projects / deliverable / binutils-gdb.git / blame
? search:
summary | shortlog | log | commit | commitdiff | tree
blob | blame (incremental) | history | HEAD
2004-11-05 Jon Beniston <jon@beniston.com>
[deliverable/binutils-gdb.git] / gas / ChangeLog
CommitLineData
1d19a770
BW		 12004-11-10 Bob Wilson <bob.wilson@acm.org>
2
3 * config/tc-xtensa.c (update_next_frag_state): Always add a NOP if
4 relaxing at the end of a loop. Don't mark frags as UNREACHABLE or
5 MAYBE_UNREACHABLE.
6 (relax_frag_immed): Update call to update_next_frag_state.
7
c04f5787
AM		 82004-11-10 Alan Modra <amodra@bigpond.net.au>
9
10 * obj.h (struct format_ops <app_file>): Add int param.
11 * read.h (s_app_file_string): Likewise.
12 * read.c (s_app_file_string): Likewise.
13 (s_app_file): Adjust s_app_file_string call.
14 * config/tc-mips.c (s_mips_file): Likewise.
15 * config/obj-coff.h (obj_app_file): Add app param.
16 * config/obj-ecoff.h (obj_app_file): Likewise.
17 * config/obj-multi.h (obj_app_file): Likewise.
18 * config/obj-elf.h (elf_file_symbol): Likewise.
19 * config/obj-elf.c (elf_file_symbol): Only emit one file symbol
20 if called for # preprocessor lines.
21
a77a9021
L		 222004-11-08 H.J. Lu <hongjiu.lu@intel.com>
23
24 PR 528
25 * symbols.c (resolve_symbol_value): Convert weak symbols only
26 for Windows PECOFF.
27 (symbol_equated_reloc_p): Don't equate weaks when relocating
28 only for Windows PECOFF.
29
aea44f62
TL		 302004-11-08 Tomer Levi <Tomer.Levi@nsc.com>
31
32 * config/tc-crx.c (print_insn): Check and set insn_addr.
33 * config/tc-crx.h (md_frag_check): Define.
34
7499d566
NC		 352004-11-08 Inderpreet Singh <inderpreetb@nioda.hcltech.com>
36 Vineet Sharma <vineets@noida.hcltech.com>
37
38 * configure.in: Add support for new target maxq-coff.
39 * configure: Regenerate.
40 * NEWS: Mention new support.
41 * config/tc-maxq.c: New file.
42 * config/tc-maxq.h: New file.
43 * config/obj-coff.h: Add support for maxq-coff.
44
977cdf5a
NC		 452004-11-08 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
46
47 * symbols.c (any_external_name): Define.
a77a9021 48 (resolve_symbol_value): Convert weak symbols.
977cdf5a
NC		 49 (S_SET_EXTERNAL): Support any_external_name.
50 (S_SET_NAME): Qualify parameter const.
316f3bf8 51 (symbol_equated_reloc_p): Equate to weaks when relocating.
977cdf5a
NC		 52 * symbols.h (S_SET_NAME): Qualfiy parameter const.
53 * tc.h (any_external_name): Declare.
54 * config/obj-coff.c ("coff/pe.h"): Include for BFD
55 assemblers also.
56 (weak_is_altname): Declare and define.
57 (weak_name2altname): Same.
58 (weak_altname2name): Same.
59 (weak_uniquify): Same.
60 (weak_altprefix): Define.
61 (obj_coff_weak): Change .weak syntax and handling.
62 (coff_frob_symbol): Fix PE weak symbol alternates.
63 * config/obj-coff.h (USE_UNIQUE): Define.
64 * config/tc-i386.c (md_apply_fix3): Assume weak symbols
65 are in another segment.
66 (tc_gen_reloc): Remove broken addend hack.
67 doc/as.texinfo: Update.
68
34e41783
BW		 692004-11-05 Sterling Augustine <sterling@tensilica.com>
70
71 * config/tc-xtensa.c (total_frag_text_expansion): New.
72 (md_estimate_size_before_relax): Use it.
73 (find_address_of_next_align_frag): Likewise.
74
dcd46b4e
TL		 752004-11-05 Tomer Levi <Tomer.Levi@nsc.com>
76
77 * config/tc-crx.c: Rename argument types.
78 (processing_arg_number): Rename to 'cur_arg_num'.
79 (get_number_of_bits): Rename to 'set_operand_size'.
80 (get_operandtype): Rename to 'parse_operand', totally rewrite.
81 (set_cons_rparams): Rename to 'set_operand', totally rewrite.
82 (set_indexmode_parameters): Remove function, integrate its code into
83 'set_operand'.
84 (set_operand_size): Get rid of 'Operand Number' function parameter -
85 use global variable 'cur_arg_num' instead.
86 Use a local 'argument' pointer to reference the current argument.
87 (parse_operand): Likewise.
88 (set_operand): Likewise.
89 (process_label_constant): Likewise.
90
9456465c
BW		 912004-11-04 Bob Wilson <bob.wilson@acm.org>
92
93 * config/tc-xtensa.c: Remove XTENSA_SECTION_RENAME ifdefs.
94 (add_section_rename): Delete. Inlined into...
95 (build_section_rename): ...here. Use xstrdup instead of strdup.
96 (xtensa_section_rename): Drop "const" from argument and return types.
97 (md_show_usage): Indent to match show_usage().
98 * config/tc-xtensa.h: Remove XTENSA_SECTION_RENAME ifdefs.
99 (tc_canonicalize_section_name): Define.
100 (md_elf_section_rename): Remove unused macro.
101 * doc/as.texinfo (Overview): Document Xtensa --rename-section option.
102 * doc/c-xtensa.texi (Xtensa Options): Likewise.
103 (Frame Directive): Delete.
104
7f266840
DJ		 1052004-11-04 Daniel Jacobowitz <dan@debian.org>
106
107 * configure.in: Remove arm-*-oabi and thumb-*-oabi.
108 * config/tc-arm.c (target_oabi): Delete.
109 (md_apply_fix3, elf32_arm_target_format): Remove target_oabi checks.
110 (arm_opts): Remove moabi.
111 * doc/as.texinfo (Overview): Remove documentation of -moabi.
112 * doc/c-arm.texi (ARM Options): Likewise.
113 * configure: Regenerated.
114
ae57792d
HPN		 1152004-11-04 Hans-Peter Nilsson <hp@axis.com>
116
117 * configure.in (crisv32): Recognize. AC_DEFINE_UNQUOTED
118 DEFAULT_CRIS_ARCH. Handle crisv32-*-linux-gnu* like
119 cris-*-linux-gnu* and crisv32-*-* like cris-*-*.
120 * configure: Regenerate.
121 * config/tc-cris.c (enum cris_archs): New.
122 (cris_mach, cris_arch_from_string, s_cris_arch, get_sup_reg)
123 (cris_insn_ver_valid_for_arch): New functions.
124 (DEFAULT_CRIS_ARCH): New macro, default to cris_any_v0_v10.
125 (cris_arch): New variable.
126 (md_pseudo_table): New pseudo .arch.
127 (err_for_dangerous_mul_placement): Initialize according to
128 DEFAULT_CRIS_ARCH.
129 (STATE_COND_BRANCH): Renamed from STATE_CONDITIONAL_BRANCH.
130 All users changed.
131 (STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON)
132 (STATE_ABS_BRANCH_V32, STATE_LAPC, BRANCH_BF_V32, BRANCH_BB_V32)
133 (BRANCH_WF_V32, BRANCH_WB_V32): New.
134 (BRANCH_BF, BRANCH_BB, BRANCH_WF, BRANCH_WB): Don't undef after
135 use in md_cris_relax_table.
136 (md_cris_relax_table): Add entries for STATE_COND_BRANCH_V32,
137 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
138 Update and improve head comment.
139 (OPTION_PIC): Define in terms of previous option, OPTION_US.
140 (OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): Similar.
141 (OPTION_ARCH): New.
142 (md_longopts): New option --march=...
143 (cris_any_v0_v10_long_jump_size, crisv32_long_jump_size): New
144 macros.
145 (md_long_jump_size): Initialize in terms of DEFAULT_CRIS_ARCH.
146 (HANDLE_RELAXABLE): New macro.
147 (md_estimate_size_before_relax): Use HANDLE_RELAXABLE for common
148 cases. Check for weak symbols and assume not relaxable. Handle
149 STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON,
150 STATE_ABS_BRANCH_V32, STATE_LAPC. Use new variable symbolP, not
151 fragP->fr_symbol.
152 (md_convert_frag): Handle STATE_COND_BRANCH_V32,
153 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
154 (cris_create_short_jump): Adjust for CRISv32.
5664e9a7 155 (cris_relax_frag): Handle new states.
ae57792d
HPN		 156 (md_create_long_jump): Ditto. Emit error for common_v10_v32.
157 (md_begin): Define symbols "..asm.arch.cris.v32",
158 "..asm.arch.cris.v10", "..asm.arch.cris.common_v10_v32" and
159 "..asm.arch.cris.any_v0_v10". Use cris_insn_ver_valid_for_arch
160 when entering opcode table entry points.
161 (md_assemble): Adjust branch handling for CRISv32. Handle LAPC
162 relaxation. In fix_new_exp call for main insn, pass 1 for pcrel
163 parameter for 8, 16 and 32-bit pc-relative insns and LAPC.
164 (cris_process_instruction): Initialize out_insnp->insn_type to
165 CRIS_INSN_NONE, not CRIS_INSN_NORMAL.
166 <case ']', '[', 'A', 'd', 'Q', 'N', 'n', 'Y', 'U', 'u', 'T'>: New
167 cases.
168 <case 'm'>: Check that modified_char == '.'.
169 <invalid operands>: Consume the rest of the line.
170 When operands don't match, skip over subsequent insns with
171 non-matching version specifier but same mnemonic.
172 <immediate constant, case SIZE_SPEC_REG>: Immediate operands for
173 special registers in CRISv32 are always 32 bit long.
174 <immediate constant, case SIZE_FIELD_SIGNED, SIZE_FIELD_UNSIGNED>:
175 New cases.
176 (get_gen_reg): Only recognize "PC" when followed by "+]" for v32
177 and compatible. Recognize "ACR" for v32, unless followed by "+".
178 (get_spec_reg): Consider cris_arch when looking up register.
179 (get_autoinc_prefix_or_indir_op): Don't recognize assignment for
180 v32 or compatible.
181 (get_3op_or_dip_prefix_op): Check for ']' after seeing '[rN+'.
182 (cris_get_expression): Restore input_line_pointer if failing "early".
183 (get_flags): Consider cris_arch and recognize flags accordingly.
184 (branch_disp): Adjust for CRISv32.
185 (gen_cond_branch_32): Similar. Emit error for common_v10_v32.
186 (cris_number_to_imm): Use as_bad_where, not as_bad. Remove
187 related FIXME. Don't insist on BFD_RELOC_32_PCREL fixup to be
188 resolved. Don't enter zeros in object file for
189 BFD_RELOC_32_PCREL.
190 <case BFD_RELOC_CRIS_LAPCQ_OFFSET, BFD_RELOC_CRIS_SIGNED_16>
191 <case BFD_RELOC_CRIS_SIGNED_8>: New case.
192 (md_parse_option): Break out "return 1".
193 <OPTION_ARCH> New case.
194 (tc_gen_reloc): <case BFD_RELOC_CRIS_LAPCQ_OFFSET>
195 <case BFD_RELOC_CRIS_SIGNED_16, BFD_RELOC_CRIS_SIGNED_8>
196 <case BFD_RELOC_CRIS_UNSIGNED_8, BFD_RELOC_CRIS_UNSIGNED_16>
197 <case BFD_RELOC_32_PCREL>: New cases.
198 Addends for non-zero fx_pcrel are too in fx_offset.
199 (md_show_usage): Show --march=<arch>.
200 (md_apply_fix3): Adjust val for BFD_RELOC_CRIS_LAPCQ_OFFSET.
201 (md_pcrel_from): BFD_RELOC_CRIS_LAPCQ_OFFSET is PC-relative too.
202 (s_syntax) <struct syntaxes>: Properly constify member operand.
203 * config/tc-cris.h (TARGET_MACH): Define.
204 (cris_mach): Declare.
205 * doc/as.texinfo (Overview) <CRIS>: Add --march=...
206 * doc/c-cris.texi (CRIS-Symbols): New node for built-in symbols.
207 (CRIS-Opts): Document --march=...
208 (CRIS-Pseudos): Document .arch.
209
9306ca4a
JB		 2102004-11-04 Jan Beulich <jbeulich@novell.com>
211
212 * config/tc-i386.c (set_intel_syntax): Allow % in symbol names when
213 intel syntax and no register prefix, allow $ in symbol names when
214 intel syntax.
215 (set_16bit_gcc_code_flag): Replace literal 'l' by LONG_MNEM_SUFFIX.
216 (intel_float_operand): Add fourth return value indicating math control
217 operations. Make classification more precise.
218 (md_assemble): Complain if memory operand of mov[sz]x has no size
219 specified.
220 (parse_insn): Translate word operands to floating point instructions
221 operating on integers as well as control instructions to short ones
222 as expected by AT&T syntax. Translate 'd' suffix to short one only for
223 floating point instructions operating on non-integer operands.
224 (match_template): Remove fldcw special case. Adjust q-suffix handling
225 to permit it on fild/fistp/fisttp in AT&T mode.
226 (process_suffix): Don't guess DefaultSize insns' suffix from
227 stackop_size for certain floating point control instructions. Guess
228 suffix for branch and [ls][gi]dt based on flag_code. Split error
229 messages for Intel and AT&T syntax, and make the condition more strict
230 for the former. Adjust suppressing of generation of operand size
231 overrides.
232 (intel parser): Allow the full set of MASM operators. Add FWORD, TBYTE,
233 OWORD, and XMMWORD operand size specifiers (TBYTE replaces XWORD). Add
234 more error checking.
235 * config/tc-i386.h (BYTE_PTR WORD_PTR DWORD_PTR QWORD_PTR XWORD_PTR
236 SHORT OFFSET_FLAT FLAT NONE_FOUND): Remove unused defines.
237
2b4f075a
HPN		 2382004-11-03 Hans-Peter Nilsson <hp@axis.com>
239
240 * symbols.c (colon) [!WORKING_DOT_WORD]: Don't declare
241 md_short_jump_size, md_long_jump_size.
242 * write.c [!WORKING_DOT_WORD]: Ditto.
243 * tc.h [!WORKING_DOT_WORD]: Declare them here. Drop const
244 qualifier.
245 * config/tc-cris.h (md_short_jump_size, md_long_jump_size): Don't
246 declare.
247 * config/tc-cris.c (md_short_jump_size, md_long_jump_size): Drop
248 const qualifier in these definitions.
249 * config/tc-i370.c, config/tc-m68k.c, config/tc-pdp11.c,
250 config/tc-s390.c, config/tc-tahoe.c, config/tc-vax.c: Ditto.
251
ddbc47ac
NC		 2522004-11-02 Nick Clifton <nickc@redhat.com>
253
254 * dwarf2dbg.c (dwarf2_finish): Check for the existence of a file
255 table before deciding to produce a .debug_line section to match up
256 with a user provided .debug_info section.
257
3ad3f5ad
TL		 2582004-10-28 Tomer Levi <Tomer.Levi@nsc.com>
259
260 * config/tc-crx.c (getreg_image): Bug fix, a return value was
261 mistakenly omitted from CRX_C_REGTYPE and CRX_CS_REGTYPE cases.
262
c815a616
TL		 2632004-10-27 Tomer Levi <Tomer.Levi@nsc.com>
264
265 * config/tc-crx.c: Remove global variable 'post_inc_mode'.
266 (get_flags): New function.
c04f5787
AM		 267 (get_number_of_bits): Edit comments, update numeric values to
268 supported sizes.
269 (process_label_constant): Don't support the colon
270 format (SYMBOL:[s|m|l]).
c815a616
TL		 271 (set_cons_rparams): Support argument type 'arg_rbase'.
272 (get_operandtype): Bug fix in 'rbase' operand type parsing.
273 (handle_LoadStor): Bug fix, first handle post-increment mode.
c04f5787
AM		 274 (getreg_image): Remove redundant code, update according to latest
275 CRX spec.
c815a616 276 (print_constant): Bug fix relate to 3-word instructions.
c04f5787 277 (assemble_insn): Bug fix, when matching instructions, verify also
c815a616
TL		 278 instruction type (not only mnemonic).
279 Add various error checking.
280 (preprocess_reglist): Support HI/LO and user registers.
281
ed7af9f9
L		 2822004-10-25 David Mosberger-Tang <davidm@hpl.hp.com>
283
c04f5787 284 * config/tc-ia64.c (fixup_unw_records): Don't let the "t" value
ed7af9f9
L		 285 in an epilogue directive go negative.
