Add support for the new R_AVR_LDI, R_AVR_6 and R_AVR_6_ADIW relocs for the
[deliverable/binutils-gdb.git] / gas / ChangeLog
1 2004-12-22 Klaus Rudolph <lts-rudolph@gmx.de>
2
3 * config/tc-avr.c: Add support for the new R_AVR_LDI, R_AVR_6 and
4 R_AVR_6_ADIW relocs for the LDI, ADIW/SBIW and LDD/STD
5 instructions.
6 (avr_offset_expression): New function to parse offsets for LDI
7 instructions.
8 (avr_operand): Use it.
9 (md_apply_fix3): Generate the relocs.
10
11 2004-12-16 Andrew Stubbs <andrew.stubbs@st.com>
12
13 * config/tc-sh64.c (shmedia_md_apply_fix3): Add missing
14 BFD_RELOC_SH_IMMS10BY8 relocation.
15
16 * config/tc-sh64.c (shmedia_build_Mytes): Emit an error message rather
17 than just ignoring bad code.
18
19 2004-12-16 Richard Sandiford <rsandifo@redhat.com>
20
21 * config/tc-v850.c (handle_lo16): New function.
22 (v850_reloc_prefix): Use it to check lo().
23 (md_assemble, md_apply_fix3): Handle BFD_RELOC_V850_LO16_SPLIT_OFFSET.
24
25 2004-12-14 P.J. Darcy <darcypj@us.ibm.com>
26
27 * configure.in: Add s390x-ibm-tpf support.
28 * configure: Regenerate.
29
30 2004-12-15 Jan Beulich <jbeulich@novell.com>
31
32 * config/obj-elf.c (obj_elf_change_section): Only set type and
33 attributes on new sections. Emit warning when type of re-declared
34 section doesn't match.
35
36 2004-12-15 Jan Beulich <jbeulich@novell.com>
37
38 * dw2gencfi.c (dot.cfi.startproc): Clear cur_cfa_offset so
39 '.cfi_startproc simple' doesn't inherit the old value.
40
41 2004-12-15 Jan Beulich <jbeulich@novell.com>
42
43 * dw2gencfi.c (output_cfi_insn): Adjust DW_CFA_def_cfa_sf generation
44 to emit a signed and factored offset. Adjust DW_CFA_def_cfa_offset_sf
45 generation to emit a factored offset.
46
47 2004-12-10 Ian Lance Taylor <ian@wasabisystems.com>
48
49 * config/tc-mips.c (macro) [M_LA_AB]: Give an error for a offset
50 which is too large in the case of NO_PIC without 64-bit
51 addresses.
52
53 * config/tc-mips.c (mips_in_shared): New static variable.
54 (macro_build_lui): Permit "_gp" if !mips_in_shared.
55 (md_longopts): Add -mshared and -mno-shared.
56 (md_parse_option): Handle OPTION_MSHARED and OPTION_MNO_SHARED.
57 (s_cpload): Implement !mips_in_shared case.
58 (s_cpsetup): Likewise.
59 * doc/c-mips.texi (MIPS Opts): Document -mno-shared.
60 * NEWS: Mention -mno-shared.
61
62 2004-12-09 Paul Brook <paul@codesourcery.com>
63
64 * config/tc-arm.c (s_arm_unwind_fnend): Use R_ARM_PREL31 relocation
65 for function start.
66
67 2004-12-09 Ian Lance Taylor <ian@wasabisystems.com>
68
69 * config/tc-mips.c (append_insn): If we emit a nop during a relax
70 sequence, increase the size of the sequence.
71
72 * config/tc-mips.c (mips_cpu_info_table): Change "9000" entry to
73 use CPU_RM9000.
74
75 2004-12-07 Ben Elliston <bje@gnu.org>
76
77 * read.c (s_align): Use an align_limit temporary to allay a GCC
78 signed/unsigned comparison warning.
79
80 2004-12-01 Mark Mitchell <mark@codesourcery.com>
81
82 * Makefile.am (TARG_ENV_HFILES): Add te-armlinuxeabi.h.
83 * configure.in: Use it for arm*-*-linux-gnueabi*.
84 * config/tc-arm.c: Allow emulation file to set FPU_DEFAULT.
85 * config/te-armlinuxeabi.h: New file.
86 * Makefile.in: Regenerated.
87 * aclocal.m4: Likewise.
88 * configure: Likewise.
89 * doc/Makefile.in: Regenerated.
90
91 2004-12-02 Bob Wilson <bob.wilson@acm.org>
92
93 * config/tc-xtensa.c (xtensa_switch_section_emit_state): Use subseg_set.
94 (xtensa_restore_emit_state): Likewise.
95
96 2004-12-02 Alan Modra <amodra@bigpond.net.au>
97
98 * read.c (ALIGN_LIMIT): Define, increasing limit for BFD_ASSEMBLER.
99 (s_align): Use it.
100
101 2004-11-30 Tero Niemela <tero_niemela@yahoo.com>
102
103 * Makefile.am: Change LOCALEDIR to $(datadir)/share.
104 * Makefile.in: Regenerate.
105
106 2004-11-29 Tomer Levi <Tomer.Levi@nsc.com>
107
108 * config/tc-crx.c: Major code cleanup. Remove unused variables and
109 functions, give functions a meaningful name, add comments.
110 (check_range): New function - Replace operand size calculation
111 with range checking.
112 (assemble_insn): Update Algorithm, improve error issuing.
113 (enum op_err): New.
114 (process_label_constant): Bug fix regarding COP_BRANCH_INS relocation
115 handling.
116
117 2004-11-29 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
118
119 * config/tc-m32r.c (md_pcrel_from_section): Fixed a pcrel relocte
120 miss between different section in the same module.
121 (tc_gen_reloc): Likewise.
122
123 2004-11-25 Theodore A. Roth <troth@openavr.org>
124
125 * gas/config/tc-avr.c (mcu_types): Add support for atmega165,
126 atmega325, atmega3250, atmega645 and atmega6450.
127
128 2004-11-25 Jan Beulich <jbeulich@novell.com>
129
130 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
131 permissible for the selected instruction suffix.
132 (process_suffix): For DefaultSize instructions, suppressing the
133 guessing of a 'q' suffix if the instruction doesn't support it is
134 pointless, because only an 'l' suffix can be guessed in this place.
135
136 2004-11-24 Nick Clifton <nickc@redhat.com>
137
138 * config/tc-iq2000.c: Remove support for IQ10 processor.
139 Convert to ISO C90 formatting.
140 * config/tc-iq2000.h: Likewise.
141
142 2004-11-23 Nick Clifton <nickc@redhat.com>
143
144 * config/tc-mn10300.c (md_relax_table): More fixes to the offsets
145 in this table. They should be correct now.
146
147 2004-11-23 Jan Beulich <jbeulich@novell.com>
148
149 * config/tc-i386.h (CpuMMX2): Declare. Artificial classifier to
150 indicate the MMX extensions added by both SSE and 3DNow!A.
151 (Cpu3dnowA): Declare.
152 (CpuUnknownFlags): Update.
153 * config/tc-i386.c (cpu_sub_arch_name): Declare.
154 (cpu_arch): i586 and pentium do not imply MMX. i686 and pentiumpro do
155 neither imply SSE nor MMX. k6 implies MMX. k6_2 additionally implies
156 3DNow!. Athlon additionally implies 3DNow!A. Several new
157 entries (those starting with a dot are for sub-arch specification).
158 (set_cpu_arch): Handle sub-arch specifications.
159 (parse_insn): Distinguish between instructions not supported because
160 of insufficient CPU features and because of 64-bit mode.
161 * doc/c-i386.texi: Describe enhanced .arch directive.
162
163 2004-11-22 Bob Wilson <bob.wilson@acm.org>
164
165 * config/tc-xtensa.c (xg_add_opcode_fix): Set fx_no_overflow.
166
167 2004-11-22 Bob Wilson <bob.wilson@acm.org>
168
169 * dwarf2dbg.c (dwarf2_finish): Don't write a .debug_line section
170 without a corresponding .debug_info section.
171
172 2004-11-22 Hans-Peter Nilsson <hp@axis.com>
173
174 * read.c (potable): Add "error" and "warning".
175 (s_errwarn): New function.
176 * read.h (s_errwarn): Declare.
177 * doc/as.texinfo (Error, Warning): Document .error and .warning.
178
179 2004-11-22 Nick Clifton <nickc@redhat.com>
180
181 * config/tc-tic54x.c (tic54x_adjust_symtab): Adjust call to
182 c_dot_file_symbol.
183
184 2004-11-19 Alan Modra <amodra@bigpond.net.au>
185
186 * config/tc-msp430.c (struct rcodes_s, MSP430_RLC, msp430_rcodes,
187 struct hcodes_s, msp430_hcodes): From include/opcode/msp430.h.
188
189 2004-11-19 Alan Modra <amodra@bigpond.net.au>
190
191 * config/obj-coff.c (c_dot_file_symbol): Add "app" param.
192 (coff_adjust_symtab): Adjust call.
193 (crawl_symbols): Likewise.
194 * config/obj-coff.h (c_dot_file_symbol): Add "app" param.
195 (obj_app_file): Adjust.
196
197 2004-11-18 Inderpreet Singh <inderpreetb@nioda.hcltech.com>
198
199 * configure.in: Enable bfd_assember for the MAXQ port.
200 * configure: Regenerate.
201
202 2004-11-12 Bob Wilson <bob.wilson@acm.org>
203 Sterling Augustine <sterling@tensilica.com>
204
205 * config/tc-xtensa.c (finish_vinsn): Clear pending instruction if
206 there is a conflict.
207 (check_t1_t2_reads_and_writes): Check for both reads and writes to
208 interfaces that are related as determined by xtensa_interface_class_id.
209
210 2004-11-12 Nick Clifton <nickc@redhat.com>
211
212 * config/tc-mn10300.c (md_relax_table): Fix off by one negative
213 offsets for conditional branches.
214
215 2004-11-11 Bob Wilson <bob.wilson@acm.org>
216
217 * config/tc-xtensa.c (MAX_IMMED6): Change value to 65.
218
219 2004-11-10 Bob Wilson <bob.wilson@acm.org>
220
221 * config/tc-xtensa.c (update_next_frag_state): Always add a NOP if
222 relaxing at the end of a loop. Don't mark frags as UNREACHABLE or
223 MAYBE_UNREACHABLE.
224 (relax_frag_immed): Update call to update_next_frag_state.
225
226 2004-11-10 Alan Modra <amodra@bigpond.net.au>
227
228 * obj.h (struct format_ops <app_file>): Add int param.
229 * read.h (s_app_file_string): Likewise.
230 * read.c (s_app_file_string): Likewise.
231 (s_app_file): Adjust s_app_file_string call.
232 * config/tc-mips.c (s_mips_file): Likewise.
233 * config/obj-coff.h (obj_app_file): Add app param.
234 * config/obj-ecoff.h (obj_app_file): Likewise.
235 * config/obj-multi.h (obj_app_file): Likewise.
236 * config/obj-elf.h (elf_file_symbol): Likewise.
237 * config/obj-elf.c (elf_file_symbol): Only emit one file symbol
238 if called for # preprocessor lines.
239
240 2004-11-08 H.J. Lu <hongjiu.lu@intel.com>
241
242 PR 528
243 * symbols.c (resolve_symbol_value): Convert weak symbols only
244 for Windows PECOFF.
245 (symbol_equated_reloc_p): Don't equate weaks when relocating
246 only for Windows PECOFF.
247
248 2004-11-08 Tomer Levi <Tomer.Levi@nsc.com>
249
250 * config/tc-crx.c (print_insn): Check and set insn_addr.
251 * config/tc-crx.h (md_frag_check): Define.
252
253 2004-11-08 Inderpreet Singh <inderpreetb@nioda.hcltech.com>
254 Vineet Sharma <vineets@noida.hcltech.com>
255
256 * configure.in: Add support for new target maxq-coff.
257 * configure: Regenerate.
258 * NEWS: Mention new support.
259 * config/tc-maxq.c: New file.
260 * config/tc-maxq.h: New file.
261 * config/obj-coff.h: Add support for maxq-coff.
262
263 2004-11-08 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
264
265 * symbols.c (any_external_name): Define.
266 (resolve_symbol_value): Convert weak symbols.
267 (S_SET_EXTERNAL): Support any_external_name.
268 (S_SET_NAME): Qualify parameter const.
269 (symbol_equated_reloc_p): Equate to weaks when relocating.
270 * symbols.h (S_SET_NAME): Qualfiy parameter const.
271 * tc.h (any_external_name): Declare.
272 * config/obj-coff.c ("coff/pe.h"): Include for BFD
273 assemblers also.
274 (weak_is_altname): Declare and define.
275 (weak_name2altname): Same.
276 (weak_altname2name): Same.
277 (weak_uniquify): Same.
278 (weak_altprefix): Define.
279 (obj_coff_weak): Change .weak syntax and handling.
280 (coff_frob_symbol): Fix PE weak symbol alternates.
281 * config/obj-coff.h (USE_UNIQUE): Define.
282 * config/tc-i386.c (md_apply_fix3): Assume weak symbols
283 are in another segment.
284 (tc_gen_reloc): Remove broken addend hack.
285 doc/as.texinfo: Update.
286
287 2004-11-05 Sterling Augustine <sterling@tensilica.com>
288
289 * config/tc-xtensa.c (total_frag_text_expansion): New.
290 (md_estimate_size_before_relax): Use it.
291 (find_address_of_next_align_frag): Likewise.
292
293 2004-11-05 Tomer Levi <Tomer.Levi@nsc.com>
294
295 * config/tc-crx.c: Rename argument types.
296 (processing_arg_number): Rename to 'cur_arg_num'.
297 (get_number_of_bits): Rename to 'set_operand_size'.
298 (get_operandtype): Rename to 'parse_operand', totally rewrite.
299 (set_cons_rparams): Rename to 'set_operand', totally rewrite.
300 (set_indexmode_parameters): Remove function, integrate its code into
301 'set_operand'.
302 (set_operand_size): Get rid of 'Operand Number' function parameter -
303 use global variable 'cur_arg_num' instead.
304 Use a local 'argument' pointer to reference the current argument.
305 (parse_operand): Likewise.
306 (set_operand): Likewise.
307 (process_label_constant): Likewise.
308
309 2004-11-04 Bob Wilson <bob.wilson@acm.org>
310
311 * config/tc-xtensa.c: Remove XTENSA_SECTION_RENAME ifdefs.
312 (add_section_rename): Delete. Inlined into...
313 (build_section_rename): ...here. Use xstrdup instead of strdup.
314 (xtensa_section_rename): Drop "const" from argument and return types.
315 (md_show_usage): Indent to match show_usage().
316 * config/tc-xtensa.h: Remove XTENSA_SECTION_RENAME ifdefs.
317 (tc_canonicalize_section_name): Define.
318 (md_elf_section_rename): Remove unused macro.
319 * doc/as.texinfo (Overview): Document Xtensa --rename-section option.
320 * doc/c-xtensa.texi (Xtensa Options): Likewise.
321 (Frame Directive): Delete.
322
323 2004-11-04 Daniel Jacobowitz <dan@debian.org>
324
325 * configure.in: Remove arm-*-oabi and thumb-*-oabi.
326 * config/tc-arm.c (target_oabi): Delete.
327 (md_apply_fix3, elf32_arm_target_format): Remove target_oabi checks.
328 (arm_opts): Remove moabi.
329 * doc/as.texinfo (Overview): Remove documentation of -moabi.
330 * doc/c-arm.texi (ARM Options): Likewise.
331 * configure: Regenerated.
332
333 2004-11-04 Hans-Peter Nilsson <hp@axis.com>
334
335 * configure.in (crisv32): Recognize. AC_DEFINE_UNQUOTED
336 DEFAULT_CRIS_ARCH. Handle crisv32-*-linux-gnu* like
337 cris-*-linux-gnu* and crisv32-*-* like cris-*-*.
338 * configure: Regenerate.
339 * config/tc-cris.c (enum cris_archs): New.
340 (cris_mach, cris_arch_from_string, s_cris_arch, get_sup_reg)
341 (cris_insn_ver_valid_for_arch): New functions.
342 (DEFAULT_CRIS_ARCH): New macro, default to cris_any_v0_v10.
343 (cris_arch): New variable.
344 (md_pseudo_table): New pseudo .arch.
345 (err_for_dangerous_mul_placement): Initialize according to
346 DEFAULT_CRIS_ARCH.
347 (STATE_COND_BRANCH): Renamed from STATE_CONDITIONAL_BRANCH.
348 All users changed.
349 (STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON)
350 (STATE_ABS_BRANCH_V32, STATE_LAPC, BRANCH_BF_V32, BRANCH_BB_V32)
351 (BRANCH_WF_V32, BRANCH_WB_V32): New.
352 (BRANCH_BF, BRANCH_BB, BRANCH_WF, BRANCH_WB): Don't undef after
353 use in md_cris_relax_table.
354 (md_cris_relax_table): Add entries for STATE_COND_BRANCH_V32,
355 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
356 Update and improve head comment.
357 (OPTION_PIC): Define in terms of previous option, OPTION_US.
358 (OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): Similar.
359 (OPTION_ARCH): New.
360 (md_longopts): New option --march=...
