1 2002-09-11 Jakub Jelinek <jakub@redhat.com>
3 * config/tc-i386.c (md_apply_fix3): Allow addend for
4 BFD_RELOC_386_TLS_LDO_32, BFD_RELOC_386_TLS_LE and
5 BFD_RELOC_386_TLS_LE_32.
7 2002-09-05 Jeff Law <law@redhat.com>
9 * config/tc-hppa.c (md_apply_fix3): Don't set fx_done for
10 marker relocations such as ENTRY/EXIT.
11 * config/tc-hppa.h (MD_APPLY_SYM_VALUE): Definition applies
12 to both OBJ_ELF and OBJ_SOM.
14 2002-09-05 Alan Modra <amodra@bigpond.net.au>
16 * doc/internals.texi (md_apply_fix3): Expand.
17 (TC_VALIDATE_FIX, TC_FORCE_RELOCATION, TC_FORCE_RELOCATION_ABS,
18 TC_FORCE_RELOCATION_LOCAL, TC_FORCE_RELOCATION_SUB_SAME,
19 TC_FORCE_RELOCATION_SUB_ABS, TC_FORCE_RELOCATION_SUB_LOCAL,
20 TC_VALIDATE_FIX_SUB, MD_APPLY_SYM_VALUE, S_FORCE_RELOC,
21 EXTERN_FORCE_RELOC): Document.
22 (TC_HANDLES_FX_DONE, obj_fix_adjustable): Remove.
23 * as.h: Don't include struc-symbol.h for arc.
25 * cgen.c (gas_cgen_md_apply_fix3): Remove *valP fudges and code to
26 subtract absolute symbol.
27 * obj.h (struct format_ops): Add frob_file_before_fix.
28 * subsegs.c (section_symbol): Set BSF_SECTION_SYM flag.
29 * symbols.c (S_FORCE_RELOC): New function.
30 * symbols.h (S_FORCE_RELOC): Declare.
31 * write.c (TC_FORCE_RELOCATION): Change default.
32 (TC_FORCE_RELOCATION_ABS): Define.
33 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
34 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
35 (TC_FORCE_RELOCATION_SECTION): Don't define.
36 (TC_FORCE_RELOCATION_SUB_SAME): Define this instead.
37 (TC_FORCE_RELOCATION_SUB_ABS): Define.
38 (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
39 (TC_VALIDATE_FIX_SUB): Define.
40 (TC_FIX_ADJUSTABLE): Don't define.
41 (MD_APPLY_SYM_VALUE): Define this instead.
42 (abs_section_sym): New variable.
43 (adjust_reloc_syms): Use S_FORCE_RELOC. Remove obj_fix_adjustable
44 call. Don't symbol_mark_used_in_reloc here. Simplify link_once tests.
45 Don't put the absolute section sym on fixups here.
46 (fix_segment): New function.
47 (write_relocs): Don't call fixup_segment from here.
48 (write_object_file): Instead call tc_frob_file_before_fix,
49 obj_frob_file_before_fix, and fix_segment prior to symbol table code.
50 Don't output the absolute section symbol.
51 (fixup_segment): Rewrite.
52 * write.h (abs_section_sym): Declare.
53 * config/obj-aout.c (obj_aout_frob_file_before_fix): Rename from
55 (aout_format_ops): Adjust to suit.
56 * config/obj-aout.h (obj_frob_file): Don't define.
57 (obj_frob_file_before_fix): Define.
58 (obj_aout_frob_file_before_fix): Rename from obj_aout_frob_file.
59 (S_FORCE_RELOC): Define.
60 * config/obj-bout.h (S_FORCE_RELOC): Define.
61 * config/obj-coff.c (coff_format_ops): Init new field.
62 * config/obj-coff.h: Formatting fixes.
63 (obj_sec_sym_ok_for_reloc): Define.
64 (S_FORCE_RELOC): Define.
65 * config/obj-ecoff.c (ecoff_frob_file_before_fix): Split out ..
66 (ecoff_frob_file): .. from here.
67 (ecoff_format_ops): Add new function.
68 * config/obj-ecoff.h (ecoff_frob_file_before_fix): Declare.
69 (obj_frob_file_before_fix): Define.
70 * config/obj-elf.c (elf_format_ops): Init new field.
71 * config/obj-elf.h (obj_sec_sym_ok_for_reloc): Expand comment.
72 * config/obj-ieee.h: Formatting fixes.
73 (S_FORCE_RELOC): Define.
74 * config/obj-multi.h (obj_frob_file_before_fix): Define.
75 * config/obj-vms.h (S_FORCE_RELOC): Define.
76 * config/tc-alpha.c (md_apply_fix3): Correct GPDISP comment.
77 (alpha_force_relocation): Use S_FORCE_RELOC, and don't return 0
78 for BFD_RELOC_32 and BFD_RELOC_64.
79 (alpha_fix_adjustable): Remove extern and weak tests.
80 (alpha_before_fix): Rename from alpha_adjust_symtab.
81 (alpha_adjust_relocs): Rename from alpha_adjust_symtab_relocs.
82 * config/tc-alpha.h (struct fix, struct alpha_reloc_tag): Declare.
83 (TC_VALIDATE_FIX): Tweak param name.
84 (TC_FORCE_RELOCATION, tc_fix_adjustable): Likewise.
85 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
86 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
87 (MD_APPLY_SYM_VALUE): Define.
88 (tc_adjust_symtab): Don't define.
89 (alpha_adjust_symtab): Don't declare.
90 (tc_frob_file_before_fix): Define.
91 (alpha_before_fix): Declare.
92 (TC_INIT_FIX_DATA): Tweak param names.
93 * config/tc-arc.c: Include "struc-symbol.h".
94 (md_pcrel_from): Remove undefined sym fudge.
95 (md_apply_fix3): Remove *valP fudges and code to subtract abs sym.
96 Don't set fx_addnumber.
97 (tc_gen_reloc): Remove spurious fx_addnumber comment.
98 * config/tc-arc.h (MD_APPLY_SYM_VALUE): Define.
99 (EXTERN_FORCE_RELOC): Define.
100 * config/tc-arm.c (md_apply_fix3 <case BFD_RELOC_ARM_GOTPC>): Remove.
101 (tc_gen_reloc): Fudge ARM_GOTPC addend.
102 (arm_validate_fix): Return void.
103 (arm_fix_adjustable <elf version>): Remove extern and weak tests.
104 Add plt and got reloc tests.
105 (arm_force_relocation): Call S_FORCE_RELOC.
106 * config/tc-arm.h (struct fix): Forward declare.
107 (TC_VALIDATE_FIX): No longer set add_symbolP.
108 (arm_validate_fix): Adjust declaration.
109 (TC_FORCE_RELOCATION <pe version): Call S_FORCE_RELOC.
110 (TC_FORCE_RELOCATION): Tweak param name.
111 (TC_INIT_FIX_DATA): Likewise.
112 (obj_fix_adjustable): Don't define.
113 (tc_fix_adjustable): Call arm_fix_adjustable.
114 (TC_FIX_ADJUSTABLE): Don't define.
115 (MD_APPLY_SYM_VALUE): Define.
116 (EXTERN_FORCE_RELOC): Define.
117 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
118 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
119 * config/tc-avr.c (md_apply_fix3): Don't cast valP pointer type.
120 Remove *valP fudges and code to subtract abs sym. Don't set
122 * config/tc-avr.h: Formatting.
123 (EXTERN_FORCE_RELOC): Define.
124 (MD_APPLY_SYM_VALUE): Define.
125 * config/tc-cris.c (tc_gen_reloc): Don't use fx_addnumber.
126 (md_apply_fix3): Remove code to subtract abs sym.
127 (md_cris_force_relocation): Update comment. Call S_FORCE_RELOC.
128 * config/tc-cris.h (TC_FORCE_RELOCATION): Tweak param name.
129 (IS_CRIS_PIC_RELOC, tc_fix_adjustable): Likewise.
130 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
131 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
132 (tc_fix_adjustable): Remove extern, weak tests.
133 * config/tc-d10v.c (tc_gen_reloc): Don't fiddle address of
134 BFD_RELOC_VTABLE_INHERIT relocs. Don't use fx_addnumber.
135 (md_apply_fix3): Don't cast valP pointer type. Remove *valP fudges
136 and code to subtract abs sym.
137 (d10v_fix_adjustable): Remove extern, weak, SEC_MERGE tests.
138 (d10v_force_relocation): Call S_FORCE_RELOC.
139 * config/tc-d10v.h: Don't include write.h.
140 (struct fix): Instead, forward declare.
141 (MD_PCREL_FROM_SECTION): Tweak param names.
142 (md_pcrel_from_section): Use "struct fix" instead of "fixS".
143 (d10v_fix_adjustable): Likewise.
144 (obj_fix_adjustable): Don't define.
145 (tc_fix_adjustable): Define.
146 (TC_FORCE_RELOCATION): Tweak param name.
147 (MD_APPLY_SYM_VALUE): Define.
148 (EXTERN_FORCE_RELOC): Define.
149 * config/tc-d30v.c (tc_gen_reloc): Don't use fx_addnumber.
150 (md_apply_fix3): Don't cast valP pointer type. Remove *valP fudges
151 and code to subtract abs sym.
152 * config/tc-d30v.h: Don't include write.h.
153 (struct fix): Instead, forward declare.
154 (MD_PCREL_FROM_SECTION): Tweak param names.
155 (md_pcrel_from_section): Use "struct fix" instead of "fixS".
156 (MD_APPLY_SYM_VALUE): Define.
157 (EXTERN_FORCE_RELOC): Define.
158 * config/tc-dlx.c (md_dlx_force_relocation): Call S_FORCE_RELOC.
159 (md_dlx_fix_adjustable): Don't test for NULL fx_addsy.
160 (md_apply_fix3): No need to test fx_pcrel before setting fx_done.
161 (tc_gen_reloc): Don't fiddle with BFD_RELOC_VTABLE_INHERIT relocs.
162 Adjust the address for BFD_RELOC_VTABLE_ENTRY, not the addend.
163 Don't use fx_addnumber.
164 * config/tc-dlx.h: Don't include write.h.
165 (md_convert_frag): We're not alpha twice over.
166 (EXTERN_FORCE_RELOC): Define.
167 (obj_fix_adjustable): Don't define.
168 (tc_fix_adjustable): Define.
169 (TC_FIX_ADJUSTABLE): Don't define.
170 (MD_APPLY_SYM_VALUE): Define.
171 (LOCAL_LABELS_DOLLAR): Undef always.
172 * config/tc-fr30.c (fr30_force_relocation): Call S_FORCE_RELOC.
173 (fr30_fix_adjustable): Remove extern, weak tests.
174 * config/tc-fr30.h (MD_APPLY_SYM_VALUE): Define.
175 (obj_fix_adjustable): Don't define.
176 (tc_fix_adjustable): Define.
177 (struct fix): Forward declare.
178 (MD_PCREL_FROM_SECTION): Tweak param name. Remove duplicate.
179 * config/tc-frv.c (frv_force_relocation): Call S_FORCE_RELOC.
180 (frv_fix_adjustable): Don't do extern, weak tests.
181 * config/tc-frv.h (MD_APPLY_SYM_VALUE): Define.
182 (TC_FIX_ADJUSTABLE): Don't define.
183 (obj_fix_adjustable): Don't define.
184 (tc_fix_adjustable): Define.
185 (struct fix): Forward declare.
186 (MD_PCREL_FROM_SECTION): Tweak param name. Remove duplicate.\
187 * config/tc-h8300.c (md_apply_fix3): Don't cast valP pointer type.
188 * config/tc-h8300.h (struct fix): Forward declare.
189 (EXTERN_FORCE_RELOC): Define.
190 * config/tc-hppa.c (md_apply_fix3): Remove subtraction of sym value.
191 (hppa_fix_adjustable): Don't test extern or weak. Don't
192 symbol_mark_used_in_reloc.
193 (hppa_force_relocation): Use S_FORCE_RELOC instead of S_IS_EXTERNAL
195 * config/tc-hppa.h (EXTERN_FORCE_RELOC): Define
196 (TC_FORCE_RELOCATION, TC_INIT_FIX_DATA): Tweak param name.
197 (TC_FORCE_RELOCATION_SECTION): Don't define.
198 (MD_APPLY_SYM_VALUE): Define.
199 * config/tc-i370.c (md_apply_fix3): Remove *valP fudges and code
201 * config/tc-i370.h: Remove ifdef OBJ_ELF tests.
202 (MD_APPLY_SYM_VALUE): Define.
203 (MD_PCREL_FROM_SECTION): Tweak param name.
204 * config/tc-i386.c (tc_i386_fix_adjustable): Remove weak, extern tests.
205 (md_apply_fix3): Test fx_addsy, not fx_pcrel.
206 (i386_force_relocation): New function.
207 * config/tc-i386.h (TC_COFF_FIX2RTYPE): Tweak param name.
208 (TC_VALIDATE_FIX): Likewise.
209 (TC_FIX_ADJUSTABLE): Don't define.
210 (MD_APPLY_SYM_VALUE): Define.
211 (TC_FORCE_RELOCATION <elf version>): Call i386_force_relocation.
212 (i386_force_relocation): Declare.
213 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
214 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
215 (TC_FORCE_RELOCATION <coff version>): Call S_FORCE_RELOC.
216 * config/tc-i860.c (md_apply_fix3): Don't cast valP pointer type.
217 * config/tc-i860.h (MD_APPLY_SYM_VALUE): Define.
218 (EXTERN_FORCE_RELOC): Define.
219 * config/tc-i960.c (reloc_callj): Remove declaration. Return false.
220 (md_apply_fix3): Don't cast valP pointer type. Move code here from
221 old fixup_segment. No need to test fx_pcrel before setting fx_done.
222 (i960_validate_fix): Remove add_symbolPP arg and add_symbolP macro.
223 Use fx_addsy instead of add_symbolP, as_bad_where instead of as_bad.
224 Remove #if 0 code. Invert return boolean.
225 * config/tc-i960.h (TC_COUNT_RELOC): Tweak param name.
226 (TC_COFF_FIX2RTYPE, TC_ADJUST_RELOC_COUNT, TC_VALIDATE_FIX): Likewise.
227 (tc_headers_hook, tc_coff_fix2rtype): Remove declaration.
228 (tc_coff_sizemachdep): Prototype.
229 (i960_handle_align): Likewise.
230 (i960_validate_fix): Adjust declaration.
231 (reloc_callj): Likewise.
232 (EXTERN_FORCE_RELOC): Define.
233 (TC_FORCE_RELOCATION_SUB_SAME): Define.
234 (TC_FORCE_RELOCATION_ABS): Define.
235 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
236 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
237 (TC_FIX_ADJUSTABLE): Don't define.
238 (MD_APPLY_SYM_VALUE): Define.
239 * config/tc-ia64.c (ia64_force_relocation): Call S_FORCE_RELOC.
240 * config/tc-ia64.h (MD_APPLY_SYM_VALUE): Define.
241 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
242 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
243 * config/tc-ip2k.c (ip2k_force_relocation): Call S_FORCE_RELOC.
244 * config/tc-ip2k.h (MD_APPLY_FIX3): Don't define.
245 (MD_APPLY_SYM_VALUE): Define.
246 (EXTERN_FORCE_RELOC): Define.
247 (TC_FORCE_RELOCATION): Tweak param name.
248 * config/tc-m32r.c (m32r_force_relocation): Call S_FORCE_RELOC.
249 (m32r_fix_adjustable): Don't test extern, weak.
250 * config/tc-m32r.h (MD_PCREL_FROM_SECTION): Tweak param name.
251 (MD_APPLY_SYM_VALUE): Define.
252 (obj_fix_adjustable): Don't define.
253 (tc_fix_adjustable): Define.
254 (tc_frob_file): Don't define.
255 (tc_frob_file_before_fix): Define.
256 (EXTERN_FORCE_RELOC): Define.
257 * config/tc-m68hc11.c (tc_gen_reloc): Set addend to zero. Adjust
258 BFD_RELOC_VTABLE_ENTRY address.
259 (tc_m68hc11_force_relocation): Call S_FORCE_RELOC.
260 (tc_m68hc11_fix_adjustable): Don't test relaxable_symbol.
261 (md_apply_fix3): Remove *valP fudges and code to subtract abs sym.
262 Remove duplicated fx_done code.
263 * config/tc-m68hc11.h (MD_APPLY_SYM_VALUE): Define.
264 (EXTERN_FORCE_RELOC): Define.
265 * config/tc-m68k.c (tc_m68k_fix_adjustable): Don't test
267 * config/tc-m68k.h (TC_COFF_FIX2RTYPE): Tweak param name.
268 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
269 (TC_FIX_ADJUSTABLE): Don't define.
270 (EXTERN_FORCE_RELOC): Define.
271 (MD_APPLY_SYM_VALUE): Define.
272 (TC_FORCE_RELOCATION): Call S_FORCE_RELOC.
273 * config/tc-mcore.c (md_apply_fix3): Don't cast valP pointer type.
274 Remove fx_addsy tests.
275 (mcore_force_relocation): Call S_FORCE_RELOC.
276 (mcore_fix_adjustable): Don't test fx_addsy.
277 * config/tc-mcore.h (MD_PCREL_FROM_SECTION): Tweak param name.
278 (EXTERN_FORCE_RELOC): Define.
279 (obj_fix_adjustable): Don't define.
280 (tc_fix_adjustable): Define.
281 (MD_APPLY_SYM_VALUE): Define.
282 * config/tc-mips.c (enum mips_pic_level): Move to tc-mips.h.
283 (mips_pic): No longer static.
284 (mips_force_relocation): Call S_FORCE_RELOC.
285 (mips_fix_adjustable): Remove extern, weak tests.
286 * config/tc-mips.h (enum mips_pic_level): Declare.
288 (tc_frob_file): Don't define.
289 (tc_frob_file_before_fix): Define this instead.
290 (EXTERN_FORCE_RELOC): Define.
291 * config/tc-mmix.c (md_apply_fix3): Replace real_reg_section tests
292 with reg_section tests. Set fx_done instead of calling
293 symbol_clear_used_in_reloc on bad relocs.
294 (tc_gen_reloc): Zero fx_addsy on bad relocs.
295 (mmix_force_relocation): Remove weak sym test. Call S_FORCE_RELOC.
296 (mmix_adjust_symtab): Simplify list handling. Abort on any
298 * config/tc-mmix.h (tc_fix_adjustable): Remove weak tests. Check
299 BFD_RELOC_MMIX_LOCAL.
300 (tc_frob_symbol): Keep user defined syms in reg_section. Don't punt.
301 (EXTERN_FORCE_RELOC): Define.
302 (MD_PCREL_FROM_SECTION): Tweak param name.
303 (tc_frob_file): Don't define.
304 (tc_frob_file_before_fix): Define this instead.
305 * config/tc-mn10300.c (mn10300_force_relocation): Call S_FORCE_RELOC.
306 Remove SEC_CODE checks.
307 (mn10300_fix_adjustable): Remove extern and weak tests.
308 * config/tc-mn10300.h (EXTERN_FORCE_RELOC): Define.
309 (TC_FORCE_RELOCATION): Tweak param name.
310 (obj_fix_adjustable): Don't define.
311 (TC_FORCE_RELOCATION_SUB_SAME): Define to handle SEC_CODE.
312 * config/tc-ns32k.h (TC_FIX_DATA_PRINT): Tweak param name.
313 * config/tc-openrisc.c (openrisc_force_relocation): Call S_FORCE_RELOC.
314 (openrisc_fix_adjustable): Don't test fx_addsy.
315 * config/tc-openrisc.h (MD_APPLY_SYM_VALUE): Define.
316 (obj_fix_adjustable): Don't define.
317 (tc_fix_adjustable): Define this instead.
318 (MD_PCREL_FROM_SECTION): Remove duplicate. Tweak param name.
319 * config/tc-or32.c (md_apply_fix3): Don't cast valP pointer type.
320 (tc_gen_reloc): Don't fiddle with BFD_RELOC_VTABLE_INHERIT relocs.
321 Adjust the address for BFD_RELOC_VTABLE_ENTRY, not the addend.
322 * config/tc-or32.h (EXTERN_FORCE_RELOC): Define.
323 (MD_APPLY_SYM_VALUE): Define.
324 * config/tc-pj.c (md_apply_fix3): Don't cast valP pointer type.