286
ac025970
L		 2872004-10-25 H.J. Lu <hongjiu.lu@intel.com>
288
289 PR 474
290 * config/tc-ia64.c (emit_one_bundle): Decrement md.num_slots_in_use
291 after reporting template error during manual bundling. Reported
292 by Michael Dupont, michaelx.dupont@intel.com.
293
a394c00f
NC		 2942004-10-25 Daniel Jacobowitz <dan@debian.org>
295
296 * Makefile.am: Run dep-am.
297 * aclocal.m4: Regenerate with automake 1.9.2.
298 * Makefile.in: Regenerate with automake 1.9.2.
299 * doc/Makefile.in: Likewise.
300
301 * config/tc-arm.c: Include "dw2gencfi.h".
302 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
303 New functions.
304 * config/tc-arm.h (TARGET_USE_CFIPOP, DWARF2_DEFAULT_RETURN_COLUMN)
305 (DWARF2_CIE_DATA_ALIGNMENT, tc_regname_to_dw2regnum)
306 (tc_cfi_frame_initial_instructions): Define.
307 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
308 Add prototypes.
309
23d83918
TL		 3102004-10-21 Tomer Levi <Tomer.Levi@nsc.com>
311
312 * config/tc-crx.c (assemble_insn): Check unsigned immediate
313 operands validity.
314 Update coprocessor id to be unsigned immediate.
315
bb41ade5
AM		 3162004-10-18 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
317
318 * config/tc-i386.c (O_secrel): Delete.
319 (tc_pe_dwarf2_emit_offset): New function.
320 * config/tc-i386.h (O_secrel): Define as O_md1.
321 (TC_DWARF2_EMIT_OFFSET): Define.
322
87e62348
NC		 3232004-10-18 Nick Clifton <nickc@redhat.com>
324
325 * config/tc-xstormy16.c (xstormy16_cons_fix_new): Accept and
326 ignore @fptr() directives for 4-byte fixups.
327
3c55da70
AM		 3282004-10-15 Alan Modra <amodra@bigpond.net.au>
329
330 * Makefile.am: Run "make dep-am"
331 * Makefile.in: Regenerate.
332
40e7b7b9
BW		 3332004-10-14 Bob Wilson <bob.wilson@acm.org>
334
335 * doc/c-xtensa.texi (Xtensa Options, Absolute Literals Directive):
336 Remove comments about placement of literal pools.
337 (Literal Directive): Update description of literal placement.
338 (Literal Prefix Directive): Remove statement that this does not apply
339 to absolute-mode literals. Describe new section naming scheme.
340
b08b5071
BW		 3412004-10-12 Bob Wilson <bob.wilson@acm.org>
342
343 * config/tc-xtensa.c (xtensa_frequency_pseudo): Use set_subseg_freq.
344 (is_entry_opcode, is_movi_opcode, is_the_loop_opcode, is_jx_opcode,
345 is_windowed_return_opcode): Delete.
346 (xtensa_frob_label): Use get_subseg_target_freq.
347 (md_assemble): Inline call to is_entry_opcode.
348 (xtensa_handle_align): Inline call to get_frag_is_literal.
349 (relaxation_requirements): Inline call to is_jx_opcode.
350 (emit_single_op): Inline call to is_movi_opcode.
351 (xg_assemble_vliw_tokens): Inline calls to get_frag_is_insn,
352 get_frag_is_no_transform, is_entry_opcode, and
353 set_frag_is_specific_opcode. Use get_subseg_total_freq.
354 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags,
355 xtensa_fix_close_loop_end_frags, relax_frag_immed, convert_frag_immed):
356 Inline calls to get_frag_is_no_transform.
357 (next_instrs_are_b_retw): Inline call to is_windowed_return_opcode.
358 (xtensa_fix_short_loop_frags): Inline calls to is_the_loop_opcode and
359 get_frag_is_no_transform.
360 (convert_frag_immed_finish_loop): Inline calls to get_expression_value
361 and set_frag_is_no_transform.
362 (get_expression_value): Delete.
363 (subseg_map struct): Rename cur_total_freq to total_freq. Rename
364 cur_target_freq to target_freq.
365 (get_subseg_info): Split out code to create a new map entry into ...
366 (add_subseg_info): ... this new function.
367 (get_last_insn_flags): Check if get_subseg_info succeeded.
368 (set_last_insn_flags): Call add_subseg_info if needed.
369 (get_subseg_total_freq, get_subseg_target_freq, set_subseg_freq): New.
370 (xtensa_reorder_segments): Compute last_sec while counting sections.
371 Remove call to get_last_sec.
372 (get_last_sec): Delete.
373 (cache_literal_section): Inline call to retrieve_literal_seg and its
374 callees, seg_present and add_seg_list.
375 (retrieve_literal_seg, seg_present, add_seg_list): Delete.
376 (get_frag_is_insn, get_frag_is_no_transform,
377 set_frag_is_specific_opcode, set_frag_is_no_transform): Delete.
378 * config/tc-xtensa.h (MAX_SLOTS): Reduce from 31 to 15.
379
7fa3d080
BW		 3802004-10-12 Bob Wilson <bob.wilson@acm.org>
381
382 * config/tc-xtensa.c: Use ISO C90 formatting.
383 * config/tc-xtensa.h: Likewise.
384 * config/xtensa-istack.h: Likewise.
385 * config/xtensa-relax.c: Likewise.
386 * config/xtensa-relax.h: Likewise.
387
8cb51566
PB		 3882004-10-12 Paul Brook <paul@codesourcery.com>
389
390 * config/tc-arm.c (md_begin): Change EF_ARM_EABI_VER3 to
391 EF_ARM_EABI_VER4.
392 (arm_eabis): Ditto.
393 * doc/c-arm.texi: Document that we actually support -meabi=4, not
394 -meabi=3.
395
653cfe85
BW		 3962004-10-08 Bob Wilson <bob.wilson@acm.org>
397
398 * doc/as.texinfo (VTableEntry, VTableInherit): Add "directive" to index
399 entries.
400 (Acknowledgements): Use "GAS" instead of AS variable.
401
d2b2c203
DJ		 4022004-10-08 Daniel Jacobowitz <dan@debian.org>
403
404 * config/tc-i386.c: Include "elf/x86-64.h".
405 (i386_elf_section_type): New function.
406 * config/tc-i386.h (md_elf_section_type): Define.
407 (i386_elf_section_type): New prototype.
408
e80e0390
NC		 4092004-10-08 Linus Nielsen Feltzing <linus@haxx.se>
410
411 * config/m68k-parse.h (enum m68k_register): New control register,
412 MBAR2 (for MCF5249)
413 * config/tc-m68k.c: Correct control register set for MCF5249.
414
43cd72b9 4152004-10-07 Bob Wilson <bob.wilson@acm.org>
85a5ac80 416 Sterling Augustine <sterling@tensilica.com>
43cd72b9
BW		 417
418 * config/tc-xtensa.c (absolute_literals_supported): New global flag.
419 (UNREACHABLE_MAX_WIDTH): Define.
420 (XTENSA_FETCH_WIDTH): Delete.
421 (cur_vinsn, xtensa_fetch_width, xt_saved_debug_type, past_xtensa_end,
422 prefer_const16, prefer_l32r): New global variables.
423 (LIT4_SECTION_NAME): Define.
424 (lit4_state struct): Add lit4_seg_name and lit4_seg fields.
425 (XTENSA_PROP_*, GET_XTENSA_PROP_*, SET_XTENSA_PROP_*): Define.
426 (frag_flags struct): New.
c04f5787 427 (xtensa_block_info struct): Move from tc-xtensa.h. Add flags field.
43cd72b9
BW		 428 (subseg_map struct): Add cur_total_freq and cur_target_freq fields.
429 (bitfield, bit_is_set, set_bit, clear_bit): Define.
430 (MAX_FORMATS): Define.
431 (op_placement_info struct, op_placement_table): New.
432 (O_pltrel, O_hi16, O_lo16): Define.
433 (directiveE enum): Rename directive_generics to directive_transform.
434 Delete directive_relax. Add directive_schedule,
435 directive_absolute_literals, and directive_last_directive.
436 (directive_info): Rename "generics" to "transform". Delete "relax".
437 Add "schedule" and "absolute-literals".
438 (directive_state): Adjust entries to match changes in directive_info.
439 (xtensa_relax_statesE, RELAX_IMMED_MAXSTEPS): Move to tc-xtensa.h.
440 (xtensa_const16_opcode, xtensa_movi_opcode, xtensa_movi_n_opcode,
441 xtensa_l32r_opcode, xtensa_nop_opcode, xtensa_rsr_lcount_opcode): New.
442 (xtensa_j_opcode, xtensa_rsr_opcode): Delete.
443 (align_only_targets, software_a0_b_retw_interlock,
444 software_avoid_b_j_loop_end, maybe_has_b_j_loop_end,
445 software_avoid_short_loop, software_avoid_close_loop_end,
446 software_avoid_all_short_loops, specific_opcode): Delete.
447 (warn_unaligned_branch_targets): New.
448 (workaround_a0_b_retw, workaround_b_j_loop_end, workaround_short_loop,
449 workaround_close_loop_end, workaround_all_short_loops): Default FALSE.
450 (option_[no_]link_relax, option_[no_]transform,
451 option_[no_]absolute_literals, option_warn_unaligned_targets,
452 option_prefer_l32r, option_prefer_const16, option_target_hardware):
453 New enum values.
454 (option_[no_]align_only_targets, option_literal_section_name,
455 option_text_section_name, option_data_section_name,
456 option_bss_section_name, option_eb, option_el): Delete.
457 (md_longopts): Add entries for: [no-]transform, [no-]absolute-literals,
458 warn-unaligned-targets, prefer-l32r, prefer-const16, [no-]link-relax,
459 and target-hardware. Delete entries for [no-]target-align-only,
460 literal-section-name, text-section-name, data-section-name, and
461 bss-section-name.
462 (md_parse_option): Handle new options and remove old ones. Accept but
463 ignore [no-]density options. Warn for [no-]generics and [no-]relax
464 and treat them as [no-]transform.
465 (md_show_usage): Add new options and remove old ones.
466 (xtensa_setup_hw_workarounds): New.
467 (md_pseudo_table): Change "word" entry to use xtensa_elf_cons. Add
468 "long", "short", "loc" and "frequency" entries.
469 (use_generics): Rename to ...
470 (use_transform): ... this function. Add past_xtensa_end check.
471 (use_longcalls): Add past_xtensa_end check.
472 (code_density_available, can_relax): Delete.
473 (do_align_targets): New.
474 (get_directive): Accept dashes in directive names. Warn about
475 [no-]generics and [no-]relax directives and treat them as
c04f5787 476 [no-]transform.
43cd72b9
BW		 477 (xtensa_begin_directive): Call md_flush_pending_output only for some
478 directives. Check for directives inside instruction bundles. Warn
479 about deprecated ".begin literal" usage. Warn and ignore [no-]density
480 directives. Handle new directives. Check generating_literals flag
481 for literal_prefix.
482 (xtensa_end_directive): Check for directives inside instruction
483 bundles. Warn and ignore [no-]density directives. Handle new
484 directives. Call xtensa_set_frag_assembly_state.
485 (xtensa_loc_directive_seen, xtensa_dwarf2_directive_loc,
486 xtensa_dwarf2_emit_insn): New.
487 (xtensa_literal_position): Call md_flush_pending_output. Do not check
488 use_literal_section flag.
489 (xtensa_literal_pseudo): Call md_flush_pending_output. Handle absolute
490 literals. Use xtensa_elf_cons to parse the expression.
491 (xtensa_literal_prefix): Do not check use_literal_section. Support
492 ".lit4" sections for absolute literals. Change prefix convention to
493 replace ".text" (or ".t" in a linkonce section). No need to call
494 subseg_set.
495 (xtensa_frequency_pseudo, xtensa_elf_cons, xtensa_elf_suffix): New.
496 (expression_end): Handle closing braces and colons.
497 (PLT_SUFFIX, plt_suffix): Delete.
498 (expression_maybe_register): Use new xtensa-isa.h functions. Use
499 xtensa_elf_suffix instead of checking for plt suffix, and handle O_lo16
500 and O_hi16 expressions as well.
501 (tokenize_arguments): Handle closing braces and colons.
502 (parse_arguments): Use new xtensa-isa.h functions. Handle "invisible"
503 operands and paired register syntax.
504 (get_invisible_operands): New.
505 (xg_translate_sysreg_op): Handle new Xtensa LX RSR/WSR/XSR syntax. Use
506 new xtensa-isa.h functions.
507 (xtensa_translate_old_userreg_ops, xtensa_translate_zero_immed): New.
508 (xg_translate_idioms): Check if inside bundle. Use use_transform.
509 Handle new Xtensa LX RSR/WSR/XSR syntax. Remove code to widen density
510 instructions. Use xtensa_translate_zero_immed.
511 (operand_is_immed, operand_is_pcrel_label): Delete.
512 (get_relaxable_immed): Use new xtensa-isa.h functions.
513 (get_opcode_from_buf): Add slot parameter. Use new xtensa-isa.h
514 functions.
515 (xtensa_print_insn_table, print_vliw_insn): New.
516 (is_direct_call_opcode): Use new xtensa-isa.h functions.
517 (is_call_opcode, is_loop_opcode, is_conditional_branch_opcode,
518 is_branch_or_jump_opcode): Delete.
519 (is_movi_opcode, decode_reloc, encode_reloc, encode_alt_reloc): New.
520 (opnum_to_reloc, reloc_to_opnum): Delete.
521 (xtensa_insnbuf_set_operand, xtensa_insnbuf_get_operand): Use new
522 xtensa-isa.h functions. Operate on one slot of an instruction.
523 (xtensa_insnbuf_set_immediate_field, is_negatable_branch,
524 xg_get_insn_size): Delete.
525 (xg_get_build_instr_size): Use xg_get_single_size.
526 (xg_is_narrow_insn, xg_is_single_relaxable_insn): Update calls to
527 xg_build_widen_table. Use xg_get_single_size.
528 (xg_get_max_narrow_insn_size): Delete.
529 (xg_get_max_insn_widen_size, xg_get_max_insn_widen_literal_size,
530 xg_is_relaxable_insn): Update calls to xg_build_widen_table. Use
531 xg_get_single_size.
532 (xg_build_to_insn): Record the loc field. Handle OP_OPERAND_HI16U and
533 OP_OPERAND_LOW16U. Check xg_valid_literal_expression.
534 (xg_expand_to_stack, xg_expand_narrow): Update calls to
535 xg_build_widen_table. Use xg_get_single_size.
536 (xg_immeds_fit): Use new xtensa-isa.h functions. Update call to
537 xg_check_operand.
538 (xg_symbolic_immeds_fit): Likewise. Also handle O_lo16 and O_hi16, and
539 treat weak symbols conservatively.
540 (xg_check_operand): Use new xtensa-isa.h functions.
541 (is_dnrange): Delete.
542 (xg_assembly_relax): Inline previous calls to tinsn_copy.
543 (xg_finish_frag): Specify separate relax states for the frag and slot0.
544 (is_branch_jmp_to_next, xg_add_branch_and_loop_targets): Use new
545 xtensa-isa.h functions.
546 (xg_instruction_matches_option_term, xg_instruction_matches_or_options,
547 xg_instruction_matches_options): New.
548 (xg_instruction_matches_rule): Handle O_register expressions. Call
549 xg_instruction_matches_options.
550 (transition_rule_cmp): New.
551 (xg_instruction_match): Update call to xg_build_simplify_table.
552 (xg_build_token_insn): Record loc fields.
553 (xg_simplify_insn): Check is_specific_opcode field and
554 density_supported flag.
555 (xg_expand_assembly_insn): Skip checking code_density_available. Use
556 new xtensa-isa.h functions. Call use_transform instead of can_relax.
557 (xg_assemble_literal): Add error handling for O_big. Call
558 record_alignment. Handle O_pltrel.
559 (xg_valid_literal_expression): New.
560 (xg_assemble_literal_space): Add slot parameter. Remove call to
561 set_expr_symbol_offset. Add call to record_alignment. Update call to
562 xg_finish_frag.
563 (xg_emit_insn): Delete.
564 (xg_emit_insn_to_buf): Add format parameter. Update calls to
565 xg_add_opcode_fix and xtensa_insnbuf_to_chars.
566 (xg_add_opcode_fix): Change opcode parameter to tinsn and add format
567 and slot parameters. Handle new "alternate" relocations for absolute
568 literals and CONST16 instructions. Check for bad uses of O_lo16 and
569 O_hi16. Use new xtensa-isa.h functions.
570 (xg_assemble_tokens): Delete.
571 (is_register_writer): Use new xtensa-isa.h functions.
572 (is_bad_loopend_opcode): Check for xtensa_rsr_lcount_opcode instead of
573 old-style RSR from LCOUNT.
