projects / deliverable / binutils-gdb.git / blame
? search:
summary | shortlog | log | commit | commitdiff | tree
blob | blame (incremental) | history | HEAD
In last change: macros not rules.
[deliverable/binutils-gdb.git] / gas / ChangeLog
CommitLineData
a1ace8d8 12004-11-12 Bob Wilson <bob.wilson@acm.org>
88ac794e 2 Sterling Augustine <sterling@tensilica.com>
a1ace8d8
BW		 3
4 * config/tc-xtensa.c (finish_vinsn): Clear pending instruction if
5 there is a conflict.
6 (check_t1_t2_reads_and_writes): Check for both reads and writes to
7 interfaces that are related as determined by xtensa_interface_class_id.
8
30e857fc
NC		 92004-11-12 Nick Clifton <nickc@redhat.com>
10
11 * config/tc-mn10300.c (md_relax_table): Fix off by one negative
12 offsets for conditional branches.
13
a67517f4
BW		 142004-11-11 Bob Wilson <bob.wilson@acm.org>
15
16 * config/tc-xtensa.c (MAX_IMMED6): Change value to 65.
17
1d19a770
BW		 182004-11-10 Bob Wilson <bob.wilson@acm.org>
19
20 * config/tc-xtensa.c (update_next_frag_state): Always add a NOP if
21 relaxing at the end of a loop. Don't mark frags as UNREACHABLE or
22 MAYBE_UNREACHABLE.
23 (relax_frag_immed): Update call to update_next_frag_state.
24
c04f5787
AM		 252004-11-10 Alan Modra <amodra@bigpond.net.au>
26
27 * obj.h (struct format_ops <app_file>): Add int param.
28 * read.h (s_app_file_string): Likewise.
29 * read.c (s_app_file_string): Likewise.
30 (s_app_file): Adjust s_app_file_string call.
31 * config/tc-mips.c (s_mips_file): Likewise.
32 * config/obj-coff.h (obj_app_file): Add app param.
33 * config/obj-ecoff.h (obj_app_file): Likewise.
34 * config/obj-multi.h (obj_app_file): Likewise.
35 * config/obj-elf.h (elf_file_symbol): Likewise.
36 * config/obj-elf.c (elf_file_symbol): Only emit one file symbol
37 if called for # preprocessor lines.
38
a77a9021
L		 392004-11-08 H.J. Lu <hongjiu.lu@intel.com>
40
41 PR 528
42 * symbols.c (resolve_symbol_value): Convert weak symbols only
43 for Windows PECOFF.
44 (symbol_equated_reloc_p): Don't equate weaks when relocating
45 only for Windows PECOFF.
46
aea44f62
TL		 472004-11-08 Tomer Levi <Tomer.Levi@nsc.com>
48
49 * config/tc-crx.c (print_insn): Check and set insn_addr.
50 * config/tc-crx.h (md_frag_check): Define.
51
7499d566
NC		 522004-11-08 Inderpreet Singh <inderpreetb@nioda.hcltech.com>
53 Vineet Sharma <vineets@noida.hcltech.com>
54
55 * configure.in: Add support for new target maxq-coff.
56 * configure: Regenerate.
57 * NEWS: Mention new support.
58 * config/tc-maxq.c: New file.
59 * config/tc-maxq.h: New file.
60 * config/obj-coff.h: Add support for maxq-coff.
61
977cdf5a
NC		 622004-11-08 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
63
64 * symbols.c (any_external_name): Define.
a77a9021 65 (resolve_symbol_value): Convert weak symbols.
977cdf5a
NC		 66 (S_SET_EXTERNAL): Support any_external_name.
67 (S_SET_NAME): Qualify parameter const.
316f3bf8 68 (symbol_equated_reloc_p): Equate to weaks when relocating.
977cdf5a
NC		 69 * symbols.h (S_SET_NAME): Qualfiy parameter const.
70 * tc.h (any_external_name): Declare.
71 * config/obj-coff.c ("coff/pe.h"): Include for BFD
72 assemblers also.
73 (weak_is_altname): Declare and define.
74 (weak_name2altname): Same.
75 (weak_altname2name): Same.
76 (weak_uniquify): Same.
77 (weak_altprefix): Define.
78 (obj_coff_weak): Change .weak syntax and handling.
79 (coff_frob_symbol): Fix PE weak symbol alternates.
80 * config/obj-coff.h (USE_UNIQUE): Define.
81 * config/tc-i386.c (md_apply_fix3): Assume weak symbols
82 are in another segment.
83 (tc_gen_reloc): Remove broken addend hack.
84 doc/as.texinfo: Update.
85
34e41783
BW		 862004-11-05 Sterling Augustine <sterling@tensilica.com>
87
88 * config/tc-xtensa.c (total_frag_text_expansion): New.
89 (md_estimate_size_before_relax): Use it.
90 (find_address_of_next_align_frag): Likewise.
91
dcd46b4e
TL		 922004-11-05 Tomer Levi <Tomer.Levi@nsc.com>
93
94 * config/tc-crx.c: Rename argument types.
95 (processing_arg_number): Rename to 'cur_arg_num'.
96 (get_number_of_bits): Rename to 'set_operand_size'.
97 (get_operandtype): Rename to 'parse_operand', totally rewrite.
98 (set_cons_rparams): Rename to 'set_operand', totally rewrite.
99 (set_indexmode_parameters): Remove function, integrate its code into
100 'set_operand'.
101 (set_operand_size): Get rid of 'Operand Number' function parameter -
102 use global variable 'cur_arg_num' instead.
103 Use a local 'argument' pointer to reference the current argument.
104 (parse_operand): Likewise.
105 (set_operand): Likewise.
106 (process_label_constant): Likewise.
107
9456465c
BW		 1082004-11-04 Bob Wilson <bob.wilson@acm.org>
109
110 * config/tc-xtensa.c: Remove XTENSA_SECTION_RENAME ifdefs.
111 (add_section_rename): Delete. Inlined into...
112 (build_section_rename): ...here. Use xstrdup instead of strdup.
113 (xtensa_section_rename): Drop "const" from argument and return types.
114 (md_show_usage): Indent to match show_usage().
115 * config/tc-xtensa.h: Remove XTENSA_SECTION_RENAME ifdefs.
116 (tc_canonicalize_section_name): Define.
117 (md_elf_section_rename): Remove unused macro.
118 * doc/as.texinfo (Overview): Document Xtensa --rename-section option.
119 * doc/c-xtensa.texi (Xtensa Options): Likewise.
120 (Frame Directive): Delete.
121
7f266840
DJ		 1222004-11-04 Daniel Jacobowitz <dan@debian.org>
123
124 * configure.in: Remove arm-*-oabi and thumb-*-oabi.
125 * config/tc-arm.c (target_oabi): Delete.
126 (md_apply_fix3, elf32_arm_target_format): Remove target_oabi checks.
127 (arm_opts): Remove moabi.
128 * doc/as.texinfo (Overview): Remove documentation of -moabi.
129 * doc/c-arm.texi (ARM Options): Likewise.
130 * configure: Regenerated.
131
ae57792d
HPN		 1322004-11-04 Hans-Peter Nilsson <hp@axis.com>
133
134 * configure.in (crisv32): Recognize. AC_DEFINE_UNQUOTED
135 DEFAULT_CRIS_ARCH. Handle crisv32-*-linux-gnu* like
136 cris-*-linux-gnu* and crisv32-*-* like cris-*-*.
137 * configure: Regenerate.
138 * config/tc-cris.c (enum cris_archs): New.
139 (cris_mach, cris_arch_from_string, s_cris_arch, get_sup_reg)
140 (cris_insn_ver_valid_for_arch): New functions.
141 (DEFAULT_CRIS_ARCH): New macro, default to cris_any_v0_v10.
142 (cris_arch): New variable.
143 (md_pseudo_table): New pseudo .arch.
144 (err_for_dangerous_mul_placement): Initialize according to
145 DEFAULT_CRIS_ARCH.
146 (STATE_COND_BRANCH): Renamed from STATE_CONDITIONAL_BRANCH.
147 All users changed.
148 (STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON)
149 (STATE_ABS_BRANCH_V32, STATE_LAPC, BRANCH_BF_V32, BRANCH_BB_V32)
150 (BRANCH_WF_V32, BRANCH_WB_V32): New.
151 (BRANCH_BF, BRANCH_BB, BRANCH_WF, BRANCH_WB): Don't undef after
152 use in md_cris_relax_table.
153 (md_cris_relax_table): Add entries for STATE_COND_BRANCH_V32,
154 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
155 Update and improve head comment.
156 (OPTION_PIC): Define in terms of previous option, OPTION_US.
157 (OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): Similar.
158 (OPTION_ARCH): New.
159 (md_longopts): New option --march=...
160 (cris_any_v0_v10_long_jump_size, crisv32_long_jump_size): New
161 macros.
162 (md_long_jump_size): Initialize in terms of DEFAULT_CRIS_ARCH.
163 (HANDLE_RELAXABLE): New macro.
164 (md_estimate_size_before_relax): Use HANDLE_RELAXABLE for common
165 cases. Check for weak symbols and assume not relaxable. Handle
166 STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON,
167 STATE_ABS_BRANCH_V32, STATE_LAPC. Use new variable symbolP, not
168 fragP->fr_symbol.
169 (md_convert_frag): Handle STATE_COND_BRANCH_V32,
170 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
171 (cris_create_short_jump): Adjust for CRISv32.
5664e9a7 172 (cris_relax_frag): Handle new states.
ae57792d
HPN		 173 (md_create_long_jump): Ditto. Emit error for common_v10_v32.
174 (md_begin): Define symbols "..asm.arch.cris.v32",
175 "..asm.arch.cris.v10", "..asm.arch.cris.common_v10_v32" and
176 "..asm.arch.cris.any_v0_v10". Use cris_insn_ver_valid_for_arch
177 when entering opcode table entry points.
178 (md_assemble): Adjust branch handling for CRISv32. Handle LAPC
179 relaxation. In fix_new_exp call for main insn, pass 1 for pcrel
180 parameter for 8, 16 and 32-bit pc-relative insns and LAPC.
181 (cris_process_instruction): Initialize out_insnp->insn_type to
182 CRIS_INSN_NONE, not CRIS_INSN_NORMAL.
183 <case ']', '[', 'A', 'd', 'Q', 'N', 'n', 'Y', 'U', 'u', 'T'>: New
184 cases.
185 <case 'm'>: Check that modified_char == '.'.
186 <invalid operands>: Consume the rest of the line.
187 When operands don't match, skip over subsequent insns with
188 non-matching version specifier but same mnemonic.
189 <immediate constant, case SIZE_SPEC_REG>: Immediate operands for
190 special registers in CRISv32 are always 32 bit long.
191 <immediate constant, case SIZE_FIELD_SIGNED, SIZE_FIELD_UNSIGNED>:
192 New cases.
193 (get_gen_reg): Only recognize "PC" when followed by "+]" for v32
194 and compatible. Recognize "ACR" for v32, unless followed by "+".
195 (get_spec_reg): Consider cris_arch when looking up register.
196 (get_autoinc_prefix_or_indir_op): Don't recognize assignment for
197 v32 or compatible.
198 (get_3op_or_dip_prefix_op): Check for ']' after seeing '[rN+'.
199 (cris_get_expression): Restore input_line_pointer if failing "early".
200 (get_flags): Consider cris_arch and recognize flags accordingly.
201 (branch_disp): Adjust for CRISv32.
202 (gen_cond_branch_32): Similar. Emit error for common_v10_v32.
203 (cris_number_to_imm): Use as_bad_where, not as_bad. Remove
204 related FIXME. Don't insist on BFD_RELOC_32_PCREL fixup to be
205 resolved. Don't enter zeros in object file for
206 BFD_RELOC_32_PCREL.
207 <case BFD_RELOC_CRIS_LAPCQ_OFFSET, BFD_RELOC_CRIS_SIGNED_16>
208 <case BFD_RELOC_CRIS_SIGNED_8>: New case.
209 (md_parse_option): Break out "return 1".
210 <OPTION_ARCH> New case.
211 (tc_gen_reloc): <case BFD_RELOC_CRIS_LAPCQ_OFFSET>
212 <case BFD_RELOC_CRIS_SIGNED_16, BFD_RELOC_CRIS_SIGNED_8>
213 <case BFD_RELOC_CRIS_UNSIGNED_8, BFD_RELOC_CRIS_UNSIGNED_16>
214 <case BFD_RELOC_32_PCREL>: New cases.
215 Addends for non-zero fx_pcrel are too in fx_offset.
216 (md_show_usage): Show --march=<arch>.
217 (md_apply_fix3): Adjust val for BFD_RELOC_CRIS_LAPCQ_OFFSET.
218 (md_pcrel_from): BFD_RELOC_CRIS_LAPCQ_OFFSET is PC-relative too.
219 (s_syntax) <struct syntaxes>: Properly constify member operand.
220 * config/tc-cris.h (TARGET_MACH): Define.
221 (cris_mach): Declare.
222 * doc/as.texinfo (Overview) <CRIS>: Add --march=...
223 * doc/c-cris.texi (CRIS-Symbols): New node for built-in symbols.
224 (CRIS-Opts): Document --march=...
225 (CRIS-Pseudos): Document .arch.
226
9306ca4a
JB		 2272004-11-04 Jan Beulich <jbeulich@novell.com>
228
229 * config/tc-i386.c (set_intel_syntax): Allow % in symbol names when
230 intel syntax and no register prefix, allow $ in symbol names when
231 intel syntax.
232 (set_16bit_gcc_code_flag): Replace literal 'l' by LONG_MNEM_SUFFIX.
233 (intel_float_operand): Add fourth return value indicating math control
234 operations. Make classification more precise.
235 (md_assemble): Complain if memory operand of mov[sz]x has no size
236 specified.
237 (parse_insn): Translate word operands to floating point instructions
238 operating on integers as well as control instructions to short ones
239 as expected by AT&T syntax. Translate 'd' suffix to short one only for
240 floating point instructions operating on non-integer operands.
241 (match_template): Remove fldcw special case. Adjust q-suffix handling
242 to permit it on fild/fistp/fisttp in AT&T mode.
243 (process_suffix): Don't guess DefaultSize insns' suffix from
244 stackop_size for certain floating point control instructions. Guess
245 suffix for branch and [ls][gi]dt based on flag_code. Split error
246 messages for Intel and AT&T syntax, and make the condition more strict
247 for the former. Adjust suppressing of generation of operand size
248 overrides.
249 (intel parser): Allow the full set of MASM operators. Add FWORD, TBYTE,
250 OWORD, and XMMWORD operand size specifiers (TBYTE replaces XWORD). Add
251 more error checking.
252 * config/tc-i386.h (BYTE_PTR WORD_PTR DWORD_PTR QWORD_PTR XWORD_PTR
253 SHORT OFFSET_FLAT FLAT NONE_FOUND): Remove unused defines.
254
2b4f075a
HPN		 2552004-11-03 Hans-Peter Nilsson <hp@axis.com>
256
257 * symbols.c (colon) [!WORKING_DOT_WORD]: Don't declare
258 md_short_jump_size, md_long_jump_size.
259 * write.c [!WORKING_DOT_WORD]: Ditto.
260 * tc.h [!WORKING_DOT_WORD]: Declare them here. Drop const
261 qualifier.
262 * config/tc-cris.h (md_short_jump_size, md_long_jump_size): Don't
263 declare.
264 * config/tc-cris.c (md_short_jump_size, md_long_jump_size): Drop
265 const qualifier in these definitions.
266 * config/tc-i370.c, config/tc-m68k.c, config/tc-pdp11.c,
267 config/tc-s390.c, config/tc-tahoe.c, config/tc-vax.c: Ditto.
268
ddbc47ac
NC		 2692004-11-02 Nick Clifton <nickc@redhat.com>
270
271 * dwarf2dbg.c (dwarf2_finish): Check for the existence of a file
272 table before deciding to produce a .debug_line section to match up
273 with a user provided .debug_info section.
274
3ad3f5ad
TL		 2752004-10-28 Tomer Levi <Tomer.Levi@nsc.com>
276
277 * config/tc-crx.c (getreg_image): Bug fix, a return value was
278 mistakenly omitted from CRX_C_REGTYPE and CRX_CS_REGTYPE cases.
279
c815a616
TL		 2802004-10-27 Tomer Levi <Tomer.Levi@nsc.com>
281
282 * config/tc-crx.c: Remove global variable 'post_inc_mode'.
283 (get_flags): New function.
c04f5787
AM		 284 (get_number_of_bits): Edit comments, update numeric values to
285 supported sizes.
286 (process_label_constant): Don't support the colon
287 format (SYMBOL:[s|m|l]).
c815a616
TL		 288 (set_cons_rparams): Support argument type 'arg_rbase'.
289 (get_operandtype): Bug fix in 'rbase' operand type parsing.