325 Don't subtract symbol value.
326 * config/tc-pj.h (md_pcrel_from): Tweak param name.
327 (EXTERN_FORCE_RELOC): Define.
328 (TC_FORCE_RELOCATION): Call S_FORCE_RELOC.
329 (MD_APPLY_SYM_VALUE): Define.
330 (obj_fix_adjustable): Don't define.
331 (tc_fix_adjustable): Define this instead.
332 * config/tc-ppc.c (ppc_frob_symbol <coff version>): Ignore absolute
334 (ppc_force_relocation <coff version>): Call S_FORCE_RELOC.
335 (ppc_force_relocation <elf version>): New.
336 (ppc_fix_adjustable <elf version>): Remove extern and weak tests.
337 (md_apply_fix3): Don't subtract symbol values for ELF. Update
338 comments. Don't subtract fx_subsy as that is already done.
339 * config/tc-ppc.h (tc_fix_adjustable): Tweak param name.
340 (MD_PCREL_FROM_SECTION): Likewise.
341 (TC_FORCE_RELOCATION): Define for both ELF and XCOFF as calling
342 ppc_force_relocation.
343 (TC_FORCE_RELOCATION_SECTION): Delete.
344 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
345 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
346 (MD_APPLY_SYM_VALUE): Define.
347 * config/tc-s390.c: #include "dwarf2dbg.h".
348 (s390_insn): Remove excess parens.
349 (tc_s390_fix_adjustable): Remove extern, weak, SEC_MERGE tests.
350 (tc_s390_force_relocation): Call S_FORCE_RELOC.
351 (md_apply_fix3): Add ATTRIBUTE_UNUSED on "seg". Abort when fx_subsy
352 non-NULL. Don't subtract off fx_addsy value.
353 * config/tc-s390.h (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
354 (TC_FORCE_RELOCATION, MD_PCREL_FROM_SECTION): Tweak param name.
355 (TC_FIX_ADJUSTABLE): Don't define.
356 (MD_APPLY_SYM_VALUE): Define.
357 * config/tc-sh.c (SWITCH_TABLE_CONS): Move to tc-sh.h.
358 (SWITCH_TABLE): Likewise.
359 (sh_force_relocation): Call S_FORCE_RELOC.
360 (sh_fix_adjustable): Remove "return 1" cases handled by the default.
361 Replace TC_RELOC_RTSYM_LOC_FIXUP with reloc type tests.
362 (md_apply_fix3 <case BFD_RELOC_32_PLT_PCREL>): Simplify,
363 fx_addnumber is zero on entry. Save val in fx_addnumber.
364 (tc_gen_reloc): Don't subtract fx_subsy.
365 * config/tc-sh.h (struct fix): Move.
366 (obj_fix_adjustable): Don't define.
367 (tc_fix_adjustable): Define this instead.
368 (TC_FIX_ADJUSTABLE): Don't define.
369 (MD_APPLY_SYM_VALUE): Define.
370 (SWITCH_TABLE_CONS): Define.
371 (SWITCH_TABLE): Define.
372 (TC_FORCE_RELOCATION_SUB_SAME): Define.
373 (TC_VALIDATE_FIX_SUB): Define.
374 (MD_PCREL_FROM_SECTION): Tweak param name.
375 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
376 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
377 (TC_FORCE_RELOCATION_SUB_ABS): Define.
378 * config/tc-sh64.h (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
379 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
380 (TC_FORCE_RELOCATION_SUB_SAME): Define.
381 (TC_VALIDATE_FIX_SUB): Define.
382 (MD_PCREL_FROM_SECTION): Tweak param name.
383 * config/tc-sparc.c (md_apply_fix3): Don't subtract off symbol value.
384 (tc_gen_reloc): Use S_FORCE_RELOC.
385 (elf32_sparc_force_relocation): Call S_FORCE_RELOC.
386 * config/tc-sparc.h (TC_FORCE_RELOCATION <coff version>): Remove.
387 (TC_FORCE_RELOCATION_ABS): Define this instead.
388 (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
389 (TC_FORCE_RELOCATION_LOCAL): Define this instead.
390 (tc_fix_adjustable): Remove extern and weak tests. Use S_FORCE_RELOC.
391 (MD_APPLY_SYM_VALUE): Define.
392 (TC_FIX_DATA_PRINT): Tweak param name.
393 * config/tc-tic30.c (USE_STDOUT): Don't define.
394 (md_parse_option): Remove stupid debug code.
395 (tc_gen_reloc): Don't use fx_addnumber.
396 * config/tc-v850.c (v850_fix_adjustable): Remove extern and weak tests.
397 (v850_force_relocation): Remove weak test. Call S_FORCE_RELOC.
398 * config/tc-v850.h (obj_fix_adjustable): Don't define.
399 (tc_fix_adjustable): Define this instead.
400 (TC_FIX_ADJUSTABLE): Don't define.
401 (MD_APPLY_SYM_VALUE): Define this instead.
402 (TC_FORCE_RELOCATION, MD_PCREL_FROM_SECTION): Tweak param name.
403 * config/tc-vax.h (TC_RELOC_RTSYM_LOC_FIXUP): Don't define.
404 (TC_FIX_ADJUSTABLE): Don't define.
405 (MD_APPLY_SYM_VALUE): Define this instead.
406 (tc_fix_adjustable): Remove extern and weak tests.
407 * config/tc-w65.h (struct fix): Forward declare.
408 * config/tc-xstormy16.c (xstormy16_force_relocation): Call
410 (xstormy16_fix_adjustable): Remove extern and weak tests. Don't
411 call xstormy16_force_relocation; Instead test for FPTR16 reloc.
412 (xstormy16_md_apply_fix3): Remove *valP fudges and code to subtract
414 * config/tc-xstormy16.h (MD_APPLY_FIX3): Don't define.
415 (MD_APPLY_SYM_VALUE): Define.
416 (obj_fix_adjustable): Don't define.
417 (tc_fix_adjustable): Define this instead.
418 (MD_PCREL_FROM_SECTION): Remove duplicate. Tweak param name.
420 2002-09-04 Alan Modra <amodra@bigpond.net.au>
422 * config/tc-ppc.c (ppc_frob_symbol): Formatting, warning fix.
423 (ppc_fix_adjustable <coff version>): Cleanup.
425 * config/tc-ppc.c (PPC_HIGHER, PPC_HIGHEST): Fix warning.
426 (md_parse_option): No -a64 without BFD64.
427 (ppc_set_cpu): Select appropriate cpu when ppc_obj64.
428 (ppc_arch): Use bfd_mach_rs6k for bfd_arch_rs6000.
430 2002-09-04 Nick Clifton <nickc@redhat.com>
432 * config/tc-ppc.c (md_begin): Do not insert non-BookE32
433 instructions into the hash table if the target cpu is the BookE32.
435 2002-08-31 Hans-Peter Nilsson <hp@bitrange.com>
437 * read.c (do_align): Use ATTRIBUTE_UNUSED_LABEL for label, not
440 2000-08-28 Catherine Moore <clm@redhat.com>
442 * tc-v850.c (v850_relax): Declare.
443 (v850_longcode): New routine.
444 (v850_handle_align): New routine.
445 (md_pseudo_table): Add longcall and longjump.
446 (md_parse_option): Check for relax option.
447 (tc_gen_reloc): Handle BFD_RELOC_V850_LONGCALL,
448 BFD_RELOC_V850_LONGJUMP, and BFD_RELOC_V850_ALIGN.
449 (md_apply_fix3): Likewise.
450 (v850_force_relocation): Likewise.
451 (v850_comm): Change the current section.
452 (md_assemble): Ensure that the correct value is put in the
454 (v850_sdata, v850_tdata, v850_zdata, v850_sbss, v850_tbss,
455 v850_zbss, v850_rosdata, v850_rozdata): Fix section book keeping.
456 Remove redundant v850ea support.
457 * tc-v850.h (HANDLE_ALIGN): Define.
458 (v850_handle_align): Declare.
459 * doc/c-v850.c: Document -mrelax, .longcall and .longjump.
461 2002-08-28 Svein E. Seldal <Svein.Seldal@solidas.com>
463 * configure.in: Add tic4x-coff* and c4x-coff*-coff-coff targets.
464 * configure: Regenerate.
465 * NEWS: Mention new port.
467 2002-08-28 Michael Hayes <m.hayes@elec.canterbury.ac.nz>
469 * config/obj-coff.c: Add sdef definition.
470 * config/obj-coff.h: Add tic4x include file and set
472 * config/tc-tic4x.c: New file.
473 * config/tc-tic4x.h: New file.
475 2002-08-28 Alan Modra <amodra@bigpond.net.au>
477 * write.c (BFD_FAST_SECTION_FILL): Remove unused macro.
478 (TC_ADJUST_RELOC_COUNT): Tweak param name.
479 (TC_FORCE_RELOCATION, TC_FORCE_RELOCATION_SECTION): Likewise.
480 (TC_FIX_ADJUSTABLE, MD_PCREL_FROM_SECTION): Likewise.
481 (RELOC_ENUM): Define.
482 (fix_new_internal): Use RELOC_ENUM.
483 (fix_new, fix_new_exp): Likewise.
484 (adjust_reloc_syms): Comment. Remove unnecessary tests on sym != NULL.
485 Replace gotos with continue.
486 (write_relocs): Formatting. Avoid symbol loops in
487 RELOC_EXPANSION_POSSIBLE case too. Report bfd_reloc_outofrange
488 errors, and error number in other cases.
489 (fixup_segment): Remove param names from prototype. Rename
490 "this_segment_type" to "this_segment". Update linkrelax comment.
491 Remove "size, "place" and "where" local vars. Formatting. Update
492 "no symbol" comment. Remove #if 0 and #if 1.
494 * app.c (do_scrub_chars): Don't test IGNORE_NONSTANDARD_ESCAPES. Tidy.
496 2002-08-27 Alan Modra <amodra@bigpond.net.au>
498 * dwarf2dbg.c: Always include dwarf2dbg.h.
499 (dwarf2_directive_file): Adjust dummy version args.
500 * ecoff.c (ecoff_directive_weakext): Add ATTRIBUTE_UNUSED.
501 * expr.c (clean_up_expression <O_subtract>): Allow subtraction
502 when symbol values differ.
503 * read.c (do_align): Add ATTRIBUTE_UNUSED to label.
504 (pseudo_set <O_subtract>): Remove unnecessary segment test.
505 * config/obj-bout.c (obj_pseudo_table): Warning fix.
507 2002-08-26 Alan Modra <amodra@bigpond.net.au>
509 * config/tc-w65.c (md_section_align): Fix typo.
510 (md_parse_option): Return 0, not 1.
512 2002-08-22 Nick Clifton <nickc@redhat.com>
514 * doc/as.texinfo (Section): Note that if '@' is a comment
515 character then another symbol is used to prefix the section's
518 2002-08-22 Christian Groessler <chris@groessler.org>
520 * config/tc-z8k.c (get_operands): Adjust ptr variable also in
523 2002-08-12 Graeme Peterson <gp@qnx.com>
525 * configure.in: Add support for sh-**-nto* target.
526 * configure: Regenerate.
528 2002-08-21 Nitin Gupta <niting@noida.hcltech.com>
530 * config/tc-h8300.h (TC_LINKRELAX_FIXUP): Define.
532 2002-08-21 Elena Zannoni <ezannoni@redhat.com>
534 * config/tc-ppc.c (ppc_cleanup): Do something only if format
536 (ppc_apuinfo_section_add): Define only if format is ELF.
537 (md_assemble): Emit APUinfo section only if format is ELF.
540 2002-08-21 Alan Modra <amodra@bigpond.net.au>
542 * config/tc-arc.c (md_pseudo_table <dwarf2_directive_file>): Cast.
543 * config/tc-frv.c: Likewise.
544 * config/tc-hppa.c: Likewise.
545 * config/tc-ia64.c: Likewise.
546 * config/tc-ip2k.c: Likewise.
547 * config/tc-m68hc11.c: Likewise.
548 * config/tc-m68k.c: Likewise.
549 * config/tc-mmix.c: Likewise.
550 * config/tc-mn10300.c: Likewise.
551 * config/tc-sh.c: Likewise.
552 * config/tc-sparc.c: Likewise.
553 * config/tc-v850.c: Likewise.
555 2002-08-20 Richard Sandiford <rsandifo@redhat.com>
557 * config/tc-mips.c (macro2): Implement rotates by zero using shifts
560 2002-08-19 Elena Zannoni <ezannoni@redhat.com>
562 From matthew green <mrg@redhat.com>
564 * config/tc-ppc.c (PPC_OPCODE_CLASSIC): Enable this everywhere
565 PPC_OPCODE_PPC is, except for BookE architectures.
566 (md_parse_option): Add support for -mspe.
567 (md_show_usage): Add -mspe.
568 (md_parse_option): Add support for -me500 and
569 -me500x2 to generate code for Motorola e500 core complex.
570 (md_show_usage): Add -me500 and -me500x2.
572 (PPC_APUINFO_ISEL, PPC_APUINFO_PMR, PPC_APUINFO_RFMCI,
573 PPC_APUINFO_CACHELCK, PPC_APUINFO_SPE, PPC_APUINFO_EFS,
574 PPC_APUINFO_BRLOCK): New macros.
576 (ppc_cleanup): New function.
577 (ppc_apuinfo_section_add): New function.
579 (md_assemble): Collect info and write the APUinfo section.
581 * config/tc-ppc.h (md_cleanup): Define.
582 (ppc_cleanup): Export.
583 (ELF_TC_SPECIAL_SECTIONS): Add .PPC.EMB.apuinfo section.
585 2002-08-17 Stan Cox <scox@redhat.com>
587 * config/obj-elf.c (obj_elf_change_section): Make non-static.
588 config/tc-mips.c (s_change_section): New function to support
589 IRIX .section pseudo-op.
591 2002-08-16 Nick Clifton <nickc@redhat.com>
593 * config/tc-v850.c (md_assemble): Fix assembling of "callt 0x3f".
595 2002-08-15 Alexandre Oliva <aoliva@redhat.com>
597 * config/tc-mips.c (macro_build_jalr): Make sure we generate
598 the fix-up against on the right frag.
599 (s_cpsetup): Likewise. Parse third argument as expression, to
600 handle global symbols and forward/backward labels correctly.
602 2002-08-14 Nick Clifton <nickc@redhat.com>
604 * read.c (stringer): Catch attempts to create strings in the abs
607 * config/tc-alpha.c: Fix compiling for COFF targets.
608 Some minor formatting tidyups.
610 2002-08-13 Stephane Carrez <stcarrez@nerim.fr>
612 * config/tc-m68hc11.h (MD_PCREL_FROM_SECTION): Remove.
613 (TC_HANDLES_FX_DONE): Define to let md_apply_fix3 set fx_done flag
614 according to the reloc.
615 (tc_fix_adjustable, tc_m68hc11_fix_adjustable): Define.
616 (TC_FORCE_RELOCATION): Define.
617 (tc_m68hc11_force_relocation): Declare.
619 * config/tc-m68hc11.c (md_pseudo_table): Add relax command.
620 (s_m68hc11_relax): New function for relax group.
621 (build_insn, build_jump_insn): Emit a M68HC11_RL_JUMP reloc at
622 beginning of jump instruction.
623 (md_pcrel_from): Rename from md_pcrel_from_section and fix
625 (tc-gen_reloc): Update.
626 (md_estimate_size_before_relax): Create the BFD_RELOC_16_PCREL as
628 (tc_m68hc11_force_relocation): New function, handle new relocs.
629 (tc_m68hc11_fix_adjustable): New to make sure there are enough
630 reloc for the linker relax pass.
631 (md_apply_fix3): Handle M68HC11_RL_JUMP, M68HC11_RL_GROUP
634 2002-08-13 Stephane Carrez <stcarrez@nerim.fr>
636 * config/tc-m68hc11.c (m68hc11_elf_final_processing): New function.
637 (md_pseudo_table): Add .mode, .far and .interrupt pseudo op.
638 (s_m68hc11_mode): New function for .mode pseudo op.
639 (s_m68hc11_mark_symbol): New function for .far and .interrupt
641 * config/tc-m68hc11.h (elf_tc_final_processing): Define.
642 (m68hc11_elf_final_processing): Declare.
644 2002-08-13 Stephane Carrez <stcarrez@nerim.fr>
646 * config/tc-m68hc11.c (md_begin): Take into account additional
647 page operand for call instruction.
648 (print_opcode_format): Likewise.
649 (check_range): Likewise for page range checking.
650 (get_operand): Don't skip a possible comma in operands.
651 (fixup8): Generate BFD_RELOC_M68HC11_PAGE reloc.
652 (fixup16): Likwise with BFD_RELOC_M68HC11_LO16.
653 (fixup24): New to handle call reloc.
654 (build_insn): Handle missing page operand for call instruction.
656 (md_apply_fix3): Take into account new relocs.
657 (get_operand): Fix the mode for indexed indirect addressing.
658 (build_indexed_byte): Fix post index byte for indexed indirect mode.
660 2002-08-12 Richard Sandiford <rsandifo@redhat.com>
662 * config/tc-mips.c (mips_ip): Don't work out the value of
663 constant %hi()s here.
665 2002-08-10 Alan Modra <amodra@bigpond.net.au>
667 * config/tc-i386.c (tc_i386_fix_adjustable): Test OUTPUT_FLAVOR
668 for ELF, and don't bother checking ELF relocs when non-ELF.
669 (i386_immediate): Allow absolute_section expressions for aout.
670 (i386_displacement): Likewise. Also test bfd_is_com_section.
671 (md_estimate_size_before_relax): Test OUTPUT_FLAVOR for ELF.
672 (md_apply_fix3): Hack for bfd_install_relocation when fx_pcrel,
673 not when fx_addsy. Remove dead code.
675 2002-08-09 Graeme Peterson <gp@qnx.com>
677 * configure.in: Add support for ppc-*-nto* target.
678 * configure: Regenerate.
680 2002-08-09 Alan Modra <amodra@bigpond.net.au>
682 * config/tc-i386.h: Reorganize.
684 2002-08-09 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
686 * config/tc-mips.c (macro): Handle a register plus a 16-bit
687 immediate offset in "dla" and "la" expansions.
689 2002-08-09 Alan Modra <amodra@bigpond.net.au>
691 * configure.in: bfd_gas=yes for all i386 targets. Formatting.
692 Remove "bfd_gas=yes" from target table when covered later.
693 Consolidate some entries.
694 * configure: Regenerate
696 2002-08-09 Jakub Jelinek <jakub@redhat.com>
698 * config/tc-i386.c (output_insn): Save frag_now and frag_now_fix ()
699 at start of insn, pass it to output_disp and output_imm.
700 (output_disp): Added arguments. If _GLOBAL_OFFSET_TABLE_ is seen
701 in displacement for R_386_32 reloc, use R_386_GOTPC and compute
703 (output_imm): Added arguments. Compute properly addend for
705 (md_apply_fix3): Remove R_386_GOTPC handling.
706 * testsuite/gas/i386/gotpc.s: New.
707 * testsuite/gas/i386/gotpc.d: New.
708 * testsuite/gas/i386/i386.exp: Add gotpc test.
710 2002-08-06 George France <france@handhelds.org>
712 * config/tc-alpha.c (cpu_types): Enabled ev67, ev68, -m21264a
713 and m21264b processor names and cpu types.
714 * doc/c-alpha.texi: Documented new types.
716 2002-08-06 Alan Modra <amodra@bigpond.net.au>
718 * config/tc-ppc.c (md_apply_fix3): Adjust 16 bit XCOFF reloc offset.
720 2002-08-01 Richard Sandiford <rsandifo@redhat.com>
722 * config/tc-mips.c (tc_gen_reloc): Extend GP-relative addend
723 handling to BFD_RELOC_MIPS16_GPREL.
725 2002-08-01 Nick Clifton <nickc@redhat.com>
727 * config/tc-arm.c (add_to_lit_pool): Ensure that offset to literal
728 pool is computed using signed arithmetic so that proper sign
729 extension is performed if X_add_number is a 64-bit integer.
731 2002-08-01 H.J. Lu <hjl@gnu.org>
732 Daniel Jacobowitz <drow@mvista.com>
734 * dwarf2dbg.c (dwarf2_finish): Don't emit unreferenced
735 .debug_line section unless it has line information.
737 2002-07-31 Ian Dall <ian@sibyl.beware.dropbear.id.au>
739 * config/tc-ns32k.h (md_pcrel_adjust): Supply prototype.
