projects / deliverable / binutils-gdb.git / blame
? search:
summary | shortlog | log | commit | commitdiff | tree
blob | blame (incremental) | history | HEAD
Fix support for PECOFF weak symbols
[deliverable/binutils-gdb.git] / gas / ChangeLog
CommitLineData
977cdf5a
NC		 12004-11-08 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
2
3 * symbols.c (any_external_name): Define.
4 (resolve_symbol_value): Do not convert weak symbols.
5 (S_SET_EXTERNAL): Support any_external_name.
6 (S_SET_NAME): Qualify parameter const.
7 (symbol_equated_reloc_p): Don't equate weaks when relocating.
8 * symbols.h (S_SET_NAME): Qualfiy parameter const.
9 * tc.h (any_external_name): Declare.
10 * config/obj-coff.c ("coff/pe.h"): Include for BFD
11 assemblers also.
12 (weak_is_altname): Declare and define.
13 (weak_name2altname): Same.
14 (weak_altname2name): Same.
15 (weak_uniquify): Same.
16 (weak_altprefix): Define.
17 (obj_coff_weak): Change .weak syntax and handling.
18 (coff_frob_symbol): Fix PE weak symbol alternates.
19 * config/obj-coff.h (USE_UNIQUE): Define.
20 * config/tc-i386.c (md_apply_fix3): Assume weak symbols
21 are in another segment.
22 (tc_gen_reloc): Remove broken addend hack.
23 doc/as.texinfo: Update.
24
34e41783
BW		 252004-11-05 Sterling Augustine <sterling@tensilica.com>
26
27 * config/tc-xtensa.c (total_frag_text_expansion): New.
28 (md_estimate_size_before_relax): Use it.
29 (find_address_of_next_align_frag): Likewise.
30
dcd46b4e
TL		 312004-11-05 Tomer Levi <Tomer.Levi@nsc.com>
32
33 * config/tc-crx.c: Rename argument types.
34 (processing_arg_number): Rename to 'cur_arg_num'.
35 (get_number_of_bits): Rename to 'set_operand_size'.
36 (get_operandtype): Rename to 'parse_operand', totally rewrite.
37 (set_cons_rparams): Rename to 'set_operand', totally rewrite.
38 (set_indexmode_parameters): Remove function, integrate its code into
39 'set_operand'.
40 (set_operand_size): Get rid of 'Operand Number' function parameter -
41 use global variable 'cur_arg_num' instead.
42 Use a local 'argument' pointer to reference the current argument.
43 (parse_operand): Likewise.
44 (set_operand): Likewise.
45 (process_label_constant): Likewise.
46
9456465c
BW		 472004-11-04 Bob Wilson <bob.wilson@acm.org>
48
49 * config/tc-xtensa.c: Remove XTENSA_SECTION_RENAME ifdefs.
50 (add_section_rename): Delete. Inlined into...
51 (build_section_rename): ...here. Use xstrdup instead of strdup.
52 (xtensa_section_rename): Drop "const" from argument and return types.
53 (md_show_usage): Indent to match show_usage().
54 * config/tc-xtensa.h: Remove XTENSA_SECTION_RENAME ifdefs.
55 (tc_canonicalize_section_name): Define.
56 (md_elf_section_rename): Remove unused macro.
57 * doc/as.texinfo (Overview): Document Xtensa --rename-section option.
58 * doc/c-xtensa.texi (Xtensa Options): Likewise.
59 (Frame Directive): Delete.
60
7f266840
DJ		 612004-11-04 Daniel Jacobowitz <dan@debian.org>
62
63 * configure.in: Remove arm-*-oabi and thumb-*-oabi.
64 * config/tc-arm.c (target_oabi): Delete.
65 (md_apply_fix3, elf32_arm_target_format): Remove target_oabi checks.
66 (arm_opts): Remove moabi.
67 * doc/as.texinfo (Overview): Remove documentation of -moabi.
68 * doc/c-arm.texi (ARM Options): Likewise.
69 * configure: Regenerated.
70
ae57792d
HPN		 712004-11-04 Hans-Peter Nilsson <hp@axis.com>
72
73 * configure.in (crisv32): Recognize. AC_DEFINE_UNQUOTED
74 DEFAULT_CRIS_ARCH. Handle crisv32-*-linux-gnu* like
75 cris-*-linux-gnu* and crisv32-*-* like cris-*-*.
76 * configure: Regenerate.
77 * config/tc-cris.c (enum cris_archs): New.
78 (cris_mach, cris_arch_from_string, s_cris_arch, get_sup_reg)
79 (cris_insn_ver_valid_for_arch): New functions.
80 (DEFAULT_CRIS_ARCH): New macro, default to cris_any_v0_v10.
81 (cris_arch): New variable.
82 (md_pseudo_table): New pseudo .arch.
83 (err_for_dangerous_mul_placement): Initialize according to
84 DEFAULT_CRIS_ARCH.
85 (STATE_COND_BRANCH): Renamed from STATE_CONDITIONAL_BRANCH.
86 All users changed.
87 (STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON)
88 (STATE_ABS_BRANCH_V32, STATE_LAPC, BRANCH_BF_V32, BRANCH_BB_V32)
89 (BRANCH_WF_V32, BRANCH_WB_V32): New.
90 (BRANCH_BF, BRANCH_BB, BRANCH_WF, BRANCH_WB): Don't undef after
91 use in md_cris_relax_table.
92 (md_cris_relax_table): Add entries for STATE_COND_BRANCH_V32,
93 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
94 Update and improve head comment.
95 (OPTION_PIC): Define in terms of previous option, OPTION_US.
96 (OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): Similar.
97 (OPTION_ARCH): New.
98 (md_longopts): New option --march=...
99 (cris_any_v0_v10_long_jump_size, crisv32_long_jump_size): New
100 macros.
101 (md_long_jump_size): Initialize in terms of DEFAULT_CRIS_ARCH.
102 (HANDLE_RELAXABLE): New macro.
103 (md_estimate_size_before_relax): Use HANDLE_RELAXABLE for common
104 cases. Check for weak symbols and assume not relaxable. Handle
105 STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON,
106 STATE_ABS_BRANCH_V32, STATE_LAPC. Use new variable symbolP, not
107 fragP->fr_symbol.
108 (md_convert_frag): Handle STATE_COND_BRANCH_V32,
109 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
110 (cris_create_short_jump): Adjust for CRISv32.
5664e9a7 111 (cris_relax_frag): Handle new states.
ae57792d
HPN		 112 (md_create_long_jump): Ditto. Emit error for common_v10_v32.
113 (md_begin): Define symbols "..asm.arch.cris.v32",
114 "..asm.arch.cris.v10", "..asm.arch.cris.common_v10_v32" and
115 "..asm.arch.cris.any_v0_v10". Use cris_insn_ver_valid_for_arch
116 when entering opcode table entry points.
117 (md_assemble): Adjust branch handling for CRISv32. Handle LAPC
118 relaxation. In fix_new_exp call for main insn, pass 1 for pcrel
119 parameter for 8, 16 and 32-bit pc-relative insns and LAPC.
120 (cris_process_instruction): Initialize out_insnp->insn_type to
121 CRIS_INSN_NONE, not CRIS_INSN_NORMAL.
122 <case ']', '[', 'A', 'd', 'Q', 'N', 'n', 'Y', 'U', 'u', 'T'>: New
123 cases.
124 <case 'm'>: Check that modified_char == '.'.
125 <invalid operands>: Consume the rest of the line.
126 When operands don't match, skip over subsequent insns with
127 non-matching version specifier but same mnemonic.
128 <immediate constant, case SIZE_SPEC_REG>: Immediate operands for
129 special registers in CRISv32 are always 32 bit long.
130 <immediate constant, case SIZE_FIELD_SIGNED, SIZE_FIELD_UNSIGNED>:
131 New cases.
132 (get_gen_reg): Only recognize "PC" when followed by "+]" for v32
133 and compatible. Recognize "ACR" for v32, unless followed by "+".
134 (get_spec_reg): Consider cris_arch when looking up register.
135 (get_autoinc_prefix_or_indir_op): Don't recognize assignment for
136 v32 or compatible.
137 (get_3op_or_dip_prefix_op): Check for ']' after seeing '[rN+'.
138 (cris_get_expression): Restore input_line_pointer if failing "early".
139 (get_flags): Consider cris_arch and recognize flags accordingly.
140 (branch_disp): Adjust for CRISv32.
141 (gen_cond_branch_32): Similar. Emit error for common_v10_v32.
142 (cris_number_to_imm): Use as_bad_where, not as_bad. Remove
143 related FIXME. Don't insist on BFD_RELOC_32_PCREL fixup to be
144 resolved. Don't enter zeros in object file for
145 BFD_RELOC_32_PCREL.
146 <case BFD_RELOC_CRIS_LAPCQ_OFFSET, BFD_RELOC_CRIS_SIGNED_16>
147 <case BFD_RELOC_CRIS_SIGNED_8>: New case.
148 (md_parse_option): Break out "return 1".
149 <OPTION_ARCH> New case.
150 (tc_gen_reloc): <case BFD_RELOC_CRIS_LAPCQ_OFFSET>
151 <case BFD_RELOC_CRIS_SIGNED_16, BFD_RELOC_CRIS_SIGNED_8>
152 <case BFD_RELOC_CRIS_UNSIGNED_8, BFD_RELOC_CRIS_UNSIGNED_16>
153 <case BFD_RELOC_32_PCREL>: New cases.
154 Addends for non-zero fx_pcrel are too in fx_offset.
155 (md_show_usage): Show --march=<arch>.
156 (md_apply_fix3): Adjust val for BFD_RELOC_CRIS_LAPCQ_OFFSET.
157 (md_pcrel_from): BFD_RELOC_CRIS_LAPCQ_OFFSET is PC-relative too.
158 (s_syntax) <struct syntaxes>: Properly constify member operand.
159 * config/tc-cris.h (TARGET_MACH): Define.
160 (cris_mach): Declare.
161 * doc/as.texinfo (Overview) <CRIS>: Add --march=...
162 * doc/c-cris.texi (CRIS-Symbols): New node for built-in symbols.
163 (CRIS-Opts): Document --march=...
164 (CRIS-Pseudos): Document .arch.
165
9306ca4a
JB		 1662004-11-04 Jan Beulich <jbeulich@novell.com>
167
168 * config/tc-i386.c (set_intel_syntax): Allow % in symbol names when
169 intel syntax and no register prefix, allow $ in symbol names when
170 intel syntax.
171 (set_16bit_gcc_code_flag): Replace literal 'l' by LONG_MNEM_SUFFIX.
172 (intel_float_operand): Add fourth return value indicating math control
173 operations. Make classification more precise.
174 (md_assemble): Complain if memory operand of mov[sz]x has no size
175 specified.
176 (parse_insn): Translate word operands to floating point instructions
177 operating on integers as well as control instructions to short ones
178 as expected by AT&T syntax. Translate 'd' suffix to short one only for
179 floating point instructions operating on non-integer operands.
180 (match_template): Remove fldcw special case. Adjust q-suffix handling
181 to permit it on fild/fistp/fisttp in AT&T mode.
182 (process_suffix): Don't guess DefaultSize insns' suffix from
183 stackop_size for certain floating point control instructions. Guess
184 suffix for branch and [ls][gi]dt based on flag_code. Split error
185 messages for Intel and AT&T syntax, and make the condition more strict
186 for the former. Adjust suppressing of generation of operand size
187 overrides.
188 (intel parser): Allow the full set of MASM operators. Add FWORD, TBYTE,
189 OWORD, and XMMWORD operand size specifiers (TBYTE replaces XWORD). Add
190 more error checking.
191 * config/tc-i386.h (BYTE_PTR WORD_PTR DWORD_PTR QWORD_PTR XWORD_PTR
192 SHORT OFFSET_FLAT FLAT NONE_FOUND): Remove unused defines.
193
2b4f075a
HPN		 1942004-11-03 Hans-Peter Nilsson <hp@axis.com>
195
196 * symbols.c (colon) [!WORKING_DOT_WORD]: Don't declare
197 md_short_jump_size, md_long_jump_size.
198 * write.c [!WORKING_DOT_WORD]: Ditto.
199 * tc.h [!WORKING_DOT_WORD]: Declare them here. Drop const
200 qualifier.
201 * config/tc-cris.h (md_short_jump_size, md_long_jump_size): Don't
202 declare.
203 * config/tc-cris.c (md_short_jump_size, md_long_jump_size): Drop
204 const qualifier in these definitions.
205 * config/tc-i370.c, config/tc-m68k.c, config/tc-pdp11.c,
206 config/tc-s390.c, config/tc-tahoe.c, config/tc-vax.c: Ditto.
207
ddbc47ac
NC		 2082004-11-02 Nick Clifton <nickc@redhat.com>
209
210 * dwarf2dbg.c (dwarf2_finish): Check for the existence of a file
211 table before deciding to produce a .debug_line section to match up
212 with a user provided .debug_info section.
213
3ad3f5ad
TL		 2142004-10-28 Tomer Levi <Tomer.Levi@nsc.com>
215
216 * config/tc-crx.c (getreg_image): Bug fix, a return value was
217 mistakenly omitted from CRX_C_REGTYPE and CRX_CS_REGTYPE cases.
218
c815a616
TL		 2192004-10-27 Tomer Levi <Tomer.Levi@nsc.com>
220
221 * config/tc-crx.c: Remove global variable 'post_inc_mode'.
222 (get_flags): New function.
223 (get_number_of_bits): Edit comments, update numeric values to supported
224 sizes.
225 (process_label_constant): Don't support the colon format (SYMBOL:[s|m|l]).
226 (set_cons_rparams): Support argument type 'arg_rbase'.
227 (get_operandtype): Bug fix in 'rbase' operand type parsing.
228 (handle_LoadStor): Bug fix, first handle post-increment mode.
229 (getreg_image): Remove redundant code, update according to latest CRX spec.
230 (print_constant): Bug fix relate to 3-word instructions.
231 (assemble_insn): Bug fix, when matching instructions, verify also
232 instruction type (not only mnemonic).
233 Add various error checking.
234 (preprocess_reglist): Support HI/LO and user registers.
235
ed7af9f9
L		 2362004-10-25 David Mosberger-Tang <davidm@hpl.hp.com>
237
238 * config/tc-ia64.c (fixup_unw_records): Don't let the "t" value
239 in an epilogue directive go negative.
240
ac025970
L		 2412004-10-25 H.J. Lu <hongjiu.lu@intel.com>
242
243 PR 474
244 * config/tc-ia64.c (emit_one_bundle): Decrement md.num_slots_in_use
245 after reporting template error during manual bundling. Reported
246 by Michael Dupont, michaelx.dupont@intel.com.
247
a394c00f
NC		 2482004-10-25 Daniel Jacobowitz <dan@debian.org>
249
250 * Makefile.am: Run dep-am.
251 * aclocal.m4: Regenerate with automake 1.9.2.
252 * Makefile.in: Regenerate with automake 1.9.2.
253 * doc/Makefile.in: Likewise.
254
255 * config/tc-arm.c: Include "dw2gencfi.h".
256 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
257 New functions.
258 * config/tc-arm.h (TARGET_USE_CFIPOP, DWARF2_DEFAULT_RETURN_COLUMN)
259 (DWARF2_CIE_DATA_ALIGNMENT, tc_regname_to_dw2regnum)
260 (tc_cfi_frame_initial_instructions): Define.
261 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
262 Add prototypes.
263
23d83918
TL		 2642004-10-21 Tomer Levi <Tomer.Levi@nsc.com>
265
266 * config/tc-crx.c (assemble_insn): Check unsigned immediate
267 operands validity.
268 Update coprocessor id to be unsigned immediate.
269
bb41ade5
AM		 2702004-10-18 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
271
272 * config/tc-i386.c (O_secrel): Delete.
273 (tc_pe_dwarf2_emit_offset): New function.
274 * config/tc-i386.h (O_secrel): Define as O_md1.
275 (TC_DWARF2_EMIT_OFFSET): Define.
276
87e62348
NC		 2772004-10-18 Nick Clifton <nickc@redhat.com>
278
279 * config/tc-xstormy16.c (xstormy16_cons_fix_new): Accept and
280 ignore @fptr() directives for 4-byte fixups.
281
3c55da70
AM		 2822004-10-15 Alan Modra <amodra@bigpond.net.au>
283
284 * Makefile.am: Run "make dep-am"
285 * Makefile.in: Regenerate.
