1 2004-11-23 Jan Beulich <jbeulich@novell.com>
3 * config/tc-i386.h (CpuMMX2): Declare. Artificial classifier to
4 indicate the MMX extensions added by both SSE and 3DNow!A.
6 (CpuUnknownFlags): Update.
7 * config/tc-i386.c (cpu_sub_arch_name): Declare.
8 (cpu_arch): i586 and pentium do not imply MMX. i686 and pentiumpro do
9 neither imply SSE nor MMX. k6 implies MMX. k6_2 additionally implies
10 3DNow!. Athlon additionally implies 3DNow!A. Several new
11 entries (those starting with a dot are for sub-arch specification).
12 (set_cpu_arch): Handle sub-arch specifications.
13 (parse_insn): Distinguish between instructions not supported because
14 of insufficient CPU features and because of 64-bit mode.
15 * doc/c-i386.texi: Describe enhanced .arch directive.
17 2004-11-22 Bob Wilson <bob.wilson@acm.org>
19 * config/tc-xtensa.c (xg_add_opcode_fix): Set fx_no_overflow.
21 2004-11-22 Bob Wilson <bob.wilson@acm.org>
23 * dwarf2dbg.c (dwarf2_finish): Don't write a .debug_line section
24 without a corresponding .debug_info section.
26 2004-11-22 Hans-Peter Nilsson <hp@axis.com>
28 * read.c (potable): Add "error" and "warning".
29 (s_errwarn): New function.
30 * read.h (s_errwarn): Declare.
31 * doc/as.texinfo (Error, Warning): Document .error and .warning.
33 2004-11-22 Nick Clifton <nickc@redhat.com>
35 * config/tc-tic54x.c (tic54x_adjust_symtab): Adjust call to
38 2004-11-19 Alan Modra <amodra@bigpond.net.au>
40 * config/tc-msp430.c (struct rcodes_s, MSP430_RLC, msp430_rcodes,
41 struct hcodes_s, msp430_hcodes): From include/opcode/msp430.h.
43 2004-11-19 Alan Modra <amodra@bigpond.net.au>
45 * config/obj-coff.c (c_dot_file_symbol): Add "app" param.
46 (coff_adjust_symtab): Adjust call.
47 (crawl_symbols): Likewise.
48 * config/obj-coff.h (c_dot_file_symbol): Add "app" param.
49 (obj_app_file): Adjust.
51 2004-11-18 Inderpreet Singh <inderpreetb@nioda.hcltech.com>
53 * configure.in: Enable bfd_assember for the MAXQ port.
54 * configure: Regenerate.
56 2004-11-12 Bob Wilson <bob.wilson@acm.org>
57 Sterling Augustine <sterling@tensilica.com>
59 * config/tc-xtensa.c (finish_vinsn): Clear pending instruction if
61 (check_t1_t2_reads_and_writes): Check for both reads and writes to
62 interfaces that are related as determined by xtensa_interface_class_id.
64 2004-11-12 Nick Clifton <nickc@redhat.com>
66 * config/tc-mn10300.c (md_relax_table): Fix off by one negative
67 offsets for conditional branches.
69 2004-11-11 Bob Wilson <bob.wilson@acm.org>
71 * config/tc-xtensa.c (MAX_IMMED6): Change value to 65.
73 2004-11-10 Bob Wilson <bob.wilson@acm.org>
75 * config/tc-xtensa.c (update_next_frag_state): Always add a NOP if
76 relaxing at the end of a loop. Don't mark frags as UNREACHABLE or
78 (relax_frag_immed): Update call to update_next_frag_state.
80 2004-11-10 Alan Modra <amodra@bigpond.net.au>
82 * obj.h (struct format_ops <app_file>): Add int param.
83 * read.h (s_app_file_string): Likewise.
84 * read.c (s_app_file_string): Likewise.
85 (s_app_file): Adjust s_app_file_string call.
86 * config/tc-mips.c (s_mips_file): Likewise.
87 * config/obj-coff.h (obj_app_file): Add app param.
88 * config/obj-ecoff.h (obj_app_file): Likewise.
89 * config/obj-multi.h (obj_app_file): Likewise.
90 * config/obj-elf.h (elf_file_symbol): Likewise.
91 * config/obj-elf.c (elf_file_symbol): Only emit one file symbol
92 if called for # preprocessor lines.
94 2004-11-08 H.J. Lu <hongjiu.lu@intel.com>
97 * symbols.c (resolve_symbol_value): Convert weak symbols only
99 (symbol_equated_reloc_p): Don't equate weaks when relocating
100 only for Windows PECOFF.
102 2004-11-08 Tomer Levi <Tomer.Levi@nsc.com>
104 * config/tc-crx.c (print_insn): Check and set insn_addr.
105 * config/tc-crx.h (md_frag_check): Define.
107 2004-11-08 Inderpreet Singh <inderpreetb@nioda.hcltech.com>
108 Vineet Sharma <vineets@noida.hcltech.com>
110 * configure.in: Add support for new target maxq-coff.
111 * configure: Regenerate.
112 * NEWS: Mention new support.
113 * config/tc-maxq.c: New file.
114 * config/tc-maxq.h: New file.
115 * config/obj-coff.h: Add support for maxq-coff.
117 2004-11-08 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
119 * symbols.c (any_external_name): Define.
120 (resolve_symbol_value): Convert weak symbols.
121 (S_SET_EXTERNAL): Support any_external_name.
122 (S_SET_NAME): Qualify parameter const.
123 (symbol_equated_reloc_p): Equate to weaks when relocating.
124 * symbols.h (S_SET_NAME): Qualfiy parameter const.
125 * tc.h (any_external_name): Declare.
126 * config/obj-coff.c ("coff/pe.h"): Include for BFD
128 (weak_is_altname): Declare and define.
129 (weak_name2altname): Same.
130 (weak_altname2name): Same.
131 (weak_uniquify): Same.
132 (weak_altprefix): Define.
133 (obj_coff_weak): Change .weak syntax and handling.
134 (coff_frob_symbol): Fix PE weak symbol alternates.
135 * config/obj-coff.h (USE_UNIQUE): Define.
136 * config/tc-i386.c (md_apply_fix3): Assume weak symbols
137 are in another segment.
138 (tc_gen_reloc): Remove broken addend hack.
139 doc/as.texinfo: Update.
141 2004-11-05 Sterling Augustine <sterling@tensilica.com>
143 * config/tc-xtensa.c (total_frag_text_expansion): New.
144 (md_estimate_size_before_relax): Use it.
145 (find_address_of_next_align_frag): Likewise.
147 2004-11-05 Tomer Levi <Tomer.Levi@nsc.com>
149 * config/tc-crx.c: Rename argument types.
150 (processing_arg_number): Rename to 'cur_arg_num'.
151 (get_number_of_bits): Rename to 'set_operand_size'.
152 (get_operandtype): Rename to 'parse_operand', totally rewrite.
153 (set_cons_rparams): Rename to 'set_operand', totally rewrite.
154 (set_indexmode_parameters): Remove function, integrate its code into
156 (set_operand_size): Get rid of 'Operand Number' function parameter -
157 use global variable 'cur_arg_num' instead.
158 Use a local 'argument' pointer to reference the current argument.
159 (parse_operand): Likewise.
160 (set_operand): Likewise.
161 (process_label_constant): Likewise.
163 2004-11-04 Bob Wilson <bob.wilson@acm.org>
165 * config/tc-xtensa.c: Remove XTENSA_SECTION_RENAME ifdefs.
166 (add_section_rename): Delete. Inlined into...
167 (build_section_rename): ...here. Use xstrdup instead of strdup.
168 (xtensa_section_rename): Drop "const" from argument and return types.
169 (md_show_usage): Indent to match show_usage().
170 * config/tc-xtensa.h: Remove XTENSA_SECTION_RENAME ifdefs.
171 (tc_canonicalize_section_name): Define.
172 (md_elf_section_rename): Remove unused macro.
173 * doc/as.texinfo (Overview): Document Xtensa --rename-section option.
174 * doc/c-xtensa.texi (Xtensa Options): Likewise.
175 (Frame Directive): Delete.
177 2004-11-04 Daniel Jacobowitz <dan@debian.org>
179 * configure.in: Remove arm-*-oabi and thumb-*-oabi.
180 * config/tc-arm.c (target_oabi): Delete.
181 (md_apply_fix3, elf32_arm_target_format): Remove target_oabi checks.
182 (arm_opts): Remove moabi.
183 * doc/as.texinfo (Overview): Remove documentation of -moabi.
184 * doc/c-arm.texi (ARM Options): Likewise.
185 * configure: Regenerated.
187 2004-11-04 Hans-Peter Nilsson <hp@axis.com>
189 * configure.in (crisv32): Recognize. AC_DEFINE_UNQUOTED
190 DEFAULT_CRIS_ARCH. Handle crisv32-*-linux-gnu* like
191 cris-*-linux-gnu* and crisv32-*-* like cris-*-*.
192 * configure: Regenerate.
193 * config/tc-cris.c (enum cris_archs): New.
194 (cris_mach, cris_arch_from_string, s_cris_arch, get_sup_reg)
195 (cris_insn_ver_valid_for_arch): New functions.
196 (DEFAULT_CRIS_ARCH): New macro, default to cris_any_v0_v10.
197 (cris_arch): New variable.
198 (md_pseudo_table): New pseudo .arch.
199 (err_for_dangerous_mul_placement): Initialize according to
201 (STATE_COND_BRANCH): Renamed from STATE_CONDITIONAL_BRANCH.
203 (STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON)
204 (STATE_ABS_BRANCH_V32, STATE_LAPC, BRANCH_BF_V32, BRANCH_BB_V32)
205 (BRANCH_WF_V32, BRANCH_WB_V32): New.
206 (BRANCH_BF, BRANCH_BB, BRANCH_WF, BRANCH_WB): Don't undef after
207 use in md_cris_relax_table.
208 (md_cris_relax_table): Add entries for STATE_COND_BRANCH_V32,
209 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
210 Update and improve head comment.
211 (OPTION_PIC): Define in terms of previous option, OPTION_US.
212 (OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): Similar.
214 (md_longopts): New option --march=...
215 (cris_any_v0_v10_long_jump_size, crisv32_long_jump_size): New
217 (md_long_jump_size): Initialize in terms of DEFAULT_CRIS_ARCH.
218 (HANDLE_RELAXABLE): New macro.
219 (md_estimate_size_before_relax): Use HANDLE_RELAXABLE for common
220 cases. Check for weak symbols and assume not relaxable. Handle
221 STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON,
222 STATE_ABS_BRANCH_V32, STATE_LAPC. Use new variable symbolP, not
224 (md_convert_frag): Handle STATE_COND_BRANCH_V32,
225 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
226 (cris_create_short_jump): Adjust for CRISv32.
227 (cris_relax_frag): Handle new states.
228 (md_create_long_jump): Ditto. Emit error for common_v10_v32.
229 (md_begin): Define symbols "..asm.arch.cris.v32",
230 "..asm.arch.cris.v10", "..asm.arch.cris.common_v10_v32" and
231 "..asm.arch.cris.any_v0_v10". Use cris_insn_ver_valid_for_arch
232 when entering opcode table entry points.
233 (md_assemble): Adjust branch handling for CRISv32. Handle LAPC
234 relaxation. In fix_new_exp call for main insn, pass 1 for pcrel
235 parameter for 8, 16 and 32-bit pc-relative insns and LAPC.
236 (cris_process_instruction): Initialize out_insnp->insn_type to
237 CRIS_INSN_NONE, not CRIS_INSN_NORMAL.
238 <case ']', '[', 'A', 'd', 'Q', 'N', 'n', 'Y', 'U', 'u', 'T'>: New
240 <case 'm'>: Check that modified_char == '.'.
241 <invalid operands>: Consume the rest of the line.
242 When operands don't match, skip over subsequent insns with
243 non-matching version specifier but same mnemonic.
244 <immediate constant, case SIZE_SPEC_REG>: Immediate operands for
245 special registers in CRISv32 are always 32 bit long.
246 <immediate constant, case SIZE_FIELD_SIGNED, SIZE_FIELD_UNSIGNED>:
248 (get_gen_reg): Only recognize "PC" when followed by "+]" for v32
249 and compatible. Recognize "ACR" for v32, unless followed by "+".
250 (get_spec_reg): Consider cris_arch when looking up register.
251 (get_autoinc_prefix_or_indir_op): Don't recognize assignment for
253 (get_3op_or_dip_prefix_op): Check for ']' after seeing '[rN+'.
254 (cris_get_expression): Restore input_line_pointer if failing "early".
255 (get_flags): Consider cris_arch and recognize flags accordingly.
256 (branch_disp): Adjust for CRISv32.
257 (gen_cond_branch_32): Similar. Emit error for common_v10_v32.
258 (cris_number_to_imm): Use as_bad_where, not as_bad. Remove
259 related FIXME. Don't insist on BFD_RELOC_32_PCREL fixup to be
260 resolved. Don't enter zeros in object file for
262 <case BFD_RELOC_CRIS_LAPCQ_OFFSET, BFD_RELOC_CRIS_SIGNED_16>
263 <case BFD_RELOC_CRIS_SIGNED_8>: New case.
264 (md_parse_option): Break out "return 1".
265 <OPTION_ARCH> New case.
266 (tc_gen_reloc): <case BFD_RELOC_CRIS_LAPCQ_OFFSET>
267 <case BFD_RELOC_CRIS_SIGNED_16, BFD_RELOC_CRIS_SIGNED_8>
268 <case BFD_RELOC_CRIS_UNSIGNED_8, BFD_RELOC_CRIS_UNSIGNED_16>
269 <case BFD_RELOC_32_PCREL>: New cases.
270 Addends for non-zero fx_pcrel are too in fx_offset.
271 (md_show_usage): Show --march=<arch>.
272 (md_apply_fix3): Adjust val for BFD_RELOC_CRIS_LAPCQ_OFFSET.
273 (md_pcrel_from): BFD_RELOC_CRIS_LAPCQ_OFFSET is PC-relative too.
274 (s_syntax) <struct syntaxes>: Properly constify member operand.
275 * config/tc-cris.h (TARGET_MACH): Define.
276 (cris_mach): Declare.
277 * doc/as.texinfo (Overview) <CRIS>: Add --march=...
278 * doc/c-cris.texi (CRIS-Symbols): New node for built-in symbols.
279 (CRIS-Opts): Document --march=...
280 (CRIS-Pseudos): Document .arch.
282 2004-11-04 Jan Beulich <jbeulich@novell.com>
284 * config/tc-i386.c (set_intel_syntax): Allow % in symbol names when
285 intel syntax and no register prefix, allow $ in symbol names when
287 (set_16bit_gcc_code_flag): Replace literal 'l' by LONG_MNEM_SUFFIX.
288 (intel_float_operand): Add fourth return value indicating math control
289 operations. Make classification more precise.
290 (md_assemble): Complain if memory operand of mov[sz]x has no size
292 (parse_insn): Translate word operands to floating point instructions
293 operating on integers as well as control instructions to short ones
294 as expected by AT&T syntax. Translate 'd' suffix to short one only for
295 floating point instructions operating on non-integer operands.
296 (match_template): Remove fldcw special case. Adjust q-suffix handling
297 to permit it on fild/fistp/fisttp in AT&T mode.