290 (handle_LoadStor): Bug fix, first handle post-increment mode.
c04f5787
AM		 291 (getreg_image): Remove redundant code, update according to latest
292 CRX spec.
c815a616 293 (print_constant): Bug fix relate to 3-word instructions.
c04f5787 294 (assemble_insn): Bug fix, when matching instructions, verify also
c815a616
TL		 295 instruction type (not only mnemonic).
296 Add various error checking.
297 (preprocess_reglist): Support HI/LO and user registers.
298
ed7af9f9
L		 2992004-10-25 David Mosberger-Tang <davidm@hpl.hp.com>
300
c04f5787 301 * config/tc-ia64.c (fixup_unw_records): Don't let the "t" value
ed7af9f9
L		 302 in an epilogue directive go negative.
303
ac025970
L		 3042004-10-25 H.J. Lu <hongjiu.lu@intel.com>
305
306 PR 474
307 * config/tc-ia64.c (emit_one_bundle): Decrement md.num_slots_in_use
308 after reporting template error during manual bundling. Reported
309 by Michael Dupont, michaelx.dupont@intel.com.
310
a394c00f
NC		 3112004-10-25 Daniel Jacobowitz <dan@debian.org>
312
313 * Makefile.am: Run dep-am.
314 * aclocal.m4: Regenerate with automake 1.9.2.
315 * Makefile.in: Regenerate with automake 1.9.2.
316 * doc/Makefile.in: Likewise.
317
318 * config/tc-arm.c: Include "dw2gencfi.h".
319 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
320 New functions.
321 * config/tc-arm.h (TARGET_USE_CFIPOP, DWARF2_DEFAULT_RETURN_COLUMN)
322 (DWARF2_CIE_DATA_ALIGNMENT, tc_regname_to_dw2regnum)
323 (tc_cfi_frame_initial_instructions): Define.
324 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
325 Add prototypes.
326
23d83918
TL		 3272004-10-21 Tomer Levi <Tomer.Levi@nsc.com>
328
329 * config/tc-crx.c (assemble_insn): Check unsigned immediate
330 operands validity.
331 Update coprocessor id to be unsigned immediate.
332
bb41ade5
AM		 3332004-10-18 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
334
335 * config/tc-i386.c (O_secrel): Delete.
336 (tc_pe_dwarf2_emit_offset): New function.
337 * config/tc-i386.h (O_secrel): Define as O_md1.
338 (TC_DWARF2_EMIT_OFFSET): Define.
339
87e62348
NC		 3402004-10-18 Nick Clifton <nickc@redhat.com>
341
342 * config/tc-xstormy16.c (xstormy16_cons_fix_new): Accept and
343 ignore @fptr() directives for 4-byte fixups.
344
3c55da70
AM		 3452004-10-15 Alan Modra <amodra@bigpond.net.au>
346
347 * Makefile.am: Run "make dep-am"
348 * Makefile.in: Regenerate.
349
40e7b7b9
BW		 3502004-10-14 Bob Wilson <bob.wilson@acm.org>
351
352 * doc/c-xtensa.texi (Xtensa Options, Absolute Literals Directive):
353 Remove comments about placement of literal pools.
354 (Literal Directive): Update description of literal placement.
355 (Literal Prefix Directive): Remove statement that this does not apply
356 to absolute-mode literals. Describe new section naming scheme.
357
b08b5071
BW		 3582004-10-12 Bob Wilson <bob.wilson@acm.org>
359
360 * config/tc-xtensa.c (xtensa_frequency_pseudo): Use set_subseg_freq.
361 (is_entry_opcode, is_movi_opcode, is_the_loop_opcode, is_jx_opcode,
362 is_windowed_return_opcode): Delete.
363 (xtensa_frob_label): Use get_subseg_target_freq.
364 (md_assemble): Inline call to is_entry_opcode.
365 (xtensa_handle_align): Inline call to get_frag_is_literal.
366 (relaxation_requirements): Inline call to is_jx_opcode.
367 (emit_single_op): Inline call to is_movi_opcode.
368 (xg_assemble_vliw_tokens): Inline calls to get_frag_is_insn,
369 get_frag_is_no_transform, is_entry_opcode, and
370 set_frag_is_specific_opcode. Use get_subseg_total_freq.
371 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags,
372 xtensa_fix_close_loop_end_frags, relax_frag_immed, convert_frag_immed):
373 Inline calls to get_frag_is_no_transform.
374 (next_instrs_are_b_retw): Inline call to is_windowed_return_opcode.
375 (xtensa_fix_short_loop_frags): Inline calls to is_the_loop_opcode and
376 get_frag_is_no_transform.
377 (convert_frag_immed_finish_loop): Inline calls to get_expression_value
378 and set_frag_is_no_transform.
379 (get_expression_value): Delete.
380 (subseg_map struct): Rename cur_total_freq to total_freq. Rename
381 cur_target_freq to target_freq.
382 (get_subseg_info): Split out code to create a new map entry into ...
383 (add_subseg_info): ... this new function.
384 (get_last_insn_flags): Check if get_subseg_info succeeded.
385 (set_last_insn_flags): Call add_subseg_info if needed.
386 (get_subseg_total_freq, get_subseg_target_freq, set_subseg_freq): New.
387 (xtensa_reorder_segments): Compute last_sec while counting sections.
388 Remove call to get_last_sec.
389 (get_last_sec): Delete.
390 (cache_literal_section): Inline call to retrieve_literal_seg and its
391 callees, seg_present and add_seg_list.
392 (retrieve_literal_seg, seg_present, add_seg_list): Delete.
393 (get_frag_is_insn, get_frag_is_no_transform,
394 set_frag_is_specific_opcode, set_frag_is_no_transform): Delete.
395 * config/tc-xtensa.h (MAX_SLOTS): Reduce from 31 to 15.
396
7fa3d080
BW		 3972004-10-12 Bob Wilson <bob.wilson@acm.org>
398
399 * config/tc-xtensa.c: Use ISO C90 formatting.
400 * config/tc-xtensa.h: Likewise.
401 * config/xtensa-istack.h: Likewise.
402 * config/xtensa-relax.c: Likewise.
403 * config/xtensa-relax.h: Likewise.
404
8cb51566
PB		 4052004-10-12 Paul Brook <paul@codesourcery.com>
406
407 * config/tc-arm.c (md_begin): Change EF_ARM_EABI_VER3 to
408 EF_ARM_EABI_VER4.
409 (arm_eabis): Ditto.
410 * doc/c-arm.texi: Document that we actually support -meabi=4, not
411 -meabi=3.
412
653cfe85
BW		 4132004-10-08 Bob Wilson <bob.wilson@acm.org>
414
415 * doc/as.texinfo (VTableEntry, VTableInherit): Add "directive" to index
416 entries.
417 (Acknowledgements): Use "GAS" instead of AS variable.
418
d2b2c203
DJ		 4192004-10-08 Daniel Jacobowitz <dan@debian.org>
420
421 * config/tc-i386.c: Include "elf/x86-64.h".
422 (i386_elf_section_type): New function.
423 * config/tc-i386.h (md_elf_section_type): Define.
424 (i386_elf_section_type): New prototype.
425
e80e0390
NC		 4262004-10-08 Linus Nielsen Feltzing <linus@haxx.se>
427
428 * config/m68k-parse.h (enum m68k_register): New control register,
429 MBAR2 (for MCF5249)
430 * config/tc-m68k.c: Correct control register set for MCF5249.
431
43cd72b9 4322004-10-07 Bob Wilson <bob.wilson@acm.org>
85a5ac80 433 Sterling Augustine <sterling@tensilica.com>
43cd72b9
BW		 434
435 * config/tc-xtensa.c (absolute_literals_supported): New global flag.
436 (UNREACHABLE_MAX_WIDTH): Define.
437 (XTENSA_FETCH_WIDTH): Delete.
438 (cur_vinsn, xtensa_fetch_width, xt_saved_debug_type, past_xtensa_end,
439 prefer_const16, prefer_l32r): New global variables.
440 (LIT4_SECTION_NAME): Define.
441 (lit4_state struct): Add lit4_seg_name and lit4_seg fields.
442 (XTENSA_PROP_*, GET_XTENSA_PROP_*, SET_XTENSA_PROP_*): Define.
443 (frag_flags struct): New.
c04f5787 444 (xtensa_block_info struct): Move from tc-xtensa.h. Add flags field.
43cd72b9
BW		 445 (subseg_map struct): Add cur_total_freq and cur_target_freq fields.
446 (bitfield, bit_is_set, set_bit, clear_bit): Define.
447 (MAX_FORMATS): Define.
448 (op_placement_info struct, op_placement_table): New.
449 (O_pltrel, O_hi16, O_lo16): Define.
450 (directiveE enum): Rename directive_generics to directive_transform.
451 Delete directive_relax. Add directive_schedule,
452 directive_absolute_literals, and directive_last_directive.
453 (directive_info): Rename "generics" to "transform". Delete "relax".
454 Add "schedule" and "absolute-literals".
455 (directive_state): Adjust entries to match changes in directive_info.
456 (xtensa_relax_statesE, RELAX_IMMED_MAXSTEPS): Move to tc-xtensa.h.
457 (xtensa_const16_opcode, xtensa_movi_opcode, xtensa_movi_n_opcode,
458 xtensa_l32r_opcode, xtensa_nop_opcode, xtensa_rsr_lcount_opcode): New.
459 (xtensa_j_opcode, xtensa_rsr_opcode): Delete.
460 (align_only_targets, software_a0_b_retw_interlock,
461 software_avoid_b_j_loop_end, maybe_has_b_j_loop_end,
462 software_avoid_short_loop, software_avoid_close_loop_end,
463 software_avoid_all_short_loops, specific_opcode): Delete.
464 (warn_unaligned_branch_targets): New.
465 (workaround_a0_b_retw, workaround_b_j_loop_end, workaround_short_loop,
466 workaround_close_loop_end, workaround_all_short_loops): Default FALSE.
467 (option_[no_]link_relax, option_[no_]transform,
468 option_[no_]absolute_literals, option_warn_unaligned_targets,
469 option_prefer_l32r, option_prefer_const16, option_target_hardware):
470 New enum values.
471 (option_[no_]align_only_targets, option_literal_section_name,
472 option_text_section_name, option_data_section_name,
473 option_bss_section_name, option_eb, option_el): Delete.
474 (md_longopts): Add entries for: [no-]transform, [no-]absolute-literals,
475 warn-unaligned-targets, prefer-l32r, prefer-const16, [no-]link-relax,
476 and target-hardware. Delete entries for [no-]target-align-only,
477 literal-section-name, text-section-name, data-section-name, and
478 bss-section-name.
479 (md_parse_option): Handle new options and remove old ones. Accept but
480 ignore [no-]density options. Warn for [no-]generics and [no-]relax
481 and treat them as [no-]transform.
482 (md_show_usage): Add new options and remove old ones.
483 (xtensa_setup_hw_workarounds): New.
484 (md_pseudo_table): Change "word" entry to use xtensa_elf_cons. Add
485 "long", "short", "loc" and "frequency" entries.
486 (use_generics): Rename to ...
487 (use_transform): ... this function. Add past_xtensa_end check.
488 (use_longcalls): Add past_xtensa_end check.
489 (code_density_available, can_relax): Delete.
490 (do_align_targets): New.
491 (get_directive): Accept dashes in directive names. Warn about
492 [no-]generics and [no-]relax directives and treat them as
c04f5787 493 [no-]transform.
43cd72b9
BW		 494 (xtensa_begin_directive): Call md_flush_pending_output only for some
495 directives. Check for directives inside instruction bundles. Warn
496 about deprecated ".begin literal" usage. Warn and ignore [no-]density
497 directives. Handle new directives. Check generating_literals flag
498 for literal_prefix.
499 (xtensa_end_directive): Check for directives inside instruction
500 bundles. Warn and ignore [no-]density directives. Handle new
501 directives. Call xtensa_set_frag_assembly_state.
502 (xtensa_loc_directive_seen, xtensa_dwarf2_directive_loc,
503 xtensa_dwarf2_emit_insn): New.
504 (xtensa_literal_position): Call md_flush_pending_output. Do not check
505 use_literal_section flag.
506 (xtensa_literal_pseudo): Call md_flush_pending_output. Handle absolute
507 literals. Use xtensa_elf_cons to parse the expression.
508 (xtensa_literal_prefix): Do not check use_literal_section. Support
509 ".lit4" sections for absolute literals. Change prefix convention to
510 replace ".text" (or ".t" in a linkonce section). No need to call
511 subseg_set.
512 (xtensa_frequency_pseudo, xtensa_elf_cons, xtensa_elf_suffix): New.
513 (expression_end): Handle closing braces and colons.
514 (PLT_SUFFIX, plt_suffix): Delete.
515 (expression_maybe_register): Use new xtensa-isa.h functions. Use
516 xtensa_elf_suffix instead of checking for plt suffix, and handle O_lo16
517 and O_hi16 expressions as well.
518 (tokenize_arguments): Handle closing braces and colons.
519 (parse_arguments): Use new xtensa-isa.h functions. Handle "invisible"
520 operands and paired register syntax.
521 (get_invisible_operands): New.
522 (xg_translate_sysreg_op): Handle new Xtensa LX RSR/WSR/XSR syntax. Use
523 new xtensa-isa.h functions.
524 (xtensa_translate_old_userreg_ops, xtensa_translate_zero_immed): New.
525 (xg_translate_idioms): Check if inside bundle. Use use_transform.
526 Handle new Xtensa LX RSR/WSR/XSR syntax. Remove code to widen density
527 instructions. Use xtensa_translate_zero_immed.
528 (operand_is_immed, operand_is_pcrel_label): Delete.
529 (get_relaxable_immed): Use new xtensa-isa.h functions.
530 (get_opcode_from_buf): Add slot parameter. Use new xtensa-isa.h
531 functions.
532 (xtensa_print_insn_table, print_vliw_insn): New.
533 (is_direct_call_opcode): Use new xtensa-isa.h functions.
534 (is_call_opcode, is_loop_opcode, is_conditional_branch_opcode,
535 is_branch_or_jump_opcode): Delete.
536 (is_movi_opcode, decode_reloc, encode_reloc, encode_alt_reloc): New.
537 (opnum_to_reloc, reloc_to_opnum): Delete.
538 (xtensa_insnbuf_set_operand, xtensa_insnbuf_get_operand): Use new
539 xtensa-isa.h functions. Operate on one slot of an instruction.
540 (xtensa_insnbuf_set_immediate_field, is_negatable_branch,
541 xg_get_insn_size): Delete.
542 (xg_get_build_instr_size): Use xg_get_single_size.
543 (xg_is_narrow_insn, xg_is_single_relaxable_insn): Update calls to
544 xg_build_widen_table. Use xg_get_single_size.
545 (xg_get_max_narrow_insn_size): Delete.
546 (xg_get_max_insn_widen_size, xg_get_max_insn_widen_literal_size,
547 xg_is_relaxable_insn): Update calls to xg_build_widen_table. Use
548 xg_get_single_size.
549 (xg_build_to_insn): Record the loc field. Handle OP_OPERAND_HI16U and
550 OP_OPERAND_LOW16U. Check xg_valid_literal_expression.
551 (xg_expand_to_stack, xg_expand_narrow): Update calls to
552 xg_build_widen_table. Use xg_get_single_size.
553 (xg_immeds_fit): Use new xtensa-isa.h functions. Update call to
554 xg_check_operand.
555 (xg_symbolic_immeds_fit): Likewise. Also handle O_lo16 and O_hi16, and
556 treat weak symbols conservatively.
557 (xg_check_operand): Use new xtensa-isa.h functions.
558 (is_dnrange): Delete.
559 (xg_assembly_relax): Inline previous calls to tinsn_copy.
560 (xg_finish_frag): Specify separate relax states for the frag and slot0.
561 (is_branch_jmp_to_next, xg_add_branch_and_loop_targets): Use new
562 xtensa-isa.h functions.
563 (xg_instruction_matches_option_term, xg_instruction_matches_or_options,
564 xg_instruction_matches_options): New.
565 (xg_instruction_matches_rule): Handle O_register expressions. Call
566 xg_instruction_matches_options.
567 (transition_rule_cmp): New.
568 (xg_instruction_match): Update call to xg_build_simplify_table.
569 (xg_build_token_insn): Record loc fields.
570 (xg_simplify_insn): Check is_specific_opcode field and
571 density_supported flag.
572 (xg_expand_assembly_insn): Skip checking code_density_available. Use
573 new xtensa-isa.h functions. Call use_transform instead of can_relax.
574 (xg_assemble_literal): Add error handling for O_big. Call
575 record_alignment. Handle O_pltrel.
576 (xg_valid_literal_expression): New.
577 (xg_assemble_literal_space): Add slot parameter. Remove call to
578 set_expr_symbol_offset. Add call to record_alignment. Update call to
579 xg_finish_frag.
580 (xg_emit_insn): Delete.