286
40e7b7b9
BW		 2872004-10-14 Bob Wilson <bob.wilson@acm.org>
288
289 * doc/c-xtensa.texi (Xtensa Options, Absolute Literals Directive):
290 Remove comments about placement of literal pools.
291 (Literal Directive): Update description of literal placement.
292 (Literal Prefix Directive): Remove statement that this does not apply
293 to absolute-mode literals. Describe new section naming scheme.
294
b08b5071
BW		 2952004-10-12 Bob Wilson <bob.wilson@acm.org>
296
297 * config/tc-xtensa.c (xtensa_frequency_pseudo): Use set_subseg_freq.
298 (is_entry_opcode, is_movi_opcode, is_the_loop_opcode, is_jx_opcode,
299 is_windowed_return_opcode): Delete.
300 (xtensa_frob_label): Use get_subseg_target_freq.
301 (md_assemble): Inline call to is_entry_opcode.
302 (xtensa_handle_align): Inline call to get_frag_is_literal.
303 (relaxation_requirements): Inline call to is_jx_opcode.
304 (emit_single_op): Inline call to is_movi_opcode.
305 (xg_assemble_vliw_tokens): Inline calls to get_frag_is_insn,
306 get_frag_is_no_transform, is_entry_opcode, and
307 set_frag_is_specific_opcode. Use get_subseg_total_freq.
308 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags,
309 xtensa_fix_close_loop_end_frags, relax_frag_immed, convert_frag_immed):
310 Inline calls to get_frag_is_no_transform.
311 (next_instrs_are_b_retw): Inline call to is_windowed_return_opcode.
312 (xtensa_fix_short_loop_frags): Inline calls to is_the_loop_opcode and
313 get_frag_is_no_transform.
314 (convert_frag_immed_finish_loop): Inline calls to get_expression_value
315 and set_frag_is_no_transform.
316 (get_expression_value): Delete.
317 (subseg_map struct): Rename cur_total_freq to total_freq. Rename
318 cur_target_freq to target_freq.
319 (get_subseg_info): Split out code to create a new map entry into ...
320 (add_subseg_info): ... this new function.
321 (get_last_insn_flags): Check if get_subseg_info succeeded.
322 (set_last_insn_flags): Call add_subseg_info if needed.
323 (get_subseg_total_freq, get_subseg_target_freq, set_subseg_freq): New.
324 (xtensa_reorder_segments): Compute last_sec while counting sections.
325 Remove call to get_last_sec.
326 (get_last_sec): Delete.
327 (cache_literal_section): Inline call to retrieve_literal_seg and its
328 callees, seg_present and add_seg_list.
329 (retrieve_literal_seg, seg_present, add_seg_list): Delete.
330 (get_frag_is_insn, get_frag_is_no_transform,
331 set_frag_is_specific_opcode, set_frag_is_no_transform): Delete.
332 * config/tc-xtensa.h (MAX_SLOTS): Reduce from 31 to 15.
333
7fa3d080
BW		 3342004-10-12 Bob Wilson <bob.wilson@acm.org>
335
336 * config/tc-xtensa.c: Use ISO C90 formatting.
337 * config/tc-xtensa.h: Likewise.
338 * config/xtensa-istack.h: Likewise.
339 * config/xtensa-relax.c: Likewise.
340 * config/xtensa-relax.h: Likewise.
341
8cb51566
PB		 3422004-10-12 Paul Brook <paul@codesourcery.com>
343
344 * config/tc-arm.c (md_begin): Change EF_ARM_EABI_VER3 to
345 EF_ARM_EABI_VER4.
346 (arm_eabis): Ditto.
347 * doc/c-arm.texi: Document that we actually support -meabi=4, not
348 -meabi=3.
349
653cfe85
BW		 3502004-10-08 Bob Wilson <bob.wilson@acm.org>
351
352 * doc/as.texinfo (VTableEntry, VTableInherit): Add "directive" to index
353 entries.
354 (Acknowledgements): Use "GAS" instead of AS variable.
355
d2b2c203
DJ		 3562004-10-08 Daniel Jacobowitz <dan@debian.org>
357
358 * config/tc-i386.c: Include "elf/x86-64.h".
359 (i386_elf_section_type): New function.
360 * config/tc-i386.h (md_elf_section_type): Define.
361 (i386_elf_section_type): New prototype.
362
e80e0390
NC		 3632004-10-08 Linus Nielsen Feltzing <linus@haxx.se>
364
365 * config/m68k-parse.h (enum m68k_register): New control register,
366 MBAR2 (for MCF5249)
367 * config/tc-m68k.c: Correct control register set for MCF5249.
368
43cd72b9 3692004-10-07 Bob Wilson <bob.wilson@acm.org>
85a5ac80 370 Sterling Augustine <sterling@tensilica.com>
43cd72b9
BW		 371
372 * config/tc-xtensa.c (absolute_literals_supported): New global flag.
373 (UNREACHABLE_MAX_WIDTH): Define.
374 (XTENSA_FETCH_WIDTH): Delete.
375 (cur_vinsn, xtensa_fetch_width, xt_saved_debug_type, past_xtensa_end,
376 prefer_const16, prefer_l32r): New global variables.
377 (LIT4_SECTION_NAME): Define.
378 (lit4_state struct): Add lit4_seg_name and lit4_seg fields.
379 (XTENSA_PROP_*, GET_XTENSA_PROP_*, SET_XTENSA_PROP_*): Define.
380 (frag_flags struct): New.
381 (xtensa_block_info struct): Move from tc-xtensa.h. Add flags field.
382 (subseg_map struct): Add cur_total_freq and cur_target_freq fields.
383 (bitfield, bit_is_set, set_bit, clear_bit): Define.
384 (MAX_FORMATS): Define.
385 (op_placement_info struct, op_placement_table): New.
386 (O_pltrel, O_hi16, O_lo16): Define.
387 (directiveE enum): Rename directive_generics to directive_transform.
388 Delete directive_relax. Add directive_schedule,
389 directive_absolute_literals, and directive_last_directive.
390 (directive_info): Rename "generics" to "transform". Delete "relax".
391 Add "schedule" and "absolute-literals".
392 (directive_state): Adjust entries to match changes in directive_info.
393 (xtensa_relax_statesE, RELAX_IMMED_MAXSTEPS): Move to tc-xtensa.h.
394 (xtensa_const16_opcode, xtensa_movi_opcode, xtensa_movi_n_opcode,
395 xtensa_l32r_opcode, xtensa_nop_opcode, xtensa_rsr_lcount_opcode): New.
396 (xtensa_j_opcode, xtensa_rsr_opcode): Delete.
397 (align_only_targets, software_a0_b_retw_interlock,
398 software_avoid_b_j_loop_end, maybe_has_b_j_loop_end,
399 software_avoid_short_loop, software_avoid_close_loop_end,
400 software_avoid_all_short_loops, specific_opcode): Delete.
401 (warn_unaligned_branch_targets): New.
402 (workaround_a0_b_retw, workaround_b_j_loop_end, workaround_short_loop,
403 workaround_close_loop_end, workaround_all_short_loops): Default FALSE.
404 (option_[no_]link_relax, option_[no_]transform,
405 option_[no_]absolute_literals, option_warn_unaligned_targets,
406 option_prefer_l32r, option_prefer_const16, option_target_hardware):
407 New enum values.
408 (option_[no_]align_only_targets, option_literal_section_name,
409 option_text_section_name, option_data_section_name,
410 option_bss_section_name, option_eb, option_el): Delete.
411 (md_longopts): Add entries for: [no-]transform, [no-]absolute-literals,
412 warn-unaligned-targets, prefer-l32r, prefer-const16, [no-]link-relax,
413 and target-hardware. Delete entries for [no-]target-align-only,
414 literal-section-name, text-section-name, data-section-name, and
415 bss-section-name.
416 (md_parse_option): Handle new options and remove old ones. Accept but
417 ignore [no-]density options. Warn for [no-]generics and [no-]relax
418 and treat them as [no-]transform.
419 (md_show_usage): Add new options and remove old ones.
420 (xtensa_setup_hw_workarounds): New.
421 (md_pseudo_table): Change "word" entry to use xtensa_elf_cons. Add
422 "long", "short", "loc" and "frequency" entries.
423 (use_generics): Rename to ...
424 (use_transform): ... this function. Add past_xtensa_end check.
425 (use_longcalls): Add past_xtensa_end check.
426 (code_density_available, can_relax): Delete.
427 (do_align_targets): New.
428 (get_directive): Accept dashes in directive names. Warn about
429 [no-]generics and [no-]relax directives and treat them as
430 [no-]transform.
431 (xtensa_begin_directive): Call md_flush_pending_output only for some
432 directives. Check for directives inside instruction bundles. Warn
433 about deprecated ".begin literal" usage. Warn and ignore [no-]density
434 directives. Handle new directives. Check generating_literals flag
435 for literal_prefix.
436 (xtensa_end_directive): Check for directives inside instruction
437 bundles. Warn and ignore [no-]density directives. Handle new
438 directives. Call xtensa_set_frag_assembly_state.
439 (xtensa_loc_directive_seen, xtensa_dwarf2_directive_loc,
440 xtensa_dwarf2_emit_insn): New.
441 (xtensa_literal_position): Call md_flush_pending_output. Do not check
442 use_literal_section flag.
443 (xtensa_literal_pseudo): Call md_flush_pending_output. Handle absolute
444 literals. Use xtensa_elf_cons to parse the expression.
445 (xtensa_literal_prefix): Do not check use_literal_section. Support
446 ".lit4" sections for absolute literals. Change prefix convention to
447 replace ".text" (or ".t" in a linkonce section). No need to call
448 subseg_set.
449 (xtensa_frequency_pseudo, xtensa_elf_cons, xtensa_elf_suffix): New.
450 (expression_end): Handle closing braces and colons.
451 (PLT_SUFFIX, plt_suffix): Delete.
452 (expression_maybe_register): Use new xtensa-isa.h functions. Use
453 xtensa_elf_suffix instead of checking for plt suffix, and handle O_lo16
454 and O_hi16 expressions as well.
455 (tokenize_arguments): Handle closing braces and colons.
456 (parse_arguments): Use new xtensa-isa.h functions. Handle "invisible"
457 operands and paired register syntax.
458 (get_invisible_operands): New.
459 (xg_translate_sysreg_op): Handle new Xtensa LX RSR/WSR/XSR syntax. Use
460 new xtensa-isa.h functions.
461 (xtensa_translate_old_userreg_ops, xtensa_translate_zero_immed): New.
462 (xg_translate_idioms): Check if inside bundle. Use use_transform.
463 Handle new Xtensa LX RSR/WSR/XSR syntax. Remove code to widen density
464 instructions. Use xtensa_translate_zero_immed.
465 (operand_is_immed, operand_is_pcrel_label): Delete.
466 (get_relaxable_immed): Use new xtensa-isa.h functions.
467 (get_opcode_from_buf): Add slot parameter. Use new xtensa-isa.h
468 functions.
469 (xtensa_print_insn_table, print_vliw_insn): New.
470 (is_direct_call_opcode): Use new xtensa-isa.h functions.
471 (is_call_opcode, is_loop_opcode, is_conditional_branch_opcode,
472 is_branch_or_jump_opcode): Delete.
473 (is_movi_opcode, decode_reloc, encode_reloc, encode_alt_reloc): New.
474 (opnum_to_reloc, reloc_to_opnum): Delete.
475 (xtensa_insnbuf_set_operand, xtensa_insnbuf_get_operand): Use new
476 xtensa-isa.h functions. Operate on one slot of an instruction.
477 (xtensa_insnbuf_set_immediate_field, is_negatable_branch,
478 xg_get_insn_size): Delete.
479 (xg_get_build_instr_size): Use xg_get_single_size.
480 (xg_is_narrow_insn, xg_is_single_relaxable_insn): Update calls to
481 xg_build_widen_table. Use xg_get_single_size.
482 (xg_get_max_narrow_insn_size): Delete.
483 (xg_get_max_insn_widen_size, xg_get_max_insn_widen_literal_size,
484 xg_is_relaxable_insn): Update calls to xg_build_widen_table. Use
485 xg_get_single_size.
486 (xg_build_to_insn): Record the loc field. Handle OP_OPERAND_HI16U and
487 OP_OPERAND_LOW16U. Check xg_valid_literal_expression.
488 (xg_expand_to_stack, xg_expand_narrow): Update calls to
489 xg_build_widen_table. Use xg_get_single_size.
490 (xg_immeds_fit): Use new xtensa-isa.h functions. Update call to
491 xg_check_operand.
492 (xg_symbolic_immeds_fit): Likewise. Also handle O_lo16 and O_hi16, and
493 treat weak symbols conservatively.
494 (xg_check_operand): Use new xtensa-isa.h functions.
495 (is_dnrange): Delete.
496 (xg_assembly_relax): Inline previous calls to tinsn_copy.
497 (xg_finish_frag): Specify separate relax states for the frag and slot0.
498 (is_branch_jmp_to_next, xg_add_branch_and_loop_targets): Use new
499 xtensa-isa.h functions.
500 (xg_instruction_matches_option_term, xg_instruction_matches_or_options,
501 xg_instruction_matches_options): New.
502 (xg_instruction_matches_rule): Handle O_register expressions. Call
503 xg_instruction_matches_options.
504 (transition_rule_cmp): New.
505 (xg_instruction_match): Update call to xg_build_simplify_table.
506 (xg_build_token_insn): Record loc fields.
507 (xg_simplify_insn): Check is_specific_opcode field and
508 density_supported flag.
509 (xg_expand_assembly_insn): Skip checking code_density_available. Use
510 new xtensa-isa.h functions. Call use_transform instead of can_relax.
511 (xg_assemble_literal): Add error handling for O_big. Call
512 record_alignment. Handle O_pltrel.
513 (xg_valid_literal_expression): New.
514 (xg_assemble_literal_space): Add slot parameter. Remove call to
515 set_expr_symbol_offset. Add call to record_alignment. Update call to
516 xg_finish_frag.
517 (xg_emit_insn): Delete.
518 (xg_emit_insn_to_buf): Add format parameter. Update calls to
519 xg_add_opcode_fix and xtensa_insnbuf_to_chars.
520 (xg_add_opcode_fix): Change opcode parameter to tinsn and add format
521 and slot parameters. Handle new "alternate" relocations for absolute
522 literals and CONST16 instructions. Check for bad uses of O_lo16 and
523 O_hi16. Use new xtensa-isa.h functions.
524 (xg_assemble_tokens): Delete.
525 (is_register_writer): Use new xtensa-isa.h functions.
526 (is_bad_loopend_opcode): Check for xtensa_rsr_lcount_opcode instead of
527 old-style RSR from LCOUNT.
528 (next_frag_opcode): Delete.
529 (next_frag_opcode_is_loop, next_frag_format_size, frag_format_size,
530 update_next_frag_state): New.
531 (update_next_frag_nop_state): Delete.
532 (next_frag_pre_opcode_bytes): Use next_frag_opcode_is_loop.
533 (xtensa_mark_literal_pool_location): Check use_literal_section flag and
534 the state of the absolute-literals directive. Add calls to
535 record_alignment and xtensa_set_frag_assembly_state. Call
536 xtensa_switch_to_non_abs_literal_fragment instead of
537 xtensa_switch_to_literal_fragment.
538 (build_nop): New.
539 (assemble_nop): Use build_nop. Update call to xtensa_insnbuf_to_chars.
540 (get_expanded_loop_offset): Change check for undefined opcode to an
541 assertion.
542 (xtensa_set_frag_assembly_state, relaxable_section,
543 xtensa_find_unmarked_state_frags, xtensa_find_unaligned_branch_targets,
544 xtensa_find_unaligned_loops, xg_apply_tentative_value): New.
545 (md_begin): Update call to xtensa_isa_init. Initialize linkrelax to 1.
546 Set lit4_seg_name. Call xg_init_vinsn. Initialize new global opcodes.
547 Call init_op_placement_info_table and xtensa_set_frag_assembly_state.
548 (xtensa_init_fix_data): New.
549 (xtensa_frob_label): Reset label symbol to the current frag. Check
550 do_align_targets and generating_literals flag. Propagate frequency
551 info to new alignment frag. Call xtensa_set_frag_assembly_state.
552 (xtensa_unrecognized_line): New.
553 (xtensa_flush_pending_output): Check if inside a bundle. Add a call
554 to xtensa_set_frag_assembly_state.
555 (error_reset_cur_vinsn): New.
