* dwarf2dbg.c (dwarf2_finish): Don't write a .debug_line section
[deliverable/binutils-gdb.git] / gas / ChangeLog
1 2004-11-22 Bob Wilson <bob.wilson@acm.org>
2
3 * dwarf2dbg.c (dwarf2_finish): Don't write a .debug_line section
4 without a corresponding .debug_info section.
5
6 2004-11-22 Hans-Peter Nilsson <hp@axis.com>
7
8 * read.c (potable): Add "error" and "warning".
9 (s_errwarn): New function.
10 * read.h (s_errwarn): Declare.
11 * doc/as.texinfo (Error, Warning): Document .error and .warning.
12
13 2004-11-22 Nick Clifton <nickc@redhat.com>
14
15 * config/tc-tic54x.c (tic54x_adjust_symtab): Adjust call to
16 c_dot_file_symbol.
17
18 2004-11-19 Alan Modra <amodra@bigpond.net.au>
19
20 * config/tc-msp430.c (struct rcodes_s, MSP430_RLC, msp430_rcodes,
21 struct hcodes_s, msp430_hcodes): From include/opcode/msp430.h.
22
23 2004-11-19 Alan Modra <amodra@bigpond.net.au>
24
25 * config/obj-coff.c (c_dot_file_symbol): Add "app" param.
26 (coff_adjust_symtab): Adjust call.
27 (crawl_symbols): Likewise.
28 * config/obj-coff.h (c_dot_file_symbol): Add "app" param.
29 (obj_app_file): Adjust.
30
31 2004-11-18 Inderpreet Singh <inderpreetb@nioda.hcltech.com>
32
33 * configure.in: Enable bfd_assember for the MAXQ port.
34 * configure: Regenerate.
35
36 2004-11-12 Bob Wilson <bob.wilson@acm.org>
37 Sterling Augustine <sterling@tensilica.com>
38
39 * config/tc-xtensa.c (finish_vinsn): Clear pending instruction if
40 there is a conflict.
41 (check_t1_t2_reads_and_writes): Check for both reads and writes to
42 interfaces that are related as determined by xtensa_interface_class_id.
43
44 2004-11-12 Nick Clifton <nickc@redhat.com>
45
46 * config/tc-mn10300.c (md_relax_table): Fix off by one negative
47 offsets for conditional branches.
48
49 2004-11-11 Bob Wilson <bob.wilson@acm.org>
50
51 * config/tc-xtensa.c (MAX_IMMED6): Change value to 65.
52
53 2004-11-10 Bob Wilson <bob.wilson@acm.org>
54
55 * config/tc-xtensa.c (update_next_frag_state): Always add a NOP if
56 relaxing at the end of a loop. Don't mark frags as UNREACHABLE or
57 MAYBE_UNREACHABLE.
58 (relax_frag_immed): Update call to update_next_frag_state.
59
60 2004-11-10 Alan Modra <amodra@bigpond.net.au>
61
62 * obj.h (struct format_ops <app_file>): Add int param.
63 * read.h (s_app_file_string): Likewise.
64 * read.c (s_app_file_string): Likewise.
65 (s_app_file): Adjust s_app_file_string call.
66 * config/tc-mips.c (s_mips_file): Likewise.
67 * config/obj-coff.h (obj_app_file): Add app param.
68 * config/obj-ecoff.h (obj_app_file): Likewise.
69 * config/obj-multi.h (obj_app_file): Likewise.
70 * config/obj-elf.h (elf_file_symbol): Likewise.
71 * config/obj-elf.c (elf_file_symbol): Only emit one file symbol
72 if called for # preprocessor lines.
73
74 2004-11-08 H.J. Lu <hongjiu.lu@intel.com>
75
76 PR 528
77 * symbols.c (resolve_symbol_value): Convert weak symbols only
78 for Windows PECOFF.
79 (symbol_equated_reloc_p): Don't equate weaks when relocating
80 only for Windows PECOFF.
81
82 2004-11-08 Tomer Levi <Tomer.Levi@nsc.com>
83
84 * config/tc-crx.c (print_insn): Check and set insn_addr.
85 * config/tc-crx.h (md_frag_check): Define.
86
87 2004-11-08 Inderpreet Singh <inderpreetb@nioda.hcltech.com>
88 Vineet Sharma <vineets@noida.hcltech.com>
89
90 * configure.in: Add support for new target maxq-coff.
91 * configure: Regenerate.
92 * NEWS: Mention new support.
93 * config/tc-maxq.c: New file.
94 * config/tc-maxq.h: New file.
95 * config/obj-coff.h: Add support for maxq-coff.
96
97 2004-11-08 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
98
99 * symbols.c (any_external_name): Define.
100 (resolve_symbol_value): Convert weak symbols.
101 (S_SET_EXTERNAL): Support any_external_name.
102 (S_SET_NAME): Qualify parameter const.
103 (symbol_equated_reloc_p): Equate to weaks when relocating.
104 * symbols.h (S_SET_NAME): Qualfiy parameter const.
105 * tc.h (any_external_name): Declare.
106 * config/obj-coff.c ("coff/pe.h"): Include for BFD
107 assemblers also.
108 (weak_is_altname): Declare and define.
109 (weak_name2altname): Same.
110 (weak_altname2name): Same.
111 (weak_uniquify): Same.
112 (weak_altprefix): Define.
113 (obj_coff_weak): Change .weak syntax and handling.
114 (coff_frob_symbol): Fix PE weak symbol alternates.
115 * config/obj-coff.h (USE_UNIQUE): Define.
116 * config/tc-i386.c (md_apply_fix3): Assume weak symbols
117 are in another segment.
118 (tc_gen_reloc): Remove broken addend hack.
119 doc/as.texinfo: Update.
120
121 2004-11-05 Sterling Augustine <sterling@tensilica.com>
122
123 * config/tc-xtensa.c (total_frag_text_expansion): New.
124 (md_estimate_size_before_relax): Use it.
125 (find_address_of_next_align_frag): Likewise.
126
127 2004-11-05 Tomer Levi <Tomer.Levi@nsc.com>
128
129 * config/tc-crx.c: Rename argument types.
130 (processing_arg_number): Rename to 'cur_arg_num'.
131 (get_number_of_bits): Rename to 'set_operand_size'.
132 (get_operandtype): Rename to 'parse_operand', totally rewrite.
133 (set_cons_rparams): Rename to 'set_operand', totally rewrite.
134 (set_indexmode_parameters): Remove function, integrate its code into
135 'set_operand'.
136 (set_operand_size): Get rid of 'Operand Number' function parameter -
137 use global variable 'cur_arg_num' instead.
138 Use a local 'argument' pointer to reference the current argument.
139 (parse_operand): Likewise.
140 (set_operand): Likewise.
141 (process_label_constant): Likewise.
142
143 2004-11-04 Bob Wilson <bob.wilson@acm.org>
144
145 * config/tc-xtensa.c: Remove XTENSA_SECTION_RENAME ifdefs.
146 (add_section_rename): Delete. Inlined into...
147 (build_section_rename): ...here. Use xstrdup instead of strdup.
148 (xtensa_section_rename): Drop "const" from argument and return types.
149 (md_show_usage): Indent to match show_usage().
150 * config/tc-xtensa.h: Remove XTENSA_SECTION_RENAME ifdefs.
151 (tc_canonicalize_section_name): Define.
152 (md_elf_section_rename): Remove unused macro.
153 * doc/as.texinfo (Overview): Document Xtensa --rename-section option.
154 * doc/c-xtensa.texi (Xtensa Options): Likewise.
155 (Frame Directive): Delete.
156
157 2004-11-04 Daniel Jacobowitz <dan@debian.org>
158
159 * configure.in: Remove arm-*-oabi and thumb-*-oabi.
160 * config/tc-arm.c (target_oabi): Delete.
161 (md_apply_fix3, elf32_arm_target_format): Remove target_oabi checks.
162 (arm_opts): Remove moabi.
163 * doc/as.texinfo (Overview): Remove documentation of -moabi.
164 * doc/c-arm.texi (ARM Options): Likewise.
165 * configure: Regenerated.
166
167 2004-11-04 Hans-Peter Nilsson <hp@axis.com>
168
169 * configure.in (crisv32): Recognize. AC_DEFINE_UNQUOTED
170 DEFAULT_CRIS_ARCH. Handle crisv32-*-linux-gnu* like
171 cris-*-linux-gnu* and crisv32-*-* like cris-*-*.
172 * configure: Regenerate.
173 * config/tc-cris.c (enum cris_archs): New.
174 (cris_mach, cris_arch_from_string, s_cris_arch, get_sup_reg)
175 (cris_insn_ver_valid_for_arch): New functions.
176 (DEFAULT_CRIS_ARCH): New macro, default to cris_any_v0_v10.
177 (cris_arch): New variable.
178 (md_pseudo_table): New pseudo .arch.
179 (err_for_dangerous_mul_placement): Initialize according to
180 DEFAULT_CRIS_ARCH.
181 (STATE_COND_BRANCH): Renamed from STATE_CONDITIONAL_BRANCH.
182 All users changed.
183 (STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON)
184 (STATE_ABS_BRANCH_V32, STATE_LAPC, BRANCH_BF_V32, BRANCH_BB_V32)
185 (BRANCH_WF_V32, BRANCH_WB_V32): New.
186 (BRANCH_BF, BRANCH_BB, BRANCH_WF, BRANCH_WB): Don't undef after
187 use in md_cris_relax_table.
188 (md_cris_relax_table): Add entries for STATE_COND_BRANCH_V32,
189 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
190 Update and improve head comment.
191 (OPTION_PIC): Define in terms of previous option, OPTION_US.
192 (OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): Similar.
193 (OPTION_ARCH): New.
194 (md_longopts): New option --march=...
195 (cris_any_v0_v10_long_jump_size, crisv32_long_jump_size): New
196 macros.
197 (md_long_jump_size): Initialize in terms of DEFAULT_CRIS_ARCH.
198 (HANDLE_RELAXABLE): New macro.
199 (md_estimate_size_before_relax): Use HANDLE_RELAXABLE for common
200 cases. Check for weak symbols and assume not relaxable. Handle
201 STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON,
202 STATE_ABS_BRANCH_V32, STATE_LAPC. Use new variable symbolP, not
203 fragP->fr_symbol.
204 (md_convert_frag): Handle STATE_COND_BRANCH_V32,
205 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
206 (cris_create_short_jump): Adjust for CRISv32.
207 (cris_relax_frag): Handle new states.
208 (md_create_long_jump): Ditto. Emit error for common_v10_v32.
209 (md_begin): Define symbols "..asm.arch.cris.v32",
210 "..asm.arch.cris.v10", "..asm.arch.cris.common_v10_v32" and
211 "..asm.arch.cris.any_v0_v10". Use cris_insn_ver_valid_for_arch
212 when entering opcode table entry points.
213 (md_assemble): Adjust branch handling for CRISv32. Handle LAPC
214 relaxation. In fix_new_exp call for main insn, pass 1 for pcrel
215 parameter for 8, 16 and 32-bit pc-relative insns and LAPC.
216 (cris_process_instruction): Initialize out_insnp->insn_type to
217 CRIS_INSN_NONE, not CRIS_INSN_NORMAL.
218 <case ']', '[', 'A', 'd', 'Q', 'N', 'n', 'Y', 'U', 'u', 'T'>: New
219 cases.
220 <case 'm'>: Check that modified_char == '.'.
221 <invalid operands>: Consume the rest of the line.
222 When operands don't match, skip over subsequent insns with
223 non-matching version specifier but same mnemonic.
224 <immediate constant, case SIZE_SPEC_REG>: Immediate operands for
225 special registers in CRISv32 are always 32 bit long.
226 <immediate constant, case SIZE_FIELD_SIGNED, SIZE_FIELD_UNSIGNED>:
227 New cases.
228 (get_gen_reg): Only recognize "PC" when followed by "+]" for v32
229 and compatible. Recognize "ACR" for v32, unless followed by "+".
230 (get_spec_reg): Consider cris_arch when looking up register.
231 (get_autoinc_prefix_or_indir_op): Don't recognize assignment for
232 v32 or compatible.
233 (get_3op_or_dip_prefix_op): Check for ']' after seeing '[rN+'.
234 (cris_get_expression): Restore input_line_pointer if failing "early".
235 (get_flags): Consider cris_arch and recognize flags accordingly.
236 (branch_disp): Adjust for CRISv32.
237 (gen_cond_branch_32): Similar. Emit error for common_v10_v32.
238 (cris_number_to_imm): Use as_bad_where, not as_bad. Remove
239 related FIXME. Don't insist on BFD_RELOC_32_PCREL fixup to be
240 resolved. Don't enter zeros in object file for
241 BFD_RELOC_32_PCREL.
242 <case BFD_RELOC_CRIS_LAPCQ_OFFSET, BFD_RELOC_CRIS_SIGNED_16>
243 <case BFD_RELOC_CRIS_SIGNED_8>: New case.
244 (md_parse_option): Break out "return 1".
245 <OPTION_ARCH> New case.
246 (tc_gen_reloc): <case BFD_RELOC_CRIS_LAPCQ_OFFSET>
247 <case BFD_RELOC_CRIS_SIGNED_16, BFD_RELOC_CRIS_SIGNED_8>
248 <case BFD_RELOC_CRIS_UNSIGNED_8, BFD_RELOC_CRIS_UNSIGNED_16>
249 <case BFD_RELOC_32_PCREL>: New cases.
250 Addends for non-zero fx_pcrel are too in fx_offset.
251 (md_show_usage): Show --march=<arch>.
252 (md_apply_fix3): Adjust val for BFD_RELOC_CRIS_LAPCQ_OFFSET.
253 (md_pcrel_from): BFD_RELOC_CRIS_LAPCQ_OFFSET is PC-relative too.
254 (s_syntax) <struct syntaxes>: Properly constify member operand.
255 * config/tc-cris.h (TARGET_MACH): Define.
256 (cris_mach): Declare.
257 * doc/as.texinfo (Overview) <CRIS>: Add --march=...
258 * doc/c-cris.texi (CRIS-Symbols): New node for built-in symbols.
259 (CRIS-Opts): Document --march=...
260 (CRIS-Pseudos): Document .arch.
261
262 2004-11-04 Jan Beulich <jbeulich@novell.com>
263
264 * config/tc-i386.c (set_intel_syntax): Allow % in symbol names when
265 intel syntax and no register prefix, allow $ in symbol names when
266 intel syntax.
267 (set_16bit_gcc_code_flag): Replace literal 'l' by LONG_MNEM_SUFFIX.
268 (intel_float_operand): Add fourth return value indicating math control
269 operations. Make classification more precise.
270 (md_assemble): Complain if memory operand of mov[sz]x has no size
271 specified.
272 (parse_insn): Translate word operands to floating point instructions
273 operating on integers as well as control instructions to short ones
274 as expected by AT&T syntax. Translate 'd' suffix to short one only for
275 floating point instructions operating on non-integer operands.
276 (match_template): Remove fldcw special case. Adjust q-suffix handling
277 to permit it on fild/fistp/fisttp in AT&T mode.
278 (process_suffix): Don't guess DefaultSize insns' suffix from
279 stackop_size for certain floating point control instructions. Guess
280 suffix for branch and [ls][gi]dt based on flag_code. Split error
281 messages for Intel and AT&T syntax, and make the condition more strict
282 for the former. Adjust suppressing of generation of operand size
283 overrides.
284 (intel parser): Allow the full set of MASM operators. Add FWORD, TBYTE,
285 OWORD, and XMMWORD operand size specifiers (TBYTE replaces XWORD). Add
286 more error checking.
287 * config/tc-i386.h (BYTE_PTR WORD_PTR DWORD_PTR QWORD_PTR XWORD_PTR
288 SHORT OFFSET_FLAT FLAT NONE_FOUND): Remove unused defines.
289
290 2004-11-03 Hans-Peter Nilsson <hp@axis.com>
291
292 * symbols.c (colon) [!WORKING_DOT_WORD]: Don't declare
293 md_short_jump_size, md_long_jump_size.
294 * write.c [!WORKING_DOT_WORD]: Ditto.
295 * tc.h [!WORKING_DOT_WORD]: Declare them here. Drop const
296 qualifier.
297 * config/tc-cris.h (md_short_jump_size, md_long_jump_size): Don't
298 declare.