574 (next_frag_opcode): Delete.
575 (next_frag_opcode_is_loop, next_frag_format_size, frag_format_size,
576 update_next_frag_state): New.
577 (update_next_frag_nop_state): Delete.
578 (next_frag_pre_opcode_bytes): Use next_frag_opcode_is_loop.
579 (xtensa_mark_literal_pool_location): Check use_literal_section flag and
580 the state of the absolute-literals directive. Add calls to
581 record_alignment and xtensa_set_frag_assembly_state. Call
582 xtensa_switch_to_non_abs_literal_fragment instead of
583 xtensa_switch_to_literal_fragment.
584 (build_nop): New.
585 (assemble_nop): Use build_nop. Update call to xtensa_insnbuf_to_chars.
586 (get_expanded_loop_offset): Change check for undefined opcode to an
587 assertion.
588 (xtensa_set_frag_assembly_state, relaxable_section,
589 xtensa_find_unmarked_state_frags, xtensa_find_unaligned_branch_targets,
590 xtensa_find_unaligned_loops, xg_apply_tentative_value): New.
591 (md_begin): Update call to xtensa_isa_init. Initialize linkrelax to 1.
592 Set lit4_seg_name. Call xg_init_vinsn. Initialize new global opcodes.
593 Call init_op_placement_info_table and xtensa_set_frag_assembly_state.
594 (xtensa_init_fix_data): New.
595 (xtensa_frob_label): Reset label symbol to the current frag. Check
596 do_align_targets and generating_literals flag. Propagate frequency
597 info to new alignment frag. Call xtensa_set_frag_assembly_state.
598 (xtensa_unrecognized_line): New.
599 (xtensa_flush_pending_output): Check if inside a bundle. Add a call
600 to xtensa_set_frag_assembly_state.
601 (error_reset_cur_vinsn): New.
602 (md_assemble): Remove check for literal frag. Remove call to
603 istack_init. Call use_transform instead of use_generics. Parse
604 explicit instruction format specifiers. Move code for
605 a0_b_retw_interlock workaround to xg_assemble_vliw_tokens. Call
606 error_reset_cur_vinsn on errors. Add call to get_invisible_operands.
607 Add dwarf2_where call. Remote automatic alignment for ENTRY
608 instructions. Move call to xtensa_clear_insn_labels to the end.
609 Rearrange to handle bundles.
610 (xtensa_cons_fix_new): Delete.
611 (xtensa_handle_align): New.
612 (xtensa_frag_init): Call xtensa_set_frag_assembly_state. Remove
613 assignment to is_no_density field.
614 (md_pcrel_from): Use new xtensa-isa.h functions. Use decode_reloc
615 instead of reloc_to_opnum. Handle "alternate" relocations.
616 (xtensa_force_relocation, xtensa_check_inside_bundle,
617 xtensa_elf_section_change_hook): New.
618 (xtensa_symbol_new_hook): Delete.
619 (xtensa_fix_adjustable): Check for difference of symbols with an
620 offset. Check for external and weak symbols.
621 (md_apply_fix3): Remove cases for XTENSA_OP{0,1,2} relocs.
622 (md_estimate_size_before_relax): Return expansion for the first slot.
623 (tc_gen_reloc): Handle difference of symbols by producing
624 XTENSA_DIFF{8,16,32} relocs and by writing the value of the difference
625 into the output. Handle new XTENSA_SLOT*_OP relocs by storing the
626 tentative values into the output when linkrelax is set.
627 (XTENSA_PROP_SEC_NAME): Define.
628 (xtensa_post_relax_hook): Call xtensa_find_unmarked_state_frags.
629 Create literal tables only if using literal sections. Create new
630 property tables instead of old instruction tables. Check for unaligned
631 branch targets and loops.
632 (finish_vinsn, find_vinsn_conflicts, check_t1_t2_reads_and_writes,
633 new_resource_table, clear_resource_table, resize_resource_table,
634 resources_available, reserve_resources, release_resources,
635 opcode_funcUnit_use_unit, opcode_funcUnit_use_stage,
636 resources_conflict, xg_find_narrowest_format, relaxation_requirements,
637 bundle_single_op, emit_single_op, xg_assemble_vliw_tokens): New.
638 (xtensa_end): Call xtensa_flush_pending_output. Set past_xtensa_end
639 flag. Update checks for workaround options. Call
640 xtensa_mark_narrow_branches and xtensa_mark_zcl_first_insns.
641 (xtensa_cleanup_align_frags): Add special case for branch targets.
642 Check for and mark unreachable frags.
643 (xtensa_fix_target_frags): Remove use of align_only_targets flag.
644 Use RELAX_LOOP_END_BYTES in special case for negatable branch at the
645 end of a zero-overhead loop body.
646 (frag_can_negate_branch): Handle instructions with multiple slots.
647 Use new xtensa-isa.h functions
648 (xtensa_mark_narrow_branches, is_narrow_branch_guaranteed_in_range,
649 xtensa_mark_zcl_first_insns): New.
650 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags): Error if
651 transformations are disabled.
652 (next_instrs_are_b_retw): Use new xtensa-isa.h functions. Handle
653 multislot instructions.
654 (xtensa_fix_close_loop_end_frags, xtensa_fix_short_loop_frags):
655 Likewise. Also error if transformations are disabled.
656 (unrelaxed_frag_max_size): New.
657 (unrelaxed_frag_min_insn_count, unrelax_frag_has_b_j): Use new
658 xtensa-isa.h functions.
659 (xtensa_sanity_check, is_empty_loop, is_local_forward_loop): Use
660 xtensa_opcode_is_loop instead of is_loop_opcode.
661 (get_text_align_power): Replace as_fatal with assertion.
662 (get_text_align_fill_size): Iterate instead of using modulus when
663 use_nops is false.
664 (get_noop_aligned_address): Assert that this is for a machine-dependent
665 RELAX_ALIGN_NEXT_OPCODE frag. Use next_frag_opcode_is_loop,
666 xg_get_single_size, and frag_format_size.
667 (get_widen_aligned_address): Rename to ...
668 (get_aligned_diff): ... this function. Add max_diff parameter.
669 Remove handling of rs_align/rs_align_code frags. Use
670 next_frag_format_size, get_text_align_power, get_text_align_fill_size,
671 next_frag_opcode_is_loop, and xg_get_single_size. Compute max_diff
672 and pass it back to caller.
673 (xtensa_relax_frag): Use relax_frag_loop_align. Add code for new
674 RELAX_SLOTS, RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN,
675 RELAX_FILL_NOP, and RELAX_UNREACHABLE frag types. Check relax_seen.
676 (relax_frag_text_align): Rename to ...
677 (relax_frag_loop_align): ... this function. Assume loops can only be
678 in the first slot of an instruction.
679 (relax_frag_add_nop): Use assemble_nop instead of constructing an OR
680 instruction. Remove call to frag_wane.
681 (relax_frag_narrow): Rename to ...
682 (relax_frag_for_align): ... this function. Extend to handle
683 RELAX_FILL_NOP and RELAX_UNREACHABLE, as well as RELAX_SLOTS with
684 RELAX_NARROW for the first slot.
685 (find_address_of_next_align_frag, bytes_to_stretch): New.
686 (future_alignment_required): Use find_address_of_next_align_frag and
687 bytes_to_stretch. Look ahead to subsequent frags to make smarter
688 alignment decisions.
689 (relax_frag_immed): Add format, slot, and estimate_only parameters.
690 Check if transformations are enabled for b_j_loop_end workaround.
691 Use new xtensa-isa.h functions and handle multislot instructions.
692 Update call to xg_assembly_relax.
693 (md_convert_frag): Handle new RELAX_SLOTS, RELAX_UNREACHABLE,
694 RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN, and RELAX_FILL_NOP
695 frag types.
696 (convert_frag_narrow): Add segP, format and slot parameters. Call
697 convert_frag_immed for branch instructions. Adjust calls to
698 tinsn_from_chars, tinsn_immed_from_frag, and xg_emit_insn_to_buf. Use
699 xg_get_single_size and xg_get_single_format.
700 (convert_frag_fill_nop): New.
701 (convert_frag_immed): Add format and slot parameters. Handle multislot
702 instructions and use new xtensa-isa.h functions. Update calls to
703 tinsn_immed_from_frag and xg_assembly_relax. Check if transformations
704 enabled for b_j_loop_end workaround. Use build_nop instead of
705 assemble_nop. Check is_specific_opcode flag. Check for unreachable
706 frags. Use xg_get_single_size. Handle O_pltrel.
707 (fix_new_exp_in_seg): Remove check for old plt flag.
708 (convert_frag_immed_finish_loop): Update calls to tinsn_from_chars and
709 xtensa_insnbuf_to_chars. Call tinsn_immed_from_frag. Change check
710 for loop opcode to an assertion. Mark all frags up to the end of the
711 loop as not transformable.
712 (get_last_insn_flags, set_last_insn_flags): Use get_subseg_info.
713 (get_subseg_info): New.
714 (xtensa_move_literals): Call xtensa_set_frag_assembly_state. Add null
715 check for dest_seg.
716 (xtensa_switch_to_literal_fragment): Rewrite to handle absolute
717 literals and use xtensa_switch_to_non_abs_literal_fragment otherwise.
718 (xtensa_switch_to_non_abs_literal_fragment): New.
719 (cache_literal_section): Add is_code parameter and pass it through to
720 retrieve_literal_seg.
721 (retrieve_literal_seg): Add is_code parameter and use it to set the
722 flags on the literal section. Handle case where head parameter is 0.
723 (get_frag_is_no_transform, set_frag_is_specific_opcode,
724 set_frag_is_no_transform): New.
725 (xtensa_create_property_segments): Add end_property_function parameter
726 and pass it through to add_xt_block_frags. Call bfd_get_section_flags
727 and skip SEC_DEBUGGING and !SEC_ALLOC sections.
728 (xtensa_create_xproperty_segments, section_has_xproperty): New.
729 (add_xt_block_frags): Add end_property_function parameter and call it
730 if it is non-zero. Call xtensa_frag_flags_init.
731 (xtensa_frag_flags_is_empty, xtensa_frag_flags_init,
732 get_frag_property_flags, frag_flags_to_number,
733 xtensa_frag_flags_combinable, xt_block_aligned_size,
734 xtensa_xt_block_combine, add_xt_prop_frags,
735 init_op_placement_info_table, opcode_fits_format_slot,
736 xg_get_single_size, xg_get_single_format): New.
737 (istack_push): Inline call to tinsn_copy.
738 (tinsn_copy): Delete.
739 (tinsn_has_invalid_symbolic_operands): Handle O_hi16 and O_lo16 and
740 CONST16 opcodes. Handle O_big, O_illegal, and O_absent.
741 (tinsn_has_complex_operands): Handle O_hi16 and O_lo16.
742 (tinsn_to_insnbuf): Use xg_get_single_format and new xtensa-isa.h
743 functions. Handle invisible operands.
744 (tinsn_to_slotbuf): New.
745 (tinsn_check_arguments): Use new xtensa-isa.h functions.
746 (tinsn_from_chars): Add slot parameter. Rewrite using xg_init_vinsn,
747 vinsn_from_chars, and xg_free_vinsn.
748 (tinsn_from_insnbuf): New.
749 (tinsn_immed_from_frag): Add slot parameter and handle multislot
750 instructions. Handle symbol differences.
751 (get_num_stack_text_bytes): Use xg_get_single_size.
752 (xg_init_vinsn, xg_clear_vinsn, vinsn_has_specific_opcodes,
753 xg_free_vinsn, vinsn_to_insnbuf, vinsn_from_chars, expr_is_register,
754 get_expr_register, set_expr_symbol_offset_diff): New.
755 * config/tc-xtensa.h (MAX_SLOTS): Define.
756 (xtensa_relax_statesE): Move from tc-xtensa.c. Add
757 RELAX_CHECK_ALIGN_NEXT_OPCODE, RELAX_MAYBE_DESIRE_ALIGN, RELAX_SLOTS,
758 RELAX_FILL_NOP, RELAX_UNREACHABLE, RELAX_MAYBE_UNREACHABLE, and
759 RELAX_NONE types.
760 (RELAX_IMMED_MAXSTEPS): Move from tc-xtensa.c.
761 (xtensa_frag_type struct): Add is_assembly_state_set,
762 use_absolute_literals, relax_seen, is_unreachable, is_specific_opcode,
763 is_align, is_text_align, alignment, and is_first_loop_insn fields.
764 Replace is_generics and is_relax fields by is_no_transform field.
765 Delete is_text and is_longcalls fields. Change text_expansion and
766 literal_expansion to arrays of MAX_SLOTS entries. Add arrays of
767 per-slot information: literal_frags, slot_subtypes, slot_symbols,
768 slot_sub_symbols, and slot_offsets. Add fr_prev field.
769 (xtensa_fix_data struct): New.
770 (xtensa_symfield_type struct): Delete plt field.
771 (xtensa_block_info struct): Move definition to tc-xtensa.h. Add
772 forward declaration here.
773 (xt_section_type enum): Delete xt_insn_sec. Add xt_prop_sec.
774 (XTENSA_SECTION_RENAME): Undefine.
775 (TC_FIX_TYPE, TC_INIT_FIX_DATA, TC_FORCE_RELOCATION, NO_PSEUDO_DOT,
776 tc_unrecognized_line, md_do_align, md_elf_section_change_hook,
777 HANDLE_ALIGN, TC_LINKRELAX_FIXUP, SUB_SEGMENT_ALIGN): Define.
778 (TC_CONS_FIX_NEW, tc_symbol_new_hook): Delete.
779 (unit_num_copies_func, opcode_num_units_func,
780 opcode_funcUnit_use_unit_func, opcode_funcUnit_use_stage_func): New.
781 (resource_table struct): New.
782 * config/xtensa-istack.h (MAX_INSN_ARGS): Increase from 6 to 10.
783 (TInsn struct): Add keep_wide, loc, fixup, record_fix, subtype,
784 literal_space, symbol, sub_symbol, offset, and literal_frag fields.
785 (tinsn_copy): Delete prototype.
786 (vliw_insn struct): New.
787 * config/xtensa-relax.c (insn_pattern_struct): Add options field.
788 (widen_spec_list): Add option conditions for density and boolean
789 instructions. Add expansions using CONST16 and conditions for using
790 CONST16 vs. L32R. Use new Xtensa LX RSR/WSR syntax. Add entries for
791 predicted branches.
792 (simplify_spec_list): Add option conditions for density instructions.
793 Add entry for NOP instruction.
794 (append_transition): Add cmp function pointer parameter and use it to
795 insert the new entry in order.
796 (operand_function_LOW16U, operand_function_HI16U): New.
797 (xg_has_userdef_op_fn, xg_apply_userdef_op_fn): Handle
798 OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
799 (enter_opname, split_string): Use xstrdup instead of strdup.
800 (init_insn_pattern): Initialize new options field.
801 (clear_req_or_option_list, clear_req_option_list,
802 clone_req_or_option_list, clone_req_option_list, parse_option_cond):
803 New.
804 (parse_insn_pattern): Parse option conditions.
805 (transition_applies): New.
806 (build_transition): Use new xtensa-isa.h functions. Fix incorrectly
807 swapped last arguments in calls to append_constant_value_condition.
808 Call clone_req_option_list. Add warning about invalid opcode.
809 Handle LOW16U and HI16U function names.
810 (build_transition_table): Add cmp parameter and use it in calls to
811 append_transition. Use new xtensa-isa.h functions. Check
812 transition_applies before adding entries.
813 (xg_build_widen_table, xg_build_simplify_table): Add cmp parameter and
814 pass it through to build_transition_table.
815 * config/xtensa-relax.h (ReqOrOptionList, ReqOrOption, ReqOptionList,
816 ReqOption, transition_cmp_fn): New types.
817 (OpType enum): Add OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
818 (transition_rule struct): Add options field.
819 * doc/as.texinfo (Overview): Update Xtensa options.
820 * doc/c-xtensa.texi (Xtensa Options): Delete --[no-]density,
821 --[no-]relax, and --[no-]generics options. Update descriptions of
822 --text-section-literals and --[no-]longcalls. Add
823 --[no-]absolute-literals and --[no-]transform.
824 (Xtensa Syntax): Add description of syntax for FLIX instructions.
825 Remove use of "generic" and "specific" terminology for opcodes.
826 (Xtensa Registers): Generalize the syntax description to include
827 user-defined register files.
828 (Xtensa Automatic Alignment): Update.
829 (Xtensa Branch Relaxation): Mention limitation of unconditional jumps.
830 (Xtensa Call Relaxation): Linker can now remove most of the overhead.
831 (Xtensa Directives): Remove confusing rules about precedence.
832 (Density Directive, Relax Directive): Delete.
833 (Schedule Directive): New.