556 (md_assemble): Remove check for literal frag. Remove call to
557 istack_init. Call use_transform instead of use_generics. Parse
558 explicit instruction format specifiers. Move code for
559 a0_b_retw_interlock workaround to xg_assemble_vliw_tokens. Call
560 error_reset_cur_vinsn on errors. Add call to get_invisible_operands.
561 Add dwarf2_where call. Remote automatic alignment for ENTRY
562 instructions. Move call to xtensa_clear_insn_labels to the end.
563 Rearrange to handle bundles.
564 (xtensa_cons_fix_new): Delete.
565 (xtensa_handle_align): New.
566 (xtensa_frag_init): Call xtensa_set_frag_assembly_state. Remove
567 assignment to is_no_density field.
568 (md_pcrel_from): Use new xtensa-isa.h functions. Use decode_reloc
569 instead of reloc_to_opnum. Handle "alternate" relocations.
570 (xtensa_force_relocation, xtensa_check_inside_bundle,
571 xtensa_elf_section_change_hook): New.
572 (xtensa_symbol_new_hook): Delete.
573 (xtensa_fix_adjustable): Check for difference of symbols with an
574 offset. Check for external and weak symbols.
575 (md_apply_fix3): Remove cases for XTENSA_OP{0,1,2} relocs.
576 (md_estimate_size_before_relax): Return expansion for the first slot.
577 (tc_gen_reloc): Handle difference of symbols by producing
578 XTENSA_DIFF{8,16,32} relocs and by writing the value of the difference
579 into the output. Handle new XTENSA_SLOT*_OP relocs by storing the
580 tentative values into the output when linkrelax is set.
581 (XTENSA_PROP_SEC_NAME): Define.
582 (xtensa_post_relax_hook): Call xtensa_find_unmarked_state_frags.
583 Create literal tables only if using literal sections. Create new
584 property tables instead of old instruction tables. Check for unaligned
585 branch targets and loops.
586 (finish_vinsn, find_vinsn_conflicts, check_t1_t2_reads_and_writes,
587 new_resource_table, clear_resource_table, resize_resource_table,
588 resources_available, reserve_resources, release_resources,
589 opcode_funcUnit_use_unit, opcode_funcUnit_use_stage,
590 resources_conflict, xg_find_narrowest_format, relaxation_requirements,
591 bundle_single_op, emit_single_op, xg_assemble_vliw_tokens): New.
592 (xtensa_end): Call xtensa_flush_pending_output. Set past_xtensa_end
593 flag. Update checks for workaround options. Call
594 xtensa_mark_narrow_branches and xtensa_mark_zcl_first_insns.
595 (xtensa_cleanup_align_frags): Add special case for branch targets.
596 Check for and mark unreachable frags.
597 (xtensa_fix_target_frags): Remove use of align_only_targets flag.
598 Use RELAX_LOOP_END_BYTES in special case for negatable branch at the
599 end of a zero-overhead loop body.
600 (frag_can_negate_branch): Handle instructions with multiple slots.
601 Use new xtensa-isa.h functions
602 (xtensa_mark_narrow_branches, is_narrow_branch_guaranteed_in_range,
603 xtensa_mark_zcl_first_insns): New.
604 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags): Error if
605 transformations are disabled.
606 (next_instrs_are_b_retw): Use new xtensa-isa.h functions. Handle
607 multislot instructions.
608 (xtensa_fix_close_loop_end_frags, xtensa_fix_short_loop_frags):
609 Likewise. Also error if transformations are disabled.
610 (unrelaxed_frag_max_size): New.
611 (unrelaxed_frag_min_insn_count, unrelax_frag_has_b_j): Use new
612 xtensa-isa.h functions.
613 (xtensa_sanity_check, is_empty_loop, is_local_forward_loop): Use
614 xtensa_opcode_is_loop instead of is_loop_opcode.
615 (get_text_align_power): Replace as_fatal with assertion.
616 (get_text_align_fill_size): Iterate instead of using modulus when
617 use_nops is false.
618 (get_noop_aligned_address): Assert that this is for a machine-dependent
619 RELAX_ALIGN_NEXT_OPCODE frag. Use next_frag_opcode_is_loop,
620 xg_get_single_size, and frag_format_size.
621 (get_widen_aligned_address): Rename to ...
622 (get_aligned_diff): ... this function. Add max_diff parameter.
623 Remove handling of rs_align/rs_align_code frags. Use
624 next_frag_format_size, get_text_align_power, get_text_align_fill_size,
625 next_frag_opcode_is_loop, and xg_get_single_size. Compute max_diff
626 and pass it back to caller.
627 (xtensa_relax_frag): Use relax_frag_loop_align. Add code for new
628 RELAX_SLOTS, RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN,
629 RELAX_FILL_NOP, and RELAX_UNREACHABLE frag types. Check relax_seen.
630 (relax_frag_text_align): Rename to ...
631 (relax_frag_loop_align): ... this function. Assume loops can only be
632 in the first slot of an instruction.
633 (relax_frag_add_nop): Use assemble_nop instead of constructing an OR
634 instruction. Remove call to frag_wane.
635 (relax_frag_narrow): Rename to ...
636 (relax_frag_for_align): ... this function. Extend to handle
637 RELAX_FILL_NOP and RELAX_UNREACHABLE, as well as RELAX_SLOTS with
638 RELAX_NARROW for the first slot.
639 (find_address_of_next_align_frag, bytes_to_stretch): New.
640 (future_alignment_required): Use find_address_of_next_align_frag and
641 bytes_to_stretch. Look ahead to subsequent frags to make smarter
642 alignment decisions.
643 (relax_frag_immed): Add format, slot, and estimate_only parameters.
644 Check if transformations are enabled for b_j_loop_end workaround.
645 Use new xtensa-isa.h functions and handle multislot instructions.
646 Update call to xg_assembly_relax.
647 (md_convert_frag): Handle new RELAX_SLOTS, RELAX_UNREACHABLE,
648 RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN, and RELAX_FILL_NOP
649 frag types.
650 (convert_frag_narrow): Add segP, format and slot parameters. Call
651 convert_frag_immed for branch instructions. Adjust calls to
652 tinsn_from_chars, tinsn_immed_from_frag, and xg_emit_insn_to_buf. Use
653 xg_get_single_size and xg_get_single_format.
654 (convert_frag_fill_nop): New.
655 (convert_frag_immed): Add format and slot parameters. Handle multislot
656 instructions and use new xtensa-isa.h functions. Update calls to
657 tinsn_immed_from_frag and xg_assembly_relax. Check if transformations
658 enabled for b_j_loop_end workaround. Use build_nop instead of
659 assemble_nop. Check is_specific_opcode flag. Check for unreachable
660 frags. Use xg_get_single_size. Handle O_pltrel.
661 (fix_new_exp_in_seg): Remove check for old plt flag.
662 (convert_frag_immed_finish_loop): Update calls to tinsn_from_chars and
663 xtensa_insnbuf_to_chars. Call tinsn_immed_from_frag. Change check
664 for loop opcode to an assertion. Mark all frags up to the end of the
665 loop as not transformable.
666 (get_last_insn_flags, set_last_insn_flags): Use get_subseg_info.
667 (get_subseg_info): New.
668 (xtensa_move_literals): Call xtensa_set_frag_assembly_state. Add null
669 check for dest_seg.
670 (xtensa_switch_to_literal_fragment): Rewrite to handle absolute
671 literals and use xtensa_switch_to_non_abs_literal_fragment otherwise.
672 (xtensa_switch_to_non_abs_literal_fragment): New.
673 (cache_literal_section): Add is_code parameter and pass it through to
674 retrieve_literal_seg.
675 (retrieve_literal_seg): Add is_code parameter and use it to set the
676 flags on the literal section. Handle case where head parameter is 0.
677 (get_frag_is_no_transform, set_frag_is_specific_opcode,
678 set_frag_is_no_transform): New.
679 (xtensa_create_property_segments): Add end_property_function parameter
680 and pass it through to add_xt_block_frags. Call bfd_get_section_flags
681 and skip SEC_DEBUGGING and !SEC_ALLOC sections.
682 (xtensa_create_xproperty_segments, section_has_xproperty): New.
683 (add_xt_block_frags): Add end_property_function parameter and call it
684 if it is non-zero. Call xtensa_frag_flags_init.
685 (xtensa_frag_flags_is_empty, xtensa_frag_flags_init,
686 get_frag_property_flags, frag_flags_to_number,
687 xtensa_frag_flags_combinable, xt_block_aligned_size,
688 xtensa_xt_block_combine, add_xt_prop_frags,
689 init_op_placement_info_table, opcode_fits_format_slot,
690 xg_get_single_size, xg_get_single_format): New.
691 (istack_push): Inline call to tinsn_copy.
692 (tinsn_copy): Delete.
693 (tinsn_has_invalid_symbolic_operands): Handle O_hi16 and O_lo16 and
694 CONST16 opcodes. Handle O_big, O_illegal, and O_absent.
695 (tinsn_has_complex_operands): Handle O_hi16 and O_lo16.
696 (tinsn_to_insnbuf): Use xg_get_single_format and new xtensa-isa.h
697 functions. Handle invisible operands.
698 (tinsn_to_slotbuf): New.
699 (tinsn_check_arguments): Use new xtensa-isa.h functions.
700 (tinsn_from_chars): Add slot parameter. Rewrite using xg_init_vinsn,
701 vinsn_from_chars, and xg_free_vinsn.
702 (tinsn_from_insnbuf): New.
703 (tinsn_immed_from_frag): Add slot parameter and handle multislot
704 instructions. Handle symbol differences.
705 (get_num_stack_text_bytes): Use xg_get_single_size.
706 (xg_init_vinsn, xg_clear_vinsn, vinsn_has_specific_opcodes,
707 xg_free_vinsn, vinsn_to_insnbuf, vinsn_from_chars, expr_is_register,
708 get_expr_register, set_expr_symbol_offset_diff): New.
709 * config/tc-xtensa.h (MAX_SLOTS): Define.
710 (xtensa_relax_statesE): Move from tc-xtensa.c. Add
711 RELAX_CHECK_ALIGN_NEXT_OPCODE, RELAX_MAYBE_DESIRE_ALIGN, RELAX_SLOTS,
712 RELAX_FILL_NOP, RELAX_UNREACHABLE, RELAX_MAYBE_UNREACHABLE, and
713 RELAX_NONE types.
714 (RELAX_IMMED_MAXSTEPS): Move from tc-xtensa.c.
715 (xtensa_frag_type struct): Add is_assembly_state_set,
716 use_absolute_literals, relax_seen, is_unreachable, is_specific_opcode,
717 is_align, is_text_align, alignment, and is_first_loop_insn fields.
718 Replace is_generics and is_relax fields by is_no_transform field.
719 Delete is_text and is_longcalls fields. Change text_expansion and
720 literal_expansion to arrays of MAX_SLOTS entries. Add arrays of
721 per-slot information: literal_frags, slot_subtypes, slot_symbols,
722 slot_sub_symbols, and slot_offsets. Add fr_prev field.
723 (xtensa_fix_data struct): New.
724 (xtensa_symfield_type struct): Delete plt field.
725 (xtensa_block_info struct): Move definition to tc-xtensa.h. Add
726 forward declaration here.
727 (xt_section_type enum): Delete xt_insn_sec. Add xt_prop_sec.
728 (XTENSA_SECTION_RENAME): Undefine.
729 (TC_FIX_TYPE, TC_INIT_FIX_DATA, TC_FORCE_RELOCATION, NO_PSEUDO_DOT,
730 tc_unrecognized_line, md_do_align, md_elf_section_change_hook,
731 HANDLE_ALIGN, TC_LINKRELAX_FIXUP, SUB_SEGMENT_ALIGN): Define.
732 (TC_CONS_FIX_NEW, tc_symbol_new_hook): Delete.
733 (unit_num_copies_func, opcode_num_units_func,
734 opcode_funcUnit_use_unit_func, opcode_funcUnit_use_stage_func): New.
735 (resource_table struct): New.
736 * config/xtensa-istack.h (MAX_INSN_ARGS): Increase from 6 to 10.
737 (TInsn struct): Add keep_wide, loc, fixup, record_fix, subtype,
738 literal_space, symbol, sub_symbol, offset, and literal_frag fields.
739 (tinsn_copy): Delete prototype.
740 (vliw_insn struct): New.
741 * config/xtensa-relax.c (insn_pattern_struct): Add options field.
742 (widen_spec_list): Add option conditions for density and boolean
743 instructions. Add expansions using CONST16 and conditions for using
744 CONST16 vs. L32R. Use new Xtensa LX RSR/WSR syntax. Add entries for
745 predicted branches.
746 (simplify_spec_list): Add option conditions for density instructions.
747 Add entry for NOP instruction.
748 (append_transition): Add cmp function pointer parameter and use it to
749 insert the new entry in order.
750 (operand_function_LOW16U, operand_function_HI16U): New.
751 (xg_has_userdef_op_fn, xg_apply_userdef_op_fn): Handle
752 OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
753 (enter_opname, split_string): Use xstrdup instead of strdup.
754 (init_insn_pattern): Initialize new options field.
755 (clear_req_or_option_list, clear_req_option_list,
756 clone_req_or_option_list, clone_req_option_list, parse_option_cond):
757 New.
758 (parse_insn_pattern): Parse option conditions.
759 (transition_applies): New.
760 (build_transition): Use new xtensa-isa.h functions. Fix incorrectly
761 swapped last arguments in calls to append_constant_value_condition.
762 Call clone_req_option_list. Add warning about invalid opcode.
763 Handle LOW16U and HI16U function names.
764 (build_transition_table): Add cmp parameter and use it in calls to
765 append_transition. Use new xtensa-isa.h functions. Check
766 transition_applies before adding entries.
767 (xg_build_widen_table, xg_build_simplify_table): Add cmp parameter and
768 pass it through to build_transition_table.
769 * config/xtensa-relax.h (ReqOrOptionList, ReqOrOption, ReqOptionList,
770 ReqOption, transition_cmp_fn): New types.
771 (OpType enum): Add OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
772 (transition_rule struct): Add options field.
773 * doc/as.texinfo (Overview): Update Xtensa options.
774 * doc/c-xtensa.texi (Xtensa Options): Delete --[no-]density,
775 --[no-]relax, and --[no-]generics options. Update descriptions of
776 --text-section-literals and --[no-]longcalls. Add
777 --[no-]absolute-literals and --[no-]transform.
778 (Xtensa Syntax): Add description of syntax for FLIX instructions.
779 Remove use of "generic" and "specific" terminology for opcodes.
780 (Xtensa Registers): Generalize the syntax description to include
781 user-defined register files.
782 (Xtensa Automatic Alignment): Update.
783 (Xtensa Branch Relaxation): Mention limitation of unconditional jumps.
784 (Xtensa Call Relaxation): Linker can now remove most of the overhead.
785 (Xtensa Directives): Remove confusing rules about precedence.
786 (Density Directive, Relax Directive): Delete.
787 (Schedule Directive): New.
788 (Generics Directive): Rename to ...
789 (Transform Directive): ... this node.
790 (Literal Directive): Update for absolute literals. Missing
791 literal_position directive is now an error.
792 (Literal Position Directive): Update for absolute literals.
793 (Freeregs Directive): Delete.
794 (Absolute Literals Directive): New.
795 (Frame Directive): Minor editing.
796 * Makefile.am (DEPTC_xtensa_elf, DEPOBJ_xtensa_elf, DEP_xtensa_elf):
797 Update dependencies.
798 * Makefile.in: Regenerate.
799
6e1304d8
RS		 8002004-10-07 Richard Sandiford <rsandifo@redhat.com>
801
802 * config/tc-mips.c (append_insn): Use fix_new rather than fix_new_exp
803 to build the second and third fixups for a composite relocation.
804 (macro_read_relocs): New function.
805 (macro_build): Use it.
806 (s_cpsetup): Pass all three composite relocation codes to macro_build.
807 Simplify fragging code accordingly.
808 (s_gpdword): Use fix_new rather than fix_new_exp for the second part
809 of the composite relocation. Set fx_tcbit in both fixups.
810
b1dca8ee
RS		 8112004-10-07 Richard Sandiford <rsandifo@redhat.com>
812
813 * config/tc-mips.c (append_insn): Set fx_tcbit for composite relocs.
814 (md_apply_fix3): Don't treat composite relocs as done.
815
89658e52
NC		 8162004-10-07 Jan Beulich <jbeulich@novell.com>
817
818 * macro.c (macro_expand_body): When ELF, use .LL rather than LL as
819 prefix for symbol names generated from the LOCAL macro directive.
