slot of an instruction.
609 (relax_frag_add_nop): Use assemble_nop instead of constructing an OR
610 instruction. Remove call to frag_wane.
611 (relax_frag_narrow): Rename to ...
612 (relax_frag_for_align): ... this function. Extend to handle
613 RELAX_FILL_NOP and RELAX_UNREACHABLE, as well as RELAX_SLOTS with
614 RELAX_NARROW for the first slot.
615 (find_address_of_next_align_frag, bytes_to_stretch): New.
616 (future_alignment_required): Use find_address_of_next_align_frag and
617 bytes_to_stretch. Look ahead to subsequent frags to make smarter
619 (relax_frag_immed): Add format, slot, and estimate_only parameters.
620 Check if transformations are enabled for b_j_loop_end workaround.
621 Use new xtensa-isa.h functions and handle multislot instructions.
622 Update call to xg_assembly_relax.
623 (md_convert_frag): Handle new RELAX_SLOTS, RELAX_UNREACHABLE,
624 RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN, and RELAX_FILL_NOP
626 (convert_frag_narrow): Add segP, format and slot parameters. Call
627 convert_frag_immed for branch instructions. Adjust calls to
628 tinsn_from_chars, tinsn_immed_from_frag, and xg_emit_insn_to_buf. Use
629 xg_get_single_size and xg_get_single_format.
630 (convert_frag_fill_nop): New.
631 (convert_frag_immed): Add format and slot parameters. Handle multislot
632 instructions and use new xtensa-isa.h functions. Update calls to
633 tinsn_immed_from_frag and xg_assembly_relax. Check if transformations
634 enabled for b_j_loop_end workaround. Use build_nop instead of
635 assemble_nop. Check is_specific_opcode flag. Check for unreachable
636 frags. Use xg_get_single_size. Handle O_pltrel.
637 (fix_new_exp_in_seg): Remove check for old plt flag.
638 (convert_frag_immed_finish_loop): Update calls to tinsn_from_chars and
639 xtensa_insnbuf_to_chars. Call tinsn_immed_from_frag. Change check
640 for loop opcode to an assertion. Mark all frags up to the end of the
641 loop as not transformable.
642 (get_last_insn_flags, set_last_insn_flags): Use get_subseg_info.
643 (get_subseg_info): New.
644 (xtensa_move_literals): Call xtensa_set_frag_assembly_state. Add null
646 (xtensa_switch_to_literal_fragment): Rewrite to handle absolute
647 literals and use xtensa_switch_to_non_abs_literal_fragment otherwise.
648 (xtensa_switch_to_non_abs_literal_fragment): New.
649 (cache_literal_section): Add is_code parameter and pass it through to
650 retrieve_literal_seg.
651 (retrieve_literal_seg): Add is_code parameter and use it to set the
652 flags on the literal section. Handle case where head parameter is 0.
653 (get_frag_is_no_transform, set_frag_is_specific_opcode,
654 set_frag_is_no_transform): New.
655 (xtensa_create_property_segments): Add end_property_function parameter
656 and pass it through to add_xt_block_frags. Call bfd_get_section_flags
657 and skip SEC_DEBUGGING and !SEC_ALLOC sections.
658 (xtensa_create_xproperty_segments, section_has_xproperty): New.
659 (add_xt_block_frags): Add end_property_function parameter and call it
660 if it is non-zero. Call xtensa_frag_flags_init.
661 (xtensa_frag_flags_is_empty, xtensa_frag_flags_init,
662 get_frag_property_flags, frag_flags_to_number,
663 xtensa_frag_flags_combinable, xt_block_aligned_size,
664 xtensa_xt_block_combine, add_xt_prop_frags,
665 init_op_placement_info_table, opcode_fits_format_slot,
666 xg_get_single_size, xg_get_single_format): New.
667 (istack_push): Inline call to tinsn_copy.
668 (tinsn_copy): Delete.
669 (tinsn_has_invalid_symbolic_operands): Handle O_hi16 and O_lo16 and
670 CONST16 opcodes. Handle O_big, O_illegal, and O_absent.
671 (tinsn_has_complex_operands): Handle O_hi16 and O_lo16.
672 (tinsn_to_insnbuf): Use xg_get_single_format and new xtensa-isa.h
673 functions. Handle invisible operands.
674 (tinsn_to_slotbuf): New.
675 (tinsn_check_arguments): Use new xtensa-isa.h functions.
676 (tinsn_from_chars): Add slot parameter. Rewrite using xg_init_vinsn,
677 vinsn_from_chars, and xg_free_vinsn.
678 (tinsn_from_insnbuf): New.
679 (tinsn_immed_from_frag): Add slot parameter and handle multislot
680 instructions. Handle symbol differences.
681 (get_num_stack_text_bytes): Use xg_get_single_size.
682 (xg_init_vinsn, xg_clear_vinsn, vinsn_has_specific_opcodes,
683 xg_free_vinsn, vinsn_to_insnbuf, vinsn_from_chars, expr_is_register,
684 get_expr_register, set_expr_symbol_offset_diff): New.
685 * config/tc-xtensa.h (MAX_SLOTS): Define.
686 (xtensa_relax_statesE): Move from tc-xtensa.c. Add
687 RELAX_CHECK_ALIGN_NEXT_OPCODE, RELAX_MAYBE_DESIRE_ALIGN, RELAX_SLOTS,
688 RELAX_FILL_NOP, RELAX_UNREACHABLE, RELAX_MAYBE_UNREACHABLE, and
690 (RELAX_IMMED_MAXSTEPS): Move from tc-xtensa.c.
691 (xtensa_frag_type struct): Add is_assembly_state_set,
692 use_absolute_literals, relax_seen, is_unreachable, is_specific_opcode,
693 is_align, is_text_align, alignment, and is_first_loop_insn fields.
694 Replace is_generics and is_relax fields by is_no_transform field.
695 Delete is_text and is_longcalls fields. Change text_expansion and
696 literal_expansion to arrays of MAX_SLOTS entries. Add arrays of
697 per-slot information: literal_frags, slot_subtypes, slot_symbols,
698 slot_sub_symbols, and slot_offsets. Add fr_prev field.
699 (xtensa_fix_data struct): New.
700 (xtensa_symfield_type struct): Delete plt field.
701 (xtensa_block_info struct): Move definition to tc-xtensa.h. Add
702 forward declaration here.
703 (xt_section_type enum): Delete xt_insn_sec. Add xt_prop_sec.
704 (XTENSA_SECTION_RENAME): Undefine.
705 (TC_FIX_TYPE, TC_INIT_FIX_DATA, TC_FORCE_RELOCATION, NO_PSEUDO_DOT,
706 tc_unrecognized_line, md_do_align, md_elf_section_change_hook,
707 HANDLE_ALIGN, TC_LINKRELAX_FIXUP, SUB_SEGMENT_ALIGN): Define.
708 (TC_CONS_FIX_NEW, tc_symbol_new_hook): Delete.
709 (unit_num_copies_func, opcode_num_units_func,
710 opcode_funcUnit_use_unit_func, opcode_funcUnit_use_stage_func): New.
711 (resource_table struct): New.
712 * config/xtensa-istack.h (MAX_INSN_ARGS): Increase from 6 to 10.
713 (TInsn struct): Add keep_wide, loc, fixup, record_fix, subtype,
714 literal_space, symbol, sub_symbol, offset, and literal_frag fields.
715 (tinsn_copy): Delete prototype.
716 (vliw_insn struct): New.
717 * config/xtensa-relax.c (insn_pattern_struct): Add options field.
718 (widen_spec_list): Add option conditions for density and boolean
719 instructions. Add expansions using CONST16 and conditions for using
720 CONST16 vs. L32R. Use new Xtensa LX RSR/WSR syntax. Add entries for
722 (simplify_spec_list): Add option conditions for density instructions.
723 Add entry for NOP instruction.
724 (append_transition): Add cmp function pointer parameter and use it to
725 insert the new entry in order.
726 (operand_function_LOW16U, operand_function_HI16U): New.
727 (xg_has_userdef_op_fn, xg_apply_userdef_op_fn): Handle
728 OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
729 (enter_opname, split_string): Use xstrdup instead of strdup.
730 (init_insn_pattern): Initialize new options field.
731 (clear_req_or_option_list, clear_req_option_list,
732 clone_req_or_option_list, clone_req_option_list, parse_option_cond):
734 (parse_insn_pattern): Parse option conditions.
735 (transition_applies): New.
736 (build_transition): Use new xtensa-isa.h functions. Fix incorrectly
737 swapped last arguments in calls to append_constant_value_condition.
738 Call clone_req_option_list. Add warning about invalid opcode.
739 Handle LOW16U and HI16U function names.
740 (build_transition_table): Add cmp parameter and use it in calls to
741 append_transition. Use new xtensa-isa.h functions. Check
742 transition_applies before adding entries.
743 (xg_build_widen_table, xg_build_simplify_table): Add cmp parameter and
744 pass it through to build_transition_table.
745 * config/xtensa-relax.h (ReqOrOptionList, ReqOrOption, ReqOptionList,
746 ReqOption, transition_cmp_fn): New types.
747 (OpType enum): Add OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
748 (transition_rule struct): Add options field.
749 * doc/as.texinfo (Overview): Update Xtensa options.
750 * doc/c-xtensa.texi (Xtensa Options): Delete --[no-]density,
751 --[no-]relax, and --[no-]generics options. Update descriptions of
752 --text-section-literals and --[no-]longcalls. Add
753 --[no-]absolute-literals and --[no-]transform.
754 (Xtensa Syntax): Add description of syntax for FLIX instructions.
755 Remove use of "generic" and "specific" terminology for opcodes.
756 (Xtensa Registers): Generalize the syntax description to include
757 user-defined register files.
758 (Xtensa Automatic Alignment): Update.
759 (Xtensa Branch Relaxation): Mention limitation of unconditional jumps.
760 (Xtensa Call Relaxation): Linker can now remove most of the overhead.
761 (Xtensa Directives): Remove confusing rules about precedence.
762 (Density Directive, Relax Directive): Delete.
763 (Schedule Directive): New.
764 (Generics Directive): Rename to ...
765 (Transform Directive): ... this node.
766 (Literal Directive): Update for absolute literals. Missing
767 literal_position directive is now an error.
768 (Literal Position Directive): Update for absolute literals.
769 (Freeregs Directive): Delete.
770 (Absolute Literals Directive): New.
771 (Frame Directive): Minor editing.
772 * Makefile.am (DEPTC_xtensa_elf, DEPOBJ_xtensa_elf, DEP_xtensa_elf):
774 * Makefile.in: Regenerate.
