1 2004-11-05 Tomer Levi <Tomer.Levi@nsc.com>
3 * config/tc-crx.c: Rename argument types.
4 (processing_arg_number): Rename to 'cur_arg_num'.
5 (get_number_of_bits): Rename to 'set_operand_size'.
6 (get_operandtype): Rename to 'parse_operand', totally rewrite.
7 (set_cons_rparams): Rename to 'set_operand', totally rewrite.
8 (set_indexmode_parameters): Remove function, integrate its code into
10 (set_operand_size): Get rid of 'Operand Number' function parameter -
11 use global variable 'cur_arg_num' instead.
12 Use a local 'argument' pointer to reference the current argument.
13 (parse_operand): Likewise.
14 (set_operand): Likewise.
15 (process_label_constant): Likewise.
17 2004-11-04 Bob Wilson <bob.wilson@acm.org>
19 * config/tc-xtensa.c: Remove XTENSA_SECTION_RENAME ifdefs.
20 (add_section_rename): Delete. Inlined into...
21 (build_section_rename): ...here. Use xstrdup instead of strdup.
22 (xtensa_section_rename): Drop "const" from argument and return types.
23 (md_show_usage): Indent to match show_usage().
24 * config/tc-xtensa.h: Remove XTENSA_SECTION_RENAME ifdefs.
25 (tc_canonicalize_section_name): Define.
26 (md_elf_section_rename): Remove unused macro.
27 * doc/as.texinfo (Overview): Document Xtensa --rename-section option.
28 * doc/c-xtensa.texi (Xtensa Options): Likewise.
29 (Frame Directive): Delete.
31 2004-11-04 Daniel Jacobowitz <dan@debian.org>
33 * configure.in: Remove arm-*-oabi and thumb-*-oabi.
34 * config/tc-arm.c (target_oabi): Delete.
35 (md_apply_fix3, elf32_arm_target_format): Remove target_oabi checks.
36 (arm_opts): Remove moabi.
37 * doc/as.texinfo (Overview): Remove documentation of -moabi.
38 * doc/c-arm.texi (ARM Options): Likewise.
39 * configure: Regenerated.
41 2004-11-04 Hans-Peter Nilsson <hp@axis.com>
43 * configure.in (crisv32): Recognize. AC_DEFINE_UNQUOTED
44 DEFAULT_CRIS_ARCH. Handle crisv32-*-linux-gnu* like
45 cris-*-linux-gnu* and crisv32-*-* like cris-*-*.
46 * configure: Regenerate.
47 * config/tc-cris.c (enum cris_archs): New.
48 (cris_mach, cris_arch_from_string, s_cris_arch, get_sup_reg)
49 (cris_insn_ver_valid_for_arch): New functions.
50 (DEFAULT_CRIS_ARCH): New macro, default to cris_any_v0_v10.
51 (cris_arch): New variable.
52 (md_pseudo_table): New pseudo .arch.
53 (err_for_dangerous_mul_placement): Initialize according to
55 (STATE_COND_BRANCH): Renamed from STATE_CONDITIONAL_BRANCH.
57 (STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON)
58 (STATE_ABS_BRANCH_V32, STATE_LAPC, BRANCH_BF_V32, BRANCH_BB_V32)
59 (BRANCH_WF_V32, BRANCH_WB_V32): New.
60 (BRANCH_BF, BRANCH_BB, BRANCH_WF, BRANCH_WB): Don't undef after
61 use in md_cris_relax_table.
62 (md_cris_relax_table): Add entries for STATE_COND_BRANCH_V32,
63 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
64 Update and improve head comment.
65 (OPTION_PIC): Define in terms of previous option, OPTION_US.
66 (OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): Similar.
68 (md_longopts): New option --march=...
69 (cris_any_v0_v10_long_jump_size, crisv32_long_jump_size): New
71 (md_long_jump_size): Initialize in terms of DEFAULT_CRIS_ARCH.
72 (HANDLE_RELAXABLE): New macro.
73 (md_estimate_size_before_relax): Use HANDLE_RELAXABLE for common
74 cases. Check for weak symbols and assume not relaxable. Handle
75 STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON,
76 STATE_ABS_BRANCH_V32, STATE_LAPC. Use new variable symbolP, not
78 (md_convert_frag): Handle STATE_COND_BRANCH_V32,
79 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
80 (cris_create_short_jump): Adjust for CRISv32.
81 (cris_relax_frag): Handle new states.
82 (md_create_long_jump): Ditto. Emit error for common_v10_v32.
83 (md_begin): Define symbols "..asm.arch.cris.v32",
84 "..asm.arch.cris.v10", "..asm.arch.cris.common_v10_v32" and
85 "..asm.arch.cris.any_v0_v10". Use cris_insn_ver_valid_for_arch
86 when entering opcode table entry points.
87 (md_assemble): Adjust branch handling for CRISv32. Handle LAPC
88 relaxation. In fix_new_exp call for main insn, pass 1 for pcrel
89 parameter for 8, 16 and 32-bit pc-relative insns and LAPC.
90 (cris_process_instruction): Initialize out_insnp->insn_type to
91 CRIS_INSN_NONE, not CRIS_INSN_NORMAL.
92 <case ']', '[', 'A', 'd', 'Q', 'N', 'n', 'Y', 'U', 'u', 'T'>: New
94 <case 'm'>: Check that modified_char == '.'.
95 <invalid operands>: Consume the rest of the line.
96 When operands don't match, skip over subsequent insns with
97 non-matching version specifier but same mnemonic.
98 <immediate constant, case SIZE_SPEC_REG>: Immediate operands for
99 special registers in CRISv32 are always 32 bit long.
100 <immediate constant, case SIZE_FIELD_SIGNED, SIZE_FIELD_UNSIGNED>:
102 (get_gen_reg): Only recognize "PC" when followed by "+]" for v32
103 and compatible. Recognize "ACR" for v32, unless followed by "+".
104 (get_spec_reg): Consider cris_arch when looking up register.
105 (get_autoinc_prefix_or_indir_op): Don't recognize assignment for
107 (get_3op_or_dip_prefix_op): Check for ']' after seeing '[rN+'.
108 (cris_get_expression): Restore input_line_pointer if failing "early".
109 (get_flags): Consider cris_arch and recognize flags accordingly.
110 (branch_disp): Adjust for CRISv32.
111 (gen_cond_branch_32): Similar. Emit error for common_v10_v32.
112 (cris_number_to_imm): Use as_bad_where, not as_bad. Remove
113 related FIXME. Don't insist on BFD_RELOC_32_PCREL fixup to be
114 resolved. Don't enter zeros in object file for
116 <case BFD_RELOC_CRIS_LAPCQ_OFFSET, BFD_RELOC_CRIS_SIGNED_16>
117 <case BFD_RELOC_CRIS_SIGNED_8>: New case.
118 (md_parse_option): Break out "return 1".
119 <OPTION_ARCH> New case.
120 (tc_gen_reloc): <case BFD_RELOC_CRIS_LAPCQ_OFFSET>
121 <case BFD_RELOC_CRIS_SIGNED_16, BFD_RELOC_CRIS_SIGNED_8>
122 <case BFD_RELOC_CRIS_UNSIGNED_8, BFD_RELOC_CRIS_UNSIGNED_16>
123 <case BFD_RELOC_32_PCREL>: New cases.
124 Addends for non-zero fx_pcrel are too in fx_offset.
125 (md_show_usage): Show --march=<arch>.
126 (md_apply_fix3): Adjust val for BFD_RELOC_CRIS_LAPCQ_OFFSET.
127 (md_pcrel_from): BFD_RELOC_CRIS_LAPCQ_OFFSET is PC-relative too.
128 (s_syntax) <struct syntaxes>: Properly constify member operand.
129 * config/tc-cris.h (TARGET_MACH): Define.
130 (cris_mach): Declare.
131 * doc/as.texinfo (Overview) <CRIS>: Add --march=...
132 * doc/c-cris.texi (CRIS-Symbols): New node for built-in symbols.
133 (CRIS-Opts): Document --march=...
134 (CRIS-Pseudos): Document .arch.
136 2004-11-04 Jan Beulich <jbeulich@novell.com>
138 * config/tc-i386.c (set_intel_syntax): Allow % in symbol names when
139 intel syntax and no register prefix, allow $ in symbol names when
141 (set_16bit_gcc_code_flag): Replace literal 'l' by LONG_MNEM_SUFFIX.
142 (intel_float_operand): Add fourth return value indicating math control
143 operations. Make classification more precise.
144 (md_assemble): Complain if memory operand of mov[sz]x has no size
146 (parse_insn): Translate word operands to floating point instructions
147 operating on integers as well as control instructions to short ones
148 as expected by AT&T syntax. Translate 'd' suffix to short one only for
149 floating point instructions operating on non-integer operands.
150 (match_template): Remove fldcw special case. Adjust q-suffix handling
151 to permit it on fild/fistp/fisttp in AT&T mode.
152 (process_suffix): Don't guess DefaultSize insns' suffix from
153 stackop_size for certain floating point control instructions. Guess
154 suffix for branch and [ls][gi]dt based on flag_code. Split error
155 messages for Intel and AT&T syntax, and make the condition more strict
156 for the former. Adjust suppressing of generation of operand size
158 (intel parser): Allow the full set of MASM operators. Add FWORD, TBYTE,
159 OWORD, and XMMWORD operand size specifiers (TBYTE replaces XWORD). Add
161 * config/tc-i386.h (BYTE_PTR WORD_PTR DWORD_PTR QWORD_PTR XWORD_PTR
162 SHORT OFFSET_FLAT FLAT NONE_FOUND): Remove unused defines.
164 2004-11-03 Hans-Peter Nilsson <hp@axis.com>
166 * symbols.c (colon) [!WORKING_DOT_WORD]: Don't declare
167 md_short_jump_size, md_long_jump_size.
168 * write.c [!WORKING_DOT_WORD]: Ditto.
169 * tc.h [!WORKING_DOT_WORD]: Declare them here. Drop const
171 * config/tc-cris.h (md_short_jump_size, md_long_jump_size): Don't
173 * config/tc-cris.c (md_short_jump_size, md_long_jump_size): Drop
174 const qualifier in these definitions.
175 * config/tc-i370.c, config/tc-m68k.c, config/tc-pdp11.c,
176 config/tc-s390.c, config/tc-tahoe.c, config/tc-vax.c: Ditto.
178 2004-11-02 Nick Clifton <nickc@redhat.com>
180 * dwarf2dbg.c (dwarf2_finish): Check for the existence of a file
181 table before deciding to produce a .debug_line section to match up
182 with a user provided .debug_info section.
184 2004-10-28 Tomer Levi <Tomer.Levi@nsc.com>
186 * config/tc-crx.c (getreg_image): Bug fix, a return value was
187 mistakenly omitted from CRX_C_REGTYPE and CRX_CS_REGTYPE cases.
189 2004-10-27 Tomer Levi <Tomer.Levi@nsc.com>
191 * config/tc-crx.c: Remove global variable 'post_inc_mode'.
192 (get_flags): New function.
193 (get_number_of_bits): Edit comments, update numeric values to supported
195 (process_label_constant): Don't support the colon format (SYMBOL:[s|m|l]).
196 (set_cons_rparams): Support argument type 'arg_rbase'.
197 (get_operandtype): Bug fix in 'rbase' operand type parsing.
198 (handle_LoadStor): Bug fix, first handle post-increment mode.
199 (getreg_image): Remove redundant code, update according to latest CRX spec.
200 (print_constant): Bug fix relate to 3-word instructions.
201 (assemble_insn): Bug fix, when matching instructions, verify also
202 instruction type (not only mnemonic).
203 Add various error checking.
204 (preprocess_reglist): Support HI/LO and user registers.
206 2004-10-25 David Mosberger-Tang <davidm@hpl.hp.com>
208 * config/tc-ia64.c (fixup_unw_records): Don't let the "t" value
209 in an epilogue directive go negative.
211 2004-10-25 H.J. Lu <hongjiu.lu@intel.com>
214 * config/tc-ia64.c (emit_one_bundle): Decrement md.num_slots_in_use
215 after reporting template error during manual bundling. Reported
216 by Michael Dupont, michaelx.dupont@intel.com.
218 2004-10-25 Daniel Jacobowitz <dan@debian.org>
220 * Makefile.am: Run dep-am.
221 * aclocal.m4: Regenerate with automake 1.9.2.
222 * Makefile.in: Regenerate with automake 1.9.2.
223 * doc/Makefile.in: Likewise.
225 * config/tc-arm.c: Include "dw2gencfi.h".
226 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
228 * config/tc-arm.h (TARGET_USE_CFIPOP, DWARF2_DEFAULT_RETURN_COLUMN)
229 (DWARF2_CIE_DATA_ALIGNMENT, tc_regname_to_dw2regnum)
230 (tc_cfi_frame_initial_instructions): Define.
231 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
234 2004-10-21 Tomer Levi <Tomer.Levi@nsc.com>
236 * config/tc-crx.c (assemble_insn): Check unsigned immediate
238 Update coprocessor id to be unsigned immediate.
240 2004-10-18 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
242 * config/tc-i386.c (O_secrel): Delete.
243 (tc_pe_dwarf2_emit_offset): New function.
244 * config/tc-i386.h (O_secrel): Define as O_md1.
245 (TC_DWARF2_EMIT_OFFSET): Define.
247 2004-10-18 Nick Clifton <nickc@redhat.com>
249 * config/tc-xstormy16.c (xstormy16_cons_fix_new): Accept and
250 ignore @fptr() directives for 4-byte fixups.
252 2004-10-15 Alan Modra <amodra@bigpond.net.au>
254 * Makefile.am: Run "make dep-am"
255 * Makefile.in: Regenerate.
257 2004-10-14 Bob Wilson <bob.wilson@acm.org>
259 * doc/c-xtensa.texi (Xtensa Options, Absolute Literals Directive):
260 Remove comments about placement of literal pools.
261 (Literal Directive): Update description of literal placement.
262 (Literal Prefix Directive): Remove statement that this does not apply
263 to absolute-mode literals. Describe new section naming scheme.
265 2004-10-12 Bob Wilson <bob.wilson@acm.org>
267 * config/tc-xtensa.c (xtensa_frequency_pseudo): Use set_subseg_freq.
268 (is_entry_opcode, is_movi_opcode, is_the_loop_opcode, is_jx_opcode,
269 is_windowed_return_opcode): Delete.
270 (xtensa_frob_label): Use get_subseg_target_freq.
271 (md_assemble): Inline call to is_entry_opcode.
272 (xtensa_handle_align): Inline call to get_frag_is_literal.
273 (relaxation_requirements): Inline call to is_jx_opcode.
274 (emit_single_op): Inline call to is_movi_opcode.