834 (Generics Directive): Rename to ...
835 (Transform Directive): ... this node.
836 (Literal Directive): Update for absolute literals. Missing
837 literal_position directive is now an error.
838 (Literal Position Directive): Update for absolute literals.
839 (Freeregs Directive): Delete.
840 (Absolute Literals Directive): New.
841 (Frame Directive): Minor editing.
842 * Makefile.am (DEPTC_xtensa_elf, DEPOBJ_xtensa_elf, DEP_xtensa_elf):
843 Update dependencies.
844 * Makefile.in: Regenerate.
845
6e1304d8
RS		 8462004-10-07 Richard Sandiford <rsandifo@redhat.com>
847
848 * config/tc-mips.c (append_insn): Use fix_new rather than fix_new_exp
849 to build the second and third fixups for a composite relocation.
850 (macro_read_relocs): New function.
851 (macro_build): Use it.
852 (s_cpsetup): Pass all three composite relocation codes to macro_build.
853 Simplify fragging code accordingly.
854 (s_gpdword): Use fix_new rather than fix_new_exp for the second part
855 of the composite relocation. Set fx_tcbit in both fixups.
856
b1dca8ee
RS		 8572004-10-07 Richard Sandiford <rsandifo@redhat.com>
858
859 * config/tc-mips.c (append_insn): Set fx_tcbit for composite relocs.
860 (md_apply_fix3): Don't treat composite relocs as done.
861
89658e52
NC		 8622004-10-07 Jan Beulich <jbeulich@novell.com>
863
864 * macro.c (macro_expand_body): When ELF, use .LL rather than LL as
865 prefix for symbol names generated from the LOCAL macro directive.
866
289040ca
NC		 867 * dw2gencfi.c (select_cie_for_fde): When separating CIE out from
868 FDE, treat a DW_CFA_remember_state as we do a DW_CFA_advance_loc.
869
48c9f030
NC		 8702004-10-07 Tomer Levi <Tomer.Levi@nsc.com>
871
872 * config/tc-crx.c (preprocess_reglist): Handle Co-processor
c04f5787 873 Special registers.
5c5054ab 874 (md_assemble): Add error checking for Co-Processor instructions.
48c9f030
NC		 875 (get_cinv_parameters): Add 'b' option to invalidate the
876 branch-target cache.
877
7ed4c4c5
NC		 8782004-10-05 Paul Brook <paul@codesourcery.com>
879
880 * config/tc-arm.c (unwind): New variable.
881 (vfp_sp_encode_reg): New function.
882 (vfp_sp_reg_required_here): Use it.
883 (vfp_sp_reg_list, vfp_dp_reg_list): Remove.
884 (vfp_parse_reg_list): New function.
885 (s_arm_unwind_fnstart, s_arm_unwind_fnend, s_arm_unwind_cantunwind,
886 s_arm_unwind_personality, s_arm_unwind_personalityindex,
887 s_arm_unwind_handlerdata, s_arm_unwind_save, s_arm_unwind_movsp,
888 s_arm_unwind_pad, s_arm_unwind_setfp, s_arm_unwind_raw): New
889 functions.
890 (md_pseudo_table): Add them.
891 (do_vfp_reg2_from_sp2): Use vfp_parse_reg_list and vfp_sp_encode_reg.
892 (do_vfp_sp2_from_reg2, vfp_sp_ldstm, vfp_dp_ldstm): Ditto.
893 (set_section, add_unwind_adjustsp, flush_pending_unwind,
894 finish_unwind_opcodes, start_unwind_section, create_unwind_entry,
895 require_hashconst, add_unwind_opcode): New functions.
896 * doc/tc-arm.text: Document unwinding opcodes.
897 * NEWS: Mention the new feature.
898
04ca8703
EC		 8992004-10-04 Eric Christopher <echristo@redhat.com>
900
901 * config/tc-mips.c (md_apply_fix3): Remove erroneous assert.
902
7c1d0959
L		 9032004-10-01 H.J. Lu <hongjiu.lu@intel.com>
904
905 * config/tc-ppc.c (md_apply_fix3): Call S_SET_THREAD_LOCAL for
906 TLS relocations.
907 * config/tc-s390.c (md_apply_fix3): Likewise.
908 * config/tc-sparc.c (md_apply_fix3): Likewise.
909
40a18ebd
NC		 9102004-10-01 Paul Brook <paul@codesourcery.com>
911
912 * config/tc-arm.c (arm_elf_section_type): New function.
913 (arm_elf_change_section): Set section link for exidx sections.
914 * config/tc-arm.h (arm_elf_section_type): Add prototype.
915 (md_elf_section_type): Define.
916
14127cc4
NC		 9172004-10-01 Bill Farmer <Bill@the-farmers.freeserve.co.uk>
918
919 * config/tc-pdp11.c (md_apply_fix3): Change to sign of the SOB
920 instruction's offset.
921
b7c92712
NC		 9222004-10-01 Adam Nemet <anemet@lnxw.com>
923
924 * (TARGET_FORMAT): Remove LynxOS COFF definition.
925
65555e64
AM		 9262004-10-01 Ravi Ramaseshan <ravi.ramaseshan@codito.com>
927
928 * config/tc-arc.c (tc_gen_reloc): Don't assume fixP->fx_addsy is an
929 asymbol *, instead use symbol_get_bfdsym.
930
9fe0b840
NC		 9312004-09-30 Linus Nielsen Feltzing <linus@haxx.se>
932
933 * config/tc-m68k.c (select_control_regs): Add mcf5249.
934
0dd132b6
NC		 9352004-09-30 Paul Brook <paul@codesourcery.com>
936
937 * config/tc-arm.c (do_smi, do_nop): New functions.
938 (insns): Add ARMv6ZK instructions.
939 (md_apply_fix3): Handle BFD_RELOC_ARM_SMI.
940 (tc_gen_reloc): Ditto.
941 (arm_cpus): Add mpcore and arm1176.
942 (arm_archs): Add armv6{k,z,zk}.
943 * doc/c-arm.texi: Document new cores and architectures.
944
a737bd4d
NC		 9452004-09-30 Nick Clifton <nickc@redhat.com>
946
947 * config/tc-arm.c: Use ISO C90 formatting.
948
3631a3c8
NC		 9492004-09-30 Vladimir Ivanov <vladitx@nucleusys.com>
950
a737bd4d 951 * config/tc-arm.c (mav_reg_required_here): Allow REG_TYPE_CN
3631a3c8
NC		 952 as alternative when REG_TYPE_MVF, REG_TYPE_MVD, REG_TYPE_MVFX or
953 REG_TYPE_MVDX is expected.
954
d5f0cf92
AM		 9552004-09-29 Marc Bevand <m.bevand@gmail.com>
956
957 * doc/c-i386.texi (i386-Mnemonics): Fix typo.
958
9f9a069e
JW		 9592004-09-21 James E Wilson <wilson@specifixinc.com>
960
961 * config/tc-ia64.c (ENCODED_PSP_OFFSET): New.
962 (output_rp_psprel, output_pfs_psprel, output_preds_psprel,
963 output_spill_base, output_unat_psprel, output_lc_psprel,
964 output_fpsr_psprel, output_priunat_psprel, output_bsp_psprel,
965 output_bsprestore_psprel, output_rnat_psprel, output_spill_psprel,
966 output_spill_psprel_p): Use it.
967
4b778c88
TL		 9682004-09-20 Tomer Levi <Tomer.Levi@nsc.com>
969
c04f5787 970 * config/tc-crx.c (handle_LoadStor): New function.
04ca8703 971 Handle load/stor unique instructions before parsing.
4b778c88 972
eb043451
PB		 9732004-09-17 Paul Brook <paul@codesourcery.com>
974
975 * config/tc-arm.c (s_arm_rel31): New funciton.
976 (md_pseudo_table): Add .rel31.
04ca8703 977 (md_apply_fix3): Handle BFD_RELOC_ARM_TARGET2,
eb043451
PB		 978 BFD_RELOC_32_PCREL and BFD_RELOC_ARM_PREL31.
979 (tc_gen_reloc): Handle BFD_RELOC_ARM_PREL31 and BFD_RELOC_ARM_TARGET2.
980 (arm_fix_adjustable): Return 0 for BFD_RELOC_ARM_TARGET2.
981 (arm_parse_reloc): Add (target2).
982
8ea2b00f
AM		 9832004-09-17 Alan Modra <amodra@bigpond.net.au>
984
985 * Makefile.am: Run "make dep-am".
986 * Makefile.in: Regenerate.
987 * aclocal.m4: Regenerate.
988 * configure: Regenerate.
989 * doc/Makefile.in: Regenerate.
990 * po/POTFILES.in: Regenerate.
991 * po/gas.pot: Regenerate.
992
7e1792e0
HPN		 9932004-09-14 Hideki IWAMOTO <h-iwamoto@kit.hi-ho.ne.jp>
994
995 * config/tc-mmix.c [!LLONG_MIN]: Correct #elsif to #elif.
996 [!LLONG_MAX]: Ditto.
997
9c504268
PB		 9982004-09-13 Paul Brook <paul@codesourcery.com>
999
1000 * config/tc-arm.c: Rename RELABS to TARGET1.
1001
2159ac21
AM		 10022004-09-13 Alan Modra <amodra@bigpond.net.au>
1003
1004 * messages.c (as_internal_value_out_of_range): Cast values passed
1005 to as_bad_where or as_warn_where to proper type.
1006
23794b24
MM		 10072004-09-11 Theodore A. Roth <troth@openavr.org>
1008
1009 * config/tc-avr.c: Add support for
1010 atmega48, atmega88, atmega168, attiny13, attiny2313, at90can128.
1011
e9fad691
AM		 10122004-09-09 Alan Modra <amodra@bigpond.net.au>
1013
1014 * dw2gencfi.c (select_cie_for_fde): When separating CIE out
1015 from FDE, treat a CFI_escape as we do a DW_CFA_advance_loc.
1016
10b016c2
PB		 10172004-09-08 Paul Brook <paul@codesourcery.com>
1018
1019 * config/obj-elf.c (obj_elf_section_type): Handle init_array,
1020 fini_array and preinit_array section types.
1021 * config/tc-ia64.c (ia64_elf_section_type): Remove init_array
1022 and fini_array.
1023 * doc/as.texinfo: Document extra section types.
1024
e5a52504
MM		 10252004-09-02 Mark Mitchell <mark@codesourcery.com>
1026
1027 * Makefile.am (TARG_ENV_HFILES): Add te-symbian.h.
1028 * Makefile.in: Regenerated.
1029 * configure.in: Set em for arm*-*-symbianelf*.
1030 * configure: Regenerated.
1031 * config/tc-arm.c (elf32_arm_target_format): Use Symbian target
1032 vectors when appropriate.
1033 * config/te-symbian.h: New file.
1034
42851540
NC		 10352004-09-03 Tomer Levi <Tomer.Levi@nsc.com>
1036
1037 * config/tc-crx.c (gettrap): Exception vector can be case
1038 insensitive.
04ca8703 1039 (process_label_constant): Fix a 32-bit displacement bug in branch
42851540
NC		 1040 instructions.
1041 (get_operandtype) : Bug fix, wrong operand was used.
04ca8703 1042 (process_label_constant): Initialize relocation type to
42851540
NC		 1043 BFD_RELOC_NONE
1044
8783612f
RE		 10452004-09-01 Richard Earnshaw < reanrsha@arm.com>
1046
1047 * tc-arm.c (arm_cpus, arm_fpus): Allow <cpu>-s as well as <cpu>s
1048 for synthesizable cores.
1049
1ff4677c
RE		 1050 * doc/c-arm.texi (ARM Options): Document canonical names of CPUs.
1051
b18c562e
NC		 10522004-08-25 Dmitry Diky <diwil@spec.ru>
1053
1054 * config/tc-msp430.c: Clean-up the code.
1055 (md_relax_table): New relax table.
1056 (mcu_types): Sort MCU types.
1057 (md_pseudo_table): Add .profiler pseudo handler.
1058 (pow2value): New function.
1059 (msp430_profiler): New function.
1060 (msp430_operands): Add new insns handlers.
1061 (msp430_srcoperand): Add register operand handler, allow complex
1062 expressions.
1063 (md_estimate_size_before_relax): Rewritten.
1064 (md_convert_frag): Rewritten.
1065 (msp430_relax_frag): New function.
1066 * config/tc-msp430.h (md_relax_frag): define macro
1067 * doc/c-msp430.texi: Update information.
1068
8f94ae4d
NC		 10692004-08-24 Nick Clifton <nickc@redhat.com>
1070
1071 * as.c (std_shortopts): Allow -g to take an optional argument.
1072 (parse_args): Pass any switch starting with -g on to the backend
1073 for parsing.
1074
e51dace8
MM		 10752004-08-18 Mark Mitchell <mark@codesourcery.com>
1076
1077 * configure.in (arm*-*-symbianelf*): New target.
1078 (arm*-*-eabi*): Likewise.
1079 * configure: Regenerated.
1080
f7870c8d
TS		 10812004-08-18 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1082 * config/tc-mips.c (append_insn): Handle delay slots in branch likely
1083 correctly.
1084
d6afba4b
JJ		 10852004-08-18 Jakub Jelinek <jakub@redhat.com>
1086
1087 * config/tc-ia64.c (start_unwind_section): Add linkonce_empty
1088 argument, don't do anything if current section is not
1089 .gnu.linkonce.t.* and linkonce_empty is set.
1090 (generate_unwind_image, dot_endp): Adjust callers, call
1091 start_unwind_section (*, 1) if nothing will be put into the
1092 section.
1093
329e276d
NC		 10942004-08-17 Nick Clifton <nickc@redhat.com>
1095
1096 * as.c (MD_DEBUG_FORMAT_SELECTOR): Provide default definition.
1097 (show_usage): Add -g.
1098 (std_longopts): Add --gen-debug. Alpha sort the table.
1099 (parse_args): Print an error message if a switch is not handled.
1100 Handle the -g switch, calling md_debug_format_selector() if
1101 necessary.
1102 * NEWS: Mention new feature.
1103 * doc/as.texinfo: Document new switch.
1104 * doc/internals.texi: Document behaviour of md_parse_option.
04ca8703 1105
329e276d
NC		 1106 * config/tc-arm.c (md_parse_option): Do not issue an error message
1107 if the switch is not recognised.
1108 * config/tc-m68k.c (md_parse_option): Likewise.
1109 * config/tc-pdp11.c (md_parse_option): Likewise.
1110 * config/tc-v850.c (md_parse_option): Likewise.
1111
1112 * as.h: Fix up formatting.
1113 * tc.h: Likewise.
1114
4101eec8
NC		 11152004-08-16 Nick Clifton <nickc@redhat.com>
1116
2766e5e4
NC		 1117 * macro.c (macro_set_alternate): Use ISO C90 formatting.
1118
4101eec8
NC		 1119 * configure.in: Sort architecture based tables alphabetically.
1120 * configure: Regenerate.
1121
80f846b6
AM		 11222004-08-16 Alan Modra <amodra@bigpond.net.au>
1123
1124 * config/tc-ppc.c (tc_ppc_regname_to_dw2regnum <regnames>): Replace
1125 { "cc", 68 }, with { "cr", 70 }.
1126
caa32fe5
NC		 11272004-08-13 Jan Beulich <jbeulich@novell.com>
1128 Nick Clifton <nickc@redhat.com>
1129
1130 * as.c: Add and handle new --alternate command line option.
1131 * macro.c (macro_set_alternate): New.
1132 * macro.h (macro_set_alternate): Declare.
1133 * read.c: Add and handle new .altmacro and .noaltmacro directives.
1134 * doc/as.texinfo: Document new command line option and pseudo-ops
1135 as well as insert documentation originating from gasp about
1136 alternate macro syntax.
1137 * NEWS: Mention new command line option and pseudo-ops.
1138
b2221023
MM		 11392004-08-10 Mark Mitchell <mark@codesourcery.com>
1140
1141 * expr.c (operand): Handle the "~", "-", and "!" operators applied
1142 to bignums.
1143
db6579d4
PB		 11442004-08-06 Paul Brook <paul@codesourcery.com>
1145
1146 * config/tc-arm.c (md_apply_fix3, tc_gen_reloc, arm_parse_reloc):
1147 Handle new relocations.
1148 * include/elf/arm.h (elf_arm_reloc_type): Add new EABI relocations.
1149
e750405d
BW		 11502004-08-05 Bob Wilson <bob.wilson@acm.org>
1151
1152 * write.c (relax_segment): Use was_address instead of address when
1153 setting fr_fix field for align frag due to backwards .org.
1154
1d70c7fb
AO		 11552004-07-29 Alexandre Oliva <aoliva@redhat.com>
1156
1157 Introduce SH2a support.
1158 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
1159 * config/tc-sh.c (get_specific): Change arch_sh2a_up to
1160 arch_sh2a_nofpu_up.