299 * config/tc-cris.c (md_short_jump_size, md_long_jump_size): Drop
300 const qualifier in these definitions.
301 * config/tc-i370.c, config/tc-m68k.c, config/tc-pdp11.c,
302 config/tc-s390.c, config/tc-tahoe.c, config/tc-vax.c: Ditto.
303
304 2004-11-02 Nick Clifton <nickc@redhat.com>
305
306 * dwarf2dbg.c (dwarf2_finish): Check for the existence of a file
307 table before deciding to produce a .debug_line section to match up
308 with a user provided .debug_info section.
309
310 2004-10-28 Tomer Levi <Tomer.Levi@nsc.com>
311
312 * config/tc-crx.c (getreg_image): Bug fix, a return value was
313 mistakenly omitted from CRX_C_REGTYPE and CRX_CS_REGTYPE cases.
314
315 2004-10-27 Tomer Levi <Tomer.Levi@nsc.com>
316
317 * config/tc-crx.c: Remove global variable 'post_inc_mode'.
318 (get_flags): New function.
319 (get_number_of_bits): Edit comments, update numeric values to
320 supported sizes.
321 (process_label_constant): Don't support the colon
322 format (SYMBOL:[s|m|l]).
323 (set_cons_rparams): Support argument type 'arg_rbase'.
324 (get_operandtype): Bug fix in 'rbase' operand type parsing.
325 (handle_LoadStor): Bug fix, first handle post-increment mode.
326 (getreg_image): Remove redundant code, update according to latest
327 CRX spec.
328 (print_constant): Bug fix relate to 3-word instructions.
329 (assemble_insn): Bug fix, when matching instructions, verify also
330 instruction type (not only mnemonic).
331 Add various error checking.
332 (preprocess_reglist): Support HI/LO and user registers.
333
334 2004-10-25 David Mosberger-Tang <davidm@hpl.hp.com>
335
336 * config/tc-ia64.c (fixup_unw_records): Don't let the "t" value
337 in an epilogue directive go negative.
338
339 2004-10-25 H.J. Lu <hongjiu.lu@intel.com>
340
341 PR 474
342 * config/tc-ia64.c (emit_one_bundle): Decrement md.num_slots_in_use
343 after reporting template error during manual bundling. Reported
344 by Michael Dupont, michaelx.dupont@intel.com.
345
346 2004-10-25 Daniel Jacobowitz <dan@debian.org>
347
348 * Makefile.am: Run dep-am.
349 * aclocal.m4: Regenerate with automake 1.9.2.
350 * Makefile.in: Regenerate with automake 1.9.2.
351 * doc/Makefile.in: Likewise.
352
353 * config/tc-arm.c: Include "dw2gencfi.h".
354 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
355 New functions.
356 * config/tc-arm.h (TARGET_USE_CFIPOP, DWARF2_DEFAULT_RETURN_COLUMN)
357 (DWARF2_CIE_DATA_ALIGNMENT, tc_regname_to_dw2regnum)
358 (tc_cfi_frame_initial_instructions): Define.
359 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
360 Add prototypes.
361
362 2004-10-21 Tomer Levi <Tomer.Levi@nsc.com>
363
364 * config/tc-crx.c (assemble_insn): Check unsigned immediate
365 operands validity.
366 Update coprocessor id to be unsigned immediate.
367
368 2004-10-18 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
369
370 * config/tc-i386.c (O_secrel): Delete.
371 (tc_pe_dwarf2_emit_offset): New function.
372 * config/tc-i386.h (O_secrel): Define as O_md1.
373 (TC_DWARF2_EMIT_OFFSET): Define.
374
375 2004-10-18 Nick Clifton <nickc@redhat.com>
376
377 * config/tc-xstormy16.c (xstormy16_cons_fix_new): Accept and
378 ignore @fptr() directives for 4-byte fixups.
379
380 2004-10-15 Alan Modra <amodra@bigpond.net.au>
381
382 * Makefile.am: Run "make dep-am"
383 * Makefile.in: Regenerate.
384
385 2004-10-14 Bob Wilson <bob.wilson@acm.org>
386
387 * doc/c-xtensa.texi (Xtensa Options, Absolute Literals Directive):
388 Remove comments about placement of literal pools.
389 (Literal Directive): Update description of literal placement.
390 (Literal Prefix Directive): Remove statement that this does not apply
391 to absolute-mode literals. Describe new section naming scheme.
392
393 2004-10-12 Bob Wilson <bob.wilson@acm.org>
394
395 * config/tc-xtensa.c (xtensa_frequency_pseudo): Use set_subseg_freq.
396 (is_entry_opcode, is_movi_opcode, is_the_loop_opcode, is_jx_opcode,
397 is_windowed_return_opcode): Delete.
398 (xtensa_frob_label): Use get_subseg_target_freq.
399 (md_assemble): Inline call to is_entry_opcode.
400 (xtensa_handle_align): Inline call to get_frag_is_literal.
401 (relaxation_requirements): Inline call to is_jx_opcode.
402 (emit_single_op): Inline call to is_movi_opcode.
403 (xg_assemble_vliw_tokens): Inline calls to get_frag_is_insn,
404 get_frag_is_no_transform, is_entry_opcode, and
405 set_frag_is_specific_opcode. Use get_subseg_total_freq.
406 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags,
407 xtensa_fix_close_loop_end_frags, relax_frag_immed, convert_frag_immed):
408 Inline calls to get_frag_is_no_transform.
409 (next_instrs_are_b_retw): Inline call to is_windowed_return_opcode.
410 (xtensa_fix_short_loop_frags): Inline calls to is_the_loop_opcode and
411 get_frag_is_no_transform.
412 (convert_frag_immed_finish_loop): Inline calls to get_expression_value
413 and set_frag_is_no_transform.
414 (get_expression_value): Delete.
415 (subseg_map struct): Rename cur_total_freq to total_freq. Rename
416 cur_target_freq to target_freq.
417 (get_subseg_info): Split out code to create a new map entry into ...
418 (add_subseg_info): ... this new function.
419 (get_last_insn_flags): Check if get_subseg_info succeeded.
420 (set_last_insn_flags): Call add_subseg_info if needed.
421 (get_subseg_total_freq, get_subseg_target_freq, set_subseg_freq): New.
422 (xtensa_reorder_segments): Compute last_sec while counting sections.
423 Remove call to get_last_sec.
424 (get_last_sec): Delete.
425 (cache_literal_section): Inline call to retrieve_literal_seg and its
426 callees, seg_present and add_seg_list.
427 (retrieve_literal_seg, seg_present, add_seg_list): Delete.
428 (get_frag_is_insn, get_frag_is_no_transform,
429 set_frag_is_specific_opcode, set_frag_is_no_transform): Delete.
430 * config/tc-xtensa.h (MAX_SLOTS): Reduce from 31 to 15.
431
432 2004-10-12 Bob Wilson <bob.wilson@acm.org>
433
434 * config/tc-xtensa.c: Use ISO C90 formatting.
435 * config/tc-xtensa.h: Likewise.
436 * config/xtensa-istack.h: Likewise.
437 * config/xtensa-relax.c: Likewise.
438 * config/xtensa-relax.h: Likewise.
439
440 2004-10-12 Paul Brook <paul@codesourcery.com>
441
442 * config/tc-arm.c (md_begin): Change EF_ARM_EABI_VER3 to
443 EF_ARM_EABI_VER4.
444 (arm_eabis): Ditto.
445 * doc/c-arm.texi: Document that we actually support -meabi=4, not
446 -meabi=3.
447
448 2004-10-08 Bob Wilson <bob.wilson@acm.org>
449
450 * doc/as.texinfo (VTableEntry, VTableInherit): Add "directive" to index
451 entries.
452 (Acknowledgements): Use "GAS" instead of AS variable.
453
454 2004-10-08 Daniel Jacobowitz <dan@debian.org>
455
456 * config/tc-i386.c: Include "elf/x86-64.h".
457 (i386_elf_section_type): New function.
458 * config/tc-i386.h (md_elf_section_type): Define.
459 (i386_elf_section_type): New prototype.
460
461 2004-10-08 Linus Nielsen Feltzing <linus@haxx.se>
462
463 * config/m68k-parse.h (enum m68k_register): New control register,
464 MBAR2 (for MCF5249)
465 * config/tc-m68k.c: Correct control register set for MCF5249.
466
467 2004-10-07 Bob Wilson <bob.wilson@acm.org>
468 Sterling Augustine <sterling@tensilica.com>
469
470 * config/tc-xtensa.c (absolute_literals_supported): New global flag.
471 (UNREACHABLE_MAX_WIDTH): Define.
472 (XTENSA_FETCH_WIDTH): Delete.
473 (cur_vinsn, xtensa_fetch_width, xt_saved_debug_type, past_xtensa_end,
474 prefer_const16, prefer_l32r): New global variables.
475 (LIT4_SECTION_NAME): Define.
476 (lit4_state struct): Add lit4_seg_name and lit4_seg fields.
477 (XTENSA_PROP_*, GET_XTENSA_PROP_*, SET_XTENSA_PROP_*): Define.
478 (frag_flags struct): New.
479 (xtensa_block_info struct): Move from tc-xtensa.h. Add flags field.
480 (subseg_map struct): Add cur_total_freq and cur_target_freq fields.
481 (bitfield, bit_is_set, set_bit, clear_bit): Define.
482 (MAX_FORMATS): Define.
483 (op_placement_info struct, op_placement_table): New.
484 (O_pltrel, O_hi16, O_lo16): Define.
485 (directiveE enum): Rename directive_generics to directive_transform.
486 Delete directive_relax. Add directive_schedule,
487 directive_absolute_literals, and directive_last_directive.
488 (directive_info): Rename "generics" to "transform". Delete "relax".
489 Add "schedule" and "absolute-literals".
490 (directive_state): Adjust entries to match changes in directive_info.
491 (xtensa_relax_statesE, RELAX_IMMED_MAXSTEPS): Move to tc-xtensa.h.
492 (xtensa_const16_opcode, xtensa_movi_opcode, xtensa_movi_n_opcode,
493 xtensa_l32r_opcode, xtensa_nop_opcode, xtensa_rsr_lcount_opcode): New.
494 (xtensa_j_opcode, xtensa_rsr_opcode): Delete.
495 (align_only_targets, software_a0_b_retw_interlock,
496 software_avoid_b_j_loop_end, maybe_has_b_j_loop_end,
497 software_avoid_short_loop, software_avoid_close_loop_end,
498 software_avoid_all_short_loops, specific_opcode): Delete.
499 (warn_unaligned_branch_targets): New.
500 (workaround_a0_b_retw, workaround_b_j_loop_end, workaround_short_loop,
501 workaround_close_loop_end, workaround_all_short_loops): Default FALSE.
502 (option_[no_]link_relax, option_[no_]transform,
503 option_[no_]absolute_literals, option_warn_unaligned_targets,
504 option_prefer_l32r, option_prefer_const16, option_target_hardware):
505 New enum values.
506 (option_[no_]align_only_targets, option_literal_section_name,
507 option_text_section_name, option_data_section_name,
508 option_bss_section_name, option_eb, option_el): Delete.
509 (md_longopts): Add entries for: [no-]transform, [no-]absolute-literals,
510 warn-unaligned-targets, prefer-l32r, prefer-const16, [no-]link-relax,
511 and target-hardware. Delete entries for [no-]target-align-only,
512 literal-section-name, text-section-name, data-section-name, and
513 bss-section-name.
514 (md_parse_option): Handle new options and remove old ones. Accept but
515 ignore [no-]density options. Warn for [no-]generics and [no-]relax
516 and treat them as [no-]transform.
517 (md_show_usage): Add new options and remove old ones.
518 (xtensa_setup_hw_workarounds): New.
519 (md_pseudo_table): Change "word" entry to use xtensa_elf_cons. Add
520 "long", "short", "loc" and "frequency" entries.
521 (use_generics): Rename to ...
522 (use_transform): ... this function. Add past_xtensa_end check.
523 (use_longcalls): Add past_xtensa_end check.
524 (code_density_available, can_relax): Delete.
525 (do_align_targets): New.
526 (get_directive): Accept dashes in directive names. Warn about
527 [no-]generics and [no-]relax directives and treat them as
528 [no-]transform.
529 (xtensa_begin_directive): Call md_flush_pending_output only for some
530 directives. Check for directives inside instruction bundles. Warn
531 about deprecated ".begin literal" usage. Warn and ignore [no-]density
532 directives. Handle new directives. Check generating_literals flag
533 for literal_prefix.
534 (xtensa_end_directive): Check for directives inside instruction
535 bundles. Warn and ignore [no-]density directives. Handle new
536 directives. Call xtensa_set_frag_assembly_state.
537 (xtensa_loc_directive_seen, xtensa_dwarf2_directive_loc,
538 xtensa_dwarf2_emit_insn): New.
539 (xtensa_literal_position): Call md_flush_pending_output. Do not check
540 use_literal_section flag.
541 (xtensa_literal_pseudo): Call md_flush_pending_output. Handle absolute
542 literals. Use xtensa_elf_cons to parse the expression.
543 (xtensa_literal_prefix): Do not check use_literal_section. Support
544 ".lit4" sections for absolute literals. Change prefix convention to
545 replace ".text" (or ".t" in a linkonce section). No need to call
546 subseg_set.
547 (xtensa_frequency_pseudo, xtensa_elf_cons, xtensa_elf_suffix): New.
548 (expression_end): Handle closing braces and colons.
549 (PLT_SUFFIX, plt_suffix): Delete.
550 (expression_maybe_register): Use new xtensa-isa.h functions. Use
551 xtensa_elf_suffix instead of checking for plt suffix, and handle O_lo16
552 and O_hi16 expressions as well.
553 (tokenize_arguments): Handle closing braces and colons.
554 (parse_arguments): Use new xtensa-isa.h functions. Handle "invisible"
555 operands and paired register syntax.
556 (get_invisible_operands): New.
557 (xg_translate_sysreg_op): Handle new Xtensa LX RSR/WSR/XSR syntax. Use
558 new xtensa-isa.h functions.
559 (xtensa_translate_old_userreg_ops, xtensa_translate_zero_immed): New.
560 (xg_translate_idioms): Check if inside bundle. Use use_transform.
561 Handle new Xtensa LX RSR/WSR/XSR syntax. Remove code to widen density
562 instructions. Use xtensa_translate_zero_immed.
563 (operand_is_immed, operand_is_pcrel_label): Delete.
564 (get_relaxable_immed): Use new xtensa-isa.h functions.
565 (get_opcode_from_buf): Add slot parameter. Use new xtensa-isa.h
566 functions.
567 (xtensa_print_insn_table, print_vliw_insn): New.
568 (is_direct_call_opcode): Use new xtensa-isa.h functions.
569 (is_call_opcode, is_loop_opcode, is_conditional_branch_opcode,
570 is_branch_or_jump_opcode): Delete.
571 (is_movi_opcode, decode_reloc, encode_reloc, encode_alt_reloc): New.
572 (opnum_to_reloc, reloc_to_opnum): Delete.
573 (xtensa_insnbuf_set_operand, xtensa_insnbuf_get_operand): Use new
574 xtensa-isa.h functions. Operate on one slot of an instruction.
575 (xtensa_insnbuf_set_immediate_field, is_negatable_branch,
576 xg_get_insn_size): Delete.
577 (xg_get_build_instr_size): Use xg_get_single_size.
578 (xg_is_narrow_insn, xg_is_single_relaxable_insn): Update calls to
579 xg_build_widen_table. Use xg_get_single_size.
580 (xg_get_max_narrow_insn_size): Delete.
581 (xg_get_max_insn_widen_size, xg_get_max_insn_widen_literal_size,
582 xg_is_relaxable_insn): Update calls to xg_build_widen_table. Use
583 xg_get_single_size.
584 (xg_build_to_insn): Record the loc field. Handle OP_OPERAND_HI16U and
585 OP_OPERAND_LOW16U. Check xg_valid_literal_expression.
586 (xg_expand_to_stack, xg_expand_narrow): Update calls to
587 xg_build_widen_table. Use xg_get_single_size.
588 (xg_immeds_fit): Use new xtensa-isa.h functions. Update call to
589 xg_check_operand.