275 (xg_assemble_vliw_tokens): Inline calls to get_frag_is_insn,
276 get_frag_is_no_transform, is_entry_opcode, and
277 set_frag_is_specific_opcode. Use get_subseg_total_freq.
278 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags,
279 xtensa_fix_close_loop_end_frags, relax_frag_immed, convert_frag_immed):
280 Inline calls to get_frag_is_no_transform.
281 (next_instrs_are_b_retw): Inline call to is_windowed_return_opcode.
282 (xtensa_fix_short_loop_frags): Inline calls to is_the_loop_opcode and
283 get_frag_is_no_transform.
284 (convert_frag_immed_finish_loop): Inline calls to get_expression_value
285 and set_frag_is_no_transform.
286 (get_expression_value): Delete.
287 (subseg_map struct): Rename cur_total_freq to total_freq. Rename
288 cur_target_freq to target_freq.
289 (get_subseg_info): Split out code to create a new map entry into ...
290 (add_subseg_info): ... this new function.
291 (get_last_insn_flags): Check if get_subseg_info succeeded.
292 (set_last_insn_flags): Call add_subseg_info if needed.
293 (get_subseg_total_freq, get_subseg_target_freq, set_subseg_freq): New.
294 (xtensa_reorder_segments): Compute last_sec while counting sections.
295 Remove call to get_last_sec.
296 (get_last_sec): Delete.
297 (cache_literal_section): Inline call to retrieve_literal_seg and its
298 callees, seg_present and add_seg_list.
299 (retrieve_literal_seg, seg_present, add_seg_list): Delete.
300 (get_frag_is_insn, get_frag_is_no_transform,
301 set_frag_is_specific_opcode, set_frag_is_no_transform): Delete.
302 * config/tc-xtensa.h (MAX_SLOTS): Reduce from 31 to 15.
304 2004-10-12 Bob Wilson <bob.wilson@acm.org>
306 * config/tc-xtensa.c: Use ISO C90 formatting.
307 * config/tc-xtensa.h: Likewise.
308 * config/xtensa-istack.h: Likewise.
309 * config/xtensa-relax.c: Likewise.
310 * config/xtensa-relax.h: Likewise.
312 2004-10-12 Paul Brook <paul@codesourcery.com>
314 * config/tc-arm.c (md_begin): Change EF_ARM_EABI_VER3 to
317 * doc/c-arm.texi: Document that we actually support -meabi=4, not
320 2004-10-08 Bob Wilson <bob.wilson@acm.org>
322 * doc/as.texinfo (VTableEntry, VTableInherit): Add "directive" to index
324 (Acknowledgements): Use "GAS" instead of AS variable.
326 2004-10-08 Daniel Jacobowitz <dan@debian.org>
328 * config/tc-i386.c: Include "elf/x86-64.h".
329 (i386_elf_section_type): New function.
330 * config/tc-i386.h (md_elf_section_type): Define.
331 (i386_elf_section_type): New prototype.
333 2004-10-08 Linus Nielsen Feltzing <linus@haxx.se>
335 * config/m68k-parse.h (enum m68k_register): New control register,
337 * config/tc-m68k.c: Correct control register set for MCF5249.
339 2004-10-07 Bob Wilson <bob.wilson@acm.org>
340 Sterling Augustine <sterling@tensilica.com>
342 * config/tc-xtensa.c (absolute_literals_supported): New global flag.
343 (UNREACHABLE_MAX_WIDTH): Define.
344 (XTENSA_FETCH_WIDTH): Delete.
345 (cur_vinsn, xtensa_fetch_width, xt_saved_debug_type, past_xtensa_end,
346 prefer_const16, prefer_l32r): New global variables.
347 (LIT4_SECTION_NAME): Define.
348 (lit4_state struct): Add lit4_seg_name and lit4_seg fields.
349 (XTENSA_PROP_*, GET_XTENSA_PROP_*, SET_XTENSA_PROP_*): Define.
350 (frag_flags struct): New.
351 (xtensa_block_info struct): Move from tc-xtensa.h. Add flags field.
352 (subseg_map struct): Add cur_total_freq and cur_target_freq fields.
353 (bitfield, bit_is_set, set_bit, clear_bit): Define.
354 (MAX_FORMATS): Define.
355 (op_placement_info struct, op_placement_table): New.
356 (O_pltrel, O_hi16, O_lo16): Define.
357 (directiveE enum): Rename directive_generics to directive_transform.
358 Delete directive_relax. Add directive_schedule,
359 directive_absolute_literals, and directive_last_directive.
360 (directive_info): Rename "generics" to "transform". Delete "relax".
361 Add "schedule" and "absolute-literals".
362 (directive_state): Adjust entries to match changes in directive_info.
363 (xtensa_relax_statesE, RELAX_IMMED_MAXSTEPS): Move to tc-xtensa.h.
364 (xtensa_const16_opcode, xtensa_movi_opcode, xtensa_movi_n_opcode,
365 xtensa_l32r_opcode, xtensa_nop_opcode, xtensa_rsr_lcount_opcode): New.
366 (xtensa_j_opcode, xtensa_rsr_opcode): Delete.
367 (align_only_targets, software_a0_b_retw_interlock,
368 software_avoid_b_j_loop_end, maybe_has_b_j_loop_end,
369 software_avoid_short_loop, software_avoid_close_loop_end,
370 software_avoid_all_short_loops, specific_opcode): Delete.
371 (warn_unaligned_branch_targets): New.
372 (workaround_a0_b_retw, workaround_b_j_loop_end, workaround_short_loop,
373 workaround_close_loop_end, workaround_all_short_loops): Default FALSE.
374 (option_[no_]link_relax, option_[no_]transform,
375 option_[no_]absolute_literals, option_warn_unaligned_targets,
376 option_prefer_l32r, option_prefer_const16, option_target_hardware):
378 (option_[no_]align_only_targets, option_literal_section_name,
379 option_text_section_name, option_data_section_name,
380 option_bss_section_name, option_eb, option_el): Delete.
381 (md_longopts): Add entries for: [no-]transform, [no-]absolute-literals,
382 warn-unaligned-targets, prefer-l32r, prefer-const16, [no-]link-relax,
383 and target-hardware. Delete entries for [no-]target-align-only,
384 literal-section-name, text-section-name, data-section-name, and
386 (md_parse_option): Handle new options and remove old ones. Accept but
387 ignore [no-]density options. Warn for [no-]generics and [no-]relax
388 and treat them as [no-]transform.
389 (md_show_usage): Add new options and remove old ones.
390 (xtensa_setup_hw_workarounds): New.
391 (md_pseudo_table): Change "word" entry to use xtensa_elf_cons. Add
392 "long", "short", "loc" and "frequency" entries.
393 (use_generics): Rename to ...
394 (use_transform): ... this function. Add past_xtensa_end check.
395 (use_longcalls): Add past_xtensa_end check.
396 (code_density_available, can_relax): Delete.
397 (do_align_targets): New.
398 (get_directive): Accept dashes in directive names. Warn about
399 [no-]generics and [no-]relax directives and treat them as
401 (xtensa_begin_directive): Call md_flush_pending_output only for some
402 directives. Check for directives inside instruction bundles. Warn
403 about deprecated ".begin literal" usage. Warn and ignore [no-]density
404 directives. Handle new directives. Check generating_literals flag
406 (xtensa_end_directive): Check for directives inside instruction
407 bundles. Warn and ignore [no-]density directives. Handle new
408 directives. Call xtensa_set_frag_assembly_state.
409 (xtensa_loc_directive_seen, xtensa_dwarf2_directive_loc,
410 xtensa_dwarf2_emit_insn): New.
411 (xtensa_literal_position): Call md_flush_pending_output. Do not check
412 use_literal_section flag.
413 (xtensa_literal_pseudo): Call md_flush_pending_output. Handle absolute
414 literals. Use xtensa_elf_cons to parse the expression.
415 (xtensa_literal_prefix): Do not check use_literal_section. Support
416 ".lit4" sections for absolute literals. Change prefix convention to
417 replace ".text" (or ".t" in a linkonce section). No need to call
419 (xtensa_frequency_pseudo, xtensa_elf_cons, xtensa_elf_suffix): New.
420 (expression_end): Handle closing braces and colons.
421 (PLT_SUFFIX, plt_suffix): Delete.
422 (expression_maybe_register): Use new xtensa-isa.h functions. Use
423 xtensa_elf_suffix instead of checking for plt suffix, and handle O_lo16
424 and O_hi16 expressions as well.
425 (tokenize_arguments): Handle closing braces and colons.
426 (parse_arguments): Use new xtensa-isa.h functions. Handle "invisible"
427 operands and paired register syntax.
428 (get_invisible_operands): New.
429 (xg_translate_sysreg_op): Handle new Xtensa LX RSR/WSR/XSR syntax. Use
430 new xtensa-isa.h functions.
431 (xtensa_translate_old_userreg_ops, xtensa_translate_zero_immed): New.
432 (xg_translate_idioms): Check if inside bundle. Use use_transform.
433 Handle new Xtensa LX RSR/WSR/XSR syntax. Remove code to widen density
434 instructions. Use xtensa_translate_zero_immed.
435 (operand_is_immed, operand_is_pcrel_label): Delete.
436 (get_relaxable_immed): Use new xtensa-isa.h functions.
437 (get_opcode_from_buf): Add slot parameter. Use new xtensa-isa.h
439 (xtensa_print_insn_table, print_vliw_insn): New.
440 (is_direct_call_opcode): Use new xtensa-isa.h functions.
441 (is_call_opcode, is_loop_opcode, is_conditional_branch_opcode,
442 is_branch_or_jump_opcode): Delete.
443 (is_movi_opcode, decode_reloc, encode_reloc, encode_alt_reloc): New.
444 (opnum_to_reloc, reloc_to_opnum): Delete.
445 (xtensa_insnbuf_set_operand, xtensa_insnbuf_get_operand): Use new
446 xtensa-isa.h functions. Operate on one slot of an instruction.
447 (xtensa_insnbuf_set_immediate_field, is_negatable_branch,
448 xg_get_insn_size): Delete.
449 (xg_get_build_instr_size): Use xg_get_single_size.
450 (xg_is_narrow_insn, xg_is_single_relaxable_insn): Update calls to
451 xg_build_widen_table. Use xg_get_single_size.
452 (xg_get_max_narrow_insn_size): Delete.
453 (xg_get_max_insn_widen_size, xg_get_max_insn_widen_literal_size,
454 xg_is_relaxable_insn): Update calls to xg_build_widen_table. Use
456 (xg_build_to_insn): Record the loc field. Handle OP_OPERAND_HI16U and
457 OP_OPERAND_LOW16U. Check xg_valid_literal_expression.
458 (xg_expand_to_stack, xg_expand_narrow): Update calls to
459 xg_build_widen_table. Use xg_get_single_size.
460 (xg_immeds_fit): Use new xtensa-isa.h functions. Update call to
462 (xg_symbolic_immeds_fit): Likewise. Also handle O_lo16 and O_hi16, and
463 treat weak symbols conservatively.
464 (xg_check_operand): Use new xtensa-isa.h functions.
465 (is_dnrange): Delete.
466 (xg_assembly_relax): Inline previous calls to tinsn_copy.
467 (xg_finish_frag): Specify separate relax states for the frag and slot0.
468 (is_branch_jmp_to_next, xg_add_branch_and_loop_targets): Use new
469 xtensa-isa.h functions.
470 (xg_instruction_matches_option_term, xg_instruction_matches_or_options,
471 xg_instruction_matches_options): New.
472 (xg_instruction_matches_rule): Handle O_register expressions. Call
473 xg_instruction_matches_options.
474 (transition_rule_cmp): New.
475 (xg_instruction_match): Update call to xg_build_simplify_table.
476 (xg_build_token_insn): Record loc fields.
477 (xg_simplify_insn): Check is_specific_opcode field and
478 density_supported flag.
479 (xg_expand_assembly_insn): Skip checking code_density_available. Use
480 new xtensa-isa.h functions. Call use_transform instead of can_relax.
481 (xg_assemble_literal): Add error handling for O_big. Call
482 record_alignment. Handle O_pltrel.
483 (xg_valid_literal_expression): New.
484 (xg_assemble_literal_space): Add slot parameter. Remove call to
485 set_expr_symbol_offset. Add call to record_alignment. Update call to
487 (xg_emit_insn): Delete.
488 (xg_emit_insn_to_buf): Add format parameter. Update calls to
489 xg_add_opcode_fix and xtensa_insnbuf_to_chars.
490 (xg_add_opcode_fix): Change opcode parameter to tinsn and add format
491 and slot parameters. Handle new "alternate" relocations for absolute
492 literals and CONST16 instructions. Check for bad uses of O_lo16 and
493 O_hi16. Use new xtensa-isa.h functions.
494 (xg_assemble_tokens): Delete.
495 (is_register_writer): Use new xtensa-isa.h functions.
496 (is_bad_loopend_opcode): Check for xtensa_rsr_lcount_opcode instead of
497 old-style RSR from LCOUNT.
498 (next_frag_opcode): Delete.
499 (next_frag_opcode_is_loop, next_frag_format_size, frag_format_size,
500 update_next_frag_state): New.
501 (update_next_frag_nop_state): Delete.
502 (next_frag_pre_opcode_bytes): Use next_frag_opcode_is_loop.
503 (xtensa_mark_literal_pool_location): Check use_literal_section flag and
504 the state of the absolute-literals directive. Add calls to
505 record_alignment and xtensa_set_frag_assembly_state. Call
506 xtensa_switch_to_non_abs_literal_fragment instead of
507 xtensa_switch_to_literal_fragment.
509 (assemble_nop): Use build_nop. Update call to xtensa_insnbuf_to_chars.
510 (get_expanded_loop_offset): Change check for undefined opcode to an
512 (xtensa_set_frag_assembly_state, relaxable_section,
513 xtensa_find_unmarked_state_frags, xtensa_find_unaligned_branch_targets,
514 xtensa_find_unaligned_loops, xg_apply_tentative_value): New.
515 (md_begin): Update call to xtensa_isa_init. Initialize linkrelax to 1.
516 Set lit4_seg_name. Call xg_init_vinsn. Initialize new global opcodes.
517 Call init_op_placement_info_table and xtensa_set_frag_assembly_state.
518 (xtensa_init_fix_data): New.
519 (xtensa_frob_label): Reset label symbol to the current frag. Check
520 do_align_targets and generating_literals flag. Propagate frequency
521 info to new alignment frag. Call xtensa_set_frag_assembly_state.
522 (xtensa_unrecognized_line): New.
523 (xtensa_flush_pending_output): Check if inside a bundle. Add a call
524 to xtensa_set_frag_assembly_state.
525 (error_reset_cur_vinsn): New.