298 (process_suffix): Don't guess DefaultSize insns' suffix from
299 stackop_size for certain floating point control instructions. Guess
300 suffix for branch and [ls][gi]dt based on flag_code. Split error
301 messages for Intel and AT&T syntax, and make the condition more strict
302 for the former. Adjust suppressing of generation of operand size
304 (intel parser): Allow the full set of MASM operators. Add FWORD, TBYTE,
305 OWORD, and XMMWORD operand size specifiers (TBYTE replaces XWORD). Add
307 * config/tc-i386.h (BYTE_PTR WORD_PTR DWORD_PTR QWORD_PTR XWORD_PTR
308 SHORT OFFSET_FLAT FLAT NONE_FOUND): Remove unused defines.
310 2004-11-03 Hans-Peter Nilsson <hp@axis.com>
312 * symbols.c (colon) [!WORKING_DOT_WORD]: Don't declare
313 md_short_jump_size, md_long_jump_size.
314 * write.c [!WORKING_DOT_WORD]: Ditto.
315 * tc.h [!WORKING_DOT_WORD]: Declare them here. Drop const
317 * config/tc-cris.h (md_short_jump_size, md_long_jump_size): Don't
319 * config/tc-cris.c (md_short_jump_size, md_long_jump_size): Drop
320 const qualifier in these definitions.
321 * config/tc-i370.c, config/tc-m68k.c, config/tc-pdp11.c,
322 config/tc-s390.c, config/tc-tahoe.c, config/tc-vax.c: Ditto.
324 2004-11-02 Nick Clifton <nickc@redhat.com>
326 * dwarf2dbg.c (dwarf2_finish): Check for the existence of a file
327 table before deciding to produce a .debug_line section to match up
328 with a user provided .debug_info section.
330 2004-10-28 Tomer Levi <Tomer.Levi@nsc.com>
332 * config/tc-crx.c (getreg_image): Bug fix, a return value was
333 mistakenly omitted from CRX_C_REGTYPE and CRX_CS_REGTYPE cases.
335 2004-10-27 Tomer Levi <Tomer.Levi@nsc.com>
337 * config/tc-crx.c: Remove global variable 'post_inc_mode'.
338 (get_flags): New function.
339 (get_number_of_bits): Edit comments, update numeric values to
341 (process_label_constant): Don't support the colon
342 format (SYMBOL:[s|m|l]).
343 (set_cons_rparams): Support argument type 'arg_rbase'.
344 (get_operandtype): Bug fix in 'rbase' operand type parsing.
345 (handle_LoadStor): Bug fix, first handle post-increment mode.
346 (getreg_image): Remove redundant code, update according to latest
348 (print_constant): Bug fix relate to 3-word instructions.
349 (assemble_insn): Bug fix, when matching instructions, verify also
350 instruction type (not only mnemonic).
351 Add various error checking.
352 (preprocess_reglist): Support HI/LO and user registers.
354 2004-10-25 David Mosberger-Tang <davidm@hpl.hp.com>
356 * config/tc-ia64.c (fixup_unw_records): Don't let the "t" value
357 in an epilogue directive go negative.
359 2004-10-25 H.J. Lu <hongjiu.lu@intel.com>
362 * config/tc-ia64.c (emit_one_bundle): Decrement md.num_slots_in_use
363 after reporting template error during manual bundling. Reported
364 by Michael Dupont, michaelx.dupont@intel.com.
366 2004-10-25 Daniel Jacobowitz <dan@debian.org>
368 * Makefile.am: Run dep-am.
369 * aclocal.m4: Regenerate with automake 1.9.2.
370 * Makefile.in: Regenerate with automake 1.9.2.
371 * doc/Makefile.in: Likewise.
373 * config/tc-arm.c: Include "dw2gencfi.h".
374 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
376 * config/tc-arm.h (TARGET_USE_CFIPOP, DWARF2_DEFAULT_RETURN_COLUMN)
377 (DWARF2_CIE_DATA_ALIGNMENT, tc_regname_to_dw2regnum)
378 (tc_cfi_frame_initial_instructions): Define.
379 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
382 2004-10-21 Tomer Levi <Tomer.Levi@nsc.com>
384 * config/tc-crx.c (assemble_insn): Check unsigned immediate
386 Update coprocessor id to be unsigned immediate.
388 2004-10-18 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
390 * config/tc-i386.c (O_secrel): Delete.
391 (tc_pe_dwarf2_emit_offset): New function.
392 * config/tc-i386.h (O_secrel): Define as O_md1.
393 (TC_DWARF2_EMIT_OFFSET): Define.
395 2004-10-18 Nick Clifton <nickc@redhat.com>
397 * config/tc-xstormy16.c (xstormy16_cons_fix_new): Accept and
398 ignore @fptr() directives for 4-byte fixups.
400 2004-10-15 Alan Modra <amodra@bigpond.net.au>
402 * Makefile.am: Run "make dep-am"
403 * Makefile.in: Regenerate.
405 2004-10-14 Bob Wilson <bob.wilson@acm.org>
407 * doc/c-xtensa.texi (Xtensa Options, Absolute Literals Directive):
408 Remove comments about placement of literal pools.
409 (Literal Directive): Update description of literal placement.
410 (Literal Prefix Directive): Remove statement that this does not apply
411 to absolute-mode literals. Describe new section naming scheme.
413 2004-10-12 Bob Wilson <bob.wilson@acm.org>
415 * config/tc-xtensa.c (xtensa_frequency_pseudo): Use set_subseg_freq.
416 (is_entry_opcode, is_movi_opcode, is_the_loop_opcode, is_jx_opcode,
417 is_windowed_return_opcode): Delete.
418 (xtensa_frob_label): Use get_subseg_target_freq.
419 (md_assemble): Inline call to is_entry_opcode.
420 (xtensa_handle_align): Inline call to get_frag_is_literal.
421 (relaxation_requirements): Inline call to is_jx_opcode.
422 (emit_single_op): Inline call to is_movi_opcode.
423 (xg_assemble_vliw_tokens): Inline calls to get_frag_is_insn,
424 get_frag_is_no_transform, is_entry_opcode, and
425 set_frag_is_specific_opcode. Use get_subseg_total_freq.
426 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags,
427 xtensa_fix_close_loop_end_frags, relax_frag_immed, convert_frag_immed):
428 Inline calls to get_frag_is_no_transform.
429 (next_instrs_are_b_retw): Inline call to is_windowed_return_opcode.
430 (xtensa_fix_short_loop_frags): Inline calls to is_the_loop_opcode and
431 get_frag_is_no_transform.
432 (convert_frag_immed_finish_loop): Inline calls to get_expression_value
433 and set_frag_is_no_transform.
434 (get_expression_value): Delete.
435 (subseg_map struct): Rename cur_total_freq to total_freq. Rename
436 cur_target_freq to target_freq.
437 (get_subseg_info): Split out code to create a new map entry into ...
438 (add_subseg_info): ... this new function.
439 (get_last_insn_flags): Check if get_subseg_info succeeded.
440 (set_last_insn_flags): Call add_subseg_info if needed.
441 (get_subseg_total_freq, get_subseg_target_freq, set_subseg_freq): New.
442 (xtensa_reorder_segments): Compute last_sec while counting sections.
443 Remove call to get_last_sec.
444 (get_last_sec): Delete.
445 (cache_literal_section): Inline call to retrieve_literal_seg and its
446 callees, seg_present and add_seg_list.
447 (retrieve_literal_seg, seg_present, add_seg_list): Delete.
448 (get_frag_is_insn, get_frag_is_no_transform,
449 set_frag_is_specific_opcode, set_frag_is_no_transform): Delete.
450 * config/tc-xtensa.h (MAX_SLOTS): Reduce from 31 to 15.
452 2004-10-12 Bob Wilson <bob.wilson@acm.org>
454 * config/tc-xtensa.c: Use ISO C90 formatting.
455 * config/tc-xtensa.h: Likewise.
456 * config/xtensa-istack.h: Likewise.
457 * config/xtensa-relax.c: Likewise.
458 * config/xtensa-relax.h: Likewise.
460 2004-10-12 Paul Brook <paul@codesourcery.com>
462 * config/tc-arm.c (md_begin): Change EF_ARM_EABI_VER3 to
465 * doc/c-arm.texi: Document that we actually support -meabi=4, not
468 2004-10-08 Bob Wilson <bob.wilson@acm.org>
470 * doc/as.texinfo (VTableEntry, VTableInherit): Add "directive" to index
472 (Acknowledgements): Use "GAS" instead of AS variable.
474 2004-10-08 Daniel Jacobowitz <dan@debian.org>
476 * config/tc-i386.c: Include "elf/x86-64.h".
477 (i386_elf_section_type): New function.
478 * config/tc-i386.h (md_elf_section_type): Define.
479 (i386_elf_section_type): New prototype.
481 2004-10-08 Linus Nielsen Feltzing <linus@haxx.se>
483 * config/m68k-parse.h (enum m68k_register): New control register,
485 * config/tc-m68k.c: Correct control register set for MCF5249.
487 2004-10-07 Bob Wilson <bob.wilson@acm.org>
488 Sterling Augustine <sterling@tensilica.com>
490 * config/tc-xtensa.c (absolute_literals_supported): New global flag.
491 (UNREACHABLE_MAX_WIDTH): Define.
492 (XTENSA_FETCH_WIDTH): Delete.
493 (cur_vinsn, xtensa_fetch_width, xt_saved_debug_type, past_xtensa_end,
494 prefer_const16, prefer_l32r): New global variables.
495 (LIT4_SECTION_NAME): Define.
496 (lit4_state struct): Add lit4_seg_name and lit4_seg fields.
497 (XTENSA_PROP_*, GET_XTENSA_PROP_*, SET_XTENSA_PROP_*): Define.
498 (frag_flags struct): New.
499 (xtensa_block_info struct): Move from tc-xtensa.h. Add flags field.
500 (subseg_map struct): Add cur_total_freq and cur_target_freq fields.
501 (bitfield, bit_is_set, set_bit, clear_bit): Define.
502 (MAX_FORMATS): Define.
503 (op_placement_info struct, op_placement_table): New.
504 (O_pltrel, O_hi16, O_lo16): Define.
505 (directiveE enum): Rename directive_generics to directive_transform.
506 Delete directive_relax. Add directive_schedule,
507 directive_absolute_literals, and directive_last_directive.
508 (directive_info): Rename "generics" to "transform". Delete "relax".
509 Add "schedule" and "absolute-literals".
510 (directive_state): Adjust entries to match changes in directive_info.
511 (xtensa_relax_statesE, RELAX_IMMED_MAXSTEPS): Move to tc-xtensa.h.
512 (xtensa_const16_opcode, xtensa_movi_opcode, xtensa_movi_n_opcode,
513 xtensa_l32r_opcode, xtensa_nop_opcode, xtensa_rsr_lcount_opcode): New.
514 (xtensa_j_opcode, xtensa_rsr_opcode): Delete.
515 (align_only_targets, software_a0_b_retw_interlock,
516 software_avoid_b_j_loop_end, maybe_has_b_j_loop_end,
517 software_avoid_short_loop, software_avoid_close_loop_end,
518 software_avoid_all_short_loops, specific_opcode): Delete.
519 (warn_unaligned_branch_targets): New.
520 (workaround_a0_b_retw, workaround_b_j_loop_end, workaround_short_loop,
521 workaround_close_loop_end, workaround_all_short_loops): Default FALSE.
522 (option_[no_]link_relax, option_[no_]transform,
523 option_[no_]absolute_literals, option_warn_unaligned_targets,
524 option_prefer_l32r, option_prefer_const16, option_target_hardware):
526 (option_[no_]align_only_targets, option_literal_section_name,
527 option_text_section_name, option_data_section_name,
528 option_bss_section_name, option_eb, option_el): Delete.
529 (md_longopts): Add entries for: [no-]transform, [no-]absolute-literals,
530 warn-unaligned-targets, prefer-l32r, prefer-const16, [no-]link-relax,
531 and target-hardware. Delete entries for [no-]target-align-only,
532 literal-section-name, text-section-name, data-section-name, and
534 (md_parse_option): Handle new options and remove old ones. Accept but
535 ignore [no-]density options. Warn for [no-]generics and [no-]relax
536 and treat them as [no-]transform.
537 (md_show_usage): Add new options and remove old ones.
538 (xtensa_setup_hw_workarounds): New.
539 (md_pseudo_table): Change "word" entry to use xtensa_elf_cons. Add
540 "long", "short", "loc" and "frequency" entries.
541 (use_generics): Rename to ...
542 (use_transform): ... this function. Add past_xtensa_end check.
543 (use_longcalls): Add past_xtensa_end check.
544 (code_density_available, can_relax): Delete.
545 (do_align_targets): New.
546 (get_directive): Accept dashes in directive names. Warn about
547 [no-]generics and [no-]relax directives and treat them as
549 (xtensa_begin_directive): Call md_flush_pending_output only for some
550 directives. Check for directives inside instruction bundles. Warn
551 about deprecated ".begin literal" usage. Warn and ignore [no-]density
552 directives. Handle new directives. Check generating_literals flag
554 (xtensa_end_directive): Check for directives inside instruction
555 bundles. Warn and ignore [no-]density directives. Handle new
556 directives. Call xtensa_set_frag_assembly_state.
557 (xtensa_loc_directive_seen, xtensa_dwarf2_directive_loc,
558 xtensa_dwarf2_emit_insn): New.
559 (xtensa_literal_position): Call md_flush_pending_output. Do not check
560 use_literal_section flag.
561 (xtensa_literal_pseudo): Call md_flush_pending_output. Handle absolute
562 literals. Use xtensa_elf_cons to parse the expression.
563 (xtensa_literal_prefix): Do not check use_literal_section. Support
564 ".lit4" sections for absolute literals. Change prefix convention to
565 replace ".text" (or ".t" in a linkonce section). No need to call
567 (xtensa_frequency_pseudo, xtensa_elf_cons, xtensa_elf_suffix): New.
568 (expression_end): Handle closing braces and colons.
569 (PLT_SUFFIX, plt_suffix): Delete.
570 (expression_maybe_register): Use new xtensa-isa.h functions. Use
571 xtensa_elf_suffix instead of checking for plt suffix, and handle O_lo16
572 and O_hi16 expressions as well.
573 (tokenize_arguments): Handle closing braces and colons.
574 (parse_arguments): Use new xtensa-isa.h functions. Handle "invisible"
575 operands and paired register syntax.
576 (get_invisible_operands): New.
577 (xg_translate_sysreg_op): Handle new Xtensa LX RSR/WSR/XSR syntax. Use
578 new xtensa-isa.h functions.
579 (xtensa_translate_old_userreg_ops, xtensa_translate_zero_immed): New.
580 (xg_translate_idioms): Check if inside bundle. Use use_transform.
581 Handle new Xtensa LX RSR/WSR/XSR syntax. Remove code to widen density
582 instructions. Use xtensa_translate_zero_immed.
583 (operand_is_immed, operand_is_pcrel_label): Delete.
584 (get_relaxable_immed): Use new xtensa-isa.h functions.
585 (get_opcode_from_buf): Add slot parameter. Use new xtensa-isa.h
587 (xtensa_print_insn_table, print_vliw_insn): New.
588 (is_direct_call_opcode): Use new xtensa-isa.h functions.
589 (is_call_opcode, is_loop_opcode, is_conditional_branch_opcode,
590 is_branch_or_jump_opcode): Delete.
591 (is_movi_opcode, decode_reloc, encode_reloc, encode_alt_reloc): New.
592 (opnum_to_reloc, reloc_to_opnum): Delete.
593 (xtensa_insnbuf_set_operand, xtensa_insnbuf_get_operand): Use new
594 xtensa-isa.h functions. Operate on one slot of an instruction.
595 (xtensa_insnbuf_set_immediate_field, is_negatable_branch,
596 xg_get_insn_size): Delete.