820
289040ca
NC		 821 * dw2gencfi.c (select_cie_for_fde): When separating CIE out from
822 FDE, treat a DW_CFA_remember_state as we do a DW_CFA_advance_loc.
823
48c9f030
NC		 8242004-10-07 Tomer Levi <Tomer.Levi@nsc.com>
825
826 * config/tc-crx.c (preprocess_reglist): Handle Co-processor
827 Special registers.
5c5054ab 828 (md_assemble): Add error checking for Co-Processor instructions.
48c9f030
NC		 829 (get_cinv_parameters): Add 'b' option to invalidate the
830 branch-target cache.
831
7ed4c4c5
NC		 8322004-10-05 Paul Brook <paul@codesourcery.com>
833
834 * config/tc-arm.c (unwind): New variable.
835 (vfp_sp_encode_reg): New function.
836 (vfp_sp_reg_required_here): Use it.
837 (vfp_sp_reg_list, vfp_dp_reg_list): Remove.
838 (vfp_parse_reg_list): New function.
839 (s_arm_unwind_fnstart, s_arm_unwind_fnend, s_arm_unwind_cantunwind,
840 s_arm_unwind_personality, s_arm_unwind_personalityindex,
841 s_arm_unwind_handlerdata, s_arm_unwind_save, s_arm_unwind_movsp,
842 s_arm_unwind_pad, s_arm_unwind_setfp, s_arm_unwind_raw): New
843 functions.
844 (md_pseudo_table): Add them.
845 (do_vfp_reg2_from_sp2): Use vfp_parse_reg_list and vfp_sp_encode_reg.
846 (do_vfp_sp2_from_reg2, vfp_sp_ldstm, vfp_dp_ldstm): Ditto.
847 (set_section, add_unwind_adjustsp, flush_pending_unwind,
848 finish_unwind_opcodes, start_unwind_section, create_unwind_entry,
849 require_hashconst, add_unwind_opcode): New functions.
850 * doc/tc-arm.text: Document unwinding opcodes.
851 * NEWS: Mention the new feature.
852
04ca8703
EC		 8532004-10-04 Eric Christopher <echristo@redhat.com>
854
855 * config/tc-mips.c (md_apply_fix3): Remove erroneous assert.
856
7c1d0959
L		 8572004-10-01 H.J. Lu <hongjiu.lu@intel.com>
858
859 * config/tc-ppc.c (md_apply_fix3): Call S_SET_THREAD_LOCAL for
860 TLS relocations.
861 * config/tc-s390.c (md_apply_fix3): Likewise.
862 * config/tc-sparc.c (md_apply_fix3): Likewise.
863
40a18ebd
NC		 8642004-10-01 Paul Brook <paul@codesourcery.com>
865
866 * config/tc-arm.c (arm_elf_section_type): New function.
867 (arm_elf_change_section): Set section link for exidx sections.
868 * config/tc-arm.h (arm_elf_section_type): Add prototype.
869 (md_elf_section_type): Define.
870
14127cc4
NC		 8712004-10-01 Bill Farmer <Bill@the-farmers.freeserve.co.uk>
872
873 * config/tc-pdp11.c (md_apply_fix3): Change to sign of the SOB
874 instruction's offset.
875
b7c92712
NC		 8762004-10-01 Adam Nemet <anemet@lnxw.com>
877
878 * (TARGET_FORMAT): Remove LynxOS COFF definition.
879
65555e64
AM		 8802004-10-01 Ravi Ramaseshan <ravi.ramaseshan@codito.com>
881
882 * config/tc-arc.c (tc_gen_reloc): Don't assume fixP->fx_addsy is an
883 asymbol *, instead use symbol_get_bfdsym.
884
9fe0b840
NC		 8852004-09-30 Linus Nielsen Feltzing <linus@haxx.se>
886
887 * config/tc-m68k.c (select_control_regs): Add mcf5249.
888
0dd132b6
NC		 8892004-09-30 Paul Brook <paul@codesourcery.com>
890
891 * config/tc-arm.c (do_smi, do_nop): New functions.
892 (insns): Add ARMv6ZK instructions.
893 (md_apply_fix3): Handle BFD_RELOC_ARM_SMI.
894 (tc_gen_reloc): Ditto.
895 (arm_cpus): Add mpcore and arm1176.
896 (arm_archs): Add armv6{k,z,zk}.
897 * doc/c-arm.texi: Document new cores and architectures.
898
a737bd4d
NC		 8992004-09-30 Nick Clifton <nickc@redhat.com>
900
901 * config/tc-arm.c: Use ISO C90 formatting.
902
3631a3c8
NC		 9032004-09-30 Vladimir Ivanov <vladitx@nucleusys.com>
904
a737bd4d 905 * config/tc-arm.c (mav_reg_required_here): Allow REG_TYPE_CN
3631a3c8
NC		 906 as alternative when REG_TYPE_MVF, REG_TYPE_MVD, REG_TYPE_MVFX or
907 REG_TYPE_MVDX is expected.
908
d5f0cf92
AM		 9092004-09-29 Marc Bevand <m.bevand@gmail.com>
910
911 * doc/c-i386.texi (i386-Mnemonics): Fix typo.
912
9f9a069e
JW		 9132004-09-21 James E Wilson <wilson@specifixinc.com>
914
915 * config/tc-ia64.c (ENCODED_PSP_OFFSET): New.
916 (output_rp_psprel, output_pfs_psprel, output_preds_psprel,
917 output_spill_base, output_unat_psprel, output_lc_psprel,
918 output_fpsr_psprel, output_priunat_psprel, output_bsp_psprel,
919 output_bsprestore_psprel, output_rnat_psprel, output_spill_psprel,
920 output_spill_psprel_p): Use it.
921
4b778c88
TL		 9222004-09-20 Tomer Levi <Tomer.Levi@nsc.com>
923
04ca8703
EC		 924 * config/tc-crx.c (handle_LoadStor): New function.
925 Handle load/stor unique instructions before parsing.
4b778c88 926
eb043451
PB		 9272004-09-17 Paul Brook <paul@codesourcery.com>
928
929 * config/tc-arm.c (s_arm_rel31): New funciton.
930 (md_pseudo_table): Add .rel31.
04ca8703 931 (md_apply_fix3): Handle BFD_RELOC_ARM_TARGET2,
eb043451
PB		 932 BFD_RELOC_32_PCREL and BFD_RELOC_ARM_PREL31.
933 (tc_gen_reloc): Handle BFD_RELOC_ARM_PREL31 and BFD_RELOC_ARM_TARGET2.
934 (arm_fix_adjustable): Return 0 for BFD_RELOC_ARM_TARGET2.
935 (arm_parse_reloc): Add (target2).
936
8ea2b00f
AM		 9372004-09-17 Alan Modra <amodra@bigpond.net.au>
938
939 * Makefile.am: Run "make dep-am".
940 * Makefile.in: Regenerate.
941 * aclocal.m4: Regenerate.
942 * configure: Regenerate.
943 * doc/Makefile.in: Regenerate.
944 * po/POTFILES.in: Regenerate.
945 * po/gas.pot: Regenerate.
946
7e1792e0
HPN		 9472004-09-14 Hideki IWAMOTO <h-iwamoto@kit.hi-ho.ne.jp>
948
949 * config/tc-mmix.c [!LLONG_MIN]: Correct #elsif to #elif.
950 [!LLONG_MAX]: Ditto.
951
9c504268
PB		 9522004-09-13 Paul Brook <paul@codesourcery.com>
953
954 * config/tc-arm.c: Rename RELABS to TARGET1.
955
2159ac21
AM		 9562004-09-13 Alan Modra <amodra@bigpond.net.au>
957
958 * messages.c (as_internal_value_out_of_range): Cast values passed
959 to as_bad_where or as_warn_where to proper type.
960
23794b24
MM		 9612004-09-11 Theodore A. Roth <troth@openavr.org>
962
963 * config/tc-avr.c: Add support for
964 atmega48, atmega88, atmega168, attiny13, attiny2313, at90can128.
965
e9fad691
AM		 9662004-09-09 Alan Modra <amodra@bigpond.net.au>
967
968 * dw2gencfi.c (select_cie_for_fde): When separating CIE out
969 from FDE, treat a CFI_escape as we do a DW_CFA_advance_loc.
970
10b016c2
PB		 9712004-09-08 Paul Brook <paul@codesourcery.com>
972
973 * config/obj-elf.c (obj_elf_section_type): Handle init_array,
974 fini_array and preinit_array section types.
975 * config/tc-ia64.c (ia64_elf_section_type): Remove init_array
976 and fini_array.
977 * doc/as.texinfo: Document extra section types.
978
e5a52504
MM		 9792004-09-02 Mark Mitchell <mark@codesourcery.com>
980
981 * Makefile.am (TARG_ENV_HFILES): Add te-symbian.h.
982 * Makefile.in: Regenerated.
983 * configure.in: Set em for arm*-*-symbianelf*.
984 * configure: Regenerated.
985 * config/tc-arm.c (elf32_arm_target_format): Use Symbian target
986 vectors when appropriate.
987 * config/te-symbian.h: New file.
988
42851540
NC		 9892004-09-03 Tomer Levi <Tomer.Levi@nsc.com>
990
991 * config/tc-crx.c (gettrap): Exception vector can be case
992 insensitive.
04ca8703 993 (process_label_constant): Fix a 32-bit displacement bug in branch
42851540
NC		 994 instructions.
995 (get_operandtype) : Bug fix, wrong operand was used.
04ca8703 996 (process_label_constant): Initialize relocation type to
42851540
NC		 997 BFD_RELOC_NONE
998
8783612f
RE		 9992004-09-01 Richard Earnshaw < reanrsha@arm.com>
1000
1001 * tc-arm.c (arm_cpus, arm_fpus): Allow <cpu>-s as well as <cpu>s
1002 for synthesizable cores.
1003
1ff4677c
RE		 1004 * doc/c-arm.texi (ARM Options): Document canonical names of CPUs.
1005
b18c562e
NC		 10062004-08-25 Dmitry Diky <diwil@spec.ru>
1007
1008 * config/tc-msp430.c: Clean-up the code.
1009 (md_relax_table): New relax table.
1010 (mcu_types): Sort MCU types.
1011 (md_pseudo_table): Add .profiler pseudo handler.
1012 (pow2value): New function.
1013 (msp430_profiler): New function.
1014 (msp430_operands): Add new insns handlers.
1015 (msp430_srcoperand): Add register operand handler, allow complex
1016 expressions.
1017 (md_estimate_size_before_relax): Rewritten.
1018 (md_convert_frag): Rewritten.
1019 (msp430_relax_frag): New function.
1020 * config/tc-msp430.h (md_relax_frag): define macro
1021 * doc/c-msp430.texi: Update information.
1022
8f94ae4d
NC		 10232004-08-24 Nick Clifton <nickc@redhat.com>
1024
1025 * as.c (std_shortopts): Allow -g to take an optional argument.
1026 (parse_args): Pass any switch starting with -g on to the backend
1027 for parsing.
1028
e51dace8
MM		 10292004-08-18 Mark Mitchell <mark@codesourcery.com>
1030
1031 * configure.in (arm*-*-symbianelf*): New target.
1032 (arm*-*-eabi*): Likewise.
1033 * configure: Regenerated.
1034
f7870c8d
TS		 10352004-08-18 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1036 * config/tc-mips.c (append_insn): Handle delay slots in branch likely
1037 correctly.
1038
d6afba4b
JJ		 10392004-08-18 Jakub Jelinek <jakub@redhat.com>
1040
1041 * config/tc-ia64.c (start_unwind_section): Add linkonce_empty
1042 argument, don't do anything if current section is not
1043 .gnu.linkonce.t.* and linkonce_empty is set.
1044 (generate_unwind_image, dot_endp): Adjust callers, call
1045 start_unwind_section (*, 1) if nothing will be put into the
1046 section.
1047
329e276d
NC		 10482004-08-17 Nick Clifton <nickc@redhat.com>
1049
1050 * as.c (MD_DEBUG_FORMAT_SELECTOR): Provide default definition.
1051 (show_usage): Add -g.
1052 (std_longopts): Add --gen-debug. Alpha sort the table.
1053 (parse_args): Print an error message if a switch is not handled.
1054 Handle the -g switch, calling md_debug_format_selector() if
1055 necessary.
1056 * NEWS: Mention new feature.
1057 * doc/as.texinfo: Document new switch.
1058 * doc/internals.texi: Document behaviour of md_parse_option.
04ca8703 1059
329e276d
NC		 1060 * config/tc-arm.c (md_parse_option): Do not issue an error message
1061 if the switch is not recognised.
1062 * config/tc-m68k.c (md_parse_option): Likewise.
1063 * config/tc-pdp11.c (md_parse_option): Likewise.
1064 * config/tc-v850.c (md_parse_option): Likewise.
1065
1066 * as.h: Fix up formatting.
1067 * tc.h: Likewise.
1068
4101eec8
NC		 10692004-08-16 Nick Clifton <nickc@redhat.com>
1070
2766e5e4
NC		 1071 * macro.c (macro_set_alternate): Use ISO C90 formatting.
1072
4101eec8
NC		 1073 * configure.in: Sort architecture based tables alphabetically.
1074 * configure: Regenerate.
1075
80f846b6
AM		 10762004-08-16 Alan Modra <amodra@bigpond.net.au>
1077
1078 * config/tc-ppc.c (tc_ppc_regname_to_dw2regnum <regnames>): Replace
1079 { "cc", 68 }, with { "cr", 70 }.
1080
caa32fe5
NC		 10812004-08-13 Jan Beulich <jbeulich@novell.com>
1082 Nick Clifton <nickc@redhat.com>
1083
1084 * as.c: Add and handle new --alternate command line option.
1085 * macro.c (macro_set_alternate): New.
1086 * macro.h (macro_set_alternate): Declare.
1087 * read.c: Add and handle new .altmacro and .noaltmacro directives.
1088 * doc/as.texinfo: Document new command line option and pseudo-ops
1089 as well as insert documentation originating from gasp about
1090 alternate macro syntax.
1091 * NEWS: Mention new command line option and pseudo-ops.
1092
b2221023
MM		 10932004-08-10 Mark Mitchell <mark@codesourcery.com>
1094
1095 * expr.c (operand): Handle the "~", "-", and "!" operators applied
1096 to bignums.
1097
db6579d4
PB		 10982004-08-06 Paul Brook <paul@codesourcery.com>
1099
1100 * config/tc-arm.c (md_apply_fix3, tc_gen_reloc, arm_parse_reloc):
1101 Handle new relocations.
1102 * include/elf/arm.h (elf_arm_reloc_type): Add new EABI relocations.
1103
e750405d
BW		 11042004-08-05 Bob Wilson <bob.wilson@acm.org>
1105
1106 * write.c (relax_segment): Use was_address instead of address when
1107 setting fr_fix field for align frag due to backwards .org.
1108
1d70c7fb
AO		 11092004-07-29 Alexandre Oliva <aoliva@redhat.com>
1110
1111 Introduce SH2a support.
1112 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
1113 * config/tc-sh.c (get_specific): Change arch_sh2a_up to
1114 arch_sh2a_nofpu_up.
1115 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
1116 * config/tc-sh.c (md_parse_option): Add sh2a-nofpu ISA handling.
1117 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
1118 * config/tc-sh.c (sh_elf_final_processing): Move sh2a recognition
1119 to end of conditional expression.
1120 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
1121 * config/tc-sh.c: Add sh2a-nofpu support.
1122 2003-12-29 DJ Delorie <dj@redhat.com>
1123 * tc-sh.c: Add sh2a support.
1124 (parse_reg): Add tbr.
1125 (parse_at): Support @@(disp,tbr).
1126 (get_specific): Support sh2a opcodes.
1127 (insert4): New, for 4 byte relocs.
1128 (build_Mytes): Support sh2a opcodes.
1129 (md_apply_fix3_Mytes): Support sh2a opcodes.
1130 2003-12-02 Michael Snyder <msnyder@redhat.com>
1131 * config/tc-sh.c (md_parse_option): Handle sh2a.
1132 (sh_elf_final_processing): Ditto.
1133
e7fbc2bf
JT		 11342004-07-27 Jason Thorpe <thorpej@wasabisystems.com>
1135
1136 * config/tc-hppa.h (TARGET_FORMAT): Set to "elf32-hppa-netbsd"
1137 for TE_NetBSD.