526 (md_assemble): Remove check for literal frag. Remove call to
527 istack_init. Call use_transform instead of use_generics. Parse
528 explicit instruction format specifiers. Move code for
529 a0_b_retw_interlock workaround to xg_assemble_vliw_tokens. Call
530 error_reset_cur_vinsn on errors. Add call to get_invisible_operands.
531 Add dwarf2_where call. Remote automatic alignment for ENTRY
532 instructions. Move call to xtensa_clear_insn_labels to the end.
533 Rearrange to handle bundles.
534 (xtensa_cons_fix_new): Delete.
535 (xtensa_handle_align): New.
536 (xtensa_frag_init): Call xtensa_set_frag_assembly_state. Remove
537 assignment to is_no_density field.
538 (md_pcrel_from): Use new xtensa-isa.h functions. Use decode_reloc
539 instead of reloc_to_opnum. Handle "alternate" relocations.
540 (xtensa_force_relocation, xtensa_check_inside_bundle,
541 xtensa_elf_section_change_hook): New.
542 (xtensa_symbol_new_hook): Delete.
543 (xtensa_fix_adjustable): Check for difference of symbols with an
544 offset. Check for external and weak symbols.
545 (md_apply_fix3): Remove cases for XTENSA_OP{0,1,2} relocs.
546 (md_estimate_size_before_relax): Return expansion for the first slot.
547 (tc_gen_reloc): Handle difference of symbols by producing
548 XTENSA_DIFF{8,16,32} relocs and by writing the value of the difference
549 into the output. Handle new XTENSA_SLOT*_OP relocs by storing the
550 tentative values into the output when linkrelax is set.
551 (XTENSA_PROP_SEC_NAME): Define.
552 (xtensa_post_relax_hook): Call xtensa_find_unmarked_state_frags.
553 Create literal tables only if using literal sections. Create new
554 property tables instead of old instruction tables. Check for unaligned
555 branch targets and loops.
556 (finish_vinsn, find_vinsn_conflicts, check_t1_t2_reads_and_writes,
557 new_resource_table, clear_resource_table, resize_resource_table,
558 resources_available, reserve_resources, release_resources,
559 opcode_funcUnit_use_unit, opcode_funcUnit_use_stage,
560 resources_conflict, xg_find_narrowest_format, relaxation_requirements,
561 bundle_single_op, emit_single_op, xg_assemble_vliw_tokens): New.
562 (xtensa_end): Call xtensa_flush_pending_output. Set past_xtensa_end
563 flag. Update checks for workaround options. Call
564 xtensa_mark_narrow_branches and xtensa_mark_zcl_first_insns.
565 (xtensa_cleanup_align_frags): Add special case for branch targets.
566 Check for and mark unreachable frags.
567 (xtensa_fix_target_frags): Remove use of align_only_targets flag.
568 Use RELAX_LOOP_END_BYTES in special case for negatable branch at the
569 end of a zero-overhead loop body.
570 (frag_can_negate_branch): Handle instructions with multiple slots.
571 Use new xtensa-isa.h functions
572 (xtensa_mark_narrow_branches, is_narrow_branch_guaranteed_in_range,
573 xtensa_mark_zcl_first_insns): New.
574 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags): Error if
575 transformations are disabled.
576 (next_instrs_are_b_retw): Use new xtensa-isa.h functions. Handle
577 multislot instructions.
578 (xtensa_fix_close_loop_end_frags, xtensa_fix_short_loop_frags):
579 Likewise. Also error if transformations are disabled.
580 (unrelaxed_frag_max_size): New.
581 (unrelaxed_frag_min_insn_count, unrelax_frag_has_b_j): Use new
582 xtensa-isa.h functions.
583 (xtensa_sanity_check, is_empty_loop, is_local_forward_loop): Use
584 xtensa_opcode_is_loop instead of is_loop_opcode.
585 (get_text_align_power): Replace as_fatal with assertion.
586 (get_text_align_fill_size): Iterate instead of using modulus when
588 (get_noop_aligned_address): Assert that this is for a machine-dependent
589 RELAX_ALIGN_NEXT_OPCODE frag. Use next_frag_opcode_is_loop,
590 xg_get_single_size, and frag_format_size.
591 (get_widen_aligned_address): Rename to ...
592 (get_aligned_diff): ... this function. Add max_diff parameter.
593 Remove handling of rs_align/rs_align_code frags. Use
594 next_frag_format_size, get_text_align_power, get_text_align_fill_size,
595 next_frag_opcode_is_loop, and xg_get_single_size. Compute max_diff
596 and pass it back to caller.
597 (xtensa_relax_frag): Use relax_frag_loop_align. Add code for new
598 RELAX_SLOTS, RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN,
599 RELAX_FILL_NOP, and RELAX_UNREACHABLE frag types. Check relax_seen.
600 (relax_frag_text_align): Rename to ...
601 (relax_frag_loop_align): ... this function. Assume loops can only be
602 in the first slot of an instruction.
603 (relax_frag_add_nop): Use assemble_nop instead of constructing an OR
604 instruction. Remove call to frag_wane.
605 (relax_frag_narrow): Rename to ...
606 (relax_frag_for_align): ... this function. Extend to handle
607 RELAX_FILL_NOP and RELAX_UNREACHABLE, as well as RELAX_SLOTS with
608 RELAX_NARROW for the first slot.
609 (find_address_of_next_align_frag, bytes_to_stretch): New.
610 (future_alignment_required): Use find_address_of_next_align_frag and
611 bytes_to_stretch. Look ahead to subsequent frags to make smarter
613 (relax_frag_immed): Add format, slot, and estimate_only parameters.
614 Check if transformations are enabled for b_j_loop_end workaround.
615 Use new xtensa-isa.h functions and handle multislot instructions.
616 Update call to xg_assembly_relax.
617 (md_convert_frag): Handle new RELAX_SLOTS, RELAX_UNREACHABLE,
618 RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN, and RELAX_FILL_NOP
620 (convert_frag_narrow): Add segP, format and slot parameters. Call
621 convert_frag_immed for branch instructions. Adjust calls to
622 tinsn_from_chars, tinsn_immed_from_frag, and xg_emit_insn_to_buf. Use
623 xg_get_single_size and xg_get_single_format.
624 (convert_frag_fill_nop): New.
625 (convert_frag_immed): Add format and slot parameters. Handle multislot
626 instructions and use new xtensa-isa.h functions. Update calls to
627 tinsn_immed_from_frag and xg_assembly_relax. Check if transformations
628 enabled for b_j_loop_end workaround. Use build_nop instead of
629 assemble_nop. Check is_specific_opcode flag. Check for unreachable
630 frags. Use xg_get_single_size. Handle O_pltrel.
631 (fix_new_exp_in_seg): Remove check for old plt flag.
632 (convert_frag_immed_finish_loop): Update calls to tinsn_from_chars and
633 xtensa_insnbuf_to_chars. Call tinsn_immed_from_frag. Change check
634 for loop opcode to an assertion. Mark all frags up to the end of the
635 loop as not transformable.
636 (get_last_insn_flags, set_last_insn_flags): Use get_subseg_info.
637 (get_subseg_info): New.
638 (xtensa_move_literals): Call xtensa_set_frag_assembly_state. Add null
640 (xtensa_switch_to_literal_fragment): Rewrite to handle absolute
641 literals and use xtensa_switch_to_non_abs_literal_fragment otherwise.
642 (xtensa_switch_to_non_abs_literal_fragment): New.
643 (cache_literal_section): Add is_code parameter and pass it through to
644 retrieve_literal_seg.
645 (retrieve_literal_seg): Add is_code parameter and use it to set the
646 flags on the literal section. Handle case where head parameter is 0.
647 (get_frag_is_no_transform, set_frag_is_specific_opcode,
648 set_frag_is_no_transform): New.
649 (xtensa_create_property_segments): Add end_property_function parameter
650 and pass it through to add_xt_block_frags. Call bfd_get_section_flags
651 and skip SEC_DEBUGGING and !SEC_ALLOC sections.
652 (xtensa_create_xproperty_segments, section_has_xproperty): New.
653 (add_xt_block_frags): Add end_property_function parameter and call it
654 if it is non-zero. Call xtensa_frag_flags_init.
655 (xtensa_frag_flags_is_empty, xtensa_frag_flags_init,
656 get_frag_property_flags, frag_flags_to_number,
657 xtensa_frag_flags_combinable, xt_block_aligned_size,
658 xtensa_xt_block_combine, add_xt_prop_frags,
659 init_op_placement_info_table, opcode_fits_format_slot,
660 xg_get_single_size, xg_get_single_format): New.
661 (istack_push): Inline call to tinsn_copy.
662 (tinsn_copy): Delete.
663 (tinsn_has_invalid_symbolic_operands): Handle O_hi16 and O_lo16 and
664 CONST16 opcodes. Handle O_big, O_illegal, and O_absent.
665 (tinsn_has_complex_operands): Handle O_hi16 and O_lo16.
666 (tinsn_to_insnbuf): Use xg_get_single_format and new xtensa-isa.h
667 functions. Handle invisible operands.
668 (tinsn_to_slotbuf): New.
669 (tinsn_check_arguments): Use new xtensa-isa.h functions.
670 (tinsn_from_chars): Add slot parameter. Rewrite using xg_init_vinsn,
671 vinsn_from_chars, and xg_free_vinsn.
672 (tinsn_from_insnbuf): New.
673 (tinsn_immed_from_frag): Add slot parameter and handle multislot
674 instructions. Handle symbol differences.
675 (get_num_stack_text_bytes): Use xg_get_single_size.
676 (xg_init_vinsn, xg_clear_vinsn, vinsn_has_specific_opcodes,
677 xg_free_vinsn, vinsn_to_insnbuf, vinsn_from_chars, expr_is_register,
678 get_expr_register, set_expr_symbol_offset_diff): New.
679 * config/tc-xtensa.h (MAX_SLOTS): Define.
680 (xtensa_relax_statesE): Move from tc-xtensa.c. Add
681 RELAX_CHECK_ALIGN_NEXT_OPCODE, RELAX_MAYBE_DESIRE_ALIGN, RELAX_SLOTS,
682 RELAX_FILL_NOP, RELAX_UNREACHABLE, RELAX_MAYBE_UNREACHABLE, and
684 (RELAX_IMMED_MAXSTEPS): Move from tc-xtensa.c.
685 (xtensa_frag_type struct): Add is_assembly_state_set,
686 use_absolute_literals, relax_seen, is_unreachable, is_specific_opcode,
687 is_align, is_text_align, alignment, and is_first_loop_insn fields.
688 Replace is_generics and is_relax fields by is_no_transform field.
689 Delete is_text and is_longcalls fields. Change text_expansion and
690 literal_expansion to arrays of MAX_SLOTS entries. Add arrays of
691 per-slot information: literal_frags, slot_subtypes, slot_symbols,
692 slot_sub_symbols, and slot_offsets. Add fr_prev field.
693 (xtensa_fix_data struct): New.
694 (xtensa_symfield_type struct): Delete plt field.
695 (xtensa_block_info struct): Move definition to tc-xtensa.h. Add
696 forward declaration here.
697 (xt_section_type enum): Delete xt_insn_sec. Add xt_prop_sec.
698 (XTENSA_SECTION_RENAME): Undefine.
699 (TC_FIX_TYPE, TC_INIT_FIX_DATA, TC_FORCE_RELOCATION, NO_PSEUDO_DOT,
700 tc_unrecognized_line, md_do_align, md_elf_section_change_hook,
701 HANDLE_ALIGN, TC_LINKRELAX_FIXUP, SUB_SEGMENT_ALIGN): Define.
702 (TC_CONS_FIX_NEW, tc_symbol_new_hook): Delete.
703 (unit_num_copies_func, opcode_num_units_func,
704 opcode_funcUnit_use_unit_func, opcode_funcUnit_use_stage_func): New.
705 (resource_table struct): New.
706 * config/xtensa-istack.h (MAX_INSN_ARGS): Increase from 6 to 10.
707 (TInsn struct): Add keep_wide, loc, fixup, record_fix, subtype,
708 literal_space, symbol, sub_symbol, offset, and literal_frag fields.
709 (tinsn_copy): Delete prototype.
710 (vliw_insn struct): New.
711 * config/xtensa-relax.c (insn_pattern_struct): Add options field.
712 (widen_spec_list): Add option conditions for density and boolean
713 instructions. Add expansions using CONST16 and conditions for using
714 CONST16 vs. L32R. Use new Xtensa LX RSR/WSR syntax. Add entries for
716 (simplify_spec_list): Add option conditions for density instructions.
717 Add entry for NOP instruction.
718 (append_transition): Add cmp function pointer parameter and use it to
719 insert the new entry in order.
720 (operand_function_LOW16U, operand_function_HI16U): New.
721 (xg_has_userdef_op_fn, xg_apply_userdef_op_fn): Handle
722 OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
723 (enter_opname, split_string): Use xstrdup instead of strdup.
724 (init_insn_pattern): Initialize new options field.
725 (clear_req_or_option_list, clear_req_option_list,
726 clone_req_or_option_list, clone_req_option_list, parse_option_cond):
728 (parse_insn_pattern): Parse option conditions.
729 (transition_applies): New.
730 (build_transition): Use new xtensa-isa.h functions. Fix incorrectly
731 swapped last arguments in calls to append_constant_value_condition.
732 Call clone_req_option_list. Add warning about invalid opcode.
733 Handle LOW16U and HI16U function names.
734 (build_transition_table): Add cmp parameter and use it in calls to
735 append_transition. Use new xtensa-isa.h functions. Check
736 transition_applies before adding entries.
737 (xg_build_widen_table, xg_build_simplify_table): Add cmp parameter and
738 pass it through to build_transition_table.
739 * config/xtensa-relax.h (ReqOrOptionList, ReqOrOption, ReqOptionList,
740 ReqOption, transition_cmp_fn): New types.
741 (OpType enum): Add OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
742 (transition_rule struct): Add options field.
743 * doc/as.texinfo (Overview): Update Xtensa options.
744 * doc/c-xtensa.texi (Xtensa Options): Delete --[no-]density,
745 --[no-]relax, and --[no-]generics options. Update descriptions of
746 --text-section-literals and --[no-]longcalls. Add
747 --[no-]absolute-literals and --[no-]transform.
748 (Xtensa Syntax): Add description of syntax for FLIX instructions.
749 Remove use of "generic" and "specific" terminology for opcodes.
750 (Xtensa Registers): Generalize the syntax description to include
751 user-defined register files.
752 (Xtensa Automatic Alignment): Update.
753 (Xtensa Branch Relaxation): Mention limitation of unconditional jumps.
754 (Xtensa Call Relaxation): Linker can now remove most of the overhead.
755 (Xtensa Directives): Remove confusing rules about precedence.
756 (Density Directive, Relax Directive): Delete.