597 (xg_get_build_instr_size): Use xg_get_single_size.
598 (xg_is_narrow_insn, xg_is_single_relaxable_insn): Update calls to
599 xg_build_widen_table. Use xg_get_single_size.
600 (xg_get_max_narrow_insn_size): Delete.
601 (xg_get_max_insn_widen_size, xg_get_max_insn_widen_literal_size,
602 xg_is_relaxable_insn): Update calls to xg_build_widen_table. Use
604 (xg_build_to_insn): Record the loc field. Handle OP_OPERAND_HI16U and
605 OP_OPERAND_LOW16U. Check xg_valid_literal_expression.
606 (xg_expand_to_stack, xg_expand_narrow): Update calls to
607 xg_build_widen_table. Use xg_get_single_size.
608 (xg_immeds_fit): Use new xtensa-isa.h functions. Update call to
610 (xg_symbolic_immeds_fit): Likewise. Also handle O_lo16 and O_hi16, and
611 treat weak symbols conservatively.
612 (xg_check_operand): Use new xtensa-isa.h functions.
613 (is_dnrange): Delete.
614 (xg_assembly_relax): Inline previous calls to tinsn_copy.
615 (xg_finish_frag): Specify separate relax states for the frag and slot0.
616 (is_branch_jmp_to_next, xg_add_branch_and_loop_targets): Use new
617 xtensa-isa.h functions.
618 (xg_instruction_matches_option_term, xg_instruction_matches_or_options,
619 xg_instruction_matches_options): New.
620 (xg_instruction_matches_rule): Handle O_register expressions. Call
621 xg_instruction_matches_options.
622 (transition_rule_cmp): New.
623 (xg_instruction_match): Update call to xg_build_simplify_table.
624 (xg_build_token_insn): Record loc fields.
625 (xg_simplify_insn): Check is_specific_opcode field and
626 density_supported flag.
627 (xg_expand_assembly_insn): Skip checking code_density_available. Use
628 new xtensa-isa.h functions. Call use_transform instead of can_relax.
629 (xg_assemble_literal): Add error handling for O_big. Call
630 record_alignment. Handle O_pltrel.
631 (xg_valid_literal_expression): New.
632 (xg_assemble_literal_space): Add slot parameter. Remove call to
633 set_expr_symbol_offset. Add call to record_alignment. Update call to
635 (xg_emit_insn): Delete.
636 (xg_emit_insn_to_buf): Add format parameter. Update calls to
637 xg_add_opcode_fix and xtensa_insnbuf_to_chars.
638 (xg_add_opcode_fix): Change opcode parameter to tinsn and add format
639 and slot parameters. Handle new "alternate" relocations for absolute
640 literals and CONST16 instructions. Check for bad uses of O_lo16 and
641 O_hi16. Use new xtensa-isa.h functions.
642 (xg_assemble_tokens): Delete.
643 (is_register_writer): Use new xtensa-isa.h functions.
644 (is_bad_loopend_opcode): Check for xtensa_rsr_lcount_opcode instead of
645 old-style RSR from LCOUNT.
646 (next_frag_opcode): Delete.
647 (next_frag_opcode_is_loop, next_frag_format_size, frag_format_size,
648 update_next_frag_state): New.
649 (update_next_frag_nop_state): Delete.
650 (next_frag_pre_opcode_bytes): Use next_frag_opcode_is_loop.
651 (xtensa_mark_literal_pool_location): Check use_literal_section flag and
652 the state of the absolute-literals directive. Add calls to
653 record_alignment and xtensa_set_frag_assembly_state. Call
654 xtensa_switch_to_non_abs_literal_fragment instead of
655 xtensa_switch_to_literal_fragment.
657 (assemble_nop): Use build_nop. Update call to xtensa_insnbuf_to_chars.
658 (get_expanded_loop_offset): Change check for undefined opcode to an
660 (xtensa_set_frag_assembly_state, relaxable_section,
661 xtensa_find_unmarked_state_frags, xtensa_find_unaligned_branch_targets,
662 xtensa_find_unaligned_loops, xg_apply_tentative_value): New.
663 (md_begin): Update call to xtensa_isa_init. Initialize linkrelax to 1.
664 Set lit4_seg_name. Call xg_init_vinsn. Initialize new global opcodes.
665 Call init_op_placement_info_table and xtensa_set_frag_assembly_state.
666 (xtensa_init_fix_data): New.
667 (xtensa_frob_label): Reset label symbol to the current frag. Check
668 do_align_targets and generating_literals flag. Propagate frequency
669 info to new alignment frag. Call xtensa_set_frag_assembly_state.
670 (xtensa_unrecognized_line): New.
671 (xtensa_flush_pending_output): Check if inside a bundle. Add a call
672 to xtensa_set_frag_assembly_state.
673 (error_reset_cur_vinsn): New.
674 (md_assemble): Remove check for literal frag. Remove call to
675 istack_init. Call use_transform instead of use_generics. Parse
676 explicit instruction format specifiers. Move code for
677 a0_b_retw_interlock workaround to xg_assemble_vliw_tokens. Call
678 error_reset_cur_vinsn on errors. Add call to get_invisible_operands.
679 Add dwarf2_where call. Remote automatic alignment for ENTRY
680 instructions. Move call to xtensa_clear_insn_labels to the end.
681 Rearrange to handle bundles.
682 (xtensa_cons_fix_new): Delete.
683 (xtensa_handle_align): New.
684 (xtensa_frag_init): Call xtensa_set_frag_assembly_state. Remove
685 assignment to is_no_density field.
686 (md_pcrel_from): Use new xtensa-isa.h functions. Use decode_reloc
687 instead of reloc_to_opnum. Handle "alternate" relocations.
688 (xtensa_force_relocation, xtensa_check_inside_bundle,
689 xtensa_elf_section_change_hook): New.
690 (xtensa_symbol_new_hook): Delete.
691 (xtensa_fix_adjustable): Check for difference of symbols with an
692 offset. Check for external and weak symbols.
693 (md_apply_fix3): Remove cases for XTENSA_OP{0,1,2} relocs.
694 (md_estimate_size_before_relax): Return expansion for the first slot.
695 (tc_gen_reloc): Handle difference of symbols by producing
696 XTENSA_DIFF{8,16,32} relocs and by writing the value of the difference
697 into the output. Handle new XTENSA_SLOT*_OP relocs by storing the
698 tentative values into the output when linkrelax is set.
699 (XTENSA_PROP_SEC_NAME): Define.
700 (xtensa_post_relax_hook): Call xtensa_find_unmarked_state_frags.
701 Create literal tables only if using literal sections. Create new
702 property tables instead of old instruction tables. Check for unaligned
703 branch targets and loops.
704 (finish_vinsn, find_vinsn_conflicts, check_t1_t2_reads_and_writes,
705 new_resource_table, clear_resource_table, resize_resource_table,
706 resources_available, reserve_resources, release_resources,
707 opcode_funcUnit_use_unit, opcode_funcUnit_use_stage,
708 resources_conflict, xg_find_narrowest_format, relaxation_requirements,
709 bundle_single_op, emit_single_op, xg_assemble_vliw_tokens): New.
710 (xtensa_end): Call xtensa_flush_pending_output. Set past_xtensa_end
711 flag. Update checks for workaround options. Call
712 xtensa_mark_narrow_branches and xtensa_mark_zcl_first_insns.
713 (xtensa_cleanup_align_frags): Add special case for branch targets.
714 Check for and mark unreachable frags.
715 (xtensa_fix_target_frags): Remove use of align_only_targets flag.
716 Use RELAX_LOOP_END_BYTES in special case for negatable branch at the
717 end of a zero-overhead loop body.
718 (frag_can_negate_branch): Handle instructions with multiple slots.
719 Use new xtensa-isa.h functions
720 (xtensa_mark_narrow_branches, is_narrow_branch_guaranteed_in_range,
721 xtensa_mark_zcl_first_insns): New.
722 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags): Error if
723 transformations are disabled.
724 (next_instrs_are_b_retw): Use new xtensa-isa.h functions. Handle
725 multislot instructions.
726 (xtensa_fix_close_loop_end_frags, xtensa_fix_short_loop_frags):
727 Likewise. Also error if transformations are disabled.
728 (unrelaxed_frag_max_size): New.
729 (unrelaxed_frag_min_insn_count, unrelax_frag_has_b_j): Use new
730 xtensa-isa.h functions.
731 (xtensa_sanity_check, is_empty_loop, is_local_forward_loop): Use
732 xtensa_opcode_is_loop instead of is_loop_opcode.
733 (get_text_align_power): Replace as_fatal with assertion.
734 (get_text_align_fill_size): Iterate instead of using modulus when
736 (get_noop_aligned_address): Assert that this is for a machine-dependent
737 RELAX_ALIGN_NEXT_OPCODE frag. Use next_frag_opcode_is_loop,
738 xg_get_single_size, and frag_format_size.
739 (get_widen_aligned_address): Rename to ...
740 (get_aligned_diff): ... this function. Add max_diff parameter.
741 Remove handling of rs_align/rs_align_code frags. Use
742 next_frag_format_size, get_text_align_power, get_text_align_fill_size,
743 next_frag_opcode_is_loop, and xg_get_single_size. Compute max_diff
744 and pass it back to caller.
745 (xtensa_relax_frag): Use relax_frag_loop_align. Add code for new
746 RELAX_SLOTS, RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN,
747 RELAX_FILL_NOP, and RELAX_UNREACHABLE frag types. Check relax_seen.
748 (relax_frag_text_align): Rename to ...
749 (relax_frag_loop_align): ... this function. Assume loops can only be
750 in the first slot of an instruction.
751 (relax_frag_add_nop): Use assemble_nop instead of constructing an OR
752 instruction. Remove call to frag_wane.
753 (relax_frag_narrow): Rename to ...
754 (relax_frag_for_align): ... this function. Extend to handle
755 RELAX_FILL_NOP and RELAX_UNREACHABLE, as well as RELAX_SLOTS with
756 RELAX_NARROW for the first slot.
757 (find_address_of_next_align_frag, bytes_to_stretch): New.
758 (future_alignment_required): Use find_address_of_next_align_frag and
759 bytes_to_stretch. Look ahead to subsequent frags to make smarter
761 (relax_frag_immed): Add format, slot, and estimate_only parameters.
762 Check if transformations are enabled for b_j_loop_end workaround.
763 Use new xtensa-isa.h functions and handle multislot instructions.
764 Update call to xg_assembly_relax.
765 (md_convert_frag): Handle new RELAX_SLOTS, RELAX_UNREACHABLE,
766 RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN, and RELAX_FILL_NOP
768 (convert_frag_narrow): Add segP, format and slot parameters. Call
769 convert_frag_immed for branch instructions. Adjust calls to
770 tinsn_from_chars, tinsn_immed_from_frag, and xg_emit_insn_to_buf. Use
771 xg_get_single_size and xg_get_single_format.
772 (convert_frag_fill_nop): New.
773 (convert_frag_immed): Add format and slot parameters. Handle multislot
774 instructions and use new xtensa-isa.h functions. Update calls to
775 tinsn_immed_from_frag and xg_assembly_relax. Check if transformations
776 enabled for b_j_loop_end workaround. Use build_nop instead of
777 assemble_nop. Check is_specific_opcode flag. Check for unreachable
778 frags. Use xg_get_single_size. Handle O_pltrel.
779 (fix_new_exp_in_seg): Remove check for old plt flag.
780 (convert_frag_immed_finish_loop): Update calls to tinsn_from_chars and
781 xtensa_insnbuf_to_chars. Call tinsn_immed_from_frag. Change check
782 for loop opcode to an assertion. Mark all frags up to the end of the
783 loop as not transformable.
784 (get_last_insn_flags, set_last_insn_flags): Use get_subseg_info.
785 (get_subseg_info): New.
786 (xtensa_move_literals): Call xtensa_set_frag_assembly_state. Add null
788 (xtensa_switch_to_literal_fragment): Rewrite to handle absolute
789 literals and use xtensa_switch_to_non_abs_literal_fragment otherwise.
790 (xtensa_switch_to_non_abs_literal_fragment): New.
791 (cache_literal_section): Add is_code parameter and pass it through to
792 retrieve_literal_seg.
793 (retrieve_literal_seg): Add is_code parameter and use it to set the
794 flags on the literal section. Handle case where head parameter is 0.
795 (get_frag_is_no_transform, set_frag_is_specific_opcode,
796 set_frag_is_no_transform): New.
797 (xtensa_create_property_segments): Add end_property_function parameter
798 and pass it through to add_xt_block_frags. Call bfd_get_section_flags
799 and skip SEC_DEBUGGING and !SEC_ALLOC sections.
800 (xtensa_create_xproperty_segments, section_has_xproperty): New.
801 (add_xt_block_frags): Add end_property_function parameter and call it
802 if it is non-zero. Call xtensa_frag_flags_init.
803 (xtensa_frag_flags_is_empty, xtensa_frag_flags_init,
804 get_frag_property_flags, frag_flags_to_number,
805 xtensa_frag_flags_combinable, xt_block_aligned_size,
806 xtensa_xt_block_combine, add_xt_prop_frags,
807 init_op_placement_info_table, opcode_fits_format_slot,
808 xg_get_single_size, xg_get_single_format): New.
809 (istack_push): Inline call to tinsn_copy.
810 (tinsn_copy): Delete.
811 (tinsn_has_invalid_symbolic_operands): Handle O_hi16 and O_lo16 and
812 CONST16 opcodes. Handle O_big, O_illegal, and O_absent.
813 (tinsn_has_complex_operands): Handle O_hi16 and O_lo16.
814 (tinsn_to_insnbuf): Use xg_get_single_format and new xtensa-isa.h
815 functions. Handle invisible operands.
816 (tinsn_to_slotbuf): New.
817 (tinsn_check_arguments): Use new xtensa-isa.h functions.
818 (tinsn_from_chars): Add slot parameter. Rewrite using xg_init_vinsn,
819 vinsn_from_chars, and xg_free_vinsn.
820 (tinsn_from_insnbuf): New.
821 (tinsn_immed_from_frag): Add slot parameter and handle multislot
822 instructions. Handle symbol differences.
823 (get_num_stack_text_bytes): Use xg_get_single_size.
824 (xg_init_vinsn, xg_clear_vinsn, vinsn_has_specific_opcodes,
825 xg_free_vinsn, vinsn_to_insnbuf, vinsn_from_chars, expr_is_register,
826 get_expr_register, set_expr_symbol_offset_diff): New.
827 * config/tc-xtensa.h (MAX_SLOTS): Define.
828 (xtensa_relax_statesE): Move from tc-xtensa.c. Add
829 RELAX_CHECK_ALIGN_NEXT_OPCODE, RELAX_MAYBE_DESIRE_ALIGN, RELAX_SLOTS,
830 RELAX_FILL_NOP, RELAX_UNREACHABLE, RELAX_MAYBE_UNREACHABLE, and
832 (RELAX_IMMED_MAXSTEPS): Move from tc-xtensa.c.
833 (xtensa_frag_type struct): Add is_assembly_state_set,
834 use_absolute_literals, relax_seen, is_unreachable, is_specific_opcode,
835 is_align, is_text_align, alignment, and is_first_loop_insn fields.
836 Replace is_generics and is_relax fields by is_no_transform field.
837 Delete is_text and is_longcalls fields. Change text_expansion and
838 literal_expansion to arrays of MAX_SLOTS entries. Add arrays of
839 per-slot information: literal_frags, slot_subtypes, slot_symbols,
840 slot_sub_symbols, and slot_offsets. Add fr_prev field.