361 (cris_any_v0_v10_long_jump_size, crisv32_long_jump_size): New
362 macros.
363 (md_long_jump_size): Initialize in terms of DEFAULT_CRIS_ARCH.
364 (HANDLE_RELAXABLE): New macro.
365 (md_estimate_size_before_relax): Use HANDLE_RELAXABLE for common
366 cases. Check for weak symbols and assume not relaxable. Handle
367 STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON,
368 STATE_ABS_BRANCH_V32, STATE_LAPC. Use new variable symbolP, not
369 fragP->fr_symbol.
370 (md_convert_frag): Handle STATE_COND_BRANCH_V32,
371 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
372 (cris_create_short_jump): Adjust for CRISv32.
373 (cris_relax_frag): Handle new states.
374 (md_create_long_jump): Ditto. Emit error for common_v10_v32.
375 (md_begin): Define symbols "..asm.arch.cris.v32",
376 "..asm.arch.cris.v10", "..asm.arch.cris.common_v10_v32" and
377 "..asm.arch.cris.any_v0_v10". Use cris_insn_ver_valid_for_arch
378 when entering opcode table entry points.
379 (md_assemble): Adjust branch handling for CRISv32. Handle LAPC
380 relaxation. In fix_new_exp call for main insn, pass 1 for pcrel
381 parameter for 8, 16 and 32-bit pc-relative insns and LAPC.
382 (cris_process_instruction): Initialize out_insnp->insn_type to
383 CRIS_INSN_NONE, not CRIS_INSN_NORMAL.
384 <case ']', '[', 'A', 'd', 'Q', 'N', 'n', 'Y', 'U', 'u', 'T'>: New
385 cases.
386 <case 'm'>: Check that modified_char == '.'.
387 <invalid operands>: Consume the rest of the line.
388 When operands don't match, skip over subsequent insns with
389 non-matching version specifier but same mnemonic.
390 <immediate constant, case SIZE_SPEC_REG>: Immediate operands for
391 special registers in CRISv32 are always 32 bit long.
392 <immediate constant, case SIZE_FIELD_SIGNED, SIZE_FIELD_UNSIGNED>:
393 New cases.
394 (get_gen_reg): Only recognize "PC" when followed by "+]" for v32
395 and compatible. Recognize "ACR" for v32, unless followed by "+".
396 (get_spec_reg): Consider cris_arch when looking up register.
397 (get_autoinc_prefix_or_indir_op): Don't recognize assignment for
398 v32 or compatible.
399 (get_3op_or_dip_prefix_op): Check for ']' after seeing '[rN+'.
400 (cris_get_expression): Restore input_line_pointer if failing "early".
401 (get_flags): Consider cris_arch and recognize flags accordingly.
402 (branch_disp): Adjust for CRISv32.
403 (gen_cond_branch_32): Similar. Emit error for common_v10_v32.
404 (cris_number_to_imm): Use as_bad_where, not as_bad. Remove
405 related FIXME. Don't insist on BFD_RELOC_32_PCREL fixup to be
406 resolved. Don't enter zeros in object file for
407 BFD_RELOC_32_PCREL.
408 <case BFD_RELOC_CRIS_LAPCQ_OFFSET, BFD_RELOC_CRIS_SIGNED_16>
409 <case BFD_RELOC_CRIS_SIGNED_8>: New case.
410 (md_parse_option): Break out "return 1".
411 <OPTION_ARCH> New case.
412 (tc_gen_reloc): <case BFD_RELOC_CRIS_LAPCQ_OFFSET>
413 <case BFD_RELOC_CRIS_SIGNED_16, BFD_RELOC_CRIS_SIGNED_8>
414 <case BFD_RELOC_CRIS_UNSIGNED_8, BFD_RELOC_CRIS_UNSIGNED_16>
415 <case BFD_RELOC_32_PCREL>: New cases.
416 Addends for non-zero fx_pcrel are too in fx_offset.
417 (md_show_usage): Show --march=<arch>.
418 (md_apply_fix3): Adjust val for BFD_RELOC_CRIS_LAPCQ_OFFSET.
419 (md_pcrel_from): BFD_RELOC_CRIS_LAPCQ_OFFSET is PC-relative too.
420 (s_syntax) <struct syntaxes>: Properly constify member operand.
421 * config/tc-cris.h (TARGET_MACH): Define.
422 (cris_mach): Declare.
423 * doc/as.texinfo (Overview) <CRIS>: Add --march=...
424 * doc/c-cris.texi (CRIS-Symbols): New node for built-in symbols.
425 (CRIS-Opts): Document --march=...
426 (CRIS-Pseudos): Document .arch.
427
428 2004-11-04 Jan Beulich <jbeulich@novell.com>
429
430 * config/tc-i386.c (set_intel_syntax): Allow % in symbol names when
431 intel syntax and no register prefix, allow $ in symbol names when
432 intel syntax.
433 (set_16bit_gcc_code_flag): Replace literal 'l' by LONG_MNEM_SUFFIX.
434 (intel_float_operand): Add fourth return value indicating math control
435 operations. Make classification more precise.
436 (md_assemble): Complain if memory operand of mov[sz]x has no size
437 specified.
438 (parse_insn): Translate word operands to floating point instructions
439 operating on integers as well as control instructions to short ones
440 as expected by AT&T syntax. Translate 'd' suffix to short one only for
441 floating point instructions operating on non-integer operands.
442 (match_template): Remove fldcw special case. Adjust q-suffix handling
443 to permit it on fild/fistp/fisttp in AT&T mode.
444 (process_suffix): Don't guess DefaultSize insns' suffix from
445 stackop_size for certain floating point control instructions. Guess
446 suffix for branch and [ls][gi]dt based on flag_code. Split error
447 messages for Intel and AT&T syntax, and make the condition more strict
448 for the former. Adjust suppressing of generation of operand size
449 overrides.
450 (intel parser): Allow the full set of MASM operators. Add FWORD, TBYTE,
451 OWORD, and XMMWORD operand size specifiers (TBYTE replaces XWORD). Add
452 more error checking.
453 * config/tc-i386.h (BYTE_PTR WORD_PTR DWORD_PTR QWORD_PTR XWORD_PTR
454 SHORT OFFSET_FLAT FLAT NONE_FOUND): Remove unused defines.
455
456 2004-11-03 Hans-Peter Nilsson <hp@axis.com>
457
458 * symbols.c (colon) [!WORKING_DOT_WORD]: Don't declare
459 md_short_jump_size, md_long_jump_size.
460 * write.c [!WORKING_DOT_WORD]: Ditto.
461 * tc.h [!WORKING_DOT_WORD]: Declare them here. Drop const
462 qualifier.
463 * config/tc-cris.h (md_short_jump_size, md_long_jump_size): Don't
464 declare.
465 * config/tc-cris.c (md_short_jump_size, md_long_jump_size): Drop
466 const qualifier in these definitions.
467 * config/tc-i370.c, config/tc-m68k.c, config/tc-pdp11.c,
468 config/tc-s390.c, config/tc-tahoe.c, config/tc-vax.c: Ditto.
469
470 2004-11-02 Nick Clifton <nickc@redhat.com>
471
472 * dwarf2dbg.c (dwarf2_finish): Check for the existence of a file
473 table before deciding to produce a .debug_line section to match up
474 with a user provided .debug_info section.
475
476 2004-10-28 Tomer Levi <Tomer.Levi@nsc.com>
477
478 * config/tc-crx.c (getreg_image): Bug fix, a return value was
479 mistakenly omitted from CRX_C_REGTYPE and CRX_CS_REGTYPE cases.
480
481 2004-10-27 Tomer Levi <Tomer.Levi@nsc.com>
482
483 * config/tc-crx.c: Remove global variable 'post_inc_mode'.
484 (get_flags): New function.
485 (get_number_of_bits): Edit comments, update numeric values to
486 supported sizes.
487 (process_label_constant): Don't support the colon
488 format (SYMBOL:[s|m|l]).
489 (set_cons_rparams): Support argument type 'arg_rbase'.
490 (get_operandtype): Bug fix in 'rbase' operand type parsing.
491 (handle_LoadStor): Bug fix, first handle post-increment mode.
492 (getreg_image): Remove redundant code, update according to latest
493 CRX spec.
494 (print_constant): Bug fix relate to 3-word instructions.
495 (assemble_insn): Bug fix, when matching instructions, verify also
496 instruction type (not only mnemonic).
497 Add various error checking.
498 (preprocess_reglist): Support HI/LO and user registers.
499
500 2004-10-25 David Mosberger-Tang <davidm@hpl.hp.com>
501
502 * config/tc-ia64.c (fixup_unw_records): Don't let the "t" value
503 in an epilogue directive go negative.
504
505 2004-10-25 H.J. Lu <hongjiu.lu@intel.com>
506
507 PR 474
508 * config/tc-ia64.c (emit_one_bundle): Decrement md.num_slots_in_use
509 after reporting template error during manual bundling. Reported
510 by Michael Dupont, michaelx.dupont@intel.com.
511
512 2004-10-25 Daniel Jacobowitz <dan@debian.org>
513
514 * Makefile.am: Run dep-am.
515 * aclocal.m4: Regenerate with automake 1.9.2.
516 * Makefile.in: Regenerate with automake 1.9.2.
517 * doc/Makefile.in: Likewise.
518
519 * config/tc-arm.c: Include "dw2gencfi.h".
520 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
521 New functions.
522 * config/tc-arm.h (TARGET_USE_CFIPOP, DWARF2_DEFAULT_RETURN_COLUMN)
523 (DWARF2_CIE_DATA_ALIGNMENT, tc_regname_to_dw2regnum)
524 (tc_cfi_frame_initial_instructions): Define.
525 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
526 Add prototypes.
527
528 2004-10-21 Tomer Levi <Tomer.Levi@nsc.com>
529
530 * config/tc-crx.c (assemble_insn): Check unsigned immediate
531 operands validity.
532 Update coprocessor id to be unsigned immediate.
533
534 2004-10-18 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
535
536 * config/tc-i386.c (O_secrel): Delete.
537 (tc_pe_dwarf2_emit_offset): New function.
538 * config/tc-i386.h (O_secrel): Define as O_md1.
539 (TC_DWARF2_EMIT_OFFSET): Define.
540
541 2004-10-18 Nick Clifton <nickc@redhat.com>
542
543 * config/tc-xstormy16.c (xstormy16_cons_fix_new): Accept and
544 ignore @fptr() directives for 4-byte fixups.
545
546 2004-10-15 Alan Modra <amodra@bigpond.net.au>
547
548 * Makefile.am: Run "make dep-am"
549 * Makefile.in: Regenerate.
550
551 2004-10-14 Bob Wilson <bob.wilson@acm.org>
552
553 * doc/c-xtensa.texi (Xtensa Options, Absolute Literals Directive):
554 Remove comments about placement of literal pools.
555 (Literal Directive): Update description of literal placement.
556 (Literal Prefix Directive): Remove statement that this does not apply
557 to absolute-mode literals. Describe new section naming scheme.
558
559 2004-10-12 Bob Wilson <bob.wilson@acm.org>
560
561 * config/tc-xtensa.c (xtensa_frequency_pseudo): Use set_subseg_freq.
562 (is_entry_opcode, is_movi_opcode, is_the_loop_opcode, is_jx_opcode,
563 is_windowed_return_opcode): Delete.
564 (xtensa_frob_label): Use get_subseg_target_freq.
565 (md_assemble): Inline call to is_entry_opcode.
566 (xtensa_handle_align): Inline call to get_frag_is_literal.
567 (relaxation_requirements): Inline call to is_jx_opcode.
568 (emit_single_op): Inline call to is_movi_opcode.
569 (xg_assemble_vliw_tokens): Inline calls to get_frag_is_insn,
570 get_frag_is_no_transform, is_entry_opcode, and
571 set_frag_is_specific_opcode. Use get_subseg_total_freq.
572 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags,
573 xtensa_fix_close_loop_end_frags, relax_frag_immed, convert_frag_immed):
574 Inline calls to get_frag_is_no_transform.
575 (next_instrs_are_b_retw): Inline call to is_windowed_return_opcode.
576 (xtensa_fix_short_loop_frags): Inline calls to is_the_loop_opcode and
577 get_frag_is_no_transform.
578 (convert_frag_immed_finish_loop): Inline calls to get_expression_value
579 and set_frag_is_no_transform.
580 (get_expression_value): Delete.
581 (subseg_map struct): Rename cur_total_freq to total_freq. Rename
582 cur_target_freq to target_freq.
583 (get_subseg_info): Split out code to create a new map entry into ...
584 (add_subseg_info): ... this new function.
585 (get_last_insn_flags): Check if get_subseg_info succeeded.
586 (set_last_insn_flags): Call add_subseg_info if needed.
587 (get_subseg_total_freq, get_subseg_target_freq, set_subseg_freq): New.
588 (xtensa_reorder_segments): Compute last_sec while counting sections.
589 Remove call to get_last_sec.
590 (get_last_sec): Delete.
591 (cache_literal_section): Inline call to retrieve_literal_seg and its
592 callees, seg_present and add_seg_list.
593 (retrieve_literal_seg, seg_present, add_seg_list): Delete.
594 (get_frag_is_insn, get_frag_is_no_transform,
595 set_frag_is_specific_opcode, set_frag_is_no_transform): Delete.
596 * config/tc-xtensa.h (MAX_SLOTS): Reduce from 31 to 15.
597
598 2004-10-12 Bob Wilson <bob.wilson@acm.org>
599
600 * config/tc-xtensa.c: Use ISO C90 formatting.
601 * config/tc-xtensa.h: Likewise.
602 * config/xtensa-istack.h: Likewise.
603 * config/xtensa-relax.c: Likewise.
604 * config/xtensa-relax.h: Likewise.
605
606 2004-10-12 Paul Brook <paul@codesourcery.com>
607
608 * config/tc-arm.c (md_begin): Change EF_ARM_EABI_VER3 to
609 EF_ARM_EABI_VER4.
610 (arm_eabis): Ditto.
611 * doc/c-arm.texi: Document that we actually support -meabi=4, not
612 -meabi=3.
613
614 2004-10-08 Bob Wilson <bob.wilson@acm.org>
615
616 * doc/as.texinfo (VTableEntry, VTableInherit): Add "directive" to index
617 entries.
618 (Acknowledgements): Use "GAS" instead of AS variable.
619
620 2004-10-08 Daniel Jacobowitz <dan@debian.org>
621
622 * config/tc-i386.c: Include "elf/x86-64.h".
623 (i386_elf_section_type): New function.
624 * config/tc-i386.h (md_elf_section_type): Define.
625 (i386_elf_section_type): New prototype.
626
627 2004-10-08 Linus Nielsen Feltzing <linus@haxx.se>
628
629 * config/m68k-parse.h (enum m68k_register): New control register,
630 MBAR2 (for MCF5249)
631 * config/tc-m68k.c: Correct control register set for MCF5249.
632
633 2004-10-07 Bob Wilson <bob.wilson@acm.org>
634 Sterling Augustine <sterling@tensilica.com>
635
636 * config/tc-xtensa.c (absolute_literals_supported): New global flag.
637 (UNREACHABLE_MAX_WIDTH): Define.
638 (XTENSA_FETCH_WIDTH): Delete.
639 (cur_vinsn, xtensa_fetch_width, xt_saved_debug_type, past_xtensa_end,
640 prefer_const16, prefer_l32r): New global variables.
641 (LIT4_SECTION_NAME): Define.
642 (lit4_state struct): Add lit4_seg_name and lit4_seg fields.
643 (XTENSA_PROP_*, GET_XTENSA_PROP_*, SET_XTENSA_PROP_*): Define.
644 (frag_flags struct): New.
645 (xtensa_block_info struct): Move from tc-xtensa.h. Add flags field.
646 (subseg_map struct): Add cur_total_freq and cur_target_freq fields.
647 (bitfield, bit_is_set, set_bit, clear_bit): Define.
648 (MAX_FORMATS): Define.
649 (op_placement_info struct, op_placement_table): New.
650 (O_pltrel, O_hi16, O_lo16): Define.
651 (directiveE enum): Rename directive_generics to directive_transform.
652 Delete directive_relax. Add directive_schedule,
653 directive_absolute_literals, and directive_last_directive.
654 (directive_info): Rename "generics" to "transform". Delete "relax".
655 Add "schedule" and "absolute-literals".
656 (directive_state): Adjust entries to match changes in directive_info.
657 (xtensa_relax_statesE, RELAX_IMMED_MAXSTEPS): Move to tc-xtensa.h.
658 (xtensa_const16_opcode, xtensa_movi_opcode, xtensa_movi_n_opcode,
659 xtensa_l32r_opcode, xtensa_nop_opcode, xtensa_rsr_lcount_opcode): New.
660 (xtensa_j_opcode, xtensa_rsr_opcode): Delete.
661 (align_only_targets, software_a0_b_retw_interlock,
662 software_avoid_b_j_loop_end, maybe_has_b_j_loop_end,
663 software_avoid_short_loop, software_avoid_close_loop_end,
664 software_avoid_all_short_loops, specific_opcode): Delete.
665 (warn_unaligned_branch_targets): New.
666 (workaround_a0_b_retw, workaround_b_j_loop_end, workaround_short_loop,
667 workaround_close_loop_end, workaround_all_short_loops): Default FALSE.