581 (xg_emit_insn_to_buf): Add format parameter. Update calls to
582 xg_add_opcode_fix and xtensa_insnbuf_to_chars.
583 (xg_add_opcode_fix): Change opcode parameter to tinsn and add format
584 and slot parameters. Handle new "alternate" relocations for absolute
585 literals and CONST16 instructions. Check for bad uses of O_lo16 and
586 O_hi16. Use new xtensa-isa.h functions.
587 (xg_assemble_tokens): Delete.
588 (is_register_writer): Use new xtensa-isa.h functions.
589 (is_bad_loopend_opcode): Check for xtensa_rsr_lcount_opcode instead of
590 old-style RSR from LCOUNT.
591 (next_frag_opcode): Delete.
592 (next_frag_opcode_is_loop, next_frag_format_size, frag_format_size,
593 update_next_frag_state): New.
594 (update_next_frag_nop_state): Delete.
595 (next_frag_pre_opcode_bytes): Use next_frag_opcode_is_loop.
596 (xtensa_mark_literal_pool_location): Check use_literal_section flag and
597 the state of the absolute-literals directive. Add calls to
598 record_alignment and xtensa_set_frag_assembly_state. Call
599 xtensa_switch_to_non_abs_literal_fragment instead of
600 xtensa_switch_to_literal_fragment.
601 (build_nop): New.
602 (assemble_nop): Use build_nop. Update call to xtensa_insnbuf_to_chars.
603 (get_expanded_loop_offset): Change check for undefined opcode to an
604 assertion.
605 (xtensa_set_frag_assembly_state, relaxable_section,
606 xtensa_find_unmarked_state_frags, xtensa_find_unaligned_branch_targets,
607 xtensa_find_unaligned_loops, xg_apply_tentative_value): New.
608 (md_begin): Update call to xtensa_isa_init. Initialize linkrelax to 1.
609 Set lit4_seg_name. Call xg_init_vinsn. Initialize new global opcodes.
610 Call init_op_placement_info_table and xtensa_set_frag_assembly_state.
611 (xtensa_init_fix_data): New.
612 (xtensa_frob_label): Reset label symbol to the current frag. Check
613 do_align_targets and generating_literals flag. Propagate frequency
614 info to new alignment frag. Call xtensa_set_frag_assembly_state.
615 (xtensa_unrecognized_line): New.
616 (xtensa_flush_pending_output): Check if inside a bundle. Add a call
617 to xtensa_set_frag_assembly_state.
618 (error_reset_cur_vinsn): New.
619 (md_assemble): Remove check for literal frag. Remove call to
620 istack_init. Call use_transform instead of use_generics. Parse
621 explicit instruction format specifiers. Move code for
622 a0_b_retw_interlock workaround to xg_assemble_vliw_tokens. Call
623 error_reset_cur_vinsn on errors. Add call to get_invisible_operands.
624 Add dwarf2_where call. Remote automatic alignment for ENTRY
625 instructions. Move call to xtensa_clear_insn_labels to the end.
626 Rearrange to handle bundles.
627 (xtensa_cons_fix_new): Delete.
628 (xtensa_handle_align): New.
629 (xtensa_frag_init): Call xtensa_set_frag_assembly_state. Remove
630 assignment to is_no_density field.
631 (md_pcrel_from): Use new xtensa-isa.h functions. Use decode_reloc
632 instead of reloc_to_opnum. Handle "alternate" relocations.
633 (xtensa_force_relocation, xtensa_check_inside_bundle,
634 xtensa_elf_section_change_hook): New.
635 (xtensa_symbol_new_hook): Delete.
636 (xtensa_fix_adjustable): Check for difference of symbols with an
637 offset. Check for external and weak symbols.
638 (md_apply_fix3): Remove cases for XTENSA_OP{0,1,2} relocs.
639 (md_estimate_size_before_relax): Return expansion for the first slot.
640 (tc_gen_reloc): Handle difference of symbols by producing
641 XTENSA_DIFF{8,16,32} relocs and by writing the value of the difference
642 into the output. Handle new XTENSA_SLOT*_OP relocs by storing the
643 tentative values into the output when linkrelax is set.
644 (XTENSA_PROP_SEC_NAME): Define.
645 (xtensa_post_relax_hook): Call xtensa_find_unmarked_state_frags.
646 Create literal tables only if using literal sections. Create new
647 property tables instead of old instruction tables. Check for unaligned
648 branch targets and loops.
649 (finish_vinsn, find_vinsn_conflicts, check_t1_t2_reads_and_writes,
650 new_resource_table, clear_resource_table, resize_resource_table,
651 resources_available, reserve_resources, release_resources,
652 opcode_funcUnit_use_unit, opcode_funcUnit_use_stage,
653 resources_conflict, xg_find_narrowest_format, relaxation_requirements,
654 bundle_single_op, emit_single_op, xg_assemble_vliw_tokens): New.
655 (xtensa_end): Call xtensa_flush_pending_output. Set past_xtensa_end
656 flag. Update checks for workaround options. Call
657 xtensa_mark_narrow_branches and xtensa_mark_zcl_first_insns.
658 (xtensa_cleanup_align_frags): Add special case for branch targets.
659 Check for and mark unreachable frags.
660 (xtensa_fix_target_frags): Remove use of align_only_targets flag.
661 Use RELAX_LOOP_END_BYTES in special case for negatable branch at the
662 end of a zero-overhead loop body.
663 (frag_can_negate_branch): Handle instructions with multiple slots.
664 Use new xtensa-isa.h functions
665 (xtensa_mark_narrow_branches, is_narrow_branch_guaranteed_in_range,
666 xtensa_mark_zcl_first_insns): New.
667 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags): Error if
668 transformations are disabled.
669 (next_instrs_are_b_retw): Use new xtensa-isa.h functions. Handle
670 multislot instructions.
671 (xtensa_fix_close_loop_end_frags, xtensa_fix_short_loop_frags):
672 Likewise. Also error if transformations are disabled.
673 (unrelaxed_frag_max_size): New.
674 (unrelaxed_frag_min_insn_count, unrelax_frag_has_b_j): Use new
675 xtensa-isa.h functions.
676 (xtensa_sanity_check, is_empty_loop, is_local_forward_loop): Use
677 xtensa_opcode_is_loop instead of is_loop_opcode.
678 (get_text_align_power): Replace as_fatal with assertion.
679 (get_text_align_fill_size): Iterate instead of using modulus when
680 use_nops is false.
681 (get_noop_aligned_address): Assert that this is for a machine-dependent
682 RELAX_ALIGN_NEXT_OPCODE frag. Use next_frag_opcode_is_loop,
683 xg_get_single_size, and frag_format_size.
684 (get_widen_aligned_address): Rename to ...
685 (get_aligned_diff): ... this function. Add max_diff parameter.
686 Remove handling of rs_align/rs_align_code frags. Use
687 next_frag_format_size, get_text_align_power, get_text_align_fill_size,
688 next_frag_opcode_is_loop, and xg_get_single_size. Compute max_diff
689 and pass it back to caller.
690 (xtensa_relax_frag): Use relax_frag_loop_align. Add code for new
691 RELAX_SLOTS, RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN,
692 RELAX_FILL_NOP, and RELAX_UNREACHABLE frag types. Check relax_seen.
693 (relax_frag_text_align): Rename to ...
694 (relax_frag_loop_align): ... this function. Assume loops can only be
695 in the first slot of an instruction.
696 (relax_frag_add_nop): Use assemble_nop instead of constructing an OR
697 instruction. Remove call to frag_wane.
698 (relax_frag_narrow): Rename to ...
699 (relax_frag_for_align): ... this function. Extend to handle
700 RELAX_FILL_NOP and RELAX_UNREACHABLE, as well as RELAX_SLOTS with
701 RELAX_NARROW for the first slot.
702 (find_address_of_next_align_frag, bytes_to_stretch): New.
703 (future_alignment_required): Use find_address_of_next_align_frag and
704 bytes_to_stretch. Look ahead to subsequent frags to make smarter
705 alignment decisions.
706 (relax_frag_immed): Add format, slot, and estimate_only parameters.
707 Check if transformations are enabled for b_j_loop_end workaround.
708 Use new xtensa-isa.h functions and handle multislot instructions.
709 Update call to xg_assembly_relax.
710 (md_convert_frag): Handle new RELAX_SLOTS, RELAX_UNREACHABLE,
711 RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN, and RELAX_FILL_NOP
712 frag types.
713 (convert_frag_narrow): Add segP, format and slot parameters. Call
714 convert_frag_immed for branch instructions. Adjust calls to
715 tinsn_from_chars, tinsn_immed_from_frag, and xg_emit_insn_to_buf. Use
716 xg_get_single_size and xg_get_single_format.
717 (convert_frag_fill_nop): New.
718 (convert_frag_immed): Add format and slot parameters. Handle multislot
719 instructions and use new xtensa-isa.h functions. Update calls to
720 tinsn_immed_from_frag and xg_assembly_relax. Check if transformations
721 enabled for b_j_loop_end workaround. Use build_nop instead of
722 assemble_nop. Check is_specific_opcode flag. Check for unreachable
723 frags. Use xg_get_single_size. Handle O_pltrel.
724 (fix_new_exp_in_seg): Remove check for old plt flag.
725 (convert_frag_immed_finish_loop): Update calls to tinsn_from_chars and
726 xtensa_insnbuf_to_chars. Call tinsn_immed_from_frag. Change check
727 for loop opcode to an assertion. Mark all frags up to the end of the
728 loop as not transformable.
729 (get_last_insn_flags, set_last_insn_flags): Use get_subseg_info.
730 (get_subseg_info): New.
731 (xtensa_move_literals): Call xtensa_set_frag_assembly_state. Add null
732 check for dest_seg.
733 (xtensa_switch_to_literal_fragment): Rewrite to handle absolute
734 literals and use xtensa_switch_to_non_abs_literal_fragment otherwise.
735 (xtensa_switch_to_non_abs_literal_fragment): New.
736 (cache_literal_section): Add is_code parameter and pass it through to
737 retrieve_literal_seg.
738 (retrieve_literal_seg): Add is_code parameter and use it to set the
739 flags on the literal section. Handle case where head parameter is 0.
740 (get_frag_is_no_transform, set_frag_is_specific_opcode,
741 set_frag_is_no_transform): New.
742 (xtensa_create_property_segments): Add end_property_function parameter
743 and pass it through to add_xt_block_frags. Call bfd_get_section_flags
744 and skip SEC_DEBUGGING and !SEC_ALLOC sections.
745 (xtensa_create_xproperty_segments, section_has_xproperty): New.
746 (add_xt_block_frags): Add end_property_function parameter and call it
747 if it is non-zero. Call xtensa_frag_flags_init.
748 (xtensa_frag_flags_is_empty, xtensa_frag_flags_init,
749 get_frag_property_flags, frag_flags_to_number,
750 xtensa_frag_flags_combinable, xt_block_aligned_size,
751 xtensa_xt_block_combine, add_xt_prop_frags,
752 init_op_placement_info_table, opcode_fits_format_slot,
753 xg_get_single_size, xg_get_single_format): New.
754 (istack_push): Inline call to tinsn_copy.
755 (tinsn_copy): Delete.
756 (tinsn_has_invalid_symbolic_operands): Handle O_hi16 and O_lo16 and
757 CONST16 opcodes. Handle O_big, O_illegal, and O_absent.
758 (tinsn_has_complex_operands): Handle O_hi16 and O_lo16.
759 (tinsn_to_insnbuf): Use xg_get_single_format and new xtensa-isa.h
760 functions. Handle invisible operands.
761 (tinsn_to_slotbuf): New.
762 (tinsn_check_arguments): Use new xtensa-isa.h functions.
763 (tinsn_from_chars): Add slot parameter. Rewrite using xg_init_vinsn,
764 vinsn_from_chars, and xg_free_vinsn.
765 (tinsn_from_insnbuf): New.
766 (tinsn_immed_from_frag): Add slot parameter and handle multislot
767 instructions. Handle symbol differences.
768 (get_num_stack_text_bytes): Use xg_get_single_size.
769 (xg_init_vinsn, xg_clear_vinsn, vinsn_has_specific_opcodes,
770 xg_free_vinsn, vinsn_to_insnbuf, vinsn_from_chars, expr_is_register,
771 get_expr_register, set_expr_symbol_offset_diff): New.
772 * config/tc-xtensa.h (MAX_SLOTS): Define.
773 (xtensa_relax_statesE): Move from tc-xtensa.c. Add
774 RELAX_CHECK_ALIGN_NEXT_OPCODE, RELAX_MAYBE_DESIRE_ALIGN, RELAX_SLOTS,
775 RELAX_FILL_NOP, RELAX_UNREACHABLE, RELAX_MAYBE_UNREACHABLE, and
776 RELAX_NONE types.
777 (RELAX_IMMED_MAXSTEPS): Move from tc-xtensa.c.
778 (xtensa_frag_type struct): Add is_assembly_state_set,
779 use_absolute_literals, relax_seen, is_unreachable, is_specific_opcode,
780 is_align, is_text_align, alignment, and is_first_loop_insn fields.
781 Replace is_generics and is_relax fields by is_no_transform field.
782 Delete is_text and is_longcalls fields. Change text_expansion and
783 literal_expansion to arrays of MAX_SLOTS entries. Add arrays of
784 per-slot information: literal_frags, slot_subtypes, slot_symbols,
785 slot_sub_symbols, and slot_offsets. Add fr_prev field.
786 (xtensa_fix_data struct): New.
787 (xtensa_symfield_type struct): Delete plt field.
788 (xtensa_block_info struct): Move definition to tc-xtensa.h. Add
789 forward declaration here.
790 (xt_section_type enum): Delete xt_insn_sec. Add xt_prop_sec.
791 (XTENSA_SECTION_RENAME): Undefine.
792 (TC_FIX_TYPE, TC_INIT_FIX_DATA, TC_FORCE_RELOCATION, NO_PSEUDO_DOT,
793 tc_unrecognized_line, md_do_align, md_elf_section_change_hook,
794 HANDLE_ALIGN, TC_LINKRELAX_FIXUP, SUB_SEGMENT_ALIGN): Define.
795 (TC_CONS_FIX_NEW, tc_symbol_new_hook): Delete.
796 (unit_num_copies_func, opcode_num_units_func,
797 opcode_funcUnit_use_unit_func, opcode_funcUnit_use_stage_func): New.
798 (resource_table struct): New.
799 * config/xtensa-istack.h (MAX_INSN_ARGS): Increase from 6 to 10.
800 (TInsn struct): Add keep_wide, loc, fixup, record_fix, subtype,
801 literal_space, symbol, sub_symbol, offset, and literal_frag fields.
802 (tinsn_copy): Delete prototype.
803 (vliw_insn struct): New.
804 * config/xtensa-relax.c (insn_pattern_struct): Add options field.
805 (widen_spec_list): Add option conditions for density and boolean
806 instructions. Add expansions using CONST16 and conditions for using
807 CONST16 vs. L32R. Use new Xtensa LX RSR/WSR syntax. Add entries for
808 predicted branches.
809 (simplify_spec_list): Add option conditions for density instructions.
810 Add entry for NOP instruction.
811 (append_transition): Add cmp function pointer parameter and use it to
812 insert the new entry in order.
813 (operand_function_LOW16U, operand_function_HI16U): New.
814 (xg_has_userdef_op_fn, xg_apply_userdef_op_fn): Handle
815 OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
816 (enter_opname, split_string): Use xstrdup instead of strdup.
817 (init_insn_pattern): Initialize new options field.
818 (clear_req_or_option_list, clear_req_option_list,
819 clone_req_or_option_list, clone_req_option_list, parse_option_cond):
820 New.
821 (parse_insn_pattern): Parse option conditions.
822 (transition_applies): New.
823 (build_transition): Use new xtensa-isa.h functions. Fix incorrectly
824 swapped last arguments in calls to append_constant_value_condition.
825 Call clone_req_option_list. Add warning about invalid opcode.
826 Handle LOW16U and HI16U function names.
827 (build_transition_table): Add cmp parameter and use it in calls to
828 append_transition. Use new xtensa-isa.h functions. Check
829 transition_applies before adding entries.
830 (xg_build_widen_table, xg_build_simplify_table): Add cmp parameter and
831 pass it through to build_transition_table.
832 * config/xtensa-relax.h (ReqOrOptionList, ReqOrOption, ReqOptionList,
833 ReqOption, transition_cmp_fn): New types.
834 (OpType enum): Add OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
835 (transition_rule struct): Add options field.
836 * doc/as.texinfo (Overview): Update Xtensa options.
837 * doc/c-xtensa.texi (Xtensa Options): Delete --[no-]density,
838 --[no-]relax, and --[no-]generics options. Update descriptions of
839 --text-section-literals and --[no-]longcalls. Add
840 --[no-]absolute-literals and --[no-]transform.
841 (Xtensa Syntax): Add description of syntax for FLIX instructions.