776 2004-10-07 Richard Sandiford <rsandifo@redhat.com>
778 * config/tc-mips.c (append_insn): Use fix_new rather than fix_new_exp
779 to build the second and third fixups for a composite relocation.
780 (macro_read_relocs): New function.
781 (macro_build): Use it.
782 (s_cpsetup): Pass all three composite relocation codes to macro_build.
783 Simplify fragging code accordingly.
784 (s_gpdword): Use fix_new rather than fix_new_exp for the second part
785 of the composite relocation. Set fx_tcbit in both fixups.
787 2004-10-07 Richard Sandiford <rsandifo@redhat.com>
789 * config/tc-mips.c (append_insn): Set fx_tcbit for composite relocs.
790 (md_apply_fix3): Don't treat composite relocs as done.
792 2004-10-07 Jan Beulich <jbeulich@novell.com>
794 * macro.c (macro_expand_body): When ELF, use .LL rather than LL as
795 prefix for symbol names generated from the LOCAL macro directive.
797 * dw2gencfi.c (select_cie_for_fde): When separating CIE out from
798 FDE, treat a DW_CFA_remember_state as we do a DW_CFA_advance_loc.
800 2004-10-07 Tomer Levi <Tomer.Levi@nsc.com>
802 * config/tc-crx.c (preprocess_reglist): Handle Co-processor
804 (md_assemble): Add error checking for Co-Processor instructions.
805 (get_cinv_parameters): Add 'b' option to invalidate the
808 2004-10-05 Paul Brook <paul@codesourcery.com>
810 * config/tc-arm.c (unwind): New variable.
811 (vfp_sp_encode_reg): New function.
812 (vfp_sp_reg_required_here): Use it.
813 (vfp_sp_reg_list, vfp_dp_reg_list): Remove.
814 (vfp_parse_reg_list): New function.
815 (s_arm_unwind_fnstart, s_arm_unwind_fnend, s_arm_unwind_cantunwind,
816 s_arm_unwind_personality, s_arm_unwind_personalityindex,
817 s_arm_unwind_handlerdata, s_arm_unwind_save, s_arm_unwind_movsp,
818 s_arm_unwind_pad, s_arm_unwind_setfp, s_arm_unwind_raw): New
820 (md_pseudo_table): Add them.
821 (do_vfp_reg2_from_sp2): Use vfp_parse_reg_list and vfp_sp_encode_reg.
822 (do_vfp_sp2_from_reg2, vfp_sp_ldstm, vfp_dp_ldstm): Ditto.
823 (set_section, add_unwind_adjustsp, flush_pending_unwind,
824 finish_unwind_opcodes, start_unwind_section, create_unwind_entry,
825 require_hashconst, add_unwind_opcode): New functions.
826 * doc/tc-arm.text: Document unwinding opcodes.
827 * NEWS: Mention the new feature.
829 2004-10-04 Eric Christopher <echristo@redhat.com>
831 * config/tc-mips.c (md_apply_fix3): Remove erroneous assert.
833 2004-10-01 H.J. Lu <hongjiu.lu@intel.com>
835 * config/tc-ppc.c (md_apply_fix3): Call S_SET_THREAD_LOCAL for
837 * config/tc-s390.c (md_apply_fix3): Likewise.
838 * config/tc-sparc.c (md_apply_fix3): Likewise.
840 2004-10-01 Paul Brook <paul@codesourcery.com>
842 * config/tc-arm.c (arm_elf_section_type): New function.
843 (arm_elf_change_section): Set section link for exidx sections.
844 * config/tc-arm.h (arm_elf_section_type): Add prototype.
845 (md_elf_section_type): Define.
847 2004-10-01 Bill Farmer <Bill@the-farmers.freeserve.co.uk>
849 * config/tc-pdp11.c (md_apply_fix3): Change to sign of the SOB
850 instruction's offset.
852 2004-10-01 Adam Nemet <anemet@lnxw.com>
854 * (TARGET_FORMAT): Remove LynxOS COFF definition.
856 2004-10-01 Ravi Ramaseshan <ravi.ramaseshan@codito.com>
858 * config/tc-arc.c (tc_gen_reloc): Don't assume fixP->fx_addsy is an
859 asymbol *, instead use symbol_get_bfdsym.
861 2004-09-30 Linus Nielsen Feltzing <linus@haxx.se>
863 * config/tc-m68k.c (select_control_regs): Add mcf5249.
865 2004-09-30 Paul Brook <paul@codesourcery.com>
867 * config/tc-arm.c (do_smi, do_nop): New functions.
868 (insns): Add ARMv6ZK instructions.
869 (md_apply_fix3): Handle BFD_RELOC_ARM_SMI.
870 (tc_gen_reloc): Ditto.
871 (arm_cpus): Add mpcore and arm1176.
872 (arm_archs): Add armv6{k,z,zk}.
873 * doc/c-arm.texi: Document new cores and architectures.
875 2004-09-30 Nick Clifton <nickc@redhat.com>
877 * config/tc-arm.c: Use ISO C90 formatting.
879 2004-09-30 Vladimir Ivanov <vladitx@nucleusys.com>
881 * config/tc-arm.c (mav_reg_required_here): Allow REG_TYPE_CN
882 as alternative when REG_TYPE_MVF, REG_TYPE_MVD, REG_TYPE_MVFX or
883 REG_TYPE_MVDX is expected.
885 2004-09-29 Marc Bevand <m.bevand@gmail.com>
887 * doc/c-i386.texi (i386-Mnemonics): Fix typo.
889 2004-09-21 James E Wilson <wilson@specifixinc.com>
891 * config/tc-ia64.c (ENCODED_PSP_OFFSET): New.
892 (output_rp_psprel, output_pfs_psprel, output_preds_psprel,
893 output_spill_base, output_unat_psprel, output_lc_psprel,
894 output_fpsr_psprel, output_priunat_psprel, output_bsp_psprel,
895 output_bsprestore_psprel, output_rnat_psprel, output_spill_psprel,
896 output_spill_psprel_p): Use it.
898 2004-09-20 Tomer Levi <Tomer.Levi@nsc.com>
900 * config/tc-crx.c (handle_LoadStor): New function.
901 Handle load/stor unique instructions before parsing.
903 2004-09-17 Paul Brook <paul@codesourcery.com>
905 * config/tc-arm.c (s_arm_rel31): New funciton.
906 (md_pseudo_table): Add .rel31.
907 (md_apply_fix3): Handle BFD_RELOC_ARM_TARGET2,
908 BFD_RELOC_32_PCREL and BFD_RELOC_ARM_PREL31.
909 (tc_gen_reloc): Handle BFD_RELOC_ARM_PREL31 and BFD_RELOC_ARM_TARGET2.
910 (arm_fix_adjustable): Return 0 for BFD_RELOC_ARM_TARGET2.
911 (arm_parse_reloc): Add (target2).
913 2004-09-17 Alan Modra <amodra@bigpond.net.au>
915 * Makefile.am: Run "make dep-am".
916 * Makefile.in: Regenerate.
917 * aclocal.m4: Regenerate.
918 * configure: Regenerate.
919 * doc/Makefile.in: Regenerate.
920 * po/POTFILES.in: Regenerate.
921 * po/gas.pot: Regenerate.
923 2004-09-14 Hideki IWAMOTO <h-iwamoto@kit.hi-ho.ne.jp>
925 * config/tc-mmix.c [!LLONG_MIN]: Correct #elsif to #elif.
928 2004-09-13 Paul Brook <paul@codesourcery.com>
930 * config/tc-arm.c: Rename RELABS to TARGET1.
932 2004-09-13 Alan Modra <amodra@bigpond.net.au>
934 * messages.c (as_internal_value_out_of_range): Cast values passed
935 to as_bad_where or as_warn_where to proper type.
937 2004-09-11 Theodore A. Roth <troth@openavr.org>
939 * config/tc-avr.c: Add support for
940 atmega48, atmega88, atmega168, attiny13, attiny2313, at90can128.
942 2004-09-09 Alan Modra <amodra@bigpond.net.au>
944 * dw2gencfi.c (select_cie_for_fde): When separating CIE out
945 from FDE, treat a CFI_escape as we do a DW_CFA_advance_loc.
947 2004-09-08 Paul Brook <paul@codesourcery.com>
949 * config/obj-elf.c (obj_elf_section_type): Handle init_array,
950 fini_array and preinit_array section types.
951 * config/tc-ia64.c (ia64_elf_section_type): Remove init_array
953 * doc/as.texinfo: Document extra section types.
955 2004-09-02 Mark Mitchell <mark@codesourcery.com>
957 * Makefile.am (TARG_ENV_HFILES): Add te-symbian.h.
958 * Makefile.in: Regenerated.
959 * configure.in: Set em for arm*-*-symbianelf*.
960 * configure: Regenerated.
961 * config/tc-arm.c (elf32_arm_target_format): Use Symbian target
962 vectors when appropriate.
963 * config/te-symbian.h: New file.
965 2004-09-03 Tomer Levi <Tomer.Levi@nsc.com>
967 * config/tc-crx.c (gettrap): Exception vector can be case
969 (process_label_constant): Fix a 32-bit displacement bug in branch
971 (get_operandtype) : Bug fix, wrong operand was used.
972 (process_label_constant): Initialize relocation type to
975 2004-09-01 Richard Earnshaw < reanrsha@arm.com>
977 * tc-arm.c (arm_cpus, arm_fpus): Allow <cpu>-s as well as <cpu>s
978 for synthesizable cores.
980 * doc/c-arm.texi (ARM Options): Document canonical names of CPUs.
982 2004-08-25 Dmitry Diky <diwil@spec.ru>
984 * config/tc-msp430.c: Clean-up the code.
985 (md_relax_table): New relax table.
986 (mcu_types): Sort MCU types.
987 (md_pseudo_table): Add .profiler pseudo handler.
988 (pow2value): New function.
989 (msp430_profiler): New function.
990 (msp430_operands): Add new insns handlers.
991 (msp430_srcoperand): Add register operand handler, allow complex
993 (md_estimate_size_before_relax): Rewritten.
994 (md_convert_frag): Rewritten.
995 (msp430_relax_frag): New function.
996 * config/tc-msp430.h (md_relax_frag): define macro
997 * doc/c-msp430.texi: Update information.
999 2004-08-24 Nick Clifton <nickc@redhat.com>
1001 * as.c (std_shortopts): Allow -g to take an optional argument.