757 (Schedule Directive): New.
758 (Generics Directive): Rename to ...
759 (Transform Directive): ... this node.
760 (Literal Directive): Update for absolute literals. Missing
761 literal_position directive is now an error.
762 (Literal Position Directive): Update for absolute literals.
763 (Freeregs Directive): Delete.
764 (Absolute Literals Directive): New.
765 (Frame Directive): Minor editing.
766 * Makefile.am (DEPTC_xtensa_elf, DEPOBJ_xtensa_elf, DEP_xtensa_elf):
768 * Makefile.in: Regenerate.
770 2004-10-07 Richard Sandiford <rsandifo@redhat.com>
772 * config/tc-mips.c (append_insn): Use fix_new rather than fix_new_exp
773 to build the second and third fixups for a composite relocation.
774 (macro_read_relocs): New function.
775 (macro_build): Use it.
776 (s_cpsetup): Pass all three composite relocation codes to macro_build.
777 Simplify fragging code accordingly.
778 (s_gpdword): Use fix_new rather than fix_new_exp for the second part
779 of the composite relocation. Set fx_tcbit in both fixups.
781 2004-10-07 Richard Sandiford <rsandifo@redhat.com>
783 * config/tc-mips.c (append_insn): Set fx_tcbit for composite relocs.
784 (md_apply_fix3): Don't treat composite relocs as done.
786 2004-10-07 Jan Beulich <jbeulich@novell.com>
788 * macro.c (macro_expand_body): When ELF, use .LL rather than LL as
789 prefix for symbol names generated from the LOCAL macro directive.
791 * dw2gencfi.c (select_cie_for_fde): When separating CIE out from
792 FDE, treat a DW_CFA_remember_state as we do a DW_CFA_advance_loc.
794 2004-10-07 Tomer Levi <Tomer.Levi@nsc.com>
796 * config/tc-crx.c (preprocess_reglist): Handle Co-processor
798 (md_assemble): Add error checking for Co-Processor instructions.
799 (get_cinv_parameters): Add 'b' option to invalidate the
802 2004-10-05 Paul Brook <paul@codesourcery.com>
804 * config/tc-arm.c (unwind): New variable.
805 (vfp_sp_encode_reg): New function.
806 (vfp_sp_reg_required_here): Use it.
807 (vfp_sp_reg_list, vfp_dp_reg_list): Remove.
808 (vfp_parse_reg_list): New function.
809 (s_arm_unwind_fnstart, s_arm_unwind_fnend, s_arm_unwind_cantunwind,
810 s_arm_unwind_personality, s_arm_unwind_personalityindex,
811 s_arm_unwind_handlerdata, s_arm_unwind_save, s_arm_unwind_movsp,
812 s_arm_unwind_pad, s_arm_unwind_setfp, s_arm_unwind_raw): New
814 (md_pseudo_table): Add them.
815 (do_vfp_reg2_from_sp2): Use vfp_parse_reg_list and vfp_sp_encode_reg.
816 (do_vfp_sp2_from_reg2, vfp_sp_ldstm, vfp_dp_ldstm): Ditto.
817 (set_section, add_unwind_adjustsp, flush_pending_unwind,
818 finish_unwind_opcodes, start_unwind_section, create_unwind_entry,
819 require_hashconst, add_unwind_opcode): New functions.
820 * doc/tc-arm.text: Document unwinding opcodes.
821 * NEWS: Mention the new feature.
823 2004-10-04 Eric Christopher <echristo@redhat.com>
825 * config/tc-mips.c (md_apply_fix3): Remove erroneous assert.
827 2004-10-01 H.J. Lu <hongjiu.lu@intel.com>
829 * config/tc-ppc.c (md_apply_fix3): Call S_SET_THREAD_LOCAL for
831 * config/tc-s390.c (md_apply_fix3): Likewise.
832 * config/tc-sparc.c (md_apply_fix3): Likewise.
834 2004-10-01 Paul Brook <paul@codesourcery.com>
836 * config/tc-arm.c (arm_elf_section_type): New function.
837 (arm_elf_change_section): Set section link for exidx sections.
838 * config/tc-arm.h (arm_elf_section_type): Add prototype.
839 (md_elf_section_type): Define.
841 2004-10-01 Bill Farmer <Bill@the-farmers.freeserve.co.uk>
843 * config/tc-pdp11.c (md_apply_fix3): Change to sign of the SOB
844 instruction's offset.
846 2004-10-01 Adam Nemet <anemet@lnxw.com>
848 * (TARGET_FORMAT): Remove LynxOS COFF definition.
850 2004-10-01 Ravi Ramaseshan <ravi.ramaseshan@codito.com>
852 * config/tc-arc.c (tc_gen_reloc): Don't assume fixP->fx_addsy is an
853 asymbol *, instead use symbol_get_bfdsym.
855 2004-09-30 Linus Nielsen Feltzing <linus@haxx.se>
857 * config/tc-m68k.c (select_control_regs): Add mcf5249.
859 2004-09-30 Paul Brook <paul@codesourcery.com>
861 * config/tc-arm.c (do_smi, do_nop): New functions.
862 (insns): Add ARMv6ZK instructions.
863 (md_apply_fix3): Handle BFD_RELOC_ARM_SMI.
864 (tc_gen_reloc): Ditto.
865 (arm_cpus): Add mpcore and arm1176.
866 (arm_archs): Add armv6{k,z,zk}.
867 * doc/c-arm.texi: Document new cores and architectures.
869 2004-09-30 Nick Clifton <nickc@redhat.com>
871 * config/tc-arm.c: Use ISO C90 formatting.
873 2004-09-30 Vladimir Ivanov <vladitx@nucleusys.com>
875 * config/tc-arm.c (mav_reg_required_here): Allow REG_TYPE_CN
876 as alternative when REG_TYPE_MVF, REG_TYPE_MVD, REG_TYPE_MVFX or
877 REG_TYPE_MVDX is expected.
879 2004-09-29 Marc Bevand <m.bevand@gmail.com>
881 * doc/c-i386.texi (i386-Mnemonics): Fix typo.
883 2004-09-21 James E Wilson <wilson@specifixinc.com>
885 * config/tc-ia64.c (ENCODED_PSP_OFFSET): New.
886 (output_rp_psprel, output_pfs_psprel, output_preds_psprel,
887 output_spill_base, output_unat_psprel, output_lc_psprel,
888 output_fpsr_psprel, output_priunat_psprel, output_bsp_psprel,
889 output_bsprestore_psprel, output_rnat_psprel, output_spill_psprel,
890 output_spill_psprel_p): Use it.
892 2004-09-20 Tomer Levi <Tomer.Levi@nsc.com>
894 * config/tc-crx.c (handle_LoadStor): New function.
895 Handle load/stor unique instructions before parsing.
897 2004-09-17 Paul Brook <paul@codesourcery.com>
899 * config/tc-arm.c (s_arm_rel31): New funciton.
900 (md_pseudo_table): Add .rel31.
901 (md_apply_fix3): Handle BFD_RELOC_ARM_TARGET2,
902 BFD_RELOC_32_PCREL and BFD_RELOC_ARM_PREL31.
903 (tc_gen_reloc): Handle BFD_RELOC_ARM_PREL31 and BFD_RELOC_ARM_TARGET2.
904 (arm_fix_adjustable): Return 0 for BFD_RELOC_ARM_TARGET2.
905 (arm_parse_reloc): Add (target2).
907 2004-09-17 Alan Modra <amodra@bigpond.net.au>
909 * Makefile.am: Run "make dep-am".
910 * Makefile.in: Regenerate.
911 * aclocal.m4: Regenerate.
912 * configure: Regenerate.
913 * doc/Makefile.in: Regenerate.
914 * po/POTFILES.in: Regenerate.
915 * po/gas.pot: Regenerate.
917 2004-09-14 Hideki IWAMOTO <h-iwamoto@kit.hi-ho.ne.jp>
919 * config/tc-mmix.c [!LLONG_MIN]: Correct #elsif to #elif.
922 2004-09-13 Paul Brook <paul@codesourcery.com>
924 * config/tc-arm.c: Rename RELABS to TARGET1.
926 2004-09-13 Alan Modra <amodra@bigpond.net.au>
928 * messages.c (as_internal_value_out_of_range): Cast values passed
929 to as_bad_where or as_warn_where to proper type.
931 2004-09-11 Theodore A. Roth <troth@openavr.org>
933 * config/tc-avr.c: Add support for
934 atmega48, atmega88, atmega168, attiny13, attiny2313, at90can128.
936 2004-09-09 Alan Modra <amodra@bigpond.net.au>
938 * dw2gencfi.c (select_cie_for_fde): When separating CIE out
939 from FDE, treat a CFI_escape as we do a DW_CFA_advance_loc.
941 2004-09-08 Paul Brook <paul@codesourcery.com>
943 * config/obj-elf.c (obj_elf_section_type): Handle init_array,
944 fini_array and preinit_array section types.
945 * config/tc-ia64.c (ia64_elf_section_type): Remove init_array
947 * doc/as.texinfo: Document extra section types.
949 2004-09-02 Mark Mitchell <mark@codesourcery.com>
951 * Makefile.am (TARG_ENV_HFILES): Add te-symbian.h.
952 * Makefile.in: Regenerated.
953 * configure.in: Set em for arm*-*-symbianelf*.
954 * configure: Regenerated.
955 * config/tc-arm.c (elf32_arm_target_format): Use Symbian target
956 vectors when appropriate.
957 * config/te-symbian.h: New file.
959 2004-09-03 Tomer Levi <Tomer.Levi@nsc.com>
961 * config/tc-crx.c (gettrap): Exception vector can be case
963 (process_label_constant): Fix a 32-bit displacement bug in branch
965 (get_operandtype) : Bug fix, wrong operand was used.
966 (process_label_constant): Initialize relocation type to
969 2004-09-01 Richard Earnshaw < reanrsha@arm.com>
971 * tc-arm.c (arm_cpus, arm_fpus): Allow <cpu>-s as well as <cpu>s
972 for synthesizable cores.
974 * doc/c-arm.texi (ARM Options): Document canonical names of CPUs.
976 2004-08-25 Dmitry Diky <diwil@spec.ru>
978 * config/tc-msp430.c: Clean-up the code.
979 (md_relax_table): New relax table.
980 (mcu_types): Sort MCU types.
981 (md_pseudo_table): Add .profiler pseudo handler.
982 (pow2value): New function.
983 (msp430_profiler): New function.
984 (msp430_operands): Add new insns handlers.
985 (msp430_srcoperand): Add register operand handler, allow complex
987 (md_estimate_size_before_relax): Rewritten.
988 (md_convert_frag): Rewritten.
989 (msp430_relax_frag): New function.
990 * config/tc-msp430.h (md_relax_frag): define macro
991 * doc/c-msp430.texi: Update information.
993 2004-08-24 Nick Clifton <nickc@redhat.com>
995 * as.c (std_shortopts): Allow -g to take an optional argument.
996 (parse_args): Pass any switch starting with -g on to the backend
999 2004-08-18 Mark Mitchell <mark@codesourcery.com>
1001 * configure.in (arm*-*-symbianelf*): New target.
1002 (arm*-*-eabi*): Likewise.
1003 * configure: Regenerated.
1005 2004-08-18 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1006 * config/tc-mips.c (append_insn): Handle delay slots in branch likely
1009 2004-08-18 Jakub Jelinek <jakub@redhat.com>
1011 * config/tc-ia64.c (start_unwind_section): Add linkonce_empty
1012 argument, don't do anything if current section is not
1013 .gnu.linkonce.t.* and linkonce_empty is set.
1014 (generate_unwind_image, dot_endp): Adjust callers, call
1015 start_unwind_section (*, 1) if nothing will be put into the
1018 2004-08-17 Nick Clifton <nickc@redhat.com>
1020 * as.c (MD_DEBUG_FORMAT_SELECTOR): Provide default definition.
1021 (show_usage): Add -g.
1022 (std_longopts): Add --gen-debug. Alpha sort the table.
1023 (parse_args): Print an error message if a switch is not handled.
1024 Handle the -g switch, calling md_debug_format_selector() if
1026 * NEWS: Mention new feature.
1027 * doc/as.texinfo: Document new switch.
1028 * doc/internals.texi: Document behaviour of md_parse_option.
1030 * config/tc-arm.c (md_parse_option): Do not issue an error message
1031 if the switch is not recognised.
1032 * config/tc-m68k.c (md_parse_option): Likewise.
1033 * config/tc-pdp11.c (md_parse_option): Likewise.
1034 * config/tc-v850.c (md_parse_option): Likewise.
1036 * as.h: Fix up formatting.
1039 2004-08-16 Nick Clifton <nickc@redhat.com>
1041 * macro.c (macro_set_alternate): Use ISO C90 formatting.
1043 * configure.in: Sort architecture based tables alphabetically.
1044 * configure: Regenerate.
1046 2004-08-16 Alan Modra <amodra@bigpond.net.au>
1048 * config/tc-ppc.c (tc_ppc_regname_to_dw2regnum <regnames>): Replace
1049 { "cc", 68 }, with { "cr", 70 }.
1051 2004-08-13 Jan Beulich <jbeulich@novell.com>
1052 Nick Clifton <nickc@redhat.com>
1054 * as.c: Add and handle new --alternate command line option.
1055 * macro.c (macro_set_alternate): New.
1056 * macro.h (macro_set_alternate): Declare.
1057 * read.c: Add and handle new .altmacro and .noaltmacro directives.
1058 * doc/as.texinfo: Document new command line option and pseudo-ops
1059 as well as insert documentation originating from gasp about
1060 alternate macro syntax.
1061 * NEWS: Mention new command line option and pseudo-ops.
1063 2004-08-10 Mark Mitchell <mark@codesourcery.com>
1065 * expr.c (operand): Handle the "~", "-", and "!" operators applied
1068 2004-08-06 Paul Brook <paul@codesourcery.com>
1070 * config/tc-arm.c (md_apply_fix3, tc_gen_reloc, arm_parse_reloc):
1071 Handle new relocations.
1072 * include/elf/arm.h (elf_arm_reloc_type): Add new EABI relocations.
1074 2004-08-05 Bob Wilson <bob.wilson@acm.org>
1076 * write.c (relax_segment): Use was_address instead of address when
1077 setting fr_fix field for align frag due to backwards .org.
1079 2004-07-29 Alexandre Oliva <aoliva@redhat.com>
1081 Introduce SH2a support.
1082 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
1083 * config/tc-sh.c (get_specific): Change arch_sh2a_up to
1085 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
1086 * config/tc-sh.c (md_parse_option): Add sh2a-nofpu ISA handling.
1087 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
1088 * config/tc-sh.c (sh_elf_final_processing): Move sh2a recognition
1089 to end of conditional expression.