841 (xtensa_fix_data struct): New.
842 (xtensa_symfield_type struct): Delete plt field.
843 (xtensa_block_info struct): Move definition to tc-xtensa.h. Add
844 forward declaration here.
845 (xt_section_type enum): Delete xt_insn_sec. Add xt_prop_sec.
846 (XTENSA_SECTION_RENAME): Undefine.
847 (TC_FIX_TYPE, TC_INIT_FIX_DATA, TC_FORCE_RELOCATION, NO_PSEUDO_DOT,
848 tc_unrecognized_line, md_do_align, md_elf_section_change_hook,
849 HANDLE_ALIGN, TC_LINKRELAX_FIXUP, SUB_SEGMENT_ALIGN): Define.
850 (TC_CONS_FIX_NEW, tc_symbol_new_hook): Delete.
851 (unit_num_copies_func, opcode_num_units_func,
852 opcode_funcUnit_use_unit_func, opcode_funcUnit_use_stage_func): New.
853 (resource_table struct): New.
854 * config/xtensa-istack.h (MAX_INSN_ARGS): Increase from 6 to 10.
855 (TInsn struct): Add keep_wide, loc, fixup, record_fix, subtype,
856 literal_space, symbol, sub_symbol, offset, and literal_frag fields.
857 (tinsn_copy): Delete prototype.
858 (vliw_insn struct): New.
859 * config/xtensa-relax.c (insn_pattern_struct): Add options field.
860 (widen_spec_list): Add option conditions for density and boolean
861 instructions. Add expansions using CONST16 and conditions for using
862 CONST16 vs. L32R. Use new Xtensa LX RSR/WSR syntax. Add entries for
864 (simplify_spec_list): Add option conditions for density instructions.
865 Add entry for NOP instruction.
866 (append_transition): Add cmp function pointer parameter and use it to
867 insert the new entry in order.
868 (operand_function_LOW16U, operand_function_HI16U): New.
869 (xg_has_userdef_op_fn, xg_apply_userdef_op_fn): Handle
870 OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
871 (enter_opname, split_string): Use xstrdup instead of strdup.
872 (init_insn_pattern): Initialize new options field.
873 (clear_req_or_option_list, clear_req_option_list,
874 clone_req_or_option_list, clone_req_option_list, parse_option_cond):
876 (parse_insn_pattern): Parse option conditions.
877 (transition_applies): New.
878 (build_transition): Use new xtensa-isa.h functions. Fix incorrectly
879 swapped last arguments in calls to append_constant_value_condition.
880 Call clone_req_option_list. Add warning about invalid opcode.
881 Handle LOW16U and HI16U function names.
882 (build_transition_table): Add cmp parameter and use it in calls to
883 append_transition. Use new xtensa-isa.h functions. Check
884 transition_applies before adding entries.
885 (xg_build_widen_table, xg_build_simplify_table): Add cmp parameter and
886 pass it through to build_transition_table.
887 * config/xtensa-relax.h (ReqOrOptionList, ReqOrOption, ReqOptionList,
888 ReqOption, transition_cmp_fn): New types.
889 (OpType enum): Add OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
890 (transition_rule struct): Add options field.
891 * doc/as.texinfo (Overview): Update Xtensa options.
892 * doc/c-xtensa.texi (Xtensa Options): Delete --[no-]density,
893 --[no-]relax, and --[no-]generics options. Update descriptions of
894 --text-section-literals and --[no-]longcalls. Add
895 --[no-]absolute-literals and --[no-]transform.
896 (Xtensa Syntax): Add description of syntax for FLIX instructions.
897 Remove use of "generic" and "specific" terminology for opcodes.
898 (Xtensa Registers): Generalize the syntax description to include
899 user-defined register files.
900 (Xtensa Automatic Alignment): Update.
901 (Xtensa Branch Relaxation): Mention limitation of unconditional jumps.
902 (Xtensa Call Relaxation): Linker can now remove most of the overhead.
903 (Xtensa Directives): Remove confusing rules about precedence.
904 (Density Directive, Relax Directive): Delete.
905 (Schedule Directive): New.
906 (Generics Directive): Rename to ...
907 (Transform Directive): ... this node.
908 (Literal Directive): Update for absolute literals. Missing
909 literal_position directive is now an error.
910 (Literal Position Directive): Update for absolute literals.
911 (Freeregs Directive): Delete.
912 (Absolute Literals Directive): New.
913 (Frame Directive): Minor editing.
914 * Makefile.am (DEPTC_xtensa_elf, DEPOBJ_xtensa_elf, DEP_xtensa_elf):
916 * Makefile.in: Regenerate.
918 2004-10-07 Richard Sandiford <rsandifo@redhat.com>
920 * config/tc-mips.c (append_insn): Use fix_new rather than fix_new_exp
921 to build the second and third fixups for a composite relocation.
922 (macro_read_relocs): New function.
923 (macro_build): Use it.
924 (s_cpsetup): Pass all three composite relocation codes to macro_build.
925 Simplify fragging code accordingly.
926 (s_gpdword): Use fix_new rather than fix_new_exp for the second part
927 of the composite relocation. Set fx_tcbit in both fixups.
929 2004-10-07 Richard Sandiford <rsandifo@redhat.com>
931 * config/tc-mips.c (append_insn): Set fx_tcbit for composite relocs.
932 (md_apply_fix3): Don't treat composite relocs as done.
934 2004-10-07 Jan Beulich <jbeulich@novell.com>
936 * macro.c (macro_expand_body): When ELF, use .LL rather than LL as
937 prefix for symbol names generated from the LOCAL macro directive.
939 * dw2gencfi.c (select_cie_for_fde): When separating CIE out from
940 FDE, treat a DW_CFA_remember_state as we do a DW_CFA_advance_loc.
942 2004-10-07 Tomer Levi <Tomer.Levi@nsc.com>
944 * config/tc-crx.c (preprocess_reglist): Handle Co-processor
946 (md_assemble): Add error checking for Co-Processor instructions.
947 (get_cinv_parameters): Add 'b' option to invalidate the
950 2004-10-05 Paul Brook <paul@codesourcery.com>
952 * config/tc-arm.c (unwind): New variable.
953 (vfp_sp_encode_reg): New function.
954 (vfp_sp_reg_required_here): Use it.
955 (vfp_sp_reg_list, vfp_dp_reg_list): Remove.
956 (vfp_parse_reg_list): New function.
957 (s_arm_unwind_fnstart, s_arm_unwind_fnend, s_arm_unwind_cantunwind,
958 s_arm_unwind_personality, s_arm_unwind_personalityindex,
959 s_arm_unwind_handlerdata, s_arm_unwind_save, s_arm_unwind_movsp,
960 s_arm_unwind_pad, s_arm_unwind_setfp, s_arm_unwind_raw): New
962 (md_pseudo_table): Add them.
963 (do_vfp_reg2_from_sp2): Use vfp_parse_reg_list and vfp_sp_encode_reg.
964 (do_vfp_sp2_from_reg2, vfp_sp_ldstm, vfp_dp_ldstm): Ditto.
965 (set_section, add_unwind_adjustsp, flush_pending_unwind,
966 finish_unwind_opcodes, start_unwind_section, create_unwind_entry,
967 require_hashconst, add_unwind_opcode): New functions.
968 * doc/tc-arm.text: Document unwinding opcodes.
969 * NEWS: Mention the new feature.
971 2004-10-04 Eric Christopher <echristo@redhat.com>
973 * config/tc-mips.c (md_apply_fix3): Remove erroneous assert.
975 2004-10-01 H.J. Lu <hongjiu.lu@intel.com>
977 * config/tc-ppc.c (md_apply_fix3): Call S_SET_THREAD_LOCAL for
979 * config/tc-s390.c (md_apply_fix3): Likewise.
980 * config/tc-sparc.c (md_apply_fix3): Likewise.
982 2004-10-01 Paul Brook <paul@codesourcery.com>
984 * config/tc-arm.c (arm_elf_section_type): New function.
985 (arm_elf_change_section): Set section link for exidx sections.
986 * config/tc-arm.h (arm_elf_section_type): Add prototype.
987 (md_elf_section_type): Define.
989 2004-10-01 Bill Farmer <Bill@the-farmers.freeserve.co.uk>
991 * config/tc-pdp11.c (md_apply_fix3): Change to sign of the SOB
992 instruction's offset.
994 2004-10-01 Adam Nemet <anemet@lnxw.com>
996 * (TARGET_FORMAT): Remove LynxOS COFF definition.
998 2004-10-01 Ravi Ramaseshan <ravi.ramaseshan@codito.com>
1000 * config/tc-arc.c (tc_gen_reloc): Don't assume fixP->fx_addsy is an
1001 asymbol *, instead use symbol_get_bfdsym.
1003 2004-09-30 Linus Nielsen Feltzing <linus@haxx.se>
1005 * config/tc-m68k.c (select_control_regs): Add mcf5249.
1007 2004-09-30 Paul Brook <paul@codesourcery.com>
1009 * config/tc-arm.c (do_smi, do_nop): New functions.
1010 (insns): Add ARMv6ZK instructions.
1011 (md_apply_fix3): Handle BFD_RELOC_ARM_SMI.
1012 (tc_gen_reloc): Ditto.
1013 (arm_cpus): Add mpcore and arm1176.
1014 (arm_archs): Add armv6{k,z,zk}.
1015 * doc/c-arm.texi: Document new cores and architectures.
1017 2004-09-30 Nick Clifton <nickc@redhat.com>
1019 * config/tc-arm.c: Use ISO C90 formatting.
1021 2004-09-30 Vladimir Ivanov <vladitx@nucleusys.com>
1023 * config/tc-arm.c (mav_reg_required_here): Allow REG_TYPE_CN
1024 as alternative when REG_TYPE_MVF, REG_TYPE_MVD, REG_TYPE_MVFX or
1025 REG_TYPE_MVDX is expected.
1027 2004-09-29 Marc Bevand <m.bevand@gmail.com>
1029 * doc/c-i386.texi (i386-Mnemonics): Fix typo.
1031 2004-09-21 James E Wilson <wilson@specifixinc.com>
1033 * config/tc-ia64.c (ENCODED_PSP_OFFSET): New.
1034 (output_rp_psprel, output_pfs_psprel, output_preds_psprel,
1035 output_spill_base, output_unat_psprel, output_lc_psprel,
1036 output_fpsr_psprel, output_priunat_psprel, output_bsp_psprel,
1037 output_bsprestore_psprel, output_rnat_psprel, output_spill_psprel,
1038 output_spill_psprel_p): Use it.
1040 2004-09-20 Tomer Levi <Tomer.Levi@nsc.com>
1042 * config/tc-crx.c (handle_LoadStor): New function.
1043 Handle load/stor unique instructions before parsing.
1045 2004-09-17 Paul Brook <paul@codesourcery.com>
1047 * config/tc-arm.c (s_arm_rel31): New funciton.
1048 (md_pseudo_table): Add .rel31.
1049 (md_apply_fix3): Handle BFD_RELOC_ARM_TARGET2,
1050 BFD_RELOC_32_PCREL and BFD_RELOC_ARM_PREL31.
1051 (tc_gen_reloc): Handle BFD_RELOC_ARM_PREL31 and BFD_RELOC_ARM_TARGET2.
1052 (arm_fix_adjustable): Return 0 for BFD_RELOC_ARM_TARGET2.
1053 (arm_parse_reloc): Add (target2).
1055 2004-09-17 Alan Modra <amodra@bigpond.net.au>
1057 * Makefile.am: Run "make dep-am".
1058 * Makefile.in: Regenerate.
1059 * aclocal.m4: Regenerate.
1060 * configure: Regenerate.
1061 * doc/Makefile.in: Regenerate.
1062 * po/POTFILES.in: Regenerate.
1063 * po/gas.pot: Regenerate.
1065 2004-09-14 Hideki IWAMOTO <h-iwamoto@kit.hi-ho.ne.jp>
1067 * config/tc-mmix.c [!LLONG_MIN]: Correct #elsif to #elif.
1068 [!LLONG_MAX]: Ditto.
1070 2004-09-13 Paul Brook <paul@codesourcery.com>
1072 * config/tc-arm.c: Rename RELABS to TARGET1.
1074 2004-09-13 Alan Modra <amodra@bigpond.net.au>
1076 * messages.c (as_internal_value_out_of_range): Cast values passed
1077 to as_bad_where or as_warn_where to proper type.
1079 2004-09-11 Theodore A. Roth <troth@openavr.org>
1081 * config/tc-avr.c: Add support for
1082 atmega48, atmega88, atmega168, attiny13, attiny2313, at90can128.
1084 2004-09-09 Alan Modra <amodra@bigpond.net.au>
1086 * dw2gencfi.c (select_cie_for_fde): When separating CIE out
1087 from FDE, treat a CFI_escape as we do a DW_CFA_advance_loc.
1089 2004-09-08 Paul Brook <paul@codesourcery.com>
1091 * config/obj-elf.c (obj_elf_section_type): Handle init_array,
1092 fini_array and preinit_array section types.
1093 * config/tc-ia64.c (ia64_elf_section_type): Remove init_array
1095 * doc/as.texinfo: Document extra section types.
1097 2004-09-02 Mark Mitchell <mark@codesourcery.com>
1099 * Makefile.am (TARG_ENV_HFILES): Add te-symbian.h.
1100 * Makefile.in: Regenerated.
1101 * configure.in: Set em for arm*-*-symbianelf*.
1102 * configure: Regenerated.
1103 * config/tc-arm.c (elf32_arm_target_format): Use Symbian target
1104 vectors when appropriate.
1105 * config/te-symbian.h: New file.
1107 2004-09-03 Tomer Levi <Tomer.Levi@nsc.com>
1109 * config/tc-crx.c (gettrap): Exception vector can be case
1111 (process_label_constant): Fix a 32-bit displacement bug in branch
1113 (get_operandtype) : Bug fix, wrong operand was used.
1114 (process_label_constant): Initialize relocation type to
1117 2004-09-01 Richard Earnshaw < reanrsha@arm.com>
1119 * tc-arm.c (arm_cpus, arm_fpus): Allow <cpu>-s as well as <cpu>s
1120 for synthesizable cores.
1122 * doc/c-arm.texi (ARM Options): Document canonical names of CPUs.
1124 2004-08-25 Dmitry Diky <diwil@spec.ru>
1126 * config/tc-msp430.c: Clean-up the code.
1127 (md_relax_table): New relax table.
1128 (mcu_types): Sort MCU types.
1129 (md_pseudo_table): Add .profiler pseudo handler.
1130 (pow2value): New function.
1131 (msp430_profiler): New function.
1132 (msp430_operands): Add new insns handlers.
1133 (msp430_srcoperand): Add register operand handler, allow complex
1135 (md_estimate_size_before_relax): Rewritten.
1136 (md_convert_frag): Rewritten.
1137 (msp430_relax_frag): New function.
1138 * config/tc-msp430.h (md_relax_frag): define macro
1139 * doc/c-msp430.texi: Update information.
1141 2004-08-24 Nick Clifton <nickc@redhat.com>
1143 * as.c (std_shortopts): Allow -g to take an optional argument.
1144 (parse_args): Pass any switch starting with -g on to the backend
1147 2004-08-18 Mark Mitchell <mark@codesourcery.com>
1149 * configure.in (arm*-*-symbianelf*): New target.
1150 (arm*-*-eabi*): Likewise.
1151 * configure: Regenerated.