590 (xg_symbolic_immeds_fit): Likewise. Also handle O_lo16 and O_hi16, and
591 treat weak symbols conservatively.
592 (xg_check_operand): Use new xtensa-isa.h functions.
593 (is_dnrange): Delete.
594 (xg_assembly_relax): Inline previous calls to tinsn_copy.
595 (xg_finish_frag): Specify separate relax states for the frag and slot0.
596 (is_branch_jmp_to_next, xg_add_branch_and_loop_targets): Use new
597 xtensa-isa.h functions.
598 (xg_instruction_matches_option_term, xg_instruction_matches_or_options,
599 xg_instruction_matches_options): New.
600 (xg_instruction_matches_rule): Handle O_register expressions. Call
601 xg_instruction_matches_options.
602 (transition_rule_cmp): New.
603 (xg_instruction_match): Update call to xg_build_simplify_table.
604 (xg_build_token_insn): Record loc fields.
605 (xg_simplify_insn): Check is_specific_opcode field and
606 density_supported flag.
607 (xg_expand_assembly_insn): Skip checking code_density_available. Use
608 new xtensa-isa.h functions. Call use_transform instead of can_relax.
609 (xg_assemble_literal): Add error handling for O_big. Call
610 record_alignment. Handle O_pltrel.
611 (xg_valid_literal_expression): New.
612 (xg_assemble_literal_space): Add slot parameter. Remove call to
613 set_expr_symbol_offset. Add call to record_alignment. Update call to
614 xg_finish_frag.
615 (xg_emit_insn): Delete.
616 (xg_emit_insn_to_buf): Add format parameter. Update calls to
617 xg_add_opcode_fix and xtensa_insnbuf_to_chars.
618 (xg_add_opcode_fix): Change opcode parameter to tinsn and add format
619 and slot parameters. Handle new "alternate" relocations for absolute
620 literals and CONST16 instructions. Check for bad uses of O_lo16 and
621 O_hi16. Use new xtensa-isa.h functions.
622 (xg_assemble_tokens): Delete.
623 (is_register_writer): Use new xtensa-isa.h functions.
624 (is_bad_loopend_opcode): Check for xtensa_rsr_lcount_opcode instead of
625 old-style RSR from LCOUNT.
626 (next_frag_opcode): Delete.
627 (next_frag_opcode_is_loop, next_frag_format_size, frag_format_size,
628 update_next_frag_state): New.
629 (update_next_frag_nop_state): Delete.
630 (next_frag_pre_opcode_bytes): Use next_frag_opcode_is_loop.
631 (xtensa_mark_literal_pool_location): Check use_literal_section flag and
632 the state of the absolute-literals directive. Add calls to
633 record_alignment and xtensa_set_frag_assembly_state. Call
634 xtensa_switch_to_non_abs_literal_fragment instead of
635 xtensa_switch_to_literal_fragment.
636 (build_nop): New.
637 (assemble_nop): Use build_nop. Update call to xtensa_insnbuf_to_chars.
638 (get_expanded_loop_offset): Change check for undefined opcode to an
639 assertion.
640 (xtensa_set_frag_assembly_state, relaxable_section,
641 xtensa_find_unmarked_state_frags, xtensa_find_unaligned_branch_targets,
642 xtensa_find_unaligned_loops, xg_apply_tentative_value): New.
643 (md_begin): Update call to xtensa_isa_init. Initialize linkrelax to 1.
644 Set lit4_seg_name. Call xg_init_vinsn. Initialize new global opcodes.
645 Call init_op_placement_info_table and xtensa_set_frag_assembly_state.
646 (xtensa_init_fix_data): New.
647 (xtensa_frob_label): Reset label symbol to the current frag. Check
648 do_align_targets and generating_literals flag. Propagate frequency
649 info to new alignment frag. Call xtensa_set_frag_assembly_state.
650 (xtensa_unrecognized_line): New.
651 (xtensa_flush_pending_output): Check if inside a bundle. Add a call
652 to xtensa_set_frag_assembly_state.
653 (error_reset_cur_vinsn): New.
654 (md_assemble): Remove check for literal frag. Remove call to
655 istack_init. Call use_transform instead of use_generics. Parse
656 explicit instruction format specifiers. Move code for
657 a0_b_retw_interlock workaround to xg_assemble_vliw_tokens. Call
658 error_reset_cur_vinsn on errors. Add call to get_invisible_operands.
659 Add dwarf2_where call. Remote automatic alignment for ENTRY
660 instructions. Move call to xtensa_clear_insn_labels to the end.
661 Rearrange to handle bundles.
662 (xtensa_cons_fix_new): Delete.
663 (xtensa_handle_align): New.
664 (xtensa_frag_init): Call xtensa_set_frag_assembly_state. Remove
665 assignment to is_no_density field.
666 (md_pcrel_from): Use new xtensa-isa.h functions. Use decode_reloc
667 instead of reloc_to_opnum. Handle "alternate" relocations.
668 (xtensa_force_relocation, xtensa_check_inside_bundle,
669 xtensa_elf_section_change_hook): New.
670 (xtensa_symbol_new_hook): Delete.
671 (xtensa_fix_adjustable): Check for difference of symbols with an
672 offset. Check for external and weak symbols.
673 (md_apply_fix3): Remove cases for XTENSA_OP{0,1,2} relocs.
674 (md_estimate_size_before_relax): Return expansion for the first slot.
675 (tc_gen_reloc): Handle difference of symbols by producing
676 XTENSA_DIFF{8,16,32} relocs and by writing the value of the difference
677 into the output. Handle new XTENSA_SLOT*_OP relocs by storing the
678 tentative values into the output when linkrelax is set.
679 (XTENSA_PROP_SEC_NAME): Define.
680 (xtensa_post_relax_hook): Call xtensa_find_unmarked_state_frags.
681 Create literal tables only if using literal sections. Create new
682 property tables instead of old instruction tables. Check for unaligned
683 branch targets and loops.
684 (finish_vinsn, find_vinsn_conflicts, check_t1_t2_reads_and_writes,
685 new_resource_table, clear_resource_table, resize_resource_table,
686 resources_available, reserve_resources, release_resources,
687 opcode_funcUnit_use_unit, opcode_funcUnit_use_stage,
688 resources_conflict, xg_find_narrowest_format, relaxation_requirements,
689 bundle_single_op, emit_single_op, xg_assemble_vliw_tokens): New.
690 (xtensa_end): Call xtensa_flush_pending_output. Set past_xtensa_end
691 flag. Update checks for workaround options. Call
692 xtensa_mark_narrow_branches and xtensa_mark_zcl_first_insns.
693 (xtensa_cleanup_align_frags): Add special case for branch targets.
694 Check for and mark unreachable frags.
695 (xtensa_fix_target_frags): Remove use of align_only_targets flag.
696 Use RELAX_LOOP_END_BYTES in special case for negatable branch at the
697 end of a zero-overhead loop body.
698 (frag_can_negate_branch): Handle instructions with multiple slots.
699 Use new xtensa-isa.h functions
700 (xtensa_mark_narrow_branches, is_narrow_branch_guaranteed_in_range,
701 xtensa_mark_zcl_first_insns): New.
702 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags): Error if
703 transformations are disabled.
704 (next_instrs_are_b_retw): Use new xtensa-isa.h functions. Handle
705 multislot instructions.
706 (xtensa_fix_close_loop_end_frags, xtensa_fix_short_loop_frags):
707 Likewise. Also error if transformations are disabled.
708 (unrelaxed_frag_max_size): New.
709 (unrelaxed_frag_min_insn_count, unrelax_frag_has_b_j): Use new
710 xtensa-isa.h functions.
711 (xtensa_sanity_check, is_empty_loop, is_local_forward_loop): Use
712 xtensa_opcode_is_loop instead of is_loop_opcode.
713 (get_text_align_power): Replace as_fatal with assertion.
714 (get_text_align_fill_size): Iterate instead of using modulus when
715 use_nops is false.
716 (get_noop_aligned_address): Assert that this is for a machine-dependent
717 RELAX_ALIGN_NEXT_OPCODE frag. Use next_frag_opcode_is_loop,
718 xg_get_single_size, and frag_format_size.
719 (get_widen_aligned_address): Rename to ...
720 (get_aligned_diff): ... this function. Add max_diff parameter.
721 Remove handling of rs_align/rs_align_code frags. Use
722 next_frag_format_size, get_text_align_power, get_text_align_fill_size,
723 next_frag_opcode_is_loop, and xg_get_single_size. Compute max_diff
724 and pass it back to caller.
725 (xtensa_relax_frag): Use relax_frag_loop_align. Add code for new
726 RELAX_SLOTS, RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN,
727 RELAX_FILL_NOP, and RELAX_UNREACHABLE frag types. Check relax_seen.
728 (relax_frag_text_align): Rename to ...
729 (relax_frag_loop_align): ... this function. Assume loops can only be
730 in the first slot of an instruction.
731 (relax_frag_add_nop): Use assemble_nop instead of constructing an OR
732 instruction. Remove call to frag_wane.
733 (relax_frag_narrow): Rename to ...
734 (relax_frag_for_align): ... this function. Extend to handle
735 RELAX_FILL_NOP and RELAX_UNREACHABLE, as well as RELAX_SLOTS with
736 RELAX_NARROW for the first slot.
737 (find_address_of_next_align_frag, bytes_to_stretch): New.
738 (future_alignment_required): Use find_address_of_next_align_frag and
739 bytes_to_stretch. Look ahead to subsequent frags to make smarter
740 alignment decisions.
741 (relax_frag_immed): Add format, slot, and estimate_only parameters.
742 Check if transformations are enabled for b_j_loop_end workaround.
743 Use new xtensa-isa.h functions and handle multislot instructions.
744 Update call to xg_assembly_relax.
745 (md_convert_frag): Handle new RELAX_SLOTS, RELAX_UNREACHABLE,
746 RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN, and RELAX_FILL_NOP
747 frag types.
748 (convert_frag_narrow): Add segP, format and slot parameters. Call
749 convert_frag_immed for branch instructions. Adjust calls to
750 tinsn_from_chars, tinsn_immed_from_frag, and xg_emit_insn_to_buf. Use
751 xg_get_single_size and xg_get_single_format.
752 (convert_frag_fill_nop): New.
753 (convert_frag_immed): Add format and slot parameters. Handle multislot
754 instructions and use new xtensa-isa.h functions. Update calls to
755 tinsn_immed_from_frag and xg_assembly_relax. Check if transformations
756 enabled for b_j_loop_end workaround. Use build_nop instead of
757 assemble_nop. Check is_specific_opcode flag. Check for unreachable
758 frags. Use xg_get_single_size. Handle O_pltrel.
759 (fix_new_exp_in_seg): Remove check for old plt flag.
760 (convert_frag_immed_finish_loop): Update calls to tinsn_from_chars and
761 xtensa_insnbuf_to_chars. Call tinsn_immed_from_frag. Change check
762 for loop opcode to an assertion. Mark all frags up to the end of the
763 loop as not transformable.
764 (get_last_insn_flags, set_last_insn_flags): Use get_subseg_info.
765 (get_subseg_info): New.
766 (xtensa_move_literals): Call xtensa_set_frag_assembly_state. Add null
767 check for dest_seg.
768 (xtensa_switch_to_literal_fragment): Rewrite to handle absolute
769 literals and use xtensa_switch_to_non_abs_literal_fragment otherwise.
770 (xtensa_switch_to_non_abs_literal_fragment): New.
771 (cache_literal_section): Add is_code parameter and pass it through to
772 retrieve_literal_seg.
773 (retrieve_literal_seg): Add is_code parameter and use it to set the
774 flags on the literal section. Handle case where head parameter is 0.
775 (get_frag_is_no_transform, set_frag_is_specific_opcode,
776 set_frag_is_no_transform): New.
777 (xtensa_create_property_segments): Add end_property_function parameter
778 and pass it through to add_xt_block_frags. Call bfd_get_section_flags
779 and skip SEC_DEBUGGING and !SEC_ALLOC sections.
780 (xtensa_create_xproperty_segments, section_has_xproperty): New.
781 (add_xt_block_frags): Add end_property_function parameter and call it
782 if it is non-zero. Call xtensa_frag_flags_init.
783 (xtensa_frag_flags_is_empty, xtensa_frag_flags_init,
784 get_frag_property_flags, frag_flags_to_number,
785 xtensa_frag_flags_combinable, xt_block_aligned_size,
786 xtensa_xt_block_combine, add_xt_prop_frags,
787 init_op_placement_info_table, opcode_fits_format_slot,
788 xg_get_single_size, xg_get_single_format): New.
789 (istack_push): Inline call to tinsn_copy.
790 (tinsn_copy): Delete.
791 (tinsn_has_invalid_symbolic_operands): Handle O_hi16 and O_lo16 and
792 CONST16 opcodes. Handle O_big, O_illegal, and O_absent.
793 (tinsn_has_complex_operands): Handle O_hi16 and O_lo16.
794 (tinsn_to_insnbuf): Use xg_get_single_format and new xtensa-isa.h
795 functions. Handle invisible operands.
796 (tinsn_to_slotbuf): New.
797 (tinsn_check_arguments): Use new xtensa-isa.h functions.
798 (tinsn_from_chars): Add slot parameter. Rewrite using xg_init_vinsn,
799 vinsn_from_chars, and xg_free_vinsn.
800 (tinsn_from_insnbuf): New.
801 (tinsn_immed_from_frag): Add slot parameter and handle multislot
802 instructions. Handle symbol differences.
803 (get_num_stack_text_bytes): Use xg_get_single_size.
804 (xg_init_vinsn, xg_clear_vinsn, vinsn_has_specific_opcodes,
805 xg_free_vinsn, vinsn_to_insnbuf, vinsn_from_chars, expr_is_register,
806 get_expr_register, set_expr_symbol_offset_diff): New.
807 * config/tc-xtensa.h (MAX_SLOTS): Define.
808 (xtensa_relax_statesE): Move from tc-xtensa.c. Add
809 RELAX_CHECK_ALIGN_NEXT_OPCODE, RELAX_MAYBE_DESIRE_ALIGN, RELAX_SLOTS,
810 RELAX_FILL_NOP, RELAX_UNREACHABLE, RELAX_MAYBE_UNREACHABLE, and
811 RELAX_NONE types.
812 (RELAX_IMMED_MAXSTEPS): Move from tc-xtensa.c.
813 (xtensa_frag_type struct): Add is_assembly_state_set,
814 use_absolute_literals, relax_seen, is_unreachable, is_specific_opcode,
815 is_align, is_text_align, alignment, and is_first_loop_insn fields.
816 Replace is_generics and is_relax fields by is_no_transform field.
817 Delete is_text and is_longcalls fields. Change text_expansion and
818 literal_expansion to arrays of MAX_SLOTS entries. Add arrays of
819 per-slot information: literal_frags, slot_subtypes, slot_symbols,
820 slot_sub_symbols, and slot_offsets. Add fr_prev field.
821 (xtensa_fix_data struct): New.
822 (xtensa_symfield_type struct): Delete plt field.
823 (xtensa_block_info struct): Move definition to tc-xtensa.h. Add
824 forward declaration here.
825 (xt_section_type enum): Delete xt_insn_sec. Add xt_prop_sec.
826 (XTENSA_SECTION_RENAME): Undefine.
827 (TC_FIX_TYPE, TC_INIT_FIX_DATA, TC_FORCE_RELOCATION, NO_PSEUDO_DOT,
828 tc_unrecognized_line, md_do_align, md_elf_section_change_hook,
829 HANDLE_ALIGN, TC_LINKRELAX_FIXUP, SUB_SEGMENT_ALIGN): Define.
830 (TC_CONS_FIX_NEW, tc_symbol_new_hook): Delete.
831 (unit_num_copies_func, opcode_num_units_func,
832 opcode_funcUnit_use_unit_func, opcode_funcUnit_use_stage_func): New.
833 (resource_table struct): New.
834 * config/xtensa-istack.h (MAX_INSN_ARGS): Increase from 6 to 10.
835 (TInsn struct): Add keep_wide, loc, fixup, record_fix, subtype,
836 literal_space, symbol, sub_symbol, offset, and literal_frag fields.
837 (tinsn_copy): Delete prototype.
838 (vliw_insn struct): New.
839 * config/xtensa-relax.c (insn_pattern_struct): Add options field.