1138
9232bbb0
AM		 11392004-07-27 Alan Modra <amodra@bigpond.net.au>
1140
1141 * config/tc-ppc.c (ppc_frob_file_before_adjust): Warn if .toc too big.
1142 (ppc_arch): Expand comment.
1143
670ec21d
NC		 11442004-07-27 Tomer Levi <Tomer.Levi@nsc.com>
1145
1146 * config/tc-crx.c: Support evaluating the difference between two
1147 symbols.
1148 * config/tc-crx.h: Likewise.
1149
406f8ff9 11502004-07-26 H.J. Lu <hongjiu.lu@intel.com>
38ce5b11
L		 1151
1152 * config/tc-ia64.c (start_unwind_section): Set the linked-to
1153 section.
1154 (ia64_elf_section_change_hook): Set the linked-to section for
1155 SHT_IA_64_UNWIND.
1156
c05e9f04
NC		 11572004-07-26 Dmitry Diky <diwil@spec.ru>
1158
1159 * config/tc-msp430.c: Add new subtargets: msp430x1610,
1160 msp430x1611, msp430x1612, msp430x415, msp430x417, msp430xG437,
1161 msp430xG438, msp430xG439.
1162
e9863d7f
DJ		 11632004-07-25 Daniel Jacobowitz <dan@debian.org>
1164
1165 * doc/as.texinfo (Section, PushSection): Correct documentation
1166 for ELF.
1167
543613e9 11682004-07-21 Jan Beulich <jbeulich@novell.com>
20f0a1fc
NC		 1169
1170 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
1171 permissible for the selected instruction suffix.
1172 (match_template): Don't permit 64-bit general purpose operands in
1173 32-bit mode.
1174 (finalize_imm): Permit 64-bit immediates.
1175 (build_modrm_byte): Don't treat 32-bit addressing in 64-bit mode
1176 specially except for the width of the used base and/or index
1177 registers. For 32-bit displacements, use sign-extended
1178 relocations only when using 64-bit addressing.
1179 Force zero displacement on rip-relative addressing when there is
1180 no other displacement.
1181 (i386_index_check): Don't treat 32-bit addressing in 64-bit mode
1182 specially except for the width of the used base and/or index
1183 registers.
1184 (parse_register): Disallow Reg64 registers in 32-bit mode.
1185
543613e9
NC		 1186 * config/tc-i386.c: For DefaultSize instructions, don't guess a 'q'
1187 suffix if the instruction doesn't support it.
04ca8703 1188
01a3f561
MR		 11892004-07-20 Maciej W. Rozycki <macro@linux-mips.org>
1190
1191 * config/tc-mips.c (append_insn): Handle constant expressions with
1192 no associated relocation.
1193 (mips_ip): Cancel the expression after use for the Q format
1194 specifier.
1195 (parse_relocation): Return no relocation for unsupported
1196 operators.
1197 (my_getSmallExpression): Return no relocation if no relocation
1198 operators are used.
1199
e12904d2
DA		 12002004-07-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1201
1202 * config/obj-som.c (adjust_stab_sections): Add prototype.
1203 (obj_som_compiler, obj_som_version, obj_som_copyright,
1204 adjust_stab_sections): Add ATTRIBUTE_UNUSED to unused arguments.
1205 * config/tc-hppa.c (update_subspace): Likewise.
1206 (is_defined_subspace): Amplify comment.
1207 * config/obj-som.h (som_frob_file): Add prototype.
1208
90c1602c
L		 12092004-07-19 Christopher Faylor <cgf@timesys.com>
1210 H.J. Lu <hongjiu.lu@intel.com>
1211
1212 * subsegs.c (section_symbol): Don't create a new segment when
1213 existing segment is undefined.
1214
f263249b
RE		 12152004-07-16 Richard Earnshaw <rearnsha@arm.com>
1216
1217 * config/tc-arm.c: Include include/opcode/arm.h.
1218 (ARM_EXT_*, ARM_ARCH_*, ARM_ANY, ARM_ALL, COPROC_ANY): Delete.
1219 (FPU_FPA_EXT_* FPU_VFP_EXT_*, FPU_ANY, FPU_NONE, FPU_MAVERICK): Delete.
1220 (FPU_ARCH_*): Delete.
1221 * Makefile.am: Update dependencies.
1222 * Makefile.in: Regenerate.
1223
152c65b9
NC		 12242004-07-15 Nick Clifton <nickc@redhat.com>
1225
1226 * configure.in: Accept armbe as a big-endian arm configuration.
1227 * configure: Regenerate.
1228
0477af35
NC		 12292004-07-13 Thomas Nystrom <thn@saeab.se>
1230
1231 * config/tc-i386.c (T_SHIFTOP): New constant.
1232 (intel_e05_1): Handle '&', '|' and T_SHIFTOP.
1233 (intel_el1): Handle '~'.
1234 (intel_get_token): Handle '<>', '&', '|' and '~'.
1235
12362004-07-13 Nick Clifton <nickc@redhat.com>
1237
1238 (md_assemble): Remove spurious newline from end of as_bad error
1239 message.
1240 (intel_e05_1): Likewise.
1241 (intel_e11): Likewise.
1242 (intel_match_token): Likewise.
1243
31302357
AS		 12442004-07-11 Andreas Schwab <schwab@suse.de>
1245
1246 * config/tc-m68k.c: Convert to C90. Remove redundant
1247 declarations. Indentation fixup.
1248 [M68KCOFF]: Include "obj-coff.h" instead of declaring
1249 obj_coff_section ourselves.
04ca8703 1250
549f748d
JW		 12512004-07-09 James E Wilson <wilson@specifixinc.com>
1252
1253 * config/tc-ia64.c (default_big_endian): New.
1254 (dot_byteorder, md_begin): Use it.
1255 (md_parse_option): Set it.
04ca8703 1256
1a320fbb
NC		 12572004-07-09 Nick Clifton <nickc@redhat.com>
1258
1259 * configure.in: Change sh-sybmian-elf to sh-*-symbianelf.
1260 * configure: Regenerate.
1261 * NEWS: Change sh-sybmian-elf to sh-*-symbianelf.
1262 * config/tc-sh.c (sh_elf_final_processing): Use renamed version of
1263 sh_find_elf_flags if necessary.
1264
de7e6852
RS		 12652004-07-08 Richard Sandiford <rsandifo@redhat.com>
1266
1267 * config/tc-mips.c (mips_fix_adjustable): If the full addend is
1268 going to be split into more than one in-place addend, return 0
1269 for relocations against mergeable sections. Associate comments
1270 with code.
1271
1fe1f39c
NC		 12722004-07-07 Tomer Levi <Tomer.Levi@nsc.com>
1273
1274 * Makefile.am (CPU_TYPES): Add crx.
1275 (TARGET_CPU_CFILES): Add config/tc-crx.c.
1276 (TARGET_CPU_HFILES): Add config/tc-crx.h.
1277 (DEPTC_crx_elf): New target.
1278 (DEPOBJ_crx_elf): Likewise.
1279 (DEP_crx_elf): Likewise.
1280 * Makefile.in: Regenerate.
1281 * configure.in: Add crx* target.
1282 * configure: Regenerate.
1283 * config/tc-crx.c: New file.
1284 * config/tc-crx.h: New file.
1285 * NEWS: Mention new target.
1286
d9895649
NC		 12872004-07-06 Nick Clifton <nickc@redhat.com>
1288
85fbca6a
NC		 1289 * config.in: Undefine TARGET_SYMBIAN by default.
1290 * configure.in:
1291 * configure: Regenerate. Add sh-symbian-elf target. If
1292 selected define TARGET_SYMBIAN.
1293 * config/tc-sh.h (TARGET_FORMAT): Select a Symbian target
1294 format if TARGET_SYMBIAN has been defined.
04ca8703 1295
e7bd9ea0
NC		 1296 * output-file.c (output_file_create): Report the target format
1297 chosen when bfd_openw reports that it is invalid.
1298
d9895649
NC		 1299 * config/obj-coff.c (coff_pseudo_table): Only define the weak
1300 pseudo for BFD based assemblers.
1301
871ec896
JR		 13022004-07-05 Andrew Stubbs <andrew.stubbs@superh.com>
1303
1304 gas:
1305 * config/tc-sh.c (md_assemble): Change isspace to ISSPACE.
1306 (md_parse_option): Remove redundant -isa testing.
1307 Make bfd_arch variable const.
1308 (md_show_usage): Make bfd_arch variable const.
1309
5a9ff93d
JW		 13102004-07-03 James E Wilson <wilson@specifixinc.com>
1311
1312 * config/tc-ia64.c (emit_one_bundle): Check and set insn_addr.
1313 * config/tc-ia64.h (md_frag_check): Define.
1314
c87db184
CF		 13152004-07-03 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
1316
1317 * config/obj-coff.c (obj_coff_weak): New .weak syntax for PE weak
1318 externals.
1319 * doc/as.texinfo (Weak): Document PE weak symbols.
1320
3b91255e
RS		 13212004-07-03 Richard Sandiford <rsandifo@redhat.com>
1322
1323 * config/tc-mips.c (HAVE_IN_PLACE_ADDENDS): New macro.
1324 (reloc_needs_lo_p): Only return true if HAVE_IN_PLACE_ADDENDS.
1325 (mips_frob_file): Rework so that only a single pass through the
1326 relocs is needed. Allow %lo()s to have higher offsets than their
1327 corresponding %hi()s or %got()s.
1328
310ea308
NC		 13292004-07-02 Nick Clifton <nickc@redhat.com>
1330
1331 * config/tc-arm.c (md_apply_fix3:BFD_RELOC_ARM_IMMEDIATE): Do not
1332 allow values which have come from undefined symbols.
1333 Always consider this fixup to have been processed as a reloc
1334 cannot be generated for it.
1335
09b935ac
AM		 13362004-07-02 Alan Modra <amodra@bigpond.net.au>
1337
1338 * frags.h (struct frag): Add has_code and insn_addr fields.
1339 * write.c (cvt_frag_to_fill): Invoke md_frag_check.
1340 * config/tc-ppc.c (md_assemble): Check and set insn_addr.
1341 * config/tc-ppc.h (md_frag_check): Define.
1342
4d1cd5c7
MR		 13432004-06-28 Maciej W. Rozycki <macro@linux-mips.org>
1344
1345 * doc/Makefile.am (info): Rename goal to...
1346 (info-local): ... this, to preserve implicit dependencies.
1347 * doc/Makefile.in: Regenerate with automake 1.8.5.
1348
097f809a
NC		 13492004-06-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1350
1351 * config/tc-m32r.c (md_convert_frag): Changed for @PLT.
1352 (m32r_cgen_record_fixup_exp): Changed for @GOTOFF, @GOT.
1353 (m32r_fix_adjustable): Changed for @GOTOFF, @GOT, @PLT.
1354 (tc_gen_reloc): Likewise.
1355 (m32r_end_of_match): Add for @GOTOFF, @GOT, @PLT.
1356 (m32r_parse_name): Likewise.
1357 (m32r_cgen_parse_fix_exp): Likewise.
1358 * config/tc-m32r.h (md_parse_name): Define for @GOTOFF, @GOT, @PLT.
1359 (O_PIC_reloc): Likewise.
1360 (TC_CGEN_PARSE_FIX_EXP): Likewise..
1361 * cgen.c (gas_cgen_parse_operand): Add TC_CGEN_PARSE_FIX_EXP
1362 for @GOTOFF, @GOT, @PLT.
1363
bb82af9f
NC		 13642004-06-21 Jan Beulich <jbeulich@novell.com>
1365
1366 * gas/symbols.c: While discarding ordinary local absolute symbols
1367 when --strip-local-absolute is in effect, retain file symbols.
1368
990ae22a
AS		 13692004-06-20 Andreas Schwab <schwab@suse.de>
1370
1371 * config/tc-m68k.c (mri_chip): Replace current_chip, not augment.
1372 (md_parse_option): Likewise.
1373
32137342
NC		 13742004-06-17 Jan Beulich <jbeulich@novell.com>
1375
1376 * config/tc-i386.c: Deal with LEX_QM the same way as with LEX_AT.
1377 * config/te-netware.h: New file.
1378 * config/te-ppcnw.h: Delete: Obsolete.
1379 * configure.in: Eliminate ill NetWare targets. Make generic
1380 NetWare target use proper emulation.
1381 * Makefile.am: Eliminate reference to obsolete te-ppcnw.h, add
1382 reference to new te-netware.h.
1383 * configure: Regenerate.
1384 * Makefile.in: Regenerate.
1385
f84d8e26
MS		 13862004-06-15 Martin Schwidefsky <schwidefsky@de.ibm.com>
1387
1388 * config/tc-s390.c (s390_insn): Avoid incorrect signed/unsigned
1389 comparison in .insn pseudo operation.
1390
587aac4e
AM		 13912004-06-15 Alan Modra <amodra@bigpond.net.au>
1392
1393 * config/obj-coff.c (coff_adjust_section_syms): Use
1394 bfd_get_section_size instead of bfd_get_section_size_before_reloc.
1395 (coff_frob_section): Likewise.
1396 * config/tc-mips.c (md_apply_fix3): Likewise.
1397 * config/obj-elf.c (elf_frob_file): Use bfd_set_section_size.
1398 (elf_frob_file_after_relocs): Likewise.
1399
2337eb33
DA		 14002004-06-10 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1401
b8fc22bc
DA		 1402 * config/tc-hppa.c (log2): Rename to exact_log2.
1403 (pa_next_subseg): Delete unused function.
1404 (create_new_space): Mark unused arguments with ATTRIBUTE_UNUSED.
1405 (create_new_subspace): Likewise.
1406
2337eb33
DA		 1407 Bug gas/213
1408 * config/tc-hppa.c (hppa_fix_adjustable): Allow reduction of fake
1409 labels. Fix warning.
1410
53cb0362
DD		 14112004-05-28 DJ Delorie <dj@redhat.com>
1412
1413 * config/tc-mn10300.h (tc_fix_adjustable): Define.
1414 * config/tc-mn10300.c (mn10300_fix_adjustable): Don't adjust debug
1415 or non-merged symbols.
1416
888a75be
L		 14172004-05-28 H.J. Lu <hongjiu.lu@intel.com>
1418
1419 * config/tc-ia64.c (remove_marked_resource): Save, clear and
1420 restore the old slot when inserting srlz.i/srlz.d.
1421
f6f9408f
JR		 14222004-05-28 Andrew Stubbs <andrew.stubbs@superh.com>
1423
1424 * Makefile.am: Regenerate dependecies.
1425 * Makefile.in: Regenerate.
1426 * config/tc-sh.c (valid_arch): Make unsigned.
1427 (preset_target_arch): Likewise.
1428 (md_begin): Use new architecture flags system.
1429 (get_specific): Likewise.
1430 (assemble_ppi): Likewise.
1431 (md_assemble): Likewise. Also fix error check for bad opcodes.
1432 (md_parse_option): Likewise. Also generate -isa values according
1433 to the table in bfd/cpu-sh.c instead of just constants. Also
1434 allow <arch>-up ISA variants.
1435 (sh_elf_final_processing): Replace if-else chain with a call to
1436 sh_find_elf_flags().
1437 * testsuite/gas/sh/arch: New directory.
1438 * testsuite/gas/sh/arch/arch.exp: New test script.
1439 * testsuite/gas/sh/arch/arch_expected.txt: New file.
1440 * testsuite/gas/sh/arch/sh.s: New file.
1441 * testsuite/gas/sh/arch/sh2.s: New file.
1442 * testsuite/gas/sh/arch/sh-dsp.s: New file.
1443 * testsuite/gas/sh/arch/sh2e.s: New file.
1444 * testsuite/gas/sh/arch/sh3-nommu.s: New file.
1445 * testsuite/gas/sh/arch/sh3.s: New file.
1446 * testsuite/gas/sh/arch/sh3-dsp.s: New file.
1447 * testsuite/gas/sh/arch/sh3e.s: New file.
1448 * testsuite/gas/sh/arch/sh4-nommu-nofpu.s: New file.
1449 * testsuite/gas/sh/arch/sh4-nofpu.s: New file.
1450 * testsuite/gas/sh/arch/sh4.s: New file.
1451 * testsuite/gas/sh/arch/sh4a-nofpu.s: New file.
1452 * testsuite/gas/sh/arch/sh4al-dsp.s: New file.
1453 * testsuite/gas/sh/arch/sh4a.s: New file.