740 * config/tc-ns32k.c (convert_iif, md_parse_option, md_show_usage):
741 Allow default displacement size to be an option "--disp-size-default".
742 (md_number_to_disp): Make error messages include value. Use %d to
743 print integers, not %s.
744 (fix_new_ns32k): Conditionally set fx_no_overflow so we don't
745 get duplicate messages sometimes.
746 (convert_iif): Grow frag to max possible instruction size. Avoid
747 creating unnecessary fixes.
748 (md_number_to_field) Add prototype.
749 (encode_operand, parse, convert_iif, md_fix_pcrel_adjust): Add
750 prototypes and make static.
751 (struct addr_mode): Make mode and scaled_mode explicitly signed.
752 (evaluate_expr): Use resultP argument instead of exprP global.
753 (get_addr_mode): Quiten compiler warnings.
754 (encode_operand): eliminate unused variables. Quiten compiler
755 warnings. Eliminate nul character in format strings.
756 (parse): argc is unsigned.
757 (reloc): Type cast index to quieten compiler.
758 (md_pcrel_adjust, md_apply_fix3): Remove unused variable.
759 (md_convert_frag): Note unused parameters. Remove unused
761 (md_create_long_jump, md_create_short_jump,
762 md_undefined_symbol_name, md_section_align, tc_gen_reloc): Note
765 2002-07-31 Nick Clifton <nickc@redhat.com>
767 * NEWS: Retroactively add entry for Lars Brinkhoff's contribution
768 of the PDP-11 and 2.11BSD a.out support.
770 2002-07-31 Momchil Velikov <velco@fadata.bg>
772 * config/tc-v850.c (md_assemble): Fix range check for immediate
775 2002-07-30 Chris Demetriou <cgd@broadcom.com>
777 * config/tc-mips.c (mips_cpu_info_table): Clean up entries related
778 to Broadcom SB-1 core support.
780 2002-07-31 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
782 * config/tc-mips.c (mips_target_format): Fix formatting.
783 Add recognition of n32 ABI formats.
785 2002-07-30 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
787 * tc-mips.c (load_address): Don't clobber $at when loading a
788 64-bit address in non-PIC code if noat is in effect.
791 2002-07-30 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
793 * config/tc-mips.c (macro): Use codes 6 and 7 in trap instructions
794 used in division/multiply macro expansions similarly to how they
795 are used in the variants with break instructions.
798 2002-07-30 Graeme Peterson <gp@qnx.com>
800 * configure.in: Add support for arm-*-nto target.
801 * configure: Regenerate.
803 2002-07-30 Nick Clifton <nickc@redhat.com>
805 * config/tc-arm.c (struct literal_pool): Add fields to allow
806 multiple literal pools to be maintained.
807 (find_literal_pool): New function.
808 (find_or_make_literal_pool): New function.
809 (add_to_literal_pool): Use find_or_make_literal_pool.
810 (arm_s_text, arm_s_data, arm_s_section): Remove - no longer
812 (s_ltorg): Use find_literal_pool.
813 (arm_cleanup): Dump all literal pools.
814 * doc/c-arm.texi: Document new behaviour of only dumping literal
817 2002-07-26 Alan Modra <amodra@bigpond.net.au>
819 * config/tc-ppc.c (ppc_set_cpu): Use PPC_OPCODE_64 as the default
820 rather than PPC_OPCODE_32 for powerpc64*.
822 2002-07-25 Nick Clifton <nickc@redhat.com>
824 * po/es.po: Updated Spanish translation.
825 * po/fr.po: Updated French translation.
827 2002-07-25 Richard Sandiford <rsandifo@redhat.com>
829 * doc/c-mips.texi: Remove -mcpu. Document -mabi.
830 * configure.in (MIPS_CPU_STRING_DEFAULT): New configuration macro.
831 (USE_E_MIPS_ABI_O32, MIPS_DEFAULT_64BIT): New configuration macros.
832 * configure, config.in: Regenerate.
833 * config/tc-mips.c (file_mips_abi): Rename to mips_abi.
834 (mips_set_options): Remove "abi" field.
835 (mips_opts): Update accordingly. Replace all uses of mips_opts.abi
838 (mips_arch_string, mips_arch_info): New vars.
839 (mips_tune_string, mips_tune_info): New vars.
840 (ABI_NEEDS_32BIT_REGS, ABI_NEEDS_64BIT_REGS): New macros.
841 (HAVE_32BIT_GPRS, HAVE_32BIT_FPRS): Don't check the ABI.
842 (mips_isa_to_str, mips_cpu_to_str): Remove.
843 (mips_ip): If the selected architecture is a generic ISA rather
844 than a processor, only mention the ISA level in error messages.
845 (OPTION_MCPU): Remove.
847 (md_longopts): Add -mfp64, remove -mcpu.
848 (mips_set_option_string): New fn.
849 (md_parse_option): Make -mipsN update file_mips_isa rather than
850 mips_opts.isa. Use mips_set_option_string to set -march or -mtune.
851 Don't let -mgp32 and -mfp32 change the ABI.
852 (show): Move to end of file. Constify string argument.
853 (md_show_usage): Move to the end of the file. Read available
854 architectures from mips_cpu_info_table.
855 (mips_set_architecture): New fn.
856 (mips_after_parse_args): Rework. Remove -mcpu handling. -mipsN
857 is an alias for -march=mipsN. Don't change the ABI based on other
858 flags. Infer the register size from the ABI as well as the
859 architecture. Complain about more conflicting arguments.
860 Unify logic with GCC.
861 (s_mipsset): Don't change the ABI.
862 (mips_elf_final_processing): Check USE_E_MIPS_ABI_O32.
863 (mips_cpu_info_table): Remove Generic-MIPS* entries, keeping just
864 "mipsN"-type entries. Remove entries that vary only in the
865 manufacturer's prefix, or that have "000" replaced by "k".
866 Remove TARGET_CPU entries. Make r2000 entry use CPU_R3000.
867 (mips_strict_matching_cpu_name_p, mips_matching_cpu_name_p): New fns.
868 (mips_parse_cpu): New fn.
869 (mips_cpu_info_from_name, mips_cpu_info_from_cpu): Remove.
870 (mips_cpu_info_from_isa): Minor formatting tweak.
872 2002-07-24 Nick Clifton <nickc@redhat.com>
874 * po/sv.po: Updated Swedish translation.
875 * po/es.po: Updated Spanish translation.
877 2002-07-23 Alan Modra <amodra@bigpond.net.au>
879 * Makefile.am: Run "make dep-am".
880 * Makefile.in: Regenerate.
881 * doc/Makefile.in: Regenerate.
882 * config.in: Regenerate.
883 * po/POTFILES.in: Regenerate.
885 2002-07-23 Nick Clifton <nickc@redhat.com>
887 * po/fr.po: Updated French translation.
889 2002-07-23 Nick Clifton <nickc@redhat.com>
891 * CONTRIBUTORS: Ken Raeburn is no longer the maintainer for
894 2002-07-19 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
896 * config/tc-mips.c (mips_need_elf_addend_fixup): Use S_IS_EXTERNAL
897 instead of S_IS_EXTERN.
898 (md_estimate_size_before_relax): Likewise.
899 (mips_fix_adjustable): Likewise.
901 2002-07-19 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
903 * config/tc-mips.c (mips_pic_level): Remove IRIX4_PIC.
905 2002-07-19 Miroslav Tichy <tichm9am@ss1000.ms.mff.cuni.cz>
906 Nick Clifton <nickc@redhat.com>
908 * cond.c (s_ifdef): Treat a referenced but not yet defined
909 symbol as if it were undefined, in exactly the same way as
912 * doc/as.texinfo: Document that .ifdef, .ifndef and .equiv
913 consider referenced bug not yet defined symbols to be
916 2002-07-18 Denis Chertykov <denisc@overta.ru>
917 Frank Ch. Eigler <fche@redhat.com>
918 Alan Lehotsky <alehotsky@cygnus.com>
919 John Healy <jhealy@redhat.com>
920 Jeff Johnston <jjohnstn@redhat.com>
922 * configure.in: Add ip2k configuraton.
923 * configure: Regenerate.
924 * Makefile.am: Add ip2k configuraton.
925 * Makefile.in: Regenerate.
926 * configure: Regenerate.
927 * Makefile.in: Regenerate.
928 * config/tc-ip2k.c: New file.
929 * config/tc-ip2k.h: New files.
930 * NEWS: Mention new support.
931 * doc/Makefile.am (CPU_DOCS): Add c-ip2k.texi.
932 * doc/Makefile.in: Regenerate.
933 * doc/all.texi: Set IP2K
934 * doc/as.texinfo: Add IP2K description.
935 * doc/c-ip2k.texi: New file.
937 2002-07-19 Nick Clifton <nickc@cambridge.redhat.com>
939 * NEWS: Reformat to match style of other NEWS files.
941 2002-07-17 Jan Hubicka <jh@suse.cz>
943 * tc-i386.c (i386_align_code): Implement x86_64 neutral code fillers.
945 2002-07-16 Moritz Jodeit <moritz@jodeit.org>
947 * config/tc-z8k.c (build_bytes): Correct order of memset args.
949 2002-07-16 Nick Clifton <nickc@cambridge.redhat.com>
951 * NEWS: Add 'Changes in 2.13'.
953 2002-07-15 Matt Fredette <fredette@netbsd.org>
955 * config/tc-hppa.h (LABELS_WITHOUT_COLONS): Define if TE_NetBSD.
957 2002-07-12 Alan Modra <amodra@bigpond.net.au>
959 * config/tc-i386.c (process_suffix): Merge CODE_64BIT JumpByte
960 case with non CODE_64BIT case. Don't warn on "qword ptr" if
963 2002-07-11 Alan Modra <amodra@bigpond.net.au>
965 * config/tc-ppc.c (ppc_elf_frob_symbol): Delete.
966 (ppc_frob_file_before_adjust): New function.
967 * config/tc-ppc.h (tc_frob_symbol): Don't define.
968 (ppc_elf_frob_symbol): Don't declare.
969 (tc_frob_file_before_adjust): Define.
970 (ppc_frob_file_before_adjust): Declare.
972 * config/tc-ppc.c (md_pseudo_table): Warning fix.
973 (ppc_cpu): Make it unsigned long to agree with struct powerpc_opcode
976 (ppc_xcoff64): Rename to ppc_obj64.
977 (md_parse_option <m>): Encode old ppc_size value in ppc_cpu.
978 (ppc_set_cpu): Set PPC_OPCODE_32 too.
979 (ppc_arch): Use ppc_obj64 instead of ppc_size to select bfd_mach_ppc64
981 (ppc_target_format): Use ppc_obj64 to select format.
982 (md_begin): Adjust for PPC_OPCODE_32/64 in ppc_cpu.
983 (ppc_insert_operand): Use ppc_obj64 instead of ppc_size.
984 (ppc_elf_suffix): Likewise. Don't depend on BFD_DEFAULT_TARGET_SIZE.
985 (tc_frob_symbol): Likewise.
986 (md_assemble): Use ppc_obj64 instead of ppc_size. Don't depend on
987 BFD_DEFAULT_TARGET_SIZE.
989 (ppc_is_toc_sym): Likewise.
990 (md_apply_fix3): Likewise.
991 * config/tc-ppc.h (TC_FORCE_RELOCATION): Don't depend on
992 BFD_DEFAULT_TARGET_SIZE.
993 (ELF_TC_SPECIAL_SECTIONS): Likewise.
994 (tc_frob_symbol): Likewise.
996 2002-07-09 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
998 * config/tc-mips.c (macro_build): Handle MIPS16 insns.
1001 2002-07-09 Alan Modra <amodra@bigpond.net.au>
1003 * config/tc-i386.c (md_pseudo_table <file>): Warning fix.
1004 (BFD_RELOC_8, BFD_RELOC_8_PCREL): Define for non-BFD.
1005 (md_apply_fix3): Formatting. Remove redundant test.
1006 (tc_gen_reloc): Remove redundant code.
1007 (tc_i386_force_relocation): Delete. Movy body of function to..
1008 * config/tc-i386.h (TC_FORCE_RELOCATION): .. here.
1010 2002-07-09 Federico G. Schwindt <fgsch@olimpo.com.br>
1012 * configure.in: Add hppa-*-openbsd* target, change
1013 alpha*-*-openbsd* format to elf, and use elf for sparc-*-openbsd*
1015 * configure: Regenerate.
1017 2002-07-08 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
1019 * config/tc-mips.c (macro): Shift the 32-bit address range
1020 accessible with a lone "lui" down by 32768.
1022 2002-07-08 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
1024 * config/tc-mips.c (load_address): Use non-trapping "daddu"
1025 instead of "dadd" in address calculations.
1028 2002-07-08 Alan Modra <amodra@bigpond.net.au>
1030 * config/tc-i386.c (process_suffix): Remove intel mode movsx and
1032 (md_assemble): Instead, zap the suffix here.
1034 2002-07-03 Nick Clifton <nickc@cambridge.redhat.com>
1036 * NEWS: Remove next release number until the release is actually
1039 2002-07-03 Alan Modra <amodra@bigpond.net.au>
1041 * Makefile.am (check-DEJAGNU): Revert 2002-06-25 change.
1043 * Makefile.in: Regenerate.
1045 2002-07-02 Martin Schwidefsky <schwidefsky@de.ibm.com>
1047 * config/tc-s390.c (tc_s390_fix_adjustable): Prevent any adjustment
1048 to symbols in merge sections, even non pc-relative ones.
1050 2002-06-29 Stephane Carrez <stcarrez@nerim.fr>
1052 * config/tc-m68hc11.h (m68hc11_listing_header): Fix warning.
1054 2002-06-29 Stephane Carrez <stcarrez@nerim.fr>
1056 * config/tc-m68hc11.h (ELF_TC_SPECIAL_SECTIONS): New sections
1057 .softregs and .eeprom.
1059 2002-06-28 David O'Brien <obrien@FreeBSD.org>
1061 * NEWS: Note the next release is 2.13.
1063 2002-06-26 Nick Clifton <nickc@cambridge.redhat.com>
1065 * po/tr.po: New translation imported.
1067 2002-06-26 Elias Athanasopoulos <eathan@otenet.gr>
1069 * ecoff.c: (get_tag): Replace strcpy with xstrdup.
1070 (ecoff_directive_def): Likewise.
1071 (ecoff_directive_tag): Likewise.
1072 * listing.c (file_info): Likewise.
1073 * hash.c (what): Likewise.
1075 2002-06-25 H.J. Lu <hjl@gnu.org>
1077 * Makefile.am (check-DEJAGNU): Set LC_ALL=C and export it.
1078 * Makefile.in: Regenerated.
1080 2002-06-19 Dhananjay R. Deshpande <dhananjayd@kpit.com>
1082 * config/tc-sh.c (get_specific): Revert 2002-05-01 change.
1083 (assemble_ppi): Generate warning if the same register is used
1084 twice as destination in the same padd / pmuls instruction.
1086 2002-06-18 Dave Brolley <brolley@redhat.com>
1088 From Catherine Moore, Michael Meissner, Richard Sandiford, Dave Brolley
1089 * po/POTFILES.in: Add tc-frv.c, tc-frv.h.
1090 * configure.in: Support frv-*-*.
1091 * Makefile.am (CPU_TYPES): Add frv.
1092 (TARGET_CPU_CFILES): Add tc-frv.c.
1093 (TARGET_CPU_HFILES): Add tc-frv.h.
1094 (DEPTC_frv_coff): New variable.
1095 (DEPTC_frv_elf): New variable.
1096 (DEPOBJ_frv_coff): New variable.
1097 (DEPOBJ_frv_elf): New variable.
1098 (DEP_frv_coff): New variable.
1099 (DEP_frv_elf): New variable.
1100 * tc-frv.c: New file.
1101 * tc-frv.h: New file.
1103 2002-06-17 Catherine Moore <clm@redhat.com>
1105 * config/obj-elf.h (TARGET_SYMBOL_FIELDS): Conditionally define.
1107 2002-06-17 J"orn Rennecke <joern.rennecke@superh.com>
1109 * config/tc-sh.c (assemble_ppi): Initialize reg_n.
1111 2002-06-17 Tom Rix <trix@redhat.com>
1113 * config/tc-i370.h (tc_comment_chars): Define for i370-elf.
1115 2002-06-14 H.J. Lu <hjl@gnu.org>
1116 Daniel Jacobowitz <drow@mvista.com>
1118 * dwarf2dbg.h (dwarf2_directive_file): Return char *.
1119 * dwarf2dbg.c (dwarf2_directive_file): Return filename.
1120 * config/tc-mips.c (s_mips_file): Call s_app_file_string
1121 and new_logical_line for the first .file directive.
1122 * read.c (s_app_file_string): New function.
1123 (s_app_file): Call it.
1124 * read.h (s_app_file_string): Add declaration.
1126 2002-06-14 Daniel Jacobowitz <drow@mvista.com>
1128 * configure.in: Remove MIPS_STABS_ELF.
1129 * configure: Regenerated.
1130 * config.in: Regenerated.
1131 * config/obj-elf.h (ECOFF_DEBUGGING): Define to mips_flag_mdebug
1133 * config/tc-mips.c (mips_pseudo_table): Remove #ifdef around
1135 (pdr_seg): Declare unconditionally.
1136 (md_begin): Always generate .pdr unless ECOFF_DEBUGGING or not ELF.
1137 (s_mips_end): Likewise. Generate stabs function markers.
1138 (s_mips_ent): Generate stabs function markers.
1139 (s_mips_frame): Always generate .pdr unless ECOFF_DEBUGGING or not
1141 (s_mips_mask): Likewise.
1142 (mips_flag_mdebug): New.
1143 (md_longopts): Add "mdebug" and "no-mdebug".
1144 (md_parse_options): Add OPTION_MDEBUG and OPTION_NO_MDEBUG.
1145 (mips_after_parse_args): Set mips_flag_mdebug.
1146 * doc/as.texinfo: Add "-mdebug" and "-no-mdebug" for MIPS.
1148 2002-06-13 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
1150 * config/tc-mips.c (md_apply_fix3): Don't subtract the symbol's
1151 value twice for RELA relocations.
1153 2002-06-12 Ben Elliston <bje@redhat.com>
1155 * symbols.c (resolve_symbol_value): Initialise final_val.
1157 * subsegs.c (subsegs_print_statistics): Cast frchp to void *.
1159 2002-06-11 Tom Rix <trix@redhat.com>
1161 * config/tc-ppc.c (ppc_subseg_align): Delete.
1162 (ppc_change_csect): Default csect align is 2.
1163 * config/tc-ppc.h (SUB_SEGMENT_ALIGN): Delete
1165 2002-06-09 Marek Michalkiewicz <marekm@amelek.gda.pl>
1167 * config/tc-avr.c (mcu_types): Update.
1169 2002-06-08 Matt Thomas <matt@3am-software.com>
1171 * configure.in (vax-*-netbsdelf*, vax-*-netbsdaout*)
1172 (vax-*-netbsd*): New targets.
1173 * configure: Regenerate.
1174 * config/aout_gnu.h (enum machine_type): Add M_VAX4K_NETBSD.
1175 * config/tc-vax.c: Add support for ELF and PIC.
1176 (flag_want_pic): New flag.
1177 (float_cons): Fix prototype.
1178 (md_apply_fix3): Adjust for BFD_ASSEMBLER.
1179 (md_assemble): Introduce a new is_absolute local, and use it
1180 rather than repeating the test. Make fatal errors actually
1181 fatal by using as_fatal as appropriate. Adjust for BFD_ASSEMBLER.
1182 Add support for ELF. Add support for PIC.
1183 (md_convert_frag): Adjust for BFD_ASSEMBLER.
1184 (tc_aout_fix_to_chars): Only include if OBJ_AOUT and not
1186 (vax_reg_parse): Make the % register prefix mandatory for ELF,
1187 optional for a.out, and not allowed for VMS. Adjust all callers.
1188 (md_create_short_jump): Add ATTRIBUTE_UNUSED to unused arguments.
1189 (md_create_long_jump): Likewise.
1190 (md_undefined_symbol): Likewise.
1191 (md_section_align): Likewise.
1192 (md_shortopts): Allow -k and -K for ELF.
1193 (md_parse_option): Set flag_want_pic if -k or -K.
1194 (tc_headers_hook): New function if OBJ_AOUT and not BFD_ASSEMBLER.