840 (widen_spec_list): Add option conditions for density and boolean
841 instructions. Add expansions using CONST16 and conditions for using
842 CONST16 vs. L32R. Use new Xtensa LX RSR/WSR syntax. Add entries for
843 predicted branches.
844 (simplify_spec_list): Add option conditions for density instructions.
845 Add entry for NOP instruction.
846 (append_transition): Add cmp function pointer parameter and use it to
847 insert the new entry in order.
848 (operand_function_LOW16U, operand_function_HI16U): New.
849 (xg_has_userdef_op_fn, xg_apply_userdef_op_fn): Handle
850 OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
851 (enter_opname, split_string): Use xstrdup instead of strdup.
852 (init_insn_pattern): Initialize new options field.
853 (clear_req_or_option_list, clear_req_option_list,
854 clone_req_or_option_list, clone_req_option_list, parse_option_cond):
855 New.
856 (parse_insn_pattern): Parse option conditions.
857 (transition_applies): New.
858 (build_transition): Use new xtensa-isa.h functions. Fix incorrectly
859 swapped last arguments in calls to append_constant_value_condition.
860 Call clone_req_option_list. Add warning about invalid opcode.
861 Handle LOW16U and HI16U function names.
862 (build_transition_table): Add cmp parameter and use it in calls to
863 append_transition. Use new xtensa-isa.h functions. Check
864 transition_applies before adding entries.
865 (xg_build_widen_table, xg_build_simplify_table): Add cmp parameter and
866 pass it through to build_transition_table.
867 * config/xtensa-relax.h (ReqOrOptionList, ReqOrOption, ReqOptionList,
868 ReqOption, transition_cmp_fn): New types.
869 (OpType enum): Add OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
870 (transition_rule struct): Add options field.
871 * doc/as.texinfo (Overview): Update Xtensa options.
872 * doc/c-xtensa.texi (Xtensa Options): Delete --[no-]density,
873 --[no-]relax, and --[no-]generics options. Update descriptions of
874 --text-section-literals and --[no-]longcalls. Add
875 --[no-]absolute-literals and --[no-]transform.
876 (Xtensa Syntax): Add description of syntax for FLIX instructions.
877 Remove use of "generic" and "specific" terminology for opcodes.
878 (Xtensa Registers): Generalize the syntax description to include
879 user-defined register files.
880 (Xtensa Automatic Alignment): Update.
881 (Xtensa Branch Relaxation): Mention limitation of unconditional jumps.
882 (Xtensa Call Relaxation): Linker can now remove most of the overhead.
883 (Xtensa Directives): Remove confusing rules about precedence.
884 (Density Directive, Relax Directive): Delete.
885 (Schedule Directive): New.
886 (Generics Directive): Rename to ...
887 (Transform Directive): ... this node.
888 (Literal Directive): Update for absolute literals. Missing
889 literal_position directive is now an error.
890 (Literal Position Directive): Update for absolute literals.
891 (Freeregs Directive): Delete.
892 (Absolute Literals Directive): New.
893 (Frame Directive): Minor editing.
894 * Makefile.am (DEPTC_xtensa_elf, DEPOBJ_xtensa_elf, DEP_xtensa_elf):
895 Update dependencies.
896 * Makefile.in: Regenerate.
897
898 2004-10-07 Richard Sandiford <rsandifo@redhat.com>
899
900 * config/tc-mips.c (append_insn): Use fix_new rather than fix_new_exp
901 to build the second and third fixups for a composite relocation.
902 (macro_read_relocs): New function.
903 (macro_build): Use it.
904 (s_cpsetup): Pass all three composite relocation codes to macro_build.
905 Simplify fragging code accordingly.
906 (s_gpdword): Use fix_new rather than fix_new_exp for the second part
907 of the composite relocation. Set fx_tcbit in both fixups.
908
909 2004-10-07 Richard Sandiford <rsandifo@redhat.com>
910
911 * config/tc-mips.c (append_insn): Set fx_tcbit for composite relocs.
912 (md_apply_fix3): Don't treat composite relocs as done.
913
914 2004-10-07 Jan Beulich <jbeulich@novell.com>
915
916 * macro.c (macro_expand_body): When ELF, use .LL rather than LL as
917 prefix for symbol names generated from the LOCAL macro directive.
918
919 * dw2gencfi.c (select_cie_for_fde): When separating CIE out from
920 FDE, treat a DW_CFA_remember_state as we do a DW_CFA_advance_loc.
921
922 2004-10-07 Tomer Levi <Tomer.Levi@nsc.com>
923
924 * config/tc-crx.c (preprocess_reglist): Handle Co-processor
925 Special registers.
926 (md_assemble): Add error checking for Co-Processor instructions.
927 (get_cinv_parameters): Add 'b' option to invalidate the
928 branch-target cache.
929
930 2004-10-05 Paul Brook <paul@codesourcery.com>
931
932 * config/tc-arm.c (unwind): New variable.
933 (vfp_sp_encode_reg): New function.
934 (vfp_sp_reg_required_here): Use it.
935 (vfp_sp_reg_list, vfp_dp_reg_list): Remove.
936 (vfp_parse_reg_list): New function.
937 (s_arm_unwind_fnstart, s_arm_unwind_fnend, s_arm_unwind_cantunwind,
938 s_arm_unwind_personality, s_arm_unwind_personalityindex,
939 s_arm_unwind_handlerdata, s_arm_unwind_save, s_arm_unwind_movsp,
940 s_arm_unwind_pad, s_arm_unwind_setfp, s_arm_unwind_raw): New
941 functions.
942 (md_pseudo_table): Add them.
943 (do_vfp_reg2_from_sp2): Use vfp_parse_reg_list and vfp_sp_encode_reg.
944 (do_vfp_sp2_from_reg2, vfp_sp_ldstm, vfp_dp_ldstm): Ditto.
945 (set_section, add_unwind_adjustsp, flush_pending_unwind,
946 finish_unwind_opcodes, start_unwind_section, create_unwind_entry,
947 require_hashconst, add_unwind_opcode): New functions.
948 * doc/tc-arm.text: Document unwinding opcodes.
949 * NEWS: Mention the new feature.
950
951 2004-10-04 Eric Christopher <echristo@redhat.com>
952
953 * config/tc-mips.c (md_apply_fix3): Remove erroneous assert.
954
955 2004-10-01 H.J. Lu <hongjiu.lu@intel.com>
956
957 * config/tc-ppc.c (md_apply_fix3): Call S_SET_THREAD_LOCAL for
958 TLS relocations.
959 * config/tc-s390.c (md_apply_fix3): Likewise.
960 * config/tc-sparc.c (md_apply_fix3): Likewise.
961
962 2004-10-01 Paul Brook <paul@codesourcery.com>
963
964 * config/tc-arm.c (arm_elf_section_type): New function.
965 (arm_elf_change_section): Set section link for exidx sections.
966 * config/tc-arm.h (arm_elf_section_type): Add prototype.
967 (md_elf_section_type): Define.
968
969 2004-10-01 Bill Farmer <Bill@the-farmers.freeserve.co.uk>
970
971 * config/tc-pdp11.c (md_apply_fix3): Change to sign of the SOB
972 instruction's offset.
973
974 2004-10-01 Adam Nemet <anemet@lnxw.com>
975
976 * (TARGET_FORMAT): Remove LynxOS COFF definition.
977
978 2004-10-01 Ravi Ramaseshan <ravi.ramaseshan@codito.com>
979
980 * config/tc-arc.c (tc_gen_reloc): Don't assume fixP->fx_addsy is an
981 asymbol *, instead use symbol_get_bfdsym.
982
983 2004-09-30 Linus Nielsen Feltzing <linus@haxx.se>
984
985 * config/tc-m68k.c (select_control_regs): Add mcf5249.
986
987 2004-09-30 Paul Brook <paul@codesourcery.com>
988
989 * config/tc-arm.c (do_smi, do_nop): New functions.
990 (insns): Add ARMv6ZK instructions.
991 (md_apply_fix3): Handle BFD_RELOC_ARM_SMI.
992 (tc_gen_reloc): Ditto.
993 (arm_cpus): Add mpcore and arm1176.
994 (arm_archs): Add armv6{k,z,zk}.
995 * doc/c-arm.texi: Document new cores and architectures.
996
997 2004-09-30 Nick Clifton <nickc@redhat.com>
998
999 * config/tc-arm.c: Use ISO C90 formatting.
1000
1001 2004-09-30 Vladimir Ivanov <vladitx@nucleusys.com>
1002
1003 * config/tc-arm.c (mav_reg_required_here): Allow REG_TYPE_CN
1004 as alternative when REG_TYPE_MVF, REG_TYPE_MVD, REG_TYPE_MVFX or
1005 REG_TYPE_MVDX is expected.
1006
1007 2004-09-29 Marc Bevand <m.bevand@gmail.com>
1008
1009 * doc/c-i386.texi (i386-Mnemonics): Fix typo.
1010
1011 2004-09-21 James E Wilson <wilson@specifixinc.com>
1012
1013 * config/tc-ia64.c (ENCODED_PSP_OFFSET): New.
1014 (output_rp_psprel, output_pfs_psprel, output_preds_psprel,
1015 output_spill_base, output_unat_psprel, output_lc_psprel,
1016 output_fpsr_psprel, output_priunat_psprel, output_bsp_psprel,
1017 output_bsprestore_psprel, output_rnat_psprel, output_spill_psprel,
1018 output_spill_psprel_p): Use it.
1019
1020 2004-09-20 Tomer Levi <Tomer.Levi@nsc.com>
1021
1022 * config/tc-crx.c (handle_LoadStor): New function.
1023 Handle load/stor unique instructions before parsing.
1024
1025 2004-09-17 Paul Brook <paul@codesourcery.com>
1026
1027 * config/tc-arm.c (s_arm_rel31): New funciton.
1028 (md_pseudo_table): Add .rel31.
1029 (md_apply_fix3): Handle BFD_RELOC_ARM_TARGET2,
1030 BFD_RELOC_32_PCREL and BFD_RELOC_ARM_PREL31.
1031 (tc_gen_reloc): Handle BFD_RELOC_ARM_PREL31 and BFD_RELOC_ARM_TARGET2.
1032 (arm_fix_adjustable): Return 0 for BFD_RELOC_ARM_TARGET2.
1033 (arm_parse_reloc): Add (target2).
1034
1035 2004-09-17 Alan Modra <amodra@bigpond.net.au>
1036
1037 * Makefile.am: Run "make dep-am".
1038 * Makefile.in: Regenerate.
1039 * aclocal.m4: Regenerate.
1040 * configure: Regenerate.
1041 * doc/Makefile.in: Regenerate.
1042 * po/POTFILES.in: Regenerate.
1043 * po/gas.pot: Regenerate.
1044
1045 2004-09-14 Hideki IWAMOTO <h-iwamoto@kit.hi-ho.ne.jp>
1046
1047 * config/tc-mmix.c [!LLONG_MIN]: Correct #elsif to #elif.
1048 [!LLONG_MAX]: Ditto.
1049
1050 2004-09-13 Paul Brook <paul@codesourcery.com>
1051
1052 * config/tc-arm.c: Rename RELABS to TARGET1.
1053
1054 2004-09-13 Alan Modra <amodra@bigpond.net.au>
1055
1056 * messages.c (as_internal_value_out_of_range): Cast values passed
1057 to as_bad_where or as_warn_where to proper type.
1058
1059 2004-09-11 Theodore A. Roth <troth@openavr.org>
1060
1061 * config/tc-avr.c: Add support for
1062 atmega48, atmega88, atmega168, attiny13, attiny2313, at90can128.
1063
1064 2004-09-09 Alan Modra <amodra@bigpond.net.au>
1065
1066 * dw2gencfi.c (select_cie_for_fde): When separating CIE out
1067 from FDE, treat a CFI_escape as we do a DW_CFA_advance_loc.
1068
1069 2004-09-08 Paul Brook <paul@codesourcery.com>
1070
1071 * config/obj-elf.c (obj_elf_section_type): Handle init_array,
1072 fini_array and preinit_array section types.
1073 * config/tc-ia64.c (ia64_elf_section_type): Remove init_array
1074 and fini_array.
1075 * doc/as.texinfo: Document extra section types.
1076
1077 2004-09-02 Mark Mitchell <mark@codesourcery.com>
1078
1079 * Makefile.am (TARG_ENV_HFILES): Add te-symbian.h.
1080 * Makefile.in: Regenerated.
1081 * configure.in: Set em for arm*-*-symbianelf*.
1082 * configure: Regenerated.
1083 * config/tc-arm.c (elf32_arm_target_format): Use Symbian target
1084 vectors when appropriate.
1085 * config/te-symbian.h: New file.
1086
1087 2004-09-03 Tomer Levi <Tomer.Levi@nsc.com>
1088
1089 * config/tc-crx.c (gettrap): Exception vector can be case
1090 insensitive.
1091 (process_label_constant): Fix a 32-bit displacement bug in branch
1092 instructions.
1093 (get_operandtype) : Bug fix, wrong operand was used.
1094 (process_label_constant): Initialize relocation type to
1095 BFD_RELOC_NONE
1096
1097 2004-09-01 Richard Earnshaw < reanrsha@arm.com>
1098
1099 * tc-arm.c (arm_cpus, arm_fpus): Allow <cpu>-s as well as <cpu>s
1100 for synthesizable cores.
1101
1102 * doc/c-arm.texi (ARM Options): Document canonical names of CPUs.
1103
1104 2004-08-25 Dmitry Diky <diwil@spec.ru>
1105
1106 * config/tc-msp430.c: Clean-up the code.
1107 (md_relax_table): New relax table.
1108 (mcu_types): Sort MCU types.
1109 (md_pseudo_table): Add .profiler pseudo handler.
1110 (pow2value): New function.
1111 (msp430_profiler): New function.
1112 (msp430_operands): Add new insns handlers.
1113 (msp430_srcoperand): Add register operand handler, allow complex
1114 expressions.
1115 (md_estimate_size_before_relax): Rewritten.
1116 (md_convert_frag): Rewritten.
1117 (msp430_relax_frag): New function.
1118 * config/tc-msp430.h (md_relax_frag): define macro
1119 * doc/c-msp430.texi: Update information.
1120
1121 2004-08-24 Nick Clifton <nickc@redhat.com>
1122
1123 * as.c (std_shortopts): Allow -g to take an optional argument.
1124 (parse_args): Pass any switch starting with -g on to the backend
1125 for parsing.
1126
1127 2004-08-18 Mark Mitchell <mark@codesourcery.com>
1128
1129 * configure.in (arm*-*-symbianelf*): New target.
1130 (arm*-*-eabi*): Likewise.
1131 * configure: Regenerated.
1132
1133 2004-08-18 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1134 * config/tc-mips.c (append_insn): Handle delay slots in branch likely
1135 correctly.
1136
1137 2004-08-18 Jakub Jelinek <jakub@redhat.com>
1138
1139 * config/tc-ia64.c (start_unwind_section): Add linkonce_empty
1140 argument, don't do anything if current section is not
1141 .gnu.linkonce.t.* and linkonce_empty is set.
1142 (generate_unwind_image, dot_endp): Adjust callers, call
1143 start_unwind_section (*, 1) if nothing will be put into the
1144 section.
1145
1146 2004-08-17 Nick Clifton <nickc@redhat.com>
1147
1148 * as.c (MD_DEBUG_FORMAT_SELECTOR): Provide default definition.
1149 (show_usage): Add -g.
1150 (std_longopts): Add --gen-debug. Alpha sort the table.
1151 (parse_args): Print an error message if a switch is not handled.
1152 Handle the -g switch, calling md_debug_format_selector() if
1153 necessary.
1154 * NEWS: Mention new feature.
1155 * doc/as.texinfo: Document new switch.
1156 * doc/internals.texi: Document behaviour of md_parse_option.
1157
1158 * config/tc-arm.c (md_parse_option): Do not issue an error message
1159 if the switch is not recognised.
1160 * config/tc-m68k.c (md_parse_option): Likewise.
1161 * config/tc-pdp11.c (md_parse_option): Likewise.
1162 * config/tc-v850.c (md_parse_option): Likewise.
1163
1164 * as.h: Fix up formatting.
1165 * tc.h: Likewise.