1161 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
1162 * config/tc-sh.c (md_parse_option): Add sh2a-nofpu ISA handling.
1163 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
1164 * config/tc-sh.c (sh_elf_final_processing): Move sh2a recognition
1165 to end of conditional expression.
1166 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
1167 * config/tc-sh.c: Add sh2a-nofpu support.
1168 2003-12-29 DJ Delorie <dj@redhat.com>
1169 * tc-sh.c: Add sh2a support.
1170 (parse_reg): Add tbr.
1171 (parse_at): Support @@(disp,tbr).
1172 (get_specific): Support sh2a opcodes.
1173 (insert4): New, for 4 byte relocs.
1174 (build_Mytes): Support sh2a opcodes.
1175 (md_apply_fix3_Mytes): Support sh2a opcodes.
1176 2003-12-02 Michael Snyder <msnyder@redhat.com>
1177 * config/tc-sh.c (md_parse_option): Handle sh2a.
1178 (sh_elf_final_processing): Ditto.
1179
e7fbc2bf
JT		 11802004-07-27 Jason Thorpe <thorpej@wasabisystems.com>
1181
1182 * config/tc-hppa.h (TARGET_FORMAT): Set to "elf32-hppa-netbsd"
1183 for TE_NetBSD.
1184
9232bbb0
AM		 11852004-07-27 Alan Modra <amodra@bigpond.net.au>
1186
1187 * config/tc-ppc.c (ppc_frob_file_before_adjust): Warn if .toc too big.
1188 (ppc_arch): Expand comment.
1189
670ec21d
NC		 11902004-07-27 Tomer Levi <Tomer.Levi@nsc.com>
1191
1192 * config/tc-crx.c: Support evaluating the difference between two
1193 symbols.
1194 * config/tc-crx.h: Likewise.
1195
406f8ff9 11962004-07-26 H.J. Lu <hongjiu.lu@intel.com>
38ce5b11
L		 1197
1198 * config/tc-ia64.c (start_unwind_section): Set the linked-to
1199 section.
1200 (ia64_elf_section_change_hook): Set the linked-to section for
1201 SHT_IA_64_UNWIND.
1202
c05e9f04
NC		 12032004-07-26 Dmitry Diky <diwil@spec.ru>
1204
1205 * config/tc-msp430.c: Add new subtargets: msp430x1610,
1206 msp430x1611, msp430x1612, msp430x415, msp430x417, msp430xG437,
1207 msp430xG438, msp430xG439.
1208
e9863d7f
DJ		 12092004-07-25 Daniel Jacobowitz <dan@debian.org>
1210
1211 * doc/as.texinfo (Section, PushSection): Correct documentation
1212 for ELF.
1213
543613e9 12142004-07-21 Jan Beulich <jbeulich@novell.com>
20f0a1fc
NC		 1215
1216 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
1217 permissible for the selected instruction suffix.
1218 (match_template): Don't permit 64-bit general purpose operands in
1219 32-bit mode.
1220 (finalize_imm): Permit 64-bit immediates.
1221 (build_modrm_byte): Don't treat 32-bit addressing in 64-bit mode
1222 specially except for the width of the used base and/or index
1223 registers. For 32-bit displacements, use sign-extended
1224 relocations only when using 64-bit addressing.
1225 Force zero displacement on rip-relative addressing when there is
1226 no other displacement.
1227 (i386_index_check): Don't treat 32-bit addressing in 64-bit mode
1228 specially except for the width of the used base and/or index
1229 registers.
1230 (parse_register): Disallow Reg64 registers in 32-bit mode.
1231
543613e9
NC		 1232 * config/tc-i386.c: For DefaultSize instructions, don't guess a 'q'
1233 suffix if the instruction doesn't support it.
04ca8703 1234
01a3f561
MR		 12352004-07-20 Maciej W. Rozycki <macro@linux-mips.org>
1236
1237 * config/tc-mips.c (append_insn): Handle constant expressions with
1238 no associated relocation.
1239 (mips_ip): Cancel the expression after use for the Q format
1240 specifier.
1241 (parse_relocation): Return no relocation for unsupported
1242 operators.
1243 (my_getSmallExpression): Return no relocation if no relocation
1244 operators are used.
1245
e12904d2
DA		 12462004-07-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1247
1248 * config/obj-som.c (adjust_stab_sections): Add prototype.
1249 (obj_som_compiler, obj_som_version, obj_som_copyright,
1250 adjust_stab_sections): Add ATTRIBUTE_UNUSED to unused arguments.
1251 * config/tc-hppa.c (update_subspace): Likewise.
1252 (is_defined_subspace): Amplify comment.
1253 * config/obj-som.h (som_frob_file): Add prototype.
1254
90c1602c
L		 12552004-07-19 Christopher Faylor <cgf@timesys.com>
1256 H.J. Lu <hongjiu.lu@intel.com>
1257
1258 * subsegs.c (section_symbol): Don't create a new segment when
1259 existing segment is undefined.
1260
f263249b
RE		 12612004-07-16 Richard Earnshaw <rearnsha@arm.com>
1262
1263 * config/tc-arm.c: Include include/opcode/arm.h.
1264 (ARM_EXT_*, ARM_ARCH_*, ARM_ANY, ARM_ALL, COPROC_ANY): Delete.
1265 (FPU_FPA_EXT_* FPU_VFP_EXT_*, FPU_ANY, FPU_NONE, FPU_MAVERICK): Delete.
1266 (FPU_ARCH_*): Delete.
1267 * Makefile.am: Update dependencies.
1268 * Makefile.in: Regenerate.
1269
152c65b9
NC		 12702004-07-15 Nick Clifton <nickc@redhat.com>
1271
1272 * configure.in: Accept armbe as a big-endian arm configuration.
1273 * configure: Regenerate.
1274
0477af35
NC		 12752004-07-13 Thomas Nystrom <thn@saeab.se>
1276
1277 * config/tc-i386.c (T_SHIFTOP): New constant.
1278 (intel_e05_1): Handle '&', '|' and T_SHIFTOP.
1279 (intel_el1): Handle '~'.
1280 (intel_get_token): Handle '<>', '&', '|' and '~'.
1281
12822004-07-13 Nick Clifton <nickc@redhat.com>
1283
1284 (md_assemble): Remove spurious newline from end of as_bad error
1285 message.
1286 (intel_e05_1): Likewise.
1287 (intel_e11): Likewise.
1288 (intel_match_token): Likewise.
1289
31302357
AS		 12902004-07-11 Andreas Schwab <schwab@suse.de>
1291
1292 * config/tc-m68k.c: Convert to C90. Remove redundant
1293 declarations. Indentation fixup.
1294 [M68KCOFF]: Include "obj-coff.h" instead of declaring
1295 obj_coff_section ourselves.
04ca8703 1296
549f748d
JW		 12972004-07-09 James E Wilson <wilson@specifixinc.com>
1298
1299 * config/tc-ia64.c (default_big_endian): New.
1300 (dot_byteorder, md_begin): Use it.
1301 (md_parse_option): Set it.
04ca8703 1302
1a320fbb
NC		 13032004-07-09 Nick Clifton <nickc@redhat.com>
1304
1305 * configure.in: Change sh-sybmian-elf to sh-*-symbianelf.
1306 * configure: Regenerate.
1307 * NEWS: Change sh-sybmian-elf to sh-*-symbianelf.
1308 * config/tc-sh.c (sh_elf_final_processing): Use renamed version of
1309 sh_find_elf_flags if necessary.
1310
de7e6852
RS		 13112004-07-08 Richard Sandiford <rsandifo@redhat.com>
1312
1313 * config/tc-mips.c (mips_fix_adjustable): If the full addend is
1314 going to be split into more than one in-place addend, return 0
1315 for relocations against mergeable sections. Associate comments
1316 with code.
1317
1fe1f39c
NC		 13182004-07-07 Tomer Levi <Tomer.Levi@nsc.com>
1319
1320 * Makefile.am (CPU_TYPES): Add crx.
1321 (TARGET_CPU_CFILES): Add config/tc-crx.c.
1322 (TARGET_CPU_HFILES): Add config/tc-crx.h.
1323 (DEPTC_crx_elf): New target.
1324 (DEPOBJ_crx_elf): Likewise.
1325 (DEP_crx_elf): Likewise.
1326 * Makefile.in: Regenerate.
1327 * configure.in: Add crx* target.
1328 * configure: Regenerate.
1329 * config/tc-crx.c: New file.
1330 * config/tc-crx.h: New file.
1331 * NEWS: Mention new target.
1332
d9895649
NC		 13332004-07-06 Nick Clifton <nickc@redhat.com>
1334
85fbca6a 1335 * config.in: Undefine TARGET_SYMBIAN by default.
c04f5787
AM		 1336 * configure.in:
1337 * configure: Regenerate. Add sh-symbian-elf target. If
1338 selected define TARGET_SYMBIAN.
1339 * config/tc-sh.h (TARGET_FORMAT): Select a Symbian target
1340 format if TARGET_SYMBIAN has been defined.
04ca8703 1341
e7bd9ea0
NC		 1342 * output-file.c (output_file_create): Report the target format
1343 chosen when bfd_openw reports that it is invalid.
1344
d9895649
NC		 1345 * config/obj-coff.c (coff_pseudo_table): Only define the weak
1346 pseudo for BFD based assemblers.
1347
871ec896
JR		 13482004-07-05 Andrew Stubbs <andrew.stubbs@superh.com>
1349
1350 gas:
1351 * config/tc-sh.c (md_assemble): Change isspace to ISSPACE.
1352 (md_parse_option): Remove redundant -isa testing.
1353 Make bfd_arch variable const.
1354 (md_show_usage): Make bfd_arch variable const.
1355
5a9ff93d
JW		 13562004-07-03 James E Wilson <wilson@specifixinc.com>
1357
1358 * config/tc-ia64.c (emit_one_bundle): Check and set insn_addr.
1359 * config/tc-ia64.h (md_frag_check): Define.
1360
c87db184
CF		 13612004-07-03 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
1362
1363 * config/obj-coff.c (obj_coff_weak): New .weak syntax for PE weak
1364 externals.
1365 * doc/as.texinfo (Weak): Document PE weak symbols.
1366
3b91255e
RS		 13672004-07-03 Richard Sandiford <rsandifo@redhat.com>
1368
1369 * config/tc-mips.c (HAVE_IN_PLACE_ADDENDS): New macro.
1370 (reloc_needs_lo_p): Only return true if HAVE_IN_PLACE_ADDENDS.
1371 (mips_frob_file): Rework so that only a single pass through the
1372 relocs is needed. Allow %lo()s to have higher offsets than their
1373 corresponding %hi()s or %got()s.
1374
310ea308
NC		 13752004-07-02 Nick Clifton <nickc@redhat.com>
1376
1377 * config/tc-arm.c (md_apply_fix3:BFD_RELOC_ARM_IMMEDIATE): Do not
1378 allow values which have come from undefined symbols.
1379 Always consider this fixup to have been processed as a reloc
1380 cannot be generated for it.
1381
09b935ac
AM		 13822004-07-02 Alan Modra <amodra@bigpond.net.au>
1383
1384 * frags.h (struct frag): Add has_code and insn_addr fields.
1385 * write.c (cvt_frag_to_fill): Invoke md_frag_check.
1386 * config/tc-ppc.c (md_assemble): Check and set insn_addr.
1387 * config/tc-ppc.h (md_frag_check): Define.
1388
4d1cd5c7
MR		 13892004-06-28 Maciej W. Rozycki <macro@linux-mips.org>
1390
1391 * doc/Makefile.am (info): Rename goal to...
1392 (info-local): ... this, to preserve implicit dependencies.
1393 * doc/Makefile.in: Regenerate with automake 1.8.5.
1394
097f809a
NC		 13952004-06-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1396
1397 * config/tc-m32r.c (md_convert_frag): Changed for @PLT.
1398 (m32r_cgen_record_fixup_exp): Changed for @GOTOFF, @GOT.
1399 (m32r_fix_adjustable): Changed for @GOTOFF, @GOT, @PLT.
1400 (tc_gen_reloc): Likewise.
1401 (m32r_end_of_match): Add for @GOTOFF, @GOT, @PLT.
1402 (m32r_parse_name): Likewise.
1403 (m32r_cgen_parse_fix_exp): Likewise.
1404 * config/tc-m32r.h (md_parse_name): Define for @GOTOFF, @GOT, @PLT.
1405 (O_PIC_reloc): Likewise.
1406 (TC_CGEN_PARSE_FIX_EXP): Likewise..
1407 * cgen.c (gas_cgen_parse_operand): Add TC_CGEN_PARSE_FIX_EXP
1408 for @GOTOFF, @GOT, @PLT.
1409
bb82af9f
NC		 14102004-06-21 Jan Beulich <jbeulich@novell.com>
1411
1412 * gas/symbols.c: While discarding ordinary local absolute symbols
1413 when --strip-local-absolute is in effect, retain file symbols.
1414
990ae22a
AS		 14152004-06-20 Andreas Schwab <schwab@suse.de>
1416
1417 * config/tc-m68k.c (mri_chip): Replace current_chip, not augment.
1418 (md_parse_option): Likewise.
1419
32137342
NC		 14202004-06-17 Jan Beulich <jbeulich@novell.com>
1421
1422 * config/tc-i386.c: Deal with LEX_QM the same way as with LEX_AT.
1423 * config/te-netware.h: New file.
1424 * config/te-ppcnw.h: Delete: Obsolete.
1425 * configure.in: Eliminate ill NetWare targets. Make generic
1426 NetWare target use proper emulation.
1427 * Makefile.am: Eliminate reference to obsolete te-ppcnw.h, add
1428 reference to new te-netware.h.
1429 * configure: Regenerate.
1430 * Makefile.in: Regenerate.
1431
f84d8e26
MS		 14322004-06-15 Martin Schwidefsky <schwidefsky@de.ibm.com>
1433
1434 * config/tc-s390.c (s390_insn): Avoid incorrect signed/unsigned
1435 comparison in .insn pseudo operation.
1436
587aac4e
AM		 14372004-06-15 Alan Modra <amodra@bigpond.net.au>
1438
1439 * config/obj-coff.c (coff_adjust_section_syms): Use
1440 bfd_get_section_size instead of bfd_get_section_size_before_reloc.
1441 (coff_frob_section): Likewise.
1442 * config/tc-mips.c (md_apply_fix3): Likewise.
1443 * config/obj-elf.c (elf_frob_file): Use bfd_set_section_size.
1444 (elf_frob_file_after_relocs): Likewise.
1445
2337eb33
DA		 14462004-06-10 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1447
b8fc22bc
DA		 1448 * config/tc-hppa.c (log2): Rename to exact_log2.
1449 (pa_next_subseg): Delete unused function.
1450 (create_new_space): Mark unused arguments with ATTRIBUTE_UNUSED.
1451 (create_new_subspace): Likewise.
1452
2337eb33
DA		 1453 Bug gas/213
1454 * config/tc-hppa.c (hppa_fix_adjustable): Allow reduction of fake
1455 labels. Fix warning.
1456
53cb0362
DD		 14572004-05-28 DJ Delorie <dj@redhat.com>
1458
1459 * config/tc-mn10300.h (tc_fix_adjustable): Define.
1460 * config/tc-mn10300.c (mn10300_fix_adjustable): Don't adjust debug
1461 or non-merged symbols.
1462
888a75be
L		 14632004-05-28 H.J. Lu <hongjiu.lu@intel.com>
1464
1465 * config/tc-ia64.c (remove_marked_resource): Save, clear and
1466 restore the old slot when inserting srlz.i/srlz.d.
1467
f6f9408f
JR		 14682004-05-28 Andrew Stubbs <andrew.stubbs@superh.com>
1469
1470 * Makefile.am: Regenerate dependecies.
1471 * Makefile.in: Regenerate.
1472 * config/tc-sh.c (valid_arch): Make unsigned.
1473 (preset_target_arch): Likewise.
1474 (md_begin): Use new architecture flags system.
1475 (get_specific): Likewise.
1476 (assemble_ppi): Likewise.
1477 (md_assemble): Likewise. Also fix error check for bad opcodes.
1478 (md_parse_option): Likewise. Also generate -isa values according
1479 to the table in bfd/cpu-sh.c instead of just constants. Also
1480 allow <arch>-up ISA variants.
1481 (sh_elf_final_processing): Replace if-else chain with a call to
1482 sh_find_elf_flags().
1483 * testsuite/gas/sh/arch: New directory.
1484 * testsuite/gas/sh/arch/arch.exp: New test script.
1485 * testsuite/gas/sh/arch/arch_expected.txt: New file.
1486 * testsuite/gas/sh/arch/sh.s: New file.
1487 * testsuite/gas/sh/arch/sh2.s: New file.