668 (option_[no_]link_relax, option_[no_]transform,
669 option_[no_]absolute_literals, option_warn_unaligned_targets,
670 option_prefer_l32r, option_prefer_const16, option_target_hardware):
671 New enum values.
672 (option_[no_]align_only_targets, option_literal_section_name,
673 option_text_section_name, option_data_section_name,
674 option_bss_section_name, option_eb, option_el): Delete.
675 (md_longopts): Add entries for: [no-]transform, [no-]absolute-literals,
676 warn-unaligned-targets, prefer-l32r, prefer-const16, [no-]link-relax,
677 and target-hardware. Delete entries for [no-]target-align-only,
678 literal-section-name, text-section-name, data-section-name, and
679 bss-section-name.
680 (md_parse_option): Handle new options and remove old ones. Accept but
681 ignore [no-]density options. Warn for [no-]generics and [no-]relax
682 and treat them as [no-]transform.
683 (md_show_usage): Add new options and remove old ones.
684 (xtensa_setup_hw_workarounds): New.
685 (md_pseudo_table): Change "word" entry to use xtensa_elf_cons. Add
686 "long", "short", "loc" and "frequency" entries.
687 (use_generics): Rename to ...
688 (use_transform): ... this function. Add past_xtensa_end check.
689 (use_longcalls): Add past_xtensa_end check.
690 (code_density_available, can_relax): Delete.
691 (do_align_targets): New.
692 (get_directive): Accept dashes in directive names. Warn about
693 [no-]generics and [no-]relax directives and treat them as
694 [no-]transform.
695 (xtensa_begin_directive): Call md_flush_pending_output only for some
696 directives. Check for directives inside instruction bundles. Warn
697 about deprecated ".begin literal" usage. Warn and ignore [no-]density
698 directives. Handle new directives. Check generating_literals flag
699 for literal_prefix.
700 (xtensa_end_directive): Check for directives inside instruction
701 bundles. Warn and ignore [no-]density directives. Handle new
702 directives. Call xtensa_set_frag_assembly_state.
703 (xtensa_loc_directive_seen, xtensa_dwarf2_directive_loc,
704 xtensa_dwarf2_emit_insn): New.
705 (xtensa_literal_position): Call md_flush_pending_output. Do not check
706 use_literal_section flag.
707 (xtensa_literal_pseudo): Call md_flush_pending_output. Handle absolute
708 literals. Use xtensa_elf_cons to parse the expression.
709 (xtensa_literal_prefix): Do not check use_literal_section. Support
710 ".lit4" sections for absolute literals. Change prefix convention to
711 replace ".text" (or ".t" in a linkonce section). No need to call
712 subseg_set.
713 (xtensa_frequency_pseudo, xtensa_elf_cons, xtensa_elf_suffix): New.
714 (expression_end): Handle closing braces and colons.
715 (PLT_SUFFIX, plt_suffix): Delete.
716 (expression_maybe_register): Use new xtensa-isa.h functions. Use
717 xtensa_elf_suffix instead of checking for plt suffix, and handle O_lo16
718 and O_hi16 expressions as well.
719 (tokenize_arguments): Handle closing braces and colons.
720 (parse_arguments): Use new xtensa-isa.h functions. Handle "invisible"
721 operands and paired register syntax.
722 (get_invisible_operands): New.
723 (xg_translate_sysreg_op): Handle new Xtensa LX RSR/WSR/XSR syntax. Use
724 new xtensa-isa.h functions.
725 (xtensa_translate_old_userreg_ops, xtensa_translate_zero_immed): New.
726 (xg_translate_idioms): Check if inside bundle. Use use_transform.
727 Handle new Xtensa LX RSR/WSR/XSR syntax. Remove code to widen density
728 instructions. Use xtensa_translate_zero_immed.
729 (operand_is_immed, operand_is_pcrel_label): Delete.
730 (get_relaxable_immed): Use new xtensa-isa.h functions.
731 (get_opcode_from_buf): Add slot parameter. Use new xtensa-isa.h
732 functions.
733 (xtensa_print_insn_table, print_vliw_insn): New.
734 (is_direct_call_opcode): Use new xtensa-isa.h functions.
735 (is_call_opcode, is_loop_opcode, is_conditional_branch_opcode,
736 is_branch_or_jump_opcode): Delete.
737 (is_movi_opcode, decode_reloc, encode_reloc, encode_alt_reloc): New.
738 (opnum_to_reloc, reloc_to_opnum): Delete.
739 (xtensa_insnbuf_set_operand, xtensa_insnbuf_get_operand): Use new
740 xtensa-isa.h functions. Operate on one slot of an instruction.
741 (xtensa_insnbuf_set_immediate_field, is_negatable_branch,
742 xg_get_insn_size): Delete.
743 (xg_get_build_instr_size): Use xg_get_single_size.
744 (xg_is_narrow_insn, xg_is_single_relaxable_insn): Update calls to
745 xg_build_widen_table. Use xg_get_single_size.
746 (xg_get_max_narrow_insn_size): Delete.
747 (xg_get_max_insn_widen_size, xg_get_max_insn_widen_literal_size,
748 xg_is_relaxable_insn): Update calls to xg_build_widen_table. Use
749 xg_get_single_size.
750 (xg_build_to_insn): Record the loc field. Handle OP_OPERAND_HI16U and
751 OP_OPERAND_LOW16U. Check xg_valid_literal_expression.
752 (xg_expand_to_stack, xg_expand_narrow): Update calls to
753 xg_build_widen_table. Use xg_get_single_size.
754 (xg_immeds_fit): Use new xtensa-isa.h functions. Update call to
755 xg_check_operand.
756 (xg_symbolic_immeds_fit): Likewise. Also handle O_lo16 and O_hi16, and
757 treat weak symbols conservatively.
758 (xg_check_operand): Use new xtensa-isa.h functions.
759 (is_dnrange): Delete.
760 (xg_assembly_relax): Inline previous calls to tinsn_copy.
761 (xg_finish_frag): Specify separate relax states for the frag and slot0.
762 (is_branch_jmp_to_next, xg_add_branch_and_loop_targets): Use new
763 xtensa-isa.h functions.
764 (xg_instruction_matches_option_term, xg_instruction_matches_or_options,
765 xg_instruction_matches_options): New.
766 (xg_instruction_matches_rule): Handle O_register expressions. Call
767 xg_instruction_matches_options.
768 (transition_rule_cmp): New.
769 (xg_instruction_match): Update call to xg_build_simplify_table.
770 (xg_build_token_insn): Record loc fields.
771 (xg_simplify_insn): Check is_specific_opcode field and
772 density_supported flag.
773 (xg_expand_assembly_insn): Skip checking code_density_available. Use
774 new xtensa-isa.h functions. Call use_transform instead of can_relax.
775 (xg_assemble_literal): Add error handling for O_big. Call
776 record_alignment. Handle O_pltrel.
777 (xg_valid_literal_expression): New.
778 (xg_assemble_literal_space): Add slot parameter. Remove call to
779 set_expr_symbol_offset. Add call to record_alignment. Update call to
780 xg_finish_frag.
781 (xg_emit_insn): Delete.
782 (xg_emit_insn_to_buf): Add format parameter. Update calls to
783 xg_add_opcode_fix and xtensa_insnbuf_to_chars.
784 (xg_add_opcode_fix): Change opcode parameter to tinsn and add format
785 and slot parameters. Handle new "alternate" relocations for absolute
786 literals and CONST16 instructions. Check for bad uses of O_lo16 and
787 O_hi16. Use new xtensa-isa.h functions.
788 (xg_assemble_tokens): Delete.
789 (is_register_writer): Use new xtensa-isa.h functions.
790 (is_bad_loopend_opcode): Check for xtensa_rsr_lcount_opcode instead of
791 old-style RSR from LCOUNT.
792 (next_frag_opcode): Delete.
793 (next_frag_opcode_is_loop, next_frag_format_size, frag_format_size,
794 update_next_frag_state): New.
795 (update_next_frag_nop_state): Delete.
796 (next_frag_pre_opcode_bytes): Use next_frag_opcode_is_loop.
797 (xtensa_mark_literal_pool_location): Check use_literal_section flag and
798 the state of the absolute-literals directive. Add calls to
799 record_alignment and xtensa_set_frag_assembly_state. Call
800 xtensa_switch_to_non_abs_literal_fragment instead of
801 xtensa_switch_to_literal_fragment.
802 (build_nop): New.
803 (assemble_nop): Use build_nop. Update call to xtensa_insnbuf_to_chars.
804 (get_expanded_loop_offset): Change check for undefined opcode to an
805 assertion.
806 (xtensa_set_frag_assembly_state, relaxable_section,
807 xtensa_find_unmarked_state_frags, xtensa_find_unaligned_branch_targets,
808 xtensa_find_unaligned_loops, xg_apply_tentative_value): New.
809 (md_begin): Update call to xtensa_isa_init. Initialize linkrelax to 1.
810 Set lit4_seg_name. Call xg_init_vinsn. Initialize new global opcodes.
811 Call init_op_placement_info_table and xtensa_set_frag_assembly_state.
812 (xtensa_init_fix_data): New.
813 (xtensa_frob_label): Reset label symbol to the current frag. Check
814 do_align_targets and generating_literals flag. Propagate frequency
815 info to new alignment frag. Call xtensa_set_frag_assembly_state.
816 (xtensa_unrecognized_line): New.
817 (xtensa_flush_pending_output): Check if inside a bundle. Add a call
818 to xtensa_set_frag_assembly_state.
819 (error_reset_cur_vinsn): New.
820 (md_assemble): Remove check for literal frag. Remove call to
821 istack_init. Call use_transform instead of use_generics. Parse
822 explicit instruction format specifiers. Move code for
823 a0_b_retw_interlock workaround to xg_assemble_vliw_tokens. Call
824 error_reset_cur_vinsn on errors. Add call to get_invisible_operands.
825 Add dwarf2_where call. Remote automatic alignment for ENTRY
826 instructions. Move call to xtensa_clear_insn_labels to the end.
827 Rearrange to handle bundles.
828 (xtensa_cons_fix_new): Delete.
829 (xtensa_handle_align): New.
830 (xtensa_frag_init): Call xtensa_set_frag_assembly_state. Remove
831 assignment to is_no_density field.
832 (md_pcrel_from): Use new xtensa-isa.h functions. Use decode_reloc
833 instead of reloc_to_opnum. Handle "alternate" relocations.
834 (xtensa_force_relocation, xtensa_check_inside_bundle,
835 xtensa_elf_section_change_hook): New.
836 (xtensa_symbol_new_hook): Delete.
837 (xtensa_fix_adjustable): Check for difference of symbols with an
838 offset. Check for external and weak symbols.
839 (md_apply_fix3): Remove cases for XTENSA_OP{0,1,2} relocs.
840 (md_estimate_size_before_relax): Return expansion for the first slot.
841 (tc_gen_reloc): Handle difference of symbols by producing
842 XTENSA_DIFF{8,16,32} relocs and by writing the value of the difference
843 into the output. Handle new XTENSA_SLOT*_OP relocs by storing the
844 tentative values into the output when linkrelax is set.
845 (XTENSA_PROP_SEC_NAME): Define.
846 (xtensa_post_relax_hook): Call xtensa_find_unmarked_state_frags.
847 Create literal tables only if using literal sections. Create new
848 property tables instead of old instruction tables. Check for unaligned
849 branch targets and loops.
850 (finish_vinsn, find_vinsn_conflicts, check_t1_t2_reads_and_writes,
851 new_resource_table, clear_resource_table, resize_resource_table,
852 resources_available, reserve_resources, release_resources,
853 opcode_funcUnit_use_unit, opcode_funcUnit_use_stage,
854 resources_conflict, xg_find_narrowest_format, relaxation_requirements,
855 bundle_single_op, emit_single_op, xg_assemble_vliw_tokens): New.
856 (xtensa_end): Call xtensa_flush_pending_output. Set past_xtensa_end
857 flag. Update checks for workaround options. Call
858 xtensa_mark_narrow_branches and xtensa_mark_zcl_first_insns.
859 (xtensa_cleanup_align_frags): Add special case for branch targets.
860 Check for and mark unreachable frags.
861 (xtensa_fix_target_frags): Remove use of align_only_targets flag.
862 Use RELAX_LOOP_END_BYTES in special case for negatable branch at the
863 end of a zero-overhead loop body.
864 (frag_can_negate_branch): Handle instructions with multiple slots.
865 Use new xtensa-isa.h functions
866 (xtensa_mark_narrow_branches, is_narrow_branch_guaranteed_in_range,
867 xtensa_mark_zcl_first_insns): New.
868 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags): Error if
869 transformations are disabled.
870 (next_instrs_are_b_retw): Use new xtensa-isa.h functions. Handle
871 multislot instructions.
872 (xtensa_fix_close_loop_end_frags, xtensa_fix_short_loop_frags):
873 Likewise. Also error if transformations are disabled.
874 (unrelaxed_frag_max_size): New.
875 (unrelaxed_frag_min_insn_count, unrelax_frag_has_b_j): Use new
876 xtensa-isa.h functions.
877 (xtensa_sanity_check, is_empty_loop, is_local_forward_loop): Use
878 xtensa_opcode_is_loop instead of is_loop_opcode.
879 (get_text_align_power): Replace as_fatal with assertion.
880 (get_text_align_fill_size): Iterate instead of using modulus when
881 use_nops is false.
882 (get_noop_aligned_address): Assert that this is for a machine-dependent
883 RELAX_ALIGN_NEXT_OPCODE frag. Use next_frag_opcode_is_loop,
884 xg_get_single_size, and frag_format_size.
885 (get_widen_aligned_address): Rename to ...
886 (get_aligned_diff): ... this function. Add max_diff parameter.
887 Remove handling of rs_align/rs_align_code frags. Use
888 next_frag_format_size, get_text_align_power, get_text_align_fill_size,
889 next_frag_opcode_is_loop, and xg_get_single_size. Compute max_diff
890 and pass it back to caller.
891 (xtensa_relax_frag): Use relax_frag_loop_align. Add code for new
892 RELAX_SLOTS, RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN,
893 RELAX_FILL_NOP, and RELAX_UNREACHABLE frag types. Check relax_seen.
894 (relax_frag_text_align): Rename to ...
895 (relax_frag_loop_align): ... this function. Assume loops can only be
896 in the first slot of an instruction.
897 (relax_frag_add_nop): Use assemble_nop instead of constructing an OR
898 instruction. Remove call to frag_wane.
899 (relax_frag_narrow): Rename to ...
900 (relax_frag_for_align): ... this function. Extend to handle
901 RELAX_FILL_NOP and RELAX_UNREACHABLE, as well as RELAX_SLOTS with
902 RELAX_NARROW for the first slot.
903 (find_address_of_next_align_frag, bytes_to_stretch): New.
904 (future_alignment_required): Use find_address_of_next_align_frag and
905 bytes_to_stretch. Look ahead to subsequent frags to make smarter
906 alignment decisions.
907 (relax_frag_immed): Add format, slot, and estimate_only parameters.
908 Check if transformations are enabled for b_j_loop_end workaround.
909 Use new xtensa-isa.h functions and handle multislot instructions.
910 Update call to xg_assembly_relax.
911 (md_convert_frag): Handle new RELAX_SLOTS, RELAX_UNREACHABLE,
912 RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN, and RELAX_FILL_NOP
913 frag types.
914 (convert_frag_narrow): Add segP, format and slot parameters. Call
915 convert_frag_immed for branch instructions. Adjust calls to
916 tinsn_from_chars, tinsn_immed_from_frag, and xg_emit_insn_to_buf. Use
917 xg_get_single_size and xg_get_single_format.
918 (convert_frag_fill_nop): New.
919 (convert_frag_immed): Add format and slot parameters. Handle multislot
920 instructions and use new xtensa-isa.h functions. Update calls to
921 tinsn_immed_from_frag and xg_assembly_relax. Check if transformations
922 enabled for b_j_loop_end workaround. Use build_nop instead of
923 assemble_nop. Check is_specific_opcode flag. Check for unreachable
924 frags. Use xg_get_single_size. Handle O_pltrel.
925 (fix_new_exp_in_seg): Remove check for old plt flag.
926 (convert_frag_immed_finish_loop): Update calls to tinsn_from_chars and
927 xtensa_insnbuf_to_chars. Call tinsn_immed_from_frag. Change check
928 for loop opcode to an assertion. Mark all frags up to the end of the
929 loop as not transformable.
930 (get_last_insn_flags, set_last_insn_flags): Use get_subseg_info.
931 (get_subseg_info): New.
932 (xtensa_move_literals): Call xtensa_set_frag_assembly_state. Add null
933 check for dest_seg.
934 (xtensa_switch_to_literal_fragment): Rewrite to handle absolute
935 literals and use xtensa_switch_to_non_abs_literal_fragment otherwise.
936 (xtensa_switch_to_non_abs_literal_fragment): New.
937 (cache_literal_section): Add is_code parameter and pass it through to
938 retrieve_literal_seg.
939 (retrieve_literal_seg): Add is_code parameter and use it to set the
940 flags on the literal section. Handle case where head parameter is 0.
941 (get_frag_is_no_transform, set_frag_is_specific_opcode,
942 set_frag_is_no_transform): New.