1166
1167 2004-08-16 Nick Clifton <nickc@redhat.com>
1168
1169 * macro.c (macro_set_alternate): Use ISO C90 formatting.
1170
1171 * configure.in: Sort architecture based tables alphabetically.
1172 * configure: Regenerate.
1173
1174 2004-08-16 Alan Modra <amodra@bigpond.net.au>
1175
1176 * config/tc-ppc.c (tc_ppc_regname_to_dw2regnum <regnames>): Replace
1177 { "cc", 68 }, with { "cr", 70 }.
1178
1179 2004-08-13 Jan Beulich <jbeulich@novell.com>
1180 Nick Clifton <nickc@redhat.com>
1181
1182 * as.c: Add and handle new --alternate command line option.
1183 * macro.c (macro_set_alternate): New.
1184 * macro.h (macro_set_alternate): Declare.
1185 * read.c: Add and handle new .altmacro and .noaltmacro directives.
1186 * doc/as.texinfo: Document new command line option and pseudo-ops
1187 as well as insert documentation originating from gasp about
1188 alternate macro syntax.
1189 * NEWS: Mention new command line option and pseudo-ops.
1190
1191 2004-08-10 Mark Mitchell <mark@codesourcery.com>
1192
1193 * expr.c (operand): Handle the "~", "-", and "!" operators applied
1194 to bignums.
1195
1196 2004-08-06 Paul Brook <paul@codesourcery.com>
1197
1198 * config/tc-arm.c (md_apply_fix3, tc_gen_reloc, arm_parse_reloc):
1199 Handle new relocations.
1200 * include/elf/arm.h (elf_arm_reloc_type): Add new EABI relocations.
1201
1202 2004-08-05 Bob Wilson <bob.wilson@acm.org>
1203
1204 * write.c (relax_segment): Use was_address instead of address when
1205 setting fr_fix field for align frag due to backwards .org.
1206
1207 2004-07-29 Alexandre Oliva <aoliva@redhat.com>
1208
1209 Introduce SH2a support.
1210 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
1211 * config/tc-sh.c (get_specific): Change arch_sh2a_up to
1212 arch_sh2a_nofpu_up.
1213 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
1214 * config/tc-sh.c (md_parse_option): Add sh2a-nofpu ISA handling.
1215 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
1216 * config/tc-sh.c (sh_elf_final_processing): Move sh2a recognition
1217 to end of conditional expression.
1218 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
1219 * config/tc-sh.c: Add sh2a-nofpu support.
1220 2003-12-29 DJ Delorie <dj@redhat.com>
1221 * tc-sh.c: Add sh2a support.
1222 (parse_reg): Add tbr.
1223 (parse_at): Support @@(disp,tbr).
1224 (get_specific): Support sh2a opcodes.
1225 (insert4): New, for 4 byte relocs.
1226 (build_Mytes): Support sh2a opcodes.
1227 (md_apply_fix3_Mytes): Support sh2a opcodes.
1228 2003-12-02 Michael Snyder <msnyder@redhat.com>
1229 * config/tc-sh.c (md_parse_option): Handle sh2a.
1230 (sh_elf_final_processing): Ditto.
1231
1232 2004-07-27 Jason Thorpe <thorpej@wasabisystems.com>
1233
1234 * config/tc-hppa.h (TARGET_FORMAT): Set to "elf32-hppa-netbsd"
1235 for TE_NetBSD.
1236
1237 2004-07-27 Alan Modra <amodra@bigpond.net.au>
1238
1239 * config/tc-ppc.c (ppc_frob_file_before_adjust): Warn if .toc too big.
1240 (ppc_arch): Expand comment.
1241
1242 2004-07-27 Tomer Levi <Tomer.Levi@nsc.com>
1243
1244 * config/tc-crx.c: Support evaluating the difference between two
1245 symbols.
1246 * config/tc-crx.h: Likewise.
1247
1248 2004-07-26 H.J. Lu <hongjiu.lu@intel.com>
1249
1250 * config/tc-ia64.c (start_unwind_section): Set the linked-to
1251 section.
1252 (ia64_elf_section_change_hook): Set the linked-to section for
1253 SHT_IA_64_UNWIND.
1254
1255 2004-07-26 Dmitry Diky <diwil@spec.ru>
1256
1257 * config/tc-msp430.c: Add new subtargets: msp430x1610,
1258 msp430x1611, msp430x1612, msp430x415, msp430x417, msp430xG437,
1259 msp430xG438, msp430xG439.
1260
1261 2004-07-25 Daniel Jacobowitz <dan@debian.org>
1262
1263 * doc/as.texinfo (Section, PushSection): Correct documentation
1264 for ELF.
1265
1266 2004-07-21 Jan Beulich <jbeulich@novell.com>
1267
1268 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
1269 permissible for the selected instruction suffix.
1270 (match_template): Don't permit 64-bit general purpose operands in
1271 32-bit mode.
1272 (finalize_imm): Permit 64-bit immediates.
1273 (build_modrm_byte): Don't treat 32-bit addressing in 64-bit mode
1274 specially except for the width of the used base and/or index
1275 registers. For 32-bit displacements, use sign-extended
1276 relocations only when using 64-bit addressing.
1277 Force zero displacement on rip-relative addressing when there is
1278 no other displacement.
1279 (i386_index_check): Don't treat 32-bit addressing in 64-bit mode
1280 specially except for the width of the used base and/or index
1281 registers.
1282 (parse_register): Disallow Reg64 registers in 32-bit mode.
1283
1284 * config/tc-i386.c: For DefaultSize instructions, don't guess a 'q'
1285 suffix if the instruction doesn't support it.
1286
1287 2004-07-20 Maciej W. Rozycki <macro@linux-mips.org>
1288
1289 * config/tc-mips.c (append_insn): Handle constant expressions with
1290 no associated relocation.
1291 (mips_ip): Cancel the expression after use for the Q format
1292 specifier.
1293 (parse_relocation): Return no relocation for unsupported
1294 operators.
1295 (my_getSmallExpression): Return no relocation if no relocation
1296 operators are used.
1297
1298 2004-07-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1299
1300 * config/obj-som.c (adjust_stab_sections): Add prototype.
1301 (obj_som_compiler, obj_som_version, obj_som_copyright,
1302 adjust_stab_sections): Add ATTRIBUTE_UNUSED to unused arguments.
1303 * config/tc-hppa.c (update_subspace): Likewise.
1304 (is_defined_subspace): Amplify comment.
1305 * config/obj-som.h (som_frob_file): Add prototype.
1306
1307 2004-07-19 Christopher Faylor <cgf@timesys.com>
1308 H.J. Lu <hongjiu.lu@intel.com>
1309
1310 * subsegs.c (section_symbol): Don't create a new segment when
1311 existing segment is undefined.
1312
1313 2004-07-16 Richard Earnshaw <rearnsha@arm.com>
1314
1315 * config/tc-arm.c: Include include/opcode/arm.h.
1316 (ARM_EXT_*, ARM_ARCH_*, ARM_ANY, ARM_ALL, COPROC_ANY): Delete.
1317 (FPU_FPA_EXT_* FPU_VFP_EXT_*, FPU_ANY, FPU_NONE, FPU_MAVERICK): Delete.
1318 (FPU_ARCH_*): Delete.
1319 * Makefile.am: Update dependencies.
1320 * Makefile.in: Regenerate.
1321
1322 2004-07-15 Nick Clifton <nickc@redhat.com>
1323
1324 * configure.in: Accept armbe as a big-endian arm configuration.
1325 * configure: Regenerate.
1326
1327 2004-07-13 Thomas Nystrom <thn@saeab.se>
1328
1329 * config/tc-i386.c (T_SHIFTOP): New constant.
1330 (intel_e05_1): Handle '&', '|' and T_SHIFTOP.
1331 (intel_el1): Handle '~'.
1332 (intel_get_token): Handle '<>', '&', '|' and '~'.
1333
1334 2004-07-13 Nick Clifton <nickc@redhat.com>
1335
1336 (md_assemble): Remove spurious newline from end of as_bad error
1337 message.
1338 (intel_e05_1): Likewise.
1339 (intel_e11): Likewise.
1340 (intel_match_token): Likewise.
1341
1342 2004-07-11 Andreas Schwab <schwab@suse.de>
1343
1344 * config/tc-m68k.c: Convert to C90. Remove redundant
1345 declarations. Indentation fixup.
1346 [M68KCOFF]: Include "obj-coff.h" instead of declaring
1347 obj_coff_section ourselves.
1348
1349 2004-07-09 James E Wilson <wilson@specifixinc.com>
1350
1351 * config/tc-ia64.c (default_big_endian): New.
1352 (dot_byteorder, md_begin): Use it.
1353 (md_parse_option): Set it.
1354
1355 2004-07-09 Nick Clifton <nickc@redhat.com>
1356
1357 * configure.in: Change sh-sybmian-elf to sh-*-symbianelf.
1358 * configure: Regenerate.
1359 * NEWS: Change sh-sybmian-elf to sh-*-symbianelf.
1360 * config/tc-sh.c (sh_elf_final_processing): Use renamed version of
1361 sh_find_elf_flags if necessary.
1362
1363 2004-07-08 Richard Sandiford <rsandifo@redhat.com>
1364
1365 * config/tc-mips.c (mips_fix_adjustable): If the full addend is
1366 going to be split into more than one in-place addend, return 0
1367 for relocations against mergeable sections. Associate comments
1368 with code.
1369
1370 2004-07-07 Tomer Levi <Tomer.Levi@nsc.com>
1371
1372 * Makefile.am (CPU_TYPES): Add crx.
1373 (TARGET_CPU_CFILES): Add config/tc-crx.c.
1374 (TARGET_CPU_HFILES): Add config/tc-crx.h.
1375 (DEPTC_crx_elf): New target.
1376 (DEPOBJ_crx_elf): Likewise.
1377 (DEP_crx_elf): Likewise.
1378 * Makefile.in: Regenerate.
1379 * configure.in: Add crx* target.
1380 * configure: Regenerate.
1381 * config/tc-crx.c: New file.
1382 * config/tc-crx.h: New file.
1383 * NEWS: Mention new target.
1384
1385 2004-07-06 Nick Clifton <nickc@redhat.com>
1386
1387 * config.in: Undefine TARGET_SYMBIAN by default.
1388 * configure.in:
1389 * configure: Regenerate. Add sh-symbian-elf target. If
1390 selected define TARGET_SYMBIAN.
1391 * config/tc-sh.h (TARGET_FORMAT): Select a Symbian target
1392 format if TARGET_SYMBIAN has been defined.
1393
1394 * output-file.c (output_file_create): Report the target format
1395 chosen when bfd_openw reports that it is invalid.
1396
1397 * config/obj-coff.c (coff_pseudo_table): Only define the weak
1398 pseudo for BFD based assemblers.
1399
1400 2004-07-05 Andrew Stubbs <andrew.stubbs@superh.com>
1401
1402 gas:
1403 * config/tc-sh.c (md_assemble): Change isspace to ISSPACE.
1404 (md_parse_option): Remove redundant -isa testing.
1405 Make bfd_arch variable const.
1406 (md_show_usage): Make bfd_arch variable const.
1407
1408 2004-07-03 James E Wilson <wilson@specifixinc.com>
1409
1410 * config/tc-ia64.c (emit_one_bundle): Check and set insn_addr.
1411 * config/tc-ia64.h (md_frag_check): Define.
1412
1413 2004-07-03 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
1414
1415 * config/obj-coff.c (obj_coff_weak): New .weak syntax for PE weak
1416 externals.
1417 * doc/as.texinfo (Weak): Document PE weak symbols.
1418
1419 2004-07-03 Richard Sandiford <rsandifo@redhat.com>
1420
1421 * config/tc-mips.c (HAVE_IN_PLACE_ADDENDS): New macro.
1422 (reloc_needs_lo_p): Only return true if HAVE_IN_PLACE_ADDENDS.
1423 (mips_frob_file): Rework so that only a single pass through the
1424 relocs is needed. Allow %lo()s to have higher offsets than their
1425 corresponding %hi()s or %got()s.
1426
1427 2004-07-02 Nick Clifton <nickc@redhat.com>
1428
1429 * config/tc-arm.c (md_apply_fix3:BFD_RELOC_ARM_IMMEDIATE): Do not
1430 allow values which have come from undefined symbols.
1431 Always consider this fixup to have been processed as a reloc
1432 cannot be generated for it.
1433
1434 2004-07-02 Alan Modra <amodra@bigpond.net.au>
1435
1436 * frags.h (struct frag): Add has_code and insn_addr fields.
1437 * write.c (cvt_frag_to_fill): Invoke md_frag_check.
1438 * config/tc-ppc.c (md_assemble): Check and set insn_addr.
1439 * config/tc-ppc.h (md_frag_check): Define.
1440
1441 2004-06-28 Maciej W. Rozycki <macro@linux-mips.org>
1442
1443 * doc/Makefile.am (info): Rename goal to...
1444 (info-local): ... this, to preserve implicit dependencies.
1445 * doc/Makefile.in: Regenerate with automake 1.8.5.
1446
1447 2004-06-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1448
1449 * config/tc-m32r.c (md_convert_frag): Changed for @PLT.
1450 (m32r_cgen_record_fixup_exp): Changed for @GOTOFF, @GOT.
1451 (m32r_fix_adjustable): Changed for @GOTOFF, @GOT, @PLT.
1452 (tc_gen_reloc): Likewise.
1453 (m32r_end_of_match): Add for @GOTOFF, @GOT, @PLT.
1454 (m32r_parse_name): Likewise.
1455 (m32r_cgen_parse_fix_exp): Likewise.
1456 * config/tc-m32r.h (md_parse_name): Define for @GOTOFF, @GOT, @PLT.
1457 (O_PIC_reloc): Likewise.
1458 (TC_CGEN_PARSE_FIX_EXP): Likewise..
1459 * cgen.c (gas_cgen_parse_operand): Add TC_CGEN_PARSE_FIX_EXP
1460 for @GOTOFF, @GOT, @PLT.
1461
1462 2004-06-21 Jan Beulich <jbeulich@novell.com>
1463
1464 * gas/symbols.c: While discarding ordinary local absolute symbols
1465 when --strip-local-absolute is in effect, retain file symbols.
1466
1467 2004-06-20 Andreas Schwab <schwab@suse.de>
1468
1469 * config/tc-m68k.c (mri_chip): Replace current_chip, not augment.
1470 (md_parse_option): Likewise.
1471
1472 2004-06-17 Jan Beulich <jbeulich@novell.com>
1473
1474 * config/tc-i386.c: Deal with LEX_QM the same way as with LEX_AT.
1475 * config/te-netware.h: New file.
1476 * config/te-ppcnw.h: Delete: Obsolete.
1477 * configure.in: Eliminate ill NetWare targets. Make generic
1478 NetWare target use proper emulation.
1479 * Makefile.am: Eliminate reference to obsolete te-ppcnw.h, add
1480 reference to new te-netware.h.
1481 * configure: Regenerate.
1482 * Makefile.in: Regenerate.
1483
1484 2004-06-15 Martin Schwidefsky <schwidefsky@de.ibm.com>
1485
1486 * config/tc-s390.c (s390_insn): Avoid incorrect signed/unsigned
1487 comparison in .insn pseudo operation.
1488
1489 2004-06-15 Alan Modra <amodra@bigpond.net.au>
1490
1491 * config/obj-coff.c (coff_adjust_section_syms): Use
1492 bfd_get_section_size instead of bfd_get_section_size_before_reloc.
1493 (coff_frob_section): Likewise.
1494 * config/tc-mips.c (md_apply_fix3): Likewise.
1495 * config/obj-elf.c (elf_frob_file): Use bfd_set_section_size.
1496 (elf_frob_file_after_relocs): Likewise.
1497
1498 2004-06-10 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1499
1500 * config/tc-hppa.c (log2): Rename to exact_log2.
1501 (pa_next_subseg): Delete unused function.
1502 (create_new_space): Mark unused arguments with ATTRIBUTE_UNUSED.
1503 (create_new_subspace): Likewise.
1504
1505 Bug gas/213
1506 * config/tc-hppa.c (hppa_fix_adjustable): Allow reduction of fake
1507 labels. Fix warning.
1508
1509 2004-05-28 DJ Delorie <dj@redhat.com>
1510
1511 * config/tc-mn10300.h (tc_fix_adjustable): Define.