1454
580b9172
NC		 14552004-05-28 Peter Barada <peter@the-baradas.com>
1456
1457 * config/gc-m68k.c(m68k_ip): Convert mode 5 addressing
1458 with zero offset into mode 2 addressing to save a word.
1459
652ca075
L		 14602004-05-27 H.J. Lu <hongjiu.lu@intel.com>
1461
1462 * config/tc-ia64.c (ar_is_in_integer_unit): Removed.
1463 (ar_is_only_in_integer_unit): New.
1464 (ar_is_only_in_memory_unit): New.
1465 (generate_unwind_image): Silence gcc on 32bit host.
1466 (md_assemble): Use ar_is_only_in_integer_unit instead of
1467 ar_is_in_integer_unit. Check AR access.
1468
60fed2d3
NC		 14692004-05-27 Peter Barada <peter@the-baradas.com>
1470
1471 * config/tc-m68k.c (md_begin): Sort the opcode table into
1472 alphabetical order.
1473 (m68k_compare_opcode): New function to do the sorting.
1474
be8c092b
NC		 14752004-05-24 Peter Barada <peter@the-baradas.com>
1476
1477 * config/m68k-parse.y(operand): Allow for MAC/EMAC mask
1478 addressing on MIT style operands.
1479 * config/m68k-parse.y(yylex): Allow '-&' for predecrement
1480 w/mask addressing.
1481 * config/tc-m68k.c(install_operand): Comment 'G' and 'H' type
1482 operands.
1483
1181c20e
AM		 14842004-05-23 Alan Modra <amodra@bigpond.net.au>
1485
1486 * expr.c (operand, operator): Don't reject '++' and '--'.
1487
20e56c33
RS		 14882004-05-20 Richard Sandiford <rsandifo@redhat.com>
1489
1490 * config/tc-mips.c (append_insn): Use ISA-encoded addresses in MIPS16
1491 dwarf tables.
1492
04ca8703 14932004-05-17 Adam Nemet <anemet@lnxw.com>
b2d65c0b
NC		 1494
1495 * configure.in: Add ppc-*-lynxos*. Update i386-*-lynxos* to ELF.
1496 * configure: Regenerate.
1497
0da76f83
NC		 14982004-05-13 Paul Brook <paul@codesourcery.com>
1499
1500 * dw2gencfi.c (output_cie): Handle dwarf3 format CIE entries.
1501
4b363073
NC		 15022004-05-13 Joel Sherrill <joel@oarcorp.com>
1503
1504 * configure.in (or32-*-rtems*): Switch to elf.
1505 * configure: Regenerate.
1506
9598fbe5
NC		 15072004-05-13 Nick Clifton <nickc@redhat.com>
1508
1509 * po/fr.po: Updated French translation.
1510
22fe14ad
NC		 15112004-05-11 Nick Clifton <nickc@redhat.com>
1512
1513 * doc/as.texinfo (Section): Document G and T flags to .section
1514 directive. Document the extra arguments that the G flag
1515 requires. Document the #tls flag.
1516
15172004-05-11 H.J. Lu <hongjiu.lu@intel.com>
1518
1519 * subsegs.c (section_symbol): Create a new section symbol if
1520 the existing one doesn't match.
1521 * symbols.c (symbol_set_bfdsym): Don't reset BFD section symbol.
1522
532c738a
RS		 15232004-05-07 Richard Sandiford <rsandifo@redhat.com>
1524
1525 * config/tc-mips.c (append_insn, mips_emit_delays): Extend -mfix-vr4120
1526 to cope with VR4181A errata MD(1) and MD(4).
1527
0503b355
BF		 15282004-05-07 Brian Ford <ford@vss.fsi.com>
1529
1530 * NEWS: Mention .secrel32 for pe[i]-i386.
1531
daeea990
AO		 15322004-05-07 Alexandre Oliva <aoliva@redhat.com>
1533
1534 * config/tc-frv.h (MAX_MEM_FOR_RS_ALIGN_CODE): New.
1535 (HANDLE_ALIGN): New.
1536
bfbdcf04
DJ		 15372004-05-06 Daniel Jacobowitz <dan@debian.org>
1538
1539 * Makefile.am (DIST_SUBDIRS): Define.
1540 * aclocal.m4: Regenerate with automake 1.8.4.
1541 * Makefile.in: Likewise.
1542 * doc/Makefile.in: Likewise.
1543
5e819f9c
JW		 15442004-05-06 David Mosberger-Tang <davidm@hpl.hp.com>
1545
1546 * config/tc-ia64.c (dot_serialize): Declare.
1547 (dot_serialize): New function.
1548 (md_pseudo_table): Add ".serialize.data" and
1549 ".serialize.instruction" directives.
1550
e5976317
NC		 15512004-05-06 Nick Clifton <nickc@redhat.com>
1552
1553 * messages (as_internal_value_out_of_range): Print a message about
1554 a value being out of range. Be consistent about whether the
1555 values are printed in decimal or hexadecimal.
1556 (as_warn_value_out_of_range): Generate a warning message about an
1557 out of range value.
1558 (as_bad_value_out_of_range): Generate an error message about an
1559 out of range value.
1560 * as.h: Prototype the new functions.
1561 * config/tc-alpha.c (insert_operand): Use new function.
1562 * config/tc-arc.c (arc_insert_operand): Likewise.
1563 * config/tc-mn10200.c (mn10200_insert_operand): Likewise.
1564 * config/tc-mn10300.c (mn10300_insert_operand): Likewise.
1565 * config/tc-ppc.c (ppc_insert_operand): Likewise.
1566 * config/tc-s390.c (s390_insert_operand): Likewise.
1567 * config/tc-v850.c (v850_insert_operand): Likewise.
1568
996991aa
AO		 15692004-05-05 Alexandre Oliva <aoliva@redhat.com>
1570
1571 * configure.in: Set em=linux for frv-*-*linux*.
1572 * configure: Rebuilt.
1573 * config/tc-frv.h (TARGET_FORMAT): Use elf32-frvfdpic if...
1574 (frv_md_fdpic_enabled): New.
1575 * config/tc-frv.c (frv_md_fdpic_enabled): New.
1576 (DEFAULT_FDPIC): New.
1577 (frv_flags): Use DEFAULT_FDPIC.
1578 (frv_pic_flag): Likewise.
1579 (OPTION_NOPIC): New.
1580 (md_longopts): Add -mnopic.
1581 (md_parse_option): Handle it.
1582 (md_show_usage): Add -mfdpic and -mnopic.
1583
6b6e92f4
NC		 15842004-05-05 Peter Barada <peter@the-baradas.com>
1585
1586 * config/tc-m68k.c: Add find_cf_chip to print list of valid
1587 chips for invalid coldfire instructions, rename selectors
1588 for ColdFire sub-variants, add 521x,5249,547x,548x and aliases,
1589 add current_chip to track which chip is referred to(including save/restore),
1590 use current_chip to select control registers, not current_arch.
1591 (md_show_usage): Add new chips.
1592 * doc/c-m68k.texi: Document new command line switches.
1593
6f079ae3
JJ		 15942004-05-05 Jakub Jelinek <jakub@redhat.com>
1595
1596 * tc-s390.h (md_do_align, HANDLE_ALIGN): Remove.
1597 (NOP_OPCODE): Define.
1598 (s390_align_code): Remove prototype.
1599 * tc-s390.c (s390_align_code): Remove.
1600
2d6ed997
L		 16012004-05-04 H.J. Lu <hongjiu.lu@intel.com>
1602
1603 * config/tc-ia64.c (make_unw_section_name): Removed.
1604 (start_unwind_section): New function.
1605 (generate_unwind_image): Take const segT instead of const
1606 char *.
1607 (dot_handlerdata): Adjusted.
1608 (dot_endp): Likewise.
1609
b9f18452
L		 16102004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1611
1612 * config/obj-elf.c (obj_elf_change_section): Allow the
1613 ".note.GNU-stack" section has SHF_EXECINSTR.
1614
fafe6678
L		 16152004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1616
1617 * config/obj-elf.c (get_section): Return bfd_boolean.
1618 (obj_elf_change_section): Call bfd_get_section_by_name_if
1619 instead of bfd_map_over_sections.
1620
86654c12
L		 16212004-04-30 H.J. Lu <hongjiu.lu@intel.com>
1622
1623 * config/obj-elf.c (get_section): New function.
1624 (obj_elf_change_section): Support multiple sections with same
1625 name.
1626
376eb240
NC		 16272004-04-30 Nick Clifton <nickc@redhat.com>
1628
1629 * config/tc-arm.c (create_register_alias): Fix typo checking for
1630 case sensitive register aliases.
1631 (co_proc_number): Use error message string in all_reg_maps[]
1632 array.
1633 (cp_reg_required_here): Likewise.
1634 (fp_reg_required_here): Likewise.
1635
8a7140c3
NC		 16362004-04-29 Brian Ford <ford@vss.fsi.com>
1637
1638 * dwarf2dbg.c (dwarf2_finish): Add SEC_DEBUGGING to section flags.
1639
3e722fb5
CD		 16402004-04-28 Chris Demetriou <cgd@broadcom.com>
1641
1642 * config/tc-mips.c (HAVE_32BIT_ADDRESSES, append_insn, macro_build)
1643 (load_address, macro, mips_ip, md_parse_option)
1644 (mips_force_relocation, mips_validate_fix, md_apply_fix3)
1645 (s_change_sec, pic_need_relax, tc_gen_reloc): Remove all
1646 embedded-PIC handling, and update comments.
1647 (SWITCH_TABLE): Remove.
1648 * config/tc-mips.h (DIFF_EXPR_OK): Delete.
1649 (enum mips_pic_level): Remove EMBEDDED_PIC.
1650 (EXTERN_FORCE_RELOC): Remove embedded-PIC handling.
1651 (TC_FORCE_RELOCATION): Update comment.
1652 * ecoff.c (ecoff_build_lineno): Add comment about some code that
1653 might be safe to remove now that MIPS embedded-PIC is gone.
1654
351e2b5a
DA		 16552004-04-28 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1656
1657 * config/obj-som.c (obj_som_init_stab_section): Add new arguments in
1658 call to obj_set_subsection_attributes.
1659 (obj_som_init_stab_section): Likewise.
1660 * config/tc-hppa.c (default_subspace_dict): Add comdat field.
1661 (pa_def_subspaces): Provide comdat default.
1662 (pa_subspace): Handle new "comdat" parameter. Set SEC_LINK_ONCE and
1663 not SEC_IS_COMMON if section is comdat, common or dup_common. Update
1664 calls to create_new_subspace and update_subspace to pass comdat flag.
1665 (create_new_subspace, update_subspace): Add new comdat argument. Use
1666 it in calls to obj_set_subsection_attributes.
1667 * doc/c-hppa.texi (.subspa, .nsubspa): Document new comdat parameter
1668 and use of comdat, common and dup_comm parameters.
1669
f5842774
L		 16702004-04-26 H.J. Lu <hongjiu.lu@intel.com>
1671
1672 * config/obj-elf.c (obj_elf_change_section): Check if the old
1673 group name is NULL before comparison.
1674
558fe4a7
CD		 16752004-04-23 Chris Demetriou <cgd@broadcom.com>
1676
1677 * config/tc-mips.h (mips_dwarf2_addr_size): Prototype.
1678
81a21e38
TS		 16792004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1680
1681 * config/tc-mips.c (s_mipsset): Set default CPU type for .set mipsN.
1682
16832004-04-23 Chris Demetriou <cgd@broadcom.com>
1ffcab4b
CD		 1684
1685 * config/tc-mips.c (md_longopts): Remove -membedded-pic option.
1686 (OPTION_MEMBEDDED_PIC): Remove.
1687 (OPTION_TRAP, OPTION_BREAK, OPTION_EB, OPTION_EL)
1688 (OPTION_FP32, OPTION_GP32, OPTION_CONSTRUCT_FLOATS)
1689 (OPTION_NO_CONSTRUCT_FLOATS, OPTIONS_FP64, OPTION_GP64)
1690 (OPTION_RELAX_BRANCH, OPTION_NO_RELAX_BRANCH)
1691 (OPTION_ELF_BASE): Renumber.
1692 (md_parse_option): Remove OPTION_MEMBEDDED_PIC handling.
1693 (md_show_usage): Remove mention of -membedded-pic.
1694 * doc/as.texinfo: Remove mention of -membedded-pic.
1695
4d0d148d
TS		 16962004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1697
1698 * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Remove.
1699 * config/tc-mips.c (RDATA_SECTION_NAME, mips_target_format): Remove
1700 a.out support.
1701 (md_begin, mips_ip, md_parse_option, s_change_sec, s_option,
1702 s_abicalls, nopic_need_relax, tc_gen_reloc): Remove uses of
1703 USE_GLOBAL_POINTER_OPT.
1704
017315e4
TS		 17052004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1706
1707 * config/tc-mips.c (macro): One more use of load_delay_nop.
1708
269137b2
TS		 17092004-04-22 Atsushi Nemoto <anemo@mba.ocn.ne.jp>
1710
1711 * config/tc-mips.c (load_delay_nop): New function.
1712 (load_address, macro): Use load_delay_nop() to build a nop
1713 which can be omitted with gpr_interlocks.
1714
24772049
TS		 17152004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1716
1717 * config/tc-mips.c (hilo_interlocks, gpr_interlocks,
1718 cop_interlocks): Remove superfluous CPU entries.
1719
06516a5e
PB		 17202004-04-22 Paul Brook <paul@codesourcery.com>
1721
1722 * config/tc-arm.c (mav_parse_offset): Value must be multiple of 4.
1723
fd99574b
NC		 17242004-04-22 Peter Barada <peter@the-baradas.com>
1725
1726 * NEWS: Added support for EMAC instructions and MAC/EMAC
1727 Motorola syntax.
1728 * config/m68k-parse.h: Add ACC[123], ACCEXT{01,23}, MAC/EMAC
1729 scale factor tokens, trailing_ampersand to mark mask addressing
1730 for MAC/EMAC instructions.
1731 * config/m68k-parse.y: Add options_ampersand clause, '<<',
1732 '>>'.
1733 (yylex): Handle '>', '<', and '&' following '+'.
1734 * config/tc-m68k.c: Set mcfmac/mcfemac on appropriate ColdFire
04ca8703 1735 architectures in archs[].
fd99574b
NC		 1736 (m68k-ip): Add '4', 'e', 'g', 'i', cases to handle mask addressing
1737 for MAC/EMAC instructions, ACC[0123], ACCEXT{01,23}, and '<<'/'>>'
1738 respectively.
1739 (m68k_ip): Handle trailing '&' on MAC/EMAC insns.
1740 (install_operand): Fix 'n' case, Add 'F', 'f', 'G', 'H', 'I', ']'
1741 cases.
1742 Add EMAC operands to init_table[].
1743
69b97547
NC		 17442004-04-22 Bruno De Bus <bdebus@elis.ugent.be>
1745
1746 * config/tc-arm.h (enum mstate): Move here, add MAP_UNDEFINED
1747 state.
1748 (TC_SEGMENT_INFO_TYPE): Define to enum mstate.
1749 * config/tc-arm.c (enum mstate): Delete from here.
1750 (mapping_state): Remove the static mapstate variable and instead
1751 store the state in the segment. This allows a per-section mapping
1752 state. Handle and ignore MAP_UNDEFINED states.
1753 (arm_elf_change_section): Get the current mapping state from the
1754 new section.
1755 (s_ltorg): Set the mapping state to MAP_DATA.
1756 (arm_cleanup): Use arm_elf_change_section to get the mapping state
1757 for each pool as it is emitted.
1758
17592004-04-22 Nick Clifton <nickc@redhat.com>
1760
1761 * config/tc-arm.h: Formatting tidy ups.
1762
7a7f4e42
CD		 17632004-04-20 Chris Demetriou <cgd@broadcom.com>
1764
1765 * NEWS: Note that MIPS -membedded-pic option is deprecated.
1766
6482c264
NC		 17672004-04-20 DJ Delorie <dj@redhat.com>
1768
1769 * config/tc-i386.h [TE_PE] (TC_CONS_FIX_NEW): Define.
1770 * config/tc-i386.c (md_pseudo_table) [TE_PE]: Add "secrel32".
1771 [TE_PE] (O_secrel): Define.
1772 [TE_PE] (x86_pe_cons_fix_new): New.
1773 [TE_PE] (pe_directive_secrel): Likewise.
1774 (tc_gen_reloc) [TE_PE]: Support BFD_RELOC_32_SECREL.