1195 (tc_gen_reloc): New function if BFD_ASSEMBLER.
1196 * config/tc-vax.h (tc_headers_hook): Remove.
1197 (TARGET_FORMAT): Set according to object format and target
1199 (BFD_ARCH, TARGET_ARCH): Define.
1200 (NO_RELOC): Adjust for BFD_ASSEMBLER.
1201 (TC_RELOC_RTSYM_LOC_FIXUP, TC_FIX_ADJUSTABLE)
1202 (tc_fix_adjustable): Define if BFD_ASSEMBLER.
1203 * config/vax-inst.h (VAX_JSB, VAX_CALLS, VAX_CALLG): Define.
1205 2002-06-08 Alan Modra <amodra@bigpond.net.au>
1207 * Makefile.am: Run "make dep-am".
1208 * Makefile.in: Regenerate.
1210 * as.c: Replace CONST with const.
1211 * write.c: Likewise.
1212 * config/obj-coff.c: Likewise.
1213 * config/tc-a29k.c: Likewise.
1214 * config/tc-arm.c: Likewise.
1215 * config/tc-dlx.c: Likewise.
1216 * config/tc-h8300.c: Likewise.
1217 * config/tc-h8500.c: Likewise.
1218 * config/tc-i370.c: Likewise.
1219 * config/tc-i860.c: Likewise.
1220 * config/tc-i960.c: Likewise.
1221 * config/tc-m68hc11.c: Likewise.
1222 * config/tc-m68k.c: Likewise.
1223 * config/tc-m88k.c: Likewise.
1224 * config/tc-mcore.c: Likewise.
1225 * config/tc-mips.c: Likewise.
1226 * config/tc-ns32k.c: Likewise.
1227 * config/tc-pdp11.c: Likewise.
1228 * config/tc-pj.c: Likewise.
1229 * config/tc-s390.c: Likewise.
1230 * config/tc-sh.c: Likewise.
1231 * config/tc-sparc.c: Likewise.
1232 * config/tc-tahoe.c: Likewise.
1233 * config/tc-tic80.c: Likewise.
1234 * config/tc-v850.c: Likewise.
1235 * config/tc-vax.c: Likewise.
1236 * config/tc-w65.c: Likewise.
1237 * config/tc-z8k.c: Likewise.
1239 2002-06-08 Daniel Jacobowitz <drow@mvista.com>
1241 Based on patch from Matt Green:
1242 * config/obj-elf.h (ECOFF_DEBUGGING): Make configurable.
1243 * config/tc-mips.c (s_mips_file): Renamed from s_file.
1244 (s_mips_loc): New function.
1245 (mips_nonecoff_pseudo_table): Call them.
1246 (append_insn): Call dwarf2_emit_insn.
1248 2002-06-08 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1250 * config/tc-mips.c (mips_opts): Fix comment, all ASE fields are set
1252 (file_ase_mips16): New veriable.
1253 (mips_eabi64): Remove.
1254 (CPU_HAS_MIPS16): New define.
1255 (CPU_HAS_MDMX): Fix data type.
1256 (md_begin): Code cleanup. Use file_ase_mips16.
1257 (mips_elf_final_processing): Handle mips16 header flag.
1258 Handle EABI flag without intermediate variable.
1260 2002-06-08 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1262 * doc/as.texinfo: Update MIPS documentation.
1264 2002-06-08 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1266 * config/tc-mips.c: Add define for $zero register.
1267 (md_begin): Add $zero as alias name for $0.
1268 (insn_uses_reg): Use ZERO define.
1269 (mips_ip): Add $zero as alias name for $0.
1270 (mips16_ip): Likewise.
1271 (s_cplocal): Demand empty rest of input line.
1272 (tc_get_register): Likewise. Add support for $kt0, kt1 register
1273 names. Use ZERO define. Fix input_line_pointer progress.
1275 2002-06-07 Alan Modra <amodra@bigpond.net.au>
1277 * symbols.c: Replace CONST by const throughout.
1278 (symbol_find_exact): Split out from..
1279 (symbol_find_base): ..here.
1280 * symbols.h: Replace CONST by const throughout.
1281 (symbol_find_exact): Declare.
1282 * config/obj-elf.c: #include "struc-symbol.h".
1283 (elf_frob_file): If group name matches an exported symbol, use that
1284 symbol for the signature and ".group" as the section name.
1286 2002-06-06 J"orn Rennecke <joern.rennecke@superh.com>
1288 * config/tc-sh.c (parse_at): @(symbol,pc) is A_DISP_PC again,
1290 * testsuite/gas/sh/pcrel.s: Also test @(symbol,pc).
1291 * testsuite/gas/sh/pcrel.d: Update.
1292 * testsuite/gas/sh/pcrel.l: New file.
1294 2002-06-06 Daniel Jacobowitz <drow@mvista.com>
1296 * tc-mips.c (mips_after_parse_args): Always set mips_opts.ase_mips3d
1297 and mips_opts.ase_mdmx if they are uninitialized.
1299 2002-06-06 John David Anglin <dave@hiauly1.hia.nrc.ca>
1301 * gas/config/tc-hppa.c (pa_ip): Replace "L$0\001" with FAKE_LABEL_NAME.
1302 (hppa_force_relocation): Check if a stub just before the start symbol
1303 of the last call_info is reachable before forcing relocation. Fix
1306 2002-06-04 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
1308 * config/tc-mips.c (mips_after_parse_args): New function.
1309 (md_begin): Move processing of defaults to mips_after_parse_args.
1310 config/tc-mips.h (md_after_parse_args): Define.
1312 2002-06-04 Jason Thorpe <thorpej@wasabisystems.com>
1314 * configure.in (sh5*): Set cpu_type to sh64 and endian to big.
1315 (sh5le*, sh64le*): Set cpu_type to sh64 and endian to little.
1316 (sh5*-*-netbsd*, sh64*-*-netbsd*): New targets.
1317 * configure: Regenerate.
1318 * config/tc-sh64.c (sh64_target_format): Add support for NetBSD
1321 2002-06-04 Jason Thorpe <thorpej@wasabisystems.com>
1323 * config/tc-sh64.h (MD_PCREL_FROM_SECTION): Undef before redefining.
1325 2002-06-04 Alan Modra <amodra@bigpond.net.au>
1327 * config/obj-elf.c (obj_elf_change_section): Set and check elf
1328 linkonce flag. Print all warnings.
1329 (obj_elf_section): Parse ",comdat" for groups.
1330 (elf_frob_file): Set SEC_LINK_ONCE on COMDAT groups. Check
1331 consistency of comdat flag.
1333 2002-06-02 Richard Henderson <rth@redhat.com>
1335 * config/tc-alpha.c (alpha_adjust_symtab_relocs): Fix thinko
1336 with LITERALs without sequence numbers.
1338 2002-06-01 Richard Henderson <rth@redhat.com>
1340 * config/tc-alpha.c: Move LITUSE constants to "elf/alpha.h".
1341 Rename them LITUSE_ALPHA_*.
1343 2002-05-31 Shrinivas Atre <ShrinivasA@kpit.com>
1345 * config/tc-h8300.c (get_operand): Allow stm.l and ldm.l insns to
1346 accept parentheses enclosed register lists.
1348 2002-05-31 Alan Modra <amodra@bigpond.net.au>
1350 * Makefile.am: Run "make dep-am".
1351 * Makefile.in: Regenerate.
1352 * po/POTFILES.in: Regenerate.
1354 2002-05-31 Graeme Peterson <gp@qnx.com>
1356 * configure.in: Add i386-*-nto-qnx*.
1357 * configure: Regenerate.
1359 2002-05-31 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1361 * config/tc-mips.c (mips_ip): Use unsigned long values for
1364 2002-05-31 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1366 * config/tc-mips.c (s_cpsetup): Fix initialization of
1367 mips_cpreturn_register and mips_cpreturn_offset.
1369 2002-05-31 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1371 * config/tc-mips.c (s_cpsetup): Fix comment.
1373 2002-05-31 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1375 * config/tc-mips.c (md_begin): Add $ra as alias name for $31.
1376 (mips_ip): Likewise.
1377 (mips16_ip): Likewise.
1378 (tc_get_register): Likewise.
1380 2002-05-30 Chris G. Demetriou <cgd@broadcom.com>
1381 Ed Satterthwaite <ehs@broadcom.com>
1383 * config/tc-mips.c (mips_set_options): New "ase_mdmx" member.
1384 (mips_opts): Initialize "ase_mdmx" member.
1385 (file_ase_mdmx): New variable.
1386 (CPU_HAS_MDMX): New macro.
1387 (md_begin): Initialize mips_opts.ase_mdmx and file_ase_mdmx
1388 based on command line options and configuration defaults.
1389 (macro_build): Note in comment that use of MDMX in macros is
1390 not currently allowed.
1391 (validate_mips_insn): Add support for the "O", "Q", "X", "Y", and
1392 "Z" MDMX operand types.
1393 (mips_ip): Accept MDMX instructions if mips_opts.ase_mdmx is set,
1394 and add support for the "O", "Q", "X", "Y", and "Z" MDMX operand
1396 (OPTION_MDMX, OPTION_NO_MDMX, md_longopts, md_parse_option):
1397 Add support for "-mdmx" and "-no-mdmx" options.
1398 (OPTION_ELF_BASE): Move to accomodate new options.
1399 (s_mipsset): Support ".set mdmx" and ".set nomdmx".
1400 (mips_elf_final_processing): Set MDMX ASE ELF header flag if
1401 file_ase_mdmx was set.
1402 * doc/as.texinfo: Document -mdmx and -no-mdmx options.
1403 * doc/c-mips.texi: Likewise, and document ".set mdmx" and ".set
1406 2002-05-31 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1408 * config/tc-mips.c (OPTION_NO_M7000_HILO_FIX): Rename to
1409 OPTION_MNO_7000_HILO_FIX. Add alternate "mno-fix7000"
1410 command line switch conforming to gcc conventions.
1411 * doc/c-mips.texi: Document -mno-fix7000 instead of no-mfix-7000.
1413 2002-05-31 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1415 * config/tc-mips.c (macro_build_jalr): New Function.
1416 (md_begin): NewABI uses big GOTs.
1417 (macro_build): Recognize BFD_RELOC_MIPS_GOT_DISP,
1418 BFD_RELOC_MIPS_GOT_PAGE, BFD_RELOC_MIPS_GOT_OFST as valid.
1419 (load_address): Add some NewABI PIC support.
1421 (md_apply_fix): Special handling for BFD_RELOC_MIPS_JALR.
1422 (tc_gen_reloc): Don't encode NewABI vtables in REL relocations.
1424 2002-05-31 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1426 * config/tc-mips.c (load_address): Use mips_gp_register instead
1427 of hardcoded value. Remove dbl parameter, use HAVE_32BIT_ADDRESSES
1429 (macro): Use mips_gp_register instead of hardcoded value.
1431 2002-05-30 Richard Henderson <rth@redhat.com>
1433 * expr.h (operatorT): Add O_md17..O_md32.
1434 * config/tc-alpha.c (O_lituse_tlsgd, O_lituse_tlsldm, O_tlsgd,
1435 O_tlsldm, O_gotdtprel, O_dtprelhi, O_dtprello, O_dtprel, O_gottprel,
1436 O_tprelhi, O_tprello, O_tprel): New.
1437 (USER_RELOC_P, alpha_reloc_op_tag, debug_exp): Include them.
1438 (DUMMY_RELOC_LITUSE_TLSGD, DUMMY_RELOC_LITUSE_TLSLDM): New.
1439 (LITUSE_TLSGD, LITUSE_TLSLDM): New.
1440 (struct alpha_reloc_tag): Add master, saw_tlsgd, saw_tlsld,
1441 saw_lu_tlsgd, saw_lu_tlsldm. Make multi_section_p a bit field.
1442 (md_apply_fix3): Handle TLS relocations.
1443 (alpha_force_relocation, alpha_fix_adjustable): Likewise.
1444 (alpha_adjust_symtab_relocs): Sort LITERAL relocs after the
1445 associated TLS reloc. Check lituse_tls relocs match up.
1446 (emit_insn): Handle TLS relocations.
1449 * doc/c-alpha.texi: Add docs for tls relocations.
1451 2002-05-30 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1453 * config/tc-mips.c (mips_gprel_offset): New variable.
1454 (s_gpvalue): Use it.
1456 2002-05-30 Diego Novillo <dnovillo@redhat.com>
1458 * gas/config/tc-d10v.c (check_resource_conflict): Only check
1459 write-write conflicts.
1460 (md_assemble): Reformat introductory comment.
1461 (parallel_ok): Prevent packing only if the first
1462 instruction cannot be packed.
1464 2002-05-30 Jason Eckhardt <jle@redhat.com>
1465 Tom Rix <trix@redhat.com>
1467 * config/tc-d10v.c (build_insn): Check for immediates.
1469 2002-05-28 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1471 * config/tc-mips.c: Replace GP in comments by $gp.
1472 (mips_big_got): Initialize.
1473 (mips_trap): Initialize.
1474 (load_address): Use mips_gp_register instead of hardcoded value.
1475 Remove dbl parameter, use HAVE_32BIT_ADDRESSES instead.
1476 (macro): Use mips_gp_register instead of hardcoded value.
1477 (macro2): Change load_address calls.
1478 (md_pcrel_from): Comment formatting.
1479 (s_cpload): Use mips_gp_register instead of hardcoded value.
1480 (s_cprestore): Likewise. Comment formatting.
1481 (s_gpword): Fix data type.
1482 (s_cpadd): Use mips_gp_register instead of hardcoded value.
1483 (nopic_need_relax): Replace GP in comments by $gp.
1484 (mips_elf_final_processing): Better comment.
1486 2002-05-28 Kuang Hwa Lin <kuang@sbcglobal.net>
1488 * configure.in: Add DLX configuraton
1489 * Makefile.am: Add DLX configuraton
1490 * configure: Regenerate.
1491 * Makefile.in: Regenerate.
1492 * config/tc-dlx.c: New file.
1493 * config/tc-dlx.h: New files.
1494 * NEWS: Mention new support.
1496 2002-05-27 Nick Clifton <nickc@cambridge.redhat.com>
1498 * config/obj-coff.c (write_object_file): Add missing semicolon.
1500 2002-05-26 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1502 * config/tc-mips.c (mips_emit_delays): Replace magic constant for RA
1503 by the define. Remove superfluous check of mips_opts.mips16.
1504 (append_insn): Likewise. Canonicalize variable increments.
1505 (macro_build): Canonicalize variable increments.
1506 (macro_build_lui): Likewise.
1507 (load_register): Likewise.
1508 (load_address): Move pointer initialization.
1509 (macro): Move pointer to a more local scope. Canonicalize variable
1510 increments. Better comments. Replace magic constant for RA by the
1512 (macro2): Replace magic constant for RA by the define. Canonicalize
1513 variable increments.
1514 (mips_ip): Canonicalize variable increments.
1515 (mips16_ip): Replace magic constant for RA by the define.
1516 (my_getSmallParser): Canonicalize variable increments/decrements.
1517 (my_getPercentOp): Likewise.
1518 (my_getSmallExpression): Likewise.
1519 (s_align): Likewise.
1520 (s_mipsset): Likewise.
1521 (s_cpsetup): Likewise.
1522 (s_insn): Remove superfluous check of mips_opts.mips16.
1523 (s_mips_stab): Likewise.
1524 (mips_handle_align): Canonicalize variable increments.
1525 (s_mips_ent): Likewise.
1526 (s_mips_end): Add comment.
1528 2002-05-26 Jason Thorpe <thorpej@wasabisystems.com>
1530 * configure.in: Use ns32k-*-netbsd* instead of ns32k-pc532-netbsd*.
1531 * configure: Regenerate.
1533 2002-05-25 Alan Modra <amodra@bigpond.net.au>
1535 * Makefile.am (OBJS): Depend on ansidecl.h and fopen-same.h.
1536 * Makefile.in: Regenerate.
1537 * dep-in.sed: Reorder to match OBJS in Makefile.am.
1538 * configure.in (ALL_OBJ_DEPS): Add symcat.h when need_bfd.
1539 * configure: Regenerate.
1540 * as.h: Use #include "" instead of <> for local header files.
1541 * flonum-konst.c: Likewise.
1542 * flonum-mult.c: Likewise.
1544 * listing.c: Likewise.
1545 * config/tc-ia64.h: Likewise.
1546 * config/tc-v850.h: Likewise.
1548 2002-05-24 TAMURA Kent <kent@netbsd.org>
1550 * configure.in: Add a target for i386-netbsdpe.
1551 * configure: Regenerate.
1553 2002-05-23 Jakub Jelinek <jakub@redhat.com>
1555 * config/obj-elf.c (elf_common): Renamed from obj_elf_common.
1556 (obj_elf_common): Call elf_common.
1557 (obj_elf_tls_common): New function.
1558 (elf_pseudo_tab): Support .tls_common.
1559 (special_sections): Add .tdata and .tbss.
1560 (obj_elf_change_section): Set SEC_THREAD_LOCAL for SHF_TLS
1562 (obj_elf_parse_section_letters): Support T in section flags (SHF_TLS).
1563 (obj_elf_parse_section_letters): Include T in error message.
1564 * config/tc-ppc.c (ppc_section_letter): Likewise.
1565 * config/tc-alpha.c (alpha_elf_section_letter): Likewise.
1566 (tc_gen_reloc): Handle SEC_THREAD_LOCAL the same way as
1568 * config/tc-sparc.c (md_apply_fix3): Likewise.
1569 * config/tc-i386.c (tc_i386_fix_adjustable): Add TLS relocs.
1570 Define them if not BFD_ASSEMBLER.
1571 (lex_got): Support @TLSGD, @TLSLDM, @GOTTPOFF, @TPOFF, @DTPOFF
1573 (md_apply_fix3): Add TLS relocs.
1574 * config/tc-ia64.c (enum reloc_func): Add FUNC_DTP_MODULE,
1575 FUNC_DTP_RELATIVE, FUNC_TP_RELATIVE, FUNC_LT_DTP_MODULE,
1576 FUNC_LT_DTP_RELATIVE, FUNC_LT_TP_RELATIVE.
1577 (pseudo_func): Support @dtpmod(), @dtprel() and @tprel().
1578 (ia64_elf_section_letter): Include T in error message.
1579 (md_begin): Support TLS operators.
1580 (md_operand): Likewise.
1581 (ia64_gen_real_reloc_type): Support TLS relocs.
1582 * testsuite/gas/i386/tlspic.s: New file.
1583 * testsuite/gas/i386/tlsd.s: New file.
1584 * testsuite/gas/i386/tlsnopic.s: New file.
1585 * testsuite/gas/i386/tlsd.d: New file.
1586 * testsuite/gas/i386/tlsnopic.d: New file.
1587 * testsuite/gas/i386/tlspic.d: New file.
1588 * testsuite/gas/i386/i386.exp: Add tlsd, tlsnopic and tlspic tests.
1589 * testsuite/gas/ia64/tls.s: New file.
1590 * testsuite/gas/ia64/tls.d: New file.
1591 * testsuite/gas/ia64/ia64.exp: Add tls test.
1592 * write.c (adjust_reloc_syms): Don't change symbols in
1593 SEC_THREAD_LOCAL sections to STT_SECTION + addend.
1595 2002-05-23 Nick Clifton <nickc@cambridge.redhat.com>
1597 * config/tc-arm.c (md_apply_fix3): For the Thumb BLX reloc
1598 round the relocation up rather than down.
1600 2002-05-23 Nick Clifton <nickc@cambridge.redhat.com>
1602 * config/obj-coff.c (obj_coff_section): Silently ignore an 'a'
1604 * doc/as.texinfo: Document that the COFF version of .section
1605 ignores the 'a' flag.
1607 2002-05-23 Alan Modra <amodra@bigpond.net.au>
1609 * config/tc-alpha.c (assemble_tokens): Protect use of
1610 ALPHA_RELOC_TABLE with #ifdef RELOC_OP_P.
1612 * write.c (size_seg): Check adjustment to last frag.
1613 (SUB_SEGMENT_ALIGN): If HANDLE_ALIGN defined, pad out last frag to
1615 * config/obj-coff.c (SUB_SEGMENT_ALIGN): Likewise.
1616 * config/obj-ieee.c (SUB_SEGMENT_ALIGN): Likewise.
1617 (write_object_file): Invoke md_do_align if available, and use
1618 frag_align_code on text sections.