1153 2004-08-18 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1154 * config/tc-mips.c (append_insn): Handle delay slots in branch likely
1157 2004-08-18 Jakub Jelinek <jakub@redhat.com>
1159 * config/tc-ia64.c (start_unwind_section): Add linkonce_empty
1160 argument, don't do anything if current section is not
1161 .gnu.linkonce.t.* and linkonce_empty is set.
1162 (generate_unwind_image, dot_endp): Adjust callers, call
1163 start_unwind_section (*, 1) if nothing will be put into the
1166 2004-08-17 Nick Clifton <nickc@redhat.com>
1168 * as.c (MD_DEBUG_FORMAT_SELECTOR): Provide default definition.
1169 (show_usage): Add -g.
1170 (std_longopts): Add --gen-debug. Alpha sort the table.
1171 (parse_args): Print an error message if a switch is not handled.
1172 Handle the -g switch, calling md_debug_format_selector() if
1174 * NEWS: Mention new feature.
1175 * doc/as.texinfo: Document new switch.
1176 * doc/internals.texi: Document behaviour of md_parse_option.
1178 * config/tc-arm.c (md_parse_option): Do not issue an error message
1179 if the switch is not recognised.
1180 * config/tc-m68k.c (md_parse_option): Likewise.
1181 * config/tc-pdp11.c (md_parse_option): Likewise.
1182 * config/tc-v850.c (md_parse_option): Likewise.
1184 * as.h: Fix up formatting.
1187 2004-08-16 Nick Clifton <nickc@redhat.com>
1189 * macro.c (macro_set_alternate): Use ISO C90 formatting.
1191 * configure.in: Sort architecture based tables alphabetically.
1192 * configure: Regenerate.
1194 2004-08-16 Alan Modra <amodra@bigpond.net.au>
1196 * config/tc-ppc.c (tc_ppc_regname_to_dw2regnum <regnames>): Replace
1197 { "cc", 68 }, with { "cr", 70 }.
1199 2004-08-13 Jan Beulich <jbeulich@novell.com>
1200 Nick Clifton <nickc@redhat.com>
1202 * as.c: Add and handle new --alternate command line option.
1203 * macro.c (macro_set_alternate): New.
1204 * macro.h (macro_set_alternate): Declare.
1205 * read.c: Add and handle new .altmacro and .noaltmacro directives.
1206 * doc/as.texinfo: Document new command line option and pseudo-ops
1207 as well as insert documentation originating from gasp about
1208 alternate macro syntax.
1209 * NEWS: Mention new command line option and pseudo-ops.
1211 2004-08-10 Mark Mitchell <mark@codesourcery.com>
1213 * expr.c (operand): Handle the "~", "-", and "!" operators applied
1216 2004-08-06 Paul Brook <paul@codesourcery.com>
1218 * config/tc-arm.c (md_apply_fix3, tc_gen_reloc, arm_parse_reloc):
1219 Handle new relocations.
1220 * include/elf/arm.h (elf_arm_reloc_type): Add new EABI relocations.
1222 2004-08-05 Bob Wilson <bob.wilson@acm.org>
1224 * write.c (relax_segment): Use was_address instead of address when
1225 setting fr_fix field for align frag due to backwards .org.
1227 2004-07-29 Alexandre Oliva <aoliva@redhat.com>
1229 Introduce SH2a support.
1230 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
1231 * config/tc-sh.c (get_specific): Change arch_sh2a_up to
1233 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
1234 * config/tc-sh.c (md_parse_option): Add sh2a-nofpu ISA handling.
1235 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
1236 * config/tc-sh.c (sh_elf_final_processing): Move sh2a recognition
1237 to end of conditional expression.
1238 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
1239 * config/tc-sh.c: Add sh2a-nofpu support.
1240 2003-12-29 DJ Delorie <dj@redhat.com>
1241 * tc-sh.c: Add sh2a support.
1242 (parse_reg): Add tbr.
1243 (parse_at): Support @@(disp,tbr).
1244 (get_specific): Support sh2a opcodes.
1245 (insert4): New, for 4 byte relocs.
1246 (build_Mytes): Support sh2a opcodes.
1247 (md_apply_fix3_Mytes): Support sh2a opcodes.
1248 2003-12-02 Michael Snyder <msnyder@redhat.com>
1249 * config/tc-sh.c (md_parse_option): Handle sh2a.
1250 (sh_elf_final_processing): Ditto.
1252 2004-07-27 Jason Thorpe <thorpej@wasabisystems.com>
1254 * config/tc-hppa.h (TARGET_FORMAT): Set to "elf32-hppa-netbsd"
1257 2004-07-27 Alan Modra <amodra@bigpond.net.au>
1259 * config/tc-ppc.c (ppc_frob_file_before_adjust): Warn if .toc too big.
1260 (ppc_arch): Expand comment.
1262 2004-07-27 Tomer Levi <Tomer.Levi@nsc.com>
1264 * config/tc-crx.c: Support evaluating the difference between two
1266 * config/tc-crx.h: Likewise.
1268 2004-07-26 H.J. Lu <hongjiu.lu@intel.com>
1270 * config/tc-ia64.c (start_unwind_section): Set the linked-to
1272 (ia64_elf_section_change_hook): Set the linked-to section for
1275 2004-07-26 Dmitry Diky <diwil@spec.ru>
1277 * config/tc-msp430.c: Add new subtargets: msp430x1610,
1278 msp430x1611, msp430x1612, msp430x415, msp430x417, msp430xG437,
1279 msp430xG438, msp430xG439.
1281 2004-07-25 Daniel Jacobowitz <dan@debian.org>
1283 * doc/as.texinfo (Section, PushSection): Correct documentation
1286 2004-07-21 Jan Beulich <jbeulich@novell.com>
1288 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
1289 permissible for the selected instruction suffix.
1290 (match_template): Don't permit 64-bit general purpose operands in
1292 (finalize_imm): Permit 64-bit immediates.
1293 (build_modrm_byte): Don't treat 32-bit addressing in 64-bit mode
1294 specially except for the width of the used base and/or index
1295 registers. For 32-bit displacements, use sign-extended
1296 relocations only when using 64-bit addressing.
1297 Force zero displacement on rip-relative addressing when there is
1298 no other displacement.
1299 (i386_index_check): Don't treat 32-bit addressing in 64-bit mode
1300 specially except for the width of the used base and/or index
1302 (parse_register): Disallow Reg64 registers in 32-bit mode.
1304 * config/tc-i386.c: For DefaultSize instructions, don't guess a 'q'
1305 suffix if the instruction doesn't support it.
1307 2004-07-20 Maciej W. Rozycki <macro@linux-mips.org>
1309 * config/tc-mips.c (append_insn): Handle constant expressions with
1310 no associated relocation.
1311 (mips_ip): Cancel the expression after use for the Q format
1313 (parse_relocation): Return no relocation for unsupported
1315 (my_getSmallExpression): Return no relocation if no relocation
1318 2004-07-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1320 * config/obj-som.c (adjust_stab_sections): Add prototype.
1321 (obj_som_compiler, obj_som_version, obj_som_copyright,
1322 adjust_stab_sections): Add ATTRIBUTE_UNUSED to unused arguments.
1323 * config/tc-hppa.c (update_subspace): Likewise.
1324 (is_defined_subspace): Amplify comment.
1325 * config/obj-som.h (som_frob_file): Add prototype.
1327 2004-07-19 Christopher Faylor <cgf@timesys.com>
1328 H.J. Lu <hongjiu.lu@intel.com>
1330 * subsegs.c (section_symbol): Don't create a new segment when
1331 existing segment is undefined.
1333 2004-07-16 Richard Earnshaw <rearnsha@arm.com>
1335 * config/tc-arm.c: Include include/opcode/arm.h.
1336 (ARM_EXT_*, ARM_ARCH_*, ARM_ANY, ARM_ALL, COPROC_ANY): Delete.
1337 (FPU_FPA_EXT_* FPU_VFP_EXT_*, FPU_ANY, FPU_NONE, FPU_MAVERICK): Delete.
1338 (FPU_ARCH_*): Delete.
1339 * Makefile.am: Update dependencies.
1340 * Makefile.in: Regenerate.
1342 2004-07-15 Nick Clifton <nickc@redhat.com>
1344 * configure.in: Accept armbe as a big-endian arm configuration.
1345 * configure: Regenerate.
1347 2004-07-13 Thomas Nystrom <thn@saeab.se>
1349 * config/tc-i386.c (T_SHIFTOP): New constant.
1350 (intel_e05_1): Handle '&', '|' and T_SHIFTOP.
1351 (intel_el1): Handle '~'.
1352 (intel_get_token): Handle '<>', '&', '|' and '~'.
1354 2004-07-13 Nick Clifton <nickc@redhat.com>
1356 (md_assemble): Remove spurious newline from end of as_bad error
1358 (intel_e05_1): Likewise.
1359 (intel_e11): Likewise.
1360 (intel_match_token): Likewise.
1362 2004-07-11 Andreas Schwab <schwab@suse.de>
1364 * config/tc-m68k.c: Convert to C90. Remove redundant
1365 declarations. Indentation fixup.
1366 [M68KCOFF]: Include "obj-coff.h" instead of declaring
1367 obj_coff_section ourselves.
1369 2004-07-09 James E Wilson <wilson@specifixinc.com>
1371 * config/tc-ia64.c (default_big_endian): New.
1372 (dot_byteorder, md_begin): Use it.
1373 (md_parse_option): Set it.
1375 2004-07-09 Nick Clifton <nickc@redhat.com>
1377 * configure.in: Change sh-sybmian-elf to sh-*-symbianelf.
1378 * configure: Regenerate.
1379 * NEWS: Change sh-sybmian-elf to sh-*-symbianelf.
1380 * config/tc-sh.c (sh_elf_final_processing): Use renamed version of
1381 sh_find_elf_flags if necessary.
1383 2004-07-08 Richard Sandiford <rsandifo@redhat.com>
1385 * config/tc-mips.c (mips_fix_adjustable): If the full addend is
1386 going to be split into more than one in-place addend, return 0
1387 for relocations against mergeable sections. Associate comments
1390 2004-07-07 Tomer Levi <Tomer.Levi@nsc.com>
1392 * Makefile.am (CPU_TYPES): Add crx.
1393 (TARGET_CPU_CFILES): Add config/tc-crx.c.
1394 (TARGET_CPU_HFILES): Add config/tc-crx.h.
1395 (DEPTC_crx_elf): New target.
1396 (DEPOBJ_crx_elf): Likewise.
1397 (DEP_crx_elf): Likewise.
1398 * Makefile.in: Regenerate.
1399 * configure.in: Add crx* target.
1400 * configure: Regenerate.
1401 * config/tc-crx.c: New file.
1402 * config/tc-crx.h: New file.
1403 * NEWS: Mention new target.
1405 2004-07-06 Nick Clifton <nickc@redhat.com>
1407 * config.in: Undefine TARGET_SYMBIAN by default.
1409 * configure: Regenerate. Add sh-symbian-elf target. If
1410 selected define TARGET_SYMBIAN.
1411 * config/tc-sh.h (TARGET_FORMAT): Select a Symbian target
1412 format if TARGET_SYMBIAN has been defined.
1414 * output-file.c (output_file_create): Report the target format
1415 chosen when bfd_openw reports that it is invalid.
1417 * config/obj-coff.c (coff_pseudo_table): Only define the weak
1418 pseudo for BFD based assemblers.
1420 2004-07-05 Andrew Stubbs <andrew.stubbs@superh.com>
1423 * config/tc-sh.c (md_assemble): Change isspace to ISSPACE.
1424 (md_parse_option): Remove redundant -isa testing.
1425 Make bfd_arch variable const.
1426 (md_show_usage): Make bfd_arch variable const.
1428 2004-07-03 James E Wilson <wilson@specifixinc.com>
1430 * config/tc-ia64.c (emit_one_bundle): Check and set insn_addr.
1431 * config/tc-ia64.h (md_frag_check): Define.
1433 2004-07-03 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
1435 * config/obj-coff.c (obj_coff_weak): New .weak syntax for PE weak
1437 * doc/as.texinfo (Weak): Document PE weak symbols.
1439 2004-07-03 Richard Sandiford <rsandifo@redhat.com>
1441 * config/tc-mips.c (HAVE_IN_PLACE_ADDENDS): New macro.
1442 (reloc_needs_lo_p): Only return true if HAVE_IN_PLACE_ADDENDS.
1443 (mips_frob_file): Rework so that only a single pass through the
1444 relocs is needed. Allow %lo()s to have higher offsets than their
1445 corresponding %hi()s or %got()s.
1447 2004-07-02 Nick Clifton <nickc@redhat.com>
1449 * config/tc-arm.c (md_apply_fix3:BFD_RELOC_ARM_IMMEDIATE): Do not
1450 allow values which have come from undefined symbols.
1451 Always consider this fixup to have been processed as a reloc
1452 cannot be generated for it.
1454 2004-07-02 Alan Modra <amodra@bigpond.net.au>
1456 * frags.h (struct frag): Add has_code and insn_addr fields.
1457 * write.c (cvt_frag_to_fill): Invoke md_frag_check.
1458 * config/tc-ppc.c (md_assemble): Check and set insn_addr.
1459 * config/tc-ppc.h (md_frag_check): Define.
1461 2004-06-28 Maciej W. Rozycki <macro@linux-mips.org>
1463 * doc/Makefile.am (info): Rename goal to...
1464 (info-local): ... this, to preserve implicit dependencies.
1465 * doc/Makefile.in: Regenerate with automake 1.8.5.
1467 2004-06-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1469 * config/tc-m32r.c (md_convert_frag): Changed for @PLT.
1470 (m32r_cgen_record_fixup_exp): Changed for @GOTOFF, @GOT.
1471 (m32r_fix_adjustable): Changed for @GOTOFF, @GOT, @PLT.
1472 (tc_gen_reloc): Likewise.
1473 (m32r_end_of_match): Add for @GOTOFF, @GOT, @PLT.
1474 (m32r_parse_name): Likewise.
1475 (m32r_cgen_parse_fix_exp): Likewise.
1476 * config/tc-m32r.h (md_parse_name): Define for @GOTOFF, @GOT, @PLT.
1477 (O_PIC_reloc): Likewise.
1478 (TC_CGEN_PARSE_FIX_EXP): Likewise..
1479 * cgen.c (gas_cgen_parse_operand): Add TC_CGEN_PARSE_FIX_EXP
1480 for @GOTOFF, @GOT, @PLT.
1482 2004-06-21 Jan Beulich <jbeulich@novell.com>
1484 * gas/symbols.c: While discarding ordinary local absolute symbols
1485 when --strip-local-absolute is in effect, retain file symbols.
1487 2004-06-20 Andreas Schwab <schwab@suse.de>
1489 * config/tc-m68k.c (mri_chip): Replace current_chip, not augment.
1490 (md_parse_option): Likewise.
1492 2004-06-17 Jan Beulich <jbeulich@novell.com>
1494 * config/tc-i386.c: Deal with LEX_QM the same way as with LEX_AT.
1495 * config/te-netware.h: New file.
1496 * config/te-ppcnw.h: Delete: Obsolete.
1497 * configure.in: Eliminate ill NetWare targets. Make generic
1498 NetWare target use proper emulation.
1499 * Makefile.am: Eliminate reference to obsolete te-ppcnw.h, add
1500 reference to new te-netware.h.
1501 * configure: Regenerate.
1502 * Makefile.in: Regenerate.
1504 2004-06-15 Martin Schwidefsky <schwidefsky@de.ibm.com>
1506 * config/tc-s390.c (s390_insn): Avoid incorrect signed/unsigned
1507 comparison in .insn pseudo operation.