1775
5bf10039
EC		 17762004-04-19 Eric Christopher <echristo@redhat.com>
1777
1778 * config/tc-mips.c (mips_dwarf2_addr_size): Revert part
1779 of previous patch for fix in gcc.
1780
77cd6497
JJ		 17812004-04-19 Jakub Jelinek <jakub@redhat.com>
1782
1783 * config/tc-xtensa.c (xg_assembler_literal): Fix a typo.
1784
b8861cfb
NS		 17852004-04-19 Nathan Sidwell <nathan@codesourcery.com>
1786
1787 * read.c (do_align): Call md_flush_pending_output, if defined.
1788
f4c724fc
AM		 17892004-04-16 Alan Modra <amodra@bigpond.net.au>
1790
1791 * expr.c (operand): Correct checks for ++ and --.
1792
a4b6e8f6
L		 17932004-04-14 H.J. Lu <hongjiu.lu@intel.com>
1794
1795 * config/tc-generic.c: Add some comments.
1796
d766e8ec
RS		 17972004-04-14 Richard Sandiford <rsandifo@redhat.com>
1798
1799 * doc/c-mips.texi (-m{no-,}fix-vr4120): Renamed from
1800 -{no-}mfix-vr4122-bugs.
1801 * config/tc-mips.c (mips_fix_vr4120): Renamed from mips_fix_4122_bugs.
1802 (append_insn, mips_emit_delays): Update accordingly.
1803 (OPTION_FIX_VR4120, OPTION_NO_FIX_VR4120): Renamed from *VR4122.
1804 (md_longopts): Change -{no-,}mfix-vr4122-bugs to -m{no-,}fix-vr4120.
1805 (md_parse_option): Update after above changes.
1806 (md_show_usage): Add -mfix-vr4120.
1807
ed9589d4
BW		 18082004-04-13 Bob Wilson <bob.wilson@acm.org>
1809
1810 * doc/as.texinfo (Sub-Sections): Conditionalize COFF-specific use
1811 of .section directive; add a reference to the ELF .subsection
1812 directive.
1813
097dba13
NC		 18142004-04-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1815
1816 * config/tc-m32r.c (md_assemble): Fixed infinite loop bug
1817 in parallel.
1818
279500de
TS		 18192004-04-11 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1820
1821 * Makefile.am: Remove mips from aout targets.
1822 * Makefile.in: Regenerate.
1823 * configure.in: Remove mips-dec-bsd* target.
1824 * configure: Regenerate.
1825
b7d7dc63
AM		 18262004-04-07 Alan Modra <amodra@bigpond.net.au>
1827
1828 PR 96
1829 * config/tc-ppc.c (ppc_elf_suffix): Add valid32 and valid64 fields
1830 to struct map_bfd. Adjust MAP macro, and define MAP32, MAP64.
1831 Update "mapping". Restrict some @ modifiers to 32 bit.
1832
f55629b8
KK		 18332004-04-01 Asgari Jinia <asgarij@kpitcummins.com>
1834 Dhananjay Deshpande <dhananjayd@kpitcummins.com>
1835
1836 * config/tc-sh.c (dont_adjust_reloc_32): New variable.
1837 (sh_fix_adjustable): Avoid adjusting BFD_RELOC_32 when
1838 dont_adjust_reloc_32 is set.
1839 (md_longopts): Add option -renesas.
1840 (md_parse_option, md_show_usage): Likewise.
1841 * doc/c-sh.texi: Likewise.
1842
a7844384
NC		 18432004-04-01 Dave Korn <dk@artimi.com>
1844
1845 * config/tc-dlx.c (md_assemble): set fx_no_overflow flag for
1846 hi16 and lo16 fixS structs.
1847 (md_assemble): generate bit_fixS for RELOC_DLX_LO16 in
1848 exactly the same way as for RELOC_DLX_REL16.
1849 (machine_ip): properly respect LO flag in the_insn and
1850 output RELOC_DLX_LO16 rather than RELOC_DLX_16.
1851 (md_apply_fix3): apply RELOC_DLX_LO16.
1852
6c71dbc4
SS		 18532004-03-30 Stan Shebs <shebs@apple.com>
1854
1855 Remove long-obsolete MPW support.
1856 * mpw-config.in, mpw-make.sed, mac-as.r: Remove files.
1857 * configure.in: Remove mention of ppc-*-mpw* config.
1858 * configure.in: Likewise.
1859
7cc69913
NC		 18602004-03-30 Nick Clifton <nickc@redhat.com>
1861
1862 * config/tc-arm.c (meabi_flags): Make its use conditional upon
1863 OBJ_ELF being defined.
1864
53fbb48a
AM		 18652004-03-27 Alan Modra <amodra@bigpond.net.au>
1866
1867 * config/obj-aout.c (obj_aout_type): Remove #ifdef BFD_ASSEMBLER code.
1868
d507cf36
PB		 18692004-03-23 Paul Brook <paul@codesourcery.com>
1870
1871 * config/tc-arm.c (meabi_flags): New variable.
1872 (arm_parse_eabi): New function.
1873 (md_begin): Set flags for EABI v3.
1874 (arm_eabis): Add.
1875 (arm_long_opts): Add meabi.
1876 * doc/as.texinf <ARM>: Document -meabi.
1877 * doc/c-arm.texi: Ditto.
1878
88d65ad6
BW		 18792004-03-22 Bob Wilson <bob.wilson@acm.org>
1880
1881 * config/tc-xtensa.c (xtensa_post_relax_hook): Create literal
1882 tables even when use_literal_section flag is not set.
1883
13ef8878
AM		 18842004-03-22 Alan Modra <amodra@bigpond.net.au>
1885
1886 * config/tc-sh.c: Remove trailing whitespace.
1887
1048a9ba
HPN		 18882004-03-22 Hans-Peter Nilsson <hp@axis.com>
1889
1890 * doc/c-cris.texi (CRIS-Opts): Document --no-mul-bug-abort,
1891 --mul-bug-abort and the default behavior.
1892 * config/tc-cris.c (cris_insn_kind): New member CRIS_INSN_MUL.
1893 (err_for_dangerous_mul_placement): New variable.
1894 (STATE_MUL, OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): New
1895 macros.
1896 (md_cris_relax_table): Have placeholder for STATE_MUL.
1897 (md_longopts): New options --mul-bug-abort and --no-mul-bug-abort.
1898 (cris_relax_frag) <case ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: New
1899 case doing nothing.
1900 (md_estimate_size_before_relax) <case ENCODE_RELAX (STATE_MUL,
1901 STATE_BYTE)>: Ditto.
1902 (md_convert_frag) <ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: Check
1903 alignment and position of this frag, emit error message if
1904 suspicious.
1905 (md_assemble): For a multiply insn and when checking it,
1906 transform the current frag into a special frag for that purpose.
1907 (md_parse_option) <case OPTION_MULBUG_ABORT_OFF, case
1908 OPTION_MULBUG_ABORT_ON>: Handle new options.
1909
a7877748
BW		 19102004-03-19 Bob Wilson <bob.wilson@acm.org>
1911
1912 * config/tc-xtensa.c (mark_literal_frags): New function.
1913 (xtensa_move_literals): Call mark_literal_frags for all literal
1914 segments, including init and fini literal segments.
1915 (xtensa_post_relax_hook): Swap use of xt_insn_sec and xt_literal_sec.
1916
9992270f
DA		 19172004-03-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1918
1919 * tc-hppa.c (cons_fix_new_hppa): Check for PC relative base type.
1920 (pa_comm): Set BSF_OBJECT in symbol flags.
1921
6e1772d5
AM		 19222004-03-19 Alan Modra <amodra@bigpond.net.au>
1923
1924 * Makefile.am: Run "make dep-am".
1925 * Makefile.in: Regenerate.
1926 * doc/Makefile.in: Regenerate.
1927 * config.in: Regenerate.
1928 * po/gas.pot: Regenerate.
1929
3e4caed2
NS		 19302004-03-18 Nathan Sidwell <nathan@codesourcery.com>
1931
c95b35a9
NS		 1932 * read.c (read_a_source_file): Use demand_empty_rest_of_line.
1933 (demand_empty_rest_of_line): Issue an error here.
1934 (ignore_rest_of_line): Silently skip to end.
1935 (demand_copy_string): Issue an error, not warning.
1936 (equals): Likewise.
1937 * config/obj-elf.c (obj_elf_section_name): Likewise.
1938 (obj_elf_section): Likewise.
1939 * config/tc-arc.c (arc_extoper): Remove bogus NULL checks.
1940 (arc_extinst): Likewise.
1941 * config/tc-ia64.c (dot_saveb): Use demand_empty_rest_of_line.
1942 (dot_spill): Likewise.
1943 (dot_unwabi): Likewise.
1944 (dot_prologue): Likewise.
1945
3e4caed2
NS		 1946 * expr.c (operand): Reject ++ and --.
1947 (operator): Likewise.
1948
2ce4cc60
KK		 19492004-03-17 Kaz Kojima <kkojima@rr.iij4u.or.jp>
1950
1951 * config/tc-sh.c: Include dw2gencfi.h.
1952 (sh_cfi_frame_initial_instructions): New function.
1953 (sh_regname_to_dw2regnum): Likewise.
1954 * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Move to the end of
1955 file.
1956 (TARGET_USE_CFIPOP): Define.
1957 (tc_cfi_frame_initial_instructions): Likewise.
1958 (tc_regname_to_dw2regnum): Likewise.
1959 (DWARF2_DEFAULT_RETURN_COLUMN, DWARF2_CIE_DATA_ALIGNMENT): Likewise.
1960 * Makefile.am: Update dependencies.
1961 * Makefile.in: Regenerate.
1962
9183ed4b
AM		 19632004-03-17 Ralf Corsepius <corsepiu@faw.uni-ulm.de>
1964
1965 * configure.in: Switch sh-*-rtems* to ELF. Add sh-*-rtemscoff*.
1966 * configure: Regenerate.
1967
00bbdfe7
BW		 19682004-03-12 Bob Wilson <bob.wilson@acm.org>
1969
1970 * read.c (s_leb128): Call md_flush_pending_output.
1971
0f10071e
ML		 19722004-03-12 Michal Ludvig <mludvig@suse.cz>
1973
1974 * config/tc-i386.c (output_insn): Handle PadLock instructions.
1975 * config/tc-i386.h (CpuPadLock): New define.
1976 (CpuUnknownFlags): Added CpuPadLock.
1977
84c84759
AS		 19782004-03-07 Andreas Schwab <schwab@suse.de>
1979
1980 * doc/c-hppa.texi (HPPA Directives): Fix typo.
1981
6ec51dba
RH		 19822004-03-07 Richard Henderson <rth@redhat.com>
1983
1984 * dw2gencfi.c (output_cie): Align length to 4 byte boundary.
1985 (cfi_finish): Likewise for fde.
1986
bae25f19
L		 19872004-03-05 H.J. Lu <hongjiu.lu@intel.com>
1988
1989 * config/tc-ia64.c (md_assemble): Properly handle NULL
1990 align_frag.
1991 (ia64_handle_align): Don't abort if failed to add a stop bit.
1992
ab509236
L		 19932004-03-04 H.J. Lu <hongjiu.lu@intel.com>
1994
1995 * Makefile.in: Regenerated.
1996 * aclocal.m4: Likewise.
1997 * configure: Likewise.
1998 * doc/Makefile.in: Likewise.
1999
d9201763
L		 20002004-03-03 H.J. Lu <hongjiu.lu@intel.com>
2001
2002 * config/tc-ia64.c (dot_align): New.
2003 (ia64_do_align): Make it static.
2004 (md_pseudo_table): Use "dot_align" for "align".
2005 (ia64_md_do_align): Don't set align_frag here.
2006 (ia64_handle_align): Add a stop bit to the previous bundle if
2007 needed.
2008
2009 * config/tc-ia64.h (ia64_do_align): Removed.
2010
ae51a426
JR		 20112003-03-03 Andrew Stubbs <andrew.stubbs@superh.com>
2012
2013 * config/tc-sh.c (md_parse_option): Add -isa=sh4-nofpu and
2014 -isa=sh4-nommu-nofpu options. Adjust help messages accordingly.
2015 (sh_elf_final_processing): Output BFD type sh4_nofpu if that is
2016 the most general type or the user specifically requested it.
2017 (md_assemble): Add a new error message for when an instruction
2018 is understood, but is not allowed due to an -isa option.
2019
9545c4ce
L		 20202004-03-02 H.J. Lu <hongjiu.lu@intel.com>
2021
2022 * config/tc-ia64.c (align_frag): New.
2023 (md_assemble): Set the tc_frag_data field in align_frag for
2024 IA64_OPCODE_FIRST instructions.
2025 (ia64_md_do_align): Set align_frag.
2026 (ia64_handle_align): Add a stop bit if needed.
2027
2028 * config/tc-ia64.h (TC_FRAG_TYPE): New.
2029 (TC_FRAG_INIT): New.
2030
676a64f4
RS		 20312004-03-01 Richard Sandiford <rsandifo@redhat.com>
2032
2033 * config/tc-frv.c (fr400_audio): New variable.
2034 (md_parse_option, md_show_usage): Add -mcpu=fr405 and -mcpu=fr450.
2035 (md_parse_option): Set fr400_audio for -mcpu=fr400 and -mcpu=fr405.
2036 (target_implements_insn_p): New function.
2037 (md_assemble): Report an error if the processor doesn't implement
2038 the instruction.
2039
bd337dde
NC		 20402004-02-27 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2041
2042 * config/tc-m32r.c (md_longopts): Added -no-bitinst option.
2043 (md_parse_option): Ditto.
2044 (OPTION_NO_SPECIAL_M32R): Added.
2045 (md_show_usage): Document it.
2046 (enable_speial_m32r): Changed a default value from 0 to 1.
2047 * doc/c-m32r.texi: Document the -no-bitinst option.
2048
78878175
NC		 20492004-02-27 Nick Clifton <nickc@redhat.com>
2050
2051 * config/tc-sh.c (get_operand): Revert previous delta.
2052 (tc_gen_reloc): Check for an unknown reloc type before processing
2053 the addend.
2054
1d6d62a4
MS		 20552004-02-27 Hannes Reinecke <hare@suse.de>
2056
2057 * config/tc-s390.c (s390_insn): Correct range check for opcode in
2058 .insn pseudo operation.
2059
4aa3e325
NC		 20602004-02-27 Anil Paranjpe <anilp1@kpitcummins.com>
2061
2062 * config/tc-sh.c (get_operand): In case of #Imm, check has been
2063 added for wrong syntax.
2064
73369e65
EC		 20652004-02-26 Eric Christopher <echristo@redhat.com>
2066
2067 * config/tc-mips.c (mips_dwarf2_addr_size): New.
2068 * config/tc-mips.h (DWARF2_ADDR_SIZE): Use.
2069
6a5709a5
JR		 20702004-02-26 Andrew Stubbs <andrew.stubbs@superh.com>
2071
2072 * config/tc-sh.c (build_Mytes): Add REG_N_D and REG_N_B01
2073 nibble types to assembler.
2074
95385380
FF		 20752004-02-25 Fred Fish <fnf@redhat.com>
2076
2077 * config/tc-iq2000.c: Add missing \n\ in multiline string literal.
2078
b5e0fabd
JW		 20792004-02-20 James E Wilson <wilson@specifixinc.com>
2080
2081 * config/tc-ia64.c (slot_index): New arg before_relax. Use instead of
2082 finalize_syms.
2083 (fixup_unw_records): New arg before_relax. Pass to slot_index.
2084 (ia64_estimate_size_before_relax): New.
2085 (ia64_convert_frag): Pass 0 to fixup_unw_records. Add comment.
2086 (generate_unwind_image): Pass 1 to fixup_unw_records.
2087 * config/tc-ia64.h (ia64_estimate_size_before_relax): Declare.
2088 (md_estimate_size_before_relax): Call ia64_estimate_size_before_relax.
2089
2f01ffbf
JJ		 20902004-02-19 Jakub Jelinek <jakub@redhat.com>
2091
2092 * stabs.c (generate_asm_file): Avoid warning about use of
2093 uninitialized variable.
2094
9c59842f
JW		 20952004-02-18 David Mosberger <davidm@hpl.hp.com>
2096
2097 * config/tc-ia64.c (ia64_flush_insns): In addition to prologue,
2098 body, and endp, allow unwind records which do not have a "t"
2099 (time/instruction) field.
2100
34920d91
NC		 21012004-02-17 Petko Manolov <petkan@nucleusys.com>
2102
2103 * config/tc-arm.c (do_mav_dspsc_1): Correct offset of CRn.