1512 * config/tc-mn10300.c (mn10300_fix_adjustable): Don't adjust debug
1513 or non-merged symbols.
1514
1515 2004-05-28 H.J. Lu <hongjiu.lu@intel.com>
1516
1517 * config/tc-ia64.c (remove_marked_resource): Save, clear and
1518 restore the old slot when inserting srlz.i/srlz.d.
1519
1520 2004-05-28 Andrew Stubbs <andrew.stubbs@superh.com>
1521
1522 * Makefile.am: Regenerate dependecies.
1523 * Makefile.in: Regenerate.
1524 * config/tc-sh.c (valid_arch): Make unsigned.
1525 (preset_target_arch): Likewise.
1526 (md_begin): Use new architecture flags system.
1527 (get_specific): Likewise.
1528 (assemble_ppi): Likewise.
1529 (md_assemble): Likewise. Also fix error check for bad opcodes.
1530 (md_parse_option): Likewise. Also generate -isa values according
1531 to the table in bfd/cpu-sh.c instead of just constants. Also
1532 allow <arch>-up ISA variants.
1533 (sh_elf_final_processing): Replace if-else chain with a call to
1534 sh_find_elf_flags().
1535 * testsuite/gas/sh/arch: New directory.
1536 * testsuite/gas/sh/arch/arch.exp: New test script.
1537 * testsuite/gas/sh/arch/arch_expected.txt: New file.
1538 * testsuite/gas/sh/arch/sh.s: New file.
1539 * testsuite/gas/sh/arch/sh2.s: New file.
1540 * testsuite/gas/sh/arch/sh-dsp.s: New file.
1541 * testsuite/gas/sh/arch/sh2e.s: New file.
1542 * testsuite/gas/sh/arch/sh3-nommu.s: New file.
1543 * testsuite/gas/sh/arch/sh3.s: New file.
1544 * testsuite/gas/sh/arch/sh3-dsp.s: New file.
1545 * testsuite/gas/sh/arch/sh3e.s: New file.
1546 * testsuite/gas/sh/arch/sh4-nommu-nofpu.s: New file.
1547 * testsuite/gas/sh/arch/sh4-nofpu.s: New file.
1548 * testsuite/gas/sh/arch/sh4.s: New file.
1549 * testsuite/gas/sh/arch/sh4a-nofpu.s: New file.
1550 * testsuite/gas/sh/arch/sh4al-dsp.s: New file.
1551 * testsuite/gas/sh/arch/sh4a.s: New file.
1552
1553 2004-05-28 Peter Barada <peter@the-baradas.com>
1554
1555 * config/gc-m68k.c(m68k_ip): Convert mode 5 addressing
1556 with zero offset into mode 2 addressing to save a word.
1557
1558 2004-05-27 H.J. Lu <hongjiu.lu@intel.com>
1559
1560 * config/tc-ia64.c (ar_is_in_integer_unit): Removed.
1561 (ar_is_only_in_integer_unit): New.
1562 (ar_is_only_in_memory_unit): New.
1563 (generate_unwind_image): Silence gcc on 32bit host.
1564 (md_assemble): Use ar_is_only_in_integer_unit instead of
1565 ar_is_in_integer_unit. Check AR access.
1566
1567 2004-05-27 Peter Barada <peter@the-baradas.com>
1568
1569 * config/tc-m68k.c (md_begin): Sort the opcode table into
1570 alphabetical order.
1571 (m68k_compare_opcode): New function to do the sorting.
1572
1573 2004-05-24 Peter Barada <peter@the-baradas.com>
1574
1575 * config/m68k-parse.y(operand): Allow for MAC/EMAC mask
1576 addressing on MIT style operands.
1577 * config/m68k-parse.y(yylex): Allow '-&' for predecrement
1578 w/mask addressing.
1579 * config/tc-m68k.c(install_operand): Comment 'G' and 'H' type
1580 operands.
1581
1582 2004-05-23 Alan Modra <amodra@bigpond.net.au>
1583
1584 * expr.c (operand, operator): Don't reject '++' and '--'.
1585
1586 2004-05-20 Richard Sandiford <rsandifo@redhat.com>
1587
1588 * config/tc-mips.c (append_insn): Use ISA-encoded addresses in MIPS16
1589 dwarf tables.
1590
1591 2004-05-17 Adam Nemet <anemet@lnxw.com>
1592
1593 * configure.in: Add ppc-*-lynxos*. Update i386-*-lynxos* to ELF.
1594 * configure: Regenerate.
1595
1596 2004-05-13 Paul Brook <paul@codesourcery.com>
1597
1598 * dw2gencfi.c (output_cie): Handle dwarf3 format CIE entries.
1599
1600 2004-05-13 Joel Sherrill <joel@oarcorp.com>
1601
1602 * configure.in (or32-*-rtems*): Switch to elf.
1603 * configure: Regenerate.
1604
1605 2004-05-13 Nick Clifton <nickc@redhat.com>
1606
1607 * po/fr.po: Updated French translation.
1608
1609 2004-05-11 Nick Clifton <nickc@redhat.com>
1610
1611 * doc/as.texinfo (Section): Document G and T flags to .section
1612 directive. Document the extra arguments that the G flag
1613 requires. Document the #tls flag.
1614
1615 2004-05-11 H.J. Lu <hongjiu.lu@intel.com>
1616
1617 * subsegs.c (section_symbol): Create a new section symbol if
1618 the existing one doesn't match.
1619 * symbols.c (symbol_set_bfdsym): Don't reset BFD section symbol.
1620
1621 2004-05-07 Richard Sandiford <rsandifo@redhat.com>
1622
1623 * config/tc-mips.c (append_insn, mips_emit_delays): Extend -mfix-vr4120
1624 to cope with VR4181A errata MD(1) and MD(4).
1625
1626 2004-05-07 Brian Ford <ford@vss.fsi.com>
1627
1628 * NEWS: Mention .secrel32 for pe[i]-i386.
1629
1630 2004-05-07 Alexandre Oliva <aoliva@redhat.com>
1631
1632 * config/tc-frv.h (MAX_MEM_FOR_RS_ALIGN_CODE): New.
1633 (HANDLE_ALIGN): New.
1634
1635 2004-05-06 Daniel Jacobowitz <dan@debian.org>
1636
1637 * Makefile.am (DIST_SUBDIRS): Define.
1638 * aclocal.m4: Regenerate with automake 1.8.4.
1639 * Makefile.in: Likewise.
1640 * doc/Makefile.in: Likewise.
1641
1642 2004-05-06 David Mosberger-Tang <davidm@hpl.hp.com>
1643
1644 * config/tc-ia64.c (dot_serialize): Declare.
1645 (dot_serialize): New function.
1646 (md_pseudo_table): Add ".serialize.data" and
1647 ".serialize.instruction" directives.
1648
1649 2004-05-06 Nick Clifton <nickc@redhat.com>
1650
1651 * messages (as_internal_value_out_of_range): Print a message about
1652 a value being out of range. Be consistent about whether the
1653 values are printed in decimal or hexadecimal.
1654 (as_warn_value_out_of_range): Generate a warning message about an
1655 out of range value.
1656 (as_bad_value_out_of_range): Generate an error message about an
1657 out of range value.
1658 * as.h: Prototype the new functions.
1659 * config/tc-alpha.c (insert_operand): Use new function.
1660 * config/tc-arc.c (arc_insert_operand): Likewise.
1661 * config/tc-mn10200.c (mn10200_insert_operand): Likewise.
1662 * config/tc-mn10300.c (mn10300_insert_operand): Likewise.
1663 * config/tc-ppc.c (ppc_insert_operand): Likewise.
1664 * config/tc-s390.c (s390_insert_operand): Likewise.
1665 * config/tc-v850.c (v850_insert_operand): Likewise.
1666
1667 2004-05-05 Alexandre Oliva <aoliva@redhat.com>
1668
1669 * configure.in: Set em=linux for frv-*-*linux*.
1670 * configure: Rebuilt.
1671 * config/tc-frv.h (TARGET_FORMAT): Use elf32-frvfdpic if...
1672 (frv_md_fdpic_enabled): New.
1673 * config/tc-frv.c (frv_md_fdpic_enabled): New.
1674 (DEFAULT_FDPIC): New.
1675 (frv_flags): Use DEFAULT_FDPIC.
1676 (frv_pic_flag): Likewise.
1677 (OPTION_NOPIC): New.
1678 (md_longopts): Add -mnopic.
1679 (md_parse_option): Handle it.
1680 (md_show_usage): Add -mfdpic and -mnopic.
1681
1682 2004-05-05 Peter Barada <peter@the-baradas.com>
1683
1684 * config/tc-m68k.c: Add find_cf_chip to print list of valid
1685 chips for invalid coldfire instructions, rename selectors
1686 for ColdFire sub-variants, add 521x,5249,547x,548x and aliases,
1687 add current_chip to track which chip is referred to(including save/restore),
1688 use current_chip to select control registers, not current_arch.
1689 (md_show_usage): Add new chips.
1690 * doc/c-m68k.texi: Document new command line switches.
1691
1692 2004-05-05 Jakub Jelinek <jakub@redhat.com>
1693
1694 * tc-s390.h (md_do_align, HANDLE_ALIGN): Remove.
1695 (NOP_OPCODE): Define.
1696 (s390_align_code): Remove prototype.
1697 * tc-s390.c (s390_align_code): Remove.
1698
1699 2004-05-04 H.J. Lu <hongjiu.lu@intel.com>
1700
1701 * config/tc-ia64.c (make_unw_section_name): Removed.
1702 (start_unwind_section): New function.
1703 (generate_unwind_image): Take const segT instead of const
1704 char *.
1705 (dot_handlerdata): Adjusted.
1706 (dot_endp): Likewise.
1707
1708 2004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1709
1710 * config/obj-elf.c (obj_elf_change_section): Allow the
1711 ".note.GNU-stack" section has SHF_EXECINSTR.
1712
1713 2004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1714
1715 * config/obj-elf.c (get_section): Return bfd_boolean.
1716 (obj_elf_change_section): Call bfd_get_section_by_name_if
1717 instead of bfd_map_over_sections.
1718
1719 2004-04-30 H.J. Lu <hongjiu.lu@intel.com>
1720
1721 * config/obj-elf.c (get_section): New function.
1722 (obj_elf_change_section): Support multiple sections with same
1723 name.
1724
1725 2004-04-30 Nick Clifton <nickc@redhat.com>
1726
1727 * config/tc-arm.c (create_register_alias): Fix typo checking for
1728 case sensitive register aliases.
1729 (co_proc_number): Use error message string in all_reg_maps[]
1730 array.
1731 (cp_reg_required_here): Likewise.
1732 (fp_reg_required_here): Likewise.
1733
1734 2004-04-29 Brian Ford <ford@vss.fsi.com>
1735
1736 * dwarf2dbg.c (dwarf2_finish): Add SEC_DEBUGGING to section flags.
1737
1738 2004-04-28 Chris Demetriou <cgd@broadcom.com>
1739
1740 * config/tc-mips.c (HAVE_32BIT_ADDRESSES, append_insn, macro_build)
1741 (load_address, macro, mips_ip, md_parse_option)
1742 (mips_force_relocation, mips_validate_fix, md_apply_fix3)
1743 (s_change_sec, pic_need_relax, tc_gen_reloc): Remove all
1744 embedded-PIC handling, and update comments.
1745 (SWITCH_TABLE): Remove.
1746 * config/tc-mips.h (DIFF_EXPR_OK): Delete.
1747 (enum mips_pic_level): Remove EMBEDDED_PIC.
1748 (EXTERN_FORCE_RELOC): Remove embedded-PIC handling.
1749 (TC_FORCE_RELOCATION): Update comment.
1750 * ecoff.c (ecoff_build_lineno): Add comment about some code that
1751 might be safe to remove now that MIPS embedded-PIC is gone.
1752
1753 2004-04-28 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1754
1755 * config/obj-som.c (obj_som_init_stab_section): Add new arguments in
1756 call to obj_set_subsection_attributes.
1757 (obj_som_init_stab_section): Likewise.
1758 * config/tc-hppa.c (default_subspace_dict): Add comdat field.
1759 (pa_def_subspaces): Provide comdat default.
1760 (pa_subspace): Handle new "comdat" parameter. Set SEC_LINK_ONCE and
1761 not SEC_IS_COMMON if section is comdat, common or dup_common. Update
1762 calls to create_new_subspace and update_subspace to pass comdat flag.
1763 (create_new_subspace, update_subspace): Add new comdat argument. Use
1764 it in calls to obj_set_subsection_attributes.
1765 * doc/c-hppa.texi (.subspa, .nsubspa): Document new comdat parameter
1766 and use of comdat, common and dup_comm parameters.
1767
1768 2004-04-26 H.J. Lu <hongjiu.lu@intel.com>
1769
1770 * config/obj-elf.c (obj_elf_change_section): Check if the old
1771 group name is NULL before comparison.
1772
1773 2004-04-23 Chris Demetriou <cgd@broadcom.com>
1774
1775 * config/tc-mips.h (mips_dwarf2_addr_size): Prototype.
1776
1777 2004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1778
1779 * config/tc-mips.c (s_mipsset): Set default CPU type for .set mipsN.
1780
1781 2004-04-23 Chris Demetriou <cgd@broadcom.com>
1782
1783 * config/tc-mips.c (md_longopts): Remove -membedded-pic option.
1784 (OPTION_MEMBEDDED_PIC): Remove.
1785 (OPTION_TRAP, OPTION_BREAK, OPTION_EB, OPTION_EL)
1786 (OPTION_FP32, OPTION_GP32, OPTION_CONSTRUCT_FLOATS)
1787 (OPTION_NO_CONSTRUCT_FLOATS, OPTIONS_FP64, OPTION_GP64)
1788 (OPTION_RELAX_BRANCH, OPTION_NO_RELAX_BRANCH)
1789 (OPTION_ELF_BASE): Renumber.
1790 (md_parse_option): Remove OPTION_MEMBEDDED_PIC handling.
1791 (md_show_usage): Remove mention of -membedded-pic.
1792 * doc/as.texinfo: Remove mention of -membedded-pic.
1793
1794 2004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1795
1796 * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Remove.
1797 * config/tc-mips.c (RDATA_SECTION_NAME, mips_target_format): Remove
1798 a.out support.
1799 (md_begin, mips_ip, md_parse_option, s_change_sec, s_option,
1800 s_abicalls, nopic_need_relax, tc_gen_reloc): Remove uses of
1801 USE_GLOBAL_POINTER_OPT.
1802
1803 2004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1804
1805 * config/tc-mips.c (macro): One more use of load_delay_nop.
1806
1807 2004-04-22 Atsushi Nemoto <anemo@mba.ocn.ne.jp>
1808
1809 * config/tc-mips.c (load_delay_nop): New function.
1810 (load_address, macro): Use load_delay_nop() to build a nop
1811 which can be omitted with gpr_interlocks.
1812
1813 2004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1814
1815 * config/tc-mips.c (hilo_interlocks, gpr_interlocks,
1816 cop_interlocks): Remove superfluous CPU entries.
1817
1818 2004-04-22 Paul Brook <paul@codesourcery.com>
1819
1820 * config/tc-arm.c (mav_parse_offset): Value must be multiple of 4.
1821
1822 2004-04-22 Peter Barada <peter@the-baradas.com>
1823
1824 * NEWS: Added support for EMAC instructions and MAC/EMAC
1825 Motorola syntax.
1826 * config/m68k-parse.h: Add ACC[123], ACCEXT{01,23}, MAC/EMAC
1827 scale factor tokens, trailing_ampersand to mark mask addressing
1828 for MAC/EMAC instructions.
1829 * config/m68k-parse.y: Add options_ampersand clause, '<<',
1830 '>>'.
1831 (yylex): Handle '>', '<', and '&' following '+'.
1832 * config/tc-m68k.c: Set mcfmac/mcfemac on appropriate ColdFire
1833 architectures in archs[].
1834 (m68k-ip): Add '4', 'e', 'g', 'i', cases to handle mask addressing
1835 for MAC/EMAC instructions, ACC[0123], ACCEXT{01,23}, and '<<'/'>>'
1836 respectively.
1837 (m68k_ip): Handle trailing '&' on MAC/EMAC insns.