943 (xtensa_create_property_segments): Add end_property_function parameter
944 and pass it through to add_xt_block_frags. Call bfd_get_section_flags
945 and skip SEC_DEBUGGING and !SEC_ALLOC sections.
946 (xtensa_create_xproperty_segments, section_has_xproperty): New.
947 (add_xt_block_frags): Add end_property_function parameter and call it
948 if it is non-zero. Call xtensa_frag_flags_init.
949 (xtensa_frag_flags_is_empty, xtensa_frag_flags_init,
950 get_frag_property_flags, frag_flags_to_number,
951 xtensa_frag_flags_combinable, xt_block_aligned_size,
952 xtensa_xt_block_combine, add_xt_prop_frags,
953 init_op_placement_info_table, opcode_fits_format_slot,
954 xg_get_single_size, xg_get_single_format): New.
955 (istack_push): Inline call to tinsn_copy.
956 (tinsn_copy): Delete.
957 (tinsn_has_invalid_symbolic_operands): Handle O_hi16 and O_lo16 and
958 CONST16 opcodes. Handle O_big, O_illegal, and O_absent.
959 (tinsn_has_complex_operands): Handle O_hi16 and O_lo16.
960 (tinsn_to_insnbuf): Use xg_get_single_format and new xtensa-isa.h
961 functions. Handle invisible operands.
962 (tinsn_to_slotbuf): New.
963 (tinsn_check_arguments): Use new xtensa-isa.h functions.
964 (tinsn_from_chars): Add slot parameter. Rewrite using xg_init_vinsn,
965 vinsn_from_chars, and xg_free_vinsn.
966 (tinsn_from_insnbuf): New.
967 (tinsn_immed_from_frag): Add slot parameter and handle multislot
968 instructions. Handle symbol differences.
969 (get_num_stack_text_bytes): Use xg_get_single_size.
970 (xg_init_vinsn, xg_clear_vinsn, vinsn_has_specific_opcodes,
971 xg_free_vinsn, vinsn_to_insnbuf, vinsn_from_chars, expr_is_register,
972 get_expr_register, set_expr_symbol_offset_diff): New.
973 * config/tc-xtensa.h (MAX_SLOTS): Define.
974 (xtensa_relax_statesE): Move from tc-xtensa.c. Add
975 RELAX_CHECK_ALIGN_NEXT_OPCODE, RELAX_MAYBE_DESIRE_ALIGN, RELAX_SLOTS,
976 RELAX_FILL_NOP, RELAX_UNREACHABLE, RELAX_MAYBE_UNREACHABLE, and
977 RELAX_NONE types.
978 (RELAX_IMMED_MAXSTEPS): Move from tc-xtensa.c.
979 (xtensa_frag_type struct): Add is_assembly_state_set,
980 use_absolute_literals, relax_seen, is_unreachable, is_specific_opcode,
981 is_align, is_text_align, alignment, and is_first_loop_insn fields.
982 Replace is_generics and is_relax fields by is_no_transform field.
983 Delete is_text and is_longcalls fields. Change text_expansion and
984 literal_expansion to arrays of MAX_SLOTS entries. Add arrays of
985 per-slot information: literal_frags, slot_subtypes, slot_symbols,
986 slot_sub_symbols, and slot_offsets. Add fr_prev field.
987 (xtensa_fix_data struct): New.
988 (xtensa_symfield_type struct): Delete plt field.
989 (xtensa_block_info struct): Move definition to tc-xtensa.h. Add
990 forward declaration here.
991 (xt_section_type enum): Delete xt_insn_sec. Add xt_prop_sec.
992 (XTENSA_SECTION_RENAME): Undefine.
993 (TC_FIX_TYPE, TC_INIT_FIX_DATA, TC_FORCE_RELOCATION, NO_PSEUDO_DOT,
994 tc_unrecognized_line, md_do_align, md_elf_section_change_hook,
995 HANDLE_ALIGN, TC_LINKRELAX_FIXUP, SUB_SEGMENT_ALIGN): Define.
996 (TC_CONS_FIX_NEW, tc_symbol_new_hook): Delete.
997 (unit_num_copies_func, opcode_num_units_func,
998 opcode_funcUnit_use_unit_func, opcode_funcUnit_use_stage_func): New.
999 (resource_table struct): New.
1000 * config/xtensa-istack.h (MAX_INSN_ARGS): Increase from 6 to 10.
1001 (TInsn struct): Add keep_wide, loc, fixup, record_fix, subtype,
1002 literal_space, symbol, sub_symbol, offset, and literal_frag fields.
1003 (tinsn_copy): Delete prototype.
1004 (vliw_insn struct): New.
1005 * config/xtensa-relax.c (insn_pattern_struct): Add options field.
1006 (widen_spec_list): Add option conditions for density and boolean
1007 instructions. Add expansions using CONST16 and conditions for using
1008 CONST16 vs. L32R. Use new Xtensa LX RSR/WSR syntax. Add entries for
1009 predicted branches.
1010 (simplify_spec_list): Add option conditions for density instructions.
1011 Add entry for NOP instruction.
1012 (append_transition): Add cmp function pointer parameter and use it to
1013 insert the new entry in order.
1014 (operand_function_LOW16U, operand_function_HI16U): New.
1015 (xg_has_userdef_op_fn, xg_apply_userdef_op_fn): Handle
1016 OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
1017 (enter_opname, split_string): Use xstrdup instead of strdup.
1018 (init_insn_pattern): Initialize new options field.
1019 (clear_req_or_option_list, clear_req_option_list,
1020 clone_req_or_option_list, clone_req_option_list, parse_option_cond):
1021 New.
1022 (parse_insn_pattern): Parse option conditions.
1023 (transition_applies): New.
1024 (build_transition): Use new xtensa-isa.h functions. Fix incorrectly
1025 swapped last arguments in calls to append_constant_value_condition.
1026 Call clone_req_option_list. Add warning about invalid opcode.
1027 Handle LOW16U and HI16U function names.
1028 (build_transition_table): Add cmp parameter and use it in calls to
1029 append_transition. Use new xtensa-isa.h functions. Check
1030 transition_applies before adding entries.
1031 (xg_build_widen_table, xg_build_simplify_table): Add cmp parameter and
1032 pass it through to build_transition_table.
1033 * config/xtensa-relax.h (ReqOrOptionList, ReqOrOption, ReqOptionList,
1034 ReqOption, transition_cmp_fn): New types.
1035 (OpType enum): Add OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
1036 (transition_rule struct): Add options field.
1037 * doc/as.texinfo (Overview): Update Xtensa options.
1038 * doc/c-xtensa.texi (Xtensa Options): Delete --[no-]density,
1039 --[no-]relax, and --[no-]generics options. Update descriptions of
1040 --text-section-literals and --[no-]longcalls. Add
1041 --[no-]absolute-literals and --[no-]transform.
1042 (Xtensa Syntax): Add description of syntax for FLIX instructions.
1043 Remove use of "generic" and "specific" terminology for opcodes.
1044 (Xtensa Registers): Generalize the syntax description to include
1045 user-defined register files.
1046 (Xtensa Automatic Alignment): Update.
1047 (Xtensa Branch Relaxation): Mention limitation of unconditional jumps.
1048 (Xtensa Call Relaxation): Linker can now remove most of the overhead.
1049 (Xtensa Directives): Remove confusing rules about precedence.
1050 (Density Directive, Relax Directive): Delete.
1051 (Schedule Directive): New.
1052 (Generics Directive): Rename to ...
1053 (Transform Directive): ... this node.
1054 (Literal Directive): Update for absolute literals. Missing
1055 literal_position directive is now an error.
1056 (Literal Position Directive): Update for absolute literals.
1057 (Freeregs Directive): Delete.
1058 (Absolute Literals Directive): New.
1059 (Frame Directive): Minor editing.
1060 * Makefile.am (DEPTC_xtensa_elf, DEPOBJ_xtensa_elf, DEP_xtensa_elf):
1061 Update dependencies.
1062 * Makefile.in: Regenerate.
1063
1064 2004-10-07 Richard Sandiford <rsandifo@redhat.com>
1065
1066 * config/tc-mips.c (append_insn): Use fix_new rather than fix_new_exp
1067 to build the second and third fixups for a composite relocation.
1068 (macro_read_relocs): New function.
1069 (macro_build): Use it.
1070 (s_cpsetup): Pass all three composite relocation codes to macro_build.
1071 Simplify fragging code accordingly.
1072 (s_gpdword): Use fix_new rather than fix_new_exp for the second part
1073 of the composite relocation. Set fx_tcbit in both fixups.
1074
1075 2004-10-07 Richard Sandiford <rsandifo@redhat.com>
1076
1077 * config/tc-mips.c (append_insn): Set fx_tcbit for composite relocs.
1078 (md_apply_fix3): Don't treat composite relocs as done.
1079
1080 2004-10-07 Jan Beulich <jbeulich@novell.com>
1081
1082 * macro.c (macro_expand_body): When ELF, use .LL rather than LL as
1083 prefix for symbol names generated from the LOCAL macro directive.
1084
1085 * dw2gencfi.c (select_cie_for_fde): When separating CIE out from
1086 FDE, treat a DW_CFA_remember_state as we do a DW_CFA_advance_loc.
1087
1088 2004-10-07 Tomer Levi <Tomer.Levi@nsc.com>
1089
1090 * config/tc-crx.c (preprocess_reglist): Handle Co-processor
1091 Special registers.
1092 (md_assemble): Add error checking for Co-Processor instructions.
1093 (get_cinv_parameters): Add 'b' option to invalidate the
1094 branch-target cache.
1095
1096 2004-10-05 Paul Brook <paul@codesourcery.com>
1097
1098 * config/tc-arm.c (unwind): New variable.
1099 (vfp_sp_encode_reg): New function.
1100 (vfp_sp_reg_required_here): Use it.
1101 (vfp_sp_reg_list, vfp_dp_reg_list): Remove.
1102 (vfp_parse_reg_list): New function.
1103 (s_arm_unwind_fnstart, s_arm_unwind_fnend, s_arm_unwind_cantunwind,
1104 s_arm_unwind_personality, s_arm_unwind_personalityindex,
1105 s_arm_unwind_handlerdata, s_arm_unwind_save, s_arm_unwind_movsp,
1106 s_arm_unwind_pad, s_arm_unwind_setfp, s_arm_unwind_raw): New
1107 functions.
1108 (md_pseudo_table): Add them.
1109 (do_vfp_reg2_from_sp2): Use vfp_parse_reg_list and vfp_sp_encode_reg.
1110 (do_vfp_sp2_from_reg2, vfp_sp_ldstm, vfp_dp_ldstm): Ditto.
1111 (set_section, add_unwind_adjustsp, flush_pending_unwind,
1112 finish_unwind_opcodes, start_unwind_section, create_unwind_entry,
1113 require_hashconst, add_unwind_opcode): New functions.
1114 * doc/tc-arm.text: Document unwinding opcodes.
1115 * NEWS: Mention the new feature.
1116
1117 2004-10-04 Eric Christopher <echristo@redhat.com>
1118
1119 * config/tc-mips.c (md_apply_fix3): Remove erroneous assert.
1120
1121 2004-10-01 H.J. Lu <hongjiu.lu@intel.com>
1122
1123 * config/tc-ppc.c (md_apply_fix3): Call S_SET_THREAD_LOCAL for
1124 TLS relocations.
1125 * config/tc-s390.c (md_apply_fix3): Likewise.
1126 * config/tc-sparc.c (md_apply_fix3): Likewise.
1127
1128 2004-10-01 Paul Brook <paul@codesourcery.com>
1129
1130 * config/tc-arm.c (arm_elf_section_type): New function.
1131 (arm_elf_change_section): Set section link for exidx sections.
1132 * config/tc-arm.h (arm_elf_section_type): Add prototype.
1133 (md_elf_section_type): Define.
1134
1135 2004-10-01 Bill Farmer <Bill@the-farmers.freeserve.co.uk>
1136
1137 * config/tc-pdp11.c (md_apply_fix3): Change to sign of the SOB
1138 instruction's offset.
1139
1140 2004-10-01 Adam Nemet <anemet@lnxw.com>
1141
1142 * (TARGET_FORMAT): Remove LynxOS COFF definition.
1143
1144 2004-10-01 Ravi Ramaseshan <ravi.ramaseshan@codito.com>
1145
1146 * config/tc-arc.c (tc_gen_reloc): Don't assume fixP->fx_addsy is an
1147 asymbol *, instead use symbol_get_bfdsym.
1148
1149 2004-09-30 Linus Nielsen Feltzing <linus@haxx.se>
1150
1151 * config/tc-m68k.c (select_control_regs): Add mcf5249.
1152
1153 2004-09-30 Paul Brook <paul@codesourcery.com>
1154
1155 * config/tc-arm.c (do_smi, do_nop): New functions.
1156 (insns): Add ARMv6ZK instructions.
1157 (md_apply_fix3): Handle BFD_RELOC_ARM_SMI.
1158 (tc_gen_reloc): Ditto.
1159 (arm_cpus): Add mpcore and arm1176.
1160 (arm_archs): Add armv6{k,z,zk}.
1161 * doc/c-arm.texi: Document new cores and architectures.
1162
1163 2004-09-30 Nick Clifton <nickc@redhat.com>
1164
1165 * config/tc-arm.c: Use ISO C90 formatting.
1166
1167 2004-09-30 Vladimir Ivanov <vladitx@nucleusys.com>
1168
1169 * config/tc-arm.c (mav_reg_required_here): Allow REG_TYPE_CN
1170 as alternative when REG_TYPE_MVF, REG_TYPE_MVD, REG_TYPE_MVFX or
1171 REG_TYPE_MVDX is expected.
1172
1173 2004-09-29 Marc Bevand <m.bevand@gmail.com>
1174
1175 * doc/c-i386.texi (i386-Mnemonics): Fix typo.
1176
1177 2004-09-21 James E Wilson <wilson@specifixinc.com>
1178
1179 * config/tc-ia64.c (ENCODED_PSP_OFFSET): New.
1180 (output_rp_psprel, output_pfs_psprel, output_preds_psprel,
1181 output_spill_base, output_unat_psprel, output_lc_psprel,
1182 output_fpsr_psprel, output_priunat_psprel, output_bsp_psprel,
1183 output_bsprestore_psprel, output_rnat_psprel, output_spill_psprel,
1184 output_spill_psprel_p): Use it.
1185
1186 2004-09-20 Tomer Levi <Tomer.Levi@nsc.com>
1187
1188 * config/tc-crx.c (handle_LoadStor): New function.
1189 Handle load/stor unique instructions before parsing.
1190
1191 2004-09-17 Paul Brook <paul@codesourcery.com>
1192
1193 * config/tc-arm.c (s_arm_rel31): New funciton.
1194 (md_pseudo_table): Add .rel31.
1195 (md_apply_fix3): Handle BFD_RELOC_ARM_TARGET2,
1196 BFD_RELOC_32_PCREL and BFD_RELOC_ARM_PREL31.
1197 (tc_gen_reloc): Handle BFD_RELOC_ARM_PREL31 and BFD_RELOC_ARM_TARGET2.
1198 (arm_fix_adjustable): Return 0 for BFD_RELOC_ARM_TARGET2.
1199 (arm_parse_reloc): Add (target2).
1200
1201 2004-09-17 Alan Modra <amodra@bigpond.net.au>
1202
1203 * Makefile.am: Run "make dep-am".
1204 * Makefile.in: Regenerate.
1205 * aclocal.m4: Regenerate.
1206 * configure: Regenerate.
1207 * doc/Makefile.in: Regenerate.
1208 * po/POTFILES.in: Regenerate.
1209 * po/gas.pot: Regenerate.
1210
1211 2004-09-14 Hideki IWAMOTO <h-iwamoto@kit.hi-ho.ne.jp>
1212
1213 * config/tc-mmix.c [!LLONG_MIN]: Correct #elsif to #elif.
1214 [!LLONG_MAX]: Ditto.
1215
1216 2004-09-13 Paul Brook <paul@codesourcery.com>
1217
1218 * config/tc-arm.c: Rename RELABS to TARGET1.
1219
1220 2004-09-13 Alan Modra <amodra@bigpond.net.au>
1221
1222 * messages.c (as_internal_value_out_of_range): Cast values passed
1223 to as_bad_where or as_warn_where to proper type.
1224
1225 2004-09-11 Theodore A. Roth <troth@openavr.org>
1226
1227 * config/tc-avr.c: Add support for
1228 atmega48, atmega88, atmega168, attiny13, attiny2313, at90can128.
1229
1230 2004-09-09 Alan Modra <amodra@bigpond.net.au>
1231
1232 * dw2gencfi.c (select_cie_for_fde): When separating CIE out
1233 from FDE, treat a CFI_escape as we do a DW_CFA_advance_loc.
1234
1235 2004-09-08 Paul Brook <paul@codesourcery.com>
1236
1237 * config/obj-elf.c (obj_elf_section_type): Handle init_array,
1238 fini_array and preinit_array section types.
1239 * config/tc-ia64.c (ia64_elf_section_type): Remove init_array
1240 and fini_array.
1241 * doc/as.texinfo: Document extra section types.