1488 * testsuite/gas/sh/arch/sh-dsp.s: New file.
1489 * testsuite/gas/sh/arch/sh2e.s: New file.
1490 * testsuite/gas/sh/arch/sh3-nommu.s: New file.
1491 * testsuite/gas/sh/arch/sh3.s: New file.
1492 * testsuite/gas/sh/arch/sh3-dsp.s: New file.
1493 * testsuite/gas/sh/arch/sh3e.s: New file.
1494 * testsuite/gas/sh/arch/sh4-nommu-nofpu.s: New file.
1495 * testsuite/gas/sh/arch/sh4-nofpu.s: New file.
1496 * testsuite/gas/sh/arch/sh4.s: New file.
1497 * testsuite/gas/sh/arch/sh4a-nofpu.s: New file.
1498 * testsuite/gas/sh/arch/sh4al-dsp.s: New file.
1499 * testsuite/gas/sh/arch/sh4a.s: New file.
1500
580b9172
NC		 15012004-05-28 Peter Barada <peter@the-baradas.com>
1502
1503 * config/gc-m68k.c(m68k_ip): Convert mode 5 addressing
1504 with zero offset into mode 2 addressing to save a word.
1505
652ca075
L		 15062004-05-27 H.J. Lu <hongjiu.lu@intel.com>
1507
1508 * config/tc-ia64.c (ar_is_in_integer_unit): Removed.
1509 (ar_is_only_in_integer_unit): New.
1510 (ar_is_only_in_memory_unit): New.
1511 (generate_unwind_image): Silence gcc on 32bit host.
1512 (md_assemble): Use ar_is_only_in_integer_unit instead of
1513 ar_is_in_integer_unit. Check AR access.
1514
60fed2d3
NC		 15152004-05-27 Peter Barada <peter@the-baradas.com>
1516
1517 * config/tc-m68k.c (md_begin): Sort the opcode table into
1518 alphabetical order.
1519 (m68k_compare_opcode): New function to do the sorting.
1520
be8c092b
NC		 15212004-05-24 Peter Barada <peter@the-baradas.com>
1522
1523 * config/m68k-parse.y(operand): Allow for MAC/EMAC mask
1524 addressing on MIT style operands.
1525 * config/m68k-parse.y(yylex): Allow '-&' for predecrement
1526 w/mask addressing.
1527 * config/tc-m68k.c(install_operand): Comment 'G' and 'H' type
1528 operands.
1529
1181c20e
AM		 15302004-05-23 Alan Modra <amodra@bigpond.net.au>
1531
1532 * expr.c (operand, operator): Don't reject '++' and '--'.
1533
20e56c33
RS		 15342004-05-20 Richard Sandiford <rsandifo@redhat.com>
1535
1536 * config/tc-mips.c (append_insn): Use ISA-encoded addresses in MIPS16
1537 dwarf tables.
1538
04ca8703 15392004-05-17 Adam Nemet <anemet@lnxw.com>
b2d65c0b
NC		 1540
1541 * configure.in: Add ppc-*-lynxos*. Update i386-*-lynxos* to ELF.
1542 * configure: Regenerate.
1543
0da76f83
NC		 15442004-05-13 Paul Brook <paul@codesourcery.com>
1545
1546 * dw2gencfi.c (output_cie): Handle dwarf3 format CIE entries.
1547
4b363073
NC		 15482004-05-13 Joel Sherrill <joel@oarcorp.com>
1549
1550 * configure.in (or32-*-rtems*): Switch to elf.
1551 * configure: Regenerate.
1552
9598fbe5
NC		 15532004-05-13 Nick Clifton <nickc@redhat.com>
1554
1555 * po/fr.po: Updated French translation.
1556
22fe14ad
NC		 15572004-05-11 Nick Clifton <nickc@redhat.com>
1558
1559 * doc/as.texinfo (Section): Document G and T flags to .section
1560 directive. Document the extra arguments that the G flag
1561 requires. Document the #tls flag.
1562
15632004-05-11 H.J. Lu <hongjiu.lu@intel.com>
1564
1565 * subsegs.c (section_symbol): Create a new section symbol if
1566 the existing one doesn't match.
1567 * symbols.c (symbol_set_bfdsym): Don't reset BFD section symbol.
1568
532c738a
RS		 15692004-05-07 Richard Sandiford <rsandifo@redhat.com>
1570
1571 * config/tc-mips.c (append_insn, mips_emit_delays): Extend -mfix-vr4120
1572 to cope with VR4181A errata MD(1) and MD(4).
1573
0503b355
BF		 15742004-05-07 Brian Ford <ford@vss.fsi.com>
1575
1576 * NEWS: Mention .secrel32 for pe[i]-i386.
1577
daeea990
AO		 15782004-05-07 Alexandre Oliva <aoliva@redhat.com>
1579
1580 * config/tc-frv.h (MAX_MEM_FOR_RS_ALIGN_CODE): New.
1581 (HANDLE_ALIGN): New.
1582
bfbdcf04
DJ		 15832004-05-06 Daniel Jacobowitz <dan@debian.org>
1584
1585 * Makefile.am (DIST_SUBDIRS): Define.
1586 * aclocal.m4: Regenerate with automake 1.8.4.
1587 * Makefile.in: Likewise.
1588 * doc/Makefile.in: Likewise.
1589
5e819f9c
JW		 15902004-05-06 David Mosberger-Tang <davidm@hpl.hp.com>
1591
1592 * config/tc-ia64.c (dot_serialize): Declare.
1593 (dot_serialize): New function.
1594 (md_pseudo_table): Add ".serialize.data" and
1595 ".serialize.instruction" directives.
1596
e5976317
NC		 15972004-05-06 Nick Clifton <nickc@redhat.com>
1598
1599 * messages (as_internal_value_out_of_range): Print a message about
1600 a value being out of range. Be consistent about whether the
1601 values are printed in decimal or hexadecimal.
1602 (as_warn_value_out_of_range): Generate a warning message about an
1603 out of range value.
1604 (as_bad_value_out_of_range): Generate an error message about an
1605 out of range value.
1606 * as.h: Prototype the new functions.
1607 * config/tc-alpha.c (insert_operand): Use new function.
1608 * config/tc-arc.c (arc_insert_operand): Likewise.
1609 * config/tc-mn10200.c (mn10200_insert_operand): Likewise.
1610 * config/tc-mn10300.c (mn10300_insert_operand): Likewise.
1611 * config/tc-ppc.c (ppc_insert_operand): Likewise.
1612 * config/tc-s390.c (s390_insert_operand): Likewise.
1613 * config/tc-v850.c (v850_insert_operand): Likewise.
1614
996991aa
AO		 16152004-05-05 Alexandre Oliva <aoliva@redhat.com>
1616
1617 * configure.in: Set em=linux for frv-*-*linux*.
1618 * configure: Rebuilt.
1619 * config/tc-frv.h (TARGET_FORMAT): Use elf32-frvfdpic if...
1620 (frv_md_fdpic_enabled): New.
1621 * config/tc-frv.c (frv_md_fdpic_enabled): New.
1622 (DEFAULT_FDPIC): New.
1623 (frv_flags): Use DEFAULT_FDPIC.
1624 (frv_pic_flag): Likewise.
1625 (OPTION_NOPIC): New.
1626 (md_longopts): Add -mnopic.
1627 (md_parse_option): Handle it.
1628 (md_show_usage): Add -mfdpic and -mnopic.
1629
6b6e92f4
NC		 16302004-05-05 Peter Barada <peter@the-baradas.com>
1631
1632 * config/tc-m68k.c: Add find_cf_chip to print list of valid
1633 chips for invalid coldfire instructions, rename selectors
1634 for ColdFire sub-variants, add 521x,5249,547x,548x and aliases,
1635 add current_chip to track which chip is referred to(including save/restore),
1636 use current_chip to select control registers, not current_arch.
1637 (md_show_usage): Add new chips.
1638 * doc/c-m68k.texi: Document new command line switches.
1639
6f079ae3
JJ		 16402004-05-05 Jakub Jelinek <jakub@redhat.com>
1641
1642 * tc-s390.h (md_do_align, HANDLE_ALIGN): Remove.
1643 (NOP_OPCODE): Define.
1644 (s390_align_code): Remove prototype.
1645 * tc-s390.c (s390_align_code): Remove.
1646
2d6ed997
L		 16472004-05-04 H.J. Lu <hongjiu.lu@intel.com>
1648
1649 * config/tc-ia64.c (make_unw_section_name): Removed.
1650 (start_unwind_section): New function.
1651 (generate_unwind_image): Take const segT instead of const
1652 char *.
1653 (dot_handlerdata): Adjusted.
1654 (dot_endp): Likewise.
1655
b9f18452
L		 16562004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1657
1658 * config/obj-elf.c (obj_elf_change_section): Allow the
1659 ".note.GNU-stack" section has SHF_EXECINSTR.
1660
fafe6678
L		 16612004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1662
1663 * config/obj-elf.c (get_section): Return bfd_boolean.
1664 (obj_elf_change_section): Call bfd_get_section_by_name_if
1665 instead of bfd_map_over_sections.
1666
86654c12
L		 16672004-04-30 H.J. Lu <hongjiu.lu@intel.com>
1668
1669 * config/obj-elf.c (get_section): New function.
1670 (obj_elf_change_section): Support multiple sections with same
1671 name.
1672
376eb240
NC		 16732004-04-30 Nick Clifton <nickc@redhat.com>
1674
1675 * config/tc-arm.c (create_register_alias): Fix typo checking for
1676 case sensitive register aliases.
1677 (co_proc_number): Use error message string in all_reg_maps[]
1678 array.
1679 (cp_reg_required_here): Likewise.
1680 (fp_reg_required_here): Likewise.
1681
8a7140c3
NC		 16822004-04-29 Brian Ford <ford@vss.fsi.com>
1683
1684 * dwarf2dbg.c (dwarf2_finish): Add SEC_DEBUGGING to section flags.
1685
3e722fb5
CD		 16862004-04-28 Chris Demetriou <cgd@broadcom.com>
1687
1688 * config/tc-mips.c (HAVE_32BIT_ADDRESSES, append_insn, macro_build)
1689 (load_address, macro, mips_ip, md_parse_option)
1690 (mips_force_relocation, mips_validate_fix, md_apply_fix3)
1691 (s_change_sec, pic_need_relax, tc_gen_reloc): Remove all
1692 embedded-PIC handling, and update comments.
1693 (SWITCH_TABLE): Remove.
1694 * config/tc-mips.h (DIFF_EXPR_OK): Delete.
1695 (enum mips_pic_level): Remove EMBEDDED_PIC.
1696 (EXTERN_FORCE_RELOC): Remove embedded-PIC handling.
1697 (TC_FORCE_RELOCATION): Update comment.
1698 * ecoff.c (ecoff_build_lineno): Add comment about some code that
1699 might be safe to remove now that MIPS embedded-PIC is gone.
1700
351e2b5a
DA		 17012004-04-28 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1702
1703 * config/obj-som.c (obj_som_init_stab_section): Add new arguments in
1704 call to obj_set_subsection_attributes.
1705 (obj_som_init_stab_section): Likewise.
1706 * config/tc-hppa.c (default_subspace_dict): Add comdat field.
1707 (pa_def_subspaces): Provide comdat default.
1708 (pa_subspace): Handle new "comdat" parameter. Set SEC_LINK_ONCE and
1709 not SEC_IS_COMMON if section is comdat, common or dup_common. Update
1710 calls to create_new_subspace and update_subspace to pass comdat flag.
1711 (create_new_subspace, update_subspace): Add new comdat argument. Use
1712 it in calls to obj_set_subsection_attributes.
1713 * doc/c-hppa.texi (.subspa, .nsubspa): Document new comdat parameter
1714 and use of comdat, common and dup_comm parameters.
1715
f5842774
L		 17162004-04-26 H.J. Lu <hongjiu.lu@intel.com>
1717
1718 * config/obj-elf.c (obj_elf_change_section): Check if the old
1719 group name is NULL before comparison.
1720
558fe4a7
CD		 17212004-04-23 Chris Demetriou <cgd@broadcom.com>
1722
1723 * config/tc-mips.h (mips_dwarf2_addr_size): Prototype.
1724
81a21e38
TS		 17252004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1726
1727 * config/tc-mips.c (s_mipsset): Set default CPU type for .set mipsN.
1728
17292004-04-23 Chris Demetriou <cgd@broadcom.com>
1ffcab4b
CD		 1730
1731 * config/tc-mips.c (md_longopts): Remove -membedded-pic option.
1732 (OPTION_MEMBEDDED_PIC): Remove.
1733 (OPTION_TRAP, OPTION_BREAK, OPTION_EB, OPTION_EL)
1734 (OPTION_FP32, OPTION_GP32, OPTION_CONSTRUCT_FLOATS)
1735 (OPTION_NO_CONSTRUCT_FLOATS, OPTIONS_FP64, OPTION_GP64)
1736 (OPTION_RELAX_BRANCH, OPTION_NO_RELAX_BRANCH)
1737 (OPTION_ELF_BASE): Renumber.
1738 (md_parse_option): Remove OPTION_MEMBEDDED_PIC handling.
1739 (md_show_usage): Remove mention of -membedded-pic.
1740 * doc/as.texinfo: Remove mention of -membedded-pic.
1741
4d0d148d
TS		 17422004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1743
1744 * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Remove.
1745 * config/tc-mips.c (RDATA_SECTION_NAME, mips_target_format): Remove
1746 a.out support.
1747 (md_begin, mips_ip, md_parse_option, s_change_sec, s_option,
1748 s_abicalls, nopic_need_relax, tc_gen_reloc): Remove uses of
1749 USE_GLOBAL_POINTER_OPT.
1750
017315e4
TS		 17512004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1752
1753 * config/tc-mips.c (macro): One more use of load_delay_nop.
1754
269137b2
TS		 17552004-04-22 Atsushi Nemoto <anemo@mba.ocn.ne.jp>
1756
1757 * config/tc-mips.c (load_delay_nop): New function.
1758 (load_address, macro): Use load_delay_nop() to build a nop
1759 which can be omitted with gpr_interlocks.
1760
24772049
TS		 17612004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1762
1763 * config/tc-mips.c (hilo_interlocks, gpr_interlocks,
1764 cop_interlocks): Remove superfluous CPU entries.
1765
06516a5e
PB		 17662004-04-22 Paul Brook <paul@codesourcery.com>
1767
1768 * config/tc-arm.c (mav_parse_offset): Value must be multiple of 4.
1769
fd99574b
NC		 17702004-04-22 Peter Barada <peter@the-baradas.com>
1771
1772 * NEWS: Added support for EMAC instructions and MAC/EMAC
1773 Motorola syntax.
1774 * config/m68k-parse.h: Add ACC[123], ACCEXT{01,23}, MAC/EMAC
1775 scale factor tokens, trailing_ampersand to mark mask addressing
1776 for MAC/EMAC instructions.
1777 * config/m68k-parse.y: Add options_ampersand clause, '<<',
1778 '>>'.
1779 (yylex): Handle '>', '<', and '&' following '+'.
1780 * config/tc-m68k.c: Set mcfmac/mcfemac on appropriate ColdFire
04ca8703 1781 architectures in archs[].
fd99574b
NC		 1782 (m68k-ip): Add '4', 'e', 'g', 'i', cases to handle mask addressing
1783 for MAC/EMAC instructions, ACC[0123], ACCEXT{01,23}, and '<<'/'>>'
1784 respectively.
1785 (m68k_ip): Handle trailing '&' on MAC/EMAC insns.
1786 (install_operand): Fix 'n' case, Add 'F', 'f', 'G', 'H', 'I', ']'
1787 cases.
1788 Add EMAC operands to init_table[].
1789
69b97547
NC		 17902004-04-22 Bruno De Bus <bdebus@elis.ugent.be>
1791
1792 * config/tc-arm.h (enum mstate): Move here, add MAP_UNDEFINED
1793 state.
1794 (TC_SEGMENT_INFO_TYPE): Define to enum mstate.
1795 * config/tc-arm.c (enum mstate): Delete from here.
1796 (mapping_state): Remove the static mapstate variable and instead
1797 store the state in the segment. This allows a per-section mapping
1798 state. Handle and ignore MAP_UNDEFINED states.
1799 (arm_elf_change_section): Get the current mapping state from the
1800 new section.
1801 (s_ltorg): Set the mapping state to MAP_DATA.
1802 (arm_cleanup): Use arm_elf_change_section to get the mapping state
1803 for each pool as it is emitted.
1804
18052004-04-22 Nick Clifton <nickc@redhat.com>
1806
1807 * config/tc-arm.h: Formatting tidy ups.
1808
7a7f4e42
CD		 18092004-04-20 Chris Demetriou <cgd@broadcom.com>
1810
1811 * NEWS: Note that MIPS -membedded-pic option is deprecated.