1509 2004-06-15 Alan Modra <amodra@bigpond.net.au>
1511 * config/obj-coff.c (coff_adjust_section_syms): Use
1512 bfd_get_section_size instead of bfd_get_section_size_before_reloc.
1513 (coff_frob_section): Likewise.
1514 * config/tc-mips.c (md_apply_fix3): Likewise.
1515 * config/obj-elf.c (elf_frob_file): Use bfd_set_section_size.
1516 (elf_frob_file_after_relocs): Likewise.
1518 2004-06-10 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1520 * config/tc-hppa.c (log2): Rename to exact_log2.
1521 (pa_next_subseg): Delete unused function.
1522 (create_new_space): Mark unused arguments with ATTRIBUTE_UNUSED.
1523 (create_new_subspace): Likewise.
1526 * config/tc-hppa.c (hppa_fix_adjustable): Allow reduction of fake
1527 labels. Fix warning.
1529 2004-05-28 DJ Delorie <dj@redhat.com>
1531 * config/tc-mn10300.h (tc_fix_adjustable): Define.
1532 * config/tc-mn10300.c (mn10300_fix_adjustable): Don't adjust debug
1533 or non-merged symbols.
1535 2004-05-28 H.J. Lu <hongjiu.lu@intel.com>
1537 * config/tc-ia64.c (remove_marked_resource): Save, clear and
1538 restore the old slot when inserting srlz.i/srlz.d.
1540 2004-05-28 Andrew Stubbs <andrew.stubbs@superh.com>
1542 * Makefile.am: Regenerate dependecies.
1543 * Makefile.in: Regenerate.
1544 * config/tc-sh.c (valid_arch): Make unsigned.
1545 (preset_target_arch): Likewise.
1546 (md_begin): Use new architecture flags system.
1547 (get_specific): Likewise.
1548 (assemble_ppi): Likewise.
1549 (md_assemble): Likewise. Also fix error check for bad opcodes.
1550 (md_parse_option): Likewise. Also generate -isa values according
1551 to the table in bfd/cpu-sh.c instead of just constants. Also
1552 allow <arch>-up ISA variants.
1553 (sh_elf_final_processing): Replace if-else chain with a call to
1554 sh_find_elf_flags().
1555 * testsuite/gas/sh/arch: New directory.
1556 * testsuite/gas/sh/arch/arch.exp: New test script.
1557 * testsuite/gas/sh/arch/arch_expected.txt: New file.
1558 * testsuite/gas/sh/arch/sh.s: New file.
1559 * testsuite/gas/sh/arch/sh2.s: New file.
1560 * testsuite/gas/sh/arch/sh-dsp.s: New file.
1561 * testsuite/gas/sh/arch/sh2e.s: New file.
1562 * testsuite/gas/sh/arch/sh3-nommu.s: New file.
1563 * testsuite/gas/sh/arch/sh3.s: New file.
1564 * testsuite/gas/sh/arch/sh3-dsp.s: New file.
1565 * testsuite/gas/sh/arch/sh3e.s: New file.
1566 * testsuite/gas/sh/arch/sh4-nommu-nofpu.s: New file.
1567 * testsuite/gas/sh/arch/sh4-nofpu.s: New file.
1568 * testsuite/gas/sh/arch/sh4.s: New file.
1569 * testsuite/gas/sh/arch/sh4a-nofpu.s: New file.
1570 * testsuite/gas/sh/arch/sh4al-dsp.s: New file.
1571 * testsuite/gas/sh/arch/sh4a.s: New file.
1573 2004-05-28 Peter Barada <peter@the-baradas.com>
1575 * config/gc-m68k.c(m68k_ip): Convert mode 5 addressing
1576 with zero offset into mode 2 addressing to save a word.
1578 2004-05-27 H.J. Lu <hongjiu.lu@intel.com>
1580 * config/tc-ia64.c (ar_is_in_integer_unit): Removed.
1581 (ar_is_only_in_integer_unit): New.
1582 (ar_is_only_in_memory_unit): New.
1583 (generate_unwind_image): Silence gcc on 32bit host.
1584 (md_assemble): Use ar_is_only_in_integer_unit instead of
1585 ar_is_in_integer_unit. Check AR access.
1587 2004-05-27 Peter Barada <peter@the-baradas.com>
1589 * config/tc-m68k.c (md_begin): Sort the opcode table into
1591 (m68k_compare_opcode): New function to do the sorting.
1593 2004-05-24 Peter Barada <peter@the-baradas.com>
1595 * config/m68k-parse.y(operand): Allow for MAC/EMAC mask
1596 addressing on MIT style operands.
1597 * config/m68k-parse.y(yylex): Allow '-&' for predecrement
1599 * config/tc-m68k.c(install_operand): Comment 'G' and 'H' type
1602 2004-05-23 Alan Modra <amodra@bigpond.net.au>
1604 * expr.c (operand, operator): Don't reject '++' and '--'.
1606 2004-05-20 Richard Sandiford <rsandifo@redhat.com>
1608 * config/tc-mips.c (append_insn): Use ISA-encoded addresses in MIPS16
1611 2004-05-17 Adam Nemet <anemet@lnxw.com>
1613 * configure.in: Add ppc-*-lynxos*. Update i386-*-lynxos* to ELF.
1614 * configure: Regenerate.
1616 2004-05-13 Paul Brook <paul@codesourcery.com>
1618 * dw2gencfi.c (output_cie): Handle dwarf3 format CIE entries.
1620 2004-05-13 Joel Sherrill <joel@oarcorp.com>
1622 * configure.in (or32-*-rtems*): Switch to elf.
1623 * configure: Regenerate.
1625 2004-05-13 Nick Clifton <nickc@redhat.com>
1627 * po/fr.po: Updated French translation.
1629 2004-05-11 Nick Clifton <nickc@redhat.com>
1631 * doc/as.texinfo (Section): Document G and T flags to .section
1632 directive. Document the extra arguments that the G flag
1633 requires. Document the #tls flag.
1635 2004-05-11 H.J. Lu <hongjiu.lu@intel.com>
1637 * subsegs.c (section_symbol): Create a new section symbol if
1638 the existing one doesn't match.
1639 * symbols.c (symbol_set_bfdsym): Don't reset BFD section symbol.
1641 2004-05-07 Richard Sandiford <rsandifo@redhat.com>
1643 * config/tc-mips.c (append_insn, mips_emit_delays): Extend -mfix-vr4120
1644 to cope with VR4181A errata MD(1) and MD(4).
1646 2004-05-07 Brian Ford <ford@vss.fsi.com>
1648 * NEWS: Mention .secrel32 for pe[i]-i386.
1650 2004-05-07 Alexandre Oliva <aoliva@redhat.com>
1652 * config/tc-frv.h (MAX_MEM_FOR_RS_ALIGN_CODE): New.
1653 (HANDLE_ALIGN): New.
1655 2004-05-06 Daniel Jacobowitz <dan@debian.org>
1657 * Makefile.am (DIST_SUBDIRS): Define.
1658 * aclocal.m4: Regenerate with automake 1.8.4.
1659 * Makefile.in: Likewise.
1660 * doc/Makefile.in: Likewise.
1662 2004-05-06 David Mosberger-Tang <davidm@hpl.hp.com>
1664 * config/tc-ia64.c (dot_serialize): Declare.
1665 (dot_serialize): New function.
1666 (md_pseudo_table): Add ".serialize.data" and
1667 ".serialize.instruction" directives.
1669 2004-05-06 Nick Clifton <nickc@redhat.com>
1671 * messages (as_internal_value_out_of_range): Print a message about
1672 a value being out of range. Be consistent about whether the
1673 values are printed in decimal or hexadecimal.
1674 (as_warn_value_out_of_range): Generate a warning message about an
1676 (as_bad_value_out_of_range): Generate an error message about an
1678 * as.h: Prototype the new functions.
1679 * config/tc-alpha.c (insert_operand): Use new function.
1680 * config/tc-arc.c (arc_insert_operand): Likewise.
1681 * config/tc-mn10200.c (mn10200_insert_operand): Likewise.
1682 * config/tc-mn10300.c (mn10300_insert_operand): Likewise.
1683 * config/tc-ppc.c (ppc_insert_operand): Likewise.
1684 * config/tc-s390.c (s390_insert_operand): Likewise.
1685 * config/tc-v850.c (v850_insert_operand): Likewise.
1687 2004-05-05 Alexandre Oliva <aoliva@redhat.com>
1689 * configure.in: Set em=linux for frv-*-*linux*.
1690 * configure: Rebuilt.
1691 * config/tc-frv.h (TARGET_FORMAT): Use elf32-frvfdpic if...
1692 (frv_md_fdpic_enabled): New.
1693 * config/tc-frv.c (frv_md_fdpic_enabled): New.
1694 (DEFAULT_FDPIC): New.
1695 (frv_flags): Use DEFAULT_FDPIC.
1696 (frv_pic_flag): Likewise.
1697 (OPTION_NOPIC): New.
1698 (md_longopts): Add -mnopic.
1699 (md_parse_option): Handle it.
1700 (md_show_usage): Add -mfdpic and -mnopic.
1702 2004-05-05 Peter Barada <peter@the-baradas.com>
1704 * config/tc-m68k.c: Add find_cf_chip to print list of valid
1705 chips for invalid coldfire instructions, rename selectors
1706 for ColdFire sub-variants, add 521x,5249,547x,548x and aliases,
1707 add current_chip to track which chip is referred to(including save/restore),
1708 use current_chip to select control registers, not current_arch.
1709 (md_show_usage): Add new chips.
1710 * doc/c-m68k.texi: Document new command line switches.
1712 2004-05-05 Jakub Jelinek <jakub@redhat.com>
1714 * tc-s390.h (md_do_align, HANDLE_ALIGN): Remove.
1715 (NOP_OPCODE): Define.
1716 (s390_align_code): Remove prototype.
1717 * tc-s390.c (s390_align_code): Remove.
1719 2004-05-04 H.J. Lu <hongjiu.lu@intel.com>
1721 * config/tc-ia64.c (make_unw_section_name): Removed.
1722 (start_unwind_section): New function.
1723 (generate_unwind_image): Take const segT instead of const
1725 (dot_handlerdata): Adjusted.
1726 (dot_endp): Likewise.
1728 2004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1730 * config/obj-elf.c (obj_elf_change_section): Allow the
1731 ".note.GNU-stack" section has SHF_EXECINSTR.
1733 2004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1735 * config/obj-elf.c (get_section): Return bfd_boolean.
1736 (obj_elf_change_section): Call bfd_get_section_by_name_if
1737 instead of bfd_map_over_sections.
1739 2004-04-30 H.J. Lu <hongjiu.lu@intel.com>
1741 * config/obj-elf.c (get_section): New function.
1742 (obj_elf_change_section): Support multiple sections with same
1745 2004-04-30 Nick Clifton <nickc@redhat.com>
1747 * config/tc-arm.c (create_register_alias): Fix typo checking for
1748 case sensitive register aliases.
1749 (co_proc_number): Use error message string in all_reg_maps[]
1751 (cp_reg_required_here): Likewise.
1752 (fp_reg_required_here): Likewise.
1754 2004-04-29 Brian Ford <ford@vss.fsi.com>
1756 * dwarf2dbg.c (dwarf2_finish): Add SEC_DEBUGGING to section flags.
1758 2004-04-28 Chris Demetriou <cgd@broadcom.com>
1760 * config/tc-mips.c (HAVE_32BIT_ADDRESSES, append_insn, macro_build)
1761 (load_address, macro, mips_ip, md_parse_option)
1762 (mips_force_relocation, mips_validate_fix, md_apply_fix3)
1763 (s_change_sec, pic_need_relax, tc_gen_reloc): Remove all
1764 embedded-PIC handling, and update comments.
1765 (SWITCH_TABLE): Remove.
1766 * config/tc-mips.h (DIFF_EXPR_OK): Delete.
1767 (enum mips_pic_level): Remove EMBEDDED_PIC.
1768 (EXTERN_FORCE_RELOC): Remove embedded-PIC handling.
1769 (TC_FORCE_RELOCATION): Update comment.
1770 * ecoff.c (ecoff_build_lineno): Add comment about some code that
1771 might be safe to remove now that MIPS embedded-PIC is gone.
1773 2004-04-28 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1775 * config/obj-som.c (obj_som_init_stab_section): Add new arguments in
1776 call to obj_set_subsection_attributes.
1777 (obj_som_init_stab_section): Likewise.
1778 * config/tc-hppa.c (default_subspace_dict): Add comdat field.
1779 (pa_def_subspaces): Provide comdat default.
1780 (pa_subspace): Handle new "comdat" parameter. Set SEC_LINK_ONCE and
1781 not SEC_IS_COMMON if section is comdat, common or dup_common. Update
1782 calls to create_new_subspace and update_subspace to pass comdat flag.
1783 (create_new_subspace, update_subspace): Add new comdat argument. Use
1784 it in calls to obj_set_subsection_attributes.
1785 * doc/c-hppa.texi (.subspa, .nsubspa): Document new comdat parameter
1786 and use of comdat, common and dup_comm parameters.
1788 2004-04-26 H.J. Lu <hongjiu.lu@intel.com>
1790 * config/obj-elf.c (obj_elf_change_section): Check if the old
1791 group name is NULL before comparison.
1793 2004-04-23 Chris Demetriou <cgd@broadcom.com>
1795 * config/tc-mips.h (mips_dwarf2_addr_size): Prototype.
1797 2004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1799 * config/tc-mips.c (s_mipsset): Set default CPU type for .set mipsN.
1801 2004-04-23 Chris Demetriou <cgd@broadcom.com>
1803 * config/tc-mips.c (md_longopts): Remove -membedded-pic option.
1804 (OPTION_MEMBEDDED_PIC): Remove.
1805 (OPTION_TRAP, OPTION_BREAK, OPTION_EB, OPTION_EL)
1806 (OPTION_FP32, OPTION_GP32, OPTION_CONSTRUCT_FLOATS)
1807 (OPTION_NO_CONSTRUCT_FLOATS, OPTIONS_FP64, OPTION_GP64)
1808 (OPTION_RELAX_BRANCH, OPTION_NO_RELAX_BRANCH)
1809 (OPTION_ELF_BASE): Renumber.
1810 (md_parse_option): Remove OPTION_MEMBEDDED_PIC handling.
1811 (md_show_usage): Remove mention of -membedded-pic.
1812 * doc/as.texinfo: Remove mention of -membedded-pic.
1814 2004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1816 * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Remove.
1817 * config/tc-mips.c (RDATA_SECTION_NAME, mips_target_format): Remove
1819 (md_begin, mips_ip, md_parse_option, s_change_sec, s_option,
1820 s_abicalls, nopic_need_relax, tc_gen_reloc): Remove uses of
1821 USE_GLOBAL_POINTER_OPT.
1823 2004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1825 * config/tc-mips.c (macro): One more use of load_delay_nop.
1827 2004-04-22 Atsushi Nemoto <anemo@mba.ocn.ne.jp>
1829 * config/tc-mips.c (load_delay_nop): New function.
1830 (load_address, macro): Use load_delay_nop() to build a nop
1831 which can be omitted with gpr_interlocks.
1833 2004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1835 * config/tc-mips.c (hilo_interlocks, gpr_interlocks,
1836 cop_interlocks): Remove superfluous CPU entries.
1838 2004-04-22 Paul Brook <paul@codesourcery.com>
1840 * config/tc-arm.c (mav_parse_offset): Value must be multiple of 4.