1838 (install_operand): Fix 'n' case, Add 'F', 'f', 'G', 'H', 'I', ']'
1839 cases.
1840 Add EMAC operands to init_table[].
1841
1842 2004-04-22 Bruno De Bus <bdebus@elis.ugent.be>
1843
1844 * config/tc-arm.h (enum mstate): Move here, add MAP_UNDEFINED
1845 state.
1846 (TC_SEGMENT_INFO_TYPE): Define to enum mstate.
1847 * config/tc-arm.c (enum mstate): Delete from here.
1848 (mapping_state): Remove the static mapstate variable and instead
1849 store the state in the segment. This allows a per-section mapping
1850 state. Handle and ignore MAP_UNDEFINED states.
1851 (arm_elf_change_section): Get the current mapping state from the
1852 new section.
1853 (s_ltorg): Set the mapping state to MAP_DATA.
1854 (arm_cleanup): Use arm_elf_change_section to get the mapping state
1855 for each pool as it is emitted.
1856
1857 2004-04-22 Nick Clifton <nickc@redhat.com>
1858
1859 * config/tc-arm.h: Formatting tidy ups.
1860
1861 2004-04-20 Chris Demetriou <cgd@broadcom.com>
1862
1863 * NEWS: Note that MIPS -membedded-pic option is deprecated.
1864
1865 2004-04-20 DJ Delorie <dj@redhat.com>
1866
1867 * config/tc-i386.h [TE_PE] (TC_CONS_FIX_NEW): Define.
1868 * config/tc-i386.c (md_pseudo_table) [TE_PE]: Add "secrel32".
1869 [TE_PE] (O_secrel): Define.
1870 [TE_PE] (x86_pe_cons_fix_new): New.
1871 [TE_PE] (pe_directive_secrel): Likewise.
1872 (tc_gen_reloc) [TE_PE]: Support BFD_RELOC_32_SECREL.
1873
1874 2004-04-19 Eric Christopher <echristo@redhat.com>
1875
1876 * config/tc-mips.c (mips_dwarf2_addr_size): Revert part
1877 of previous patch for fix in gcc.
1878
1879 2004-04-19 Jakub Jelinek <jakub@redhat.com>
1880
1881 * config/tc-xtensa.c (xg_assembler_literal): Fix a typo.
1882
1883 2004-04-19 Nathan Sidwell <nathan@codesourcery.com>
1884
1885 * read.c (do_align): Call md_flush_pending_output, if defined.
1886
1887 2004-04-16 Alan Modra <amodra@bigpond.net.au>
1888
1889 * expr.c (operand): Correct checks for ++ and --.
1890
1891 2004-04-14 H.J. Lu <hongjiu.lu@intel.com>
1892
1893 * config/tc-generic.c: Add some comments.
1894
1895 2004-04-14 Richard Sandiford <rsandifo@redhat.com>
1896
1897 * doc/c-mips.texi (-m{no-,}fix-vr4120): Renamed from
1898 -{no-}mfix-vr4122-bugs.
1899 * config/tc-mips.c (mips_fix_vr4120): Renamed from mips_fix_4122_bugs.
1900 (append_insn, mips_emit_delays): Update accordingly.
1901 (OPTION_FIX_VR4120, OPTION_NO_FIX_VR4120): Renamed from *VR4122.
1902 (md_longopts): Change -{no-,}mfix-vr4122-bugs to -m{no-,}fix-vr4120.
1903 (md_parse_option): Update after above changes.
1904 (md_show_usage): Add -mfix-vr4120.
1905
1906 2004-04-13 Bob Wilson <bob.wilson@acm.org>
1907
1908 * doc/as.texinfo (Sub-Sections): Conditionalize COFF-specific use
1909 of .section directive; add a reference to the ELF .subsection
1910 directive.
1911
1912 2004-04-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1913
1914 * config/tc-m32r.c (md_assemble): Fixed infinite loop bug
1915 in parallel.
1916
1917 2004-04-11 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1918
1919 * Makefile.am: Remove mips from aout targets.
1920 * Makefile.in: Regenerate.
1921 * configure.in: Remove mips-dec-bsd* target.
1922 * configure: Regenerate.
1923
1924 2004-04-07 Alan Modra <amodra@bigpond.net.au>
1925
1926 PR 96
1927 * config/tc-ppc.c (ppc_elf_suffix): Add valid32 and valid64 fields
1928 to struct map_bfd. Adjust MAP macro, and define MAP32, MAP64.
1929 Update "mapping". Restrict some @ modifiers to 32 bit.
1930
1931 2004-04-01 Asgari Jinia <asgarij@kpitcummins.com>
1932 Dhananjay Deshpande <dhananjayd@kpitcummins.com>
1933
1934 * config/tc-sh.c (dont_adjust_reloc_32): New variable.
1935 (sh_fix_adjustable): Avoid adjusting BFD_RELOC_32 when
1936 dont_adjust_reloc_32 is set.
1937 (md_longopts): Add option -renesas.
1938 (md_parse_option, md_show_usage): Likewise.
1939 * doc/c-sh.texi: Likewise.
1940
1941 2004-04-01 Dave Korn <dk@artimi.com>
1942
1943 * config/tc-dlx.c (md_assemble): set fx_no_overflow flag for
1944 hi16 and lo16 fixS structs.
1945 (md_assemble): generate bit_fixS for RELOC_DLX_LO16 in
1946 exactly the same way as for RELOC_DLX_REL16.
1947 (machine_ip): properly respect LO flag in the_insn and
1948 output RELOC_DLX_LO16 rather than RELOC_DLX_16.
1949 (md_apply_fix3): apply RELOC_DLX_LO16.
1950
1951 2004-03-30 Stan Shebs <shebs@apple.com>
1952
1953 Remove long-obsolete MPW support.
1954 * mpw-config.in, mpw-make.sed, mac-as.r: Remove files.
1955 * configure.in: Remove mention of ppc-*-mpw* config.
1956 * configure.in: Likewise.
1957
1958 2004-03-30 Nick Clifton <nickc@redhat.com>
1959
1960 * config/tc-arm.c (meabi_flags): Make its use conditional upon
1961 OBJ_ELF being defined.
1962
1963 2004-03-27 Alan Modra <amodra@bigpond.net.au>
1964
1965 * config/obj-aout.c (obj_aout_type): Remove #ifdef BFD_ASSEMBLER code.
1966
1967 2004-03-23 Paul Brook <paul@codesourcery.com>
1968
1969 * config/tc-arm.c (meabi_flags): New variable.
1970 (arm_parse_eabi): New function.
1971 (md_begin): Set flags for EABI v3.
1972 (arm_eabis): Add.
1973 (arm_long_opts): Add meabi.
1974 * doc/as.texinf <ARM>: Document -meabi.
1975 * doc/c-arm.texi: Ditto.
1976
1977 2004-03-22 Bob Wilson <bob.wilson@acm.org>
1978
1979 * config/tc-xtensa.c (xtensa_post_relax_hook): Create literal
1980 tables even when use_literal_section flag is not set.
1981
1982 2004-03-22 Alan Modra <amodra@bigpond.net.au>
1983
1984 * config/tc-sh.c: Remove trailing whitespace.
1985
1986 2004-03-22 Hans-Peter Nilsson <hp@axis.com>
1987
1988 * doc/c-cris.texi (CRIS-Opts): Document --no-mul-bug-abort,
1989 --mul-bug-abort and the default behavior.
1990 * config/tc-cris.c (cris_insn_kind): New member CRIS_INSN_MUL.
1991 (err_for_dangerous_mul_placement): New variable.
1992 (STATE_MUL, OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): New
1993 macros.
1994 (md_cris_relax_table): Have placeholder for STATE_MUL.
1995 (md_longopts): New options --mul-bug-abort and --no-mul-bug-abort.
1996 (cris_relax_frag) <case ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: New
1997 case doing nothing.
1998 (md_estimate_size_before_relax) <case ENCODE_RELAX (STATE_MUL,
1999 STATE_BYTE)>: Ditto.
2000 (md_convert_frag) <ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: Check
2001 alignment and position of this frag, emit error message if
2002 suspicious.
2003 (md_assemble): For a multiply insn and when checking it,
2004 transform the current frag into a special frag for that purpose.
2005 (md_parse_option) <case OPTION_MULBUG_ABORT_OFF, case
2006 OPTION_MULBUG_ABORT_ON>: Handle new options.
2007
2008 2004-03-19 Bob Wilson <bob.wilson@acm.org>
2009
2010 * config/tc-xtensa.c (mark_literal_frags): New function.
2011 (xtensa_move_literals): Call mark_literal_frags for all literal
2012 segments, including init and fini literal segments.
2013 (xtensa_post_relax_hook): Swap use of xt_insn_sec and xt_literal_sec.
2014
2015 2004-03-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2016
2017 * tc-hppa.c (cons_fix_new_hppa): Check for PC relative base type.
2018 (pa_comm): Set BSF_OBJECT in symbol flags.
2019
2020 2004-03-19 Alan Modra <amodra@bigpond.net.au>
2021
2022 * Makefile.am: Run "make dep-am".
2023 * Makefile.in: Regenerate.
2024 * doc/Makefile.in: Regenerate.
2025 * config.in: Regenerate.
2026 * po/gas.pot: Regenerate.
2027
2028 2004-03-18 Nathan Sidwell <nathan@codesourcery.com>
2029
2030 * read.c (read_a_source_file): Use demand_empty_rest_of_line.
2031 (demand_empty_rest_of_line): Issue an error here.
2032 (ignore_rest_of_line): Silently skip to end.
2033 (demand_copy_string): Issue an error, not warning.
2034 (equals): Likewise.
2035 * config/obj-elf.c (obj_elf_section_name): Likewise.
2036 (obj_elf_section): Likewise.
2037 * config/tc-arc.c (arc_extoper): Remove bogus NULL checks.
2038 (arc_extinst): Likewise.
2039 * config/tc-ia64.c (dot_saveb): Use demand_empty_rest_of_line.
2040 (dot_spill): Likewise.
2041 (dot_unwabi): Likewise.
2042 (dot_prologue): Likewise.
2043
2044 * expr.c (operand): Reject ++ and --.
2045 (operator): Likewise.
2046
2047 2004-03-17 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2048
2049 * config/tc-sh.c: Include dw2gencfi.h.
2050 (sh_cfi_frame_initial_instructions): New function.
2051 (sh_regname_to_dw2regnum): Likewise.
2052 * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Move to the end of
2053 file.
2054 (TARGET_USE_CFIPOP): Define.
2055 (tc_cfi_frame_initial_instructions): Likewise.
2056 (tc_regname_to_dw2regnum): Likewise.
2057 (DWARF2_DEFAULT_RETURN_COLUMN, DWARF2_CIE_DATA_ALIGNMENT): Likewise.
2058 * Makefile.am: Update dependencies.
2059 * Makefile.in: Regenerate.
2060
2061 2004-03-17 Ralf Corsepius <corsepiu@faw.uni-ulm.de>
2062
2063 * configure.in: Switch sh-*-rtems* to ELF. Add sh-*-rtemscoff*.
2064 * configure: Regenerate.
2065
2066 2004-03-12 Bob Wilson <bob.wilson@acm.org>
2067
2068 * read.c (s_leb128): Call md_flush_pending_output.
2069
2070 2004-03-12 Michal Ludvig <mludvig@suse.cz>
2071
2072 * config/tc-i386.c (output_insn): Handle PadLock instructions.
2073 * config/tc-i386.h (CpuPadLock): New define.
2074 (CpuUnknownFlags): Added CpuPadLock.
2075
2076 2004-03-07 Andreas Schwab <schwab@suse.de>
2077
2078 * doc/c-hppa.texi (HPPA Directives): Fix typo.
2079
2080 2004-03-07 Richard Henderson <rth@redhat.com>
2081
2082 * dw2gencfi.c (output_cie): Align length to 4 byte boundary.
2083 (cfi_finish): Likewise for fde.
2084
2085 2004-03-05 H.J. Lu <hongjiu.lu@intel.com>
2086
2087 * config/tc-ia64.c (md_assemble): Properly handle NULL
2088 align_frag.
2089 (ia64_handle_align): Don't abort if failed to add a stop bit.
2090
2091 2004-03-04 H.J. Lu <hongjiu.lu@intel.com>
2092
2093 * Makefile.in: Regenerated.
2094 * aclocal.m4: Likewise.
2095 * configure: Likewise.
2096 * doc/Makefile.in: Likewise.
2097
2098 2004-03-03 H.J. Lu <hongjiu.lu@intel.com>
2099
2100 * config/tc-ia64.c (dot_align): New.
2101 (ia64_do_align): Make it static.
2102 (md_pseudo_table): Use "dot_align" for "align".
2103 (ia64_md_do_align): Don't set align_frag here.
2104 (ia64_handle_align): Add a stop bit to the previous bundle if
2105 needed.
2106
2107 * config/tc-ia64.h (ia64_do_align): Removed.
2108
2109 2003-03-03 Andrew Stubbs <andrew.stubbs@superh.com>
2110
2111 * config/tc-sh.c (md_parse_option): Add -isa=sh4-nofpu and
2112 -isa=sh4-nommu-nofpu options. Adjust help messages accordingly.
2113 (sh_elf_final_processing): Output BFD type sh4_nofpu if that is
2114 the most general type or the user specifically requested it.
2115 (md_assemble): Add a new error message for when an instruction
2116 is understood, but is not allowed due to an -isa option.
2117
2118 2004-03-02 H.J. Lu <hongjiu.lu@intel.com>
2119
2120 * config/tc-ia64.c (align_frag): New.
2121 (md_assemble): Set the tc_frag_data field in align_frag for
2122 IA64_OPCODE_FIRST instructions.
2123 (ia64_md_do_align): Set align_frag.
2124 (ia64_handle_align): Add a stop bit if needed.
2125
2126 * config/tc-ia64.h (TC_FRAG_TYPE): New.
2127 (TC_FRAG_INIT): New.
2128
2129 2004-03-01 Richard Sandiford <rsandifo@redhat.com>
2130
2131 * config/tc-frv.c (fr400_audio): New variable.
2132 (md_parse_option, md_show_usage): Add -mcpu=fr405 and -mcpu=fr450.
2133 (md_parse_option): Set fr400_audio for -mcpu=fr400 and -mcpu=fr405.
2134 (target_implements_insn_p): New function.
2135 (md_assemble): Report an error if the processor doesn't implement
2136 the instruction.
2137
2138 2004-02-27 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2139
2140 * config/tc-m32r.c (md_longopts): Added -no-bitinst option.
2141 (md_parse_option): Ditto.
2142 (OPTION_NO_SPECIAL_M32R): Added.
2143 (md_show_usage): Document it.
2144 (enable_speial_m32r): Changed a default value from 0 to 1.
2145 * doc/c-m32r.texi: Document the -no-bitinst option.
2146
2147 2004-02-27 Nick Clifton <nickc@redhat.com>
2148
2149 * config/tc-sh.c (get_operand): Revert previous delta.
2150 (tc_gen_reloc): Check for an unknown reloc type before processing
2151 the addend.
2152
2153 2004-02-27 Hannes Reinecke <hare@suse.de>
2154
2155 * config/tc-s390.c (s390_insn): Correct range check for opcode in
2156 .insn pseudo operation.
2157
2158 2004-02-27 Anil Paranjpe <anilp1@kpitcummins.com>
2159
2160 * config/tc-sh.c (get_operand): In case of #Imm, check has been
2161 added for wrong syntax.
2162
2163 2004-02-26 Eric Christopher <echristo@redhat.com>
2164
2165 * config/tc-mips.c (mips_dwarf2_addr_size): New.
2166 * config/tc-mips.h (DWARF2_ADDR_SIZE): Use.
2167
2168 2004-02-26 Andrew Stubbs <andrew.stubbs@superh.com>
2169
2170 * config/tc-sh.c (build_Mytes): Add REG_N_D and REG_N_B01
2171 nibble types to assembler.
2172
2173 2004-02-25 Fred Fish <fnf@redhat.com>
2174
2175 * config/tc-iq2000.c: Add missing \n\ in multiline string literal.
2176
2177 2004-02-20 James E Wilson <wilson@specifixinc.com>
2178
2179 * config/tc-ia64.c (slot_index): New arg before_relax. Use instead of
2180 finalize_syms.