1242
1243 2004-09-02 Mark Mitchell <mark@codesourcery.com>
1244
1245 * Makefile.am (TARG_ENV_HFILES): Add te-symbian.h.
1246 * Makefile.in: Regenerated.
1247 * configure.in: Set em for arm*-*-symbianelf*.
1248 * configure: Regenerated.
1249 * config/tc-arm.c (elf32_arm_target_format): Use Symbian target
1250 vectors when appropriate.
1251 * config/te-symbian.h: New file.
1252
1253 2004-09-03 Tomer Levi <Tomer.Levi@nsc.com>
1254
1255 * config/tc-crx.c (gettrap): Exception vector can be case
1256 insensitive.
1257 (process_label_constant): Fix a 32-bit displacement bug in branch
1258 instructions.
1259 (get_operandtype) : Bug fix, wrong operand was used.
1260 (process_label_constant): Initialize relocation type to
1261 BFD_RELOC_NONE
1262
1263 2004-09-01 Richard Earnshaw < reanrsha@arm.com>
1264
1265 * tc-arm.c (arm_cpus, arm_fpus): Allow <cpu>-s as well as <cpu>s
1266 for synthesizable cores.
1267
1268 * doc/c-arm.texi (ARM Options): Document canonical names of CPUs.
1269
1270 2004-08-25 Dmitry Diky <diwil@spec.ru>
1271
1272 * config/tc-msp430.c: Clean-up the code.
1273 (md_relax_table): New relax table.
1274 (mcu_types): Sort MCU types.
1275 (md_pseudo_table): Add .profiler pseudo handler.
1276 (pow2value): New function.
1277 (msp430_profiler): New function.
1278 (msp430_operands): Add new insns handlers.
1279 (msp430_srcoperand): Add register operand handler, allow complex
1280 expressions.
1281 (md_estimate_size_before_relax): Rewritten.
1282 (md_convert_frag): Rewritten.
1283 (msp430_relax_frag): New function.
1284 * config/tc-msp430.h (md_relax_frag): define macro
1285 * doc/c-msp430.texi: Update information.
1286
1287 2004-08-24 Nick Clifton <nickc@redhat.com>
1288
1289 * as.c (std_shortopts): Allow -g to take an optional argument.
1290 (parse_args): Pass any switch starting with -g on to the backend
1291 for parsing.
1292
1293 2004-08-18 Mark Mitchell <mark@codesourcery.com>
1294
1295 * configure.in (arm*-*-symbianelf*): New target.
1296 (arm*-*-eabi*): Likewise.
1297 * configure: Regenerated.
1298
1299 2004-08-18 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1300 * config/tc-mips.c (append_insn): Handle delay slots in branch likely
1301 correctly.
1302
1303 2004-08-18 Jakub Jelinek <jakub@redhat.com>
1304
1305 * config/tc-ia64.c (start_unwind_section): Add linkonce_empty
1306 argument, don't do anything if current section is not
1307 .gnu.linkonce.t.* and linkonce_empty is set.
1308 (generate_unwind_image, dot_endp): Adjust callers, call
1309 start_unwind_section (*, 1) if nothing will be put into the
1310 section.
1311
1312 2004-08-17 Nick Clifton <nickc@redhat.com>
1313
1314 * as.c (MD_DEBUG_FORMAT_SELECTOR): Provide default definition.
1315 (show_usage): Add -g.
1316 (std_longopts): Add --gen-debug. Alpha sort the table.
1317 (parse_args): Print an error message if a switch is not handled.
1318 Handle the -g switch, calling md_debug_format_selector() if
1319 necessary.
1320 * NEWS: Mention new feature.
1321 * doc/as.texinfo: Document new switch.
1322 * doc/internals.texi: Document behaviour of md_parse_option.
1323
1324 * config/tc-arm.c (md_parse_option): Do not issue an error message
1325 if the switch is not recognised.
1326 * config/tc-m68k.c (md_parse_option): Likewise.
1327 * config/tc-pdp11.c (md_parse_option): Likewise.
1328 * config/tc-v850.c (md_parse_option): Likewise.
1329
1330 * as.h: Fix up formatting.
1331 * tc.h: Likewise.
1332
1333 2004-08-16 Nick Clifton <nickc@redhat.com>
1334
1335 * macro.c (macro_set_alternate): Use ISO C90 formatting.
1336
1337 * configure.in: Sort architecture based tables alphabetically.
1338 * configure: Regenerate.
1339
1340 2004-08-16 Alan Modra <amodra@bigpond.net.au>
1341
1342 * config/tc-ppc.c (tc_ppc_regname_to_dw2regnum <regnames>): Replace
1343 { "cc", 68 }, with { "cr", 70 }.
1344
1345 2004-08-13 Jan Beulich <jbeulich@novell.com>
1346 Nick Clifton <nickc@redhat.com>
1347
1348 * as.c: Add and handle new --alternate command line option.
1349 * macro.c (macro_set_alternate): New.
1350 * macro.h (macro_set_alternate): Declare.
1351 * read.c: Add and handle new .altmacro and .noaltmacro directives.
1352 * doc/as.texinfo: Document new command line option and pseudo-ops
1353 as well as insert documentation originating from gasp about
1354 alternate macro syntax.
1355 * NEWS: Mention new command line option and pseudo-ops.
1356
1357 2004-08-10 Mark Mitchell <mark@codesourcery.com>
1358
1359 * expr.c (operand): Handle the "~", "-", and "!" operators applied
1360 to bignums.
1361
1362 2004-08-06 Paul Brook <paul@codesourcery.com>
1363
1364 * config/tc-arm.c (md_apply_fix3, tc_gen_reloc, arm_parse_reloc):
1365 Handle new relocations.
1366 * include/elf/arm.h (elf_arm_reloc_type): Add new EABI relocations.
1367
1368 2004-08-05 Bob Wilson <bob.wilson@acm.org>
1369
1370 * write.c (relax_segment): Use was_address instead of address when
1371 setting fr_fix field for align frag due to backwards .org.
1372
1373 2004-07-29 Alexandre Oliva <aoliva@redhat.com>
1374
1375 Introduce SH2a support.
1376 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
1377 * config/tc-sh.c (get_specific): Change arch_sh2a_up to
1378 arch_sh2a_nofpu_up.
1379 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
1380 * config/tc-sh.c (md_parse_option): Add sh2a-nofpu ISA handling.
1381 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
1382 * config/tc-sh.c (sh_elf_final_processing): Move sh2a recognition
1383 to end of conditional expression.
1384 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
1385 * config/tc-sh.c: Add sh2a-nofpu support.
1386 2003-12-29 DJ Delorie <dj@redhat.com>
1387 * tc-sh.c: Add sh2a support.
1388 (parse_reg): Add tbr.
1389 (parse_at): Support @@(disp,tbr).
1390 (get_specific): Support sh2a opcodes.
1391 (insert4): New, for 4 byte relocs.
1392 (build_Mytes): Support sh2a opcodes.
1393 (md_apply_fix3_Mytes): Support sh2a opcodes.
1394 2003-12-02 Michael Snyder <msnyder@redhat.com>
1395 * config/tc-sh.c (md_parse_option): Handle sh2a.
1396 (sh_elf_final_processing): Ditto.
1397
1398 2004-07-27 Jason Thorpe <thorpej@wasabisystems.com>
1399
1400 * config/tc-hppa.h (TARGET_FORMAT): Set to "elf32-hppa-netbsd"
1401 for TE_NetBSD.
1402
1403 2004-07-27 Alan Modra <amodra@bigpond.net.au>
1404
1405 * config/tc-ppc.c (ppc_frob_file_before_adjust): Warn if .toc too big.
1406 (ppc_arch): Expand comment.
1407
1408 2004-07-27 Tomer Levi <Tomer.Levi@nsc.com>
1409
1410 * config/tc-crx.c: Support evaluating the difference between two
1411 symbols.
1412 * config/tc-crx.h: Likewise.
1413
1414 2004-07-26 H.J. Lu <hongjiu.lu@intel.com>
1415
1416 * config/tc-ia64.c (start_unwind_section): Set the linked-to
1417 section.
1418 (ia64_elf_section_change_hook): Set the linked-to section for
1419 SHT_IA_64_UNWIND.
1420
1421 2004-07-26 Dmitry Diky <diwil@spec.ru>
1422
1423 * config/tc-msp430.c: Add new subtargets: msp430x1610,
1424 msp430x1611, msp430x1612, msp430x415, msp430x417, msp430xG437,
1425 msp430xG438, msp430xG439.
1426
1427 2004-07-25 Daniel Jacobowitz <dan@debian.org>
1428
1429 * doc/as.texinfo (Section, PushSection): Correct documentation
1430 for ELF.
1431
1432 2004-07-21 Jan Beulich <jbeulich@novell.com>
1433
1434 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
1435 permissible for the selected instruction suffix.
1436 (match_template): Don't permit 64-bit general purpose operands in
1437 32-bit mode.
1438 (finalize_imm): Permit 64-bit immediates.
1439 (build_modrm_byte): Don't treat 32-bit addressing in 64-bit mode
1440 specially except for the width of the used base and/or index
1441 registers. For 32-bit displacements, use sign-extended
1442 relocations only when using 64-bit addressing.
1443 Force zero displacement on rip-relative addressing when there is
1444 no other displacement.
1445 (i386_index_check): Don't treat 32-bit addressing in 64-bit mode
1446 specially except for the width of the used base and/or index
1447 registers.
1448 (parse_register): Disallow Reg64 registers in 32-bit mode.
1449
1450 * config/tc-i386.c: For DefaultSize instructions, don't guess a 'q'
1451 suffix if the instruction doesn't support it.
1452
1453 2004-07-20 Maciej W. Rozycki <macro@linux-mips.org>
1454
1455 * config/tc-mips.c (append_insn): Handle constant expressions with
1456 no associated relocation.
1457 (mips_ip): Cancel the expression after use for the Q format
1458 specifier.
1459 (parse_relocation): Return no relocation for unsupported
1460 operators.
1461 (my_getSmallExpression): Return no relocation if no relocation
1462 operators are used.
1463
1464 2004-07-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1465
1466 * config/obj-som.c (adjust_stab_sections): Add prototype.
1467 (obj_som_compiler, obj_som_version, obj_som_copyright,
1468 adjust_stab_sections): Add ATTRIBUTE_UNUSED to unused arguments.
1469 * config/tc-hppa.c (update_subspace): Likewise.
1470 (is_defined_subspace): Amplify comment.
1471 * config/obj-som.h (som_frob_file): Add prototype.
1472
1473 2004-07-19 Christopher Faylor <cgf@timesys.com>
1474 H.J. Lu <hongjiu.lu@intel.com>
1475
1476 * subsegs.c (section_symbol): Don't create a new segment when
1477 existing segment is undefined.
1478
1479 2004-07-16 Richard Earnshaw <rearnsha@arm.com>
1480
1481 * config/tc-arm.c: Include include/opcode/arm.h.
1482 (ARM_EXT_*, ARM_ARCH_*, ARM_ANY, ARM_ALL, COPROC_ANY): Delete.
1483 (FPU_FPA_EXT_* FPU_VFP_EXT_*, FPU_ANY, FPU_NONE, FPU_MAVERICK): Delete.
1484 (FPU_ARCH_*): Delete.
1485 * Makefile.am: Update dependencies.
1486 * Makefile.in: Regenerate.
1487
1488 2004-07-15 Nick Clifton <nickc@redhat.com>
1489
1490 * configure.in: Accept armbe as a big-endian arm configuration.
1491 * configure: Regenerate.
1492
1493 2004-07-13 Thomas Nystrom <thn@saeab.se>
1494
1495 * config/tc-i386.c (T_SHIFTOP): New constant.
1496 (intel_e05_1): Handle '&', '|' and T_SHIFTOP.
1497 (intel_el1): Handle '~'.
1498 (intel_get_token): Handle '<>', '&', '|' and '~'.
1499
1500 2004-07-13 Nick Clifton <nickc@redhat.com>
1501
1502 (md_assemble): Remove spurious newline from end of as_bad error
1503 message.
1504 (intel_e05_1): Likewise.
1505 (intel_e11): Likewise.
1506 (intel_match_token): Likewise.
1507
1508 2004-07-11 Andreas Schwab <schwab@suse.de>
1509
1510 * config/tc-m68k.c: Convert to C90. Remove redundant
1511 declarations. Indentation fixup.
1512 [M68KCOFF]: Include "obj-coff.h" instead of declaring
1513 obj_coff_section ourselves.
1514
1515 2004-07-09 James E Wilson <wilson@specifixinc.com>
1516
1517 * config/tc-ia64.c (default_big_endian): New.
1518 (dot_byteorder, md_begin): Use it.
1519 (md_parse_option): Set it.
1520
1521 2004-07-09 Nick Clifton <nickc@redhat.com>
1522
1523 * configure.in: Change sh-sybmian-elf to sh-*-symbianelf.
1524 * configure: Regenerate.
1525 * NEWS: Change sh-sybmian-elf to sh-*-symbianelf.
1526 * config/tc-sh.c (sh_elf_final_processing): Use renamed version of
1527 sh_find_elf_flags if necessary.
1528
1529 2004-07-08 Richard Sandiford <rsandifo@redhat.com>
1530
1531 * config/tc-mips.c (mips_fix_adjustable): If the full addend is
1532 going to be split into more than one in-place addend, return 0
1533 for relocations against mergeable sections. Associate comments
1534 with code.
1535
1536 2004-07-07 Tomer Levi <Tomer.Levi@nsc.com>
1537
1538 * Makefile.am (CPU_TYPES): Add crx.
1539 (TARGET_CPU_CFILES): Add config/tc-crx.c.
1540 (TARGET_CPU_HFILES): Add config/tc-crx.h.
1541 (DEPTC_crx_elf): New target.
1542 (DEPOBJ_crx_elf): Likewise.
1543 (DEP_crx_elf): Likewise.
1544 * Makefile.in: Regenerate.
1545 * configure.in: Add crx* target.
1546 * configure: Regenerate.
1547 * config/tc-crx.c: New file.
1548 * config/tc-crx.h: New file.
1549 * NEWS: Mention new target.
1550
1551 2004-07-06 Nick Clifton <nickc@redhat.com>
1552
1553 * config.in: Undefine TARGET_SYMBIAN by default.
1554 * configure.in:
1555 * configure: Regenerate. Add sh-symbian-elf target. If
1556 selected define TARGET_SYMBIAN.
1557 * config/tc-sh.h (TARGET_FORMAT): Select a Symbian target
1558 format if TARGET_SYMBIAN has been defined.
1559
1560 * output-file.c (output_file_create): Report the target format
1561 chosen when bfd_openw reports that it is invalid.
1562
1563 * config/obj-coff.c (coff_pseudo_table): Only define the weak
1564 pseudo for BFD based assemblers.
1565
1566 2004-07-05 Andrew Stubbs <andrew.stubbs@superh.com>
1567
1568 gas:
1569 * config/tc-sh.c (md_assemble): Change isspace to ISSPACE.
1570 (md_parse_option): Remove redundant -isa testing.
1571 Make bfd_arch variable const.
1572 (md_show_usage): Make bfd_arch variable const.
1573
1574 2004-07-03 James E Wilson <wilson@specifixinc.com>
1575
1576 * config/tc-ia64.c (emit_one_bundle): Check and set insn_addr.
1577 * config/tc-ia64.h (md_frag_check): Define.
1578
1579 2004-07-03 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
1580
1581 * config/obj-coff.c (obj_coff_weak): New .weak syntax for PE weak
1582 externals.
1583 * doc/as.texinfo (Weak): Document PE weak symbols.
1584
1585 2004-07-03 Richard Sandiford <rsandifo@redhat.com>
1586
1587 * config/tc-mips.c (HAVE_IN_PLACE_ADDENDS): New macro.
1588 (reloc_needs_lo_p): Only return true if HAVE_IN_PLACE_ADDENDS.
1589 (mips_frob_file): Rework so that only a single pass through the
1590 relocs is needed. Allow %lo()s to have higher offsets than their
1591 corresponding %hi()s or %got()s.
1592
1593 2004-07-02 Nick Clifton <nickc@redhat.com>
1594
1595 * config/tc-arm.c (md_apply_fix3:BFD_RELOC_ARM_IMMEDIATE): Do not
1596 allow values which have come from undefined symbols.
1597 Always consider this fixup to have been processed as a reloc
1598 cannot be generated for it.
1599
1600 2004-07-02 Alan Modra <amodra@bigpond.net.au>
1601
1602 * frags.h (struct frag): Add has_code and insn_addr fields.
1603 * write.c (cvt_frag_to_fill): Invoke md_frag_check.
1604 * config/tc-ppc.c (md_assemble): Check and set insn_addr.
1605 * config/tc-ppc.h (md_frag_check): Define.
1606
1607 2004-06-28 Maciej W. Rozycki <macro@linux-mips.org>
1608
1609 * doc/Makefile.am (info): Rename goal to...
1610 (info-local): ... this, to preserve implicit dependencies.
1611 * doc/Makefile.in: Regenerate with automake 1.8.5.
1612
1613 2004-06-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1614
1615 * config/tc-m32r.c (md_convert_frag): Changed for @PLT.
1616 (m32r_cgen_record_fixup_exp): Changed for @GOTOFF, @GOT.
1617 (m32r_fix_adjustable): Changed for @GOTOFF, @GOT, @PLT.