1619 * config/obj-vms.h (SUB_SEGMENT_ALIGN): Now two args.
1620 * config/tc-m88k.h (SUB_SEGMENT_ALIGN): Likewise.
1621 * config/tc-ppc.h (SUB_SEGMENT_ALIGN): Likewise.
1622 * config/tc-sh.h (SUB_SEGMENT_ALIGN): Likewise.
1623 * config/tc-i386.h (SUB_SEGMENT_ALIGN): Likewise. Define for
1626 2002-05-22 H.J. Lu <hjl@gnu.org>
1628 * dwarf2dbg.c (dwarf2_directive_loc): Call listing_source_file
1631 2002-05-22 Nick Clifton <nickc@cambridge.redhat.com>
1633 * config/tc-arm.c (arm_s_section): Enable for COFF builds as well
1636 2002-05-22 H.J. Lu <hjl@gnu.org>
1638 * dwarf2dbg.c (dwarf2_emit_insn): Emit only one line symbol
1639 for one .loc for compiler.
1641 2002-05-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1643 * config/tc-mips.c (macro): Relax warning, it's toot strict for
1646 2002-05-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1648 * config/tc-mips.c (macro2): Add 64 bit drol, dror macros.
1649 Optimize the rotate by zero case.
1651 2002-05-21 Nick Clifton <nickc@cambridge.redhat.com>
1653 * configure.in: Remove accidental enabling of bfd_gas=yes for
1655 * configure: Regenerate.
1657 2002-05-18 Kazu Hirata <kazu@cs.umass.edu>
1659 * app.c: Fix formatting.
1661 * ehopt.c: Likewise.
1663 * input-file.c: Likewise.
1664 * listing.c: Likewise.
1665 * macro.h: Likewise.
1666 * stabs.c: Likewise.
1667 * symbols.c: Likewise.
1669 2002-05-17 Alan Modra <amodra@bigpond.net.au>
1671 * config/obj-generic.c: Delete file.
1672 * config/obj-generic.h: Likewise.
1674 2002-05-16 Marek Michalkiewicz <marekm@amelek.gda.pl>
1676 * config/tc-avr.c (mcu_types): Update for new devices.
1678 2002-05-15 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1680 * config/tc-mips.c (macro): Warn about wrong la/dla use.
1682 2002-05-15 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1684 * config/tc_mips.c (s_cpsetup): Fix completely bogus code which had
1685 worked sometimes by accident. Fix copy&paste comment.
1687 2002-05-15 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1689 * config/tc-mips.c (md_begin): Fix .reginfo and .MIPS.option section
1690 alignment for NewABI. Let n32 use .reginfo. Remove useless casts.
1691 (mips_elf_final_processing): Let n32 use .reginfo.
1693 2002-05-15 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1695 * config/tc-mips.c (append_insn): Fix too small range of variable.
1697 2002-05-14 Nick Clifton <nickc@cambridge.redhat.com>
1699 * config/tc-arm.c (arm_cleanup): Remove redundant call to
1700 listing_prev_line().
1702 2002-05-13 Nick Clifton <nickc@cambridge.redhat.com>
1704 * config/tc-arm.c (md_assemble): Remove redundant call to
1707 * dwarf2dbg.c (dwarf2_emit_insn): Do not reset
1710 * stabs.c (s_stab_generic): Fix grammatical error in warning
1713 2002-05-13 Alan Modra <amodra@bigpond.net.au>
1715 * write.c (subsegs_finish): Don't specially align last subseg.
1717 2002-05-11 Nick Clifton <nickc@cambridge.redhat.com>
1719 * stabs.c (s_stab_generic): Warn about a description field that is
1722 * config/obj-coff.c: Fix compile time warnings when compiling
1723 without BFD_ASSEMBLER defined.
1726 * config/tc-sh.c (md_pcrel_from): Define for use with sh-hms
1728 (md_pcrel_from_section): Use md_pcrel_from().
1730 2002-05-11 Bruno Haible <bruno@clisp.org>
1732 * dwarf2dbg.c (dwarf2_emit_insn): Use the 'current' struct filled
1733 by dwarf2_directive_loc, instead of calling dwarf2_where.
1735 2002-05-11 Kazu Hirata <kazu@cs.umass.edu>
1737 * config/obj-coff.h: Fix formatting.
1738 * config/tc-mcore.c: Likewise.
1739 * config/tc-mn10300.c: Likewise.
1740 * config/tc-openrisc.c: Likewise.
1741 * config/tc-or32.c: Likewise.
1742 * config/tc-pdp11.c: Likewise.
1743 * config/tc-ppc.c: Likewise.
1744 * config/tc-ppc.h: Likewise.
1745 * config/tc-sh64.c: Likewise.
1746 * config/tc-sh.c: Likewise.
1747 * config/tc-tic54x.c: Likewise.
1748 * config/tc-xstormy16.c: Likewise.
1749 * config/tc-xstormy16.h: Likewise.
1751 2002-05-09 Kazu Hirata <kazu@cs.umass.edu>
1753 * config/obj-coff.c: Fix formatting.
1754 * config/obj-elf.c: Likewise.
1755 * config/tc-alpha.c: Likewise.
1756 * config/tc-arm.c: Likewise.
1757 * config/tc-d10v.c: Likewise.
1758 * config/tc-d30v.c: Likewise.
1759 * config/tc-h8300.c: Likewise.
1760 * config/tc-hppa.c: Likewise.
1762 2002-05-09 Alan Modra <amodra@bigpond.net.au>
1764 * config/tc-i386.c (md_estimate_size_before_relax) Don't lose
1765 reloc when no_cond_jump_promotion.
1767 2002-05-08 Jim Wilson <wilson@redhat.com>
1769 * config/tc-i960.c (md_estimate_size_before_relax): Return size of
1770 current variable part of frag.
1772 2002-05-08 Kazu Hirata <kazu@cs.umass.edu>
1774 * config/tc-mmix.c: Fix formatting.
1775 * config/tc-mmix.h: Likewise.
1777 2002-05-08 Alan Modra <amodra@bigpond.net.au>
1779 * configure: Regenerate.
1781 2002-05-07 Kazu Hirata <kazu@cs.umass.edu>
1783 * config/tc-m68k.c: Fix formatting.
1785 2002-05-07 Federico G. Schwindt <fgsch@olimpo.com.br>
1787 * Makefile.am: Honour DESTDIR.
1788 * Makefile.in: Regenerate.
1790 2002-05-06 Kazu Hirata <kazu@cs.umass.edu>
1792 * config/tc-ia64.c: Fix formatting.
1793 * config/tc-ia64.h: Likewise.
1795 2002-05-04 Kazu Hirata <kazu@cs.umass.edu>
1797 * config/tc-mips.c: Fix formatting.
1798 * config/tc-s390.c: Likewise.
1799 * config/tc-s390.h: Likewise.
1801 2002-05-03 Alexandre Oliva <aoliva@redhat.com>
1803 * config/tc-s390.c (md_gather_operands): Emit dwarf2 line-number
1804 information for instructions.
1806 2002-05-02 Kazu Hirata <kazu@cs.umass.edu>
1808 * as.h: Fix formatting.
1811 * dwarf2dbg.c: Likewise.
1812 * frags.h: Likewise.
1814 * macro.c: Likewise.
1816 * stabs.c: Likewise.
1817 * symbols.c: Likewise.
1819 2002-05-02 Alan Modra <amodra@bigpond.net.au>
1821 * app.c (mri_pseudo): Only declare for TC_M68K.
1823 * config/tc-ppc.c (mapping): Map sectoff to BFD_RELOC_16_BASEREL.
1824 (ppc_elf_validate_fix): Replace BFD_RELOC_32_BASEREL with
1825 BFD_RELOC_16_BASEREL.
1826 (md_assemble): Likewise.
1827 (md_apply_fix3): Likewise.
1829 2002-05-02 Nick Clifton <nickc@cambridge.redhat.com>
1831 * config/tc-arm.c (thumb_add_sub): Do not convert a subtract of
1832 zero into an add of zero - it is not the same.
1834 2002-05-01 Arati Dikey <aratid@kpit.com>
1836 * tc-sh.c (get_specific): Generate warning if the same
1837 destination register is used in parallel instructions.
1839 2002-05-01 Andrew Macleod <amacleod@cygnus.com>
1841 * config/tc-i386.c (extra_symbol_chars): Add '[' to the list.
1843 2002-05-01 Alan Modra <amodra@bigpond.net.au>
1845 * write.c (cvt_frag_to_fill): Set fr_offset to zero on .org
1846 backwards to prevent cascading errors.
1848 2002-04-30 Mark Mitchell <mark@codesourcery.com>
1850 * configure.in: Add support for powerpc-*-windiss.
1851 * configure: Regenerated.
1853 2002-04-28 Alan Modra <amodra@bigpond.net.au>
1855 * config/tc-s390.c (md_parse_option): Formatting.
1857 * config/tc-i386.c: Formatting fixes, add missing space in error
1860 2002-04-24 Christian Groessler <chris@groessler.org>
1862 * config/tc-z8k.c (build_bytes): Add support for new cases:
1863 CLASS_IGNORE and ARG_NIM4.
1864 (md_assemble): Prevent destruction of input_line_pointer if
1865 get_operands returns failure.
1867 2002-04-24 Chris G. Demetriou <cgd@broadcom.com>
1869 * config/tc-mips.c (macro_build): Do _not_ allow MIPS-3D
1870 instructions to be generated by macros.
1872 2002-04-24 Andreas Schwab <schwab@suse.de>
1874 * config/tc-i386.c (output_jump, output_disp)
1875 (md_estimate_size_before_relax): Don't set fx_pcrel_adjust any
1877 (md_apply_fix3): Remember addend value for rela relocations.
1878 (tc_gen_reloc): Correctly compute pc-relative relocation addend.
1880 2002-04-22 Chris Demetriou <cgd@broadcom.com>
1882 * config/tc-mips.c (macro_build): Add close-parenthesis missing
1883 from previous change.
1885 2002-04-22 Eric Christopher <echristo@redhat.com>
1887 * config/tc-mips.c (macro_build): Add warning if macro instructions
1888 are expanded into a branch delay slot.
1890 2002-04-17 Geoffrey Keating <geoffk@redhat.com>
1892 * dwarf2dbg.c (dwarf2_gen_line_info): Do emit duplicate line
1893 numbers, gdb relies on them to detect the start of the prologue.
1895 2002-04-17 Martin Schwidefsky <schwidefsky@de.ibm.com>
1897 * config/tc-s390.c (tc_s390_fix_adjustable): Prevent adjustments to
1898 symbols in merge sections.
1900 2002-04-16 Alan Modra <amodra@bigpond.net.au>
1902 * as.c (main): Don't reference _bfd_chunksize.
1904 2002-04-15 Tom Rix <trix@redhat.com>
1906 * config/tc-d10v.c (d10v_fix_adjustable): Prevent adjustments to
1907 symbols in merge sections.
1909 2002-04-11 Richard Sandiford <rsandifo@redhat.com>
1911 * doc/invoke.texi (TC_LARGEST_EXPONENT_IS_NORMAL): Document.
1912 * config/atof-ieee.c (TC_LARGEST_EXPONENT_IS_NORMAL): Add an
1913 argument for the precision.
1914 (gen_to_words): Update accordingly.
1916 2002-04-10 Alan Modra <amodra@bigpond.net.au>
1918 * as.c (parse_args <OPTION_VERSION>): Use VERSION is
1919 BFD_VERSION_STRING unavailable.
1920 * config/tc-i386.c (INLINE): Define (for non-BFD assembler).
1922 2002-04-09 J"orn Rennecke <joern.rennecke@superh.com>
1924 * config/tc-sh.h (TC_FIX_ADJUSTABLE): Disable adjusting if
1925 symbol_used_in_reloc_p is true.
1927 * config/tc-sh.c (md_apply_fix3): Don't zero relocations on big
1930 2002-04-04 Alan Modra <amodra@bigpond.net.au>
1932 * dep-in.sed: Cope with absolute paths.
1933 * Makefile.am (dep.sed): Subst TOPDIR.
1935 * Makefile.in: Regenerate.
1937 2002-04-04 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1939 * config/tc-mips.c (mips16_macro_build): Cast type mismatch.
1940 (mips_ip): Remove unused variable.
1941 (md_apply_fix3): Cast signed/unsignes mismatches. Replace
1942 unsigned char with bfd_byte.
1943 (s_file): Remove unused variable.
1944 (s_mips_ent): Likewise.
1946 2002-04-03 Tom Rix <trix@redhat.com>
1948 * config/tc-d10v.c (d10v_insert_operand): Fix warning in as_bad_where.
1950 (find_opcode): Fix warning in as_warn.
1951 * config/tc-d10v.h: Update Copyright.
1953 2002-04-03 Alan Matsuoka <alanm@redhat.com>
1954 Tom Rix <trix@redhat.com>
1956 From Jeff Knaggs <jknaggs@redhat.com>
1957 * config/tc-d10v.c (check_resource_conflict): New function to
1958 check for resource conflicts.
1960 From Jason Eckhardt <jle@redhat.com>
1961 * config/tc-d10v.c (build_insn): Check for unresolved imm4 or
1963 * config/tc-d10v.c (find_opcode): Emit a warning if one of the
1964 reserved control registers is used.
1965 * config/tc-d10v.c (build_insn): Check for unresolved imm4 or
1968 From 2001-03-28 Diego Novillo <dnovillo@redhat.com>
1969 * tc-d10v.c (parallel_ok): Prevent packing only if the first
1970 instruction cannot be packed.
1972 From 2001-03-30 Diego Novillo <dnovillo@redhat.com>
1973 * gas/config/tc-d10v.c (check_resource_conflict): Only check
1974 write-write conflicts.
1975 (md_assemble): Reformat introductory comment.
1976 * opcodes/d10v-opc.c (d10v_opcodes): `btsti' does not modify its
1979 2002-04-03 Alan Modra <amodra@bigpond.net.au>
1981 * symbols.c (resolve_symbol_value <O_uminus, O_bit_not,
1982 O_logical_not>): Derive final_seg from add_symbol.
1983 <O_multiply..O_logical_or>: More final_seg twiddles.
1985 2002-04-01 Jessica Han <jessica@cup.hp.com>
1987 * config/tc-ia64.c (ia64_cons_fix_new): Handle 8 byte iplt reloc
1990 2002-03-27 Andreas Schwab <schwab@suse.de>
1992 * config/tc-i386.c (output_jump): Set fx_pcrel_adjust to size of
1993 field for pc-relative fixups.
1994 (output_disp): Likewise.
1995 (md_estimate_size_before_relax): Likewise.
1996 (tc_gen_reloc): Subtract fx_pcrel_adjust instead of fx_size for
1997 pc-relative fixups in 64bit mode.
1999 2002-03-22 Alan Modra <amodra@bigpond.net.au>
2001 * config/te-aix5.h: Typo fix.
2003 2002-03-21 Alan Modra <amodra@bigpond.net.au>
2005 * Makefile.am: Run "make dep-am".
2006 * Makefile.in: Regenerate.
2007 * doc/Makefile.in: Regenerate.
2009 2002-03-20 Albert Chin-A-Young <china@thewrittenword.com>
2011 * config/tc-arm.c (vfp_dp_reg_required_here): Fix typo
2012 (vfp_sp_reg_pos -> vfp_dp_reg_pos).
2014 2002-03-18 Alexandre Oliva <aoliva@redhat.com>
2016 * config/tc-mips.c (md_estimate_size_before_relax): Do not modify
2017 the EXTENDED bit here; report the estimate according to the
2020 2002-03-18 Tom Rix <trix@redhat.com>
2022 * configure.in: Add AIX 5 support.
2023 * config/tc-ppc.c (ppc_target_format): Add AIX 5 64 bit target.
2024 * config/te-aix5.h: New file. AIX 5 support.
2025 * configure: Regenerate.
2027 2002-03-18 Nick Clifton <nickc@cambridge.redhat.com>
2029 * po/fr.po: Updated version.
2031 2002-03-16 Andreas Jaeger <aj@suse.de>
2033 * doc/c-mips.texi (Machine Dependencies): Add new node.
2035 2002-03-15 Chris G. Demetriou <cgd@broadcom.com>
2037 * config/tc-mips.c (mips_set_options): New "ase_mips3d" member.
2038 (mips_opts): Initialize "ase_mips3d" member.
2039 (file_ase_mips3d): New variable.
2040 (CPU_HAS_MIPS3D): New macro.
2041 (md_begin): Initialize mips_opts.ase_mips3d and file_ase_mips3d
2042 based on command line options and configuration defaults.
2043 (macro_build, mips_ip): Accept MIPS-3D instructions if
2044 mips_opts.ase_mips3d is set.
2045 (OPTION_MIPS3D, OPTION_NO_MIPS3D, md_longopts, md_parse_option):
2046 Add support for "-mips3d" and "-no-mips3d" options.
2047 (OPTION_ELF_BASE): Move to accomodate new options.
2048 (s_mipsset): Support ".set mips3d" and ".set nomips3d".
2049 (mips_elf_final_processing): Add a comment indicating that a
2050 MIPS-3D ASE ELF header flag should be set, when one exists.
2051 * doc/as.texinfo: Document -mips3d and -no-mips3d options.
2052 * doc/c-mips.texi: Likewise, and document ".set mips3d" and ".set
2053 nomips3d" directives.
2055 2002-03-14 Hans-Peter Nilsson <hp@bitrange.com>
2057 * config/tc-mmix.c (md_estimate_size_before_relax): Don't consider
2058 a weak symbol in same section to be within reach.
2060 2002-03-13 Nick Clifton <nickc@cambridge.redhat.com>
2062 * po/fr.po: Updated version.
2064 2002-03-12 Andreas Schwab <schwab@suse.de>
2066 * config/tc-ia64.c (fixup_unw_records): Clear region when seeing a
2067 body record so that an error is given for misplaced .save
2070 2002-03-09 Alan Modra <amodra@bigpond.net.au>
2072 * config/tc-i386.h (REX_OPCODE): Define.
2073 (REX_MODE64, REX_EXTX, REX_EXTY, REX_EXTZ): Define.
2074 (rex_byte): typedef to int.
2075 * config/tc-i386.c: Group prototypes and vars together.
2076 Formatting fixes. Remove occurrences of "register" keyword.
2079 (mode_from_disp_size): Add INLINE keyword to prototype.
2080 (fits_in_signed_byte): Likewise.
2081 (fits_in_unsigned_byte): Likewise.
2082 (fits_in_unsigned_word): Likewise.
2083 (fits_in_signed_word): Likewise.
2084 (fits_in_unsigned_long): Likewise.
2085 (fits_in_signed_long): Likewise.
2086 (type_names): Constify.
2087 (intel_float_operand): Constify param.
2088 (add_prefix): Use REX_OPCODE.
2089 (md_assemble): Likewise. Modify for changed rex_byte.
2090 (parse_insn): Split out of md_assemble.
2091 (parse_operands): Likewise.
2092 (swap_operands): Likewise.
2093 (optimize_imm): Likewise.
2094 (optimize_disp): Likewise.
2095 (match_template): Likewise.
2096 (check_string): Likewise.
2097 (process_suffix): Likewise.
2098 (check_byte_reg): Likewise.
2099 (check_long_reg): Likewise.
2100 (check_qword_reg): Likewise.
2101 (check_word_reg): Likewise.
2102 (finalize_imm): Likewise.
2103 (process_operands): Likewise.
2104 (build_modrm_byte): Likewise.
2105 (output_insn): Likewise.
2106 (output_branch): Likewise.
2107 (output_jump): Likewise.
2108 (output_interseg_jump): Likewise.
2109 (output_disp): Likewise.
2110 (output_imm): Likewise.
2112 2002-03-07 Daniel Jacobowitz <drow@mvista.com>
2114 * doc/as.texinfo: Wrap @menu in @ifnottex, not @ifinfo.
2116 2002-03-06 Alan Modra <amodra@bigpond.net.au>
2118 * config/tc-i386.c (tc_gen_reloc): Don't attempt to handle 8 byte
2119 relocs except when BFD64.
2121 * write.c (number_to_chars_bigendian): Don't abort when N is
2122 larger than sizeof (VAL).
2123 (number_to_chars_littleendian): Likewise.
2125 2002-03-05 John David Anglin <dave@hiauly1.hia.nrc.ca>
2127 * config/tc-hppa.c (md_apply_fix3): Add cast.