1842 2004-04-22 Peter Barada <peter@the-baradas.com>
1844 * NEWS: Added support for EMAC instructions and MAC/EMAC
1846 * config/m68k-parse.h: Add ACC[123], ACCEXT{01,23}, MAC/EMAC
1847 scale factor tokens, trailing_ampersand to mark mask addressing
1848 for MAC/EMAC instructions.
1849 * config/m68k-parse.y: Add options_ampersand clause, '<<',
1851 (yylex): Handle '>', '<', and '&' following '+'.
1852 * config/tc-m68k.c: Set mcfmac/mcfemac on appropriate ColdFire
1853 architectures in archs[].
1854 (m68k-ip): Add '4', 'e', 'g', 'i', cases to handle mask addressing
1855 for MAC/EMAC instructions, ACC[0123], ACCEXT{01,23}, and '<<'/'>>'
1857 (m68k_ip): Handle trailing '&' on MAC/EMAC insns.
1858 (install_operand): Fix 'n' case, Add 'F', 'f', 'G', 'H', 'I', ']'
1860 Add EMAC operands to init_table[].
1862 2004-04-22 Bruno De Bus <bdebus@elis.ugent.be>
1864 * config/tc-arm.h (enum mstate): Move here, add MAP_UNDEFINED
1866 (TC_SEGMENT_INFO_TYPE): Define to enum mstate.
1867 * config/tc-arm.c (enum mstate): Delete from here.
1868 (mapping_state): Remove the static mapstate variable and instead
1869 store the state in the segment. This allows a per-section mapping
1870 state. Handle and ignore MAP_UNDEFINED states.
1871 (arm_elf_change_section): Get the current mapping state from the
1873 (s_ltorg): Set the mapping state to MAP_DATA.
1874 (arm_cleanup): Use arm_elf_change_section to get the mapping state
1875 for each pool as it is emitted.
1877 2004-04-22 Nick Clifton <nickc@redhat.com>
1879 * config/tc-arm.h: Formatting tidy ups.
1881 2004-04-20 Chris Demetriou <cgd@broadcom.com>
1883 * NEWS: Note that MIPS -membedded-pic option is deprecated.
1885 2004-04-20 DJ Delorie <dj@redhat.com>
1887 * config/tc-i386.h [TE_PE] (TC_CONS_FIX_NEW): Define.
1888 * config/tc-i386.c (md_pseudo_table) [TE_PE]: Add "secrel32".
1889 [TE_PE] (O_secrel): Define.
1890 [TE_PE] (x86_pe_cons_fix_new): New.
1891 [TE_PE] (pe_directive_secrel): Likewise.
1892 (tc_gen_reloc) [TE_PE]: Support BFD_RELOC_32_SECREL.
1894 2004-04-19 Eric Christopher <echristo@redhat.com>
1896 * config/tc-mips.c (mips_dwarf2_addr_size): Revert part
1897 of previous patch for fix in gcc.
1899 2004-04-19 Jakub Jelinek <jakub@redhat.com>
1901 * config/tc-xtensa.c (xg_assembler_literal): Fix a typo.
1903 2004-04-19 Nathan Sidwell <nathan@codesourcery.com>
1905 * read.c (do_align): Call md_flush_pending_output, if defined.
1907 2004-04-16 Alan Modra <amodra@bigpond.net.au>
1909 * expr.c (operand): Correct checks for ++ and --.
1911 2004-04-14 H.J. Lu <hongjiu.lu@intel.com>
1913 * config/tc-generic.c: Add some comments.
1915 2004-04-14 Richard Sandiford <rsandifo@redhat.com>
1917 * doc/c-mips.texi (-m{no-,}fix-vr4120): Renamed from
1918 -{no-}mfix-vr4122-bugs.
1919 * config/tc-mips.c (mips_fix_vr4120): Renamed from mips_fix_4122_bugs.
1920 (append_insn, mips_emit_delays): Update accordingly.
1921 (OPTION_FIX_VR4120, OPTION_NO_FIX_VR4120): Renamed from *VR4122.
1922 (md_longopts): Change -{no-,}mfix-vr4122-bugs to -m{no-,}fix-vr4120.
1923 (md_parse_option): Update after above changes.
1924 (md_show_usage): Add -mfix-vr4120.
1926 2004-04-13 Bob Wilson <bob.wilson@acm.org>
1928 * doc/as.texinfo (Sub-Sections): Conditionalize COFF-specific use
1929 of .section directive; add a reference to the ELF .subsection
1932 2004-04-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1934 * config/tc-m32r.c (md_assemble): Fixed infinite loop bug
1937 2004-04-11 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1939 * Makefile.am: Remove mips from aout targets.
1940 * Makefile.in: Regenerate.
1941 * configure.in: Remove mips-dec-bsd* target.
1942 * configure: Regenerate.
1944 2004-04-07 Alan Modra <amodra@bigpond.net.au>
1947 * config/tc-ppc.c (ppc_elf_suffix): Add valid32 and valid64 fields
1948 to struct map_bfd. Adjust MAP macro, and define MAP32, MAP64.
1949 Update "mapping". Restrict some @ modifiers to 32 bit.
1951 2004-04-01 Asgari Jinia <asgarij@kpitcummins.com>
1952 Dhananjay Deshpande <dhananjayd@kpitcummins.com>
1954 * config/tc-sh.c (dont_adjust_reloc_32): New variable.
1955 (sh_fix_adjustable): Avoid adjusting BFD_RELOC_32 when
1956 dont_adjust_reloc_32 is set.
1957 (md_longopts): Add option -renesas.
1958 (md_parse_option, md_show_usage): Likewise.
1959 * doc/c-sh.texi: Likewise.
1961 2004-04-01 Dave Korn <dk@artimi.com>
1963 * config/tc-dlx.c (md_assemble): set fx_no_overflow flag for
1964 hi16 and lo16 fixS structs.
1965 (md_assemble): generate bit_fixS for RELOC_DLX_LO16 in
1966 exactly the same way as for RELOC_DLX_REL16.
1967 (machine_ip): properly respect LO flag in the_insn and
1968 output RELOC_DLX_LO16 rather than RELOC_DLX_16.
1969 (md_apply_fix3): apply RELOC_DLX_LO16.
1971 2004-03-30 Stan Shebs <shebs@apple.com>
1973 Remove long-obsolete MPW support.
1974 * mpw-config.in, mpw-make.sed, mac-as.r: Remove files.
1975 * configure.in: Remove mention of ppc-*-mpw* config.
1976 * configure.in: Likewise.
1978 2004-03-30 Nick Clifton <nickc@redhat.com>
1980 * config/tc-arm.c (meabi_flags): Make its use conditional upon
1981 OBJ_ELF being defined.
1983 2004-03-27 Alan Modra <amodra@bigpond.net.au>
1985 * config/obj-aout.c (obj_aout_type): Remove #ifdef BFD_ASSEMBLER code.
1987 2004-03-23 Paul Brook <paul@codesourcery.com>
1989 * config/tc-arm.c (meabi_flags): New variable.
1990 (arm_parse_eabi): New function.
1991 (md_begin): Set flags for EABI v3.
1993 (arm_long_opts): Add meabi.
1994 * doc/as.texinf <ARM>: Document -meabi.
1995 * doc/c-arm.texi: Ditto.
1997 2004-03-22 Bob Wilson <bob.wilson@acm.org>
1999 * config/tc-xtensa.c (xtensa_post_relax_hook): Create literal
2000 tables even when use_literal_section flag is not set.
2002 2004-03-22 Alan Modra <amodra@bigpond.net.au>
2004 * config/tc-sh.c: Remove trailing whitespace.
2006 2004-03-22 Hans-Peter Nilsson <hp@axis.com>
2008 * doc/c-cris.texi (CRIS-Opts): Document --no-mul-bug-abort,
2009 --mul-bug-abort and the default behavior.
2010 * config/tc-cris.c (cris_insn_kind): New member CRIS_INSN_MUL.
2011 (err_for_dangerous_mul_placement): New variable.
2012 (STATE_MUL, OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): New
2014 (md_cris_relax_table): Have placeholder for STATE_MUL.
2015 (md_longopts): New options --mul-bug-abort and --no-mul-bug-abort.
2016 (cris_relax_frag) <case ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: New
2018 (md_estimate_size_before_relax) <case ENCODE_RELAX (STATE_MUL,
2019 STATE_BYTE)>: Ditto.
2020 (md_convert_frag) <ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: Check
2021 alignment and position of this frag, emit error message if
2023 (md_assemble): For a multiply insn and when checking it,
2024 transform the current frag into a special frag for that purpose.
2025 (md_parse_option) <case OPTION_MULBUG_ABORT_OFF, case
2026 OPTION_MULBUG_ABORT_ON>: Handle new options.
2028 2004-03-19 Bob Wilson <bob.wilson@acm.org>
2030 * config/tc-xtensa.c (mark_literal_frags): New function.
2031 (xtensa_move_literals): Call mark_literal_frags for all literal
2032 segments, including init and fini literal segments.
2033 (xtensa_post_relax_hook): Swap use of xt_insn_sec and xt_literal_sec.
2035 2004-03-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2037 * tc-hppa.c (cons_fix_new_hppa): Check for PC relative base type.
2038 (pa_comm): Set BSF_OBJECT in symbol flags.
2040 2004-03-19 Alan Modra <amodra@bigpond.net.au>
2042 * Makefile.am: Run "make dep-am".
2043 * Makefile.in: Regenerate.
2044 * doc/Makefile.in: Regenerate.
2045 * config.in: Regenerate.
2046 * po/gas.pot: Regenerate.
2048 2004-03-18 Nathan Sidwell <nathan@codesourcery.com>
2050 * read.c (read_a_source_file): Use demand_empty_rest_of_line.
2051 (demand_empty_rest_of_line): Issue an error here.
2052 (ignore_rest_of_line): Silently skip to end.
2053 (demand_copy_string): Issue an error, not warning.
2055 * config/obj-elf.c (obj_elf_section_name): Likewise.
2056 (obj_elf_section): Likewise.
2057 * config/tc-arc.c (arc_extoper): Remove bogus NULL checks.
2058 (arc_extinst): Likewise.
2059 * config/tc-ia64.c (dot_saveb): Use demand_empty_rest_of_line.
2060 (dot_spill): Likewise.
2061 (dot_unwabi): Likewise.
2062 (dot_prologue): Likewise.
2064 * expr.c (operand): Reject ++ and --.
2065 (operator): Likewise.
2067 2004-03-17 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2069 * config/tc-sh.c: Include dw2gencfi.h.
2070 (sh_cfi_frame_initial_instructions): New function.
2071 (sh_regname_to_dw2regnum): Likewise.
2072 * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Move to the end of
2074 (TARGET_USE_CFIPOP): Define.
2075 (tc_cfi_frame_initial_instructions): Likewise.
2076 (tc_regname_to_dw2regnum): Likewise.
2077 (DWARF2_DEFAULT_RETURN_COLUMN, DWARF2_CIE_DATA_ALIGNMENT): Likewise.
2078 * Makefile.am: Update dependencies.
2079 * Makefile.in: Regenerate.
2081 2004-03-17 Ralf Corsepius <corsepiu@faw.uni-ulm.de>
2083 * configure.in: Switch sh-*-rtems* to ELF. Add sh-*-rtemscoff*.
2084 * configure: Regenerate.
2086 2004-03-12 Bob Wilson <bob.wilson@acm.org>
2088 * read.c (s_leb128): Call md_flush_pending_output.
2090 2004-03-12 Michal Ludvig <mludvig@suse.cz>
2092 * config/tc-i386.c (output_insn): Handle PadLock instructions.
2093 * config/tc-i386.h (CpuPadLock): New define.
2094 (CpuUnknownFlags): Added CpuPadLock.
2096 2004-03-07 Andreas Schwab <schwab@suse.de>
2098 * doc/c-hppa.texi (HPPA Directives): Fix typo.
2100 2004-03-07 Richard Henderson <rth@redhat.com>
2102 * dw2gencfi.c (output_cie): Align length to 4 byte boundary.
2103 (cfi_finish): Likewise for fde.
2105 2004-03-05 H.J. Lu <hongjiu.lu@intel.com>
2107 * config/tc-ia64.c (md_assemble): Properly handle NULL
2109 (ia64_handle_align): Don't abort if failed to add a stop bit.
2111 2004-03-04 H.J. Lu <hongjiu.lu@intel.com>
2113 * Makefile.in: Regenerated.
2114 * aclocal.m4: Likewise.
2115 * configure: Likewise.
2116 * doc/Makefile.in: Likewise.
2118 2004-03-03 H.J. Lu <hongjiu.lu@intel.com>
2120 * config/tc-ia64.c (dot_align): New.
2121 (ia64_do_align): Make it static.
2122 (md_pseudo_table): Use "dot_align" for "align".
2123 (ia64_md_do_align): Don't set align_frag here.
2124 (ia64_handle_align): Add a stop bit to the previous bundle if
2127 * config/tc-ia64.h (ia64_do_align): Removed.
2129 2003-03-03 Andrew Stubbs <andrew.stubbs@superh.com>
2131 * config/tc-sh.c (md_parse_option): Add -isa=sh4-nofpu and
2132 -isa=sh4-nommu-nofpu options. Adjust help messages accordingly.
2133 (sh_elf_final_processing): Output BFD type sh4_nofpu if that is
2134 the most general type or the user specifically requested it.
2135 (md_assemble): Add a new error message for when an instruction
2136 is understood, but is not allowed due to an -isa option.
2138 2004-03-02 H.J. Lu <hongjiu.lu@intel.com>
2140 * config/tc-ia64.c (align_frag): New.
2141 (md_assemble): Set the tc_frag_data field in align_frag for
2142 IA64_OPCODE_FIRST instructions.
2143 (ia64_md_do_align): Set align_frag.
2144 (ia64_handle_align): Add a stop bit if needed.
2146 * config/tc-ia64.h (TC_FRAG_TYPE): New.
2147 (TC_FRAG_INIT): New.
2149 2004-03-01 Richard Sandiford <rsandifo@redhat.com>
2151 * config/tc-frv.c (fr400_audio): New variable.
2152 (md_parse_option, md_show_usage): Add -mcpu=fr405 and -mcpu=fr450.
2153 (md_parse_option): Set fr400_audio for -mcpu=fr400 and -mcpu=fr405.
2154 (target_implements_insn_p): New function.
2155 (md_assemble): Report an error if the processor doesn't implement
2158 2004-02-27 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2160 * config/tc-m32r.c (md_longopts): Added -no-bitinst option.
2161 (md_parse_option): Ditto.
2162 (OPTION_NO_SPECIAL_M32R): Added.
2163 (md_show_usage): Document it.
2164 (enable_speial_m32r): Changed a default value from 0 to 1.
2165 * doc/c-m32r.texi: Document the -no-bitinst option.
2167 2004-02-27 Nick Clifton <nickc@redhat.com>
2169 * config/tc-sh.c (get_operand): Revert previous delta.
2170 (tc_gen_reloc): Check for an unknown reloc type before processing
2173 2004-02-27 Hannes Reinecke <hare@suse.de>
2175 * config/tc-s390.c (s390_insn): Correct range check for opcode in
2176 .insn pseudo operation.
2178 2004-02-27 Anil Paranjpe <anilp1@kpitcummins.com>
2180 * config/tc-sh.c (get_operand): In case of #Imm, check has been
2181 added for wrong syntax.
2183 2004-02-26 Eric Christopher <echristo@redhat.com>
2185 * config/tc-mips.c (mips_dwarf2_addr_size): New.
2186 * config/tc-mips.h (DWARF2_ADDR_SIZE): Use.
2188 2004-02-26 Andrew Stubbs <andrew.stubbs@superh.com>
2190 * config/tc-sh.c (build_Mytes): Add REG_N_D and REG_N_B01
2191 nibble types to assembler.