1618 (tc_gen_reloc): Likewise.
1619 (m32r_end_of_match): Add for @GOTOFF, @GOT, @PLT.
1620 (m32r_parse_name): Likewise.
1621 (m32r_cgen_parse_fix_exp): Likewise.
1622 * config/tc-m32r.h (md_parse_name): Define for @GOTOFF, @GOT, @PLT.
1623 (O_PIC_reloc): Likewise.
1624 (TC_CGEN_PARSE_FIX_EXP): Likewise..
1625 * cgen.c (gas_cgen_parse_operand): Add TC_CGEN_PARSE_FIX_EXP
1626 for @GOTOFF, @GOT, @PLT.
1627
1628 2004-06-21 Jan Beulich <jbeulich@novell.com>
1629
1630 * gas/symbols.c: While discarding ordinary local absolute symbols
1631 when --strip-local-absolute is in effect, retain file symbols.
1632
1633 2004-06-20 Andreas Schwab <schwab@suse.de>
1634
1635 * config/tc-m68k.c (mri_chip): Replace current_chip, not augment.
1636 (md_parse_option): Likewise.
1637
1638 2004-06-17 Jan Beulich <jbeulich@novell.com>
1639
1640 * config/tc-i386.c: Deal with LEX_QM the same way as with LEX_AT.
1641 * config/te-netware.h: New file.
1642 * config/te-ppcnw.h: Delete: Obsolete.
1643 * configure.in: Eliminate ill NetWare targets. Make generic
1644 NetWare target use proper emulation.
1645 * Makefile.am: Eliminate reference to obsolete te-ppcnw.h, add
1646 reference to new te-netware.h.
1647 * configure: Regenerate.
1648 * Makefile.in: Regenerate.
1649
1650 2004-06-15 Martin Schwidefsky <schwidefsky@de.ibm.com>
1651
1652 * config/tc-s390.c (s390_insn): Avoid incorrect signed/unsigned
1653 comparison in .insn pseudo operation.
1654
1655 2004-06-15 Alan Modra <amodra@bigpond.net.au>
1656
1657 * config/obj-coff.c (coff_adjust_section_syms): Use
1658 bfd_get_section_size instead of bfd_get_section_size_before_reloc.
1659 (coff_frob_section): Likewise.
1660 * config/tc-mips.c (md_apply_fix3): Likewise.
1661 * config/obj-elf.c (elf_frob_file): Use bfd_set_section_size.
1662 (elf_frob_file_after_relocs): Likewise.
1663
1664 2004-06-10 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1665
1666 * config/tc-hppa.c (log2): Rename to exact_log2.
1667 (pa_next_subseg): Delete unused function.
1668 (create_new_space): Mark unused arguments with ATTRIBUTE_UNUSED.
1669 (create_new_subspace): Likewise.
1670
1671 Bug gas/213
1672 * config/tc-hppa.c (hppa_fix_adjustable): Allow reduction of fake
1673 labels. Fix warning.
1674
1675 2004-05-28 DJ Delorie <dj@redhat.com>
1676
1677 * config/tc-mn10300.h (tc_fix_adjustable): Define.
1678 * config/tc-mn10300.c (mn10300_fix_adjustable): Don't adjust debug
1679 or non-merged symbols.
1680
1681 2004-05-28 H.J. Lu <hongjiu.lu@intel.com>
1682
1683 * config/tc-ia64.c (remove_marked_resource): Save, clear and
1684 restore the old slot when inserting srlz.i/srlz.d.
1685
1686 2004-05-28 Andrew Stubbs <andrew.stubbs@superh.com>
1687
1688 * Makefile.am: Regenerate dependecies.
1689 * Makefile.in: Regenerate.
1690 * config/tc-sh.c (valid_arch): Make unsigned.
1691 (preset_target_arch): Likewise.
1692 (md_begin): Use new architecture flags system.
1693 (get_specific): Likewise.
1694 (assemble_ppi): Likewise.
1695 (md_assemble): Likewise. Also fix error check for bad opcodes.
1696 (md_parse_option): Likewise. Also generate -isa values according
1697 to the table in bfd/cpu-sh.c instead of just constants. Also
1698 allow <arch>-up ISA variants.
1699 (sh_elf_final_processing): Replace if-else chain with a call to
1700 sh_find_elf_flags().
1701 * testsuite/gas/sh/arch: New directory.
1702 * testsuite/gas/sh/arch/arch.exp: New test script.
1703 * testsuite/gas/sh/arch/arch_expected.txt: New file.
1704 * testsuite/gas/sh/arch/sh.s: New file.
1705 * testsuite/gas/sh/arch/sh2.s: New file.
1706 * testsuite/gas/sh/arch/sh-dsp.s: New file.
1707 * testsuite/gas/sh/arch/sh2e.s: New file.
1708 * testsuite/gas/sh/arch/sh3-nommu.s: New file.
1709 * testsuite/gas/sh/arch/sh3.s: New file.
1710 * testsuite/gas/sh/arch/sh3-dsp.s: New file.
1711 * testsuite/gas/sh/arch/sh3e.s: New file.
1712 * testsuite/gas/sh/arch/sh4-nommu-nofpu.s: New file.
1713 * testsuite/gas/sh/arch/sh4-nofpu.s: New file.
1714 * testsuite/gas/sh/arch/sh4.s: New file.
1715 * testsuite/gas/sh/arch/sh4a-nofpu.s: New file.
1716 * testsuite/gas/sh/arch/sh4al-dsp.s: New file.
1717 * testsuite/gas/sh/arch/sh4a.s: New file.
1718
1719 2004-05-28 Peter Barada <peter@the-baradas.com>
1720
1721 * config/gc-m68k.c(m68k_ip): Convert mode 5 addressing
1722 with zero offset into mode 2 addressing to save a word.
1723
1724 2004-05-27 H.J. Lu <hongjiu.lu@intel.com>
1725
1726 * config/tc-ia64.c (ar_is_in_integer_unit): Removed.
1727 (ar_is_only_in_integer_unit): New.
1728 (ar_is_only_in_memory_unit): New.
1729 (generate_unwind_image): Silence gcc on 32bit host.
1730 (md_assemble): Use ar_is_only_in_integer_unit instead of
1731 ar_is_in_integer_unit. Check AR access.
1732
1733 2004-05-27 Peter Barada <peter@the-baradas.com>
1734
1735 * config/tc-m68k.c (md_begin): Sort the opcode table into
1736 alphabetical order.
1737 (m68k_compare_opcode): New function to do the sorting.
1738
1739 2004-05-24 Peter Barada <peter@the-baradas.com>
1740
1741 * config/m68k-parse.y(operand): Allow for MAC/EMAC mask
1742 addressing on MIT style operands.
1743 * config/m68k-parse.y(yylex): Allow '-&' for predecrement
1744 w/mask addressing.
1745 * config/tc-m68k.c(install_operand): Comment 'G' and 'H' type
1746 operands.
1747
1748 2004-05-23 Alan Modra <amodra@bigpond.net.au>
1749
1750 * expr.c (operand, operator): Don't reject '++' and '--'.
1751
1752 2004-05-20 Richard Sandiford <rsandifo@redhat.com>
1753
1754 * config/tc-mips.c (append_insn): Use ISA-encoded addresses in MIPS16
1755 dwarf tables.
1756
1757 2004-05-17 Adam Nemet <anemet@lnxw.com>
1758
1759 * configure.in: Add ppc-*-lynxos*. Update i386-*-lynxos* to ELF.
1760 * configure: Regenerate.
1761
1762 2004-05-13 Paul Brook <paul@codesourcery.com>
1763
1764 * dw2gencfi.c (output_cie): Handle dwarf3 format CIE entries.
1765
1766 2004-05-13 Joel Sherrill <joel@oarcorp.com>
1767
1768 * configure.in (or32-*-rtems*): Switch to elf.
1769 * configure: Regenerate.
1770
1771 2004-05-13 Nick Clifton <nickc@redhat.com>
1772
1773 * po/fr.po: Updated French translation.
1774
1775 2004-05-11 Nick Clifton <nickc@redhat.com>
1776
1777 * doc/as.texinfo (Section): Document G and T flags to .section
1778 directive. Document the extra arguments that the G flag
1779 requires. Document the #tls flag.
1780
1781 2004-05-11 H.J. Lu <hongjiu.lu@intel.com>
1782
1783 * subsegs.c (section_symbol): Create a new section symbol if
1784 the existing one doesn't match.
1785 * symbols.c (symbol_set_bfdsym): Don't reset BFD section symbol.
1786
1787 2004-05-07 Richard Sandiford <rsandifo@redhat.com>
1788
1789 * config/tc-mips.c (append_insn, mips_emit_delays): Extend -mfix-vr4120
1790 to cope with VR4181A errata MD(1) and MD(4).
1791
1792 2004-05-07 Brian Ford <ford@vss.fsi.com>
1793
1794 * NEWS: Mention .secrel32 for pe[i]-i386.
1795
1796 2004-05-07 Alexandre Oliva <aoliva@redhat.com>
1797
1798 * config/tc-frv.h (MAX_MEM_FOR_RS_ALIGN_CODE): New.
1799 (HANDLE_ALIGN): New.
1800
1801 2004-05-06 Daniel Jacobowitz <dan@debian.org>
1802
1803 * Makefile.am (DIST_SUBDIRS): Define.
1804 * aclocal.m4: Regenerate with automake 1.8.4.
1805 * Makefile.in: Likewise.
1806 * doc/Makefile.in: Likewise.
1807
1808 2004-05-06 David Mosberger-Tang <davidm@hpl.hp.com>
1809
1810 * config/tc-ia64.c (dot_serialize): Declare.
1811 (dot_serialize): New function.
1812 (md_pseudo_table): Add ".serialize.data" and
1813 ".serialize.instruction" directives.
1814
1815 2004-05-06 Nick Clifton <nickc@redhat.com>
1816
1817 * messages (as_internal_value_out_of_range): Print a message about
1818 a value being out of range. Be consistent about whether the
1819 values are printed in decimal or hexadecimal.
1820 (as_warn_value_out_of_range): Generate a warning message about an
1821 out of range value.
1822 (as_bad_value_out_of_range): Generate an error message about an
1823 out of range value.
1824 * as.h: Prototype the new functions.
1825 * config/tc-alpha.c (insert_operand): Use new function.
1826 * config/tc-arc.c (arc_insert_operand): Likewise.
1827 * config/tc-mn10200.c (mn10200_insert_operand): Likewise.
1828 * config/tc-mn10300.c (mn10300_insert_operand): Likewise.
1829 * config/tc-ppc.c (ppc_insert_operand): Likewise.
1830 * config/tc-s390.c (s390_insert_operand): Likewise.
1831 * config/tc-v850.c (v850_insert_operand): Likewise.
1832
1833 2004-05-05 Alexandre Oliva <aoliva@redhat.com>
1834
1835 * configure.in: Set em=linux for frv-*-*linux*.
1836 * configure: Rebuilt.
1837 * config/tc-frv.h (TARGET_FORMAT): Use elf32-frvfdpic if...
1838 (frv_md_fdpic_enabled): New.
1839 * config/tc-frv.c (frv_md_fdpic_enabled): New.
1840 (DEFAULT_FDPIC): New.
1841 (frv_flags): Use DEFAULT_FDPIC.
1842 (frv_pic_flag): Likewise.
1843 (OPTION_NOPIC): New.
1844 (md_longopts): Add -mnopic.
1845 (md_parse_option): Handle it.
1846 (md_show_usage): Add -mfdpic and -mnopic.
1847
1848 2004-05-05 Peter Barada <peter@the-baradas.com>
1849
1850 * config/tc-m68k.c: Add find_cf_chip to print list of valid
1851 chips for invalid coldfire instructions, rename selectors
1852 for ColdFire sub-variants, add 521x,5249,547x,548x and aliases,
1853 add current_chip to track which chip is referred to(including save/restore),
1854 use current_chip to select control registers, not current_arch.
1855 (md_show_usage): Add new chips.
1856 * doc/c-m68k.texi: Document new command line switches.
1857
1858 2004-05-05 Jakub Jelinek <jakub@redhat.com>
1859
1860 * tc-s390.h (md_do_align, HANDLE_ALIGN): Remove.
1861 (NOP_OPCODE): Define.
1862 (s390_align_code): Remove prototype.
1863 * tc-s390.c (s390_align_code): Remove.
1864
1865 2004-05-04 H.J. Lu <hongjiu.lu@intel.com>
1866
1867 * config/tc-ia64.c (make_unw_section_name): Removed.
1868 (start_unwind_section): New function.
1869 (generate_unwind_image): Take const segT instead of const
1870 char *.
1871 (dot_handlerdata): Adjusted.
1872 (dot_endp): Likewise.
1873
1874 2004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1875
1876 * config/obj-elf.c (obj_elf_change_section): Allow the
1877 ".note.GNU-stack" section has SHF_EXECINSTR.
1878
1879 2004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1880
1881 * config/obj-elf.c (get_section): Return bfd_boolean.
1882 (obj_elf_change_section): Call bfd_get_section_by_name_if
1883 instead of bfd_map_over_sections.
1884
1885 2004-04-30 H.J. Lu <hongjiu.lu@intel.com>
1886
1887 * config/obj-elf.c (get_section): New function.
1888 (obj_elf_change_section): Support multiple sections with same
1889 name.
1890
1891 2004-04-30 Nick Clifton <nickc@redhat.com>
1892
1893 * config/tc-arm.c (create_register_alias): Fix typo checking for
1894 case sensitive register aliases.
1895 (co_proc_number): Use error message string in all_reg_maps[]
1896 array.
1897 (cp_reg_required_here): Likewise.
1898 (fp_reg_required_here): Likewise.
1899
1900 2004-04-29 Brian Ford <ford@vss.fsi.com>
1901
1902 * dwarf2dbg.c (dwarf2_finish): Add SEC_DEBUGGING to section flags.
1903
1904 2004-04-28 Chris Demetriou <cgd@broadcom.com>
1905
1906 * config/tc-mips.c (HAVE_32BIT_ADDRESSES, append_insn, macro_build)
1907 (load_address, macro, mips_ip, md_parse_option)
1908 (mips_force_relocation, mips_validate_fix, md_apply_fix3)
1909 (s_change_sec, pic_need_relax, tc_gen_reloc): Remove all
1910 embedded-PIC handling, and update comments.
1911 (SWITCH_TABLE): Remove.
1912 * config/tc-mips.h (DIFF_EXPR_OK): Delete.
1913 (enum mips_pic_level): Remove EMBEDDED_PIC.
1914 (EXTERN_FORCE_RELOC): Remove embedded-PIC handling.
1915 (TC_FORCE_RELOCATION): Update comment.
1916 * ecoff.c (ecoff_build_lineno): Add comment about some code that
1917 might be safe to remove now that MIPS embedded-PIC is gone.
1918
1919 2004-04-28 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1920
1921 * config/obj-som.c (obj_som_init_stab_section): Add new arguments in
1922 call to obj_set_subsection_attributes.
1923 (obj_som_init_stab_section): Likewise.
1924 * config/tc-hppa.c (default_subspace_dict): Add comdat field.
1925 (pa_def_subspaces): Provide comdat default.
1926 (pa_subspace): Handle new "comdat" parameter. Set SEC_LINK_ONCE and
1927 not SEC_IS_COMMON if section is comdat, common or dup_common. Update
1928 calls to create_new_subspace and update_subspace to pass comdat flag.
1929 (create_new_subspace, update_subspace): Add new comdat argument. Use
1930 it in calls to obj_set_subsection_attributes.
1931 * doc/c-hppa.texi (.subspa, .nsubspa): Document new comdat parameter
1932 and use of comdat, common and dup_comm parameters.
1933
1934 2004-04-26 H.J. Lu <hongjiu.lu@intel.com>
1935
1936 * config/obj-elf.c (obj_elf_change_section): Check if the old
1937 group name is NULL before comparison.
1938
1939 2004-04-23 Chris Demetriou <cgd@broadcom.com>
1940
1941 * config/tc-mips.h (mips_dwarf2_addr_size): Prototype.
1942
1943 2004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1944
1945 * config/tc-mips.c (s_mipsset): Set default CPU type for .set mipsN.
1946
1947 2004-04-23 Chris Demetriou <cgd@broadcom.com>
1948
1949 * config/tc-mips.c (md_longopts): Remove -membedded-pic option.
1950 (OPTION_MEMBEDDED_PIC): Remove.
1951 (OPTION_TRAP, OPTION_BREAK, OPTION_EB, OPTION_EL)
1952 (OPTION_FP32, OPTION_GP32, OPTION_CONSTRUCT_FLOATS)
1953 (OPTION_NO_CONSTRUCT_FLOATS, OPTIONS_FP64, OPTION_GP64)
1954 (OPTION_RELAX_BRANCH, OPTION_NO_RELAX_BRANCH)
1955 (OPTION_ELF_BASE): Renumber.
1956 (md_parse_option): Remove OPTION_MEMBEDDED_PIC handling.
1957 (md_show_usage): Remove mention of -membedded-pic.
1958 * doc/as.texinfo: Remove mention of -membedded-pic.
1959
1960 2004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1961
1962 * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Remove.
1963 * config/tc-mips.c (RDATA_SECTION_NAME, mips_target_format): Remove
1964 a.out support.