2128 (hppa_fix_adjustable): Adjust list of selectors using e_lrsel and
2131 2002-03-05 Paul Koning <pkoning@equallogic.com>
2133 * tc-pdp11.c: Use VAX float format support for PDP-11 target.
2134 (parse_ac5): New function for parsing float regs in float operand.
2135 (parse_expression): Remove attempt to make literals be octal.
2136 (parse_op_no_deferred): Support float literals.
2137 (parse_op): Reject attempts to refer to float regs.
2138 (parse_fop): New function, like parse_op but for float operand.
2139 (md_assemble): Add cases to parse float operands. Also fix
2140 IMM3, IMM6, IMM8 cases to pick up the operand from the right spot.
2142 2002-03-04 H.J. Lu <hjl@gnu.org>
2144 * config/obj-elf.c (special_section): Add .init_array,
2145 .fini_array and .preinit_array.
2147 * config/tc-ia64.h (ELF_TC_SPECIAL_SECTIONS): Remove
2148 .init_array and .fini_array.
2150 2002-03-01 Jakub Jelinek <jakub@redhat.com>
2152 * config/obj-elf.c (elf_copy_symbol_attributes): Don't copy
2154 (obj_frob_symbol): Copy visibility.
2156 2002-02-28 Jakub Jelinek <jakub@redhat.com>
2158 * config/tc-alpha.c (s_alpha_text): Use obj_elf_text for OBJ_ELF, not
2160 (s_alpha_data): Use obj_elf_data for OBJ_ELF, not s_data.
2162 2002-02-27 Nick Clifton <nickc@cambridge.redhat.com>
2164 * po/es.po: Updated.
2166 2002-02-26 Chris Demetriou <cgd@broadcom.com>
2168 * config/tc-mips.c (mips_need_elf_addend_fixup): For embedded-PIC
2169 only, undo the changes made on 2001-06-08, with the
2170 effect being that common or extern symbols are
2171 adjusted for embedded-PIC, but weak symbols are not.
2172 (md_estimate_size_before_relax: Likewise, with the effect
2173 that extern symbols are treated the same as weak symbols
2174 only if not embedded-PIC.
2175 (mips_fix_adjustable) Likewise, with the effect that
2176 weak or extern symbols are not adjusted for embedded-PIC.
2177 (md_apply_fix3): Tweak so that the case where value is zero
2178 is handled more correctly for embedded-PIC code.
2180 2002-02-26 Nick Clifton <nickc@cambridge.redhat.com>
2182 * doc/as.texinfo (Overview): Add missing @ifset IA64
2184 * configure.in (LINGUAS): Add es.po.
2185 * configure: Regenerate.
2186 * po/es.po: New file.
2188 2002-02-25 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
2190 * config/tc-mips.c (set_at): Fix handling of 64bit register loads.
2191 (macro): Likewise. Fix la/dla address expansions for EMBEDDED_PIC
2192 and NO_PIC cases. Code cleanup.
2193 (macro2): Fix handling of 64bit register loads.
2195 2002-02-25 David Mosberger <davidm@hpl.hp.com>
2197 * doc/as.texinfo: Add entry for IA64.
2198 * doc/c-ia64.texi: New file.
2200 2002-02-25 Alan Modra <amodra@bigpond.net.au>
2202 * config/tc-hppa.c: Update copyright date.
2204 * doc/c-ppc.texi (PowerPC-Opts): Add -mpower4, -maltivec and -m7400
2205 Remove references to chip manufacturers.
2206 * config/tc-ppc.c (md_parse_option): Handle -mpower4 option.
2208 (md_show_usage): Remove references to chip manufacturers. Mention
2210 (md_begin): Test power4 opcode flag bits.
2212 2002-02-22 David Mosberger <davidm@hpl.hp.com>
2214 * config/tc-ia64.c (dot_restore): Issue error message of epilogue
2215 count exceeds prologue count.
2216 (md_show_usage): Describe -mconstant-gp and -mauto-pic.
2217 (unwind.label_prologue_count): New member.
2219 Based on a patch by Hans Boehm <hboehm@hpl.hp.com>:
2221 (get_saved_prologue_count): New function.
2222 (save_prologue_count): New function.
2223 (free_saved_prologue_count): New function.
2224 (dot_label_state): Record state label by calling save_prologue_count().
2225 (dot_copy_state): Restore prologue count by calling
2226 get_saved_prologue_count().
2227 (generate_unwind_image): Free up list of saved prologue
2228 counts by calling free_saved_prologue_counts().
2230 2002-02-22 Nick Clifton <nickc@cambridge.redhat.com>
2232 * config/tc-tic54x.c: Add missing prototypes and remove ANSI style
2233 function declarations.
2235 2002-02-21 Nick Clifton <nickc@cambridge.redhat.com>
2237 * NEWS: Note that GASP is now deprecated.
2238 * Makefile.am: Do not build gasp-new by default.
2239 * Makefile.in: Regenerate.
2240 * doc/Makefile.am: Do not install gasp.info.
2241 * doc/Makefile.in: Regenerate.
2242 * gas/gasp.texi: Note that gasp is now deprecated.
2244 2002-02-20 Nick Clifton <nickc@cambridge.redhat.com>
2246 * NEWS: Mark 2.12 branch.
2248 2002-02-19 Tom Tromey <tromey@redhat.com>
2250 * config/tc-xstormy16.h (DWARF2_LINE_MIN_INSN_LENGTH): Define.
2252 2002-02-19 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
2254 * config/tc-mips.c (md_parse_option): Complain about invalid -mabi
2257 2002-02-19 Martin Schwidefsky <schwidefsky@de.ibm.com>
2259 * config/tc-s390.c (md_parse_option): Add switches -m31 and -m64.
2260 Make bit size independent of architecture switch.
2261 (md_begin): Add warning for -m64 with -Aesa.
2262 (s390_md_end): Use renamed architecture defines.
2264 2002-02-18 Daniel Jacobowitz <drow@mvista.com>
2266 * config/obj-coff.h: Check !target_big_endian, not shl, for coff-sh.
2268 2002-02-16 Nick Clifton <nickc@cambridge.redhat.com>
2270 * doc/as.texinfo (Machine Dependencies): Fix typo: MMIX used
2273 2002-02-15 Chris Demetriou <cgd@broadcom.com>
2275 * config/tc-mips.c (md_estimate_size_before_relax): Really
2276 make sure we treat weak like extern only for ELF. (Fixes
2277 patch from 2001-07-25.)
2279 2002-02-15 Ben Elliston <bje@redhat.com>
2281 * doc/as.texinfo: Add duplicate directory entry so that "info gas"
2284 2002-02-15 Hans-Peter Nilsson <hp@bitrange.com>
2286 * NEWS: Mention support for MMIX.
2288 2002-02-13 Chris Demetriou <cgd@broadcom.com>
2290 * config/tc-mips.c (mips_need_elf_addend_fixup): Restructure into
2291 a sequence of indpendent 'if' statements for easier debugging
2292 and future modification.
2294 2002-02-13 Matt Fredette <fredette@netbsd.org>
2296 * config/tc-m68k.c (md_show_usage): No longer display a
2297 hard-coded "68020" for the default CPU, instead display the
2298 canonical name of the true, configured default CPU.
2299 (m68k_elf_final_processing): Mark objects for sub-68020
2300 CPUs with the new EF_M68000 flag.
2302 2002-02-13 Andreas Schwab <schwab@suse.de>
2304 * config/tc-i386.c (tc_i386_fix_adjustable): Don't adjust
2305 pc-relative relocations to merge sections in 64-bit mode.
2307 2002-02-13 Ben Elliston <bje@redhat.com>
2309 * NEWS: Document floating point number handling in gasp.
2310 * gasp.c: Include <assert.h> and "xregex.h".
2311 (is_flonum): New function.
2312 (chew_flownum): Likewise.
2313 (change_base): Consume flonums from the input, where possible.
2314 * doc/gasp.texi (Constants): Document floating point numbers.
2316 2002-02-12 John David Anglin <dave@hiauly1.hia.nrc.ca>
2318 * config/tc-hppa.c (hppa_fix_adjustable): Don't adjust final types
2319 that implicitly use LR and RR selectors.
2321 2002-02-12 Alexandre Oliva <aoliva@redhat.com>
2323 * config/tc-mn10300.c (other_registers): Added `epsw'. Mark `pc'
2324 and `epsw' as available on AM33 and above only.
2325 (other_register_name): Add logic to handle machine type encoded in
2328 2002-02-11 Tom Rix <trix@redhat.com>
2330 * config/tc-ppc.c (md_apply_fix3): Check for insert fop invalid for
2333 2002-06-11 Alexandre Oliva <aoliva@redhat.com>
2335 * config/tc-sparc.c (U0x80000000, U0xffffffff): New constants.
2338 2002-02-11 Jan Hubicka <jh@suse.cz>
2340 * i386.c (md_assemble): Support 32bit address prefix.
2341 (i386_displacement): Likewise.
2342 (i386_index_check): Accept 32bit addressing in 64bit mode.
2344 2002-02-11 Alexandre Oliva <aoliva@redhat.com>
2346 * config/tc-sh.c (dot): Removed unused function.
2348 2002-02-11 Alan Modra <amodra@bigpond.net.au>
2350 * Makefile.am: "make dep-am".
2351 * Makefile.in: Regenerate.
2352 * aclocal.m4: Regenerate.
2353 * config.in: Regenerate.
2354 * configure: Regenerate.
2355 * doc/Makefile.in: Regenerate.
2357 2002-02-10 Richard Henderson <rth@redhat.com>
2359 * doc/c-alpha.texi: New file.
2360 * doc/Makefile.am (CPU_DOCS): Add it.
2361 * doc/all.texi, doc/as.texinfo: Add hooks for Alpha.
2363 2002-02-09 Richard Henderson <rth@redhat.com>
2365 * config/tc-alpha.c (O_samegp): New.
2366 (USER_RELOC_P): Include it.
2367 (alpha_reloc_op_tag, debug_exp, find_macro_match): Add it.
2368 (md_apply_fix3): Handle BFD_RELOC_ALPHA_BRSGP.
2369 (alpha_force_relocation, alpha_fix_adjustable): Likewise.
2370 (alpha_validate_fix): New.
2371 * config/tc-alpha.h (TC_VALIDATE_FIX): New.
2373 2002-02-09 Hans-Peter Nilsson <hp@axis.com>
2375 * doc/c-cris.texi: New.
2376 * doc/all.texi: @set CRIS.
2377 * doc/as.texinfo: Ditto. Add CRIS gas manpage option overview.
2378 Include c-cris.texi.
2379 * doc/Makefile.am (CPU_DOCS): Add c-cris.texi
2380 * doc/Makefile.in: Regenerate.
2382 2002-02-08 Chris Demetriou <cgd@broadcom.com>
2384 * config/tc-mips.c (IS_SEXT_32BIT_NUM): New macro to
2385 determine if a number is a sign-extended 32-bit number.
2386 (load_register): Use IS_SEXT_32BIT_NUM.
2387 (macro): Check if load/store macro handling is using a
2388 constant 32-bit address on 64-bit address systems, and if
2389 so optimize the generation of that address.
2391 2002-02-08 Richard Henderson <rth@redhat.com>
2393 * config/tc-alpha.c (alpha_force_relocation): Don't assert that
2394 we've eliminated all foreign relocation types yet.
2395 (alpha_fix_adjustable): Likewise.
2397 2002-02-08 Alexandre Oliva <aoliva@redhat.com>
2399 Contribute sh64-elf.
2400 2002-02-08 Alexandre Oliva <aoliva@redhat.com>
2401 Stephen Clarke <Stephen.Clarke@st.com>
2402 * doc/c-sh64.texi: Fix citation of SH64 architecture manual.
2403 2002-01-31 Alexandre Oliva <aoliva@redhat.com>
2404 * config/tc-sh.c (md_relax_table): Added default sizes for
2405 non-PC-relative UNDEF_MOVI, and relaxation sequences for
2406 MOVI_16, MOVI_32 and MOVI_48.
2407 * config/tc-sh64.c (shmedia_md_apply_fix3): Fix warning.
2408 (shmedia_md_convert_frag): Handle non-PC-relative UNDEF_MOVI
2410 (shmedia_md_estimate_size_before_relax): Remove redundant
2411 blocks. Set fragP->fr_var even if relaxation type unchanged.
2412 Retain UNDEF_MOVI until expression decays to number.
2413 2002-01-24 Alexandre Oliva <aoliva@redhat.com>
2414 * config/tc-sh64.c (shmedia_init_reloc): Handle new SHmedia PIC
2415 relocation types. Take fixP->fx_addnumber into account too.
2416 (shmedia_md_apply_fix): Likewise.
2417 (shmedia_md_convert_frag): Likewise.
2418 (shmedia_build_Mytes): Likewise.
2419 (sh64_consume_datalabel): Complain about nested datalabel.
2420 Support PIC relocs. Call sh_parse_name.
2421 * config/tc-sh64.h (TC_RELOC_RTSYM_LOC_FIXUP): Extend definition
2422 in tc-sh.h to SHmedia reloc types.
2423 * config/tc-sh.c (SH64PCRELPLT, MOVI_PLT, MOVI_GOTOFF,
2424 MOVI_GOTPC): New relaxation constants.
2425 (md_relax_table): Introduce relaxation directives for PIC-related
2427 (sh_PIC_related_p): Handle datalabel.
2428 (sh_check_fixup): Choose SH5 PIC relocations.
2429 (sh_cons_fix_new): Added BDF_RELOC_64.
2430 (md_apply_fix3, sh_parse_name): Handle GOTPLT.
2431 2002-01-18 Alexandre Oliva <aoliva@redhat.com>
2432 * config/tc-sh64.c (sh64_max_mem_for_rs_align_code): If the
2433 current ISA is SHmedia, get 7 bytes.
2434 2001-11-28 Nick Clifton <nickc@cambridge.redhat.com>
2435 * config/tc-sh.c (md_apply_fix3): Treat shmedia_md_apply_fix3 as a
2437 * config/tc-sh64.c (shmedia_apply_fix): Rename to
2438 shmedia_apply_fix3 and make void.
2439 2001-05-17 Alexandre Oliva <aoliva@redhat.com>
2440 * config/tc-sh64.c (s_sh64_abi): Remove unused arguments passed to
2442 2001-04-12 Alexandre Oliva <aoliva@redhat.com>
2443 * config/tc-sh64.h (md_parse_name): Take &c as argument.
2444 2001-03-14 DJ Delorie <dj@redhat.com>
2445 * doc/Makefile.am (CPU_DOCS): Added c-sh64.texi
2446 * doc/Makefile.in(CPU_DOCS): Ditto.
2447 * doc/c-sh64.texi: New file.
2448 * doc/as.texinfo: Add SH64 support.
2449 2001-03-13 DJ Delorie <dj@redhat.com>
2450 * config/tc-sh64.c (shmedia_get_operands): Rename A_RESV_Fx to
2451 A_REUSE_PREV so that its purpose is more obvious.
2452 (shmedia_build_Mytes): Ditto.
2453 2001-03-07 DJ Delorie <dj@redhat.com>
2454 * config/tc-sh64.c (sh64_vtable_entry): New, strip datalabels
2456 (sh64_vtable_inherit): Ditto.
2457 (strip_datalabels): New, strip "datalabel" from given line.
2458 * config/tc-sh.c (md_pseudo_table): Add sh64-specific vtable
2460 2001-03-06 Hans-Peter Nilsson <hpn@cygnus.com>
2461 * config/tc-sh64.c (shmedia_md_assemble): Move dwarf2_emit_insn
2463 (shmedia_build_Mytes): ... to here.
2464 2001-03-06 DJ Delorie <dj@redhat.com>
2465 * config/tc-sh.c: Remove sh64-specific uaquad now that there
2467 2001-01-21 Hans-Peter Nilsson <hpn@cygnus.com>
2468 * config/tc-sh64.h (DWARF2_LINE_MIN_INSN_LENGTH): Override.
2469 * config/tc-sh64.c (shmedia_md_assemble): Offset recorded insn
2470 address by one in call to dwarf2_emit_insn.
2471 2001-01-13 Hans-Peter Nilsson <hpn@cygnus.com>
2472 Implement ".abi" pseudo and correct .cranges descriptors. Correct
2473 alignment handling broken by imported changes.
2474 * config/tc-sh64.h (HANDLE_ALIGN): Override definition in tc-sh.h.
2475 (sh64_handle_align): Declare.
2476 (MAX_MEM_FOR_RS_ALIGN_CODE): Override definition in tc-sh.h.
2477 (sh64_max_mem_for_rs_align_code): Declare.
2478 (enum sh64_isa_values): Moved here from tc-sh64.c.
2479 (md_do_align): Define.
2480 (sh64_do_align): Declare.
2481 (struct sh64_tc_frag_data): New.
2482 (TC_FRAG_TYPE): Change to struct sh64_tc_frag_data. Users
2484 (TC_FRAG_INIT): Change to set new datatype.
2485 (struct sh64_segment_info_type): Rename member
2486 last_flushed_location to last_contents_mark. All users changed.
2487 (md_elf_section_change_hook, TC_CONS_FIX_NEW): Do not define.
2488 (shmedia_elf_new_section, sh64_tc_cons_fix_new): Do not prototype.
2489 * config/tc-sh.c (md_pseudo_table): Add ".abi".
2490 (sh_elf_cons) [HAVE_SH64]: Call sh64_update_contents_mark instead
2491 of unsetting seen_insn.
2492 (md_assemble) [HAVE_SH64] <before new SHcompact sequence>: Also
2493 call sh64_update_contents_mark.
2494 (sh_handle_align): Remove HAVE_SH64-conditioned code.
2495 * config/tc-sh64.c (sh64_isa_mode): Correct type from boolean to
2496 enum sh64_isa_values.
2497 (sh64_set_contents_type): Drop segT parameter. All callers changed.
2498 (emitting_crange): Boolean guard moved to file scope from function
2499 scope in sh64_set_contents_type.
2501 (sh64_update_contents_mark): New; most split out from
2502 sh64_flush_pending_output.
2503 (shmedia_md_end): Call sh64_update_contents_mark. Set
2504 sh64_isa_mode to sh64_isa_sh5_guard unless sh64_isa_unspecified.
2505 (sh64_do_align): New function.
2506 (sh64_max_mem_for_rs_align_code): New function.
2507 (sh64_handle_align): Rename from shmedia_do_align. Make
2508 non-static. Add head comment. Emit zero bytes for n bytes modulo
2509 four. Change return-type to void.
2510 (shmedia_elf_new_section): Remove.
2511 (shmedia_md_assemble): Call sh64_update_contents_mark.
2512 (s_sh64_mode): Ditto. Do not call md_flush_pending_output. Make
2513 new frag. Call sh64_update_contents_mark after making the new
2515 (sh64_flush_pending_output): Just call sh64_update_contents_mark
2516 and sh_flush_pending_output.
2517 (sh64_flag_output): Also call md_flush_pending_output, but add
2518 condition on not emitting_crange.
2519 (sh64_tc_cons_fix_new): Remove.
2520 2001-01-12 Nick Clifton <nickc@redhat.com>
2521 * config/tc-sh64.c (shmedia_do_align): Fix to work with new
2522 alignment handling scheme imported from sourceware.
2523 2001-01-12 Hans-Peter Nilsson <hpn@cygnus.com>
2524 * config/tc-sh64.h (TARGET_FORMAT): Define.
2525 (sh64_target_format): Prototype.
2526 * config/tc-sh64.c (sh64_target_mach): New function.
2527 2001-01-07 Hans-Peter Nilsson <hpn@cygnus.com>
2528 * config/tc-sh64.c (shmedia_md_end): When equating a symbol, use
2529 zero_address_frag instead of copying the frag of the symbol.
2530 (shmedia_frob_file_before_adjust): Ditto.
2531 (shmedia_md_apply_fix) <case BFD_RELOC_SH_IMM_MEDLOW16>: Cast mask
2532 to valueT to remove signedness.
2533 (shmedia_md_convert_frag): Add parameter final. Rename parameter
2534 headers to output_bfd. Do not evaluate symbols if final is false;
2536 (shmedia_md_estimate_size_before_relax) <case C (MOVI_IMM_32,
2537 UNDEF_MOVI) et al>: If symbol cannot be modified to be PC-relative
2538 to the current frag, call shmedia_md_convert_frag to emit fixups
2539 and make frag_wane neutralize the frag. Update comments.