1090 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
1091 * config/tc-sh.c: Add sh2a-nofpu support.
1092 2003-12-29 DJ Delorie <dj@redhat.com>
1093 * tc-sh.c: Add sh2a support.
1094 (parse_reg): Add tbr.
1095 (parse_at): Support @@(disp,tbr).
1096 (get_specific): Support sh2a opcodes.
1097 (insert4): New, for 4 byte relocs.
1098 (build_Mytes): Support sh2a opcodes.
1099 (md_apply_fix3_Mytes): Support sh2a opcodes.
1100 2003-12-02 Michael Snyder <msnyder@redhat.com>
1101 * config/tc-sh.c (md_parse_option): Handle sh2a.
1102 (sh_elf_final_processing): Ditto.
1104 2004-07-27 Jason Thorpe <thorpej@wasabisystems.com>
1106 * config/tc-hppa.h (TARGET_FORMAT): Set to "elf32-hppa-netbsd"
1109 2004-07-27 Alan Modra <amodra@bigpond.net.au>
1111 * config/tc-ppc.c (ppc_frob_file_before_adjust): Warn if .toc too big.
1112 (ppc_arch): Expand comment.
1114 2004-07-27 Tomer Levi <Tomer.Levi@nsc.com>
1116 * config/tc-crx.c: Support evaluating the difference between two
1118 * config/tc-crx.h: Likewise.
1120 2004-07-26 H.J. Lu <hongjiu.lu@intel.com>
1122 * config/tc-ia64.c (start_unwind_section): Set the linked-to
1124 (ia64_elf_section_change_hook): Set the linked-to section for
1127 2004-07-26 Dmitry Diky <diwil@spec.ru>
1129 * config/tc-msp430.c: Add new subtargets: msp430x1610,
1130 msp430x1611, msp430x1612, msp430x415, msp430x417, msp430xG437,
1131 msp430xG438, msp430xG439.
1133 2004-07-25 Daniel Jacobowitz <dan@debian.org>
1135 * doc/as.texinfo (Section, PushSection): Correct documentation
1138 2004-07-21 Jan Beulich <jbeulich@novell.com>
1140 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
1141 permissible for the selected instruction suffix.
1142 (match_template): Don't permit 64-bit general purpose operands in
1144 (finalize_imm): Permit 64-bit immediates.
1145 (build_modrm_byte): Don't treat 32-bit addressing in 64-bit mode
1146 specially except for the width of the used base and/or index
1147 registers. For 32-bit displacements, use sign-extended
1148 relocations only when using 64-bit addressing.
1149 Force zero displacement on rip-relative addressing when there is
1150 no other displacement.
1151 (i386_index_check): Don't treat 32-bit addressing in 64-bit mode
1152 specially except for the width of the used base and/or index
1154 (parse_register): Disallow Reg64 registers in 32-bit mode.
1156 * config/tc-i386.c: For DefaultSize instructions, don't guess a 'q'
1157 suffix if the instruction doesn't support it.
1159 2004-07-20 Maciej W. Rozycki <macro@linux-mips.org>
1161 * config/tc-mips.c (append_insn): Handle constant expressions with
1162 no associated relocation.
1163 (mips_ip): Cancel the expression after use for the Q format
1165 (parse_relocation): Return no relocation for unsupported
1167 (my_getSmallExpression): Return no relocation if no relocation
1170 2004-07-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1172 * config/obj-som.c (adjust_stab_sections): Add prototype.
1173 (obj_som_compiler, obj_som_version, obj_som_copyright,
1174 adjust_stab_sections): Add ATTRIBUTE_UNUSED to unused arguments.
1175 * config/tc-hppa.c (update_subspace): Likewise.
1176 (is_defined_subspace): Amplify comment.
1177 * config/obj-som.h (som_frob_file): Add prototype.
1179 2004-07-19 Christopher Faylor <cgf@timesys.com>
1180 H.J. Lu <hongjiu.lu@intel.com>
1182 * subsegs.c (section_symbol): Don't create a new segment when
1183 existing segment is undefined.
1185 2004-07-16 Richard Earnshaw <rearnsha@arm.com>
1187 * config/tc-arm.c: Include include/opcode/arm.h.
1188 (ARM_EXT_*, ARM_ARCH_*, ARM_ANY, ARM_ALL, COPROC_ANY): Delete.
1189 (FPU_FPA_EXT_* FPU_VFP_EXT_*, FPU_ANY, FPU_NONE, FPU_MAVERICK): Delete.
1190 (FPU_ARCH_*): Delete.
1191 * Makefile.am: Update dependencies.
1192 * Makefile.in: Regenerate.
1194 2004-07-15 Nick Clifton <nickc@redhat.com>
1196 * configure.in: Accept armbe as a big-endian arm configuration.
1197 * configure: Regenerate.
1199 2004-07-13 Thomas Nystrom <thn@saeab.se>
1201 * config/tc-i386.c (T_SHIFTOP): New constant.
1202 (intel_e05_1): Handle '&', '|' and T_SHIFTOP.
1203 (intel_el1): Handle '~'.
1204 (intel_get_token): Handle '<>', '&', '|' and '~'.
1206 2004-07-13 Nick Clifton <nickc@redhat.com>
1208 (md_assemble): Remove spurious newline from end of as_bad error
1210 (intel_e05_1): Likewise.
1211 (intel_e11): Likewise.
1212 (intel_match_token): Likewise.
1214 2004-07-11 Andreas Schwab <schwab@suse.de>
1216 * config/tc-m68k.c: Convert to C90. Remove redundant
1217 declarations. Indentation fixup.
1218 [M68KCOFF]: Include "obj-coff.h" instead of declaring
1219 obj_coff_section ourselves.
1221 2004-07-09 James E Wilson <wilson@specifixinc.com>
1223 * config/tc-ia64.c (default_big_endian): New.
1224 (dot_byteorder, md_begin): Use it.
1225 (md_parse_option): Set it.
1227 2004-07-09 Nick Clifton <nickc@redhat.com>
1229 * configure.in: Change sh-sybmian-elf to sh-*-symbianelf.
1230 * configure: Regenerate.
1231 * NEWS: Change sh-sybmian-elf to sh-*-symbianelf.
1232 * config/tc-sh.c (sh_elf_final_processing): Use renamed version of
1233 sh_find_elf_flags if necessary.
1235 2004-07-08 Richard Sandiford <rsandifo@redhat.com>
1237 * config/tc-mips.c (mips_fix_adjustable): If the full addend is
1238 going to be split into more than one in-place addend, return 0
1239 for relocations against mergeable sections. Associate comments
1242 2004-07-07 Tomer Levi <Tomer.Levi@nsc.com>
1244 * Makefile.am (CPU_TYPES): Add crx.
1245 (TARGET_CPU_CFILES): Add config/tc-crx.c.
1246 (TARGET_CPU_HFILES): Add config/tc-crx.h.
1247 (DEPTC_crx_elf): New target.
1248 (DEPOBJ_crx_elf): Likewise.
1249 (DEP_crx_elf): Likewise.
1250 * Makefile.in: Regenerate.
1251 * configure.in: Add crx* target.
1252 * configure: Regenerate.
1253 * config/tc-crx.c: New file.
1254 * config/tc-crx.h: New file.
1255 * NEWS: Mention new target.
1257 2004-07-06 Nick Clifton <nickc@redhat.com>
1259 * config.in: Undefine TARGET_SYMBIAN by default.
1261 * configure: Regenerate. Add sh-symbian-elf target. If
1262 selected define TARGET_SYMBIAN.
1263 * config/tc-sh.h (TARGET_FORMAT): Select a Symbian target
1264 format if TARGET_SYMBIAN has been defined.
1266 * output-file.c (output_file_create): Report the target format
1267 chosen when bfd_openw reports that it is invalid.
1269 * config/obj-coff.c (coff_pseudo_table): Only define the weak
1270 pseudo for BFD based assemblers.
1272 2004-07-05 Andrew Stubbs <andrew.stubbs@superh.com>
1275 * config/tc-sh.c (md_assemble): Change isspace to ISSPACE.
1276 (md_parse_option): Remove redundant -isa testing.
1277 Make bfd_arch variable const.
1278 (md_show_usage): Make bfd_arch variable const.
1280 2004-07-03 James E Wilson <wilson@specifixinc.com>
1282 * config/tc-ia64.c (emit_one_bundle): Check and set insn_addr.
1283 * config/tc-ia64.h (md_frag_check): Define.
1285 2004-07-03 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
1287 * config/obj-coff.c (obj_coff_weak): New .weak syntax for PE weak
1289 * doc/as.texinfo (Weak): Document PE weak symbols.
1291 2004-07-03 Richard Sandiford <rsandifo@redhat.com>
1293 * config/tc-mips.c (HAVE_IN_PLACE_ADDENDS): New macro.
1294 (reloc_needs_lo_p): Only return true if HAVE_IN_PLACE_ADDENDS.
1295 (mips_frob_file): Rework so that only a single pass through the
1296 relocs is needed. Allow %lo()s to have higher offsets than their
1297 corresponding %hi()s or %got()s.
1299 2004-07-02 Nick Clifton <nickc@redhat.com>
1301 * config/tc-arm.c (md_apply_fix3:BFD_RELOC_ARM_IMMEDIATE): Do not
1302 allow values which have come from undefined symbols.
1303 Always consider this fixup to have been processed as a reloc
1304 cannot be generated for it.
1306 2004-07-02 Alan Modra <amodra@bigpond.net.au>
1308 * frags.h (struct frag): Add has_code and insn_addr fields.
1309 * write.c (cvt_frag_to_fill): Invoke md_frag_check.
1310 * config/tc-ppc.c (md_assemble): Check and set insn_addr.
1311 * config/tc-ppc.h (md_frag_check): Define.
1313 2004-06-28 Maciej W. Rozycki <macro@linux-mips.org>
1315 * doc/Makefile.am (info): Rename goal to...
1316 (info-local): ... this, to preserve implicit dependencies.
1317 * doc/Makefile.in: Regenerate with automake 1.8.5.
1319 2004-06-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1321 * config/tc-m32r.c (md_convert_frag): Changed for @PLT.
1322 (m32r_cgen_record_fixup_exp): Changed for @GOTOFF, @GOT.
1323 (m32r_fix_adjustable): Changed for @GOTOFF, @GOT, @PLT.
1324 (tc_gen_reloc): Likewise.
1325 (m32r_end_of_match): Add for @GOTOFF, @GOT, @PLT.
1326 (m32r_parse_name): Likewise.
1327 (m32r_cgen_parse_fix_exp): Likewise.
1328 * config/tc-m32r.h (md_parse_name): Define for @GOTOFF, @GOT, @PLT.
1329 (O_PIC_reloc): Likewise.
1330 (TC_CGEN_PARSE_FIX_EXP): Likewise..
1331 * cgen.c (gas_cgen_parse_operand): Add TC_CGEN_PARSE_FIX_EXP
1332 for @GOTOFF, @GOT, @PLT.
1334 2004-06-21 Jan Beulich <jbeulich@novell.com>
1336 * gas/symbols.c: While discarding ordinary local absolute symbols
1337 when --strip-local-absolute is in effect, retain file symbols.
1339 2004-06-20 Andreas Schwab <schwab@suse.de>
1341 * config/tc-m68k.c (mri_chip): Replace current_chip, not augment.
1342 (md_parse_option): Likewise.
1344 2004-06-17 Jan Beulich <jbeulich@novell.com>
1346 * config/tc-i386.c: Deal with LEX_QM the same way as with LEX_AT.
1347 * config/te-netware.h: New file.
1348 * config/te-ppcnw.h: Delete: Obsolete.
1349 * configure.in: Eliminate ill NetWare targets. Make generic
1350 NetWare target use proper emulation.
1351 * Makefile.am: Eliminate reference to obsolete te-ppcnw.h, add
1352 reference to new te-netware.h.
1353 * configure: Regenerate.
1354 * Makefile.in: Regenerate.
1356 2004-06-15 Martin Schwidefsky <schwidefsky@de.ibm.com>
1358 * config/tc-s390.c (s390_insn): Avoid incorrect signed/unsigned
1359 comparison in .insn pseudo operation.
1361 2004-06-15 Alan Modra <amodra@bigpond.net.au>
1363 * config/obj-coff.c (coff_adjust_section_syms): Use
1364 bfd_get_section_size instead of bfd_get_section_size_before_reloc.
1365 (coff_frob_section): Likewise.
1366 * config/tc-mips.c (md_apply_fix3): Likewise.
1367 * config/obj-elf.c (elf_frob_file): Use bfd_set_section_size.
1368 (elf_frob_file_after_relocs): Likewise.
1370 2004-06-10 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1372 * config/tc-hppa.c (log2): Rename to exact_log2.
1373 (pa_next_subseg): Delete unused function.
1374 (create_new_space): Mark unused arguments with ATTRIBUTE_UNUSED.
1375 (create_new_subspace): Likewise.
1378 * config/tc-hppa.c (hppa_fix_adjustable): Allow reduction of fake
1379 labels. Fix warning.
1381 2004-05-28 DJ Delorie <dj@redhat.com>
1383 * config/tc-mn10300.h (tc_fix_adjustable): Define.
1384 * config/tc-mn10300.c (mn10300_fix_adjustable): Don't adjust debug
1385 or non-merged symbols.
1387 2004-05-28 H.J. Lu <hongjiu.lu@intel.com>
1389 * config/tc-ia64.c (remove_marked_resource): Save, clear and
1390 restore the old slot when inserting srlz.i/srlz.d.
1392 2004-05-28 Andrew Stubbs <andrew.stubbs@superh.com>
1394 * Makefile.am: Regenerate dependecies.
1395 * Makefile.in: Regenerate.
1396 * config/tc-sh.c (valid_arch): Make unsigned.
1397 (preset_target_arch): Likewise.
1398 (md_begin): Use new architecture flags system.
1399 (get_specific): Likewise.
1400 (assemble_ppi): Likewise.
1401 (md_assemble): Likewise. Also fix error check for bad opcodes.
1402 (md_parse_option): Likewise. Also generate -isa values according
1403 to the table in bfd/cpu-sh.c instead of just constants. Also
1404 allow <arch>-up ISA variants.
1405 (sh_elf_final_processing): Replace if-else chain with a call to
1406 sh_find_elf_flags().
1407 * testsuite/gas/sh/arch: New directory.
1408 * testsuite/gas/sh/arch/arch.exp: New test script.
1409 * testsuite/gas/sh/arch/arch_expected.txt: New file.
1410 * testsuite/gas/sh/arch/sh.s: New file.
1411 * testsuite/gas/sh/arch/sh2.s: New file.
1412 * testsuite/gas/sh/arch/sh-dsp.s: New file.
1413 * testsuite/gas/sh/arch/sh2e.s: New file.