842 Remove use of "generic" and "specific" terminology for opcodes.
843 (Xtensa Registers): Generalize the syntax description to include
844 user-defined register files.
845 (Xtensa Automatic Alignment): Update.
846 (Xtensa Branch Relaxation): Mention limitation of unconditional jumps.
847 (Xtensa Call Relaxation): Linker can now remove most of the overhead.
848 (Xtensa Directives): Remove confusing rules about precedence.
849 (Density Directive, Relax Directive): Delete.
850 (Schedule Directive): New.
851 (Generics Directive): Rename to ...
852 (Transform Directive): ... this node.
853 (Literal Directive): Update for absolute literals. Missing
854 literal_position directive is now an error.
855 (Literal Position Directive): Update for absolute literals.
856 (Freeregs Directive): Delete.
857 (Absolute Literals Directive): New.
858 (Frame Directive): Minor editing.
859 * Makefile.am (DEPTC_xtensa_elf, DEPOBJ_xtensa_elf, DEP_xtensa_elf):
860 Update dependencies.
861 * Makefile.in: Regenerate.
862
6e1304d8
RS		 8632004-10-07 Richard Sandiford <rsandifo@redhat.com>
864
865 * config/tc-mips.c (append_insn): Use fix_new rather than fix_new_exp
866 to build the second and third fixups for a composite relocation.
867 (macro_read_relocs): New function.
868 (macro_build): Use it.
869 (s_cpsetup): Pass all three composite relocation codes to macro_build.
870 Simplify fragging code accordingly.
871 (s_gpdword): Use fix_new rather than fix_new_exp for the second part
872 of the composite relocation. Set fx_tcbit in both fixups.
873
b1dca8ee
RS		 8742004-10-07 Richard Sandiford <rsandifo@redhat.com>
875
876 * config/tc-mips.c (append_insn): Set fx_tcbit for composite relocs.
877 (md_apply_fix3): Don't treat composite relocs as done.
878
89658e52
NC		 8792004-10-07 Jan Beulich <jbeulich@novell.com>
880
881 * macro.c (macro_expand_body): When ELF, use .LL rather than LL as
882 prefix for symbol names generated from the LOCAL macro directive.
883
289040ca
NC		 884 * dw2gencfi.c (select_cie_for_fde): When separating CIE out from
885 FDE, treat a DW_CFA_remember_state as we do a DW_CFA_advance_loc.
886
48c9f030
NC		 8872004-10-07 Tomer Levi <Tomer.Levi@nsc.com>
888
889 * config/tc-crx.c (preprocess_reglist): Handle Co-processor
c04f5787 890 Special registers.
5c5054ab 891 (md_assemble): Add error checking for Co-Processor instructions.
48c9f030
NC		 892 (get_cinv_parameters): Add 'b' option to invalidate the
893 branch-target cache.
894
7ed4c4c5
NC		 8952004-10-05 Paul Brook <paul@codesourcery.com>
896
897 * config/tc-arm.c (unwind): New variable.
898 (vfp_sp_encode_reg): New function.
899 (vfp_sp_reg_required_here): Use it.
900 (vfp_sp_reg_list, vfp_dp_reg_list): Remove.
901 (vfp_parse_reg_list): New function.
902 (s_arm_unwind_fnstart, s_arm_unwind_fnend, s_arm_unwind_cantunwind,
903 s_arm_unwind_personality, s_arm_unwind_personalityindex,
904 s_arm_unwind_handlerdata, s_arm_unwind_save, s_arm_unwind_movsp,
905 s_arm_unwind_pad, s_arm_unwind_setfp, s_arm_unwind_raw): New
906 functions.
907 (md_pseudo_table): Add them.
908 (do_vfp_reg2_from_sp2): Use vfp_parse_reg_list and vfp_sp_encode_reg.
909 (do_vfp_sp2_from_reg2, vfp_sp_ldstm, vfp_dp_ldstm): Ditto.
910 (set_section, add_unwind_adjustsp, flush_pending_unwind,
911 finish_unwind_opcodes, start_unwind_section, create_unwind_entry,
912 require_hashconst, add_unwind_opcode): New functions.
913 * doc/tc-arm.text: Document unwinding opcodes.
914 * NEWS: Mention the new feature.
915
04ca8703
EC		 9162004-10-04 Eric Christopher <echristo@redhat.com>
917
918 * config/tc-mips.c (md_apply_fix3): Remove erroneous assert.
919
7c1d0959
L		 9202004-10-01 H.J. Lu <hongjiu.lu@intel.com>
921
922 * config/tc-ppc.c (md_apply_fix3): Call S_SET_THREAD_LOCAL for
923 TLS relocations.
924 * config/tc-s390.c (md_apply_fix3): Likewise.
925 * config/tc-sparc.c (md_apply_fix3): Likewise.
926
40a18ebd
NC		 9272004-10-01 Paul Brook <paul@codesourcery.com>
928
929 * config/tc-arm.c (arm_elf_section_type): New function.
930 (arm_elf_change_section): Set section link for exidx sections.
931 * config/tc-arm.h (arm_elf_section_type): Add prototype.
932 (md_elf_section_type): Define.
933
14127cc4
NC		 9342004-10-01 Bill Farmer <Bill@the-farmers.freeserve.co.uk>
935
936 * config/tc-pdp11.c (md_apply_fix3): Change to sign of the SOB
937 instruction's offset.
938
b7c92712
NC		 9392004-10-01 Adam Nemet <anemet@lnxw.com>
940
941 * (TARGET_FORMAT): Remove LynxOS COFF definition.
942
65555e64
AM		 9432004-10-01 Ravi Ramaseshan <ravi.ramaseshan@codito.com>
944
945 * config/tc-arc.c (tc_gen_reloc): Don't assume fixP->fx_addsy is an
946 asymbol *, instead use symbol_get_bfdsym.
947
9fe0b840
NC		 9482004-09-30 Linus Nielsen Feltzing <linus@haxx.se>
949
950 * config/tc-m68k.c (select_control_regs): Add mcf5249.
951
0dd132b6
NC		 9522004-09-30 Paul Brook <paul@codesourcery.com>
953
954 * config/tc-arm.c (do_smi, do_nop): New functions.
955 (insns): Add ARMv6ZK instructions.
956 (md_apply_fix3): Handle BFD_RELOC_ARM_SMI.
957 (tc_gen_reloc): Ditto.
958 (arm_cpus): Add mpcore and arm1176.
959 (arm_archs): Add armv6{k,z,zk}.
960 * doc/c-arm.texi: Document new cores and architectures.
961
a737bd4d
NC		 9622004-09-30 Nick Clifton <nickc@redhat.com>
963
964 * config/tc-arm.c: Use ISO C90 formatting.
965
3631a3c8
NC		 9662004-09-30 Vladimir Ivanov <vladitx@nucleusys.com>
967
a737bd4d 968 * config/tc-arm.c (mav_reg_required_here): Allow REG_TYPE_CN
3631a3c8
NC		 969 as alternative when REG_TYPE_MVF, REG_TYPE_MVD, REG_TYPE_MVFX or
970 REG_TYPE_MVDX is expected.
971
d5f0cf92
AM		 9722004-09-29 Marc Bevand <m.bevand@gmail.com>
973
974 * doc/c-i386.texi (i386-Mnemonics): Fix typo.
975
9f9a069e
JW		 9762004-09-21 James E Wilson <wilson@specifixinc.com>
977
978 * config/tc-ia64.c (ENCODED_PSP_OFFSET): New.
979 (output_rp_psprel, output_pfs_psprel, output_preds_psprel,
980 output_spill_base, output_unat_psprel, output_lc_psprel,
981 output_fpsr_psprel, output_priunat_psprel, output_bsp_psprel,
982 output_bsprestore_psprel, output_rnat_psprel, output_spill_psprel,
983 output_spill_psprel_p): Use it.
984
4b778c88
TL		 9852004-09-20 Tomer Levi <Tomer.Levi@nsc.com>
986
c04f5787 987 * config/tc-crx.c (handle_LoadStor): New function.
04ca8703 988 Handle load/stor unique instructions before parsing.
4b778c88 989
eb043451
PB		 9902004-09-17 Paul Brook <paul@codesourcery.com>
991
992 * config/tc-arm.c (s_arm_rel31): New funciton.
993 (md_pseudo_table): Add .rel31.
04ca8703 994 (md_apply_fix3): Handle BFD_RELOC_ARM_TARGET2,
eb043451
PB		 995 BFD_RELOC_32_PCREL and BFD_RELOC_ARM_PREL31.
996 (tc_gen_reloc): Handle BFD_RELOC_ARM_PREL31 and BFD_RELOC_ARM_TARGET2.
997 (arm_fix_adjustable): Return 0 for BFD_RELOC_ARM_TARGET2.
998 (arm_parse_reloc): Add (target2).
999
8ea2b00f
AM		 10002004-09-17 Alan Modra <amodra@bigpond.net.au>
1001
1002 * Makefile.am: Run "make dep-am".
1003 * Makefile.in: Regenerate.
1004 * aclocal.m4: Regenerate.
1005 * configure: Regenerate.
1006 * doc/Makefile.in: Regenerate.
1007 * po/POTFILES.in: Regenerate.
1008 * po/gas.pot: Regenerate.
1009
7e1792e0
HPN		 10102004-09-14 Hideki IWAMOTO <h-iwamoto@kit.hi-ho.ne.jp>
1011
1012 * config/tc-mmix.c [!LLONG_MIN]: Correct #elsif to #elif.
1013 [!LLONG_MAX]: Ditto.
1014
9c504268
PB		 10152004-09-13 Paul Brook <paul@codesourcery.com>
1016
1017 * config/tc-arm.c: Rename RELABS to TARGET1.
1018
2159ac21
AM		 10192004-09-13 Alan Modra <amodra@bigpond.net.au>
1020
1021 * messages.c (as_internal_value_out_of_range): Cast values passed
1022 to as_bad_where or as_warn_where to proper type.
1023
23794b24
MM		 10242004-09-11 Theodore A. Roth <troth@openavr.org>
1025
1026 * config/tc-avr.c: Add support for
1027 atmega48, atmega88, atmega168, attiny13, attiny2313, at90can128.
1028
e9fad691
AM		 10292004-09-09 Alan Modra <amodra@bigpond.net.au>
1030
1031 * dw2gencfi.c (select_cie_for_fde): When separating CIE out
1032 from FDE, treat a CFI_escape as we do a DW_CFA_advance_loc.
1033
10b016c2
PB		 10342004-09-08 Paul Brook <paul@codesourcery.com>
1035
1036 * config/obj-elf.c (obj_elf_section_type): Handle init_array,
1037 fini_array and preinit_array section types.
1038 * config/tc-ia64.c (ia64_elf_section_type): Remove init_array
1039 and fini_array.
1040 * doc/as.texinfo: Document extra section types.
1041
e5a52504
MM		 10422004-09-02 Mark Mitchell <mark@codesourcery.com>
1043
1044 * Makefile.am (TARG_ENV_HFILES): Add te-symbian.h.
1045 * Makefile.in: Regenerated.
1046 * configure.in: Set em for arm*-*-symbianelf*.
1047 * configure: Regenerated.
1048 * config/tc-arm.c (elf32_arm_target_format): Use Symbian target
1049 vectors when appropriate.
1050 * config/te-symbian.h: New file.
1051
42851540
NC		 10522004-09-03 Tomer Levi <Tomer.Levi@nsc.com>
1053
1054 * config/tc-crx.c (gettrap): Exception vector can be case
1055 insensitive.
04ca8703 1056 (process_label_constant): Fix a 32-bit displacement bug in branch
42851540
NC		 1057 instructions.
1058 (get_operandtype) : Bug fix, wrong operand was used.
04ca8703 1059 (process_label_constant): Initialize relocation type to
42851540
NC		 1060 BFD_RELOC_NONE
1061
8783612f
RE		 10622004-09-01 Richard Earnshaw < reanrsha@arm.com>
1063
1064 * tc-arm.c (arm_cpus, arm_fpus): Allow <cpu>-s as well as <cpu>s
1065 for synthesizable cores.
1066
1ff4677c
RE		 1067 * doc/c-arm.texi (ARM Options): Document canonical names of CPUs.
1068
b18c562e
NC		 10692004-08-25 Dmitry Diky <diwil@spec.ru>
1070
1071 * config/tc-msp430.c: Clean-up the code.
1072 (md_relax_table): New relax table.
1073 (mcu_types): Sort MCU types.
1074 (md_pseudo_table): Add .profiler pseudo handler.
1075 (pow2value): New function.
1076 (msp430_profiler): New function.
1077 (msp430_operands): Add new insns handlers.
1078 (msp430_srcoperand): Add register operand handler, allow complex
1079 expressions.
1080 (md_estimate_size_before_relax): Rewritten.
1081 (md_convert_frag): Rewritten.
1082 (msp430_relax_frag): New function.
1083 * config/tc-msp430.h (md_relax_frag): define macro
1084 * doc/c-msp430.texi: Update information.
1085
8f94ae4d
NC		 10862004-08-24 Nick Clifton <nickc@redhat.com>
1087
1088 * as.c (std_shortopts): Allow -g to take an optional argument.
1089 (parse_args): Pass any switch starting with -g on to the backend
1090 for parsing.
1091
e51dace8
MM		 10922004-08-18 Mark Mitchell <mark@codesourcery.com>
1093
1094 * configure.in (arm*-*-symbianelf*): New target.
1095 (arm*-*-eabi*): Likewise.
1096 * configure: Regenerated.
1097
f7870c8d
TS		 10982004-08-18 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1099 * config/tc-mips.c (append_insn): Handle delay slots in branch likely
1100 correctly.
1101
d6afba4b
JJ		 11022004-08-18 Jakub Jelinek <jakub@redhat.com>
1103
1104 * config/tc-ia64.c (start_unwind_section): Add linkonce_empty
1105 argument, don't do anything if current section is not
1106 .gnu.linkonce.t.* and linkonce_empty is set.
1107 (generate_unwind_image, dot_endp): Adjust callers, call
1108 start_unwind_section (*, 1) if nothing will be put into the
1109 section.
1110
329e276d
NC		 11112004-08-17 Nick Clifton <nickc@redhat.com>
1112
1113 * as.c (MD_DEBUG_FORMAT_SELECTOR): Provide default definition.
1114 (show_usage): Add -g.
1115 (std_longopts): Add --gen-debug. Alpha sort the table.
1116 (parse_args): Print an error message if a switch is not handled.
1117 Handle the -g switch, calling md_debug_format_selector() if
1118 necessary.
1119 * NEWS: Mention new feature.
1120 * doc/as.texinfo: Document new switch.
1121 * doc/internals.texi: Document behaviour of md_parse_option.
04ca8703 1122
329e276d
NC		 1123 * config/tc-arm.c (md_parse_option): Do not issue an error message
1124 if the switch is not recognised.
1125 * config/tc-m68k.c (md_parse_option): Likewise.
1126 * config/tc-pdp11.c (md_parse_option): Likewise.
1127 * config/tc-v850.c (md_parse_option): Likewise.
1128
1129 * as.h: Fix up formatting.
1130 * tc.h: Likewise.
1131
4101eec8
NC		 11322004-08-16 Nick Clifton <nickc@redhat.com>
1133
2766e5e4
NC		 1134 * macro.c (macro_set_alternate): Use ISO C90 formatting.
1135
4101eec8
NC		 1136 * configure.in: Sort architecture based tables alphabetically.
1137 * configure: Regenerate.
1138
80f846b6
AM		 11392004-08-16 Alan Modra <amodra@bigpond.net.au>
1140
1141 * config/tc-ppc.c (tc_ppc_regname_to_dw2regnum <regnames>): Replace
1142 { "cc", 68 }, with { "cr", 70 }.
1143
caa32fe5
NC		 11442004-08-13 Jan Beulich <jbeulich@novell.com>
1145 Nick Clifton <nickc@redhat.com>
1146
1147 * as.c: Add and handle new --alternate command line option.
1148 * macro.c (macro_set_alternate): New.
1149 * macro.h (macro_set_alternate): Declare.
1150 * read.c: Add and handle new .altmacro and .noaltmacro directives.
1151 * doc/as.texinfo: Document new command line option and pseudo-ops
1152 as well as insert documentation originating from gasp about
1153 alternate macro syntax.
1154 * NEWS: Mention new command line option and pseudo-ops.
1155
b2221023
MM		 11562004-08-10 Mark Mitchell <mark@codesourcery.com>
1157
1158 * expr.c (operand): Handle the "~", "-", and "!" operators applied
1159 to bignums.
1160
db6579d4
PB		 11612004-08-06 Paul Brook <paul@codesourcery.com>
1162
1163 * config/tc-arm.c (md_apply_fix3, tc_gen_reloc, arm_parse_reloc):
1164 Handle new relocations.
1165 * include/elf/arm.h (elf_arm_reloc_type): Add new EABI relocations.