1002 (parse_args): Pass any switch starting with -g on to the backend
1005 2004-08-18 Mark Mitchell <mark@codesourcery.com>
1007 * configure.in (arm*-*-symbianelf*): New target.
1008 (arm*-*-eabi*): Likewise.
1009 * configure: Regenerated.
1011 2004-08-18 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1012 * config/tc-mips.c (append_insn): Handle delay slots in branch likely
1015 2004-08-18 Jakub Jelinek <jakub@redhat.com>
1017 * config/tc-ia64.c (start_unwind_section): Add linkonce_empty
1018 argument, don't do anything if current section is not
1019 .gnu.linkonce.t.* and linkonce_empty is set.
1020 (generate_unwind_image, dot_endp): Adjust callers, call
1021 start_unwind_section (*, 1) if nothing will be put into the
1024 2004-08-17 Nick Clifton <nickc@redhat.com>
1026 * as.c (MD_DEBUG_FORMAT_SELECTOR): Provide default definition.
1027 (show_usage): Add -g.
1028 (std_longopts): Add --gen-debug. Alpha sort the table.
1029 (parse_args): Print an error message if a switch is not handled.
1030 Handle the -g switch, calling md_debug_format_selector() if
1032 * NEWS: Mention new feature.
1033 * doc/as.texinfo: Document new switch.
1034 * doc/internals.texi: Document behaviour of md_parse_option.
1036 * config/tc-arm.c (md_parse_option): Do not issue an error message
1037 if the switch is not recognised.
1038 * config/tc-m68k.c (md_parse_option): Likewise.
1039 * config/tc-pdp11.c (md_parse_option): Likewise.
1040 * config/tc-v850.c (md_parse_option): Likewise.
1042 * as.h: Fix up formatting.
1045 2004-08-16 Nick Clifton <nickc@redhat.com>
1047 * macro.c (macro_set_alternate): Use ISO C90 formatting.
1049 * configure.in: Sort architecture based tables alphabetically.
1050 * configure: Regenerate.
1052 2004-08-16 Alan Modra <amodra@bigpond.net.au>
1054 * config/tc-ppc.c (tc_ppc_regname_to_dw2regnum <regnames>): Replace
1055 { "cc", 68 }, with { "cr", 70 }.
1057 2004-08-13 Jan Beulich <jbeulich@novell.com>
1058 Nick Clifton <nickc@redhat.com>
1060 * as.c: Add and handle new --alternate command line option.
1061 * macro.c (macro_set_alternate): New.
1062 * macro.h (macro_set_alternate): Declare.
1063 * read.c: Add and handle new .altmacro and .noaltmacro directives.
1064 * doc/as.texinfo: Document new command line option and pseudo-ops
1065 as well as insert documentation originating from gasp about
1066 alternate macro syntax.
1067 * NEWS: Mention new command line option and pseudo-ops.
1069 2004-08-10 Mark Mitchell <mark@codesourcery.com>
1071 * expr.c (operand): Handle the "~", "-", and "!" operators applied
1074 2004-08-06 Paul Brook <paul@codesourcery.com>
1076 * config/tc-arm.c (md_apply_fix3, tc_gen_reloc, arm_parse_reloc):
1077 Handle new relocations.
1078 * include/elf/arm.h (elf_arm_reloc_type): Add new EABI relocations.
1080 2004-08-05 Bob Wilson <bob.wilson@acm.org>
1082 * write.c (relax_segment): Use was_address instead of address when
1083 setting fr_fix field for align frag due to backwards .org.
1085 2004-07-29 Alexandre Oliva <aoliva@redhat.com>
1087 Introduce SH2a support.
1088 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
1089 * config/tc-sh.c (get_specific): Change arch_sh2a_up to
1091 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
1092 * config/tc-sh.c (md_parse_option): Add sh2a-nofpu ISA handling.
1093 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
1094 * config/tc-sh.c (sh_elf_final_processing): Move sh2a recognition
1095 to end of conditional expression.
1096 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
1097 * config/tc-sh.c: Add sh2a-nofpu support.
1098 2003-12-29 DJ Delorie <dj@redhat.com>
1099 * tc-sh.c: Add sh2a support.
1100 (parse_reg): Add tbr.
1101 (parse_at): Support @@(disp,tbr).
1102 (get_specific): Support sh2a opcodes.
1103 (insert4): New, for 4 byte relocs.
1104 (build_Mytes): Support sh2a opcodes.
1105 (md_apply_fix3_Mytes): Support sh2a opcodes.
1106 2003-12-02 Michael Snyder <msnyder@redhat.com>
1107 * config/tc-sh.c (md_parse_option): Handle sh2a.
1108 (sh_elf_final_processing): Ditto.
1110 2004-07-27 Jason Thorpe <thorpej@wasabisystems.com>
1112 * config/tc-hppa.h (TARGET_FORMAT): Set to "elf32-hppa-netbsd"
1115 2004-07-27 Alan Modra <amodra@bigpond.net.au>
1117 * config/tc-ppc.c (ppc_frob_file_before_adjust): Warn if .toc too big.
1118 (ppc_arch): Expand comment.
1120 2004-07-27 Tomer Levi <Tomer.Levi@nsc.com>
1122 * config/tc-crx.c: Support evaluating the difference between two
1124 * config/tc-crx.h: Likewise.
1126 2004-07-26 H.J. Lu <hongjiu.lu@intel.com>
1128 * config/tc-ia64.c (start_unwind_section): Set the linked-to
1130 (ia64_elf_section_change_hook): Set the linked-to section for
1133 2004-07-26 Dmitry Diky <diwil@spec.ru>
1135 * config/tc-msp430.c: Add new subtargets: msp430x1610,
1136 msp430x1611, msp430x1612, msp430x415, msp430x417, msp430xG437,
1137 msp430xG438, msp430xG439.
1139 2004-07-25 Daniel Jacobowitz <dan@debian.org>
1141 * doc/as.texinfo (Section, PushSection): Correct documentation
1144 2004-07-21 Jan Beulich <jbeulich@novell.com>
1146 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
1147 permissible for the selected instruction suffix.
1148 (match_template): Don't permit 64-bit general purpose operands in
1150 (finalize_imm): Permit 64-bit immediates.
1151 (build_modrm_byte): Don't treat 32-bit addressing in 64-bit mode
1152 specially except for the width of the used base and/or index
1153 registers. For 32-bit displacements, use sign-extended
1154 relocations only when using 64-bit addressing.
1155 Force zero displacement on rip-relative addressing when there is
1156 no other displacement.
1157 (i386_index_check): Don't treat 32-bit addressing in 64-bit mode
1158 specially except for the width of the used base and/or index
1160 (parse_register): Disallow Reg64 registers in 32-bit mode.
1162 * config/tc-i386.c: For DefaultSize instructions, don't guess a 'q'
1163 suffix if the instruction doesn't support it.
1165 2004-07-20 Maciej W. Rozycki <macro@linux-mips.org>
1167 * config/tc-mips.c (append_insn): Handle constant expressions with
1168 no associated relocation.
1169 (mips_ip): Cancel the expression after use for the Q format
1171 (parse_relocation): Return no relocation for unsupported
1173 (my_getSmallExpression): Return no relocation if no relocation
1176 2004-07-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1178 * config/obj-som.c (adjust_stab_sections): Add prototype.
1179 (obj_som_compiler, obj_som_version, obj_som_copyright,
1180 adjust_stab_sections): Add ATTRIBUTE_UNUSED to unused arguments.
1181 * config/tc-hppa.c (update_subspace): Likewise.
1182 (is_defined_subspace): Amplify comment.
1183 * config/obj-som.h (som_frob_file): Add prototype.
1185 2004-07-19 Christopher Faylor <cgf@timesys.com>
1186 H.J. Lu <hongjiu.lu@intel.com>
1188 * subsegs.c (section_symbol): Don't create a new segment when
1189 existing segment is undefined.
1191 2004-07-16 Richard Earnshaw <rearnsha@arm.com>
1193 * config/tc-arm.c: Include include/opcode/arm.h.
1194 (ARM_EXT_*, ARM_ARCH_*, ARM_ANY, ARM_ALL, COPROC_ANY): Delete.
1195 (FPU_FPA_EXT_* FPU_VFP_EXT_*, FPU_ANY, FPU_NONE, FPU_MAVERICK): Delete.
1196 (FPU_ARCH_*): Delete.
1197 * Makefile.am: Update dependencies.
1198 * Makefile.in: Regenerate.
1200 2004-07-15 Nick Clifton <nickc@redhat.com>
1202 * configure.in: Accept armbe as a big-endian arm configuration.
1203 * configure: Regenerate.
1205 2004-07-13 Thomas Nystrom <thn@saeab.se>
1207 * config/tc-i386.c (T_SHIFTOP): New constant.
1208 (intel_e05_1): Handle '&', '|' and T_SHIFTOP.
1209 (intel_el1): Handle '~'.
1210 (intel_get_token): Handle '<>', '&', '|' and '~'.
1212 2004-07-13 Nick Clifton <nickc@redhat.com>
1214 (md_assemble): Remove spurious newline from end of as_bad error
1216 (intel_e05_1): Likewise.
1217 (intel_e11): Likewise.
1218 (intel_match_token): Likewise.
1220 2004-07-11 Andreas Schwab <schwab@suse.de>
1222 * config/tc-m68k.c: Convert to C90. Remove redundant
1223 declarations. Indentation fixup.
1224 [M68KCOFF]: Include "obj-coff.h" instead of declaring
1225 obj_coff_section ourselves.
1227 2004-07-09 James E Wilson <wilson@specifixinc.com>
1229 * config/tc-ia64.c (default_big_endian): New.
1230 (dot_byteorder, md_begin): Use it.
1231 (md_parse_option): Set it.
1233 2004-07-09 Nick Clifton <nickc@redhat.com>
1235 * configure.in: Change sh-sybmian-elf to sh-*-symbianelf.
1236 * configure: Regenerate.
1237 * NEWS: Change sh-sybmian-elf to sh-*-symbianelf.
1238 * config/tc-sh.c (sh_elf_final_processing): Use renamed version of
1239 sh_find_elf_flags if necessary.
1241 2004-07-08 Richard Sandiford <rsandifo@redhat.com>
1243 * config/tc-mips.c (mips_fix_adjustable): If the full addend is
1244 going to be split into more than one in-place addend, return 0
1245 for relocations against mergeable sections. Associate comments
1248 2004-07-07 Tomer Levi <Tomer.Levi@nsc.com>
1250 * Makefile.am (CPU_TYPES): Add crx.
1251 (TARGET_CPU_CFILES): Add config/tc-crx.c.