1965 (md_begin, mips_ip, md_parse_option, s_change_sec, s_option,
1966 s_abicalls, nopic_need_relax, tc_gen_reloc): Remove uses of
1967 USE_GLOBAL_POINTER_OPT.
1968
1969 2004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1970
1971 * config/tc-mips.c (macro): One more use of load_delay_nop.
1972
1973 2004-04-22 Atsushi Nemoto <anemo@mba.ocn.ne.jp>
1974
1975 * config/tc-mips.c (load_delay_nop): New function.
1976 (load_address, macro): Use load_delay_nop() to build a nop
1977 which can be omitted with gpr_interlocks.
1978
1979 2004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1980
1981 * config/tc-mips.c (hilo_interlocks, gpr_interlocks,
1982 cop_interlocks): Remove superfluous CPU entries.
1983
1984 2004-04-22 Paul Brook <paul@codesourcery.com>
1985
1986 * config/tc-arm.c (mav_parse_offset): Value must be multiple of 4.
1987
1988 2004-04-22 Peter Barada <peter@the-baradas.com>
1989
1990 * NEWS: Added support for EMAC instructions and MAC/EMAC
1991 Motorola syntax.
1992 * config/m68k-parse.h: Add ACC[123], ACCEXT{01,23}, MAC/EMAC
1993 scale factor tokens, trailing_ampersand to mark mask addressing
1994 for MAC/EMAC instructions.
1995 * config/m68k-parse.y: Add options_ampersand clause, '<<',
1996 '>>'.
1997 (yylex): Handle '>', '<', and '&' following '+'.
1998 * config/tc-m68k.c: Set mcfmac/mcfemac on appropriate ColdFire
1999 architectures in archs[].
2000 (m68k-ip): Add '4', 'e', 'g', 'i', cases to handle mask addressing
2001 for MAC/EMAC instructions, ACC[0123], ACCEXT{01,23}, and '<<'/'>>'
2002 respectively.
2003 (m68k_ip): Handle trailing '&' on MAC/EMAC insns.
2004 (install_operand): Fix 'n' case, Add 'F', 'f', 'G', 'H', 'I', ']'
2005 cases.
2006 Add EMAC operands to init_table[].
2007
2008 2004-04-22 Bruno De Bus <bdebus@elis.ugent.be>
2009
2010 * config/tc-arm.h (enum mstate): Move here, add MAP_UNDEFINED
2011 state.
2012 (TC_SEGMENT_INFO_TYPE): Define to enum mstate.
2013 * config/tc-arm.c (enum mstate): Delete from here.
2014 (mapping_state): Remove the static mapstate variable and instead
2015 store the state in the segment. This allows a per-section mapping
2016 state. Handle and ignore MAP_UNDEFINED states.
2017 (arm_elf_change_section): Get the current mapping state from the
2018 new section.
2019 (s_ltorg): Set the mapping state to MAP_DATA.
2020 (arm_cleanup): Use arm_elf_change_section to get the mapping state
2021 for each pool as it is emitted.
2022
2023 2004-04-22 Nick Clifton <nickc@redhat.com>
2024
2025 * config/tc-arm.h: Formatting tidy ups.
2026
2027 2004-04-20 Chris Demetriou <cgd@broadcom.com>
2028
2029 * NEWS: Note that MIPS -membedded-pic option is deprecated.
2030
2031 2004-04-20 DJ Delorie <dj@redhat.com>
2032
2033 * config/tc-i386.h [TE_PE] (TC_CONS_FIX_NEW): Define.
2034 * config/tc-i386.c (md_pseudo_table) [TE_PE]: Add "secrel32".
2035 [TE_PE] (O_secrel): Define.
2036 [TE_PE] (x86_pe_cons_fix_new): New.
2037 [TE_PE] (pe_directive_secrel): Likewise.
2038 (tc_gen_reloc) [TE_PE]: Support BFD_RELOC_32_SECREL.
2039
2040 2004-04-19 Eric Christopher <echristo@redhat.com>
2041
2042 * config/tc-mips.c (mips_dwarf2_addr_size): Revert part
2043 of previous patch for fix in gcc.
2044
2045 2004-04-19 Jakub Jelinek <jakub@redhat.com>
2046
2047 * config/tc-xtensa.c (xg_assembler_literal): Fix a typo.
2048
2049 2004-04-19 Nathan Sidwell <nathan@codesourcery.com>
2050
2051 * read.c (do_align): Call md_flush_pending_output, if defined.
2052
2053 2004-04-16 Alan Modra <amodra@bigpond.net.au>
2054
2055 * expr.c (operand): Correct checks for ++ and --.
2056
2057 2004-04-14 H.J. Lu <hongjiu.lu@intel.com>
2058
2059 * config/tc-generic.c: Add some comments.
2060
2061 2004-04-14 Richard Sandiford <rsandifo@redhat.com>
2062
2063 * doc/c-mips.texi (-m{no-,}fix-vr4120): Renamed from
2064 -{no-}mfix-vr4122-bugs.
2065 * config/tc-mips.c (mips_fix_vr4120): Renamed from mips_fix_4122_bugs.
2066 (append_insn, mips_emit_delays): Update accordingly.
2067 (OPTION_FIX_VR4120, OPTION_NO_FIX_VR4120): Renamed from *VR4122.
2068 (md_longopts): Change -{no-,}mfix-vr4122-bugs to -m{no-,}fix-vr4120.
2069 (md_parse_option): Update after above changes.
2070 (md_show_usage): Add -mfix-vr4120.
2071
2072 2004-04-13 Bob Wilson <bob.wilson@acm.org>
2073
2074 * doc/as.texinfo (Sub-Sections): Conditionalize COFF-specific use
2075 of .section directive; add a reference to the ELF .subsection
2076 directive.
2077
2078 2004-04-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2079
2080 * config/tc-m32r.c (md_assemble): Fixed infinite loop bug
2081 in parallel.
2082
2083 2004-04-11 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
2084
2085 * Makefile.am: Remove mips from aout targets.
2086 * Makefile.in: Regenerate.
2087 * configure.in: Remove mips-dec-bsd* target.
2088 * configure: Regenerate.
2089
2090 2004-04-07 Alan Modra <amodra@bigpond.net.au>
2091
2092 PR 96
2093 * config/tc-ppc.c (ppc_elf_suffix): Add valid32 and valid64 fields
2094 to struct map_bfd. Adjust MAP macro, and define MAP32, MAP64.
2095 Update "mapping". Restrict some @ modifiers to 32 bit.
2096
2097 2004-04-01 Asgari Jinia <asgarij@kpitcummins.com>
2098 Dhananjay Deshpande <dhananjayd@kpitcummins.com>
2099
2100 * config/tc-sh.c (dont_adjust_reloc_32): New variable.
2101 (sh_fix_adjustable): Avoid adjusting BFD_RELOC_32 when
2102 dont_adjust_reloc_32 is set.
2103 (md_longopts): Add option -renesas.
2104 (md_parse_option, md_show_usage): Likewise.
2105 * doc/c-sh.texi: Likewise.
2106
2107 2004-04-01 Dave Korn <dk@artimi.com>
2108
2109 * config/tc-dlx.c (md_assemble): set fx_no_overflow flag for
2110 hi16 and lo16 fixS structs.
2111 (md_assemble): generate bit_fixS for RELOC_DLX_LO16 in
2112 exactly the same way as for RELOC_DLX_REL16.
2113 (machine_ip): properly respect LO flag in the_insn and
2114 output RELOC_DLX_LO16 rather than RELOC_DLX_16.
2115 (md_apply_fix3): apply RELOC_DLX_LO16.
2116
2117 2004-03-30 Stan Shebs <shebs@apple.com>
2118
2119 Remove long-obsolete MPW support.
2120 * mpw-config.in, mpw-make.sed, mac-as.r: Remove files.
2121 * configure.in: Remove mention of ppc-*-mpw* config.
2122 * configure.in: Likewise.
2123
2124 2004-03-30 Nick Clifton <nickc@redhat.com>
2125
2126 * config/tc-arm.c (meabi_flags): Make its use conditional upon
2127 OBJ_ELF being defined.
2128
2129 2004-03-27 Alan Modra <amodra@bigpond.net.au>
2130
2131 * config/obj-aout.c (obj_aout_type): Remove #ifdef BFD_ASSEMBLER code.
2132
2133 2004-03-23 Paul Brook <paul@codesourcery.com>
2134
2135 * config/tc-arm.c (meabi_flags): New variable.
2136 (arm_parse_eabi): New function.
2137 (md_begin): Set flags for EABI v3.
2138 (arm_eabis): Add.
2139 (arm_long_opts): Add meabi.
2140 * doc/as.texinf <ARM>: Document -meabi.
2141 * doc/c-arm.texi: Ditto.
2142
2143 2004-03-22 Bob Wilson <bob.wilson@acm.org>
2144
2145 * config/tc-xtensa.c (xtensa_post_relax_hook): Create literal
2146 tables even when use_literal_section flag is not set.
2147
2148 2004-03-22 Alan Modra <amodra@bigpond.net.au>
2149
2150 * config/tc-sh.c: Remove trailing whitespace.
2151
2152 2004-03-22 Hans-Peter Nilsson <hp@axis.com>
2153
2154 * doc/c-cris.texi (CRIS-Opts): Document --no-mul-bug-abort,
2155 --mul-bug-abort and the default behavior.
2156 * config/tc-cris.c (cris_insn_kind): New member CRIS_INSN_MUL.
2157 (err_for_dangerous_mul_placement): New variable.
2158 (STATE_MUL, OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): New
2159 macros.
2160 (md_cris_relax_table): Have placeholder for STATE_MUL.
2161 (md_longopts): New options --mul-bug-abort and --no-mul-bug-abort.
2162 (cris_relax_frag) <case ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: New
2163 case doing nothing.
2164 (md_estimate_size_before_relax) <case ENCODE_RELAX (STATE_MUL,
2165 STATE_BYTE)>: Ditto.
2166 (md_convert_frag) <ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: Check
2167 alignment and position of this frag, emit error message if
2168 suspicious.
2169 (md_assemble): For a multiply insn and when checking it,
2170 transform the current frag into a special frag for that purpose.
2171 (md_parse_option) <case OPTION_MULBUG_ABORT_OFF, case
2172 OPTION_MULBUG_ABORT_ON>: Handle new options.
2173
2174 2004-03-19 Bob Wilson <bob.wilson@acm.org>
2175
2176 * config/tc-xtensa.c (mark_literal_frags): New function.
2177 (xtensa_move_literals): Call mark_literal_frags for all literal
2178 segments, including init and fini literal segments.
2179 (xtensa_post_relax_hook): Swap use of xt_insn_sec and xt_literal_sec.
2180
2181 2004-03-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2182
2183 * tc-hppa.c (cons_fix_new_hppa): Check for PC relative base type.
2184 (pa_comm): Set BSF_OBJECT in symbol flags.
2185
2186 2004-03-19 Alan Modra <amodra@bigpond.net.au>
2187
2188 * Makefile.am: Run "make dep-am".
2189 * Makefile.in: Regenerate.
2190 * doc/Makefile.in: Regenerate.
2191 * config.in: Regenerate.
2192 * po/gas.pot: Regenerate.
2193
2194 2004-03-18 Nathan Sidwell <nathan@codesourcery.com>
2195
2196 * read.c (read_a_source_file): Use demand_empty_rest_of_line.
2197 (demand_empty_rest_of_line): Issue an error here.
2198 (ignore_rest_of_line): Silently skip to end.
2199 (demand_copy_string): Issue an error, not warning.
2200 (equals): Likewise.
2201 * config/obj-elf.c (obj_elf_section_name): Likewise.
2202 (obj_elf_section): Likewise.
2203 * config/tc-arc.c (arc_extoper): Remove bogus NULL checks.
2204 (arc_extinst): Likewise.
2205 * config/tc-ia64.c (dot_saveb): Use demand_empty_rest_of_line.
2206 (dot_spill): Likewise.
2207 (dot_unwabi): Likewise.
2208 (dot_prologue): Likewise.
2209
2210 * expr.c (operand): Reject ++ and --.
2211 (operator): Likewise.
2212
2213 2004-03-17 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2214
2215 * config/tc-sh.c: Include dw2gencfi.h.
2216 (sh_cfi_frame_initial_instructions): New function.
2217 (sh_regname_to_dw2regnum): Likewise.
2218 * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Move to the end of
2219 file.
2220 (TARGET_USE_CFIPOP): Define.
2221 (tc_cfi_frame_initial_instructions): Likewise.
2222 (tc_regname_to_dw2regnum): Likewise.
2223 (DWARF2_DEFAULT_RETURN_COLUMN, DWARF2_CIE_DATA_ALIGNMENT): Likewise.
2224 * Makefile.am: Update dependencies.
2225 * Makefile.in: Regenerate.
2226
2227 2004-03-17 Ralf Corsepius <corsepiu@faw.uni-ulm.de>
2228
2229 * configure.in: Switch sh-*-rtems* to ELF. Add sh-*-rtemscoff*.
2230 * configure: Regenerate.
2231
2232 2004-03-12 Bob Wilson <bob.wilson@acm.org>
2233
2234 * read.c (s_leb128): Call md_flush_pending_output.
2235
2236 2004-03-12 Michal Ludvig <mludvig@suse.cz>
2237
2238 * config/tc-i386.c (output_insn): Handle PadLock instructions.
2239 * config/tc-i386.h (CpuPadLock): New define.
2240 (CpuUnknownFlags): Added CpuPadLock.
2241
2242 2004-03-07 Andreas Schwab <schwab@suse.de>
2243
2244 * doc/c-hppa.texi (HPPA Directives): Fix typo.
2245
2246 2004-03-07 Richard Henderson <rth@redhat.com>
2247
2248 * dw2gencfi.c (output_cie): Align length to 4 byte boundary.
2249 (cfi_finish): Likewise for fde.
2250
2251 2004-03-05 H.J. Lu <hongjiu.lu@intel.com>
2252
2253 * config/tc-ia64.c (md_assemble): Properly handle NULL
2254 align_frag.
2255 (ia64_handle_align): Don't abort if failed to add a stop bit.
2256
2257 2004-03-04 H.J. Lu <hongjiu.lu@intel.com>
2258
2259 * Makefile.in: Regenerated.
2260 * aclocal.m4: Likewise.
2261 * configure: Likewise.
2262 * doc/Makefile.in: Likewise.
2263
2264 2004-03-03 H.J. Lu <hongjiu.lu@intel.com>
2265
2266 * config/tc-ia64.c (dot_align): New.
2267 (ia64_do_align): Make it static.
2268 (md_pseudo_table): Use "dot_align" for "align".
2269 (ia64_md_do_align): Don't set align_frag here.
2270 (ia64_handle_align): Add a stop bit to the previous bundle if
2271 needed.
2272
2273 * config/tc-ia64.h (ia64_do_align): Removed.
2274
2275 2003-03-03 Andrew Stubbs <andrew.stubbs@superh.com>
2276
2277 * config/tc-sh.c (md_parse_option): Add -isa=sh4-nofpu and
2278 -isa=sh4-nommu-nofpu options. Adjust help messages accordingly.
2279 (sh_elf_final_processing): Output BFD type sh4_nofpu if that is
2280 the most general type or the user specifically requested it.
2281 (md_assemble): Add a new error message for when an instruction
2282 is understood, but is not allowed due to an -isa option.
2283
2284 2004-03-02 H.J. Lu <hongjiu.lu@intel.com>
2285
2286 * config/tc-ia64.c (align_frag): New.
2287 (md_assemble): Set the tc_frag_data field in align_frag for
2288 IA64_OPCODE_FIRST instructions.
2289 (ia64_md_do_align): Set align_frag.
2290 (ia64_handle_align): Add a stop bit if needed.
2291
2292 * config/tc-ia64.h (TC_FRAG_TYPE): New.
2293 (TC_FRAG_INIT): New.
2294
2295 2004-03-01 Richard Sandiford <rsandifo@redhat.com>
2296
2297 * config/tc-frv.c (fr400_audio): New variable.
2298 (md_parse_option, md_show_usage): Add -mcpu=fr405 and -mcpu=fr450.
2299 (md_parse_option): Set fr400_audio for -mcpu=fr400 and -mcpu=fr405.
2300 (target_implements_insn_p): New function.
2301 (md_assemble): Report an error if the processor doesn't implement
2302 the instruction.
2303
2304 2004-02-27 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2305
2306 * config/tc-m32r.c (md_longopts): Added -no-bitinst option.
2307 (md_parse_option): Ditto.
2308 (OPTION_NO_SPECIAL_M32R): Added.
2309 (md_show_usage): Document it.
2310 (enable_speial_m32r): Changed a default value from 0 to 1.
2311 * doc/c-m32r.texi: Document the -no-bitinst option.
2312
2313 2004-02-27 Nick Clifton <nickc@redhat.com>
2314
2315 * config/tc-sh.c (get_operand): Revert previous delta.
2316 (tc_gen_reloc): Check for an unknown reloc type before processing
2317 the addend.
2318
2319 2004-02-27 Hannes Reinecke <hare@suse.de>
2320
2321 * config/tc-s390.c (s390_insn): Correct range check for opcode in
2322 .insn pseudo operation.