1166
e750405d
BW		 11672004-08-05 Bob Wilson <bob.wilson@acm.org>
1168
1169 * write.c (relax_segment): Use was_address instead of address when
1170 setting fr_fix field for align frag due to backwards .org.
1171
1d70c7fb
AO		 11722004-07-29 Alexandre Oliva <aoliva@redhat.com>
1173
1174 Introduce SH2a support.
1175 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
1176 * config/tc-sh.c (get_specific): Change arch_sh2a_up to
1177 arch_sh2a_nofpu_up.
1178 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
1179 * config/tc-sh.c (md_parse_option): Add sh2a-nofpu ISA handling.
1180 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
1181 * config/tc-sh.c (sh_elf_final_processing): Move sh2a recognition
1182 to end of conditional expression.
1183 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
1184 * config/tc-sh.c: Add sh2a-nofpu support.
1185 2003-12-29 DJ Delorie <dj@redhat.com>
1186 * tc-sh.c: Add sh2a support.
1187 (parse_reg): Add tbr.
1188 (parse_at): Support @@(disp,tbr).
1189 (get_specific): Support sh2a opcodes.
1190 (insert4): New, for 4 byte relocs.
1191 (build_Mytes): Support sh2a opcodes.
1192 (md_apply_fix3_Mytes): Support sh2a opcodes.
1193 2003-12-02 Michael Snyder <msnyder@redhat.com>
1194 * config/tc-sh.c (md_parse_option): Handle sh2a.
1195 (sh_elf_final_processing): Ditto.
1196
e7fbc2bf
JT		 11972004-07-27 Jason Thorpe <thorpej@wasabisystems.com>
1198
1199 * config/tc-hppa.h (TARGET_FORMAT): Set to "elf32-hppa-netbsd"
1200 for TE_NetBSD.
1201
9232bbb0
AM		 12022004-07-27 Alan Modra <amodra@bigpond.net.au>
1203
1204 * config/tc-ppc.c (ppc_frob_file_before_adjust): Warn if .toc too big.
1205 (ppc_arch): Expand comment.
1206
670ec21d
NC		 12072004-07-27 Tomer Levi <Tomer.Levi@nsc.com>
1208
1209 * config/tc-crx.c: Support evaluating the difference between two
1210 symbols.
1211 * config/tc-crx.h: Likewise.
1212
406f8ff9 12132004-07-26 H.J. Lu <hongjiu.lu@intel.com>
38ce5b11
L		 1214
1215 * config/tc-ia64.c (start_unwind_section): Set the linked-to
1216 section.
1217 (ia64_elf_section_change_hook): Set the linked-to section for
1218 SHT_IA_64_UNWIND.
1219
c05e9f04
NC		 12202004-07-26 Dmitry Diky <diwil@spec.ru>
1221
1222 * config/tc-msp430.c: Add new subtargets: msp430x1610,
1223 msp430x1611, msp430x1612, msp430x415, msp430x417, msp430xG437,
1224 msp430xG438, msp430xG439.
1225
e9863d7f
DJ		 12262004-07-25 Daniel Jacobowitz <dan@debian.org>
1227
1228 * doc/as.texinfo (Section, PushSection): Correct documentation
1229 for ELF.
1230
543613e9 12312004-07-21 Jan Beulich <jbeulich@novell.com>
20f0a1fc
NC		 1232
1233 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
1234 permissible for the selected instruction suffix.
1235 (match_template): Don't permit 64-bit general purpose operands in
1236 32-bit mode.
1237 (finalize_imm): Permit 64-bit immediates.
1238 (build_modrm_byte): Don't treat 32-bit addressing in 64-bit mode
1239 specially except for the width of the used base and/or index
1240 registers. For 32-bit displacements, use sign-extended
1241 relocations only when using 64-bit addressing.
1242 Force zero displacement on rip-relative addressing when there is
1243 no other displacement.
1244 (i386_index_check): Don't treat 32-bit addressing in 64-bit mode
1245 specially except for the width of the used base and/or index
1246 registers.
1247 (parse_register): Disallow Reg64 registers in 32-bit mode.
1248
543613e9
NC		 1249 * config/tc-i386.c: For DefaultSize instructions, don't guess a 'q'
1250 suffix if the instruction doesn't support it.
04ca8703 1251
01a3f561
MR		 12522004-07-20 Maciej W. Rozycki <macro@linux-mips.org>
1253
1254 * config/tc-mips.c (append_insn): Handle constant expressions with
1255 no associated relocation.
1256 (mips_ip): Cancel the expression after use for the Q format
1257 specifier.
1258 (parse_relocation): Return no relocation for unsupported
1259 operators.
1260 (my_getSmallExpression): Return no relocation if no relocation
1261 operators are used.
1262
e12904d2
DA		 12632004-07-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1264
1265 * config/obj-som.c (adjust_stab_sections): Add prototype.
1266 (obj_som_compiler, obj_som_version, obj_som_copyright,
1267 adjust_stab_sections): Add ATTRIBUTE_UNUSED to unused arguments.
1268 * config/tc-hppa.c (update_subspace): Likewise.
1269 (is_defined_subspace): Amplify comment.
1270 * config/obj-som.h (som_frob_file): Add prototype.
1271
90c1602c
L		 12722004-07-19 Christopher Faylor <cgf@timesys.com>
1273 H.J. Lu <hongjiu.lu@intel.com>
1274
1275 * subsegs.c (section_symbol): Don't create a new segment when
1276 existing segment is undefined.
1277
f263249b
RE		 12782004-07-16 Richard Earnshaw <rearnsha@arm.com>
1279
1280 * config/tc-arm.c: Include include/opcode/arm.h.
1281 (ARM_EXT_*, ARM_ARCH_*, ARM_ANY, ARM_ALL, COPROC_ANY): Delete.
1282 (FPU_FPA_EXT_* FPU_VFP_EXT_*, FPU_ANY, FPU_NONE, FPU_MAVERICK): Delete.
1283 (FPU_ARCH_*): Delete.
1284 * Makefile.am: Update dependencies.
1285 * Makefile.in: Regenerate.
1286
152c65b9
NC		 12872004-07-15 Nick Clifton <nickc@redhat.com>
1288
1289 * configure.in: Accept armbe as a big-endian arm configuration.
1290 * configure: Regenerate.
1291
0477af35
NC		 12922004-07-13 Thomas Nystrom <thn@saeab.se>
1293
1294 * config/tc-i386.c (T_SHIFTOP): New constant.
1295 (intel_e05_1): Handle '&', '|' and T_SHIFTOP.
1296 (intel_el1): Handle '~'.
1297 (intel_get_token): Handle '<>', '&', '|' and '~'.
1298
12992004-07-13 Nick Clifton <nickc@redhat.com>
1300
1301 (md_assemble): Remove spurious newline from end of as_bad error
1302 message.
1303 (intel_e05_1): Likewise.
1304 (intel_e11): Likewise.
1305 (intel_match_token): Likewise.
1306
31302357
AS		 13072004-07-11 Andreas Schwab <schwab@suse.de>
1308
1309 * config/tc-m68k.c: Convert to C90. Remove redundant
1310 declarations. Indentation fixup.
1311 [M68KCOFF]: Include "obj-coff.h" instead of declaring
1312 obj_coff_section ourselves.
04ca8703 1313
549f748d
JW		 13142004-07-09 James E Wilson <wilson@specifixinc.com>
1315
1316 * config/tc-ia64.c (default_big_endian): New.
1317 (dot_byteorder, md_begin): Use it.
1318 (md_parse_option): Set it.
04ca8703 1319
1a320fbb
NC		 13202004-07-09 Nick Clifton <nickc@redhat.com>
1321
1322 * configure.in: Change sh-sybmian-elf to sh-*-symbianelf.
1323 * configure: Regenerate.
1324 * NEWS: Change sh-sybmian-elf to sh-*-symbianelf.
1325 * config/tc-sh.c (sh_elf_final_processing): Use renamed version of
1326 sh_find_elf_flags if necessary.
1327
de7e6852
RS		 13282004-07-08 Richard Sandiford <rsandifo@redhat.com>
1329
1330 * config/tc-mips.c (mips_fix_adjustable): If the full addend is
1331 going to be split into more than one in-place addend, return 0
1332 for relocations against mergeable sections. Associate comments
1333 with code.
1334
1fe1f39c
NC		 13352004-07-07 Tomer Levi <Tomer.Levi@nsc.com>
1336
1337 * Makefile.am (CPU_TYPES): Add crx.
1338 (TARGET_CPU_CFILES): Add config/tc-crx.c.
1339 (TARGET_CPU_HFILES): Add config/tc-crx.h.
1340 (DEPTC_crx_elf): New target.
1341 (DEPOBJ_crx_elf): Likewise.
1342 (DEP_crx_elf): Likewise.
1343 * Makefile.in: Regenerate.
1344 * configure.in: Add crx* target.
1345 * configure: Regenerate.
1346 * config/tc-crx.c: New file.
1347 * config/tc-crx.h: New file.
1348 * NEWS: Mention new target.
1349
d9895649
NC		 13502004-07-06 Nick Clifton <nickc@redhat.com>
1351
85fbca6a 1352 * config.in: Undefine TARGET_SYMBIAN by default.
c04f5787
AM		 1353 * configure.in:
1354 * configure: Regenerate. Add sh-symbian-elf target. If
1355 selected define TARGET_SYMBIAN.
1356 * config/tc-sh.h (TARGET_FORMAT): Select a Symbian target
1357 format if TARGET_SYMBIAN has been defined.
04ca8703 1358
e7bd9ea0
NC		 1359 * output-file.c (output_file_create): Report the target format
1360 chosen when bfd_openw reports that it is invalid.
1361
d9895649
NC		 1362 * config/obj-coff.c (coff_pseudo_table): Only define the weak
1363 pseudo for BFD based assemblers.
1364
871ec896
JR		 13652004-07-05 Andrew Stubbs <andrew.stubbs@superh.com>
1366
1367 gas:
1368 * config/tc-sh.c (md_assemble): Change isspace to ISSPACE.
1369 (md_parse_option): Remove redundant -isa testing.
1370 Make bfd_arch variable const.
1371 (md_show_usage): Make bfd_arch variable const.
1372
5a9ff93d
JW		 13732004-07-03 James E Wilson <wilson@specifixinc.com>
1374
1375 * config/tc-ia64.c (emit_one_bundle): Check and set insn_addr.
1376 * config/tc-ia64.h (md_frag_check): Define.
1377
c87db184
CF		 13782004-07-03 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
1379
1380 * config/obj-coff.c (obj_coff_weak): New .weak syntax for PE weak
1381 externals.
1382 * doc/as.texinfo (Weak): Document PE weak symbols.
1383
3b91255e
RS		 13842004-07-03 Richard Sandiford <rsandifo@redhat.com>
1385
1386 * config/tc-mips.c (HAVE_IN_PLACE_ADDENDS): New macro.
1387 (reloc_needs_lo_p): Only return true if HAVE_IN_PLACE_ADDENDS.
1388 (mips_frob_file): Rework so that only a single pass through the
1389 relocs is needed. Allow %lo()s to have higher offsets than their
1390 corresponding %hi()s or %got()s.
1391
310ea308
NC		 13922004-07-02 Nick Clifton <nickc@redhat.com>
1393
1394 * config/tc-arm.c (md_apply_fix3:BFD_RELOC_ARM_IMMEDIATE): Do not
1395 allow values which have come from undefined symbols.
1396 Always consider this fixup to have been processed as a reloc
1397 cannot be generated for it.
1398
09b935ac
AM		 13992004-07-02 Alan Modra <amodra@bigpond.net.au>
1400
1401 * frags.h (struct frag): Add has_code and insn_addr fields.
1402 * write.c (cvt_frag_to_fill): Invoke md_frag_check.
1403 * config/tc-ppc.c (md_assemble): Check and set insn_addr.
1404 * config/tc-ppc.h (md_frag_check): Define.
1405
4d1cd5c7
MR		 14062004-06-28 Maciej W. Rozycki <macro@linux-mips.org>
1407
1408 * doc/Makefile.am (info): Rename goal to...
1409 (info-local): ... this, to preserve implicit dependencies.
1410 * doc/Makefile.in: Regenerate with automake 1.8.5.
1411
097f809a
NC		 14122004-06-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1413
1414 * config/tc-m32r.c (md_convert_frag): Changed for @PLT.
1415 (m32r_cgen_record_fixup_exp): Changed for @GOTOFF, @GOT.
1416 (m32r_fix_adjustable): Changed for @GOTOFF, @GOT, @PLT.
1417 (tc_gen_reloc): Likewise.
1418 (m32r_end_of_match): Add for @GOTOFF, @GOT, @PLT.
1419 (m32r_parse_name): Likewise.
1420 (m32r_cgen_parse_fix_exp): Likewise.
1421 * config/tc-m32r.h (md_parse_name): Define for @GOTOFF, @GOT, @PLT.
1422 (O_PIC_reloc): Likewise.
1423 (TC_CGEN_PARSE_FIX_EXP): Likewise..
1424 * cgen.c (gas_cgen_parse_operand): Add TC_CGEN_PARSE_FIX_EXP
1425 for @GOTOFF, @GOT, @PLT.
1426
bb82af9f
NC		 14272004-06-21 Jan Beulich <jbeulich@novell.com>
1428
1429 * gas/symbols.c: While discarding ordinary local absolute symbols
1430 when --strip-local-absolute is in effect, retain file symbols.
1431
990ae22a
AS		 14322004-06-20 Andreas Schwab <schwab@suse.de>
1433
1434 * config/tc-m68k.c (mri_chip): Replace current_chip, not augment.
1435 (md_parse_option): Likewise.
1436
32137342
NC		 14372004-06-17 Jan Beulich <jbeulich@novell.com>
1438
1439 * config/tc-i386.c: Deal with LEX_QM the same way as with LEX_AT.
1440 * config/te-netware.h: New file.
1441 * config/te-ppcnw.h: Delete: Obsolete.
1442 * configure.in: Eliminate ill NetWare targets. Make generic
1443 NetWare target use proper emulation.
1444 * Makefile.am: Eliminate reference to obsolete te-ppcnw.h, add
1445 reference to new te-netware.h.
1446 * configure: Regenerate.
1447 * Makefile.in: Regenerate.
1448
f84d8e26
MS		 14492004-06-15 Martin Schwidefsky <schwidefsky@de.ibm.com>
1450
1451 * config/tc-s390.c (s390_insn): Avoid incorrect signed/unsigned
1452 comparison in .insn pseudo operation.
1453
587aac4e
AM		 14542004-06-15 Alan Modra <amodra@bigpond.net.au>
1455
1456 * config/obj-coff.c (coff_adjust_section_syms): Use
1457 bfd_get_section_size instead of bfd_get_section_size_before_reloc.
1458 (coff_frob_section): Likewise.
1459 * config/tc-mips.c (md_apply_fix3): Likewise.
1460 * config/obj-elf.c (elf_frob_file): Use bfd_set_section_size.
1461 (elf_frob_file_after_relocs): Likewise.
1462
2337eb33
DA		 14632004-06-10 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1464
b8fc22bc
DA		 1465 * config/tc-hppa.c (log2): Rename to exact_log2.
1466 (pa_next_subseg): Delete unused function.
1467 (create_new_space): Mark unused arguments with ATTRIBUTE_UNUSED.
1468 (create_new_subspace): Likewise.
1469
2337eb33
DA		 1470 Bug gas/213
1471 * config/tc-hppa.c (hppa_fix_adjustable): Allow reduction of fake
1472 labels. Fix warning.
1473
53cb0362
DD		 14742004-05-28 DJ Delorie <dj@redhat.com>
1475
1476 * config/tc-mn10300.h (tc_fix_adjustable): Define.
1477 * config/tc-mn10300.c (mn10300_fix_adjustable): Don't adjust debug
1478 or non-merged symbols.
1479
888a75be
L		 14802004-05-28 H.J. Lu <hongjiu.lu@intel.com>
1481
1482 * config/tc-ia64.c (remove_marked_resource): Save, clear and
1483 restore the old slot when inserting srlz.i/srlz.d.
1484
f6f9408f
JR		 14852004-05-28 Andrew Stubbs <andrew.stubbs@superh.com>
1486
1487 * Makefile.am: Regenerate dependecies.
1488 * Makefile.in: Regenerate.
1489 * config/tc-sh.c (valid_arch): Make unsigned.
1490 (preset_target_arch): Likewise.
1491 (md_begin): Use new architecture flags system.
1492 (get_specific): Likewise.
1493 (assemble_ppi): Likewise.
1494 (md_assemble): Likewise. Also fix error check for bad opcodes.
1495 (md_parse_option): Likewise. Also generate -isa values according
1496 to the table in bfd/cpu-sh.c instead of just constants. Also
1497 allow <arch>-up ISA variants.
1498 (sh_elf_final_processing): Replace if-else chain with a call to
1499 sh_find_elf_flags().