1252 (TARGET_CPU_HFILES): Add config/tc-crx.h.
1253 (DEPTC_crx_elf): New target.
1254 (DEPOBJ_crx_elf): Likewise.
1255 (DEP_crx_elf): Likewise.
1256 * Makefile.in: Regenerate.
1257 * configure.in: Add crx* target.
1258 * configure: Regenerate.
1259 * config/tc-crx.c: New file.
1260 * config/tc-crx.h: New file.
1261 * NEWS: Mention new target.
1263 2004-07-06 Nick Clifton <nickc@redhat.com>
1265 * config.in: Undefine TARGET_SYMBIAN by default.
1267 * configure: Regenerate. Add sh-symbian-elf target. If
1268 selected define TARGET_SYMBIAN.
1269 * config/tc-sh.h (TARGET_FORMAT): Select a Symbian target
1270 format if TARGET_SYMBIAN has been defined.
1272 * output-file.c (output_file_create): Report the target format
1273 chosen when bfd_openw reports that it is invalid.
1275 * config/obj-coff.c (coff_pseudo_table): Only define the weak
1276 pseudo for BFD based assemblers.
1278 2004-07-05 Andrew Stubbs <andrew.stubbs@superh.com>
1281 * config/tc-sh.c (md_assemble): Change isspace to ISSPACE.
1282 (md_parse_option): Remove redundant -isa testing.
1283 Make bfd_arch variable const.
1284 (md_show_usage): Make bfd_arch variable const.
1286 2004-07-03 James E Wilson <wilson@specifixinc.com>
1288 * config/tc-ia64.c (emit_one_bundle): Check and set insn_addr.
1289 * config/tc-ia64.h (md_frag_check): Define.
1291 2004-07-03 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
1293 * config/obj-coff.c (obj_coff_weak): New .weak syntax for PE weak
1295 * doc/as.texinfo (Weak): Document PE weak symbols.
1297 2004-07-03 Richard Sandiford <rsandifo@redhat.com>
1299 * config/tc-mips.c (HAVE_IN_PLACE_ADDENDS): New macro.
1300 (reloc_needs_lo_p): Only return true if HAVE_IN_PLACE_ADDENDS.
1301 (mips_frob_file): Rework so that only a single pass through the
1302 relocs is needed. Allow %lo()s to have higher offsets than their
1303 corresponding %hi()s or %got()s.
1305 2004-07-02 Nick Clifton <nickc@redhat.com>
1307 * config/tc-arm.c (md_apply_fix3:BFD_RELOC_ARM_IMMEDIATE): Do not
1308 allow values which have come from undefined symbols.
1309 Always consider this fixup to have been processed as a reloc
1310 cannot be generated for it.
1312 2004-07-02 Alan Modra <amodra@bigpond.net.au>
1314 * frags.h (struct frag): Add has_code and insn_addr fields.
1315 * write.c (cvt_frag_to_fill): Invoke md_frag_check.
1316 * config/tc-ppc.c (md_assemble): Check and set insn_addr.
1317 * config/tc-ppc.h (md_frag_check): Define.
1319 2004-06-28 Maciej W. Rozycki <macro@linux-mips.org>
1321 * doc/Makefile.am (info): Rename goal to...
1322 (info-local): ... this, to preserve implicit dependencies.
1323 * doc/Makefile.in: Regenerate with automake 1.8.5.
1325 2004-06-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1327 * config/tc-m32r.c (md_convert_frag): Changed for @PLT.
1328 (m32r_cgen_record_fixup_exp): Changed for @GOTOFF, @GOT.
1329 (m32r_fix_adjustable): Changed for @GOTOFF, @GOT, @PLT.
1330 (tc_gen_reloc): Likewise.
1331 (m32r_end_of_match): Add for @GOTOFF, @GOT, @PLT.
1332 (m32r_parse_name): Likewise.
1333 (m32r_cgen_parse_fix_exp): Likewise.
1334 * config/tc-m32r.h (md_parse_name): Define for @GOTOFF, @GOT, @PLT.
1335 (O_PIC_reloc): Likewise.
1336 (TC_CGEN_PARSE_FIX_EXP): Likewise..
1337 * cgen.c (gas_cgen_parse_operand): Add TC_CGEN_PARSE_FIX_EXP
1338 for @GOTOFF, @GOT, @PLT.
1340 2004-06-21 Jan Beulich <jbeulich@novell.com>
1342 * gas/symbols.c: While discarding ordinary local absolute symbols
1343 when --strip-local-absolute is in effect, retain file symbols.
1345 2004-06-20 Andreas Schwab <schwab@suse.de>
1347 * config/tc-m68k.c (mri_chip): Replace current_chip, not augment.
1348 (md_parse_option): Likewise.
1350 2004-06-17 Jan Beulich <jbeulich@novell.com>
1352 * config/tc-i386.c: Deal with LEX_QM the same way as with LEX_AT.
1353 * config/te-netware.h: New file.
1354 * config/te-ppcnw.h: Delete: Obsolete.
1355 * configure.in: Eliminate ill NetWare targets. Make generic
1356 NetWare target use proper emulation.
1357 * Makefile.am: Eliminate reference to obsolete te-ppcnw.h, add
1358 reference to new te-netware.h.
1359 * configure: Regenerate.
1360 * Makefile.in: Regenerate.
1362 2004-06-15 Martin Schwidefsky <schwidefsky@de.ibm.com>
1364 * config/tc-s390.c (s390_insn): Avoid incorrect signed/unsigned
1365 comparison in .insn pseudo operation.
1367 2004-06-15 Alan Modra <amodra@bigpond.net.au>
1369 * config/obj-coff.c (coff_adjust_section_syms): Use
1370 bfd_get_section_size instead of bfd_get_section_size_before_reloc.
1371 (coff_frob_section): Likewise.
1372 * config/tc-mips.c (md_apply_fix3): Likewise.
1373 * config/obj-elf.c (elf_frob_file): Use bfd_set_section_size.
1374 (elf_frob_file_after_relocs): Likewise.
1376 2004-06-10 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1378 * config/tc-hppa.c (log2): Rename to exact_log2.
1379 (pa_next_subseg): Delete unused function.
1380 (create_new_space): Mark unused arguments with ATTRIBUTE_UNUSED.
1381 (create_new_subspace): Likewise.
1384 * config/tc-hppa.c (hppa_fix_adjustable): Allow reduction of fake
1385 labels. Fix warning.
1387 2004-05-28 DJ Delorie <dj@redhat.com>
1389 * config/tc-mn10300.h (tc_fix_adjustable): Define.
1390 * config/tc-mn10300.c (mn10300_fix_adjustable): Don't adjust debug
1391 or non-merged symbols.
1393 2004-05-28 H.J. Lu <hongjiu.lu@intel.com>
1395 * config/tc-ia64.c (remove_marked_resource): Save, clear and
1396 restore the old slot when inserting srlz.i/srlz.d.
1398 2004-05-28 Andrew Stubbs <andrew.stubbs@superh.com>
1400 * Makefile.am: Regenerate dependecies.
1401 * Makefile.in: Regenerate.
1402 * config/tc-sh.c (valid_arch): Make unsigned.
1403 (preset_target_arch): Likewise.
1404 (md_begin): Use new architecture flags system.
1405 (get_specific): Likewise.
1406 (assemble_ppi): Likewise.
1407 (md_assemble): Likewise. Also fix error check for bad opcodes.
1408 (md_parse_option): Likewise. Also generate -isa values according
1409 to the table in bfd/cpu-sh.c instead of just constants. Also
1410 allow <arch>-up ISA variants.
1411 (sh_elf_final_processing): Replace if-else chain with a call to
1412 sh_find_elf_flags().
1413 * testsuite/gas/sh/arch: New directory.
1414 * testsuite/gas/sh/arch/arch.exp: New test script.
1415 * testsuite/gas/sh/arch/arch_expected.txt: New file.
1416 * testsuite/gas/sh/arch/sh.s: New file.
1417 * testsuite/gas/sh/arch/sh2.s: New file.
1418 * testsuite/gas/sh/arch/sh-dsp.s: New file.
1419 * testsuite/gas/sh/arch/sh2e.s: New file.
1420 * testsuite/gas/sh/arch/sh3-nommu.s: New file.
1421 * testsuite/gas/sh/arch/sh3.s: New file.
1422 * testsuite/gas/sh/arch/sh3-dsp.s: New file.
1423 * testsuite/gas/sh/arch/sh3e.s: New file.
1424 * testsuite/gas/sh/arch/sh4-nommu-nofpu.s: New file.
1425 * testsuite/gas/sh/arch/sh4-nofpu.s: New file.
1426 * testsuite/gas/sh/arch/sh4.s: New file.
1427 * testsuite/gas/sh/arch/sh4a-nofpu.s: New file.
1428 * testsuite/gas/sh/arch/sh4al-dsp.s: New file.
1429 * testsuite/gas/sh/arch/sh4a.s: New file.
1431 2004-05-28 Peter Barada <peter@the-baradas.com>
1433 * config/gc-m68k.c(m68k_ip): Convert mode 5 addressing
1434 with zero offset into mode 2 addressing to save a word.
1436 2004-05-27 H.J. Lu <hongjiu.lu@intel.com>
1438 * config/tc-ia64.c (ar_is_in_integer_unit): Removed.
1439 (ar_is_only_in_integer_unit): New.
1440 (ar_is_only_in_memory_unit): New.
1441 (generate_unwind_image): Silence gcc on 32bit host.
1442 (md_assemble): Use ar_is_only_in_integer_unit instead of
1443 ar_is_in_integer_unit. Check AR access.
1445 2004-05-27 Peter Barada <peter@the-baradas.com>
1447 * config/tc-m68k.c (md_begin): Sort the opcode table into
1449 (m68k_compare_opcode): New function to do the sorting.
1451 2004-05-24 Peter Barada <peter@the-baradas.com>
1453 * config/m68k-parse.y(operand): Allow for MAC/EMAC mask
1454 addressing on MIT style operands.
1455 * config/m68k-parse.y(yylex): Allow '-&' for predecrement
1457 * config/tc-m68k.c(install_operand): Comment 'G' and 'H' type
1460 2004-05-23 Alan Modra <amodra@bigpond.net.au>
1462 * expr.c (operand, operator): Don't reject '++' and '--'.
1464 2004-05-20 Richard Sandiford <rsandifo@redhat.com>
1466 * config/tc-mips.c (append_insn): Use ISA-encoded addresses in MIPS16
1469 2004-05-17 Adam Nemet <anemet@lnxw.com>
1471 * configure.in: Add ppc-*-lynxos*. Update i386-*-lynxos* to ELF.