2323
2324 2004-02-27 Anil Paranjpe <anilp1@kpitcummins.com>
2325
2326 * config/tc-sh.c (get_operand): In case of #Imm, check has been
2327 added for wrong syntax.
2328
2329 2004-02-26 Eric Christopher <echristo@redhat.com>
2330
2331 * config/tc-mips.c (mips_dwarf2_addr_size): New.
2332 * config/tc-mips.h (DWARF2_ADDR_SIZE): Use.
2333
2334 2004-02-26 Andrew Stubbs <andrew.stubbs@superh.com>
2335
2336 * config/tc-sh.c (build_Mytes): Add REG_N_D and REG_N_B01
2337 nibble types to assembler.
2338
2339 2004-02-25 Fred Fish <fnf@redhat.com>
2340
2341 * config/tc-iq2000.c: Add missing \n\ in multiline string literal.
2342
2343 2004-02-20 James E Wilson <wilson@specifixinc.com>
2344
2345 * config/tc-ia64.c (slot_index): New arg before_relax. Use instead of
2346 finalize_syms.
2347 (fixup_unw_records): New arg before_relax. Pass to slot_index.
2348 (ia64_estimate_size_before_relax): New.
2349 (ia64_convert_frag): Pass 0 to fixup_unw_records. Add comment.
2350 (generate_unwind_image): Pass 1 to fixup_unw_records.
2351 * config/tc-ia64.h (ia64_estimate_size_before_relax): Declare.
2352 (md_estimate_size_before_relax): Call ia64_estimate_size_before_relax.
2353
2354 2004-02-19 Jakub Jelinek <jakub@redhat.com>
2355
2356 * stabs.c (generate_asm_file): Avoid warning about use of
2357 uninitialized variable.
2358
2359 2004-02-18 David Mosberger <davidm@hpl.hp.com>
2360
2361 * config/tc-ia64.c (ia64_flush_insns): In addition to prologue,
2362 body, and endp, allow unwind records which do not have a "t"
2363 (time/instruction) field.
2364
2365 2004-02-17 Petko Manolov <petkan@nucleusys.com>
2366
2367 * config/tc-arm.c (do_mav_dspsc_1): Correct offset of CRn.
2368 (do_mav_dspsc_2): Likewise.
2369 Fix accumulator registers move opcodes.
2370
2371 2004-02-13 Hannes Reinecke <hare@suse.de>
2372 Jakub Jelinek <jakub@redhat.com>
2373
2374 * dwarf2dbg.c (get_filenum): Do not read beyond allocated memory.
2375
2376 2004-02-10 Steve Ellcey <sje@cup.hp.com>
2377
2378 * config/tc-ia64.h (ia64_frob_symbol): New declaration.
2379 (tc_frob_symbol): New macro definition.
2380 * config/tc-ia64.c (ia64_frob_symbol): New routine.
2381
2382 2004-02-09 Daniel Jacobowitz <drow@mvista.com>
2383
2384 * config/tc-arm.c (md_begin): Mark .note.gnu.arm.ident as
2385 read-only.
2386
2387 2004-02-09 Nathan Sidwell <nathan@codesourcery.com>
2388
2389 * read.h (IGNORE_OPCODE_CASE): Do not define. Replace with ...
2390 (TC_CASE_SENSITIVE): ... this.
2391 * read.c: Replace IGNORE_OPCODE_CASE with TC_CASE_SENSITIVE.
2392 * doc/internals.texi (TC_CASE_SENSITIVE): Document.
2393
2394 2004-02-06 James E Wilson <wilson@specifixinc.com>
2395
2396 * config/tc-ia64.c (dot_endp): Delete call to output_endp.
2397 (generate_unwind_image): Re-add it here.
2398
2399 2004-02-06 Nathan Sidwell <nathan@codesourcery.com>
2400
2401 * dwarf2dbg.c (DWARF2_ADDR_SIZE): Remove trailing ';'
2402 * read.h (SKIP_WHITESPACE): Turn into an expression.
2403 * read.c (read_a_source_file): A pseudo is removed by having a
2404 NULL handler.
2405
2406 2004-02-05 James E Wilson <wilson@specifixinc.com>
2407
2408 * config/tc-ia64.c (output_endp): New.
2409 (count_bits): Delete.
2410 (ia64_flush_insns, process_one_record, optimize_unw_records): Handle
2411 endp unwind records.
2412 (fixup_unw_records): Handle endp unwind records. Delete code for
2413 shortening prologue regions not followed by a body record.
2414 (dot_endp): Call add_unwind_entry to emit endp unwind record.
2415 * config/tc-ia64.h (unw_record_type): Add endp.
2416
2417 2004-02-03 James E Wilson <wilson@specifixinc.com>
2418
2419 * config/tc-ia64.c (ia64_convert_frag): Call md_number_to_chars to
2420 fill padding bytes with zeroes.
2421 (emit_one_bundle): New locals last_ptr, end_ptr. Rewrite code that
2422 sets unwind_record slot_number and slot_frag fields.
2423
2424 2004-02-02 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2425
2426 * config/tc-mips.c (add_got_offset_hilo): New function.
2427 (macro): Use load_register() and add_got_offset_hilo() to load
2428 constants instead of hardcoding code sequences throughout.
2429
2430 2004-01-28 H.J. Lu <hongjiu.lu@intel.com>
2431
2432 * config/tc-ia64.c (emit_one_bundle): Add proper indentation.
2433
2434 2004-01-26 Bernardo Innocenti <bernie@develer.com>
2435
2436 * config/tc-m68k.h (EXTERN_FORCE_RELOC): Handle m68k-uclinux specially,
2437 like m68k-elf.
2438 * config/tc-m68k.c (RELAXABLE_SYMBOL): Use EXTERN_FORCE_RELOC instead
2439 of hard-coded test for TARGET_OS=elf.
2440
2441 2004-01-24 Chris Demetriou <cgd@broadcom.com>
2442
2443 * config/tc-mips.c (hilo_interlocks): Change definition
2444 so that MIPS32, MIPS64 and later ISAs are included, along with
2445 the already-included machines. Update comments.
2446
2447 2004-01-23 Daniel Jacobowitz <drow@mvista.com>
2448
2449 * config/tc-arm.c (tc_gen_reloc): Improve error message for
2450 undefined local labels.
2451
2452 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2453
2454 * config/tc-mips.c (load_address, macro): Update comments about
2455 NewABI GP relaxation.
2456
2457 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2458
2459 * config/tc-mips.c (macro_build): Remove place and counter arguments.
2460 (mips_build_lui, macro_build_ldst_constoffset): Likewise.
2461 (mips16_macro_build, macro_build_jalr): Remove counter argument.
2462 (set_at, load_register, load_address, move_register): Likewise.
2463 (load_got_offset, add_got_offset): Likewise.
2464 Update all calls and tidy accordingly.
2465
2466 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2467
2468 * config/tc-mips.c (RELAX_ENCODE): Remove WARN argument.
2469 (RELAX_FIRST, RELAX_SECOND): Turn into 8-bit quantities.
2470 (RELAX_USE_SECOND): Bump to 0x10000.
2471 (RELAX_SECOND_LONGER, RELAX_NOMACRO, RELAX_DELAY_SLOT): New flags.
2472 (mips_macro_warning): New variable.
2473 (md_assemble): Wrap macro expansion in macro_start() and macro_end().
2474 (s_cpload, s_cpsetup, s_cprestore, s_cpreturn): Likewise.
2475 (relax_close_frag): Set mips_macro_warning.first_frag. Adjust use
2476 of RELAX_ENCODE.
2477 (append_insn): Update mips_macro_warning.sizes.
2478 (macro_start, macro_warning, macro_end): New functions.
2479 (macro_build): Don't emit warnings here.
2480 (macro_build_lui, md_estimate_size_before_relax): ...or here.
2481 (md_convert_frag): Check for cases where one macro alternative
2482 needs a warning and the other doesn't. Emit a warning if the
2483 longer sequence was chosen.
2484
2485 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2486
2487 * config/tc-mips.h (tc_frag_data_type, TC_FRAG_TYPE): Remove.
2488 * config/tc-mips.c (RELAX_ENCODE): Take three arguments: the size of
2489 the first sequence, the size of the second sequence, and a flag
2490 that says whether we should warn.
2491 (RELAX_OLD, RELAX_NEW, RELAX_RELOC[123]): Delete.
2492 (RELAX_FIRST, RELAX_SECOND): New.
2493 (mips_relax): New variable.
2494 (relax_close_frag, relax_start, relax_switch, relax_end): New fns.
2495 (append_insn): Remove "place" argument. Use mips_relax.sequence
2496 rather than "place" to check whether we're expanding the second
2497 alternative of a relaxable macro. Remove redundant check for
2498 branch relaxation. If generating a normal insn, and there
2499 is not enough room in the current frag, call relax_close_frag()
2500 to close it. Update mips_relax.sizes[]. Emit fixups for the
2501 second version of a relaxable macro. Record the first relaxable
2502 fixup in mips_relax. Remove tc_gen_reloc workaround.
2503 (macro_build): Remove all uses of "place". Use mips_relax.sequence
2504 in the same way as in append_insn.
2505 (mips16_macro_build): Remove "place" argument.
2506 (macro_build_lui): As for macro_build. Don't drop the add_symbol
2507 when generating the second version of a relaxable macro.
2508 (load_got_offset, add_got_offset): New functions.
2509 (load_address, macro): Use new relaxation machinery. Remove
2510 tc_gen_reloc workarounds.
2511 (md_estimate_size_before_relax): Set RELAX_USE_SECOND if the second
2512 version of a relaxable macro is needed. Return -RELAX_SECOND if the
2513 first version is needed.
2514 (tc_gen_reloc): Remove relaxation handling.
2515 (md_convert_frag): Go through the fixups for a relaxable macro and
2516 mark those that belong to the unneeded alternative as done. If the
2517 second alternative is needed, adjust the fixup addresses to account
2518 for the deleted first alternative.
2519
2520 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2521
2522 * frags.h (frag_room): Declare.
2523 * frags.c (frag_room): New function.
2524 * doc/internals.texi: Document it.
2525
2526 2004-01-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
2527
2528 * config/tc-mips.c (append_insn): Don't do r3900 interlock
2529 optimization for -mtune=r3900, as this will break on other CPUs.
2530
2531 2004-01-11 Tom Rix <tcrix@worldnet.att.net>
2532
2533 * config/tc-m68hc11.c (build_indexed_byte): movb and movw cannot
2534 be relaxed, use fixup.
2535 (md_apply_fix3): Use 5 bit reloc from movb and movw fixup.
2536
2537 2004-01-19 Jakub Jelinek <jakub@redhat.com>
2538
2539 * config/tc-sparc.c (sparc_ip): Disallow %f32-%f63 for single
2540 precision operands.
2541
2542 2004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2543
2544 * config/tc-mips.c (append_insn): Properly detect variant frags
2545 that preclude swapping of relaxed branches. Correctly swap
2546 instructions between frags when dealing with relaxed branches.
2547
2548 2004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2549
2550 * acinclude.m4: Quote names of macros to be defined by AC_DEFUN
2551 throughout.
2552 * aclocal.m4: Regenerate.
2553 * configure: Regenerate.
2554
2555 2004-01-12 Anil Paranjpe <anilp1@KPITCummins.com>
2556
2557 * config/tc-h8300.c (build_bytes): Apply relaxation to bit
2558 manipulation insns.
2559
2560 2004-01-12 Richard Sandiford <rsandifo@redhat.com>
2561
2562 * config/tc-mips.c (macro_build_jalr): When adding an R_MIPS_JALR
2563 reloc, reserve space for the delay slot as well as the jalr itself.
2564
2565 2004-01-09 Paul Brook <paul@codesourcery.com>
2566
2567 * config/tc-arm.c (do_vfp_reg2_from_sp2): Rename from do_vfp_sp_reg2.
2568 (do_vfp_sp2_from_reg2): New function.
2569 (insns): Use them.
2570 (do_vfp_dp_from_reg2): Check return values properly.
2571
2572 2004-01-08 Ian Lance Taylor <ian@wasabisystems.com>
2573
2574 * config/tc-mips.c (warn_nops): Remove static variable.
2575 (macro): Remove test of warn_nops.
2576 (md_shortops): Remove 'n'.
2577 (md_parse_option): Remove 'n' case.
2578 (md_show_usage): Remove -n.
2579 * doc/as.texinfo (Overview): Remove MIPS -n option.
2580 * doc/c-mips.texi (MIPS Opts): Remove mention -n.
2581 * NEWS: Mention removal of MIPS -n option.
2582
2583 * config/tc-mips.c (ISA_HAS_COPROC_DELAYS): Remove.
2584 (cop_interlocks): Check ISA level.
2585 (cop_mem_interlocks): Define.
2586 (reg_needs_delay): Check cop_interlocks rather than
2587 ISA_HAS_COPROC_DELAYS.
2588 (append_insn): Likewise. Use cop_mem_interlocks rather than
2589 directly checking mips_opts.isa.
2590 (mips_emit_delays): Likewise.
2591
2592 2004-01-07 H.J. Lu <hongjiu.lu@intel.com>
2593
2594 * config/tc-ia64.c (unwind): Move next_slot_number and
2595 next_slot_frag to ...
2596 (unw_rec_list): Here.
2597 (free_list_records): Removed.
2598 (output_unw_records): Likewise.
2599 (generate_unwind_image): Make it void.
2600 (alloc_record): Initialize next_slot_number and next_slot_frag.
2601 (slot_index): Take .org, .space and .align into account.
2602 (fixup_unw_records): Don't set slot_number to 0. Use
2603 list->next_slot_number and list->next_slot_frag instead of
2604 unwind.next_slot_number and unwind.next_slot_frag.
2605 (ia64_convert_frag): New.
2606 (generate_unwind_image): Generate a rs_machine_dependent frag
2607 for unwind record.
2608 (emit_one_bundle): Use list->next_slot_number and
2609 list->next_slot_frag instead of unwind.next_slot_number and
2610 unwind.next_slot_frag.
2611
2612 * config/tc-ia64.h (md_convert_frag): Defined as
2613 ia64_convert_frag.
2614 (md_estimate_size_before_relax): Defined as (f)->fr_var.
2615
2616 2004-01-06 Alexandre Oliva <aoliva@redhat.com>
2617
2618 2003-12-19 Alexandre Oliva <aoliva@redhat.com>
2619 * config/tc-frv.h (md_apply_fix3): Don't define.
2620 * config/tc-frv.c (md_apply_fix3): New. Shift/truncate %hi/%lo
2621 operands.
2622 * config/tc-frv.h (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
2623 2003-10-07 Alexandre Oliva <aoliva@redhat.com>
2624 * config/tc-frv.c (line_separator_chars): Add `!'.
2625 2003-09-19 Alexandre Oliva <aoliva@redhat.com>
2626 * config/tc-frv.c (md_assemble): Clear insn upfront.
2627 2003-09-18 Alexandre Oliva <aoliva@redhat.com>
2628 * config/tc-frv.c (OPTION_FDPIC): New macro.
2629 (md_longopts): Add mfdpic.
2630 (md_parse_option): Handle it.
2631 2003-08-04 Alexandre Oliva <aoliva@redhat.com>
2632 * config/tc-frv.c (md_cgen_lookup_reloc) <FRV_OPERAND_D12,
2633 FRV_OPERAND_S12>: Use reloc type encoded in fix-up.
2634 (frv_pic_ptr): Parse funcdesc.
2635
2636 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2637
2638 * doc/as.texinfo: Let texi2pod parse asconfig.texi and
2639 gasver.texi. Remove duplicate symbol definitions for texi2pod.
2640
2641 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2642
2643 * Makefile.am (Makefile): Move the dependency on
2644 $(BFDDIR)/configure.in to...
2645 (CONFIG_STATUS_DEPENDENCIES): ... here.
2646 (AUTOMAKE_OPTIONS): Require automake 1.8.
2647 * Makefile.in: Regenerate.
2648 * doc/Makefile.am (BASEDIR, BFDDIR): Define.
2649 (CONFIG_STATUS_DEPENDENCIES): Add a dependency on
2650 $(BFDDIR)/configure.in here as well.
2651 * doc/Makefile.in: Regenerate.
2652
2653 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2654
2655 * Makefile.am (install, install-info, RECURSIVE_TARGETS): Remove.
2656 * Makefile.in: Regenerate.
2657 * aclocal.m4: Regenerate.
2658 * doc/Makefile.am (install, install-info): Remove.
2659 (install-data-local): A new hook for install-info.
2660 (AUTOMAKE_OPTIONS): Require automake 1.8.
2661 * doc/Makefile.in: Regenerate.
2662
2663 2004-01-02 Nutan Singh <nutan@kpitcummins.com>
2664
2665 * doc/c-sh.texi: Update description about floating point behavior
2666 of SH family.
2667
2668 2004-01-02 Bernardo Innocenti <bernie@develer.com>
2669
2670 * configure.in: Add m68k-uClinux target.
2671 * configure: Regenerate.
2672
2673 For older changes see ChangeLog-0203
2674 \f
2675 Local Variables:
2676 mode: change-log
2677 left-margin: 8
2678 fill-column: 74
2679 version-control: never
2680 End:
This page took 0.085261 seconds and 4 git commands to generate.