2193 2004-02-25 Fred Fish <fnf@redhat.com>
2195 * config/tc-iq2000.c: Add missing \n\ in multiline string literal.
2197 2004-02-20 James E Wilson <wilson@specifixinc.com>
2199 * config/tc-ia64.c (slot_index): New arg before_relax. Use instead of
2201 (fixup_unw_records): New arg before_relax. Pass to slot_index.
2202 (ia64_estimate_size_before_relax): New.
2203 (ia64_convert_frag): Pass 0 to fixup_unw_records. Add comment.
2204 (generate_unwind_image): Pass 1 to fixup_unw_records.
2205 * config/tc-ia64.h (ia64_estimate_size_before_relax): Declare.
2206 (md_estimate_size_before_relax): Call ia64_estimate_size_before_relax.
2208 2004-02-19 Jakub Jelinek <jakub@redhat.com>
2210 * stabs.c (generate_asm_file): Avoid warning about use of
2211 uninitialized variable.
2213 2004-02-18 David Mosberger <davidm@hpl.hp.com>
2215 * config/tc-ia64.c (ia64_flush_insns): In addition to prologue,
2216 body, and endp, allow unwind records which do not have a "t"
2217 (time/instruction) field.
2219 2004-02-17 Petko Manolov <petkan@nucleusys.com>
2221 * config/tc-arm.c (do_mav_dspsc_1): Correct offset of CRn.
2222 (do_mav_dspsc_2): Likewise.
2223 Fix accumulator registers move opcodes.
2225 2004-02-13 Hannes Reinecke <hare@suse.de>
2226 Jakub Jelinek <jakub@redhat.com>
2228 * dwarf2dbg.c (get_filenum): Do not read beyond allocated memory.
2230 2004-02-10 Steve Ellcey <sje@cup.hp.com>
2232 * config/tc-ia64.h (ia64_frob_symbol): New declaration.
2233 (tc_frob_symbol): New macro definition.
2234 * config/tc-ia64.c (ia64_frob_symbol): New routine.
2236 2004-02-09 Daniel Jacobowitz <drow@mvista.com>
2238 * config/tc-arm.c (md_begin): Mark .note.gnu.arm.ident as
2241 2004-02-09 Nathan Sidwell <nathan@codesourcery.com>
2243 * read.h (IGNORE_OPCODE_CASE): Do not define. Replace with ...
2244 (TC_CASE_SENSITIVE): ... this.
2245 * read.c: Replace IGNORE_OPCODE_CASE with TC_CASE_SENSITIVE.
2246 * doc/internals.texi (TC_CASE_SENSITIVE): Document.
2248 2004-02-06 James E Wilson <wilson@specifixinc.com>
2250 * config/tc-ia64.c (dot_endp): Delete call to output_endp.
2251 (generate_unwind_image): Re-add it here.
2253 2004-02-06 Nathan Sidwell <nathan@codesourcery.com>
2255 * dwarf2dbg.c (DWARF2_ADDR_SIZE): Remove trailing ';'
2256 * read.h (SKIP_WHITESPACE): Turn into an expression.
2257 * read.c (read_a_source_file): A pseudo is removed by having a
2260 2004-02-05 James E Wilson <wilson@specifixinc.com>
2262 * config/tc-ia64.c (output_endp): New.
2263 (count_bits): Delete.
2264 (ia64_flush_insns, process_one_record, optimize_unw_records): Handle
2265 endp unwind records.
2266 (fixup_unw_records): Handle endp unwind records. Delete code for
2267 shortening prologue regions not followed by a body record.
2268 (dot_endp): Call add_unwind_entry to emit endp unwind record.
2269 * config/tc-ia64.h (unw_record_type): Add endp.
2271 2004-02-03 James E Wilson <wilson@specifixinc.com>
2273 * config/tc-ia64.c (ia64_convert_frag): Call md_number_to_chars to
2274 fill padding bytes with zeroes.
2275 (emit_one_bundle): New locals last_ptr, end_ptr. Rewrite code that
2276 sets unwind_record slot_number and slot_frag fields.
2278 2004-02-02 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2280 * config/tc-mips.c (add_got_offset_hilo): New function.
2281 (macro): Use load_register() and add_got_offset_hilo() to load
2282 constants instead of hardcoding code sequences throughout.
2284 2004-01-28 H.J. Lu <hongjiu.lu@intel.com>
2286 * config/tc-ia64.c (emit_one_bundle): Add proper indentation.
2288 2004-01-26 Bernardo Innocenti <bernie@develer.com>
2290 * config/tc-m68k.h (EXTERN_FORCE_RELOC): Handle m68k-uclinux specially,
2292 * config/tc-m68k.c (RELAXABLE_SYMBOL): Use EXTERN_FORCE_RELOC instead
2293 of hard-coded test for TARGET_OS=elf.
2295 2004-01-24 Chris Demetriou <cgd@broadcom.com>
2297 * config/tc-mips.c (hilo_interlocks): Change definition
2298 so that MIPS32, MIPS64 and later ISAs are included, along with
2299 the already-included machines. Update comments.
2301 2004-01-23 Daniel Jacobowitz <drow@mvista.com>
2303 * config/tc-arm.c (tc_gen_reloc): Improve error message for
2304 undefined local labels.
2306 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2308 * config/tc-mips.c (load_address, macro): Update comments about
2309 NewABI GP relaxation.
2311 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2313 * config/tc-mips.c (macro_build): Remove place and counter arguments.
2314 (mips_build_lui, macro_build_ldst_constoffset): Likewise.
2315 (mips16_macro_build, macro_build_jalr): Remove counter argument.
2316 (set_at, load_register, load_address, move_register): Likewise.
2317 (load_got_offset, add_got_offset): Likewise.
2318 Update all calls and tidy accordingly.
2320 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2322 * config/tc-mips.c (RELAX_ENCODE): Remove WARN argument.
2323 (RELAX_FIRST, RELAX_SECOND): Turn into 8-bit quantities.
2324 (RELAX_USE_SECOND): Bump to 0x10000.
2325 (RELAX_SECOND_LONGER, RELAX_NOMACRO, RELAX_DELAY_SLOT): New flags.
2326 (mips_macro_warning): New variable.
2327 (md_assemble): Wrap macro expansion in macro_start() and macro_end().
2328 (s_cpload, s_cpsetup, s_cprestore, s_cpreturn): Likewise.
2329 (relax_close_frag): Set mips_macro_warning.first_frag. Adjust use
2331 (append_insn): Update mips_macro_warning.sizes.
2332 (macro_start, macro_warning, macro_end): New functions.
2333 (macro_build): Don't emit warnings here.
2334 (macro_build_lui, md_estimate_size_before_relax): ...or here.
2335 (md_convert_frag): Check for cases where one macro alternative
2336 needs a warning and the other doesn't. Emit a warning if the
2337 longer sequence was chosen.
2339 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2341 * config/tc-mips.h (tc_frag_data_type, TC_FRAG_TYPE): Remove.
2342 * config/tc-mips.c (RELAX_ENCODE): Take three arguments: the size of
2343 the first sequence, the size of the second sequence, and a flag
2344 that says whether we should warn.
2345 (RELAX_OLD, RELAX_NEW, RELAX_RELOC[123]): Delete.
2346 (RELAX_FIRST, RELAX_SECOND): New.
2347 (mips_relax): New variable.
2348 (relax_close_frag, relax_start, relax_switch, relax_end): New fns.
2349 (append_insn): Remove "place" argument. Use mips_relax.sequence
2350 rather than "place" to check whether we're expanding the second
2351 alternative of a relaxable macro. Remove redundant check for
2352 branch relaxation. If generating a normal insn, and there
2353 is not enough room in the current frag, call relax_close_frag()
2354 to close it. Update mips_relax.sizes[]. Emit fixups for the
2355 second version of a relaxable macro. Record the first relaxable
2356 fixup in mips_relax. Remove tc_gen_reloc workaround.
2357 (macro_build): Remove all uses of "place". Use mips_relax.sequence
2358 in the same way as in append_insn.
2359 (mips16_macro_build): Remove "place" argument.
2360 (macro_build_lui): As for macro_build. Don't drop the add_symbol
2361 when generating the second version of a relaxable macro.
2362 (load_got_offset, add_got_offset): New functions.
2363 (load_address, macro): Use new relaxation machinery. Remove
2364 tc_gen_reloc workarounds.
2365 (md_estimate_size_before_relax): Set RELAX_USE_SECOND if the second
2366 version of a relaxable macro is needed. Return -RELAX_SECOND if the
2367 first version is needed.
2368 (tc_gen_reloc): Remove relaxation handling.
2369 (md_convert_frag): Go through the fixups for a relaxable macro and
2370 mark those that belong to the unneeded alternative as done. If the
2371 second alternative is needed, adjust the fixup addresses to account
2372 for the deleted first alternative.
2374 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2376 * frags.h (frag_room): Declare.
2377 * frags.c (frag_room): New function.
2378 * doc/internals.texi: Document it.
2380 2004-01-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
2382 * config/tc-mips.c (append_insn): Don't do r3900 interlock
2383 optimization for -mtune=r3900, as this will break on other CPUs.
2385 2004-01-11 Tom Rix <tcrix@worldnet.att.net>
2387 * config/tc-m68hc11.c (build_indexed_byte): movb and movw cannot
2388 be relaxed, use fixup.
2389 (md_apply_fix3): Use 5 bit reloc from movb and movw fixup.
2391 2004-01-19 Jakub Jelinek <jakub@redhat.com>
2393 * config/tc-sparc.c (sparc_ip): Disallow %f32-%f63 for single
2396 2004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2398 * config/tc-mips.c (append_insn): Properly detect variant frags
2399 that preclude swapping of relaxed branches. Correctly swap
2400 instructions between frags when dealing with relaxed branches.
2402 2004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2404 * acinclude.m4: Quote names of macros to be defined by AC_DEFUN
2406 * aclocal.m4: Regenerate.
2407 * configure: Regenerate.
2409 2004-01-12 Anil Paranjpe <anilp1@KPITCummins.com>
2411 * config/tc-h8300.c (build_bytes): Apply relaxation to bit
2414 2004-01-12 Richard Sandiford <rsandifo@redhat.com>
2416 * config/tc-mips.c (macro_build_jalr): When adding an R_MIPS_JALR
2417 reloc, reserve space for the delay slot as well as the jalr itself.
2419 2004-01-09 Paul Brook <paul@codesourcery.com>
2421 * config/tc-arm.c (do_vfp_reg2_from_sp2): Rename from do_vfp_sp_reg2.
2422 (do_vfp_sp2_from_reg2): New function.
2424 (do_vfp_dp_from_reg2): Check return values properly.
2426 2004-01-08 Ian Lance Taylor <ian@wasabisystems.com>
2428 * config/tc-mips.c (warn_nops): Remove static variable.
2429 (macro): Remove test of warn_nops.
2430 (md_shortops): Remove 'n'.
2431 (md_parse_option): Remove 'n' case.
2432 (md_show_usage): Remove -n.
2433 * doc/as.texinfo (Overview): Remove MIPS -n option.
2434 * doc/c-mips.texi (MIPS Opts): Remove mention -n.
2435 * NEWS: Mention removal of MIPS -n option.
2437 * config/tc-mips.c (ISA_HAS_COPROC_DELAYS): Remove.
2438 (cop_interlocks): Check ISA level.
2439 (cop_mem_interlocks): Define.
2440 (reg_needs_delay): Check cop_interlocks rather than
2441 ISA_HAS_COPROC_DELAYS.
2442 (append_insn): Likewise. Use cop_mem_interlocks rather than
2443 directly checking mips_opts.isa.
2444 (mips_emit_delays): Likewise.
2446 2004-01-07 H.J. Lu <hongjiu.lu@intel.com>
2448 * config/tc-ia64.c (unwind): Move next_slot_number and
2449 next_slot_frag to ...
2450 (unw_rec_list): Here.
2451 (free_list_records): Removed.
2452 (output_unw_records): Likewise.
2453 (generate_unwind_image): Make it void.
2454 (alloc_record): Initialize next_slot_number and next_slot_frag.
2455 (slot_index): Take .org, .space and .align into account.
2456 (fixup_unw_records): Don't set slot_number to 0. Use
2457 list->next_slot_number and list->next_slot_frag instead of
2458 unwind.next_slot_number and unwind.next_slot_frag.
2459 (ia64_convert_frag): New.
2460 (generate_unwind_image): Generate a rs_machine_dependent frag
2462 (emit_one_bundle): Use list->next_slot_number and
2463 list->next_slot_frag instead of unwind.next_slot_number and
2464 unwind.next_slot_frag.
2466 * config/tc-ia64.h (md_convert_frag): Defined as
2468 (md_estimate_size_before_relax): Defined as (f)->fr_var.
2470 2004-01-06 Alexandre Oliva <aoliva@redhat.com>
2472 2003-12-19 Alexandre Oliva <aoliva@redhat.com>
2473 * config/tc-frv.h (md_apply_fix3): Don't define.
2474 * config/tc-frv.c (md_apply_fix3): New. Shift/truncate %hi/%lo
2476 * config/tc-frv.h (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
2477 2003-10-07 Alexandre Oliva <aoliva@redhat.com>
2478 * config/tc-frv.c (line_separator_chars): Add `!'.
2479 2003-09-19 Alexandre Oliva <aoliva@redhat.com>
2480 * config/tc-frv.c (md_assemble): Clear insn upfront.
2481 2003-09-18 Alexandre Oliva <aoliva@redhat.com>
2482 * config/tc-frv.c (OPTION_FDPIC): New macro.
2483 (md_longopts): Add mfdpic.
2484 (md_parse_option): Handle it.
2485 2003-08-04 Alexandre Oliva <aoliva@redhat.com>
2486 * config/tc-frv.c (md_cgen_lookup_reloc) <FRV_OPERAND_D12,
2487 FRV_OPERAND_S12>: Use reloc type encoded in fix-up.
2488 (frv_pic_ptr): Parse funcdesc.
2490 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2492 * doc/as.texinfo: Let texi2pod parse asconfig.texi and
2493 gasver.texi. Remove duplicate symbol definitions for texi2pod.
2495 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2497 * Makefile.am (Makefile): Move the dependency on
2498 $(BFDDIR)/configure.in to...
2499 (CONFIG_STATUS_DEPENDENCIES): ... here.
2500 (AUTOMAKE_OPTIONS): Require automake 1.8.
2501 * Makefile.in: Regenerate.
2502 * doc/Makefile.am (BASEDIR, BFDDIR): Define.
2503 (CONFIG_STATUS_DEPENDENCIES): Add a dependency on
2504 $(BFDDIR)/configure.in here as well.
2505 * doc/Makefile.in: Regenerate.
2507 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2509 * Makefile.am (install, install-info, RECURSIVE_TARGETS): Remove.
2510 * Makefile.in: Regenerate.
2511 * aclocal.m4: Regenerate.
2512 * doc/Makefile.am (install, install-info): Remove.
2513 (install-data-local): A new hook for install-info.
2514 (AUTOMAKE_OPTIONS): Require automake 1.8.
2515 * doc/Makefile.in: Regenerate.
2517 2004-01-02 Nutan Singh <nutan@kpitcummins.com>
2519 * doc/c-sh.texi: Update description about floating point behavior
2522 2004-01-02 Bernardo Innocenti <bernie@develer.com>
2524 * configure.in: Add m68k-uClinux target.
2525 * configure: Regenerate.
2527 For older changes see ChangeLog-0203
2533 version-control: never