1472 * configure: Regenerate.
1474 2004-05-13 Paul Brook <paul@codesourcery.com>
1476 * dw2gencfi.c (output_cie): Handle dwarf3 format CIE entries.
1478 2004-05-13 Joel Sherrill <joel@oarcorp.com>
1480 * configure.in (or32-*-rtems*): Switch to elf.
1481 * configure: Regenerate.
1483 2004-05-13 Nick Clifton <nickc@redhat.com>
1485 * po/fr.po: Updated French translation.
1487 2004-05-11 Nick Clifton <nickc@redhat.com>
1489 * doc/as.texinfo (Section): Document G and T flags to .section
1490 directive. Document the extra arguments that the G flag
1491 requires. Document the #tls flag.
1493 2004-05-11 H.J. Lu <hongjiu.lu@intel.com>
1495 * subsegs.c (section_symbol): Create a new section symbol if
1496 the existing one doesn't match.
1497 * symbols.c (symbol_set_bfdsym): Don't reset BFD section symbol.
1499 2004-05-07 Richard Sandiford <rsandifo@redhat.com>
1501 * config/tc-mips.c (append_insn, mips_emit_delays): Extend -mfix-vr4120
1502 to cope with VR4181A errata MD(1) and MD(4).
1504 2004-05-07 Brian Ford <ford@vss.fsi.com>
1506 * NEWS: Mention .secrel32 for pe[i]-i386.
1508 2004-05-07 Alexandre Oliva <aoliva@redhat.com>
1510 * config/tc-frv.h (MAX_MEM_FOR_RS_ALIGN_CODE): New.
1511 (HANDLE_ALIGN): New.
1513 2004-05-06 Daniel Jacobowitz <dan@debian.org>
1515 * Makefile.am (DIST_SUBDIRS): Define.
1516 * aclocal.m4: Regenerate with automake 1.8.4.
1517 * Makefile.in: Likewise.
1518 * doc/Makefile.in: Likewise.
1520 2004-05-06 David Mosberger-Tang <davidm@hpl.hp.com>
1522 * config/tc-ia64.c (dot_serialize): Declare.
1523 (dot_serialize): New function.
1524 (md_pseudo_table): Add ".serialize.data" and
1525 ".serialize.instruction" directives.
1527 2004-05-06 Nick Clifton <nickc@redhat.com>
1529 * messages (as_internal_value_out_of_range): Print a message about
1530 a value being out of range. Be consistent about whether the
1531 values are printed in decimal or hexadecimal.
1532 (as_warn_value_out_of_range): Generate a warning message about an
1534 (as_bad_value_out_of_range): Generate an error message about an
1536 * as.h: Prototype the new functions.
1537 * config/tc-alpha.c (insert_operand): Use new function.
1538 * config/tc-arc.c (arc_insert_operand): Likewise.
1539 * config/tc-mn10200.c (mn10200_insert_operand): Likewise.
1540 * config/tc-mn10300.c (mn10300_insert_operand): Likewise.
1541 * config/tc-ppc.c (ppc_insert_operand): Likewise.
1542 * config/tc-s390.c (s390_insert_operand): Likewise.
1543 * config/tc-v850.c (v850_insert_operand): Likewise.
1545 2004-05-05 Alexandre Oliva <aoliva@redhat.com>
1547 * configure.in: Set em=linux for frv-*-*linux*.
1548 * configure: Rebuilt.
1549 * config/tc-frv.h (TARGET_FORMAT): Use elf32-frvfdpic if...
1550 (frv_md_fdpic_enabled): New.
1551 * config/tc-frv.c (frv_md_fdpic_enabled): New.
1552 (DEFAULT_FDPIC): New.
1553 (frv_flags): Use DEFAULT_FDPIC.
1554 (frv_pic_flag): Likewise.
1555 (OPTION_NOPIC): New.
1556 (md_longopts): Add -mnopic.
1557 (md_parse_option): Handle it.
1558 (md_show_usage): Add -mfdpic and -mnopic.
1560 2004-05-05 Peter Barada <peter@the-baradas.com>
1562 * config/tc-m68k.c: Add find_cf_chip to print list of valid
1563 chips for invalid coldfire instructions, rename selectors
1564 for ColdFire sub-variants, add 521x,5249,547x,548x and aliases,
1565 add current_chip to track which chip is referred to(including save/restore),
1566 use current_chip to select control registers, not current_arch.
1567 (md_show_usage): Add new chips.
1568 * doc/c-m68k.texi: Document new command line switches.
1570 2004-05-05 Jakub Jelinek <jakub@redhat.com>
1572 * tc-s390.h (md_do_align, HANDLE_ALIGN): Remove.
1573 (NOP_OPCODE): Define.
1574 (s390_align_code): Remove prototype.
1575 * tc-s390.c (s390_align_code): Remove.
1577 2004-05-04 H.J. Lu <hongjiu.lu@intel.com>
1579 * config/tc-ia64.c (make_unw_section_name): Removed.
1580 (start_unwind_section): New function.
1581 (generate_unwind_image): Take const segT instead of const
1583 (dot_handlerdata): Adjusted.
1584 (dot_endp): Likewise.
1586 2004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1588 * config/obj-elf.c (obj_elf_change_section): Allow the
1589 ".note.GNU-stack" section has SHF_EXECINSTR.
1591 2004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1593 * config/obj-elf.c (get_section): Return bfd_boolean.
1594 (obj_elf_change_section): Call bfd_get_section_by_name_if
1595 instead of bfd_map_over_sections.
1597 2004-04-30 H.J. Lu <hongjiu.lu@intel.com>
1599 * config/obj-elf.c (get_section): New function.
1600 (obj_elf_change_section): Support multiple sections with same
1603 2004-04-30 Nick Clifton <nickc@redhat.com>
1605 * config/tc-arm.c (create_register_alias): Fix typo checking for
1606 case sensitive register aliases.
1607 (co_proc_number): Use error message string in all_reg_maps[]
1609 (cp_reg_required_here): Likewise.
1610 (fp_reg_required_here): Likewise.
1612 2004-04-29 Brian Ford <ford@vss.fsi.com>
1614 * dwarf2dbg.c (dwarf2_finish): Add SEC_DEBUGGING to section flags.
1616 2004-04-28 Chris Demetriou <cgd@broadcom.com>
1618 * config/tc-mips.c (HAVE_32BIT_ADDRESSES, append_insn, macro_build)
1619 (load_address, macro, mips_ip, md_parse_option)
1620 (mips_force_relocation, mips_validate_fix, md_apply_fix3)
1621 (s_change_sec, pic_need_relax, tc_gen_reloc): Remove all
1622 embedded-PIC handling, and update comments.
1623 (SWITCH_TABLE): Remove.
1624 * config/tc-mips.h (DIFF_EXPR_OK): Delete.
1625 (enum mips_pic_level): Remove EMBEDDED_PIC.
1626 (EXTERN_FORCE_RELOC): Remove embedded-PIC handling.
1627 (TC_FORCE_RELOCATION): Update comment.
1628 * ecoff.c (ecoff_build_lineno): Add comment about some code that
1629 might be safe to remove now that MIPS embedded-PIC is gone.
1631 2004-04-28 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1633 * config/obj-som.c (obj_som_init_stab_section): Add new arguments in
1634 call to obj_set_subsection_attributes.
1635 (obj_som_init_stab_section): Likewise.
1636 * config/tc-hppa.c (default_subspace_dict): Add comdat field.
1637 (pa_def_subspaces): Provide comdat default.
1638 (pa_subspace): Handle new "comdat" parameter. Set SEC_LINK_ONCE and
1639 not SEC_IS_COMMON if section is comdat, common or dup_common. Update
1640 calls to create_new_subspace and update_subspace to pass comdat flag.
1641 (create_new_subspace, update_subspace): Add new comdat argument. Use
1642 it in calls to obj_set_subsection_attributes.
1643 * doc/c-hppa.texi (.subspa, .nsubspa): Document new comdat parameter
1644 and use of comdat, common and dup_comm parameters.
1646 2004-04-26 H.J. Lu <hongjiu.lu@intel.com>
1648 * config/obj-elf.c (obj_elf_change_section): Check if the old
1649 group name is NULL before comparison.
1651 2004-04-23 Chris Demetriou <cgd@broadcom.com>
1653 * config/tc-mips.h (mips_dwarf2_addr_size): Prototype.
1655 2004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1657 * config/tc-mips.c (s_mipsset): Set default CPU type for .set mipsN.
1659 2004-04-23 Chris Demetriou <cgd@broadcom.com>
1661 * config/tc-mips.c (md_longopts): Remove -membedded-pic option.
1662 (OPTION_MEMBEDDED_PIC): Remove.
1663 (OPTION_TRAP, OPTION_BREAK, OPTION_EB, OPTION_EL)
1664 (OPTION_FP32, OPTION_GP32, OPTION_CONSTRUCT_FLOATS)
1665 (OPTION_NO_CONSTRUCT_FLOATS, OPTIONS_FP64, OPTION_GP64)
1666 (OPTION_RELAX_BRANCH, OPTION_NO_RELAX_BRANCH)
1667 (OPTION_ELF_BASE): Renumber.
1668 (md_parse_option): Remove OPTION_MEMBEDDED_PIC handling.
1669 (md_show_usage): Remove mention of -membedded-pic.
1670 * doc/as.texinfo: Remove mention of -membedded-pic.
1672 2004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1674 * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Remove.
1675 * config/tc-mips.c (RDATA_SECTION_NAME, mips_target_format): Remove
1677 (md_begin, mips_ip, md_parse_option, s_change_sec, s_option,
1678 s_abicalls, nopic_need_relax, tc_gen_reloc): Remove uses of
1679 USE_GLOBAL_POINTER_OPT.
1681 2004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1683 * config/tc-mips.c (macro): One more use of load_delay_nop.
1685 2004-04-22 Atsushi Nemoto <anemo@mba.ocn.ne.jp>
1687 * config/tc-mips.c (load_delay_nop): New function.
1688 (load_address, macro): Use load_delay_nop() to build a nop
1689 which can be omitted with gpr_interlocks.
1691 2004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1693 * config/tc-mips.c (hilo_interlocks, gpr_interlocks,
1694 cop_interlocks): Remove superfluous CPU entries.
1696 2004-04-22 Paul Brook <paul@codesourcery.com>
1698 * config/tc-arm.c (mav_parse_offset): Value must be multiple of 4.