2540 * config/tc-sh.c (md_convert_frag): Change caller of
2541 shmedia_md_convert_frag.
2542 2001-01-06 Hans-Peter Nilsson <hpn@cygnus.com>
2543 * config/tc-sh64.h: Tweak comments and correct formatting.
2544 * config/tc-sh64.c: Ditto.
2545 (shmedia_md_convert_frag) <PT/PTA/PTB 32, 48 and 64 bit
2546 expansion, MOVI pcrel expansion>: Fix thinko calculating offset
2547 for the no-relocation case.
2548 (shmedia_check_limits): Fix range check being off-by-one for PTA.
2549 * config/tc-sh.c: Ditto. Add proper comments to #ifdef/#ifndef
2551 (SH64PCREL16_F): Increment for proper max-PTA handling. Update
2553 (SH64PCREL16_M, MOVI_16_M): Correct range thinko.
2554 (SH64PCREL48_M, MOVI_48_M): Similar; don't count in length of
2556 (SH64PCREL32_M, MOVI_32_M): Ditto; handle overflowing expression.
2558 2001-01-05 Hans-Peter Nilsson <hpn@cygnus.com>
2559 * config/tc-sh64.c (shmedia_md_apply_fix) <second switch, case
2560 BFD_RELOC_SH_PT_16>: Set lowest bit in field to be relocated to 1.
2561 (shmedia_md_convert_frag) <case C (SH64PCREL16_32, SH64PCREL16) et
2562 al>: Set lowest bit of field to relocate to 1 and rest to empty,
2563 if reloc is emitted.
2564 2000-12-31 Hans-Peter Nilsson <hpn@cygnus.com>
2565 New options plus bugfixes.
2566 * config/tc-sh.c (md_longopts): New options "-no-expand" and
2568 (md_parse_option): Handle new options.
2569 (md_show_usage): Add blurb for new options.
2570 * config/tc-sh64.c (SHMEDIA_BFD_RELOC_PT): New macro.
2571 (sh64_expand, sh64_pt32): New variables.
2572 (shmedia_init_reloc): Handle BFD_RELOC_SH_PT_16.
2573 (shmedia_md_apply_fix): Hold original fixP->fx_r_type in
2574 orig_fx_r_type. Change SHMEDIA_BFD_RELOC_PT into
2575 BFD_RELOC_SH_PT_16. Handle BFD_RELOC_SH_PT_16 as pc-relative.
2576 <resolved previously-pc-relative relocs>: Handle
2577 SHMEDIA_BFD_RELOC_PT and BFD_RELOC_SH_PT_16.
2578 (shmedia_md_convert_frag) <case C (SH64PCREL16PT_64, SH64PCREL16),
2579 case C (SH64PCREL16PT_32, SH64PCREL16)>: Modify to PTB if operand
2580 points to SHcompact code.
2581 <case C (SH64PCREL16_32, SH64PCREL16), case C (SH64PCREL16_64,
2582 SH64PCREL16)>: Check that ISA of what operand points at and
2583 PTA/PTB matches, or emit error.
2584 (shmedia_check_limits): Handle BFD_RELOC_SH_PT_16 and
2585 SHMEDIA_BFD_RELOC_PT.
2586 (shmedia_immediate_op): If pcrel, emit fixup also for constant
2588 (shmedia_build_Mytes) <case A_IMMS16>: Also check sh64_expand in
2589 condition for MOVI expansion.
2590 <case A_PCIMMS16BY4>: Handle expansion to 32 bits only, if
2591 sh64_pt32. Emit only a BFD_RELOC_SH_PT_16 fixup if not
2593 <case A_PCIMMS16BY4_PT>: Likewise, but emit a SHMEDIA_BFD_RELOC_PT
2595 (sh64_target_format): Error-check setting of sh64_pt32 and
2596 sh64_expand. Fix typo in check for sh64_shcompact_const_crange.
2597 (shmedia_md_pcrel_from_section): Handle BFD_RELOC_SH_PT_16 and
2598 SHMEDIA_BFD_RELOC_PT as coming from SHmedia code.
2599 2000-12-31 Hans-Peter Nilsson <hpn@cygnus.com>
2600 * config/tc-sh64.c: Improve comments.
2601 (shmedia_md_convert_frag): Remove inactive is_pt_variant code. Do
2602 not say the linker will check correctness of PTA/PTB expansion.
2603 (shmedia_md_end): Make non-static.
2604 * config/tc-sh64.h (md_end): Define to shmedia_md_end. Add
2606 * config/tc-sh.c (sh_finalize): Remove.
2607 * config/tc-sh.h (md_end): Do not define.
2608 Remove prototype for sh_finalize.
2609 2000-12-30 Hans-Peter Nilsson <hpn@cygnus.com>
2610 * config/tc-sh64.c (shmedia_frob_section_type): Use a struct
2611 sh64_section_data container when storing section type in tdata
2612 field in elf_section_data.
2613 * config/tc-sh.c (sh_elf_final_processing): Change from EF_SH64 to
2615 * Makefile.am: Update dependencies.
2616 * Makefile.in: Regenerate.
2617 2000-12-22 Hans-Peter Nilsson <hpn@cygnus.com>
2618 * config/tc-sh64.c (shmedia_md_assemble): Don't protect
2619 dwarf2_emit_insn call with test on debug_type.
2620 2000-12-19 Hans-Peter Nilsson <hpn@cygnus.com>
2621 * config/tc-sh64.c (sh64_set_contents_type): Make contents-type
2622 CRT_SH5_ISA32 sticky for 64-bit.
2623 2000-12-18 Hans-Peter Nilsson <hpn@cygnus.com>
2624 Generate .crange sections when switching ISA mode or emitting
2625 constants in same section as code.
2626 * config/tc-sh64.c: Reformat structure definitions.
2627 (sh64_end_of_assembly, sh64_mix, sh64_shcompact_const_crange): New
2629 (sh64_set_contents_type): Rename from sh64_init_section. Rewrite
2630 to emit a .cranges descriptor when contents type changes. Only
2631 emit error if changing contents type and -no-mix is in effect.
2632 (sh64_emit_crange, sh64_flush_last_crange, sh64_flag_output,
2633 sh64_flush_pending_output, sh64_tc_cons_fix_new): New functions.
2634 (shmedia_md_end): Set sh64_end_of_assembly. Pass
2635 sh64_flush_last_crange over sections.
2636 When checking main symbol of datalabel symbol, check for
2637 STO_SH5_ISA32, not ISA type of section in definition.
2638 (shmedia_frob_file_before_adjust): Check main symbol for
2639 STO_SH5_ISA32; don't check ISA type of section in definition.
2640 (shmedia_frob_section_type): Adjust for .cranges; set section flag
2641 to SHF_SH5_ISA32_MIXED or SHF_SH5_ISA32 according to whether
2642 .cranges entries have been output.
2643 (shmedia_elf_new_section): Just call md_flush_pending_output.
2644 (shmedia_md_assemble): Do not emit a BFD_RELOC_SH_SHMEDIA_CODE
2645 fix. Do not set tc_segment_info_data.in_code for section. Call
2646 sh64_set_contents_type for SHmedia code.
2647 (s_sh64_mode): Do not call sh64_init_section or set seen_insn to
2648 false. Call md_flush_pending_output.
2649 (sh64_target_format): Check that -no-mix and
2650 -shcompact-const-crange are used in sane combination with other
2652 (shmedia_md_pcrel_from_section): Check type of fix for how to
2654 (sh64_consume_datalabel): Check symbol for having STO_SH5_ISA32,
2655 not ISA type of section in definition.
2656 * config/tc-sh64.h (struct sh64_segment_info_type): Rewrite to
2657 hold contents-type state.
2658 (md_flush_pending_output): Redefine to sh64_flush_pending_output.
2659 (sh64_flush_pending_output): Declare.
2660 (TC_CONS_FIX_NEW): Define to sh64_tc_cons_fix_new.
2661 (sh64_tc_cons_fix_new): Declare.
2662 * config/tc-sh.c (sh_elf_cons) [HAVE_SH64]: Unset seen_insn and
2663 call sh64_flag_output.
2664 (md_assemble) [HAVE_SH64]: Do not emit BFD_RELOC_SH_CODE. Just
2665 call sh64_set_contents_type to mark SHcompact code and set
2667 (md_longopts): New options "-no-mix" and
2668 "-shcompact-const-crange".
2669 (md_parse_option): Handle new options.
2670 (md_show_usage): Add blurb for new options.
2671 (md_number_to_chars) [HAVE_SH64]: Call sh64_flag_output.
2672 2000-12-15 Hans-Peter Nilsson <hpn@cygnus.com>
2673 * config/tc-sh64.c: Delete investigated and obsolete fixme:s.
2674 (sh64_last_insn_frag): New.
2675 (shmedia_md_convert_frag): Use tc_frag_data field of incoming frag
2676 to get frag for insn opcode for generating fixups; do not assume it
2678 (shmedia_build_Mytes): Set sh64_last_insn_frag after growing frag
2680 * config/tc-sh64.h (ELF_TC_SPECIAL_SECTIONS): Define for .cranges
2682 (TC_FRAG_TYPE): Define as fragS *.
2683 (TC_FRAG_INIT): Define to set tc_frag_data to sh64_last_insn_frag.
2684 (sh64_last_insn_frag): Declare.
2685 (sh64_consume_datalabel): Fix typo; check for seginfo != NULL,
2686 not == NULL before dereferencing.
2687 2000-12-12 Hans-Peter Nilsson <hpn@cygnus.com>
2688 Get rid of BFD section flag and EF_SH64_ABI64.
2689 * config/tc-sh64.c (shmedia_frob_section_type): Use
2690 elf_section_data (sec)->tdata, not a specific BFD section flag, to
2691 communicate the section as containing SHmedia code. Describe why.
2692 * config/tc-sh.c (sh_elf_final_processing): Tweak comment. Set
2693 EF_SH64 regardless of ABI.
2694 * config/tc-sh64.c (shmedia_md_apply_fix): Decapitalize "invalid"
2695 in error message. Handle resolved expressions for
2696 BFD_RELOC_SH_IMMS10, BFD_RELOC_SH_IMMS10BY2,
2697 BFD_RELOC_SH_IMMS10BY4 and BFD_RELOC_64.
2698 (shmedia_check_limits): Handle BFD_RELOC_64.
2699 (sh64_adjust_symtab): Do not decrement the GAS symbol value for
2700 a STO_SH5_ISA32 symbol, only the BFD value.
2701 2000-12-11 Ben Elliston <bje@redhat.com>
2702 * config/tc-sh64.c: Call dwarf2_emit_insn, not the defunct
2703 dwarf2_generate_asm_lineno.
2704 2000-12-11 Hans-Peter Nilsson <hpn@cygnus.com>
2705 Handle PC-relative MOVI expansions with assembler relaxation.
2706 Generate PC-relative relocs from 16-bit PC-relative expressions.
2707 * config/tc-sh64.c (SHMEDIA_MD_PCREL_FROM_FIX): Break out from...
2708 (shmedia_md_pcrel_from_section): ...here.
2709 (shmedia_md_apply_fix): Handle fixups for 16-bit operands that has
2710 turned completely resolved. Adjust relocation type for 16-bit
2711 immediate operands that has turned PC-relative. Adjust back for
2712 MD_PCREL_FROM_SECTION being applied twice.
2713 (shmedia_md_convert_frag): Always emit reloc for expression with
2714 global or weak symbol. Handle relaxation result for PC-relative
2716 (shmedia_md_estimate_size_before_relax): An expression with a weak
2717 or global symbol can not be relaxed. Break out tests for
2718 relaxable symbol into variable sym_relaxable.
2719 <cases C (MOVI_IMM_64, UNDEF_MOVI) and C (MOVI_IMM_32,
2720 UNDEF_MOVI)>: Break out any PC-relative expression and change
2722 (shmedia_build_Mytes): CSE &operands->operands[j] into variable
2724 <case A_IMMS16>: Fix typo for initial minor relaxation type of
2725 MOVI expansion. If X_op_symbol of the immediate expression is
2726 set, make an expression symbol for the argument to frag_var.
2727 * config/tc-sh.c (MOVI_IMM_32_PCREL, MOVI_IMM_64_PCREL): New
2729 (END): Adjust for new relaxations.
2730 (md_relax_table): Add entries for new relaxations.
2731 2000-12-07 Ben Elliston <bje@redhat.com>
2732 * config/tc-sh64.c (shmedia_parse_reg): Initialize variable len.
2733 2000-12-07 Hans-Peter Nilsson <hpn@cygnus.com>
2734 * config/tc-sh64.c (shmedia_md_convert_frag): Correct all MOVI and
2735 SHORI operand offsets in PT/PTA/PTB expansions.
2736 2000-12-05 Hans-Peter Nilsson <hpn@cygnus.com>
2737 Implement DataLabel semantics.
2738 * config/tc-sh.c (sh_frob_file) [HAVE_SH64]: Call
2739 shmedia_frob_file_before_adjust.
2740 * config/tc-sh64.c [! OBJ_ELF]: Emit #error.
2741 (DATALABEL_SUFFIX): Define.
2742 (shmedia_md_end) <before adjusting STO_SH5_ISA32 symbols>: Walk
2743 symbol list to update "datalabel" symbols to their main symbol
2745 (shmedia_frob_file_before_adjust): New.
2746 (sh64_adjust_symtab): For remaining datalabel symbols, set to
2747 undefined and set STT_DATALABEL.
2748 (sh64_frob_label): Initialize TC symbol field.
2749 (sh64_consume_datalabel): Actually implement semantics. New
2750 parameter operandf, call it instead of expression.
2751 (sh64_exclude_symbol): New.
2752 * config/tc-sh64.h (md_parse_name): Pass on the function operand
2753 to sh64_consume_datalabel.
2754 (tc_symbol_new_hook): Define to tc_frob_symbol.
2755 (TC_SYMFIELD_TYPE): Define to symbolS *.
2756 (tc_frob_symbol): Define to call sh64_exclude_symbol.
2757 2000-12-01 Hans-Peter Nilsson <hpn@cygnus.com>
2758 * config/tc-sh64.c (shmedia_init_reloc): Tweak comment for default
2760 (shmedia_md_assemble): Call dwarf2_generate_asm_lineno if
2761 generating dwarf2 debug information.
2762 2000-11-30 Hans-Peter Nilsson <hpn@cygnus.com>
2763 * config/tc-sh64.c (sh64_target_format): Use elf64-sh64l and
2764 elf64-sh64 for the 64-bit ABI.
2765 * config/tc-sh.c (md_show_usage): Tweak usage output for -abi=*
2767 2000-11-29 Hans-Peter Nilsson <hpn@cygnus.com>
2768 * config/tc-sh.c: Remove conditionalizing on HAVE_SH64 for
2770 2000-11-27 Hans-Peter Nilsson <hpn@cygnus.com>
2771 * config/tc-sh64.c: Tweak comments, formatting and error messages.
2772 (enum sh64_abi_values): New type.
2773 (enum sh64_isa_values): New type.
2774 (sh64_isa_mode): Replace shmedia_mode. All referers changed.
2775 (seen_shcompact_mode, seen_shmedia_mode): Delete.
2776 (sh64_abi): Replace shmedia_64.
2777 (shmedia_md_convert_frag) <C (MOVI_IMM_64, MOVI_64),
2778 C (MOVI_IMM_32, MOVI_32)>: Correct register number handling.
2779 (s_sh64_mode): Check validity for this target.
2780 (sh64_target_format): Initialize defaults for ISA and ABI.
2781 Fallback to old object format if no SH64 ISA or ABI has been
2783 * config/tc-sh.c (md_parse_option): Check combinations for errors.
2784 (sh_elf_final_processing): Change to have EF_SH64_ABI64 for 64-bit
2785 ABI and EF_SH64 for 32-bit ABI, if SH64 options are specified.
2786 * config/tc-sh64.h: Fix typo in comment.
2787 2000-11-25 Hans-Peter Nilsson <hpn@cygnus.com>
2788 * config/tc-sh64.c (shmedia_md_estimate_size_before_relax)
2789 <PT fixups for absolute values>: Size will be longest, not
2791 (shmedia_md_convert_frag): Disable PTB-warning machinery. Correct
2792 all MOVI and SHORI operand offsets in PT/PTA/PTB expansions.
2793 * config/tc-sh.c (parse_reg) [HAVE_SH64]: Add local variables l0
2794 and l1 to hold lowercase of two first characters. Change all
2795 remaining TO_LOWER to tolower.
2796 * config/tc-sh64.c (TO_LOWER): Delete.
2797 (shmedia_find_cooked_opcode): Use tolower, not TO_LOWER.
2798 (md_parse_name): Define.
2799 (sh64_consume_datalabel): Declare.
2800 (DOLLAR_DOT): Define.
2801 * config/tc-sh64.c (shmedia_parse_exp): New.
2802 (sh64_consume_datalabel): New; just ignoring datalabel semantics.
2803 (shmedia_parse_reg): Remove const from src
2805 (shmedia_get_operands): Ditto for args parameter and ptr variable.
2806 (shmedia_md_assemble): Ditto for op_end variable.
2807 (shmedia_get_operand): Ditto for ptr parameter and src variable.
2808 Use shmedia_parse_exp, not parse_exp.
2809 * config/tc-sh64.c (shmedia_parse_reg): Add shmedia_arg_type
2810 parameter. All callers changed.
2811 (shmedia_get_operand): Add shmedia_arg_type parameter. All
2813 (shmedia_parse_reg): Put first two character in local variables.
2814 Use tolower, not TO_LOWER. If no register is found and argtype
2815 indicates a control register, scan shmedia_creg_table
2817 2000-11-24 Hans-Peter Nilsson <hpn@cygnus.com>
2818 * Makefile.am (CPU_TYPES): Add sh64.
2819 (TARGET_CPU_CFILES): Add config/tc-sh64.c.
2820 (TARGET_CPU_HFILES): Add config/tc-sh64.h.
2821 Regenerate dependencies.
2822 * Makefile.in: Regenerate.
2823 * configure.in: Add support for sh64-*-elf*.
2824 * configure: Regenerate.
2825 * config/tc-sh64.h: New.
2826 * config/tc-sh64.c: New.
2827 * config/tc-sh.c (md_pseudo_table) [HAVE_SH64]: New pseudos
2828 .mode, .isa and .uaquad.
2829 [HAVE_SH64] (SH64PCREL16_32, SH64PCREL16_64, SH64PCREL16PT_32,
2830 SH64PCREL16PT_64, MOVI_IMM_32, MOVI_IMM_64): Define.
2831 [HAVE_SH64] (END): Define as 10.
2832 [HAVE_SH64] (UNDEF_SH64PCREL, SH64PCREL16, SH64PCREL32,
2833 SH64PCREL48, SH64PCREL64, UNDEF_MOVI, MOVI_16, MOVI_32, MOVI_48,
2835 [HAVE_SH64] (SH64PCREL16_F, SH64PCREL16_M, SH64PCREL16_LENGTH,
2836 SH64PCREL32_F, SH64PCREL32_M, SH64PCREL32_LENGTH, SH64PCREL48_F,
2837 SH64PCREL48_M, SH64PCREL48_LENGTH, SH64PCREL64_LENGTH,
2838 MOVI_16_LENGTH, MOVI_32_LENGTH, MOVI_48_LENGTH, MOVI_64_LENGTH):
2840 (md_relax_table) [HAVE_SH64]: Provide relaxations for SHmedia.
2841 (md_begin) [HAVE_SH64]: Call shmedia_md_begin.
2842 (parse_reg) [HAVE_SH64]: Parse register names case-insensitive.
2843 (md_assemble) [HAVE_SH64]: Call shmedia_md_assemble if assembling
2844 SHmedia instructions. Handle state-change after switching to
2846 (md_longopts) [HAVE_SH64]: New options --isa=* and --abi=*.
2847 (md_parse_option) [HAVE_SH64]: Parse new options.
2848 (md_show_usage) [HAVE_SH64]: Show usage of new options.
2849 (md_convert_frag) [HAVE_SH64] <default>: Call
2850 shmedia_md_convert_frag instead of abort.
2851 (sh_force_relocation) [HAVE_SH64]: Also force relocation for
2852 BFD_RELOC_SH_SHMEDIA_CODE.
2853 (sh_elf_final_processing) [HAVE_SH64]: Set flags identifying
2854 SHcompact or SHmedia code.
2855 (md_apply_fix) [HAVE_SH64] <default>: Return result from calling
2856 shmedia_md_apply_fix instead of abort.