1500 * testsuite/gas/sh/arch: New directory.
1501 * testsuite/gas/sh/arch/arch.exp: New test script.
1502 * testsuite/gas/sh/arch/arch_expected.txt: New file.
1503 * testsuite/gas/sh/arch/sh.s: New file.
1504 * testsuite/gas/sh/arch/sh2.s: New file.
1505 * testsuite/gas/sh/arch/sh-dsp.s: New file.
1506 * testsuite/gas/sh/arch/sh2e.s: New file.
1507 * testsuite/gas/sh/arch/sh3-nommu.s: New file.
1508 * testsuite/gas/sh/arch/sh3.s: New file.
1509 * testsuite/gas/sh/arch/sh3-dsp.s: New file.
1510 * testsuite/gas/sh/arch/sh3e.s: New file.
1511 * testsuite/gas/sh/arch/sh4-nommu-nofpu.s: New file.
1512 * testsuite/gas/sh/arch/sh4-nofpu.s: New file.
1513 * testsuite/gas/sh/arch/sh4.s: New file.
1514 * testsuite/gas/sh/arch/sh4a-nofpu.s: New file.
1515 * testsuite/gas/sh/arch/sh4al-dsp.s: New file.
1516 * testsuite/gas/sh/arch/sh4a.s: New file.
1517
580b9172
NC		 15182004-05-28 Peter Barada <peter@the-baradas.com>
1519
1520 * config/gc-m68k.c(m68k_ip): Convert mode 5 addressing
1521 with zero offset into mode 2 addressing to save a word.
1522
652ca075
L		 15232004-05-27 H.J. Lu <hongjiu.lu@intel.com>
1524
1525 * config/tc-ia64.c (ar_is_in_integer_unit): Removed.
1526 (ar_is_only_in_integer_unit): New.
1527 (ar_is_only_in_memory_unit): New.
1528 (generate_unwind_image): Silence gcc on 32bit host.
1529 (md_assemble): Use ar_is_only_in_integer_unit instead of
1530 ar_is_in_integer_unit. Check AR access.
1531
60fed2d3
NC		 15322004-05-27 Peter Barada <peter@the-baradas.com>
1533
1534 * config/tc-m68k.c (md_begin): Sort the opcode table into
1535 alphabetical order.
1536 (m68k_compare_opcode): New function to do the sorting.
1537
be8c092b
NC		 15382004-05-24 Peter Barada <peter@the-baradas.com>
1539
1540 * config/m68k-parse.y(operand): Allow for MAC/EMAC mask
1541 addressing on MIT style operands.
1542 * config/m68k-parse.y(yylex): Allow '-&' for predecrement
1543 w/mask addressing.
1544 * config/tc-m68k.c(install_operand): Comment 'G' and 'H' type
1545 operands.
1546
1181c20e
AM		 15472004-05-23 Alan Modra <amodra@bigpond.net.au>
1548
1549 * expr.c (operand, operator): Don't reject '++' and '--'.
1550
20e56c33
RS		 15512004-05-20 Richard Sandiford <rsandifo@redhat.com>
1552
1553 * config/tc-mips.c (append_insn): Use ISA-encoded addresses in MIPS16
1554 dwarf tables.
1555
04ca8703 15562004-05-17 Adam Nemet <anemet@lnxw.com>
b2d65c0b
NC		 1557
1558 * configure.in: Add ppc-*-lynxos*. Update i386-*-lynxos* to ELF.
1559 * configure: Regenerate.
1560
0da76f83
NC		 15612004-05-13 Paul Brook <paul@codesourcery.com>
1562
1563 * dw2gencfi.c (output_cie): Handle dwarf3 format CIE entries.
1564
4b363073
NC		 15652004-05-13 Joel Sherrill <joel@oarcorp.com>
1566
1567 * configure.in (or32-*-rtems*): Switch to elf.
1568 * configure: Regenerate.
1569
9598fbe5
NC		 15702004-05-13 Nick Clifton <nickc@redhat.com>
1571
1572 * po/fr.po: Updated French translation.
1573
22fe14ad
NC		 15742004-05-11 Nick Clifton <nickc@redhat.com>
1575
1576 * doc/as.texinfo (Section): Document G and T flags to .section
1577 directive. Document the extra arguments that the G flag
1578 requires. Document the #tls flag.
1579
15802004-05-11 H.J. Lu <hongjiu.lu@intel.com>
1581
1582 * subsegs.c (section_symbol): Create a new section symbol if
1583 the existing one doesn't match.
1584 * symbols.c (symbol_set_bfdsym): Don't reset BFD section symbol.
1585
532c738a
RS		 15862004-05-07 Richard Sandiford <rsandifo@redhat.com>
1587
1588 * config/tc-mips.c (append_insn, mips_emit_delays): Extend -mfix-vr4120
1589 to cope with VR4181A errata MD(1) and MD(4).
1590
0503b355
BF		 15912004-05-07 Brian Ford <ford@vss.fsi.com>
1592
1593 * NEWS: Mention .secrel32 for pe[i]-i386.
1594
daeea990
AO		 15952004-05-07 Alexandre Oliva <aoliva@redhat.com>
1596
1597 * config/tc-frv.h (MAX_MEM_FOR_RS_ALIGN_CODE): New.
1598 (HANDLE_ALIGN): New.
1599
bfbdcf04
DJ		 16002004-05-06 Daniel Jacobowitz <dan@debian.org>
1601
1602 * Makefile.am (DIST_SUBDIRS): Define.
1603 * aclocal.m4: Regenerate with automake 1.8.4.
1604 * Makefile.in: Likewise.
1605 * doc/Makefile.in: Likewise.
1606
5e819f9c
JW		 16072004-05-06 David Mosberger-Tang <davidm@hpl.hp.com>
1608
1609 * config/tc-ia64.c (dot_serialize): Declare.
1610 (dot_serialize): New function.
1611 (md_pseudo_table): Add ".serialize.data" and
1612 ".serialize.instruction" directives.
1613
e5976317
NC		 16142004-05-06 Nick Clifton <nickc@redhat.com>
1615
1616 * messages (as_internal_value_out_of_range): Print a message about
1617 a value being out of range. Be consistent about whether the
1618 values are printed in decimal or hexadecimal.
1619 (as_warn_value_out_of_range): Generate a warning message about an
1620 out of range value.
1621 (as_bad_value_out_of_range): Generate an error message about an
1622 out of range value.
1623 * as.h: Prototype the new functions.
1624 * config/tc-alpha.c (insert_operand): Use new function.
1625 * config/tc-arc.c (arc_insert_operand): Likewise.
1626 * config/tc-mn10200.c (mn10200_insert_operand): Likewise.
1627 * config/tc-mn10300.c (mn10300_insert_operand): Likewise.
1628 * config/tc-ppc.c (ppc_insert_operand): Likewise.
1629 * config/tc-s390.c (s390_insert_operand): Likewise.
1630 * config/tc-v850.c (v850_insert_operand): Likewise.
1631
996991aa
AO		 16322004-05-05 Alexandre Oliva <aoliva@redhat.com>
1633
1634 * configure.in: Set em=linux for frv-*-*linux*.
1635 * configure: Rebuilt.
1636 * config/tc-frv.h (TARGET_FORMAT): Use elf32-frvfdpic if...
1637 (frv_md_fdpic_enabled): New.
1638 * config/tc-frv.c (frv_md_fdpic_enabled): New.
1639 (DEFAULT_FDPIC): New.
1640 (frv_flags): Use DEFAULT_FDPIC.
1641 (frv_pic_flag): Likewise.
1642 (OPTION_NOPIC): New.
1643 (md_longopts): Add -mnopic.
1644 (md_parse_option): Handle it.
1645 (md_show_usage): Add -mfdpic and -mnopic.
1646
6b6e92f4
NC		 16472004-05-05 Peter Barada <peter@the-baradas.com>
1648
1649 * config/tc-m68k.c: Add find_cf_chip to print list of valid
1650 chips for invalid coldfire instructions, rename selectors
1651 for ColdFire sub-variants, add 521x,5249,547x,548x and aliases,
1652 add current_chip to track which chip is referred to(including save/restore),
1653 use current_chip to select control registers, not current_arch.
1654 (md_show_usage): Add new chips.
1655 * doc/c-m68k.texi: Document new command line switches.
1656
6f079ae3
JJ		 16572004-05-05 Jakub Jelinek <jakub@redhat.com>
1658
1659 * tc-s390.h (md_do_align, HANDLE_ALIGN): Remove.
1660 (NOP_OPCODE): Define.
1661 (s390_align_code): Remove prototype.
1662 * tc-s390.c (s390_align_code): Remove.
1663
2d6ed997
L		 16642004-05-04 H.J. Lu <hongjiu.lu@intel.com>
1665
1666 * config/tc-ia64.c (make_unw_section_name): Removed.
1667 (start_unwind_section): New function.
1668 (generate_unwind_image): Take const segT instead of const
1669 char *.
1670 (dot_handlerdata): Adjusted.
1671 (dot_endp): Likewise.
1672
b9f18452
L		 16732004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1674
1675 * config/obj-elf.c (obj_elf_change_section): Allow the
1676 ".note.GNU-stack" section has SHF_EXECINSTR.
1677
fafe6678
L		 16782004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1679
1680 * config/obj-elf.c (get_section): Return bfd_boolean.
1681 (obj_elf_change_section): Call bfd_get_section_by_name_if
1682 instead of bfd_map_over_sections.
1683
86654c12
L		 16842004-04-30 H.J. Lu <hongjiu.lu@intel.com>
1685
1686 * config/obj-elf.c (get_section): New function.
1687 (obj_elf_change_section): Support multiple sections with same
1688 name.
1689
376eb240
NC		 16902004-04-30 Nick Clifton <nickc@redhat.com>
1691
1692 * config/tc-arm.c (create_register_alias): Fix typo checking for
1693 case sensitive register aliases.
1694 (co_proc_number): Use error message string in all_reg_maps[]
1695 array.
1696 (cp_reg_required_here): Likewise.
1697 (fp_reg_required_here): Likewise.
1698
8a7140c3
NC		 16992004-04-29 Brian Ford <ford@vss.fsi.com>
1700
1701 * dwarf2dbg.c (dwarf2_finish): Add SEC_DEBUGGING to section flags.
1702
3e722fb5
CD		 17032004-04-28 Chris Demetriou <cgd@broadcom.com>
1704
1705 * config/tc-mips.c (HAVE_32BIT_ADDRESSES, append_insn, macro_build)
1706 (load_address, macro, mips_ip, md_parse_option)
1707 (mips_force_relocation, mips_validate_fix, md_apply_fix3)
1708 (s_change_sec, pic_need_relax, tc_gen_reloc): Remove all
1709 embedded-PIC handling, and update comments.
1710 (SWITCH_TABLE): Remove.
1711 * config/tc-mips.h (DIFF_EXPR_OK): Delete.
1712 (enum mips_pic_level): Remove EMBEDDED_PIC.
1713 (EXTERN_FORCE_RELOC): Remove embedded-PIC handling.
1714 (TC_FORCE_RELOCATION): Update comment.
1715 * ecoff.c (ecoff_build_lineno): Add comment about some code that
1716 might be safe to remove now that MIPS embedded-PIC is gone.
1717
351e2b5a
DA		 17182004-04-28 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1719
1720 * config/obj-som.c (obj_som_init_stab_section): Add new arguments in
1721 call to obj_set_subsection_attributes.
1722 (obj_som_init_stab_section): Likewise.
1723 * config/tc-hppa.c (default_subspace_dict): Add comdat field.
1724 (pa_def_subspaces): Provide comdat default.
1725 (pa_subspace): Handle new "comdat" parameter. Set SEC_LINK_ONCE and
1726 not SEC_IS_COMMON if section is comdat, common or dup_common. Update
1727 calls to create_new_subspace and update_subspace to pass comdat flag.
1728 (create_new_subspace, update_subspace): Add new comdat argument. Use
1729 it in calls to obj_set_subsection_attributes.
1730 * doc/c-hppa.texi (.subspa, .nsubspa): Document new comdat parameter
1731 and use of comdat, common and dup_comm parameters.
1732
f5842774
L		 17332004-04-26 H.J. Lu <hongjiu.lu@intel.com>
1734
1735 * config/obj-elf.c (obj_elf_change_section): Check if the old
1736 group name is NULL before comparison.
1737
558fe4a7
CD		 17382004-04-23 Chris Demetriou <cgd@broadcom.com>
1739
1740 * config/tc-mips.h (mips_dwarf2_addr_size): Prototype.
1741
81a21e38
TS		 17422004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1743
1744 * config/tc-mips.c (s_mipsset): Set default CPU type for .set mipsN.
1745
17462004-04-23 Chris Demetriou <cgd@broadcom.com>
1ffcab4b
CD		 1747
1748 * config/tc-mips.c (md_longopts): Remove -membedded-pic option.
1749 (OPTION_MEMBEDDED_PIC): Remove.
1750 (OPTION_TRAP, OPTION_BREAK, OPTION_EB, OPTION_EL)
1751 (OPTION_FP32, OPTION_GP32, OPTION_CONSTRUCT_FLOATS)
1752 (OPTION_NO_CONSTRUCT_FLOATS, OPTIONS_FP64, OPTION_GP64)
1753 (OPTION_RELAX_BRANCH, OPTION_NO_RELAX_BRANCH)
1754 (OPTION_ELF_BASE): Renumber.
1755 (md_parse_option): Remove OPTION_MEMBEDDED_PIC handling.
1756 (md_show_usage): Remove mention of -membedded-pic.
1757 * doc/as.texinfo: Remove mention of -membedded-pic.
1758
4d0d148d
TS		 17592004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1760
1761 * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Remove.
1762 * config/tc-mips.c (RDATA_SECTION_NAME, mips_target_format): Remove
1763 a.out support.
1764 (md_begin, mips_ip, md_parse_option, s_change_sec, s_option,
1765 s_abicalls, nopic_need_relax, tc_gen_reloc): Remove uses of
1766 USE_GLOBAL_POINTER_OPT.
1767
017315e4
TS		 17682004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1769
1770 * config/tc-mips.c (macro): One more use of load_delay_nop.
1771
269137b2
TS		 17722004-04-22 Atsushi Nemoto <anemo@mba.ocn.ne.jp>
1773
1774 * config/tc-mips.c (load_delay_nop): New function.
1775 (load_address, macro): Use load_delay_nop() to build a nop
1776 which can be omitted with gpr_interlocks.
1777
24772049
TS		 17782004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1779
1780 * config/tc-mips.c (hilo_interlocks, gpr_interlocks,
1781 cop_interlocks): Remove superfluous CPU entries.
1782
06516a5e
PB		 17832004-04-22 Paul Brook <paul@codesourcery.com>
1784
1785 * config/tc-arm.c (mav_parse_offset): Value must be multiple of 4.
1786
fd99574b
NC		 17872004-04-22 Peter Barada <peter@the-baradas.com>
1788
1789 * NEWS: Added support for EMAC instructions and MAC/EMAC
1790 Motorola syntax.
1791 * config/m68k-parse.h: Add ACC[123], ACCEXT{01,23}, MAC/EMAC
1792 scale factor tokens, trailing_ampersand to mark mask addressing
1793 for MAC/EMAC instructions.
1794 * config/m68k-parse.y: Add options_ampersand clause, '<<',
1795 '>>'.
1796 (yylex): Handle '>', '<', and '&' following '+'.
1797 * config/tc-m68k.c: Set mcfmac/mcfemac on appropriate ColdFire
04ca8703 1798 architectures in archs[].
fd99574b
NC		 1799 (m68k-ip): Add '4', 'e', 'g', 'i', cases to handle mask addressing
1800 for MAC/EMAC instructions, ACC[0123], ACCEXT{01,23}, and '<<'/'>>'
1801 respectively.
1802 (m68k_ip): Handle trailing '&' on MAC/EMAC insns.
1803 (install_operand): Fix 'n' case, Add 'F', 'f', 'G', 'H', 'I', ']'
1804 cases.
1805 Add EMAC operands to init_table[].
1806
69b97547
NC		 18072004-04-22 Bruno De Bus <bdebus@elis.ugent.be>
1808
1809 * config/tc-arm.h (enum mstate): Move here, add MAP_UNDEFINED
1810 state.
1811 (TC_SEGMENT_INFO_TYPE): Define to enum mstate.
1812 * config/tc-arm.c (enum mstate): Delete from here.
1813 (mapping_state): Remove the static mapstate variable and instead
1814 store the state in the segment. This allows a per-section mapping
1815 state. Handle and ignore MAP_UNDEFINED states.
1816 (arm_elf_change_section): Get the current mapping state from the
1817 new section.
1818 (s_ltorg): Set the mapping state to MAP_DATA.
1819 (arm_cleanup): Use arm_elf_change_section to get the mapping state
1820 for each pool as it is emitted.
1821
18222004-04-22 Nick Clifton <nickc@redhat.com>
1823
1824 * config/tc-arm.h: Formatting tidy ups.