1812
6482c264
NC		 18132004-04-20 DJ Delorie <dj@redhat.com>
1814
1815 * config/tc-i386.h [TE_PE] (TC_CONS_FIX_NEW): Define.
1816 * config/tc-i386.c (md_pseudo_table) [TE_PE]: Add "secrel32".
1817 [TE_PE] (O_secrel): Define.
1818 [TE_PE] (x86_pe_cons_fix_new): New.
1819 [TE_PE] (pe_directive_secrel): Likewise.
1820 (tc_gen_reloc) [TE_PE]: Support BFD_RELOC_32_SECREL.
1821
5bf10039
EC		 18222004-04-19 Eric Christopher <echristo@redhat.com>
1823
c04f5787
AM		 1824 * config/tc-mips.c (mips_dwarf2_addr_size): Revert part
1825 of previous patch for fix in gcc.
5bf10039 1826
77cd6497
JJ		 18272004-04-19 Jakub Jelinek <jakub@redhat.com>
1828
1829 * config/tc-xtensa.c (xg_assembler_literal): Fix a typo.
1830
b8861cfb
NS		 18312004-04-19 Nathan Sidwell <nathan@codesourcery.com>
1832
1833 * read.c (do_align): Call md_flush_pending_output, if defined.
1834
f4c724fc
AM		 18352004-04-16 Alan Modra <amodra@bigpond.net.au>
1836
1837 * expr.c (operand): Correct checks for ++ and --.
1838
a4b6e8f6
L		 18392004-04-14 H.J. Lu <hongjiu.lu@intel.com>
1840
1841 * config/tc-generic.c: Add some comments.
1842
d766e8ec
RS		 18432004-04-14 Richard Sandiford <rsandifo@redhat.com>
1844
1845 * doc/c-mips.texi (-m{no-,}fix-vr4120): Renamed from
1846 -{no-}mfix-vr4122-bugs.
1847 * config/tc-mips.c (mips_fix_vr4120): Renamed from mips_fix_4122_bugs.
1848 (append_insn, mips_emit_delays): Update accordingly.
1849 (OPTION_FIX_VR4120, OPTION_NO_FIX_VR4120): Renamed from *VR4122.
1850 (md_longopts): Change -{no-,}mfix-vr4122-bugs to -m{no-,}fix-vr4120.
1851 (md_parse_option): Update after above changes.
1852 (md_show_usage): Add -mfix-vr4120.
1853
ed9589d4
BW		 18542004-04-13 Bob Wilson <bob.wilson@acm.org>
1855
1856 * doc/as.texinfo (Sub-Sections): Conditionalize COFF-specific use
1857 of .section directive; add a reference to the ELF .subsection
1858 directive.
1859
097dba13
NC		 18602004-04-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1861
1862 * config/tc-m32r.c (md_assemble): Fixed infinite loop bug
1863 in parallel.
1864
279500de
TS		 18652004-04-11 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1866
1867 * Makefile.am: Remove mips from aout targets.
1868 * Makefile.in: Regenerate.
1869 * configure.in: Remove mips-dec-bsd* target.
1870 * configure: Regenerate.
1871
b7d7dc63
AM		 18722004-04-07 Alan Modra <amodra@bigpond.net.au>
1873
1874 PR 96
1875 * config/tc-ppc.c (ppc_elf_suffix): Add valid32 and valid64 fields
1876 to struct map_bfd. Adjust MAP macro, and define MAP32, MAP64.
1877 Update "mapping". Restrict some @ modifiers to 32 bit.
1878
f55629b8
KK		 18792004-04-01 Asgari Jinia <asgarij@kpitcummins.com>
1880 Dhananjay Deshpande <dhananjayd@kpitcummins.com>
1881
1882 * config/tc-sh.c (dont_adjust_reloc_32): New variable.
1883 (sh_fix_adjustable): Avoid adjusting BFD_RELOC_32 when
1884 dont_adjust_reloc_32 is set.
1885 (md_longopts): Add option -renesas.
1886 (md_parse_option, md_show_usage): Likewise.
1887 * doc/c-sh.texi: Likewise.
1888
a7844384
NC		 18892004-04-01 Dave Korn <dk@artimi.com>
1890
1891 * config/tc-dlx.c (md_assemble): set fx_no_overflow flag for
1892 hi16 and lo16 fixS structs.
1893 (md_assemble): generate bit_fixS for RELOC_DLX_LO16 in
1894 exactly the same way as for RELOC_DLX_REL16.
1895 (machine_ip): properly respect LO flag in the_insn and
1896 output RELOC_DLX_LO16 rather than RELOC_DLX_16.
1897 (md_apply_fix3): apply RELOC_DLX_LO16.
1898
6c71dbc4
SS		 18992004-03-30 Stan Shebs <shebs@apple.com>
1900
1901 Remove long-obsolete MPW support.
1902 * mpw-config.in, mpw-make.sed, mac-as.r: Remove files.
1903 * configure.in: Remove mention of ppc-*-mpw* config.
1904 * configure.in: Likewise.
1905
7cc69913
NC		 19062004-03-30 Nick Clifton <nickc@redhat.com>
1907
1908 * config/tc-arm.c (meabi_flags): Make its use conditional upon
1909 OBJ_ELF being defined.
1910
53fbb48a
AM		 19112004-03-27 Alan Modra <amodra@bigpond.net.au>
1912
1913 * config/obj-aout.c (obj_aout_type): Remove #ifdef BFD_ASSEMBLER code.
1914
d507cf36
PB		 19152004-03-23 Paul Brook <paul@codesourcery.com>
1916
1917 * config/tc-arm.c (meabi_flags): New variable.
1918 (arm_parse_eabi): New function.
1919 (md_begin): Set flags for EABI v3.
1920 (arm_eabis): Add.
1921 (arm_long_opts): Add meabi.
1922 * doc/as.texinf <ARM>: Document -meabi.
1923 * doc/c-arm.texi: Ditto.
1924
88d65ad6
BW		 19252004-03-22 Bob Wilson <bob.wilson@acm.org>
1926
1927 * config/tc-xtensa.c (xtensa_post_relax_hook): Create literal
1928 tables even when use_literal_section flag is not set.
1929
13ef8878
AM		 19302004-03-22 Alan Modra <amodra@bigpond.net.au>
1931
1932 * config/tc-sh.c: Remove trailing whitespace.
1933
1048a9ba
HPN		 19342004-03-22 Hans-Peter Nilsson <hp@axis.com>
1935
1936 * doc/c-cris.texi (CRIS-Opts): Document --no-mul-bug-abort,
1937 --mul-bug-abort and the default behavior.
1938 * config/tc-cris.c (cris_insn_kind): New member CRIS_INSN_MUL.
1939 (err_for_dangerous_mul_placement): New variable.
1940 (STATE_MUL, OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): New
1941 macros.
1942 (md_cris_relax_table): Have placeholder for STATE_MUL.
1943 (md_longopts): New options --mul-bug-abort and --no-mul-bug-abort.
1944 (cris_relax_frag) <case ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: New
1945 case doing nothing.
1946 (md_estimate_size_before_relax) <case ENCODE_RELAX (STATE_MUL,
1947 STATE_BYTE)>: Ditto.
1948 (md_convert_frag) <ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: Check
1949 alignment and position of this frag, emit error message if
1950 suspicious.
1951 (md_assemble): For a multiply insn and when checking it,
1952 transform the current frag into a special frag for that purpose.
1953 (md_parse_option) <case OPTION_MULBUG_ABORT_OFF, case
1954 OPTION_MULBUG_ABORT_ON>: Handle new options.
1955
a7877748
BW		 19562004-03-19 Bob Wilson <bob.wilson@acm.org>
1957
1958 * config/tc-xtensa.c (mark_literal_frags): New function.
1959 (xtensa_move_literals): Call mark_literal_frags for all literal
1960 segments, including init and fini literal segments.
1961 (xtensa_post_relax_hook): Swap use of xt_insn_sec and xt_literal_sec.
1962
9992270f
DA		 19632004-03-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1964
1965 * tc-hppa.c (cons_fix_new_hppa): Check for PC relative base type.
1966 (pa_comm): Set BSF_OBJECT in symbol flags.
1967
6e1772d5
AM		 19682004-03-19 Alan Modra <amodra@bigpond.net.au>
1969
1970 * Makefile.am: Run "make dep-am".
1971 * Makefile.in: Regenerate.
1972 * doc/Makefile.in: Regenerate.
1973 * config.in: Regenerate.
1974 * po/gas.pot: Regenerate.
1975
3e4caed2
NS		 19762004-03-18 Nathan Sidwell <nathan@codesourcery.com>
1977
c95b35a9
NS		 1978 * read.c (read_a_source_file): Use demand_empty_rest_of_line.
1979 (demand_empty_rest_of_line): Issue an error here.
1980 (ignore_rest_of_line): Silently skip to end.
1981 (demand_copy_string): Issue an error, not warning.
1982 (equals): Likewise.
1983 * config/obj-elf.c (obj_elf_section_name): Likewise.
1984 (obj_elf_section): Likewise.
1985 * config/tc-arc.c (arc_extoper): Remove bogus NULL checks.
1986 (arc_extinst): Likewise.
1987 * config/tc-ia64.c (dot_saveb): Use demand_empty_rest_of_line.
1988 (dot_spill): Likewise.
1989 (dot_unwabi): Likewise.
1990 (dot_prologue): Likewise.
1991
3e4caed2
NS		 1992 * expr.c (operand): Reject ++ and --.
1993 (operator): Likewise.
1994
2ce4cc60
KK		 19952004-03-17 Kaz Kojima <kkojima@rr.iij4u.or.jp>
1996
1997 * config/tc-sh.c: Include dw2gencfi.h.
1998 (sh_cfi_frame_initial_instructions): New function.
1999 (sh_regname_to_dw2regnum): Likewise.
2000 * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Move to the end of
2001 file.
2002 (TARGET_USE_CFIPOP): Define.
2003 (tc_cfi_frame_initial_instructions): Likewise.
2004 (tc_regname_to_dw2regnum): Likewise.
2005 (DWARF2_DEFAULT_RETURN_COLUMN, DWARF2_CIE_DATA_ALIGNMENT): Likewise.
2006 * Makefile.am: Update dependencies.
2007 * Makefile.in: Regenerate.
2008
9183ed4b
AM		 20092004-03-17 Ralf Corsepius <corsepiu@faw.uni-ulm.de>
2010
2011 * configure.in: Switch sh-*-rtems* to ELF. Add sh-*-rtemscoff*.
2012 * configure: Regenerate.
2013
00bbdfe7
BW		 20142004-03-12 Bob Wilson <bob.wilson@acm.org>
2015
2016 * read.c (s_leb128): Call md_flush_pending_output.
2017
0f10071e
ML		 20182004-03-12 Michal Ludvig <mludvig@suse.cz>
2019
2020 * config/tc-i386.c (output_insn): Handle PadLock instructions.
2021 * config/tc-i386.h (CpuPadLock): New define.
2022 (CpuUnknownFlags): Added CpuPadLock.
2023
84c84759
AS		 20242004-03-07 Andreas Schwab <schwab@suse.de>
2025
2026 * doc/c-hppa.texi (HPPA Directives): Fix typo.
2027
6ec51dba
RH		 20282004-03-07 Richard Henderson <rth@redhat.com>
2029
2030 * dw2gencfi.c (output_cie): Align length to 4 byte boundary.
2031 (cfi_finish): Likewise for fde.
2032
bae25f19
L		 20332004-03-05 H.J. Lu <hongjiu.lu@intel.com>
2034
2035 * config/tc-ia64.c (md_assemble): Properly handle NULL
2036 align_frag.
2037 (ia64_handle_align): Don't abort if failed to add a stop bit.
2038
ab509236
L		 20392004-03-04 H.J. Lu <hongjiu.lu@intel.com>
2040
2041 * Makefile.in: Regenerated.
2042 * aclocal.m4: Likewise.
2043 * configure: Likewise.
2044 * doc/Makefile.in: Likewise.
2045
d9201763
L		 20462004-03-03 H.J. Lu <hongjiu.lu@intel.com>
2047
2048 * config/tc-ia64.c (dot_align): New.
2049 (ia64_do_align): Make it static.
2050 (md_pseudo_table): Use "dot_align" for "align".
2051 (ia64_md_do_align): Don't set align_frag here.
2052 (ia64_handle_align): Add a stop bit to the previous bundle if
2053 needed.
2054
2055 * config/tc-ia64.h (ia64_do_align): Removed.
2056
ae51a426
JR		 20572003-03-03 Andrew Stubbs <andrew.stubbs@superh.com>
2058
2059 * config/tc-sh.c (md_parse_option): Add -isa=sh4-nofpu and
2060 -isa=sh4-nommu-nofpu options. Adjust help messages accordingly.
2061 (sh_elf_final_processing): Output BFD type sh4_nofpu if that is
2062 the most general type or the user specifically requested it.
2063 (md_assemble): Add a new error message for when an instruction
2064 is understood, but is not allowed due to an -isa option.
2065
9545c4ce
L		 20662004-03-02 H.J. Lu <hongjiu.lu@intel.com>
2067
2068 * config/tc-ia64.c (align_frag): New.
2069 (md_assemble): Set the tc_frag_data field in align_frag for
2070 IA64_OPCODE_FIRST instructions.
2071 (ia64_md_do_align): Set align_frag.
2072 (ia64_handle_align): Add a stop bit if needed.
2073
2074 * config/tc-ia64.h (TC_FRAG_TYPE): New.
2075 (TC_FRAG_INIT): New.
2076
676a64f4
RS		 20772004-03-01 Richard Sandiford <rsandifo@redhat.com>
2078
2079 * config/tc-frv.c (fr400_audio): New variable.
2080 (md_parse_option, md_show_usage): Add -mcpu=fr405 and -mcpu=fr450.
2081 (md_parse_option): Set fr400_audio for -mcpu=fr400 and -mcpu=fr405.
2082 (target_implements_insn_p): New function.
2083 (md_assemble): Report an error if the processor doesn't implement
2084 the instruction.
2085
bd337dde
NC		 20862004-02-27 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2087
2088 * config/tc-m32r.c (md_longopts): Added -no-bitinst option.
2089 (md_parse_option): Ditto.
2090 (OPTION_NO_SPECIAL_M32R): Added.
2091 (md_show_usage): Document it.
2092 (enable_speial_m32r): Changed a default value from 0 to 1.
2093 * doc/c-m32r.texi: Document the -no-bitinst option.
2094
78878175
NC		 20952004-02-27 Nick Clifton <nickc@redhat.com>
2096
2097 * config/tc-sh.c (get_operand): Revert previous delta.
2098 (tc_gen_reloc): Check for an unknown reloc type before processing
2099 the addend.
2100
1d6d62a4
MS		 21012004-02-27 Hannes Reinecke <hare@suse.de>
2102
2103 * config/tc-s390.c (s390_insn): Correct range check for opcode in
2104 .insn pseudo operation.
2105
4aa3e325
NC		 21062004-02-27 Anil Paranjpe <anilp1@kpitcummins.com>
2107
2108 * config/tc-sh.c (get_operand): In case of #Imm, check has been
2109 added for wrong syntax.
2110
73369e65
EC		 21112004-02-26 Eric Christopher <echristo@redhat.com>
2112
2113 * config/tc-mips.c (mips_dwarf2_addr_size): New.
2114 * config/tc-mips.h (DWARF2_ADDR_SIZE): Use.
2115
6a5709a5
JR		 21162004-02-26 Andrew Stubbs <andrew.stubbs@superh.com>
2117
2118 * config/tc-sh.c (build_Mytes): Add REG_N_D and REG_N_B01
2119 nibble types to assembler.
2120
95385380
FF		 21212004-02-25 Fred Fish <fnf@redhat.com>
2122
2123 * config/tc-iq2000.c: Add missing \n\ in multiline string literal.
2124
b5e0fabd
JW		 21252004-02-20 James E Wilson <wilson@specifixinc.com>
2126
2127 * config/tc-ia64.c (slot_index): New arg before_relax. Use instead of
2128 finalize_syms.
2129 (fixup_unw_records): New arg before_relax. Pass to slot_index.
2130 (ia64_estimate_size_before_relax): New.
2131 (ia64_convert_frag): Pass 0 to fixup_unw_records. Add comment.
2132 (generate_unwind_image): Pass 1 to fixup_unw_records.
2133 * config/tc-ia64.h (ia64_estimate_size_before_relax): Declare.
2134 (md_estimate_size_before_relax): Call ia64_estimate_size_before_relax.
2135
2f01ffbf
JJ		 21362004-02-19 Jakub Jelinek <jakub@redhat.com>
2137
2138 * stabs.c (generate_asm_file): Avoid warning about use of
2139 uninitialized variable.