1700 2004-04-22 Peter Barada <peter@the-baradas.com>
1702 * NEWS: Added support for EMAC instructions and MAC/EMAC
1704 * config/m68k-parse.h: Add ACC[123], ACCEXT{01,23}, MAC/EMAC
1705 scale factor tokens, trailing_ampersand to mark mask addressing
1706 for MAC/EMAC instructions.
1707 * config/m68k-parse.y: Add options_ampersand clause, '<<',
1709 (yylex): Handle '>', '<', and '&' following '+'.
1710 * config/tc-m68k.c: Set mcfmac/mcfemac on appropriate ColdFire
1711 architectures in archs[].
1712 (m68k-ip): Add '4', 'e', 'g', 'i', cases to handle mask addressing
1713 for MAC/EMAC instructions, ACC[0123], ACCEXT{01,23}, and '<<'/'>>'
1715 (m68k_ip): Handle trailing '&' on MAC/EMAC insns.
1716 (install_operand): Fix 'n' case, Add 'F', 'f', 'G', 'H', 'I', ']'
1718 Add EMAC operands to init_table[].
1720 2004-04-22 Bruno De Bus <bdebus@elis.ugent.be>
1722 * config/tc-arm.h (enum mstate): Move here, add MAP_UNDEFINED
1724 (TC_SEGMENT_INFO_TYPE): Define to enum mstate.
1725 * config/tc-arm.c (enum mstate): Delete from here.
1726 (mapping_state): Remove the static mapstate variable and instead
1727 store the state in the segment. This allows a per-section mapping
1728 state. Handle and ignore MAP_UNDEFINED states.
1729 (arm_elf_change_section): Get the current mapping state from the
1731 (s_ltorg): Set the mapping state to MAP_DATA.
1732 (arm_cleanup): Use arm_elf_change_section to get the mapping state
1733 for each pool as it is emitted.
1735 2004-04-22 Nick Clifton <nickc@redhat.com>
1737 * config/tc-arm.h: Formatting tidy ups.
1739 2004-04-20 Chris Demetriou <cgd@broadcom.com>
1741 * NEWS: Note that MIPS -membedded-pic option is deprecated.
1743 2004-04-20 DJ Delorie <dj@redhat.com>
1745 * config/tc-i386.h [TE_PE] (TC_CONS_FIX_NEW): Define.
1746 * config/tc-i386.c (md_pseudo_table) [TE_PE]: Add "secrel32".
1747 [TE_PE] (O_secrel): Define.
1748 [TE_PE] (x86_pe_cons_fix_new): New.
1749 [TE_PE] (pe_directive_secrel): Likewise.
1750 (tc_gen_reloc) [TE_PE]: Support BFD_RELOC_32_SECREL.
1752 2004-04-19 Eric Christopher <echristo@redhat.com>
1754 * config/tc-mips.c (mips_dwarf2_addr_size): Revert part
1755 of previous patch for fix in gcc.
1757 2004-04-19 Jakub Jelinek <jakub@redhat.com>
1759 * config/tc-xtensa.c (xg_assembler_literal): Fix a typo.
1761 2004-04-19 Nathan Sidwell <nathan@codesourcery.com>
1763 * read.c (do_align): Call md_flush_pending_output, if defined.
1765 2004-04-16 Alan Modra <amodra@bigpond.net.au>
1767 * expr.c (operand): Correct checks for ++ and --.
1769 2004-04-14 H.J. Lu <hongjiu.lu@intel.com>
1771 * config/tc-generic.c: Add some comments.
1773 2004-04-14 Richard Sandiford <rsandifo@redhat.com>
1775 * doc/c-mips.texi (-m{no-,}fix-vr4120): Renamed from
1776 -{no-}mfix-vr4122-bugs.
1777 * config/tc-mips.c (mips_fix_vr4120): Renamed from mips_fix_4122_bugs.
1778 (append_insn, mips_emit_delays): Update accordingly.
1779 (OPTION_FIX_VR4120, OPTION_NO_FIX_VR4120): Renamed from *VR4122.
1780 (md_longopts): Change -{no-,}mfix-vr4122-bugs to -m{no-,}fix-vr4120.
1781 (md_parse_option): Update after above changes.
1782 (md_show_usage): Add -mfix-vr4120.
1784 2004-04-13 Bob Wilson <bob.wilson@acm.org>
1786 * doc/as.texinfo (Sub-Sections): Conditionalize COFF-specific use
1787 of .section directive; add a reference to the ELF .subsection
1790 2004-04-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1792 * config/tc-m32r.c (md_assemble): Fixed infinite loop bug
1795 2004-04-11 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1797 * Makefile.am: Remove mips from aout targets.
1798 * Makefile.in: Regenerate.
1799 * configure.in: Remove mips-dec-bsd* target.
1800 * configure: Regenerate.
1802 2004-04-07 Alan Modra <amodra@bigpond.net.au>
1805 * config/tc-ppc.c (ppc_elf_suffix): Add valid32 and valid64 fields
1806 to struct map_bfd. Adjust MAP macro, and define MAP32, MAP64.
1807 Update "mapping". Restrict some @ modifiers to 32 bit.
1809 2004-04-01 Asgari Jinia <asgarij@kpitcummins.com>
1810 Dhananjay Deshpande <dhananjayd@kpitcummins.com>
1812 * config/tc-sh.c (dont_adjust_reloc_32): New variable.
1813 (sh_fix_adjustable): Avoid adjusting BFD_RELOC_32 when
1814 dont_adjust_reloc_32 is set.
1815 (md_longopts): Add option -renesas.
1816 (md_parse_option, md_show_usage): Likewise.
1817 * doc/c-sh.texi: Likewise.
1819 2004-04-01 Dave Korn <dk@artimi.com>
1821 * config/tc-dlx.c (md_assemble): set fx_no_overflow flag for
1822 hi16 and lo16 fixS structs.
1823 (md_assemble): generate bit_fixS for RELOC_DLX_LO16 in
1824 exactly the same way as for RELOC_DLX_REL16.
1825 (machine_ip): properly respect LO flag in the_insn and
1826 output RELOC_DLX_LO16 rather than RELOC_DLX_16.
1827 (md_apply_fix3): apply RELOC_DLX_LO16.
1829 2004-03-30 Stan Shebs <shebs@apple.com>
1831 Remove long-obsolete MPW support.
1832 * mpw-config.in, mpw-make.sed, mac-as.r: Remove files.
1833 * configure.in: Remove mention of ppc-*-mpw* config.
1834 * configure.in: Likewise.
1836 2004-03-30 Nick Clifton <nickc@redhat.com>
1838 * config/tc-arm.c (meabi_flags): Make its use conditional upon
1839 OBJ_ELF being defined.
1841 2004-03-27 Alan Modra <amodra@bigpond.net.au>
1843 * config/obj-aout.c (obj_aout_type): Remove #ifdef BFD_ASSEMBLER code.
1845 2004-03-23 Paul Brook <paul@codesourcery.com>
1847 * config/tc-arm.c (meabi_flags): New variable.
1848 (arm_parse_eabi): New function.
1849 (md_begin): Set flags for EABI v3.
1851 (arm_long_opts): Add meabi.
1852 * doc/as.texinf <ARM>: Document -meabi.
1853 * doc/c-arm.texi: Ditto.
1855 2004-03-22 Bob Wilson <bob.wilson@acm.org>
1857 * config/tc-xtensa.c (xtensa_post_relax_hook): Create literal
1858 tables even when use_literal_section flag is not set.
1860 2004-03-22 Alan Modra <amodra@bigpond.net.au>
1862 * config/tc-sh.c: Remove trailing whitespace.
1864 2004-03-22 Hans-Peter Nilsson <hp@axis.com>
1866 * doc/c-cris.texi (CRIS-Opts): Document --no-mul-bug-abort,
1867 --mul-bug-abort and the default behavior.
1868 * config/tc-cris.c (cris_insn_kind): New member CRIS_INSN_MUL.
1869 (err_for_dangerous_mul_placement): New variable.
1870 (STATE_MUL, OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): New
1872 (md_cris_relax_table): Have placeholder for STATE_MUL.
1873 (md_longopts): New options --mul-bug-abort and --no-mul-bug-abort.
1874 (cris_relax_frag) <case ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: New
1876 (md_estimate_size_before_relax) <case ENCODE_RELAX (STATE_MUL,
1877 STATE_BYTE)>: Ditto.
1878 (md_convert_frag) <ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: Check
1879 alignment and position of this frag, emit error message if
1881 (md_assemble): For a multiply insn and when checking it,
1882 transform the current frag into a special frag for that purpose.
1883 (md_parse_option) <case OPTION_MULBUG_ABORT_OFF, case
1884 OPTION_MULBUG_ABORT_ON>: Handle new options.
1886 2004-03-19 Bob Wilson <bob.wilson@acm.org>
1888 * config/tc-xtensa.c (mark_literal_frags): New function.
1889 (xtensa_move_literals): Call mark_literal_frags for all literal
1890 segments, including init and fini literal segments.
1891 (xtensa_post_relax_hook): Swap use of xt_insn_sec and xt_literal_sec.
1893 2004-03-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1895 * tc-hppa.c (cons_fix_new_hppa): Check for PC relative base type.
1896 (pa_comm): Set BSF_OBJECT in symbol flags.
1898 2004-03-19 Alan Modra <amodra@bigpond.net.au>
1900 * Makefile.am: Run "make dep-am".
1901 * Makefile.in: Regenerate.
1902 * doc/Makefile.in: Regenerate.
1903 * config.in: Regenerate.
1904 * po/gas.pot: Regenerate.
1906 2004-03-18 Nathan Sidwell <nathan@codesourcery.com>
1908 * read.c (read_a_source_file): Use demand_empty_rest_of_line.
1909 (demand_empty_rest_of_line): Issue an error here.
1910 (ignore_rest_of_line): Silently skip to end.
1911 (demand_copy_string): Issue an error, not warning.
1913 * config/obj-elf.c (obj_elf_section_name): Likewise.
1914 (obj_elf_section): Likewise.
1915 * config/tc-arc.c (arc_extoper): Remove bogus NULL checks.
1916 (arc_extinst): Likewise.
1917 * config/tc-ia64.c (dot_saveb): Use demand_empty_rest_of_line.
1918 (dot_spill): Likewise.
1919 (dot_unwabi): Likewise.
1920 (dot_prologue): Likewise.
1922 * expr.c (operand): Reject ++ and --.
1923 (operator): Likewise.