1414 * testsuite/gas/sh/arch/sh3-nommu.s: New file.
1415 * testsuite/gas/sh/arch/sh3.s: New file.
1416 * testsuite/gas/sh/arch/sh3-dsp.s: New file.
1417 * testsuite/gas/sh/arch/sh3e.s: New file.
1418 * testsuite/gas/sh/arch/sh4-nommu-nofpu.s: New file.
1419 * testsuite/gas/sh/arch/sh4-nofpu.s: New file.
1420 * testsuite/gas/sh/arch/sh4.s: New file.
1421 * testsuite/gas/sh/arch/sh4a-nofpu.s: New file.
1422 * testsuite/gas/sh/arch/sh4al-dsp.s: New file.
1423 * testsuite/gas/sh/arch/sh4a.s: New file.
1425 2004-05-28 Peter Barada <peter@the-baradas.com>
1427 * config/gc-m68k.c(m68k_ip): Convert mode 5 addressing
1428 with zero offset into mode 2 addressing to save a word.
1430 2004-05-27 H.J. Lu <hongjiu.lu@intel.com>
1432 * config/tc-ia64.c (ar_is_in_integer_unit): Removed.
1433 (ar_is_only_in_integer_unit): New.
1434 (ar_is_only_in_memory_unit): New.
1435 (generate_unwind_image): Silence gcc on 32bit host.
1436 (md_assemble): Use ar_is_only_in_integer_unit instead of
1437 ar_is_in_integer_unit. Check AR access.
1439 2004-05-27 Peter Barada <peter@the-baradas.com>
1441 * config/tc-m68k.c (md_begin): Sort the opcode table into
1443 (m68k_compare_opcode): New function to do the sorting.
1445 2004-05-24 Peter Barada <peter@the-baradas.com>
1447 * config/m68k-parse.y(operand): Allow for MAC/EMAC mask
1448 addressing on MIT style operands.
1449 * config/m68k-parse.y(yylex): Allow '-&' for predecrement
1451 * config/tc-m68k.c(install_operand): Comment 'G' and 'H' type
1454 2004-05-23 Alan Modra <amodra@bigpond.net.au>
1456 * expr.c (operand, operator): Don't reject '++' and '--'.
1458 2004-05-20 Richard Sandiford <rsandifo@redhat.com>
1460 * config/tc-mips.c (append_insn): Use ISA-encoded addresses in MIPS16
1463 2004-05-17 Adam Nemet <anemet@lnxw.com>
1465 * configure.in: Add ppc-*-lynxos*. Update i386-*-lynxos* to ELF.
1466 * configure: Regenerate.
1468 2004-05-13 Paul Brook <paul@codesourcery.com>
1470 * dw2gencfi.c (output_cie): Handle dwarf3 format CIE entries.
1472 2004-05-13 Joel Sherrill <joel@oarcorp.com>
1474 * configure.in (or32-*-rtems*): Switch to elf.
1475 * configure: Regenerate.
1477 2004-05-13 Nick Clifton <nickc@redhat.com>
1479 * po/fr.po: Updated French translation.
1481 2004-05-11 Nick Clifton <nickc@redhat.com>
1483 * doc/as.texinfo (Section): Document G and T flags to .section
1484 directive. Document the extra arguments that the G flag
1485 requires. Document the #tls flag.
1487 2004-05-11 H.J. Lu <hongjiu.lu@intel.com>
1489 * subsegs.c (section_symbol): Create a new section symbol if
1490 the existing one doesn't match.
1491 * symbols.c (symbol_set_bfdsym): Don't reset BFD section symbol.
1493 2004-05-07 Richard Sandiford <rsandifo@redhat.com>
1495 * config/tc-mips.c (append_insn, mips_emit_delays): Extend -mfix-vr4120
1496 to cope with VR4181A errata MD(1) and MD(4).
1498 2004-05-07 Brian Ford <ford@vss.fsi.com>
1500 * NEWS: Mention .secrel32 for pe[i]-i386.
1502 2004-05-07 Alexandre Oliva <aoliva@redhat.com>
1504 * config/tc-frv.h (MAX_MEM_FOR_RS_ALIGN_CODE): New.
1505 (HANDLE_ALIGN): New.
1507 2004-05-06 Daniel Jacobowitz <dan@debian.org>
1509 * Makefile.am (DIST_SUBDIRS): Define.
1510 * aclocal.m4: Regenerate with automake 1.8.4.
1511 * Makefile.in: Likewise.
1512 * doc/Makefile.in: Likewise.
1514 2004-05-06 David Mosberger-Tang <davidm@hpl.hp.com>
1516 * config/tc-ia64.c (dot_serialize): Declare.
1517 (dot_serialize): New function.
1518 (md_pseudo_table): Add ".serialize.data" and
1519 ".serialize.instruction" directives.
1521 2004-05-06 Nick Clifton <nickc@redhat.com>
1523 * messages (as_internal_value_out_of_range): Print a message about
1524 a value being out of range. Be consistent about whether the
1525 values are printed in decimal or hexadecimal.
1526 (as_warn_value_out_of_range): Generate a warning message about an
1528 (as_bad_value_out_of_range): Generate an error message about an
1530 * as.h: Prototype the new functions.
1531 * config/tc-alpha.c (insert_operand): Use new function.
1532 * config/tc-arc.c (arc_insert_operand): Likewise.
1533 * config/tc-mn10200.c (mn10200_insert_operand): Likewise.
1534 * config/tc-mn10300.c (mn10300_insert_operand): Likewise.
1535 * config/tc-ppc.c (ppc_insert_operand): Likewise.
1536 * config/tc-s390.c (s390_insert_operand): Likewise.
1537 * config/tc-v850.c (v850_insert_operand): Likewise.
1539 2004-05-05 Alexandre Oliva <aoliva@redhat.com>
1541 * configure.in: Set em=linux for frv-*-*linux*.
1542 * configure: Rebuilt.
1543 * config/tc-frv.h (TARGET_FORMAT): Use elf32-frvfdpic if...
1544 (frv_md_fdpic_enabled): New.
1545 * config/tc-frv.c (frv_md_fdpic_enabled): New.
1546 (DEFAULT_FDPIC): New.
1547 (frv_flags): Use DEFAULT_FDPIC.
1548 (frv_pic_flag): Likewise.
1549 (OPTION_NOPIC): New.
1550 (md_longopts): Add -mnopic.
1551 (md_parse_option): Handle it.
1552 (md_show_usage): Add -mfdpic and -mnopic.
1554 2004-05-05 Peter Barada <peter@the-baradas.com>
1556 * config/tc-m68k.c: Add find_cf_chip to print list of valid
1557 chips for invalid coldfire instructions, rename selectors
1558 for ColdFire sub-variants, add 521x,5249,547x,548x and aliases,
1559 add current_chip to track which chip is referred to(including save/restore),
1560 use current_chip to select control registers, not current_arch.
1561 (md_show_usage): Add new chips.
1562 * doc/c-m68k.texi: Document new command line switches.
1564 2004-05-05 Jakub Jelinek <jakub@redhat.com>
1566 * tc-s390.h (md_do_align, HANDLE_ALIGN): Remove.
1567 (NOP_OPCODE): Define.
1568 (s390_align_code): Remove prototype.
1569 * tc-s390.c (s390_align_code): Remove.
1571 2004-05-04 H.J. Lu <hongjiu.lu@intel.com>
1573 * config/tc-ia64.c (make_unw_section_name): Removed.
1574 (start_unwind_section): New function.
1575 (generate_unwind_image): Take const segT instead of const
1577 (dot_handlerdata): Adjusted.
1578 (dot_endp): Likewise.
1580 2004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1582 * config/obj-elf.c (obj_elf_change_section): Allow the
1583 ".note.GNU-stack" section has SHF_EXECINSTR.
1585 2004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1587 * config/obj-elf.c (get_section): Return bfd_boolean.
1588 (obj_elf_change_section): Call bfd_get_section_by_name_if
1589 instead of bfd_map_over_sections.
1591 2004-04-30 H.J. Lu <hongjiu.lu@intel.com>
1593 * config/obj-elf.c (get_section): New function.
1594 (obj_elf_change_section): Support multiple sections with same
1597 2004-04-30 Nick Clifton <nickc@redhat.com>
1599 * config/tc-arm.c (create_register_alias): Fix typo checking for
1600 case sensitive register aliases.
1601 (co_proc_number): Use error message string in all_reg_maps[]
1603 (cp_reg_required_here): Likewise.
1604 (fp_reg_required_here): Likewise.
1606 2004-04-29 Brian Ford <ford@vss.fsi.com>
1608 * dwarf2dbg.c (dwarf2_finish): Add SEC_DEBUGGING to section flags.
1610 2004-04-28 Chris Demetriou <cgd@broadcom.com>
1612 * config/tc-mips.c (HAVE_32BIT_ADDRESSES, append_insn, macro_build)
1613 (load_address, macro, mips_ip, md_parse_option)
1614 (mips_force_relocation, mips_validate_fix, md_apply_fix3)
1615 (s_change_sec, pic_need_relax, tc_gen_reloc): Remove all
1616 embedded-PIC handling, and update comments.
1617 (SWITCH_TABLE): Remove.
1618 * config/tc-mips.h (DIFF_EXPR_OK): Delete.
1619 (enum mips_pic_level): Remove EMBEDDED_PIC.
1620 (EXTERN_FORCE_RELOC): Remove embedded-PIC handling.
1621 (TC_FORCE_RELOCATION): Update comment.
1622 * ecoff.c (ecoff_build_lineno): Add comment about some code that
1623 might be safe to remove now that MIPS embedded-PIC is gone.
1625 2004-04-28 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1627 * config/obj-som.c (obj_som_init_stab_section): Add new arguments in
1628 call to obj_set_subsection_attributes.
1629 (obj_som_init_stab_section): Likewise.
1630 * config/tc-hppa.c (default_subspace_dict): Add comdat field.
1631 (pa_def_subspaces): Provide comdat default.
1632 (pa_subspace): Handle new "comdat" parameter. Set SEC_LINK_ONCE and
1633 not SEC_IS_COMMON if section is comdat, common or dup_common. Update
1634 calls to create_new_subspace and update_subspace to pass comdat flag.
1635 (create_new_subspace, update_subspace): Add new comdat argument. Use
1636 it in calls to obj_set_subsection_attributes.
1637 * doc/c-hppa.texi (.subspa, .nsubspa): Document new comdat parameter
1638 and use of comdat, common and dup_comm parameters.
1640 2004-04-26 H.J. Lu <hongjiu.lu@intel.com>
1642 * config/obj-elf.c (obj_elf_change_section): Check if the old
1643 group name is NULL before comparison.
1645 2004-04-23 Chris Demetriou <cgd@broadcom.com>
1647 * config/tc-mips.h (mips_dwarf2_addr_size): Prototype.
1649 2004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1651 * config/tc-mips.c (s_mipsset): Set default CPU type for .set mipsN.
1653 2004-04-23 Chris Demetriou <cgd@broadcom.com>
1655 * config/tc-mips.c (md_longopts): Remove -membedded-pic option.
1656 (OPTION_MEMBEDDED_PIC): Remove.
1657 (OPTION_TRAP, OPTION_BREAK, OPTION_EB, OPTION_EL)
1658 (OPTION_FP32, OPTION_GP32, OPTION_CONSTRUCT_FLOATS)
1659 (OPTION_NO_CONSTRUCT_FLOATS, OPTIONS_FP64, OPTION_GP64)
1660 (OPTION_RELAX_BRANCH, OPTION_NO_RELAX_BRANCH)
1661 (OPTION_ELF_BASE): Renumber.
1662 (md_parse_option): Remove OPTION_MEMBEDDED_PIC handling.
1663 (md_show_usage): Remove mention of -membedded-pic.
1664 * doc/as.texinfo: Remove mention of -membedded-pic.
1666 2004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1668 * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Remove.
1669 * config/tc-mips.c (RDATA_SECTION_NAME, mips_target_format): Remove
1671 (md_begin, mips_ip, md_parse_option, s_change_sec, s_option,
1672 s_abicalls, nopic_need_relax, tc_gen_reloc): Remove uses of
1673 USE_GLOBAL_POINTER_OPT.
1675 2004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1677 * config/tc-mips.c (macro): One more use of load_delay_nop.
1679 2004-04-22 Atsushi Nemoto <anemo@mba.ocn.ne.jp>
1681 * config/tc-mips.c (load_delay_nop): New function.
1682 (load_address, macro): Use load_delay_nop() to build a nop
1683 which can be omitted with gpr_interlocks.
1685 2004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1687 * config/tc-mips.c (hilo_interlocks, gpr_interlocks,
1688 cop_interlocks): Remove superfluous CPU entries.
1690 2004-04-22 Paul Brook <paul@codesourcery.com>
1692 * config/tc-arm.c (mav_parse_offset): Value must be multiple of 4.
1694 2004-04-22 Peter Barada <peter@the-baradas.com>
1696 * NEWS: Added support for EMAC instructions and MAC/EMAC
1698 * config/m68k-parse.h: Add ACC[123], ACCEXT{01,23}, MAC/EMAC
1699 scale factor tokens, trailing_ampersand to mark mask addressing
1700 for MAC/EMAC instructions.
1701 * config/m68k-parse.y: Add options_ampersand clause, '<<',
1703 (yylex): Handle '>', '<', and '&' following '+'.
1704 * config/tc-m68k.c: Set mcfmac/mcfemac on appropriate ColdFire
1705 architectures in archs[].
1706 (m68k-ip): Add '4', 'e', 'g', 'i', cases to handle mask addressing
1707 for MAC/EMAC instructions, ACC[0123], ACCEXT{01,23}, and '<<'/'>>'
1709 (m68k_ip): Handle trailing '&' on MAC/EMAC insns.
1710 (install_operand): Fix 'n' case, Add 'F', 'f', 'G', 'H', 'I', ']'
1712 Add EMAC operands to init_table[].
1714 2004-04-22 Bruno De Bus <bdebus@elis.ugent.be>
1716 * config/tc-arm.h (enum mstate): Move here, add MAP_UNDEFINED
1718 (TC_SEGMENT_INFO_TYPE): Define to enum mstate.
1719 * config/tc-arm.c (enum mstate): Delete from here.
1720 (mapping_state): Remove the static mapstate variable and instead
1721 store the state in the segment. This allows a per-section mapping
1722 state. Handle and ignore MAP_UNDEFINED states.
1723 (arm_elf_change_section): Get the current mapping state from the
1725 (s_ltorg): Set the mapping state to MAP_DATA.
1726 (arm_cleanup): Use arm_elf_change_section to get the mapping state
1727 for each pool as it is emitted.
1729 2004-04-22 Nick Clifton <nickc@redhat.com>
1731 * config/tc-arm.h: Formatting tidy ups.