2181 (fixup_unw_records): New arg before_relax. Pass to slot_index.
2182 (ia64_estimate_size_before_relax): New.
2183 (ia64_convert_frag): Pass 0 to fixup_unw_records. Add comment.
2184 (generate_unwind_image): Pass 1 to fixup_unw_records.
2185 * config/tc-ia64.h (ia64_estimate_size_before_relax): Declare.
2186 (md_estimate_size_before_relax): Call ia64_estimate_size_before_relax.
2187
2188 2004-02-19 Jakub Jelinek <jakub@redhat.com>
2189
2190 * stabs.c (generate_asm_file): Avoid warning about use of
2191 uninitialized variable.
2192
2193 2004-02-18 David Mosberger <davidm@hpl.hp.com>
2194
2195 * config/tc-ia64.c (ia64_flush_insns): In addition to prologue,
2196 body, and endp, allow unwind records which do not have a "t"
2197 (time/instruction) field.
2198
2199 2004-02-17 Petko Manolov <petkan@nucleusys.com>
2200
2201 * config/tc-arm.c (do_mav_dspsc_1): Correct offset of CRn.
2202 (do_mav_dspsc_2): Likewise.
2203 Fix accumulator registers move opcodes.
2204
2205 2004-02-13 Hannes Reinecke <hare@suse.de>
2206 Jakub Jelinek <jakub@redhat.com>
2207
2208 * dwarf2dbg.c (get_filenum): Do not read beyond allocated memory.
2209
2210 2004-02-10 Steve Ellcey <sje@cup.hp.com>
2211
2212 * config/tc-ia64.h (ia64_frob_symbol): New declaration.
2213 (tc_frob_symbol): New macro definition.
2214 * config/tc-ia64.c (ia64_frob_symbol): New routine.
2215
2216 2004-02-09 Daniel Jacobowitz <drow@mvista.com>
2217
2218 * config/tc-arm.c (md_begin): Mark .note.gnu.arm.ident as
2219 read-only.
2220
2221 2004-02-09 Nathan Sidwell <nathan@codesourcery.com>
2222
2223 * read.h (IGNORE_OPCODE_CASE): Do not define. Replace with ...
2224 (TC_CASE_SENSITIVE): ... this.
2225 * read.c: Replace IGNORE_OPCODE_CASE with TC_CASE_SENSITIVE.
2226 * doc/internals.texi (TC_CASE_SENSITIVE): Document.
2227
2228 2004-02-06 James E Wilson <wilson@specifixinc.com>
2229
2230 * config/tc-ia64.c (dot_endp): Delete call to output_endp.
2231 (generate_unwind_image): Re-add it here.
2232
2233 2004-02-06 Nathan Sidwell <nathan@codesourcery.com>
2234
2235 * dwarf2dbg.c (DWARF2_ADDR_SIZE): Remove trailing ';'
2236 * read.h (SKIP_WHITESPACE): Turn into an expression.
2237 * read.c (read_a_source_file): A pseudo is removed by having a
2238 NULL handler.
2239
2240 2004-02-05 James E Wilson <wilson@specifixinc.com>
2241
2242 * config/tc-ia64.c (output_endp): New.
2243 (count_bits): Delete.
2244 (ia64_flush_insns, process_one_record, optimize_unw_records): Handle
2245 endp unwind records.
2246 (fixup_unw_records): Handle endp unwind records. Delete code for
2247 shortening prologue regions not followed by a body record.
2248 (dot_endp): Call add_unwind_entry to emit endp unwind record.
2249 * config/tc-ia64.h (unw_record_type): Add endp.
2250
2251 2004-02-03 James E Wilson <wilson@specifixinc.com>
2252
2253 * config/tc-ia64.c (ia64_convert_frag): Call md_number_to_chars to
2254 fill padding bytes with zeroes.
2255 (emit_one_bundle): New locals last_ptr, end_ptr. Rewrite code that
2256 sets unwind_record slot_number and slot_frag fields.
2257
2258 2004-02-02 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2259
2260 * config/tc-mips.c (add_got_offset_hilo): New function.
2261 (macro): Use load_register() and add_got_offset_hilo() to load
2262 constants instead of hardcoding code sequences throughout.
2263
2264 2004-01-28 H.J. Lu <hongjiu.lu@intel.com>
2265
2266 * config/tc-ia64.c (emit_one_bundle): Add proper indentation.
2267
2268 2004-01-26 Bernardo Innocenti <bernie@develer.com>
2269
2270 * config/tc-m68k.h (EXTERN_FORCE_RELOC): Handle m68k-uclinux specially,
2271 like m68k-elf.
2272 * config/tc-m68k.c (RELAXABLE_SYMBOL): Use EXTERN_FORCE_RELOC instead
2273 of hard-coded test for TARGET_OS=elf.
2274
2275 2004-01-24 Chris Demetriou <cgd@broadcom.com>
2276
2277 * config/tc-mips.c (hilo_interlocks): Change definition
2278 so that MIPS32, MIPS64 and later ISAs are included, along with
2279 the already-included machines. Update comments.
2280
2281 2004-01-23 Daniel Jacobowitz <drow@mvista.com>
2282
2283 * config/tc-arm.c (tc_gen_reloc): Improve error message for
2284 undefined local labels.
2285
2286 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2287
2288 * config/tc-mips.c (load_address, macro): Update comments about
2289 NewABI GP relaxation.
2290
2291 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2292
2293 * config/tc-mips.c (macro_build): Remove place and counter arguments.
2294 (mips_build_lui, macro_build_ldst_constoffset): Likewise.
2295 (mips16_macro_build, macro_build_jalr): Remove counter argument.
2296 (set_at, load_register, load_address, move_register): Likewise.
2297 (load_got_offset, add_got_offset): Likewise.
2298 Update all calls and tidy accordingly.
2299
2300 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2301
2302 * config/tc-mips.c (RELAX_ENCODE): Remove WARN argument.
2303 (RELAX_FIRST, RELAX_SECOND): Turn into 8-bit quantities.
2304 (RELAX_USE_SECOND): Bump to 0x10000.
2305 (RELAX_SECOND_LONGER, RELAX_NOMACRO, RELAX_DELAY_SLOT): New flags.
2306 (mips_macro_warning): New variable.
2307 (md_assemble): Wrap macro expansion in macro_start() and macro_end().
2308 (s_cpload, s_cpsetup, s_cprestore, s_cpreturn): Likewise.
2309 (relax_close_frag): Set mips_macro_warning.first_frag. Adjust use
2310 of RELAX_ENCODE.
2311 (append_insn): Update mips_macro_warning.sizes.
2312 (macro_start, macro_warning, macro_end): New functions.
2313 (macro_build): Don't emit warnings here.
2314 (macro_build_lui, md_estimate_size_before_relax): ...or here.
2315 (md_convert_frag): Check for cases where one macro alternative
2316 needs a warning and the other doesn't. Emit a warning if the
2317 longer sequence was chosen.
2318
2319 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2320
2321 * config/tc-mips.h (tc_frag_data_type, TC_FRAG_TYPE): Remove.
2322 * config/tc-mips.c (RELAX_ENCODE): Take three arguments: the size of
2323 the first sequence, the size of the second sequence, and a flag
2324 that says whether we should warn.
2325 (RELAX_OLD, RELAX_NEW, RELAX_RELOC[123]): Delete.
2326 (RELAX_FIRST, RELAX_SECOND): New.
2327 (mips_relax): New variable.
2328 (relax_close_frag, relax_start, relax_switch, relax_end): New fns.
2329 (append_insn): Remove "place" argument. Use mips_relax.sequence
2330 rather than "place" to check whether we're expanding the second
2331 alternative of a relaxable macro. Remove redundant check for
2332 branch relaxation. If generating a normal insn, and there
2333 is not enough room in the current frag, call relax_close_frag()
2334 to close it. Update mips_relax.sizes[]. Emit fixups for the
2335 second version of a relaxable macro. Record the first relaxable
2336 fixup in mips_relax. Remove tc_gen_reloc workaround.
2337 (macro_build): Remove all uses of "place". Use mips_relax.sequence
2338 in the same way as in append_insn.
2339 (mips16_macro_build): Remove "place" argument.
2340 (macro_build_lui): As for macro_build. Don't drop the add_symbol
2341 when generating the second version of a relaxable macro.
2342 (load_got_offset, add_got_offset): New functions.
2343 (load_address, macro): Use new relaxation machinery. Remove
2344 tc_gen_reloc workarounds.
2345 (md_estimate_size_before_relax): Set RELAX_USE_SECOND if the second
2346 version of a relaxable macro is needed. Return -RELAX_SECOND if the
2347 first version is needed.
2348 (tc_gen_reloc): Remove relaxation handling.
2349 (md_convert_frag): Go through the fixups for a relaxable macro and
2350 mark those that belong to the unneeded alternative as done. If the
2351 second alternative is needed, adjust the fixup addresses to account
2352 for the deleted first alternative.
2353
2354 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2355
2356 * frags.h (frag_room): Declare.
2357 * frags.c (frag_room): New function.
2358 * doc/internals.texi: Document it.
2359
2360 2004-01-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
2361
2362 * config/tc-mips.c (append_insn): Don't do r3900 interlock
2363 optimization for -mtune=r3900, as this will break on other CPUs.
2364
2365 2004-01-11 Tom Rix <tcrix@worldnet.att.net>
2366
2367 * config/tc-m68hc11.c (build_indexed_byte): movb and movw cannot
2368 be relaxed, use fixup.
2369 (md_apply_fix3): Use 5 bit reloc from movb and movw fixup.
2370
2371 2004-01-19 Jakub Jelinek <jakub@redhat.com>
2372
2373 * config/tc-sparc.c (sparc_ip): Disallow %f32-%f63 for single
2374 precision operands.
2375
2376 2004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2377
2378 * config/tc-mips.c (append_insn): Properly detect variant frags
2379 that preclude swapping of relaxed branches. Correctly swap
2380 instructions between frags when dealing with relaxed branches.
2381
2382 2004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2383
2384 * acinclude.m4: Quote names of macros to be defined by AC_DEFUN
2385 throughout.
2386 * aclocal.m4: Regenerate.
2387 * configure: Regenerate.
2388
2389 2004-01-12 Anil Paranjpe <anilp1@KPITCummins.com>
2390
2391 * config/tc-h8300.c (build_bytes): Apply relaxation to bit
2392 manipulation insns.
2393
2394 2004-01-12 Richard Sandiford <rsandifo@redhat.com>
2395
2396 * config/tc-mips.c (macro_build_jalr): When adding an R_MIPS_JALR
2397 reloc, reserve space for the delay slot as well as the jalr itself.
2398
2399 2004-01-09 Paul Brook <paul@codesourcery.com>
2400
2401 * config/tc-arm.c (do_vfp_reg2_from_sp2): Rename from do_vfp_sp_reg2.
2402 (do_vfp_sp2_from_reg2): New function.
2403 (insns): Use them.
2404 (do_vfp_dp_from_reg2): Check return values properly.
2405
2406 2004-01-08 Ian Lance Taylor <ian@wasabisystems.com>
2407
2408 * config/tc-mips.c (warn_nops): Remove static variable.
2409 (macro): Remove test of warn_nops.
2410 (md_shortops): Remove 'n'.
2411 (md_parse_option): Remove 'n' case.
2412 (md_show_usage): Remove -n.
2413 * doc/as.texinfo (Overview): Remove MIPS -n option.
2414 * doc/c-mips.texi (MIPS Opts): Remove mention -n.
2415 * NEWS: Mention removal of MIPS -n option.
2416
2417 * config/tc-mips.c (ISA_HAS_COPROC_DELAYS): Remove.
2418 (cop_interlocks): Check ISA level.
2419 (cop_mem_interlocks): Define.
2420 (reg_needs_delay): Check cop_interlocks rather than
2421 ISA_HAS_COPROC_DELAYS.
2422 (append_insn): Likewise. Use cop_mem_interlocks rather than
2423 directly checking mips_opts.isa.
2424 (mips_emit_delays): Likewise.
2425
2426 2004-01-07 H.J. Lu <hongjiu.lu@intel.com>
2427
2428 * config/tc-ia64.c (unwind): Move next_slot_number and
2429 next_slot_frag to ...
2430 (unw_rec_list): Here.
2431 (free_list_records): Removed.
2432 (output_unw_records): Likewise.
2433 (generate_unwind_image): Make it void.
2434 (alloc_record): Initialize next_slot_number and next_slot_frag.
2435 (slot_index): Take .org, .space and .align into account.
2436 (fixup_unw_records): Don't set slot_number to 0. Use
2437 list->next_slot_number and list->next_slot_frag instead of
2438 unwind.next_slot_number and unwind.next_slot_frag.
2439 (ia64_convert_frag): New.
2440 (generate_unwind_image): Generate a rs_machine_dependent frag
2441 for unwind record.
2442 (emit_one_bundle): Use list->next_slot_number and
2443 list->next_slot_frag instead of unwind.next_slot_number and
2444 unwind.next_slot_frag.
2445
2446 * config/tc-ia64.h (md_convert_frag): Defined as
2447 ia64_convert_frag.
2448 (md_estimate_size_before_relax): Defined as (f)->fr_var.
2449
2450 2004-01-06 Alexandre Oliva <aoliva@redhat.com>
2451
2452 2003-12-19 Alexandre Oliva <aoliva@redhat.com>
2453 * config/tc-frv.h (md_apply_fix3): Don't define.
2454 * config/tc-frv.c (md_apply_fix3): New. Shift/truncate %hi/%lo
2455 operands.
2456 * config/tc-frv.h (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
2457 2003-10-07 Alexandre Oliva <aoliva@redhat.com>
2458 * config/tc-frv.c (line_separator_chars): Add `!'.
2459 2003-09-19 Alexandre Oliva <aoliva@redhat.com>
2460 * config/tc-frv.c (md_assemble): Clear insn upfront.
2461 2003-09-18 Alexandre Oliva <aoliva@redhat.com>
2462 * config/tc-frv.c (OPTION_FDPIC): New macro.
2463 (md_longopts): Add mfdpic.
2464 (md_parse_option): Handle it.
2465 2003-08-04 Alexandre Oliva <aoliva@redhat.com>
2466 * config/tc-frv.c (md_cgen_lookup_reloc) <FRV_OPERAND_D12,
2467 FRV_OPERAND_S12>: Use reloc type encoded in fix-up.
2468 (frv_pic_ptr): Parse funcdesc.
2469
2470 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2471
2472 * doc/as.texinfo: Let texi2pod parse asconfig.texi and
2473 gasver.texi. Remove duplicate symbol definitions for texi2pod.
2474
2475 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2476
2477 * Makefile.am (Makefile): Move the dependency on
2478 $(BFDDIR)/configure.in to...
2479 (CONFIG_STATUS_DEPENDENCIES): ... here.
2480 (AUTOMAKE_OPTIONS): Require automake 1.8.
2481 * Makefile.in: Regenerate.
2482 * doc/Makefile.am (BASEDIR, BFDDIR): Define.
2483 (CONFIG_STATUS_DEPENDENCIES): Add a dependency on
2484 $(BFDDIR)/configure.in here as well.
2485 * doc/Makefile.in: Regenerate.
2486
2487 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2488
2489 * Makefile.am (install, install-info, RECURSIVE_TARGETS): Remove.
2490 * Makefile.in: Regenerate.
2491 * aclocal.m4: Regenerate.
2492 * doc/Makefile.am (install, install-info): Remove.
2493 (install-data-local): A new hook for install-info.
2494 (AUTOMAKE_OPTIONS): Require automake 1.8.
2495 * doc/Makefile.in: Regenerate.
2496
2497 2004-01-02 Nutan Singh <nutan@kpitcummins.com>
2498
2499 * doc/c-sh.texi: Update description about floating point behavior
2500 of SH family.
2501
2502 2004-01-02 Bernardo Innocenti <bernie@develer.com>
2503
2504 * configure.in: Add m68k-uClinux target.
2505 * configure: Regenerate.
2506
2507 For older changes see ChangeLog-0203
2508 \f
2509 Local Variables:
2510 mode: change-log
2511 left-margin: 8
2512 fill-column: 74
2513 version-control: never
2514 End:
This page took 0.077511 seconds and 5 git commands to generate.