1925 2004-03-17 Kaz Kojima <kkojima@rr.iij4u.or.jp>
1927 * config/tc-sh.c: Include dw2gencfi.h.
1928 (sh_cfi_frame_initial_instructions): New function.
1929 (sh_regname_to_dw2regnum): Likewise.
1930 * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Move to the end of
1932 (TARGET_USE_CFIPOP): Define.
1933 (tc_cfi_frame_initial_instructions): Likewise.
1934 (tc_regname_to_dw2regnum): Likewise.
1935 (DWARF2_DEFAULT_RETURN_COLUMN, DWARF2_CIE_DATA_ALIGNMENT): Likewise.
1936 * Makefile.am: Update dependencies.
1937 * Makefile.in: Regenerate.
1939 2004-03-17 Ralf Corsepius <corsepiu@faw.uni-ulm.de>
1941 * configure.in: Switch sh-*-rtems* to ELF. Add sh-*-rtemscoff*.
1942 * configure: Regenerate.
1944 2004-03-12 Bob Wilson <bob.wilson@acm.org>
1946 * read.c (s_leb128): Call md_flush_pending_output.
1948 2004-03-12 Michal Ludvig <mludvig@suse.cz>
1950 * config/tc-i386.c (output_insn): Handle PadLock instructions.
1951 * config/tc-i386.h (CpuPadLock): New define.
1952 (CpuUnknownFlags): Added CpuPadLock.
1954 2004-03-07 Andreas Schwab <schwab@suse.de>
1956 * doc/c-hppa.texi (HPPA Directives): Fix typo.
1958 2004-03-07 Richard Henderson <rth@redhat.com>
1960 * dw2gencfi.c (output_cie): Align length to 4 byte boundary.
1961 (cfi_finish): Likewise for fde.
1963 2004-03-05 H.J. Lu <hongjiu.lu@intel.com>
1965 * config/tc-ia64.c (md_assemble): Properly handle NULL
1967 (ia64_handle_align): Don't abort if failed to add a stop bit.
1969 2004-03-04 H.J. Lu <hongjiu.lu@intel.com>
1971 * Makefile.in: Regenerated.
1972 * aclocal.m4: Likewise.
1973 * configure: Likewise.
1974 * doc/Makefile.in: Likewise.
1976 2004-03-03 H.J. Lu <hongjiu.lu@intel.com>
1978 * config/tc-ia64.c (dot_align): New.
1979 (ia64_do_align): Make it static.
1980 (md_pseudo_table): Use "dot_align" for "align".
1981 (ia64_md_do_align): Don't set align_frag here.
1982 (ia64_handle_align): Add a stop bit to the previous bundle if
1985 * config/tc-ia64.h (ia64_do_align): Removed.
1987 2003-03-03 Andrew Stubbs <andrew.stubbs@superh.com>
1989 * config/tc-sh.c (md_parse_option): Add -isa=sh4-nofpu and
1990 -isa=sh4-nommu-nofpu options. Adjust help messages accordingly.
1991 (sh_elf_final_processing): Output BFD type sh4_nofpu if that is
1992 the most general type or the user specifically requested it.
1993 (md_assemble): Add a new error message for when an instruction
1994 is understood, but is not allowed due to an -isa option.
1996 2004-03-02 H.J. Lu <hongjiu.lu@intel.com>
1998 * config/tc-ia64.c (align_frag): New.
1999 (md_assemble): Set the tc_frag_data field in align_frag for
2000 IA64_OPCODE_FIRST instructions.
2001 (ia64_md_do_align): Set align_frag.
2002 (ia64_handle_align): Add a stop bit if needed.
2004 * config/tc-ia64.h (TC_FRAG_TYPE): New.
2005 (TC_FRAG_INIT): New.
2007 2004-03-01 Richard Sandiford <rsandifo@redhat.com>
2009 * config/tc-frv.c (fr400_audio): New variable.
2010 (md_parse_option, md_show_usage): Add -mcpu=fr405 and -mcpu=fr450.
2011 (md_parse_option): Set fr400_audio for -mcpu=fr400 and -mcpu=fr405.
2012 (target_implements_insn_p): New function.
2013 (md_assemble): Report an error if the processor doesn't implement
2016 2004-02-27 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2018 * config/tc-m32r.c (md_longopts): Added -no-bitinst option.
2019 (md_parse_option): Ditto.
2020 (OPTION_NO_SPECIAL_M32R): Added.
2021 (md_show_usage): Document it.
2022 (enable_speial_m32r): Changed a default value from 0 to 1.
2023 * doc/c-m32r.texi: Document the -no-bitinst option.
2025 2004-02-27 Nick Clifton <nickc@redhat.com>
2027 * config/tc-sh.c (get_operand): Revert previous delta.
2028 (tc_gen_reloc): Check for an unknown reloc type before processing
2031 2004-02-27 Hannes Reinecke <hare@suse.de>
2033 * config/tc-s390.c (s390_insn): Correct range check for opcode in
2034 .insn pseudo operation.
2036 2004-02-27 Anil Paranjpe <anilp1@kpitcummins.com>
2038 * config/tc-sh.c (get_operand): In case of #Imm, check has been
2039 added for wrong syntax.
2041 2004-02-26 Eric Christopher <echristo@redhat.com>
2043 * config/tc-mips.c (mips_dwarf2_addr_size): New.
2044 * config/tc-mips.h (DWARF2_ADDR_SIZE): Use.
2046 2004-02-26 Andrew Stubbs <andrew.stubbs@superh.com>
2048 * config/tc-sh.c (build_Mytes): Add REG_N_D and REG_N_B01
2049 nibble types to assembler.
2051 2004-02-25 Fred Fish <fnf@redhat.com>
2053 * config/tc-iq2000.c: Add missing \n\ in multiline string literal.
2055 2004-02-20 James E Wilson <wilson@specifixinc.com>
2057 * config/tc-ia64.c (slot_index): New arg before_relax. Use instead of
2059 (fixup_unw_records): New arg before_relax. Pass to slot_index.
2060 (ia64_estimate_size_before_relax): New.
2061 (ia64_convert_frag): Pass 0 to fixup_unw_records. Add comment.
2062 (generate_unwind_image): Pass 1 to fixup_unw_records.
2063 * config/tc-ia64.h (ia64_estimate_size_before_relax): Declare.
2064 (md_estimate_size_before_relax): Call ia64_estimate_size_before_relax.
2066 2004-02-19 Jakub Jelinek <jakub@redhat.com>
2068 * stabs.c (generate_asm_file): Avoid warning about use of
2069 uninitialized variable.
2071 2004-02-18 David Mosberger <davidm@hpl.hp.com>
2073 * config/tc-ia64.c (ia64_flush_insns): In addition to prologue,
2074 body, and endp, allow unwind records which do not have a "t"
2075 (time/instruction) field.
2077 2004-02-17 Petko Manolov <petkan@nucleusys.com>
2079 * config/tc-arm.c (do_mav_dspsc_1): Correct offset of CRn.
2080 (do_mav_dspsc_2): Likewise.
2081 Fix accumulator registers move opcodes.
2083 2004-02-13 Hannes Reinecke <hare@suse.de>
2084 Jakub Jelinek <jakub@redhat.com>
2086 * dwarf2dbg.c (get_filenum): Do not read beyond allocated memory.
2088 2004-02-10 Steve Ellcey <sje@cup.hp.com>
2090 * config/tc-ia64.h (ia64_frob_symbol): New declaration.
2091 (tc_frob_symbol): New macro definition.
2092 * config/tc-ia64.c (ia64_frob_symbol): New routine.
2094 2004-02-09 Daniel Jacobowitz <drow@mvista.com>
2096 * config/tc-arm.c (md_begin): Mark .note.gnu.arm.ident as
2099 2004-02-09 Nathan Sidwell <nathan@codesourcery.com>
2101 * read.h (IGNORE_OPCODE_CASE): Do not define. Replace with ...
2102 (TC_CASE_SENSITIVE): ... this.
2103 * read.c: Replace IGNORE_OPCODE_CASE with TC_CASE_SENSITIVE.
2104 * doc/internals.texi (TC_CASE_SENSITIVE): Document.
2106 2004-02-06 James E Wilson <wilson@specifixinc.com>
2108 * config/tc-ia64.c (dot_endp): Delete call to output_endp.
2109 (generate_unwind_image): Re-add it here.
2111 2004-02-06 Nathan Sidwell <nathan@codesourcery.com>
2113 * dwarf2dbg.c (DWARF2_ADDR_SIZE): Remove trailing ';'
2114 * read.h (SKIP_WHITESPACE): Turn into an expression.
2115 * read.c (read_a_source_file): A pseudo is removed by having a
2118 2004-02-05 James E Wilson <wilson@specifixinc.com>
2120 * config/tc-ia64.c (output_endp): New.
2121 (count_bits): Delete.
2122 (ia64_flush_insns, process_one_record, optimize_unw_records): Handle
2123 endp unwind records.
2124 (fixup_unw_records): Handle endp unwind records. Delete code for
2125 shortening prologue regions not followed by a body record.
2126 (dot_endp): Call add_unwind_entry to emit endp unwind record.
2127 * config/tc-ia64.h (unw_record_type): Add endp.
2129 2004-02-03 James E Wilson <wilson@specifixinc.com>
2131 * config/tc-ia64.c (ia64_convert_frag): Call md_number_to_chars to
2132 fill padding bytes with zeroes.
2133 (emit_one_bundle): New locals last_ptr, end_ptr. Rewrite code that
2134 sets unwind_record slot_number and slot_frag fields.
2136 2004-02-02 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2138 * config/tc-mips.c (add_got_offset_hilo): New function.
2139 (macro): Use load_register() and add_got_offset_hilo() to load
2140 constants instead of hardcoding code sequences throughout.
2142 2004-01-28 H.J. Lu <hongjiu.lu@intel.com>
2144 * config/tc-ia64.c (emit_one_bundle): Add proper indentation.
2146 2004-01-26 Bernardo Innocenti <bernie@develer.com>
2148 * config/tc-m68k.h (EXTERN_FORCE_RELOC): Handle m68k-uclinux specially,
2150 * config/tc-m68k.c (RELAXABLE_SYMBOL): Use EXTERN_FORCE_RELOC instead
2151 of hard-coded test for TARGET_OS=elf.
2153 2004-01-24 Chris Demetriou <cgd@broadcom.com>
2155 * config/tc-mips.c (hilo_interlocks): Change definition
2156 so that MIPS32, MIPS64 and later ISAs are included, along with
2157 the already-included machines. Update comments.