1733 2004-04-20 Chris Demetriou <cgd@broadcom.com>
1735 * NEWS: Note that MIPS -membedded-pic option is deprecated.
1737 2004-04-20 DJ Delorie <dj@redhat.com>
1739 * config/tc-i386.h [TE_PE] (TC_CONS_FIX_NEW): Define.
1740 * config/tc-i386.c (md_pseudo_table) [TE_PE]: Add "secrel32".
1741 [TE_PE] (O_secrel): Define.
1742 [TE_PE] (x86_pe_cons_fix_new): New.
1743 [TE_PE] (pe_directive_secrel): Likewise.
1744 (tc_gen_reloc) [TE_PE]: Support BFD_RELOC_32_SECREL.
1746 2004-04-19 Eric Christopher <echristo@redhat.com>
1748 * config/tc-mips.c (mips_dwarf2_addr_size): Revert part
1749 of previous patch for fix in gcc.
1751 2004-04-19 Jakub Jelinek <jakub@redhat.com>
1753 * config/tc-xtensa.c (xg_assembler_literal): Fix a typo.
1755 2004-04-19 Nathan Sidwell <nathan@codesourcery.com>
1757 * read.c (do_align): Call md_flush_pending_output, if defined.
1759 2004-04-16 Alan Modra <amodra@bigpond.net.au>
1761 * expr.c (operand): Correct checks for ++ and --.
1763 2004-04-14 H.J. Lu <hongjiu.lu@intel.com>
1765 * config/tc-generic.c: Add some comments.
1767 2004-04-14 Richard Sandiford <rsandifo@redhat.com>
1769 * doc/c-mips.texi (-m{no-,}fix-vr4120): Renamed from
1770 -{no-}mfix-vr4122-bugs.
1771 * config/tc-mips.c (mips_fix_vr4120): Renamed from mips_fix_4122_bugs.
1772 (append_insn, mips_emit_delays): Update accordingly.
1773 (OPTION_FIX_VR4120, OPTION_NO_FIX_VR4120): Renamed from *VR4122.
1774 (md_longopts): Change -{no-,}mfix-vr4122-bugs to -m{no-,}fix-vr4120.
1775 (md_parse_option): Update after above changes.
1776 (md_show_usage): Add -mfix-vr4120.
1778 2004-04-13 Bob Wilson <bob.wilson@acm.org>
1780 * doc/as.texinfo (Sub-Sections): Conditionalize COFF-specific use
1781 of .section directive; add a reference to the ELF .subsection
1784 2004-04-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1786 * config/tc-m32r.c (md_assemble): Fixed infinite loop bug
1789 2004-04-11 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1791 * Makefile.am: Remove mips from aout targets.
1792 * Makefile.in: Regenerate.
1793 * configure.in: Remove mips-dec-bsd* target.
1794 * configure: Regenerate.
1796 2004-04-07 Alan Modra <amodra@bigpond.net.au>
1799 * config/tc-ppc.c (ppc_elf_suffix): Add valid32 and valid64 fields
1800 to struct map_bfd. Adjust MAP macro, and define MAP32, MAP64.
1801 Update "mapping". Restrict some @ modifiers to 32 bit.
1803 2004-04-01 Asgari Jinia <asgarij@kpitcummins.com>
1804 Dhananjay Deshpande <dhananjayd@kpitcummins.com>
1806 * config/tc-sh.c (dont_adjust_reloc_32): New variable.
1807 (sh_fix_adjustable): Avoid adjusting BFD_RELOC_32 when
1808 dont_adjust_reloc_32 is set.
1809 (md_longopts): Add option -renesas.
1810 (md_parse_option, md_show_usage): Likewise.
1811 * doc/c-sh.texi: Likewise.
1813 2004-04-01 Dave Korn <dk@artimi.com>
1815 * config/tc-dlx.c (md_assemble): set fx_no_overflow flag for
1816 hi16 and lo16 fixS structs.
1817 (md_assemble): generate bit_fixS for RELOC_DLX_LO16 in
1818 exactly the same way as for RELOC_DLX_REL16.
1819 (machine_ip): properly respect LO flag in the_insn and
1820 output RELOC_DLX_LO16 rather than RELOC_DLX_16.
1821 (md_apply_fix3): apply RELOC_DLX_LO16.
1823 2004-03-30 Stan Shebs <shebs@apple.com>
1825 Remove long-obsolete MPW support.
1826 * mpw-config.in, mpw-make.sed, mac-as.r: Remove files.
1827 * configure.in: Remove mention of ppc-*-mpw* config.
1828 * configure.in: Likewise.
1830 2004-03-30 Nick Clifton <nickc@redhat.com>
1832 * config/tc-arm.c (meabi_flags): Make its use conditional upon
1833 OBJ_ELF being defined.
1835 2004-03-27 Alan Modra <amodra@bigpond.net.au>
1837 * config/obj-aout.c (obj_aout_type): Remove #ifdef BFD_ASSEMBLER code.
1839 2004-03-23 Paul Brook <paul@codesourcery.com>
1841 * config/tc-arm.c (meabi_flags): New variable.
1842 (arm_parse_eabi): New function.
1843 (md_begin): Set flags for EABI v3.
1845 (arm_long_opts): Add meabi.
1846 * doc/as.texinf <ARM>: Document -meabi.
1847 * doc/c-arm.texi: Ditto.
1849 2004-03-22 Bob Wilson <bob.wilson@acm.org>
1851 * config/tc-xtensa.c (xtensa_post_relax_hook): Create literal
1852 tables even when use_literal_section flag is not set.
1854 2004-03-22 Alan Modra <amodra@bigpond.net.au>
1856 * config/tc-sh.c: Remove trailing whitespace.
1858 2004-03-22 Hans-Peter Nilsson <hp@axis.com>
1860 * doc/c-cris.texi (CRIS-Opts): Document --no-mul-bug-abort,
1861 --mul-bug-abort and the default behavior.
1862 * config/tc-cris.c (cris_insn_kind): New member CRIS_INSN_MUL.
1863 (err_for_dangerous_mul_placement): New variable.
1864 (STATE_MUL, OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): New
1866 (md_cris_relax_table): Have placeholder for STATE_MUL.
1867 (md_longopts): New options --mul-bug-abort and --no-mul-bug-abort.
1868 (cris_relax_frag) <case ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: New
1870 (md_estimate_size_before_relax) <case ENCODE_RELAX (STATE_MUL,
1871 STATE_BYTE)>: Ditto.
1872 (md_convert_frag) <ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: Check
1873 alignment and position of this frag, emit error message if
1875 (md_assemble): For a multiply insn and when checking it,
1876 transform the current frag into a special frag for that purpose.
1877 (md_parse_option) <case OPTION_MULBUG_ABORT_OFF, case
1878 OPTION_MULBUG_ABORT_ON>: Handle new options.
1880 2004-03-19 Bob Wilson <bob.wilson@acm.org>
1882 * config/tc-xtensa.c (mark_literal_frags): New function.
1883 (xtensa_move_literals): Call mark_literal_frags for all literal
1884 segments, including init and fini literal segments.
1885 (xtensa_post_relax_hook): Swap use of xt_insn_sec and xt_literal_sec.
1887 2004-03-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1889 * tc-hppa.c (cons_fix_new_hppa): Check for PC relative base type.
1890 (pa_comm): Set BSF_OBJECT in symbol flags.
1892 2004-03-19 Alan Modra <amodra@bigpond.net.au>
1894 * Makefile.am: Run "make dep-am".
1895 * Makefile.in: Regenerate.
1896 * doc/Makefile.in: Regenerate.
1897 * config.in: Regenerate.
1898 * po/gas.pot: Regenerate.
1900 2004-03-18 Nathan Sidwell <nathan@codesourcery.com>
1902 * read.c (read_a_source_file): Use demand_empty_rest_of_line.
1903 (demand_empty_rest_of_line): Issue an error here.
1904 (ignore_rest_of_line): Silently skip to end.
1905 (demand_copy_string): Issue an error, not warning.
1907 * config/obj-elf.c (obj_elf_section_name): Likewise.
1908 (obj_elf_section): Likewise.
1909 * config/tc-arc.c (arc_extoper): Remove bogus NULL checks.
1910 (arc_extinst): Likewise.
1911 * config/tc-ia64.c (dot_saveb): Use demand_empty_rest_of_line.
1912 (dot_spill): Likewise.
1913 (dot_unwabi): Likewise.
1914 (dot_prologue): Likewise.
1916 * expr.c (operand): Reject ++ and --.
1917 (operator): Likewise.
1919 2004-03-17 Kaz Kojima <kkojima@rr.iij4u.or.jp>
1921 * config/tc-sh.c: Include dw2gencfi.h.
1922 (sh_cfi_frame_initial_instructions): New function.
1923 (sh_regname_to_dw2regnum): Likewise.
1924 * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Move to the end of
1926 (TARGET_USE_CFIPOP): Define.
1927 (tc_cfi_frame_initial_instructions): Likewise.
1928 (tc_regname_to_dw2regnum): Likewise.
1929 (DWARF2_DEFAULT_RETURN_COLUMN, DWARF2_CIE_DATA_ALIGNMENT): Likewise.
1930 * Makefile.am: Update dependencies.
1931 * Makefile.in: Regenerate.
1933 2004-03-17 Ralf Corsepius <corsepiu@faw.uni-ulm.de>
1935 * configure.in: Switch sh-*-rtems* to ELF. Add sh-*-rtemscoff*.
1936 * configure: Regenerate.
1938 2004-03-12 Bob Wilson <bob.wilson@acm.org>
1940 * read.c (s_leb128): Call md_flush_pending_output.
1942 2004-03-12 Michal Ludvig <mludvig@suse.cz>
1944 * config/tc-i386.c (output_insn): Handle PadLock instructions.
1945 * config/tc-i386.h (CpuPadLock): New define.
1946 (CpuUnknownFlags): Added CpuPadLock.
1948 2004-03-07 Andreas Schwab <schwab@suse.de>
1950 * doc/c-hppa.texi (HPPA Directives): Fix typo.
1952 2004-03-07 Richard Henderson <rth@redhat.com>
1954 * dw2gencfi.c (output_cie): Align length to 4 byte boundary.
1955 (cfi_finish): Likewise for fde.
1957 2004-03-05 H.J. Lu <hongjiu.lu@intel.com>
1959 * config/tc-ia64.c (md_assemble): Properly handle NULL
1961 (ia64_handle_align): Don't abort if failed to add a stop bit.
1963 2004-03-04 H.J. Lu <hongjiu.lu@intel.com>
1965 * Makefile.in: Regenerated.
1966 * aclocal.m4: Likewise.
1967 * configure: Likewise.
1968 * doc/Makefile.in: Likewise.
1970 2004-03-03 H.J. Lu <hongjiu.lu@intel.com>
1972 * config/tc-ia64.c (dot_align): New.
1973 (ia64_do_align): Make it static.
1974 (md_pseudo_table): Use "dot_align" for "align".
1975 (ia64_md_do_align): Don't set align_frag here.
1976 (ia64_handle_align): Add a stop bit to the previous bundle if
1979 * config/tc-ia64.h (ia64_do_align): Removed.
1981 2003-03-03 Andrew Stubbs <andrew.stubbs@superh.com>
1983 * config/tc-sh.c (md_parse_option): Add -isa=sh4-nofpu and
1984 -isa=sh4-nommu-nofpu options. Adjust help messages accordingly.
1985 (sh_elf_final_processing): Output BFD type sh4_nofpu if that is
1986 the most general type or the user specifically requested it.
1987 (md_assemble): Add a new error message for when an instruction
1988 is understood, but is not allowed due to an -isa option.
1990 2004-03-02 H.J. Lu <hongjiu.lu@intel.com>
1992 * config/tc-ia64.c (align_frag): New.
1993 (md_assemble): Set the tc_frag_data field in align_frag for
1994 IA64_OPCODE_FIRST instructions.
1995 (ia64_md_do_align): Set align_frag.
1996 (ia64_handle_align): Add a stop bit if needed.
1998 * config/tc-ia64.h (TC_FRAG_TYPE): New.
1999 (TC_FRAG_INIT): New.
2001 2004-03-01 Richard Sandiford <rsandifo@redhat.com>
2003 * config/tc-frv.c (fr400_audio): New variable.
2004 (md_parse_option, md_show_usage): Add -mcpu=fr405 and -mcpu=fr450.
2005 (md_parse_option): Set fr400_audio for -mcpu=fr400 and -mcpu=fr405.
2006 (target_implements_insn_p): New function.
2007 (md_assemble): Report an error if the processor doesn't implement
2010 2004-02-27 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2012 * config/tc-m32r.c (md_longopts): Added -no-bitinst option.
2013 (md_parse_option): Ditto.
2014 (OPTION_NO_SPECIAL_M32R): Added.
2015 (md_show_usage): Document it.
2016 (enable_speial_m32r): Changed a default value from 0 to 1.
2017 * doc/c-m32r.texi: Document the -no-bitinst option.
2019 2004-02-27 Nick Clifton <nickc@redhat.com>
2021 * config/tc-sh.c (get_operand): Revert previous delta.
2022 (tc_gen_reloc): Check for an unknown reloc type before processing
2025 2004-02-27 Hannes Reinecke <hare@suse.de>
2027 * config/tc-s390.c (s390_insn): Correct range check for opcode in
2028 .insn pseudo operation.
2030 2004-02-27 Anil Paranjpe <anilp1@kpitcummins.com>
2032 * config/tc-sh.c (get_operand): In case of #Imm, check has been
2033 added for wrong syntax.
2035 2004-02-26 Eric Christopher <echristo@redhat.com>
2037 * config/tc-mips.c (mips_dwarf2_addr_size): New.
2038 * config/tc-mips.h (DWARF2_ADDR_SIZE): Use.
2040 2004-02-26 Andrew Stubbs <andrew.stubbs@superh.com>
2042 * config/tc-sh.c (build_Mytes): Add REG_N_D and REG_N_B01
2043 nibble types to assembler.
2045 2004-02-25 Fred Fish <fnf@redhat.com>
2047 * config/tc-iq2000.c: Add missing \n\ in multiline string literal.
2049 2004-02-20 James E Wilson <wilson@specifixinc.com>
2051 * config/tc-ia64.c (slot_index): New arg before_relax. Use instead of
2053 (fixup_unw_records): New arg before_relax. Pass to slot_index.
2054 (ia64_estimate_size_before_relax): New.
2055 (ia64_convert_frag): Pass 0 to fixup_unw_records. Add comment.
2056 (generate_unwind_image): Pass 1 to fixup_unw_records.
2057 * config/tc-ia64.h (ia64_estimate_size_before_relax): Declare.
2058 (md_estimate_size_before_relax): Call ia64_estimate_size_before_relax.