1825
7a7f4e42
CD		 18262004-04-20 Chris Demetriou <cgd@broadcom.com>
1827
1828 * NEWS: Note that MIPS -membedded-pic option is deprecated.
1829
6482c264
NC		 18302004-04-20 DJ Delorie <dj@redhat.com>
1831
1832 * config/tc-i386.h [TE_PE] (TC_CONS_FIX_NEW): Define.
1833 * config/tc-i386.c (md_pseudo_table) [TE_PE]: Add "secrel32".
1834 [TE_PE] (O_secrel): Define.
1835 [TE_PE] (x86_pe_cons_fix_new): New.
1836 [TE_PE] (pe_directive_secrel): Likewise.
1837 (tc_gen_reloc) [TE_PE]: Support BFD_RELOC_32_SECREL.
1838
5bf10039
EC		 18392004-04-19 Eric Christopher <echristo@redhat.com>
1840
c04f5787
AM		 1841 * config/tc-mips.c (mips_dwarf2_addr_size): Revert part
1842 of previous patch for fix in gcc.
5bf10039 1843
77cd6497
JJ		 18442004-04-19 Jakub Jelinek <jakub@redhat.com>
1845
1846 * config/tc-xtensa.c (xg_assembler_literal): Fix a typo.
1847
b8861cfb
NS		 18482004-04-19 Nathan Sidwell <nathan@codesourcery.com>
1849
1850 * read.c (do_align): Call md_flush_pending_output, if defined.
1851
f4c724fc
AM		 18522004-04-16 Alan Modra <amodra@bigpond.net.au>
1853
1854 * expr.c (operand): Correct checks for ++ and --.
1855
a4b6e8f6
L		 18562004-04-14 H.J. Lu <hongjiu.lu@intel.com>
1857
1858 * config/tc-generic.c: Add some comments.
1859
d766e8ec
RS		 18602004-04-14 Richard Sandiford <rsandifo@redhat.com>
1861
1862 * doc/c-mips.texi (-m{no-,}fix-vr4120): Renamed from
1863 -{no-}mfix-vr4122-bugs.
1864 * config/tc-mips.c (mips_fix_vr4120): Renamed from mips_fix_4122_bugs.
1865 (append_insn, mips_emit_delays): Update accordingly.
1866 (OPTION_FIX_VR4120, OPTION_NO_FIX_VR4120): Renamed from *VR4122.
1867 (md_longopts): Change -{no-,}mfix-vr4122-bugs to -m{no-,}fix-vr4120.
1868 (md_parse_option): Update after above changes.
1869 (md_show_usage): Add -mfix-vr4120.
1870
ed9589d4
BW		 18712004-04-13 Bob Wilson <bob.wilson@acm.org>
1872
1873 * doc/as.texinfo (Sub-Sections): Conditionalize COFF-specific use
1874 of .section directive; add a reference to the ELF .subsection
1875 directive.
1876
097dba13
NC		 18772004-04-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1878
1879 * config/tc-m32r.c (md_assemble): Fixed infinite loop bug
1880 in parallel.
1881
279500de
TS		 18822004-04-11 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1883
1884 * Makefile.am: Remove mips from aout targets.
1885 * Makefile.in: Regenerate.
1886 * configure.in: Remove mips-dec-bsd* target.
1887 * configure: Regenerate.
1888
b7d7dc63
AM		 18892004-04-07 Alan Modra <amodra@bigpond.net.au>
1890
1891 PR 96
1892 * config/tc-ppc.c (ppc_elf_suffix): Add valid32 and valid64 fields
1893 to struct map_bfd. Adjust MAP macro, and define MAP32, MAP64.
1894 Update "mapping". Restrict some @ modifiers to 32 bit.
1895
f55629b8
KK		 18962004-04-01 Asgari Jinia <asgarij@kpitcummins.com>
1897 Dhananjay Deshpande <dhananjayd@kpitcummins.com>
1898
1899 * config/tc-sh.c (dont_adjust_reloc_32): New variable.
1900 (sh_fix_adjustable): Avoid adjusting BFD_RELOC_32 when
1901 dont_adjust_reloc_32 is set.
1902 (md_longopts): Add option -renesas.
1903 (md_parse_option, md_show_usage): Likewise.
1904 * doc/c-sh.texi: Likewise.
1905
a7844384
NC		 19062004-04-01 Dave Korn <dk@artimi.com>
1907
1908 * config/tc-dlx.c (md_assemble): set fx_no_overflow flag for
1909 hi16 and lo16 fixS structs.
1910 (md_assemble): generate bit_fixS for RELOC_DLX_LO16 in
1911 exactly the same way as for RELOC_DLX_REL16.
1912 (machine_ip): properly respect LO flag in the_insn and
1913 output RELOC_DLX_LO16 rather than RELOC_DLX_16.
1914 (md_apply_fix3): apply RELOC_DLX_LO16.
1915
6c71dbc4
SS		 19162004-03-30 Stan Shebs <shebs@apple.com>
1917
1918 Remove long-obsolete MPW support.
1919 * mpw-config.in, mpw-make.sed, mac-as.r: Remove files.
1920 * configure.in: Remove mention of ppc-*-mpw* config.
1921 * configure.in: Likewise.
1922
7cc69913
NC		 19232004-03-30 Nick Clifton <nickc@redhat.com>
1924
1925 * config/tc-arm.c (meabi_flags): Make its use conditional upon
1926 OBJ_ELF being defined.
1927
53fbb48a
AM		 19282004-03-27 Alan Modra <amodra@bigpond.net.au>
1929
1930 * config/obj-aout.c (obj_aout_type): Remove #ifdef BFD_ASSEMBLER code.
1931
d507cf36
PB		 19322004-03-23 Paul Brook <paul@codesourcery.com>
1933
1934 * config/tc-arm.c (meabi_flags): New variable.
1935 (arm_parse_eabi): New function.
1936 (md_begin): Set flags for EABI v3.
1937 (arm_eabis): Add.
1938 (arm_long_opts): Add meabi.
1939 * doc/as.texinf <ARM>: Document -meabi.
1940 * doc/c-arm.texi: Ditto.
1941
88d65ad6
BW		 19422004-03-22 Bob Wilson <bob.wilson@acm.org>
1943
1944 * config/tc-xtensa.c (xtensa_post_relax_hook): Create literal
1945 tables even when use_literal_section flag is not set.
1946
13ef8878
AM		 19472004-03-22 Alan Modra <amodra@bigpond.net.au>
1948
1949 * config/tc-sh.c: Remove trailing whitespace.
1950
1048a9ba
HPN		 19512004-03-22 Hans-Peter Nilsson <hp@axis.com>
1952
1953 * doc/c-cris.texi (CRIS-Opts): Document --no-mul-bug-abort,
1954 --mul-bug-abort and the default behavior.
1955 * config/tc-cris.c (cris_insn_kind): New member CRIS_INSN_MUL.
1956 (err_for_dangerous_mul_placement): New variable.
1957 (STATE_MUL, OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): New
1958 macros.
1959 (md_cris_relax_table): Have placeholder for STATE_MUL.
1960 (md_longopts): New options --mul-bug-abort and --no-mul-bug-abort.
1961 (cris_relax_frag) <case ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: New
1962 case doing nothing.
1963 (md_estimate_size_before_relax) <case ENCODE_RELAX (STATE_MUL,
1964 STATE_BYTE)>: Ditto.
1965 (md_convert_frag) <ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: Check
1966 alignment and position of this frag, emit error message if
1967 suspicious.
1968 (md_assemble): For a multiply insn and when checking it,
1969 transform the current frag into a special frag for that purpose.
1970 (md_parse_option) <case OPTION_MULBUG_ABORT_OFF, case
1971 OPTION_MULBUG_ABORT_ON>: Handle new options.
1972
a7877748
BW		 19732004-03-19 Bob Wilson <bob.wilson@acm.org>
1974
1975 * config/tc-xtensa.c (mark_literal_frags): New function.
1976 (xtensa_move_literals): Call mark_literal_frags for all literal
1977 segments, including init and fini literal segments.
1978 (xtensa_post_relax_hook): Swap use of xt_insn_sec and xt_literal_sec.
1979
9992270f
DA		 19802004-03-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1981
1982 * tc-hppa.c (cons_fix_new_hppa): Check for PC relative base type.
1983 (pa_comm): Set BSF_OBJECT in symbol flags.
1984
6e1772d5
AM		 19852004-03-19 Alan Modra <amodra@bigpond.net.au>
1986
1987 * Makefile.am: Run "make dep-am".
1988 * Makefile.in: Regenerate.
1989 * doc/Makefile.in: Regenerate.
1990 * config.in: Regenerate.
1991 * po/gas.pot: Regenerate.
1992
3e4caed2
NS		 19932004-03-18 Nathan Sidwell <nathan@codesourcery.com>
1994
c95b35a9
NS		 1995 * read.c (read_a_source_file): Use demand_empty_rest_of_line.
1996 (demand_empty_rest_of_line): Issue an error here.
1997 (ignore_rest_of_line): Silently skip to end.
1998 (demand_copy_string): Issue an error, not warning.
1999 (equals): Likewise.
2000 * config/obj-elf.c (obj_elf_section_name): Likewise.
2001 (obj_elf_section): Likewise.
2002 * config/tc-arc.c (arc_extoper): Remove bogus NULL checks.
2003 (arc_extinst): Likewise.
2004 * config/tc-ia64.c (dot_saveb): Use demand_empty_rest_of_line.
2005 (dot_spill): Likewise.
2006 (dot_unwabi): Likewise.
2007 (dot_prologue): Likewise.
2008
3e4caed2
NS		 2009 * expr.c (operand): Reject ++ and --.
2010 (operator): Likewise.
2011
2ce4cc60
KK		 20122004-03-17 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2013
2014 * config/tc-sh.c: Include dw2gencfi.h.
2015 (sh_cfi_frame_initial_instructions): New function.
2016 (sh_regname_to_dw2regnum): Likewise.
2017 * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Move to the end of
2018 file.
2019 (TARGET_USE_CFIPOP): Define.
2020 (tc_cfi_frame_initial_instructions): Likewise.
2021 (tc_regname_to_dw2regnum): Likewise.
2022 (DWARF2_DEFAULT_RETURN_COLUMN, DWARF2_CIE_DATA_ALIGNMENT): Likewise.
2023 * Makefile.am: Update dependencies.
2024 * Makefile.in: Regenerate.
2025
9183ed4b
AM		 20262004-03-17 Ralf Corsepius <corsepiu@faw.uni-ulm.de>
2027
2028 * configure.in: Switch sh-*-rtems* to ELF. Add sh-*-rtemscoff*.
2029 * configure: Regenerate.
2030
00bbdfe7
BW		 20312004-03-12 Bob Wilson <bob.wilson@acm.org>
2032
2033 * read.c (s_leb128): Call md_flush_pending_output.
2034
0f10071e
ML		 20352004-03-12 Michal Ludvig <mludvig@suse.cz>
2036
2037 * config/tc-i386.c (output_insn): Handle PadLock instructions.
2038 * config/tc-i386.h (CpuPadLock): New define.
2039 (CpuUnknownFlags): Added CpuPadLock.
2040
84c84759
AS		 20412004-03-07 Andreas Schwab <schwab@suse.de>
2042
2043 * doc/c-hppa.texi (HPPA Directives): Fix typo.
2044
6ec51dba
RH		 20452004-03-07 Richard Henderson <rth@redhat.com>
2046
2047 * dw2gencfi.c (output_cie): Align length to 4 byte boundary.
2048 (cfi_finish): Likewise for fde.
2049
bae25f19
L		 20502004-03-05 H.J. Lu <hongjiu.lu@intel.com>
2051
2052 * config/tc-ia64.c (md_assemble): Properly handle NULL
2053 align_frag.
2054 (ia64_handle_align): Don't abort if failed to add a stop bit.
2055
ab509236
L		 20562004-03-04 H.J. Lu <hongjiu.lu@intel.com>
2057
2058 * Makefile.in: Regenerated.
2059 * aclocal.m4: Likewise.
2060 * configure: Likewise.
2061 * doc/Makefile.in: Likewise.
2062
d9201763
L		 20632004-03-03 H.J. Lu <hongjiu.lu@intel.com>
2064
2065 * config/tc-ia64.c (dot_align): New.
2066 (ia64_do_align): Make it static.
2067 (md_pseudo_table): Use "dot_align" for "align".
2068 (ia64_md_do_align): Don't set align_frag here.
2069 (ia64_handle_align): Add a stop bit to the previous bundle if
2070 needed.
2071
2072 * config/tc-ia64.h (ia64_do_align): Removed.
2073
ae51a426
JR		 20742003-03-03 Andrew Stubbs <andrew.stubbs@superh.com>
2075
2076 * config/tc-sh.c (md_parse_option): Add -isa=sh4-nofpu and
2077 -isa=sh4-nommu-nofpu options. Adjust help messages accordingly.
2078 (sh_elf_final_processing): Output BFD type sh4_nofpu if that is
2079 the most general type or the user specifically requested it.
2080 (md_assemble): Add a new error message for when an instruction
2081 is understood, but is not allowed due to an -isa option.
2082
9545c4ce
L		 20832004-03-02 H.J. Lu <hongjiu.lu@intel.com>
2084
2085 * config/tc-ia64.c (align_frag): New.
2086 (md_assemble): Set the tc_frag_data field in align_frag for
2087 IA64_OPCODE_FIRST instructions.
2088 (ia64_md_do_align): Set align_frag.
2089 (ia64_handle_align): Add a stop bit if needed.
2090
2091 * config/tc-ia64.h (TC_FRAG_TYPE): New.
2092 (TC_FRAG_INIT): New.
2093
676a64f4
RS		 20942004-03-01 Richard Sandiford <rsandifo@redhat.com>
2095
2096 * config/tc-frv.c (fr400_audio): New variable.
2097 (md_parse_option, md_show_usage): Add -mcpu=fr405 and -mcpu=fr450.
2098 (md_parse_option): Set fr400_audio for -mcpu=fr400 and -mcpu=fr405.
2099 (target_implements_insn_p): New function.
2100 (md_assemble): Report an error if the processor doesn't implement
2101 the instruction.
2102
bd337dde
NC		 21032004-02-27 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2104
2105 * config/tc-m32r.c (md_longopts): Added -no-bitinst option.
2106 (md_parse_option): Ditto.
2107 (OPTION_NO_SPECIAL_M32R): Added.
2108 (md_show_usage): Document it.
2109 (enable_speial_m32r): Changed a default value from 0 to 1.
2110 * doc/c-m32r.texi: Document the -no-bitinst option.
2111
78878175
NC		 21122004-02-27 Nick Clifton <nickc@redhat.com>
2113
2114 * config/tc-sh.c (get_operand): Revert previous delta.
2115 (tc_gen_reloc): Check for an unknown reloc type before processing
2116 the addend.
2117
1d6d62a4
MS		 21182004-02-27 Hannes Reinecke <hare@suse.de>
2119
2120 * config/tc-s390.c (s390_insn): Correct range check for opcode in
2121 .insn pseudo operation.
2122
4aa3e325
NC		 21232004-02-27 Anil Paranjpe <anilp1@kpitcummins.com>
2124
2125 * config/tc-sh.c (get_operand): In case of #Imm, check has been
2126 added for wrong syntax.
2127
73369e65
EC		 21282004-02-26 Eric Christopher <echristo@redhat.com>
2129
2130 * config/tc-mips.c (mips_dwarf2_addr_size): New.
2131 * config/tc-mips.h (DWARF2_ADDR_SIZE): Use.
2132
6a5709a5
JR		 21332004-02-26 Andrew Stubbs <andrew.stubbs@superh.com>
2134
2135 * config/tc-sh.c (build_Mytes): Add REG_N_D and REG_N_B01
2136 nibble types to assembler.
2137
95385380
FF		 21382004-02-25 Fred Fish <fnf@redhat.com>
2139
2140 * config/tc-iq2000.c: Add missing \n\ in multiline string literal.
2141
b5e0fabd
JW		 21422004-02-20 James E Wilson <wilson@specifixinc.com>
2143
2144 * config/tc-ia64.c (slot_index): New arg before_relax. Use instead of
2145 finalize_syms.
2146 (fixup_unw_records): New arg before_relax. Pass to slot_index.
2147 (ia64_estimate_size_before_relax): New.
2148 (ia64_convert_frag): Pass 0 to fixup_unw_records. Add comment.
2149 (generate_unwind_image): Pass 1 to fixup_unw_records.
2150 * config/tc-ia64.h (ia64_estimate_size_before_relax): Declare.
2151 (md_estimate_size_before_relax): Call ia64_estimate_size_before_relax.
2152
2f01ffbf
JJ		 21532004-02-19 Jakub Jelinek <jakub@redhat.com>
2154
2155 * stabs.c (generate_asm_file): Avoid warning about use of
2156 uninitialized variable.