2159 2004-01-23 Daniel Jacobowitz <drow@mvista.com>
2161 * config/tc-arm.c (tc_gen_reloc): Improve error message for
2162 undefined local labels.
2164 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2166 * config/tc-mips.c (load_address, macro): Update comments about
2167 NewABI GP relaxation.
2169 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2171 * config/tc-mips.c (macro_build): Remove place and counter arguments.
2172 (mips_build_lui, macro_build_ldst_constoffset): Likewise.
2173 (mips16_macro_build, macro_build_jalr): Remove counter argument.
2174 (set_at, load_register, load_address, move_register): Likewise.
2175 (load_got_offset, add_got_offset): Likewise.
2176 Update all calls and tidy accordingly.
2178 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2180 * config/tc-mips.c (RELAX_ENCODE): Remove WARN argument.
2181 (RELAX_FIRST, RELAX_SECOND): Turn into 8-bit quantities.
2182 (RELAX_USE_SECOND): Bump to 0x10000.
2183 (RELAX_SECOND_LONGER, RELAX_NOMACRO, RELAX_DELAY_SLOT): New flags.
2184 (mips_macro_warning): New variable.
2185 (md_assemble): Wrap macro expansion in macro_start() and macro_end().
2186 (s_cpload, s_cpsetup, s_cprestore, s_cpreturn): Likewise.
2187 (relax_close_frag): Set mips_macro_warning.first_frag. Adjust use
2189 (append_insn): Update mips_macro_warning.sizes.
2190 (macro_start, macro_warning, macro_end): New functions.
2191 (macro_build): Don't emit warnings here.
2192 (macro_build_lui, md_estimate_size_before_relax): ...or here.
2193 (md_convert_frag): Check for cases where one macro alternative
2194 needs a warning and the other doesn't. Emit a warning if the
2195 longer sequence was chosen.
2197 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2199 * config/tc-mips.h (tc_frag_data_type, TC_FRAG_TYPE): Remove.
2200 * config/tc-mips.c (RELAX_ENCODE): Take three arguments: the size of
2201 the first sequence, the size of the second sequence, and a flag
2202 that says whether we should warn.
2203 (RELAX_OLD, RELAX_NEW, RELAX_RELOC[123]): Delete.
2204 (RELAX_FIRST, RELAX_SECOND): New.
2205 (mips_relax): New variable.
2206 (relax_close_frag, relax_start, relax_switch, relax_end): New fns.
2207 (append_insn): Remove "place" argument. Use mips_relax.sequence
2208 rather than "place" to check whether we're expanding the second
2209 alternative of a relaxable macro. Remove redundant check for
2210 branch relaxation. If generating a normal insn, and there
2211 is not enough room in the current frag, call relax_close_frag()
2212 to close it. Update mips_relax.sizes[]. Emit fixups for the
2213 second version of a relaxable macro. Record the first relaxable
2214 fixup in mips_relax. Remove tc_gen_reloc workaround.
2215 (macro_build): Remove all uses of "place". Use mips_relax.sequence
2216 in the same way as in append_insn.
2217 (mips16_macro_build): Remove "place" argument.
2218 (macro_build_lui): As for macro_build. Don't drop the add_symbol
2219 when generating the second version of a relaxable macro.
2220 (load_got_offset, add_got_offset): New functions.
2221 (load_address, macro): Use new relaxation machinery. Remove
2222 tc_gen_reloc workarounds.
2223 (md_estimate_size_before_relax): Set RELAX_USE_SECOND if the second
2224 version of a relaxable macro is needed. Return -RELAX_SECOND if the
2225 first version is needed.
2226 (tc_gen_reloc): Remove relaxation handling.
2227 (md_convert_frag): Go through the fixups for a relaxable macro and
2228 mark those that belong to the unneeded alternative as done. If the
2229 second alternative is needed, adjust the fixup addresses to account
2230 for the deleted first alternative.
2232 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2234 * frags.h (frag_room): Declare.
2235 * frags.c (frag_room): New function.
2236 * doc/internals.texi: Document it.
2238 2004-01-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
2240 * config/tc-mips.c (append_insn): Don't do r3900 interlock
2241 optimization for -mtune=r3900, as this will break on other CPUs.
2243 2004-01-11 Tom Rix <tcrix@worldnet.att.net>
2245 * config/tc-m68hc11.c (build_indexed_byte): movb and movw cannot
2246 be relaxed, use fixup.
2247 (md_apply_fix3): Use 5 bit reloc from movb and movw fixup.
2249 2004-01-19 Jakub Jelinek <jakub@redhat.com>
2251 * config/tc-sparc.c (sparc_ip): Disallow %f32-%f63 for single
2254 2004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2256 * config/tc-mips.c (append_insn): Properly detect variant frags
2257 that preclude swapping of relaxed branches. Correctly swap
2258 instructions between frags when dealing with relaxed branches.
2260 2004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2262 * acinclude.m4: Quote names of macros to be defined by AC_DEFUN
2264 * aclocal.m4: Regenerate.
2265 * configure: Regenerate.
2267 2004-01-12 Anil Paranjpe <anilp1@KPITCummins.com>
2269 * config/tc-h8300.c (build_bytes): Apply relaxation to bit
2272 2004-01-12 Richard Sandiford <rsandifo@redhat.com>
2274 * config/tc-mips.c (macro_build_jalr): When adding an R_MIPS_JALR
2275 reloc, reserve space for the delay slot as well as the jalr itself.
2277 2004-01-09 Paul Brook <paul@codesourcery.com>
2279 * config/tc-arm.c (do_vfp_reg2_from_sp2): Rename from do_vfp_sp_reg2.
2280 (do_vfp_sp2_from_reg2): New function.
2282 (do_vfp_dp_from_reg2): Check return values properly.
2284 2004-01-08 Ian Lance Taylor <ian@wasabisystems.com>
2286 * config/tc-mips.c (warn_nops): Remove static variable.
2287 (macro): Remove test of warn_nops.
2288 (md_shortops): Remove 'n'.
2289 (md_parse_option): Remove 'n' case.
2290 (md_show_usage): Remove -n.
2291 * doc/as.texinfo (Overview): Remove MIPS -n option.
2292 * doc/c-mips.texi (MIPS Opts): Remove mention -n.
2293 * NEWS: Mention removal of MIPS -n option.
2295 * config/tc-mips.c (ISA_HAS_COPROC_DELAYS): Remove.
2296 (cop_interlocks): Check ISA level.
2297 (cop_mem_interlocks): Define.
2298 (reg_needs_delay): Check cop_interlocks rather than
2299 ISA_HAS_COPROC_DELAYS.
2300 (append_insn): Likewise. Use cop_mem_interlocks rather than
2301 directly checking mips_opts.isa.
2302 (mips_emit_delays): Likewise.
2304 2004-01-07 H.J. Lu <hongjiu.lu@intel.com>
2306 * config/tc-ia64.c (unwind): Move next_slot_number and
2307 next_slot_frag to ...
2308 (unw_rec_list): Here.
2309 (free_list_records): Removed.
2310 (output_unw_records): Likewise.
2311 (generate_unwind_image): Make it void.
2312 (alloc_record): Initialize next_slot_number and next_slot_frag.
2313 (slot_index): Take .org, .space and .align into account.
2314 (fixup_unw_records): Don't set slot_number to 0. Use
2315 list->next_slot_number and list->next_slot_frag instead of
2316 unwind.next_slot_number and unwind.next_slot_frag.
2317 (ia64_convert_frag): New.
2318 (generate_unwind_image): Generate a rs_machine_dependent frag
2320 (emit_one_bundle): Use list->next_slot_number and
2321 list->next_slot_frag instead of unwind.next_slot_number and
2322 unwind.next_slot_frag.
2324 * config/tc-ia64.h (md_convert_frag): Defined as
2326 (md_estimate_size_before_relax): Defined as (f)->fr_var.
2328 2004-01-06 Alexandre Oliva <aoliva@redhat.com>
2330 2003-12-19 Alexandre Oliva <aoliva@redhat.com>
2331 * config/tc-frv.h (md_apply_fix3): Don't define.
2332 * config/tc-frv.c (md_apply_fix3): New. Shift/truncate %hi/%lo
2334 * config/tc-frv.h (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
2335 2003-10-07 Alexandre Oliva <aoliva@redhat.com>
2336 * config/tc-frv.c (line_separator_chars): Add `!'.
2337 2003-09-19 Alexandre Oliva <aoliva@redhat.com>
2338 * config/tc-frv.c (md_assemble): Clear insn upfront.
2339 2003-09-18 Alexandre Oliva <aoliva@redhat.com>
2340 * config/tc-frv.c (OPTION_FDPIC): New macro.
2341 (md_longopts): Add mfdpic.
2342 (md_parse_option): Handle it.
2343 2003-08-04 Alexandre Oliva <aoliva@redhat.com>
2344 * config/tc-frv.c (md_cgen_lookup_reloc) <FRV_OPERAND_D12,
2345 FRV_OPERAND_S12>: Use reloc type encoded in fix-up.
2346 (frv_pic_ptr): Parse funcdesc.
2348 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2350 * doc/as.texinfo: Let texi2pod parse asconfig.texi and
2351 gasver.texi. Remove duplicate symbol definitions for texi2pod.
2353 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2355 * Makefile.am (Makefile): Move the dependency on
2356 $(BFDDIR)/configure.in to...
2357 (CONFIG_STATUS_DEPENDENCIES): ... here.
2358 (AUTOMAKE_OPTIONS): Require automake 1.8.
2359 * Makefile.in: Regenerate.
2360 * doc/Makefile.am (BASEDIR, BFDDIR): Define.
2361 (CONFIG_STATUS_DEPENDENCIES): Add a dependency on
2362 $(BFDDIR)/configure.in here as well.
2363 * doc/Makefile.in: Regenerate.
2365 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2367 * Makefile.am (install, install-info, RECURSIVE_TARGETS): Remove.
2368 * Makefile.in: Regenerate.
2369 * aclocal.m4: Regenerate.
2370 * doc/Makefile.am (install, install-info): Remove.
2371 (install-data-local): A new hook for install-info.
2372 (AUTOMAKE_OPTIONS): Require automake 1.8.
2373 * doc/Makefile.in: Regenerate.
2375 2004-01-02 Nutan Singh <nutan@kpitcummins.com>
2377 * doc/c-sh.texi: Update description about floating point behavior
2380 2004-01-02 Bernardo Innocenti <bernie@develer.com>
2382 * configure.in: Add m68k-uClinux target.
2383 * configure: Regenerate.
2385 For older changes see ChangeLog-0203
2391 version-control: never