2104 (do_mav_dspsc_2): Likewise.
2105 Fix accumulator registers move opcodes.
2106
4dde8e61 21072004-02-13 Hannes Reinecke <hare@suse.de>
13ef8878 2108 Jakub Jelinek <jakub@redhat.com>
4dde8e61
AJ		 2109
2110 * dwarf2dbg.c (get_filenum): Do not read beyond allocated memory.
2111
936cf02e
JW		 21122004-02-10 Steve Ellcey <sje@cup.hp.com>
2113
2114 * config/tc-ia64.h (ia64_frob_symbol): New declaration.
2115 (tc_frob_symbol): New macro definition.
2116 * config/tc-ia64.c (ia64_frob_symbol): New routine.
2117
e41f12f3
DJ		 21182004-02-09 Daniel Jacobowitz <drow@mvista.com>
2119
2120 * config/tc-arm.c (md_begin): Mark .note.gnu.arm.ident as
2121 read-only.
2122
a8a3b3b2
NS		 21232004-02-09 Nathan Sidwell <nathan@codesourcery.com>
2124
2125 * read.h (IGNORE_OPCODE_CASE): Do not define. Replace with ...
2126 (TC_CASE_SENSITIVE): ... this.
2127 * read.c: Replace IGNORE_OPCODE_CASE with TC_CASE_SENSITIVE.
2128 * doc/internals.texi (TC_CASE_SENSITIVE): Document.
2129
c97b7ef6
JW		 21302004-02-06 James E Wilson <wilson@specifixinc.com>
2131
2132 * config/tc-ia64.c (dot_endp): Delete call to output_endp.
2133 (generate_unwind_image): Re-add it here.
2134
e4475e39
NS		 21352004-02-06 Nathan Sidwell <nathan@codesourcery.com>
2136
2137 * dwarf2dbg.c (DWARF2_ADDR_SIZE): Remove trailing ';'
2138 * read.h (SKIP_WHITESPACE): Turn into an expression.
2139 * read.c (read_a_source_file): A pseudo is removed by having a
2140 NULL handler.
2141
5738bc24
JW		 21422004-02-05 James E Wilson <wilson@specifixinc.com>
2143
2144 * config/tc-ia64.c (output_endp): New.
2145 (count_bits): Delete.
2146 (ia64_flush_insns, process_one_record, optimize_unw_records): Handle
2147 endp unwind records.
2148 (fixup_unw_records): Handle endp unwind records. Delete code for
2149 shortening prologue regions not followed by a body record.
2150 (dot_endp): Call add_unwind_entry to emit endp unwind record.
2151 * config/tc-ia64.h (unw_record_type): Add endp.
2152
d6e78c11
JW		 21532004-02-03 James E Wilson <wilson@specifixinc.com>
2154
2155 * config/tc-ia64.c (ia64_convert_frag): Call md_number_to_chars to
2156 fill padding bytes with zeroes.
2157 (emit_one_bundle): New locals last_ptr, end_ptr. Rewrite code that
2158 sets unwind_record slot_number and slot_frag fields.
2159
f6a22291
MR		 21602004-02-02 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2161
2162 * config/tc-mips.c (add_got_offset_hilo): New function.
2163 (macro): Use load_register() and add_got_offset_hilo() to load
2164 constants instead of hardcoding code sequences throughout.
2165
127cab00
L		 21662004-01-28 H.J. Lu <hongjiu.lu@intel.com>
2167
2168 * config/tc-ia64.c (emit_one_bundle): Add proper indentation.
2169
6358301e
NC		 21702004-01-26 Bernardo Innocenti <bernie@develer.com>
2171
2172 * config/tc-m68k.h (EXTERN_FORCE_RELOC): Handle m68k-uclinux specially,
2173 like m68k-elf.
2174 * config/tc-m68k.c (RELAXABLE_SYMBOL): Use EXTERN_FORCE_RELOC instead
2175 of hard-coded test for TARGET_OS=elf.
2176
c8978940
CD		 21772004-01-24 Chris Demetriou <cgd@broadcom.com>
2178
2179 * config/tc-mips.c (hilo_interlocks): Change definition
2180 so that MIPS32, MIPS64 and later ISAs are included, along with
2181 the already-included machines. Update comments.
2182
c3ba240c
DJ		 21832004-01-23 Daniel Jacobowitz <drow@mvista.com>
2184
2185 * config/tc-arm.c (tc_gen_reloc): Improve error message for
2186 undefined local labels.
2187
e864ceca
RS		 21882004-01-23 Richard Sandiford <rsandifo@redhat.com>
2189
2190 * config/tc-mips.c (load_address, macro): Update comments about
2191 NewABI GP relaxation.
2192
67c0d1eb
RS		 21932004-01-23 Richard Sandiford <rsandifo@redhat.com>
2194
2195 * config/tc-mips.c (macro_build): Remove place and counter arguments.
2196 (mips_build_lui, macro_build_ldst_constoffset): Likewise.
2197 (mips16_macro_build, macro_build_jalr): Remove counter argument.
2198 (set_at, load_register, load_address, move_register): Likewise.
2199 (load_got_offset, add_got_offset): Likewise.
2200 Update all calls and tidy accordingly.
2201
584892a6
RS		 22022004-01-23 Richard Sandiford <rsandifo@redhat.com>
2203
2204 * config/tc-mips.c (RELAX_ENCODE): Remove WARN argument.
2205 (RELAX_FIRST, RELAX_SECOND): Turn into 8-bit quantities.
2206 (RELAX_USE_SECOND): Bump to 0x10000.
2207 (RELAX_SECOND_LONGER, RELAX_NOMACRO, RELAX_DELAY_SLOT): New flags.
2208 (mips_macro_warning): New variable.
2209 (md_assemble): Wrap macro expansion in macro_start() and macro_end().
2210 (s_cpload, s_cpsetup, s_cprestore, s_cpreturn): Likewise.
2211 (relax_close_frag): Set mips_macro_warning.first_frag. Adjust use
2212 of RELAX_ENCODE.
2213 (append_insn): Update mips_macro_warning.sizes.
2214 (macro_start, macro_warning, macro_end): New functions.
2215 (macro_build): Don't emit warnings here.
2216 (macro_build_lui, md_estimate_size_before_relax): ...or here.
2217 (md_convert_frag): Check for cases where one macro alternative
2218 needs a warning and the other doesn't. Emit a warning if the
2219 longer sequence was chosen.
2220
4d7206a2
RS		 22212004-01-23 Richard Sandiford <rsandifo@redhat.com>
2222
2223 * config/tc-mips.h (tc_frag_data_type, TC_FRAG_TYPE): Remove.
2224 * config/tc-mips.c (RELAX_ENCODE): Take three arguments: the size of
2225 the first sequence, the size of the second sequence, and a flag
2226 that says whether we should warn.
2227 (RELAX_OLD, RELAX_NEW, RELAX_RELOC[123]): Delete.
2228 (RELAX_FIRST, RELAX_SECOND): New.
2229 (mips_relax): New variable.
2230 (relax_close_frag, relax_start, relax_switch, relax_end): New fns.
2231 (append_insn): Remove "place" argument. Use mips_relax.sequence
2232 rather than "place" to check whether we're expanding the second
2233 alternative of a relaxable macro. Remove redundant check for
2234 branch relaxation. If generating a normal insn, and there
2235 is not enough room in the current frag, call relax_close_frag()
2236 to close it. Update mips_relax.sizes[]. Emit fixups for the
2237 second version of a relaxable macro. Record the first relaxable
2238 fixup in mips_relax. Remove tc_gen_reloc workaround.
2239 (macro_build): Remove all uses of "place". Use mips_relax.sequence
2240 in the same way as in append_insn.
2241 (mips16_macro_build): Remove "place" argument.
2242 (macro_build_lui): As for macro_build. Don't drop the add_symbol
2243 when generating the second version of a relaxable macro.
2244 (load_got_offset, add_got_offset): New functions.
2245 (load_address, macro): Use new relaxation machinery. Remove
2246 tc_gen_reloc workarounds.
2247 (md_estimate_size_before_relax): Set RELAX_USE_SECOND if the second
2248 version of a relaxable macro is needed. Return -RELAX_SECOND if the
2249 first version is needed.
2250 (tc_gen_reloc): Remove relaxation handling.
2251 (md_convert_frag): Go through the fixups for a relaxable macro and
2252 mark those that belong to the unneeded alternative as done. If the
2253 second alternative is needed, adjust the fixup addresses to account
2254 for the deleted first alternative.
2255
0530d30a
RS		 22562004-01-23 Richard Sandiford <rsandifo@redhat.com>
2257
2258 * frags.h (frag_room): Declare.
2259 * frags.c (frag_room): New function.
2260 * doc/internals.texi: Document it.
2261
048cdf86
TS		 22622004-01-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
2263
2264 * config/tc-mips.c (append_insn): Don't do r3900 interlock
2265 optimization for -mtune=r3900, as this will break on other CPUs.
2266
28d39d1a
NC		 22672004-01-11 Tom Rix <tcrix@worldnet.att.net>
2268
2269 * config/tc-m68hc11.c (build_indexed_byte): movb and movw cannot
73369e65 2270 be relaxed, use fixup.
28d39d1a
NC		 2271 (md_apply_fix3): Use 5 bit reloc from movb and movw fixup.
2272
26664553
JJ		 22732004-01-19 Jakub Jelinek <jakub@redhat.com>
2274
2275 * config/tc-sparc.c (sparc_ip): Disallow %f32-%f63 for single
2276 precision operands.
2277
895921c9
MR		 22782004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2279
2280 * config/tc-mips.c (append_insn): Properly detect variant frags
2281 that preclude swapping of relaxed branches. Correctly swap
2282 instructions between frags when dealing with relaxed branches.
2283
f8b73030
MR		 22842004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2285
2286 * acinclude.m4: Quote names of macros to be defined by AC_DEFUN
2287 throughout.
2288 * aclocal.m4: Regenerate.
2289 * configure: Regenerate.
2290
ca9a79a1 22912004-01-12 Anil Paranjpe <anilp1@KPITCummins.com>
73369e65 2292
ca9a79a1
NC		 2293 * config/tc-h8300.c (build_bytes): Apply relaxation to bit
2294 manipulation insns.
2295
cc3d92a5
RS		 22962004-01-12 Richard Sandiford <rsandifo@redhat.com>
2297
2298 * config/tc-mips.c (macro_build_jalr): When adding an R_MIPS_JALR
2299 reloc, reserve space for the delay slot as well as the jalr itself.
2300
e45d0630
PB		 23012004-01-09 Paul Brook <paul@codesourcery.com>
2302
2303 * config/tc-arm.c (do_vfp_reg2_from_sp2): Rename from do_vfp_sp_reg2.
2304 (do_vfp_sp2_from_reg2): New function.
2305 (insns): Use them.
2306 (do_vfp_dp_from_reg2): Check return values properly.
2307
81912461
ILT		 23082004-01-08 Ian Lance Taylor <ian@wasabisystems.com>
2309
78849248
ILT		 2310 * config/tc-mips.c (warn_nops): Remove static variable.
2311 (macro): Remove test of warn_nops.
2312 (md_shortops): Remove 'n'.
2313 (md_parse_option): Remove 'n' case.
2314 (md_show_usage): Remove -n.
2315 * doc/as.texinfo (Overview): Remove MIPS -n option.
2316 * doc/c-mips.texi (MIPS Opts): Remove mention -n.
2317 * NEWS: Mention removal of MIPS -n option.
2318
81912461
ILT		 2319 * config/tc-mips.c (ISA_HAS_COPROC_DELAYS): Remove.
2320 (cop_interlocks): Check ISA level.
2321 (cop_mem_interlocks): Define.
2322 (reg_needs_delay): Check cop_interlocks rather than
2323 ISA_HAS_COPROC_DELAYS.
2324 (append_insn): Likewise. Use cop_mem_interlocks rather than
2325 directly checking mips_opts.isa.
2326 (mips_emit_delays): Likewise.
2327
73f20958
L		 23282004-01-07 H.J. Lu <hongjiu.lu@intel.com>
2329
2330 * config/tc-ia64.c (unwind): Move next_slot_number and
2331 next_slot_frag to ...
2332 (unw_rec_list): Here.
2333 (free_list_records): Removed.
2334 (output_unw_records): Likewise.
2335 (generate_unwind_image): Make it void.
2336 (alloc_record): Initialize next_slot_number and next_slot_frag.
2337 (slot_index): Take .org, .space and .align into account.
2338 (fixup_unw_records): Don't set slot_number to 0. Use
2339 list->next_slot_number and list->next_slot_frag instead of
2340 unwind.next_slot_number and unwind.next_slot_frag.
2341 (ia64_convert_frag): New.
2342 (generate_unwind_image): Generate a rs_machine_dependent frag
2343 for unwind record.
2344 (emit_one_bundle): Use list->next_slot_number and
2345 list->next_slot_frag instead of unwind.next_slot_number and
2346 unwind.next_slot_frag.
2347
2348 * config/tc-ia64.h (md_convert_frag): Defined as
2349 ia64_convert_frag.
2350 (md_estimate_size_before_relax): Defined as (f)->fr_var.
2351
99d09cdb
AO		 23522004-01-06 Alexandre Oliva <aoliva@redhat.com>
2353
2354 2003-12-19 Alexandre Oliva <aoliva@redhat.com>
2355 * config/tc-frv.h (md_apply_fix3): Don't define.
2356 * config/tc-frv.c (md_apply_fix3): New. Shift/truncate %hi/%lo
2357 operands.
2358 * config/tc-frv.h (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
2359 2003-10-07 Alexandre Oliva <aoliva@redhat.com>
2360 * config/tc-frv.c (line_separator_chars): Add `!'.
2361 2003-09-19 Alexandre Oliva <aoliva@redhat.com>
2362 * config/tc-frv.c (md_assemble): Clear insn upfront.
2363 2003-09-18 Alexandre Oliva <aoliva@redhat.com>
2364 * config/tc-frv.c (OPTION_FDPIC): New macro.
2365 (md_longopts): Add mfdpic.
2366 (md_parse_option): Handle it.
2367 2003-08-04 Alexandre Oliva <aoliva@redhat.com>
2368 * config/tc-frv.c (md_cgen_lookup_reloc) <FRV_OPERAND_D12,
2369 FRV_OPERAND_S12>: Use reloc type encoded in fix-up.
2370 (frv_pic_ptr): Parse funcdesc.
2371
4a4c4a1d
MR		 23722004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2373
2374 * doc/as.texinfo: Let texi2pod parse asconfig.texi and
2375 gasver.texi. Remove duplicate symbol definitions for texi2pod.
2376
2251dd89
MR		 23772004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2378
2379 * Makefile.am (Makefile): Move the dependency on
2380 $(BFDDIR)/configure.in to...
2381 (CONFIG_STATUS_DEPENDENCIES): ... here.
2382 (AUTOMAKE_OPTIONS): Require automake 1.8.
2383 * Makefile.in: Regenerate.
2384 * doc/Makefile.am (BASEDIR, BFDDIR): Define.
2385 (CONFIG_STATUS_DEPENDENCIES): Add a dependency on
2386 $(BFDDIR)/configure.in here as well.
2387 * doc/Makefile.in: Regenerate.
2388
c6581168
MR		 23892004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2390
2391 * Makefile.am (install, install-info, RECURSIVE_TARGETS): Remove.
2392 * Makefile.in: Regenerate.
2393 * aclocal.m4: Regenerate.
2394 * doc/Makefile.am (install, install-info): Remove.
2395 (install-data-local): A new hook for install-info.
2396 (AUTOMAKE_OPTIONS): Require automake 1.8.
2397 * doc/Makefile.in: Regenerate.
2398
303e7b79
NC		 23992004-01-02 Nutan Singh <nutan@kpitcummins.com>
2400
2401 * doc/c-sh.texi: Update description about floating point behavior
2402 of SH family.
2403
b50c15dc
NC		 24042004-01-02 Bernardo Innocenti <bernie@develer.com>
2405
2406 * configure.in: Add m68k-uClinux target.
2407 * configure: Regenerate.
7de9afa2 2408
c9e214e5 2409For older changes see ChangeLog-0203
08d56133
NC		 2410\f
2411Local Variables:
2412mode: change-log
2413left-margin: 8
2414fill-column: 74
2415version-control: never
2416End:
GDB Binutils - Deliverables
This page took 0.290069 seconds and 4 git commands to generate.