2157
9c59842f
JW		 21582004-02-18 David Mosberger <davidm@hpl.hp.com>
2159
2160 * config/tc-ia64.c (ia64_flush_insns): In addition to prologue,
2161 body, and endp, allow unwind records which do not have a "t"
2162 (time/instruction) field.
2163
34920d91
NC		 21642004-02-17 Petko Manolov <petkan@nucleusys.com>
2165
2166 * config/tc-arm.c (do_mav_dspsc_1): Correct offset of CRn.
2167 (do_mav_dspsc_2): Likewise.
2168 Fix accumulator registers move opcodes.
2169
4dde8e61 21702004-02-13 Hannes Reinecke <hare@suse.de>
13ef8878 2171 Jakub Jelinek <jakub@redhat.com>
4dde8e61
AJ		 2172
2173 * dwarf2dbg.c (get_filenum): Do not read beyond allocated memory.
2174
936cf02e
JW		 21752004-02-10 Steve Ellcey <sje@cup.hp.com>
2176
2177 * config/tc-ia64.h (ia64_frob_symbol): New declaration.
2178 (tc_frob_symbol): New macro definition.
2179 * config/tc-ia64.c (ia64_frob_symbol): New routine.
2180
e41f12f3
DJ		 21812004-02-09 Daniel Jacobowitz <drow@mvista.com>
2182
2183 * config/tc-arm.c (md_begin): Mark .note.gnu.arm.ident as
2184 read-only.
2185
a8a3b3b2
NS		 21862004-02-09 Nathan Sidwell <nathan@codesourcery.com>
2187
2188 * read.h (IGNORE_OPCODE_CASE): Do not define. Replace with ...
2189 (TC_CASE_SENSITIVE): ... this.
2190 * read.c: Replace IGNORE_OPCODE_CASE with TC_CASE_SENSITIVE.
2191 * doc/internals.texi (TC_CASE_SENSITIVE): Document.
2192
c97b7ef6
JW		 21932004-02-06 James E Wilson <wilson@specifixinc.com>
2194
2195 * config/tc-ia64.c (dot_endp): Delete call to output_endp.
2196 (generate_unwind_image): Re-add it here.
2197
e4475e39
NS		 21982004-02-06 Nathan Sidwell <nathan@codesourcery.com>
2199
2200 * dwarf2dbg.c (DWARF2_ADDR_SIZE): Remove trailing ';'
2201 * read.h (SKIP_WHITESPACE): Turn into an expression.
2202 * read.c (read_a_source_file): A pseudo is removed by having a
2203 NULL handler.
2204
5738bc24
JW		 22052004-02-05 James E Wilson <wilson@specifixinc.com>
2206
2207 * config/tc-ia64.c (output_endp): New.
2208 (count_bits): Delete.
2209 (ia64_flush_insns, process_one_record, optimize_unw_records): Handle
2210 endp unwind records.
2211 (fixup_unw_records): Handle endp unwind records. Delete code for
2212 shortening prologue regions not followed by a body record.
2213 (dot_endp): Call add_unwind_entry to emit endp unwind record.
2214 * config/tc-ia64.h (unw_record_type): Add endp.
2215
d6e78c11
JW		 22162004-02-03 James E Wilson <wilson@specifixinc.com>
2217
2218 * config/tc-ia64.c (ia64_convert_frag): Call md_number_to_chars to
2219 fill padding bytes with zeroes.
2220 (emit_one_bundle): New locals last_ptr, end_ptr. Rewrite code that
2221 sets unwind_record slot_number and slot_frag fields.
2222
f6a22291
MR		 22232004-02-02 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2224
2225 * config/tc-mips.c (add_got_offset_hilo): New function.
2226 (macro): Use load_register() and add_got_offset_hilo() to load
2227 constants instead of hardcoding code sequences throughout.
2228
127cab00
L		 22292004-01-28 H.J. Lu <hongjiu.lu@intel.com>
2230
2231 * config/tc-ia64.c (emit_one_bundle): Add proper indentation.
2232
6358301e
NC		 22332004-01-26 Bernardo Innocenti <bernie@develer.com>
2234
2235 * config/tc-m68k.h (EXTERN_FORCE_RELOC): Handle m68k-uclinux specially,
2236 like m68k-elf.
2237 * config/tc-m68k.c (RELAXABLE_SYMBOL): Use EXTERN_FORCE_RELOC instead
2238 of hard-coded test for TARGET_OS=elf.
2239
c8978940
CD		 22402004-01-24 Chris Demetriou <cgd@broadcom.com>
2241
2242 * config/tc-mips.c (hilo_interlocks): Change definition
2243 so that MIPS32, MIPS64 and later ISAs are included, along with
2244 the already-included machines. Update comments.
2245
c3ba240c
DJ		 22462004-01-23 Daniel Jacobowitz <drow@mvista.com>
2247
2248 * config/tc-arm.c (tc_gen_reloc): Improve error message for
2249 undefined local labels.
2250
e864ceca
RS		 22512004-01-23 Richard Sandiford <rsandifo@redhat.com>
2252
2253 * config/tc-mips.c (load_address, macro): Update comments about
2254 NewABI GP relaxation.
2255
67c0d1eb
RS		 22562004-01-23 Richard Sandiford <rsandifo@redhat.com>
2257
2258 * config/tc-mips.c (macro_build): Remove place and counter arguments.
2259 (mips_build_lui, macro_build_ldst_constoffset): Likewise.
2260 (mips16_macro_build, macro_build_jalr): Remove counter argument.
2261 (set_at, load_register, load_address, move_register): Likewise.
2262 (load_got_offset, add_got_offset): Likewise.
2263 Update all calls and tidy accordingly.
2264
584892a6
RS		 22652004-01-23 Richard Sandiford <rsandifo@redhat.com>
2266
2267 * config/tc-mips.c (RELAX_ENCODE): Remove WARN argument.
2268 (RELAX_FIRST, RELAX_SECOND): Turn into 8-bit quantities.
2269 (RELAX_USE_SECOND): Bump to 0x10000.
2270 (RELAX_SECOND_LONGER, RELAX_NOMACRO, RELAX_DELAY_SLOT): New flags.
2271 (mips_macro_warning): New variable.
2272 (md_assemble): Wrap macro expansion in macro_start() and macro_end().
2273 (s_cpload, s_cpsetup, s_cprestore, s_cpreturn): Likewise.
2274 (relax_close_frag): Set mips_macro_warning.first_frag. Adjust use
2275 of RELAX_ENCODE.
2276 (append_insn): Update mips_macro_warning.sizes.
2277 (macro_start, macro_warning, macro_end): New functions.
2278 (macro_build): Don't emit warnings here.
2279 (macro_build_lui, md_estimate_size_before_relax): ...or here.
2280 (md_convert_frag): Check for cases where one macro alternative
2281 needs a warning and the other doesn't. Emit a warning if the
2282 longer sequence was chosen.
2283
4d7206a2
RS		 22842004-01-23 Richard Sandiford <rsandifo@redhat.com>
2285
2286 * config/tc-mips.h (tc_frag_data_type, TC_FRAG_TYPE): Remove.
2287 * config/tc-mips.c (RELAX_ENCODE): Take three arguments: the size of
2288 the first sequence, the size of the second sequence, and a flag
2289 that says whether we should warn.
2290 (RELAX_OLD, RELAX_NEW, RELAX_RELOC[123]): Delete.
2291 (RELAX_FIRST, RELAX_SECOND): New.
2292 (mips_relax): New variable.
2293 (relax_close_frag, relax_start, relax_switch, relax_end): New fns.
2294 (append_insn): Remove "place" argument. Use mips_relax.sequence
2295 rather than "place" to check whether we're expanding the second
2296 alternative of a relaxable macro. Remove redundant check for
2297 branch relaxation. If generating a normal insn, and there
2298 is not enough room in the current frag, call relax_close_frag()
2299 to close it. Update mips_relax.sizes[]. Emit fixups for the
2300 second version of a relaxable macro. Record the first relaxable
2301 fixup in mips_relax. Remove tc_gen_reloc workaround.
2302 (macro_build): Remove all uses of "place". Use mips_relax.sequence
2303 in the same way as in append_insn.
2304 (mips16_macro_build): Remove "place" argument.
2305 (macro_build_lui): As for macro_build. Don't drop the add_symbol
2306 when generating the second version of a relaxable macro.
2307 (load_got_offset, add_got_offset): New functions.
2308 (load_address, macro): Use new relaxation machinery. Remove
2309 tc_gen_reloc workarounds.
2310 (md_estimate_size_before_relax): Set RELAX_USE_SECOND if the second
2311 version of a relaxable macro is needed. Return -RELAX_SECOND if the
2312 first version is needed.
2313 (tc_gen_reloc): Remove relaxation handling.
2314 (md_convert_frag): Go through the fixups for a relaxable macro and
2315 mark those that belong to the unneeded alternative as done. If the
2316 second alternative is needed, adjust the fixup addresses to account
2317 for the deleted first alternative.
2318
0530d30a
RS		 23192004-01-23 Richard Sandiford <rsandifo@redhat.com>
2320
2321 * frags.h (frag_room): Declare.
2322 * frags.c (frag_room): New function.
2323 * doc/internals.texi: Document it.
2324
048cdf86
TS		 23252004-01-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
2326
2327 * config/tc-mips.c (append_insn): Don't do r3900 interlock
2328 optimization for -mtune=r3900, as this will break on other CPUs.
2329
28d39d1a
NC		 23302004-01-11 Tom Rix <tcrix@worldnet.att.net>
2331
2332 * config/tc-m68hc11.c (build_indexed_byte): movb and movw cannot
73369e65 2333 be relaxed, use fixup.
28d39d1a
NC		 2334 (md_apply_fix3): Use 5 bit reloc from movb and movw fixup.
2335
26664553
JJ		 23362004-01-19 Jakub Jelinek <jakub@redhat.com>
2337
2338 * config/tc-sparc.c (sparc_ip): Disallow %f32-%f63 for single
2339 precision operands.
2340
895921c9
MR		 23412004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2342
2343 * config/tc-mips.c (append_insn): Properly detect variant frags
2344 that preclude swapping of relaxed branches. Correctly swap
2345 instructions between frags when dealing with relaxed branches.
2346
f8b73030
MR		 23472004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2348
2349 * acinclude.m4: Quote names of macros to be defined by AC_DEFUN
2350 throughout.
2351 * aclocal.m4: Regenerate.
2352 * configure: Regenerate.
2353
ca9a79a1 23542004-01-12 Anil Paranjpe <anilp1@KPITCummins.com>
73369e65 2355
ca9a79a1
NC		 2356 * config/tc-h8300.c (build_bytes): Apply relaxation to bit
2357 manipulation insns.
2358
cc3d92a5
RS		 23592004-01-12 Richard Sandiford <rsandifo@redhat.com>
2360
2361 * config/tc-mips.c (macro_build_jalr): When adding an R_MIPS_JALR
2362 reloc, reserve space for the delay slot as well as the jalr itself.
2363
e45d0630
PB		 23642004-01-09 Paul Brook <paul@codesourcery.com>
2365
2366 * config/tc-arm.c (do_vfp_reg2_from_sp2): Rename from do_vfp_sp_reg2.
2367 (do_vfp_sp2_from_reg2): New function.
2368 (insns): Use them.
2369 (do_vfp_dp_from_reg2): Check return values properly.
2370
81912461
ILT		 23712004-01-08 Ian Lance Taylor <ian@wasabisystems.com>
2372
78849248
ILT		 2373 * config/tc-mips.c (warn_nops): Remove static variable.
2374 (macro): Remove test of warn_nops.
2375 (md_shortops): Remove 'n'.
2376 (md_parse_option): Remove 'n' case.
2377 (md_show_usage): Remove -n.
2378 * doc/as.texinfo (Overview): Remove MIPS -n option.
2379 * doc/c-mips.texi (MIPS Opts): Remove mention -n.
2380 * NEWS: Mention removal of MIPS -n option.
2381
81912461
ILT		 2382 * config/tc-mips.c (ISA_HAS_COPROC_DELAYS): Remove.
2383 (cop_interlocks): Check ISA level.
2384 (cop_mem_interlocks): Define.
2385 (reg_needs_delay): Check cop_interlocks rather than
2386 ISA_HAS_COPROC_DELAYS.
2387 (append_insn): Likewise. Use cop_mem_interlocks rather than
2388 directly checking mips_opts.isa.
2389 (mips_emit_delays): Likewise.
2390
73f20958
L		 23912004-01-07 H.J. Lu <hongjiu.lu@intel.com>
2392
2393 * config/tc-ia64.c (unwind): Move next_slot_number and
2394 next_slot_frag to ...
2395 (unw_rec_list): Here.
2396 (free_list_records): Removed.
2397 (output_unw_records): Likewise.
2398 (generate_unwind_image): Make it void.
2399 (alloc_record): Initialize next_slot_number and next_slot_frag.
2400 (slot_index): Take .org, .space and .align into account.
2401 (fixup_unw_records): Don't set slot_number to 0. Use
2402 list->next_slot_number and list->next_slot_frag instead of
2403 unwind.next_slot_number and unwind.next_slot_frag.
2404 (ia64_convert_frag): New.
2405 (generate_unwind_image): Generate a rs_machine_dependent frag
2406 for unwind record.
2407 (emit_one_bundle): Use list->next_slot_number and
2408 list->next_slot_frag instead of unwind.next_slot_number and
2409 unwind.next_slot_frag.
2410
2411 * config/tc-ia64.h (md_convert_frag): Defined as
2412 ia64_convert_frag.
2413 (md_estimate_size_before_relax): Defined as (f)->fr_var.
2414
99d09cdb
AO		 24152004-01-06 Alexandre Oliva <aoliva@redhat.com>
2416
2417 2003-12-19 Alexandre Oliva <aoliva@redhat.com>
2418 * config/tc-frv.h (md_apply_fix3): Don't define.
2419 * config/tc-frv.c (md_apply_fix3): New. Shift/truncate %hi/%lo
2420 operands.
2421 * config/tc-frv.h (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
2422 2003-10-07 Alexandre Oliva <aoliva@redhat.com>
2423 * config/tc-frv.c (line_separator_chars): Add `!'.
2424 2003-09-19 Alexandre Oliva <aoliva@redhat.com>
2425 * config/tc-frv.c (md_assemble): Clear insn upfront.
2426 2003-09-18 Alexandre Oliva <aoliva@redhat.com>
2427 * config/tc-frv.c (OPTION_FDPIC): New macro.
2428 (md_longopts): Add mfdpic.
2429 (md_parse_option): Handle it.
2430 2003-08-04 Alexandre Oliva <aoliva@redhat.com>
2431 * config/tc-frv.c (md_cgen_lookup_reloc) <FRV_OPERAND_D12,
2432 FRV_OPERAND_S12>: Use reloc type encoded in fix-up.
2433 (frv_pic_ptr): Parse funcdesc.
2434
4a4c4a1d
MR		 24352004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2436
2437 * doc/as.texinfo: Let texi2pod parse asconfig.texi and
2438 gasver.texi. Remove duplicate symbol definitions for texi2pod.
2439
2251dd89
MR		 24402004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2441
2442 * Makefile.am (Makefile): Move the dependency on
2443 $(BFDDIR)/configure.in to...
2444 (CONFIG_STATUS_DEPENDENCIES): ... here.
2445 (AUTOMAKE_OPTIONS): Require automake 1.8.
2446 * Makefile.in: Regenerate.
2447 * doc/Makefile.am (BASEDIR, BFDDIR): Define.
2448 (CONFIG_STATUS_DEPENDENCIES): Add a dependency on
2449 $(BFDDIR)/configure.in here as well.
2450 * doc/Makefile.in: Regenerate.
2451
c6581168
MR		 24522004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2453
2454 * Makefile.am (install, install-info, RECURSIVE_TARGETS): Remove.
2455 * Makefile.in: Regenerate.
2456 * aclocal.m4: Regenerate.
2457 * doc/Makefile.am (install, install-info): Remove.
2458 (install-data-local): A new hook for install-info.
2459 (AUTOMAKE_OPTIONS): Require automake 1.8.
2460 * doc/Makefile.in: Regenerate.
2461
303e7b79
NC		 24622004-01-02 Nutan Singh <nutan@kpitcummins.com>
2463
2464 * doc/c-sh.texi: Update description about floating point behavior
2465 of SH family.
2466
b50c15dc
NC		 24672004-01-02 Bernardo Innocenti <bernie@develer.com>
2468
2469 * configure.in: Add m68k-uClinux target.
2470 * configure: Regenerate.
7de9afa2 2471
c9e214e5 2472For older changes see ChangeLog-0203
08d56133
NC		 2473\f
2474Local Variables:
2475mode: change-log
2476left-margin: 8
2477fill-column: 74
2478version-control: never
2479End:
GDB Binutils - Deliverables
This page took 0.327893 seconds and 4 git commands to generate.