2140
9c59842f
JW		 21412004-02-18 David Mosberger <davidm@hpl.hp.com>
2142
2143 * config/tc-ia64.c (ia64_flush_insns): In addition to prologue,
2144 body, and endp, allow unwind records which do not have a "t"
2145 (time/instruction) field.
2146
34920d91
NC		 21472004-02-17 Petko Manolov <petkan@nucleusys.com>
2148
2149 * config/tc-arm.c (do_mav_dspsc_1): Correct offset of CRn.
2150 (do_mav_dspsc_2): Likewise.
2151 Fix accumulator registers move opcodes.
2152
4dde8e61 21532004-02-13 Hannes Reinecke <hare@suse.de>
13ef8878 2154 Jakub Jelinek <jakub@redhat.com>
4dde8e61
AJ		 2155
2156 * dwarf2dbg.c (get_filenum): Do not read beyond allocated memory.
2157
936cf02e
JW		 21582004-02-10 Steve Ellcey <sje@cup.hp.com>
2159
2160 * config/tc-ia64.h (ia64_frob_symbol): New declaration.
2161 (tc_frob_symbol): New macro definition.
2162 * config/tc-ia64.c (ia64_frob_symbol): New routine.
2163
e41f12f3
DJ		 21642004-02-09 Daniel Jacobowitz <drow@mvista.com>
2165
2166 * config/tc-arm.c (md_begin): Mark .note.gnu.arm.ident as
2167 read-only.
2168
a8a3b3b2
NS		 21692004-02-09 Nathan Sidwell <nathan@codesourcery.com>
2170
2171 * read.h (IGNORE_OPCODE_CASE): Do not define. Replace with ...
2172 (TC_CASE_SENSITIVE): ... this.
2173 * read.c: Replace IGNORE_OPCODE_CASE with TC_CASE_SENSITIVE.
2174 * doc/internals.texi (TC_CASE_SENSITIVE): Document.
2175
c97b7ef6
JW		 21762004-02-06 James E Wilson <wilson@specifixinc.com>
2177
2178 * config/tc-ia64.c (dot_endp): Delete call to output_endp.
2179 (generate_unwind_image): Re-add it here.
2180
e4475e39
NS		 21812004-02-06 Nathan Sidwell <nathan@codesourcery.com>
2182
2183 * dwarf2dbg.c (DWARF2_ADDR_SIZE): Remove trailing ';'
2184 * read.h (SKIP_WHITESPACE): Turn into an expression.
2185 * read.c (read_a_source_file): A pseudo is removed by having a
2186 NULL handler.
2187
5738bc24
JW		 21882004-02-05 James E Wilson <wilson@specifixinc.com>
2189
2190 * config/tc-ia64.c (output_endp): New.
2191 (count_bits): Delete.
2192 (ia64_flush_insns, process_one_record, optimize_unw_records): Handle
2193 endp unwind records.
2194 (fixup_unw_records): Handle endp unwind records. Delete code for
2195 shortening prologue regions not followed by a body record.
2196 (dot_endp): Call add_unwind_entry to emit endp unwind record.
2197 * config/tc-ia64.h (unw_record_type): Add endp.
2198
d6e78c11
JW		 21992004-02-03 James E Wilson <wilson@specifixinc.com>
2200
2201 * config/tc-ia64.c (ia64_convert_frag): Call md_number_to_chars to
2202 fill padding bytes with zeroes.
2203 (emit_one_bundle): New locals last_ptr, end_ptr. Rewrite code that
2204 sets unwind_record slot_number and slot_frag fields.
2205
f6a22291
MR		 22062004-02-02 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2207
2208 * config/tc-mips.c (add_got_offset_hilo): New function.
2209 (macro): Use load_register() and add_got_offset_hilo() to load
2210 constants instead of hardcoding code sequences throughout.
2211
127cab00
L		 22122004-01-28 H.J. Lu <hongjiu.lu@intel.com>
2213
2214 * config/tc-ia64.c (emit_one_bundle): Add proper indentation.
2215
6358301e
NC		 22162004-01-26 Bernardo Innocenti <bernie@develer.com>
2217
2218 * config/tc-m68k.h (EXTERN_FORCE_RELOC): Handle m68k-uclinux specially,
2219 like m68k-elf.
2220 * config/tc-m68k.c (RELAXABLE_SYMBOL): Use EXTERN_FORCE_RELOC instead
2221 of hard-coded test for TARGET_OS=elf.
2222
c8978940
CD		 22232004-01-24 Chris Demetriou <cgd@broadcom.com>
2224
2225 * config/tc-mips.c (hilo_interlocks): Change definition
2226 so that MIPS32, MIPS64 and later ISAs are included, along with
2227 the already-included machines. Update comments.
2228
c3ba240c
DJ		 22292004-01-23 Daniel Jacobowitz <drow@mvista.com>
2230
2231 * config/tc-arm.c (tc_gen_reloc): Improve error message for
2232 undefined local labels.
2233
e864ceca
RS		 22342004-01-23 Richard Sandiford <rsandifo@redhat.com>
2235
2236 * config/tc-mips.c (load_address, macro): Update comments about
2237 NewABI GP relaxation.
2238
67c0d1eb
RS		 22392004-01-23 Richard Sandiford <rsandifo@redhat.com>
2240
2241 * config/tc-mips.c (macro_build): Remove place and counter arguments.
2242 (mips_build_lui, macro_build_ldst_constoffset): Likewise.
2243 (mips16_macro_build, macro_build_jalr): Remove counter argument.
2244 (set_at, load_register, load_address, move_register): Likewise.
2245 (load_got_offset, add_got_offset): Likewise.
2246 Update all calls and tidy accordingly.
2247
584892a6
RS		 22482004-01-23 Richard Sandiford <rsandifo@redhat.com>
2249
2250 * config/tc-mips.c (RELAX_ENCODE): Remove WARN argument.
2251 (RELAX_FIRST, RELAX_SECOND): Turn into 8-bit quantities.
2252 (RELAX_USE_SECOND): Bump to 0x10000.
2253 (RELAX_SECOND_LONGER, RELAX_NOMACRO, RELAX_DELAY_SLOT): New flags.
2254 (mips_macro_warning): New variable.
2255 (md_assemble): Wrap macro expansion in macro_start() and macro_end().
2256 (s_cpload, s_cpsetup, s_cprestore, s_cpreturn): Likewise.
2257 (relax_close_frag): Set mips_macro_warning.first_frag. Adjust use
2258 of RELAX_ENCODE.
2259 (append_insn): Update mips_macro_warning.sizes.
2260 (macro_start, macro_warning, macro_end): New functions.
2261 (macro_build): Don't emit warnings here.
2262 (macro_build_lui, md_estimate_size_before_relax): ...or here.
2263 (md_convert_frag): Check for cases where one macro alternative
2264 needs a warning and the other doesn't. Emit a warning if the
2265 longer sequence was chosen.
2266
4d7206a2
RS		 22672004-01-23 Richard Sandiford <rsandifo@redhat.com>
2268
2269 * config/tc-mips.h (tc_frag_data_type, TC_FRAG_TYPE): Remove.
2270 * config/tc-mips.c (RELAX_ENCODE): Take three arguments: the size of
2271 the first sequence, the size of the second sequence, and a flag
2272 that says whether we should warn.
2273 (RELAX_OLD, RELAX_NEW, RELAX_RELOC[123]): Delete.
2274 (RELAX_FIRST, RELAX_SECOND): New.
2275 (mips_relax): New variable.
2276 (relax_close_frag, relax_start, relax_switch, relax_end): New fns.
2277 (append_insn): Remove "place" argument. Use mips_relax.sequence
2278 rather than "place" to check whether we're expanding the second
2279 alternative of a relaxable macro. Remove redundant check for
2280 branch relaxation. If generating a normal insn, and there
2281 is not enough room in the current frag, call relax_close_frag()
2282 to close it. Update mips_relax.sizes[]. Emit fixups for the
2283 second version of a relaxable macro. Record the first relaxable
2284 fixup in mips_relax. Remove tc_gen_reloc workaround.
2285 (macro_build): Remove all uses of "place". Use mips_relax.sequence
2286 in the same way as in append_insn.
2287 (mips16_macro_build): Remove "place" argument.
2288 (macro_build_lui): As for macro_build. Don't drop the add_symbol
2289 when generating the second version of a relaxable macro.
2290 (load_got_offset, add_got_offset): New functions.
2291 (load_address, macro): Use new relaxation machinery. Remove
2292 tc_gen_reloc workarounds.
2293 (md_estimate_size_before_relax): Set RELAX_USE_SECOND if the second
2294 version of a relaxable macro is needed. Return -RELAX_SECOND if the
2295 first version is needed.
2296 (tc_gen_reloc): Remove relaxation handling.
2297 (md_convert_frag): Go through the fixups for a relaxable macro and
2298 mark those that belong to the unneeded alternative as done. If the
2299 second alternative is needed, adjust the fixup addresses to account
2300 for the deleted first alternative.
2301
0530d30a
RS		 23022004-01-23 Richard Sandiford <rsandifo@redhat.com>
2303
2304 * frags.h (frag_room): Declare.
2305 * frags.c (frag_room): New function.
2306 * doc/internals.texi: Document it.
2307
048cdf86
TS		 23082004-01-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
2309
2310 * config/tc-mips.c (append_insn): Don't do r3900 interlock
2311 optimization for -mtune=r3900, as this will break on other CPUs.
2312
28d39d1a
NC		 23132004-01-11 Tom Rix <tcrix@worldnet.att.net>
2314
2315 * config/tc-m68hc11.c (build_indexed_byte): movb and movw cannot
73369e65 2316 be relaxed, use fixup.
28d39d1a
NC		 2317 (md_apply_fix3): Use 5 bit reloc from movb and movw fixup.
2318
26664553
JJ		 23192004-01-19 Jakub Jelinek <jakub@redhat.com>
2320
2321 * config/tc-sparc.c (sparc_ip): Disallow %f32-%f63 for single
2322 precision operands.
2323
895921c9
MR		 23242004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2325
2326 * config/tc-mips.c (append_insn): Properly detect variant frags
2327 that preclude swapping of relaxed branches. Correctly swap
2328 instructions between frags when dealing with relaxed branches.
2329
f8b73030
MR		 23302004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2331
2332 * acinclude.m4: Quote names of macros to be defined by AC_DEFUN
2333 throughout.
2334 * aclocal.m4: Regenerate.
2335 * configure: Regenerate.
2336
ca9a79a1 23372004-01-12 Anil Paranjpe <anilp1@KPITCummins.com>
73369e65 2338
ca9a79a1
NC		 2339 * config/tc-h8300.c (build_bytes): Apply relaxation to bit
2340 manipulation insns.
2341
cc3d92a5
RS		 23422004-01-12 Richard Sandiford <rsandifo@redhat.com>
2343
2344 * config/tc-mips.c (macro_build_jalr): When adding an R_MIPS_JALR
2345 reloc, reserve space for the delay slot as well as the jalr itself.
2346
e45d0630
PB		 23472004-01-09 Paul Brook <paul@codesourcery.com>
2348
2349 * config/tc-arm.c (do_vfp_reg2_from_sp2): Rename from do_vfp_sp_reg2.
2350 (do_vfp_sp2_from_reg2): New function.
2351 (insns): Use them.
2352 (do_vfp_dp_from_reg2): Check return values properly.
2353
81912461
ILT		 23542004-01-08 Ian Lance Taylor <ian@wasabisystems.com>
2355
78849248
ILT		 2356 * config/tc-mips.c (warn_nops): Remove static variable.
2357 (macro): Remove test of warn_nops.
2358 (md_shortops): Remove 'n'.
2359 (md_parse_option): Remove 'n' case.
2360 (md_show_usage): Remove -n.
2361 * doc/as.texinfo (Overview): Remove MIPS -n option.
2362 * doc/c-mips.texi (MIPS Opts): Remove mention -n.
2363 * NEWS: Mention removal of MIPS -n option.
2364
81912461
ILT		 2365 * config/tc-mips.c (ISA_HAS_COPROC_DELAYS): Remove.
2366 (cop_interlocks): Check ISA level.
2367 (cop_mem_interlocks): Define.
2368 (reg_needs_delay): Check cop_interlocks rather than
2369 ISA_HAS_COPROC_DELAYS.
2370 (append_insn): Likewise. Use cop_mem_interlocks rather than
2371 directly checking mips_opts.isa.
2372 (mips_emit_delays): Likewise.
2373
73f20958
L		 23742004-01-07 H.J. Lu <hongjiu.lu@intel.com>
2375
2376 * config/tc-ia64.c (unwind): Move next_slot_number and
2377 next_slot_frag to ...
2378 (unw_rec_list): Here.
2379 (free_list_records): Removed.
2380 (output_unw_records): Likewise.
2381 (generate_unwind_image): Make it void.
2382 (alloc_record): Initialize next_slot_number and next_slot_frag.
2383 (slot_index): Take .org, .space and .align into account.
2384 (fixup_unw_records): Don't set slot_number to 0. Use
2385 list->next_slot_number and list->next_slot_frag instead of
2386 unwind.next_slot_number and unwind.next_slot_frag.
2387 (ia64_convert_frag): New.
2388 (generate_unwind_image): Generate a rs_machine_dependent frag
2389 for unwind record.
2390 (emit_one_bundle): Use list->next_slot_number and
2391 list->next_slot_frag instead of unwind.next_slot_number and
2392 unwind.next_slot_frag.
2393
2394 * config/tc-ia64.h (md_convert_frag): Defined as
2395 ia64_convert_frag.
2396 (md_estimate_size_before_relax): Defined as (f)->fr_var.
2397
99d09cdb
AO		 23982004-01-06 Alexandre Oliva <aoliva@redhat.com>
2399
2400 2003-12-19 Alexandre Oliva <aoliva@redhat.com>
2401 * config/tc-frv.h (md_apply_fix3): Don't define.
2402 * config/tc-frv.c (md_apply_fix3): New. Shift/truncate %hi/%lo
2403 operands.
2404 * config/tc-frv.h (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
2405 2003-10-07 Alexandre Oliva <aoliva@redhat.com>
2406 * config/tc-frv.c (line_separator_chars): Add `!'.
2407 2003-09-19 Alexandre Oliva <aoliva@redhat.com>
2408 * config/tc-frv.c (md_assemble): Clear insn upfront.
2409 2003-09-18 Alexandre Oliva <aoliva@redhat.com>
2410 * config/tc-frv.c (OPTION_FDPIC): New macro.
2411 (md_longopts): Add mfdpic.
2412 (md_parse_option): Handle it.
2413 2003-08-04 Alexandre Oliva <aoliva@redhat.com>
2414 * config/tc-frv.c (md_cgen_lookup_reloc) <FRV_OPERAND_D12,
2415 FRV_OPERAND_S12>: Use reloc type encoded in fix-up.
2416 (frv_pic_ptr): Parse funcdesc.
2417
4a4c4a1d
MR		 24182004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2419
2420 * doc/as.texinfo: Let texi2pod parse asconfig.texi and
2421 gasver.texi. Remove duplicate symbol definitions for texi2pod.
2422
2251dd89
MR		 24232004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2424
2425 * Makefile.am (Makefile): Move the dependency on
2426 $(BFDDIR)/configure.in to...
2427 (CONFIG_STATUS_DEPENDENCIES): ... here.
2428 (AUTOMAKE_OPTIONS): Require automake 1.8.
2429 * Makefile.in: Regenerate.
2430 * doc/Makefile.am (BASEDIR, BFDDIR): Define.
2431 (CONFIG_STATUS_DEPENDENCIES): Add a dependency on
2432 $(BFDDIR)/configure.in here as well.
2433 * doc/Makefile.in: Regenerate.
2434
c6581168
MR		 24352004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2436
2437 * Makefile.am (install, install-info, RECURSIVE_TARGETS): Remove.
2438 * Makefile.in: Regenerate.
2439 * aclocal.m4: Regenerate.
2440 * doc/Makefile.am (install, install-info): Remove.
2441 (install-data-local): A new hook for install-info.
2442 (AUTOMAKE_OPTIONS): Require automake 1.8.
2443 * doc/Makefile.in: Regenerate.
2444
303e7b79
NC		 24452004-01-02 Nutan Singh <nutan@kpitcummins.com>
2446
2447 * doc/c-sh.texi: Update description about floating point behavior
2448 of SH family.
2449
b50c15dc
NC		 24502004-01-02 Bernardo Innocenti <bernie@develer.com>
2451
2452 * configure.in: Add m68k-uClinux target.
2453 * configure: Regenerate.
7de9afa2 2454
c9e214e5 2455For older changes see ChangeLog-0203
08d56133
NC		 2456\f
2457Local Variables:
2458mode: change-log
2459left-margin: 8
2460fill-column: 74
2461version-control: never
2462End:
GDB Binutils - Deliverables
This page took 0.425242 seconds and 4 git commands to generate.