2060 2004-02-19 Jakub Jelinek <jakub@redhat.com>
2062 * stabs.c (generate_asm_file): Avoid warning about use of
2063 uninitialized variable.
2065 2004-02-18 David Mosberger <davidm@hpl.hp.com>
2067 * config/tc-ia64.c (ia64_flush_insns): In addition to prologue,
2068 body, and endp, allow unwind records which do not have a "t"
2069 (time/instruction) field.
2071 2004-02-17 Petko Manolov <petkan@nucleusys.com>
2073 * config/tc-arm.c (do_mav_dspsc_1): Correct offset of CRn.
2074 (do_mav_dspsc_2): Likewise.
2075 Fix accumulator registers move opcodes.
2077 2004-02-13 Hannes Reinecke <hare@suse.de>
2078 Jakub Jelinek <jakub@redhat.com>
2080 * dwarf2dbg.c (get_filenum): Do not read beyond allocated memory.
2082 2004-02-10 Steve Ellcey <sje@cup.hp.com>
2084 * config/tc-ia64.h (ia64_frob_symbol): New declaration.
2085 (tc_frob_symbol): New macro definition.
2086 * config/tc-ia64.c (ia64_frob_symbol): New routine.
2088 2004-02-09 Daniel Jacobowitz <drow@mvista.com>
2090 * config/tc-arm.c (md_begin): Mark .note.gnu.arm.ident as
2093 2004-02-09 Nathan Sidwell <nathan@codesourcery.com>
2095 * read.h (IGNORE_OPCODE_CASE): Do not define. Replace with ...
2096 (TC_CASE_SENSITIVE): ... this.
2097 * read.c: Replace IGNORE_OPCODE_CASE with TC_CASE_SENSITIVE.
2098 * doc/internals.texi (TC_CASE_SENSITIVE): Document.
2100 2004-02-06 James E Wilson <wilson@specifixinc.com>
2102 * config/tc-ia64.c (dot_endp): Delete call to output_endp.
2103 (generate_unwind_image): Re-add it here.
2105 2004-02-06 Nathan Sidwell <nathan@codesourcery.com>
2107 * dwarf2dbg.c (DWARF2_ADDR_SIZE): Remove trailing ';'
2108 * read.h (SKIP_WHITESPACE): Turn into an expression.
2109 * read.c (read_a_source_file): A pseudo is removed by having a
2112 2004-02-05 James E Wilson <wilson@specifixinc.com>
2114 * config/tc-ia64.c (output_endp): New.
2115 (count_bits): Delete.
2116 (ia64_flush_insns, process_one_record, optimize_unw_records): Handle
2117 endp unwind records.
2118 (fixup_unw_records): Handle endp unwind records. Delete code for
2119 shortening prologue regions not followed by a body record.
2120 (dot_endp): Call add_unwind_entry to emit endp unwind record.
2121 * config/tc-ia64.h (unw_record_type): Add endp.
2123 2004-02-03 James E Wilson <wilson@specifixinc.com>
2125 * config/tc-ia64.c (ia64_convert_frag): Call md_number_to_chars to
2126 fill padding bytes with zeroes.
2127 (emit_one_bundle): New locals last_ptr, end_ptr. Rewrite code that
2128 sets unwind_record slot_number and slot_frag fields.
2130 2004-02-02 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2132 * config/tc-mips.c (add_got_offset_hilo): New function.
2133 (macro): Use load_register() and add_got_offset_hilo() to load
2134 constants instead of hardcoding code sequences throughout.
2136 2004-01-28 H.J. Lu <hongjiu.lu@intel.com>
2138 * config/tc-ia64.c (emit_one_bundle): Add proper indentation.
2140 2004-01-26 Bernardo Innocenti <bernie@develer.com>
2142 * config/tc-m68k.h (EXTERN_FORCE_RELOC): Handle m68k-uclinux specially,
2144 * config/tc-m68k.c (RELAXABLE_SYMBOL): Use EXTERN_FORCE_RELOC instead
2145 of hard-coded test for TARGET_OS=elf.
2147 2004-01-24 Chris Demetriou <cgd@broadcom.com>
2149 * config/tc-mips.c (hilo_interlocks): Change definition
2150 so that MIPS32, MIPS64 and later ISAs are included, along with
2151 the already-included machines. Update comments.
2153 2004-01-23 Daniel Jacobowitz <drow@mvista.com>
2155 * config/tc-arm.c (tc_gen_reloc): Improve error message for
2156 undefined local labels.
2158 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2160 * config/tc-mips.c (load_address, macro): Update comments about
2161 NewABI GP relaxation.
2163 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2165 * config/tc-mips.c (macro_build): Remove place and counter arguments.
2166 (mips_build_lui, macro_build_ldst_constoffset): Likewise.
2167 (mips16_macro_build, macro_build_jalr): Remove counter argument.
2168 (set_at, load_register, load_address, move_register): Likewise.
2169 (load_got_offset, add_got_offset): Likewise.
2170 Update all calls and tidy accordingly.
2172 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2174 * config/tc-mips.c (RELAX_ENCODE): Remove WARN argument.
2175 (RELAX_FIRST, RELAX_SECOND): Turn into 8-bit quantities.
2176 (RELAX_USE_SECOND): Bump to 0x10000.
2177 (RELAX_SECOND_LONGER, RELAX_NOMACRO, RELAX_DELAY_SLOT): New flags.
2178 (mips_macro_warning): New variable.
2179 (md_assemble): Wrap macro expansion in macro_start() and macro_end().
2180 (s_cpload, s_cpsetup, s_cprestore, s_cpreturn): Likewise.
2181 (relax_close_frag): Set mips_macro_warning.first_frag. Adjust use
2183 (append_insn): Update mips_macro_warning.sizes.
2184 (macro_start, macro_warning, macro_end): New functions.
2185 (macro_build): Don't emit warnings here.
2186 (macro_build_lui, md_estimate_size_before_relax): ...or here.
2187 (md_convert_frag): Check for cases where one macro alternative
2188 needs a warning and the other doesn't. Emit a warning if the
2189 longer sequence was chosen.
2191 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2193 * config/tc-mips.h (tc_frag_data_type, TC_FRAG_TYPE): Remove.
2194 * config/tc-mips.c (RELAX_ENCODE): Take three arguments: the size of
2195 the first sequence, the size of the second sequence, and a flag
2196 that says whether we should warn.
2197 (RELAX_OLD, RELAX_NEW, RELAX_RELOC[123]): Delete.
2198 (RELAX_FIRST, RELAX_SECOND): New.
2199 (mips_relax): New variable.
2200 (relax_close_frag, relax_start, relax_switch, relax_end): New fns.
2201 (append_insn): Remove "place" argument. Use mips_relax.sequence
2202 rather than "place" to check whether we're expanding the second
2203 alternative of a relaxable macro. Remove redundant check for
2204 branch relaxation. If generating a normal insn, and there
2205 is not enough room in the current frag, call relax_close_frag()
2206 to close it. Update mips_relax.sizes[]. Emit fixups for the
2207 second version of a relaxable macro. Record the first relaxable
2208 fixup in mips_relax. Remove tc_gen_reloc workaround.
2209 (macro_build): Remove all uses of "place". Use mips_relax.sequence
2210 in the same way as in append_insn.
2211 (mips16_macro_build): Remove "place" argument.
2212 (macro_build_lui): As for macro_build. Don't drop the add_symbol
2213 when generating the second version of a relaxable macro.
2214 (load_got_offset, add_got_offset): New functions.
2215 (load_address, macro): Use new relaxation machinery. Remove
2216 tc_gen_reloc workarounds.
2217 (md_estimate_size_before_relax): Set RELAX_USE_SECOND if the second
2218 version of a relaxable macro is needed. Return -RELAX_SECOND if the
2219 first version is needed.
2220 (tc_gen_reloc): Remove relaxation handling.
2221 (md_convert_frag): Go through the fixups for a relaxable macro and
2222 mark those that belong to the unneeded alternative as done. If the
2223 second alternative is needed, adjust the fixup addresses to account
2224 for the deleted first alternative.
2226 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2228 * frags.h (frag_room): Declare.
2229 * frags.c (frag_room): New function.
2230 * doc/internals.texi: Document it.
2232 2004-01-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
2234 * config/tc-mips.c (append_insn): Don't do r3900 interlock
2235 optimization for -mtune=r3900, as this will break on other CPUs.
2237 2004-01-11 Tom Rix <tcrix@worldnet.att.net>
2239 * config/tc-m68hc11.c (build_indexed_byte): movb and movw cannot
2240 be relaxed, use fixup.
2241 (md_apply_fix3): Use 5 bit reloc from movb and movw fixup.
2243 2004-01-19 Jakub Jelinek <jakub@redhat.com>
2245 * config/tc-sparc.c (sparc_ip): Disallow %f32-%f63 for single
2248 2004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2250 * config/tc-mips.c (append_insn): Properly detect variant frags
2251 that preclude swapping of relaxed branches. Correctly swap
2252 instructions between frags when dealing with relaxed branches.
2254 2004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2256 * acinclude.m4: Quote names of macros to be defined by AC_DEFUN
2258 * aclocal.m4: Regenerate.
2259 * configure: Regenerate.
2261 2004-01-12 Anil Paranjpe <anilp1@KPITCummins.com>
2263 * config/tc-h8300.c (build_bytes): Apply relaxation to bit
2266 2004-01-12 Richard Sandiford <rsandifo@redhat.com>
2268 * config/tc-mips.c (macro_build_jalr): When adding an R_MIPS_JALR
2269 reloc, reserve space for the delay slot as well as the jalr itself.
2271 2004-01-09 Paul Brook <paul@codesourcery.com>
2273 * config/tc-arm.c (do_vfp_reg2_from_sp2): Rename from do_vfp_sp_reg2.
2274 (do_vfp_sp2_from_reg2): New function.
2276 (do_vfp_dp_from_reg2): Check return values properly.
2278 2004-01-08 Ian Lance Taylor <ian@wasabisystems.com>
2280 * config/tc-mips.c (warn_nops): Remove static variable.
2281 (macro): Remove test of warn_nops.
2282 (md_shortops): Remove 'n'.
2283 (md_parse_option): Remove 'n' case.
2284 (md_show_usage): Remove -n.
2285 * doc/as.texinfo (Overview): Remove MIPS -n option.
2286 * doc/c-mips.texi (MIPS Opts): Remove mention -n.
2287 * NEWS: Mention removal of MIPS -n option.
2289 * config/tc-mips.c (ISA_HAS_COPROC_DELAYS): Remove.
2290 (cop_interlocks): Check ISA level.
2291 (cop_mem_interlocks): Define.
2292 (reg_needs_delay): Check cop_interlocks rather than
2293 ISA_HAS_COPROC_DELAYS.
2294 (append_insn): Likewise. Use cop_mem_interlocks rather than
2295 directly checking mips_opts.isa.
2296 (mips_emit_delays): Likewise.
2298 2004-01-07 H.J. Lu <hongjiu.lu@intel.com>
2300 * config/tc-ia64.c (unwind): Move next_slot_number and
2301 next_slot_frag to ...
2302 (unw_rec_list): Here.
2303 (free_list_records): Removed.
2304 (output_unw_records): Likewise.
2305 (generate_unwind_image): Make it void.
2306 (alloc_record): Initialize next_slot_number and next_slot_frag.
2307 (slot_index): Take .org, .space and .align into account.
2308 (fixup_unw_records): Don't set slot_number to 0. Use
2309 list->next_slot_number and list->next_slot_frag instead of
2310 unwind.next_slot_number and unwind.next_slot_frag.
2311 (ia64_convert_frag): New.
2312 (generate_unwind_image): Generate a rs_machine_dependent frag
2314 (emit_one_bundle): Use list->next_slot_number and
2315 list->next_slot_frag instead of unwind.next_slot_number and
2316 unwind.next_slot_frag.
2318 * config/tc-ia64.h (md_convert_frag): Defined as
2320 (md_estimate_size_before_relax): Defined as (f)->fr_var.
2322 2004-01-06 Alexandre Oliva <aoliva@redhat.com>
2324 2003-12-19 Alexandre Oliva <aoliva@redhat.com>
2325 * config/tc-frv.h (md_apply_fix3): Don't define.
2326 * config/tc-frv.c (md_apply_fix3): New. Shift/truncate %hi/%lo
2328 * config/tc-frv.h (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
2329 2003-10-07 Alexandre Oliva <aoliva@redhat.com>
2330 * config/tc-frv.c (line_separator_chars): Add `!'.
2331 2003-09-19 Alexandre Oliva <aoliva@redhat.com>
2332 * config/tc-frv.c (md_assemble): Clear insn upfront.
2333 2003-09-18 Alexandre Oliva <aoliva@redhat.com>
2334 * config/tc-frv.c (OPTION_FDPIC): New macro.
2335 (md_longopts): Add mfdpic.
2336 (md_parse_option): Handle it.
2337 2003-08-04 Alexandre Oliva <aoliva@redhat.com>
2338 * config/tc-frv.c (md_cgen_lookup_reloc) <FRV_OPERAND_D12,
2339 FRV_OPERAND_S12>: Use reloc type encoded in fix-up.
2340 (frv_pic_ptr): Parse funcdesc.
2342 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2344 * doc/as.texinfo: Let texi2pod parse asconfig.texi and
2345 gasver.texi. Remove duplicate symbol definitions for texi2pod.
2347 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2349 * Makefile.am (Makefile): Move the dependency on
2350 $(BFDDIR)/configure.in to...
2351 (CONFIG_STATUS_DEPENDENCIES): ... here.
2352 (AUTOMAKE_OPTIONS): Require automake 1.8.
2353 * Makefile.in: Regenerate.
2354 * doc/Makefile.am (BASEDIR, BFDDIR): Define.
2355 (CONFIG_STATUS_DEPENDENCIES): Add a dependency on
2356 $(BFDDIR)/configure.in here as well.
2357 * doc/Makefile.in: Regenerate.
2359 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2361 * Makefile.am (install, install-info, RECURSIVE_TARGETS): Remove.
2362 * Makefile.in: Regenerate.
2363 * aclocal.m4: Regenerate.
2364 * doc/Makefile.am (install, install-info): Remove.
2365 (install-data-local): A new hook for install-info.
2366 (AUTOMAKE_OPTIONS): Require automake 1.8.
2367 * doc/Makefile.in: Regenerate.
2369 2004-01-02 Nutan Singh <nutan@kpitcummins.com>
2371 * doc/c-sh.texi: Update description about floating point behavior
2374 2004-01-02 Bernardo Innocenti <bernie@develer.com>
2376 * configure.in: Add m68k-uClinux target.
2377 * configure: Regenerate.
2379 For older changes see ChangeLog-0203
2385 version-control: never