2857 (md_estimate_size_before_relax) [HAVE_SH64] <default>: Return
2858 result from calling shmedia_md_estimate_size_before_relax instead
2860 (sh_do_align) [HAVE_SH64]: If shmedia_mode, let shmedia_do_align
2862 (tc_gen_reloc) [HAVE_SH64]: For unrecognized relocs, call
2863 shmedia_init_reloc and do nothing more if it returns non-zero.
2864 (sh_finalize) [HAVE_SH64]: Call shmedia_md_end.
2865 * po/POTFILES.in: Regenerate.
2866 * po/gas.pot: Regenerate.
2868 2002-02-06 Alexandre Oliva <aoliva@redhat.com>
2870 * config/tc-sh.c (parse_at): Install the correct version of
2873 * config/tc-sh.c (md_apply_fix3) <BFD_RELOC_32_PLT_PCREL>: Don't
2874 assume fixP->fx_subsy is non-NULL.
2876 2002-02-04 Alexandre Oliva <aoliva@redhat.com>
2878 * config/tc-sh.c (parse_at): Set arg type of @(expr,pc) to
2879 A_DISP_PC_ABS, and adjust it by -4.
2880 (get_specific): Accept A_DISP_PC_ABS where A_DISP_PC is
2882 (build_Mytes): Mark PCRELIMM fix-ups as pc-relative only if
2883 the operand type is not A_DISP_PC_ABS.
2885 2002-02-04 Hans-Peter Nilsson <hp@bitrange.com>
2887 * config/tc-mmix.c (tc_gen_reloc): Don't try and take the value of
2888 common and weak symbols. Handle common and weak symbols as
2889 undefined symbols with regards to GREG handling and merging.
2890 (mmix_frob_file): Ditto.
2892 2002-02-02 Jason Thorpe <thorpej@wasabisystems.com>
2894 * configure.in (hppa-*-netbsd*): New target.
2895 * configure: Regenerate.
2896 * config/tc-hppa.h: Also define WARN_COMMENTS if TE_NetBSD.
2898 2002-02-02 Alan Modra <amodra@bigpond.net.au>
2900 * config/tc-v850.c: Add missing prototypes amd use old-style
2901 function definitions.
2902 (AREA_ZDA, AREA_SDA, AREA_TDA): Delete.
2903 (sdata_section tdata_section, zdata_section, sbss_section,
2904 tbss_section, zbss_section, rosdata_section, rozdata_section,
2905 scommon_section, tcommon_section, zcommon_section,
2906 call_table_data_section, call_table_text_section): Delete.
2907 (v850_sdata, v850_tdata, v850_zdata, v850_sbss, v850_tbss,
2908 v850_zbss, v850_bss, v850_rosdata, v850_rozdata,
2909 v850_call_table_data, v850_call_table_text): Delete.
2910 (struct v850_seg_entry): New.
2911 (v850_seg_table): New.
2912 (SDATA_SECTION TDATA_SECTION, ZDATA_SECTION, SBSS_SECTION,
2913 TBSS_SECTION, ZBSS_SECTION, BSS_SECTION, ROSDATA_SECTION,
2914 ROZDATA_SECTION, SCOMMON_SECTION, TCOMMON_SECTION, ZCOMMON_SECTION,
2915 CALL_TABLE_DATA_SECTION, CALL_TABLE_TEXT_SECTION): Define.
2918 (v850_comm): Use do_v850_seg and v850_seg_table. Simplify
2919 recording of alignment.
2920 (md_pseudo_table): Use v850_seg.
2921 (md_begin): Don't init .call_table_data and .call_table_text here.
2922 Set v850_seg_table bss entry.
2924 2002-02-01 Hans-Peter Nilsson <hp@bitrange.com>
2926 Support on-demand global register allocation by passing on
2927 base-plus-offset relocs to the linker.
2928 * config/tc-mmix.c: Tweak and fix typos in comments.
2929 (allocate_undefined_gregs_in_linker): New variable.
2930 (OPTION_LINKER_ALLOCATED_GREGS): New option macro.
2931 (md_longopts): Add --linker-allocated-gregs.
2932 (md_parse_option) <case 'x'>: Imply --linker-allocated-gregs.
2933 <case OPTION_LINKER_ALLOCATED_GREGS>: New.
2934 (md_show_usage): Update text for -x. Add text for
2935 --linker-allocated-gregs.
2936 (tc_gen_reloc): Derive default value for addend from val and
2937 baddsy. Use addsec and bfd_is_abs_section in more places. Don't
2938 emit error for BFD_RELOC_MMIX_BASE_PLUS_OFFSET without suitable
2939 GREG if allocate_undefined_gregs_in_linker.
2940 * doc/as.texinfo (Overview) <Target MMIX options>: Add
2941 --linker-allocated-gregs.
2942 * doc/c-mmix.texi (MMIX-Opts): Add blurb about
2943 --linker-allocated-gregs. Mention that it's implied by -x.
2944 (MMIX-Pseudos) <GREG>: Mention when and how a GREG can be omitted.
2945 (MMIX-mmixal): Clarify dated comparison and location of MMIXware.
2947 * config/tc-mmix.h (md_parse_name): Use ISUPPER, not isupper.
2949 2002-02-01 Alan Modra <amodra@bigpond.net.au>
2951 * Makefile.am: Run "make dep-am"
2952 * Makefile.in: Regenerate.
2954 2002-01-31 Ivan Guzvinec <ivang@opencores.org>
2956 * config/tc-or32.c: New file.
2957 * config/tc-or32.h: New file.
2958 * configure.in: Add support for or32 targets.
2959 * configure: Regenerate.
2960 * config/obj-coff.c: Add support for or32 targets.
2961 * config/obj-coff.h: Add support for or32 targets.
2962 * Makefile.am: Add support for or32 targets.
2963 * Makefile.in: Regenerate.
2964 * NEWS: Mention support for OpenRISC.
2965 * doc/Makefile.in: Regenerate.
2966 * po/POTFILES.in: Regenerate.
2967 * po/gas.pot: Regenerate.
2969 2002-01-30 Richard Sandiford <rsandifo@redhat.com>
2971 * config/tc-sh.c (parse_reg): Fix end-of-word check for is, ix, iy
2974 2002-01-29 Chris Demetriou <cgd@broadcom.com>
2976 * config/tc-mips.c (tc_gen_reloc): Arrange for
2977 BFD_RELOC_PCREL_HI16_S relocations to be output relative to
2978 their LO16 parts, even for ELF.
2980 2002-01-29 Daniel Jacobowitz <drow@mvista.com>
2982 * config/tc-i386.c: Protect definitions of true and false
2985 2002-01-28 Jakub Jelinek <jakub@redhat.com>
2987 * config/obj-elf.c (elf_frob_file_before_adjust): Remove symbols
2988 made because of .weak, if they are neither defined nor used in any
2991 2002-01-27 Daniel Jacobowitz <drow@mvista.com>
2993 * configure: Regenerated.
2995 2002-01-26 Hans-Peter Nilsson <hp@bitrange.com>
2997 * doc/Makefile.am (install): Depend on install-info.
2998 * doc/Makefile.in: Regenerate.
3000 2002-01-26 Nick Clifton <nick@redhat.com>
3002 * po/fr.po: Updated version
3004 2002-01-24 Kazu Hirata <kazu@hxi.com>
3006 * config/tc-h8300.c (check_operand): Don't print a warning
3007 when a valid 24-bit address is given to a 16-bit address
3010 2002-01-24 Alexandre Oliva <aoliva@redhat.com>
3012 * config/tc-sh.c (sh_elf_suffix): Removed.
3013 (sh_PIC_related_p, sh_check_fixup, sh_cons_fix_new,
3014 sh_end_of_match, sh_parse_name): New functions.
3015 (sh_elf_cons): Simplify.
3016 (parse_exp): Reject misplaced PIC operands.
3017 (md_undefined_symbol): Simplify.
3018 (sh_fix_adjustable): Let @GOTOFF be adjusted.
3019 (md_apply_fix3): Write @PLT and @GOTOFF addends in place.
3020 (tc_gen_reloc): Move fixp subsy absolute value into addnumber.
3021 Complain if subsy remains at the end.
3022 * config/tc-sh.h (sh_parse_name, sh_cons_fix_new): Declare.
3023 (md_parse_name, TC_CONS_FIX_NEW, O_PIC_reloc): Define.
3025 2002-01-22 Alexandre Oliva <aoliva@redhat.com>
3027 * config/tc-mn10300.c (xr_registers): Move `pc'...
3028 (other_registers): ... here.
3030 2002-01-22 Alan Modra <amodra@bigpond.net.au>
3032 * Makefile.am: Run "make dep-am".
3033 * Makefile.in: Regenerate.
3034 * gas/po/POTFILES.in: Regenerate.
3036 2002-01-21 DJ Delorie <dj@redhat.com>
3038 * config/obj-coff.c (obj_coff_init_stab_section): Make the
3039 stabstr_name allocation permanent, as it will be referenced from
3042 2002-01-21 Jason Thorpe <thorpej@wasabisystems.com>
3044 * configure.in (ia64-*-netbsd*): New target.
3045 * configure: Regenerate.
3047 2002-01-21 Hans-Peter Nilsson <hp@bitrange.com>
3049 * doc/as.texinfo (Overview) <Target ARM options>: Add missing {}
3051 <Detailed description, ARM options>: Ditto.
3052 * doc/c-arm.texi (ARM Options): Ditto.
3054 2002-01-18 Richard Earnshaw <rearnsha@arm.com>
3056 * tc-arm.c (do_xsc_mia, do_xsc_mar, do_xsc_mra): Renamed from
3057 do_mia, do_mar and do_mra respectively.
3058 (do_mav_*): Renamed from do_c_*.
3059 (mav_reg_required_here, mav_parse_offset): Renamed from
3060 cirrus_reg_required_here and cirrus_parse_offset respectively.
3061 (MAV_MODE?): Renamed from CIRRUS_MODE?.
3063 2002-01-18 Richard Earnshaw <rearnsha@arm.com>
3064 Keith Walker <keith.walker@arm.com>
3066 * tc-arm.c (ARM_EXT_V5J, ARM_ARCH_V5TEJ): Define.
3067 (insns): Add pattern for bxj instruction.
3068 (do_bxj): New function.
3069 (arm_cpus): Add arm926ej.
3070 (arm_archs): Add armv5tej.
3072 2002-01-18 Richard Earnshaw <rearnsha@arm.com>
3074 * doc/c-arm.texi: Add new fpe options to list of supported flags.
3076 2002-01-19 Keith Walker <keith.walker@arm.com>
3078 * tc-arm.c (arm_fpus): Add fpe2 and fpe3.
3080 2002-01-18 Richard Earnshaw <rearnsha@arm.com>
3082 * NEWS: Mention new ARM command-line options and VFP support.
3084 * config/tc-arm.c (ARM_CEXT_XSCALE): Replaces ARM_EXT_XSCALE. All
3086 (ARM_CEXT_MAVERICK): Similarly.
3087 (ARM_ANY): Now means any core instruction.
3088 (CPU_DEFAULT): Default to ARM_ANY.
3089 (uses_apcs_26, atcps, support_interwork, uses_apcs_float)
3090 (pic_code): Declare for all object types. Make type int.
3091 (legacy_cpu, legacy_fpu, mcpu_cpu_opt, mcpu_fpu_opt, march_cpu_opt)
3092 (march_fpu_opt, mfpu_opt): Declare.
3093 (md_longopts): Tidy up conditional definitions.
3094 (arm_opts, arm_cpus, arm_archs, arm_fpus, arm_extensions)
3095 (arm_long_opts): New tables.
3096 (arm_parse_cpu, arm_parse_arch, arm_parse_fpu): New functions.
3097 (arm_parse_extension): New function.
3098 (md_parse_option): Rewrite using new table-driven system.
3099 (md_show_usage): Use new table-driven system.
3100 (md_begin): Calculate cpu_variant from command line option data.
3101 * doc/as.texinfo (ARM ISA options): Docuement new ARM-specific
3102 command-line options.
3103 * doc/c-arm.texi: Likewise.
3105 2002-01-18 Andreas Jaeger <aj@suse.de>
3107 * as.c (parse_args): Update year.
3109 2002-01-17 Timothy Wall <twall@alum.mit.edu>
3111 * config/tc-tic54x.c (encode_address): Add a more informative
3112 warning about incorrect syntax.
3114 2002-01-17 Nick Clifton <nickc@cambridge.redhat.com>
3116 * po/gas.pot: Regenerate.
3118 2002-01-17 Nick Clifton <nickc@cambridge.redhat.com>
3120 * macro.c (get_any_string): Add no-c-format comment to prevent
3121 confusion when translating string.
3122 * gasp.c (get_any_string): Add no-c-format comment to prevent
3123 confusion when translating string.
3125 2002-01-16 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
3126 Johannes Stezenbach <js@convergence.de>
3127 * config/tc-mips.c (percent_op): Ensure longer percent_op's are
3128 matched before the shorter ones.
3129 (my_getSmallParser): Fix handling of nested parentheses in
3130 percent_op's. Code cleanup.
3131 (my_getPercentOp): New function, code from my_getSmallParser.
3132 (my_getSmallExpression): Fix handling of closing parentheses.
3133 Code cleanup. Better comments.
3135 2002-01-16 Nick Clifton <nickc@redhat.com>
3137 po/tr.po: New file: Turkish translation.
3138 configure.in (LINGUAS): Add "tr".
3139 configure: Regenerate.
3141 2002-01-15 Richard Earnshaw <rearnsha@arm.com>
3143 Support for VFP instructions
3144 * tc-arm.c (CP_WB_OK, CP_NO_WB): New defines.
3145 (cp_address_required_here): New argument wb_ok. When false, do not
3146 accept write-back forms of addressing. Change all callers.
3147 (FPU_VFP_EXT_NONE, FPU_VFP_EXT_V1xD, FPU_VFP_VFP_V1)
3148 (FPU_VFP_EXT_V2): Define.
3149 (FPU_ARCH_VFP, FPU_ARCH_VFP_V1xD, FPU_ARCH_VFP_V1, FPU_ARCH_VFP_V2):
3150 Define in terms of above.
3151 (vfp_dp_reg_pos, vfp_sp_reg_pos, vfp_ldstm_type): New enums.
3152 (vfp_reg): New struct.
3153 (vfp_regs): New array of registers.
3154 (insns): Add VFP instructions.
3155 (sn_table): New array of VFP single-precision register names.
3156 (dn_table): New array of VFP double-precision register names.
3157 (all_reg_maps): Add the new register tables.
3158 (arm_reg_type): Add new values for above. Increase RET_TYPE_MAX.
3159 (vfp_sp_reg_required_here, vfp_dp_reg_required_here, do_vfp_sp_monadic)
3160 (do_vfp_dp_monadic, do_vfp_sp_dyadic, do_vfp_dp_dyadic)
3161 (do_vfp_reg_from_sp, do_vfp_sp_reg2, do_vfp_sp_from_reg)
3162 (do_vfp_reg_from_dp, do_vfp_reg2_from_dp, do_vfp_dp_from_reg)
3163 (do_vfp_dp_from_reg2, vfp_psr_parse, vfp_psr_required_here)
3164 (do_vfp_reg_from_ctrl, do_vfp_ctrl_from_reg, do_vfp_sp_ldst)
3165 (do_vfp_dp_ldst, vfp_sp_reg_list, vfp_dp_reg_list, vfp_sp_ldstm)
3166 (vfp_dp_ldstm, do_vfp_sp_ldstmia, do_vfp_sp_ldstmdb, do_vfp_ldstmia)
3167 (do_vfp_dp_ldstmdb, do_vfp_xp_ldstmia, do_vfp_xp_ldstmdb)
3168 (do_vfp_sp_compare_z, do_vfp_dp_compare_z, do_vfp_dp_sp_cvt)
3169 (do_vfp_sp_dp_cvt): New functions.
3170 (md_begin): Set soft-float flag for appropriate VFP work.
3171 (md_atof): Handle VFP-format doubles.
3172 (md_parse_option): Handle VFP command-line options.
3173 (md_show_usage): Display VFP command-line options.
3175 2002-01-15 Richard Earnshaw <rearnsha@arm.com>
3177 * tc-arm.c (md_parse_option): Tidy up setting of cpu_variant for
3178 various command line options.
3180 2002-01-15 Nick Clifton <nickc@cambridge.redhat.com>
3182 * config/tc-xstormy16.c: (xstormy16_fix_adjustable): Do not fix
3184 (xstormy16_md_apply_fix3): Do not return a value.
3186 2002-01-14 Richard Earnshaw <rearnsha@arm.com>
3188 * tc-arm.c (md_longopts): On targets that aren't bi-endian, support
3189 the -EL/-EB option that matches the target's endianness.
3190 (md_parse_option): Likewise.
3192 2002-01-14 Richard Earnshaw <rearnsha@arm.com>
3194 * tc-arm.c (md_longopts): Fix misplaced #endif -- the -oabi option
3195 is not dependent on ARM_BI_ENDIAN.
3197 2002-01-14 Richard Earnshaw <rearnsha@arm.com>
3199 * tc-arm.c (all error messages): Normalize capitalization of messages.
3201 * tc-arm.h (md_operand): Delete define.
3202 * tc-arm.c (in_my_get_expression): New static variable.
3203 (my_get_expression): Set and clear it.
3204 (md_operand): New function. If called from my_get_expression
3205 put the error in inst.error.
3206 (output_inst): Now takes argument of instruction being assembled.
3207 Print it out with any error message.
3208 (do_ldst, do_ldstv4, thumb_load_store): Fault attempt to use a store
3210 (end_of_line): Don't update inst.error if it is already set.
3212 2002-01-11 Richard Earnshaw <rearnsha@arm.com>
3214 * tc-arm.c (do_ldst): Fix handling an immediate expression pseudo
3215 op that can be translated into a mvn instruction.
3217 2002-01-11 Steve Ellcey <sje@cup.hp.com>
3219 * gas/config/tc-ia64.h (MD_FLAGS_DEFAULT): New Macro for
3220 setting default md.flags.
3221 (SHT_INIT_ARRAY): New elf special section used by HP-UX.
3222 (SHT_FINI_ARRAY): New elf special section used by HP-UX.
3223 * gas/config/tc-ia64.c (setup_unwind_header): Add support
3224 for 32 bit unwind info blocks.
3225 (generate_unwind_image): Add support for different types
3226 of unwind images (32 bits and/or big-endian).
3227 (ia64_init): Use MD_FLAGS_DEFAULT to set md.flags.
3228 (ia64_target_format): Add support for hpux target formats.
3229 (ia64_gen_real_reloc_type): Add support for FUNC_IPLT_RELOC.
3230 (ia64_elf_section_type): Add support for SHT_INIT_ARRAY and
3231 SHT_FINI_ARRAY elf section types.
3233 2002-01-10 Richard Earnshaw <rearnsha@arm.com>
3235 * tc-arm.c (struct reg_entry): Move before prototypes.
3236 (int_register, cp_register, fp_register): Delete.
3237 (reg_table): Delete. Replaced with ...
3238 (rn_table, cp_table, cn_table, fn_table, mav_mvf_table)
3239 (mav_mvd_table, mav_mvfx_table, mav_mvdx_table, mav_mvax_table)
3240 (mav_dspsc_table): ... one table per register set.
3241 (arm_reg_hsh): Delete.
3242 (struct reg_map): New structure.
3243 (all_reg_maps): New array.
3244 (enum arm_reg_type): New enums.
3245 (build_reg_hsh): New function.
3246 (insert_reg_alias): Use hash table passed by caller. Adjust all
3248 (create_register_alias): New function, split out from ...
3249 (md_assemble): ... here.
3250 (md_begin): Build new register hash tables.
3251 (arm_reg_parse): New argument for the hash table to search. Adjust all
3253 (arm_reg_parse_any): New function.
3254 (co_proc_number): Look up the processor number in the processor hash
3256 (cirrus_regtype): Delete.
3257 (cirrus_register, cirrus_mvf_register, cirrus_mvd_register)
3258 (cirrus_mvfx_register, cirrus_mvdx_register, cirrus_mvax_register)
3259 (ARM_EXT_MAVERICKsc_register): Delete.
3260 (do_c_binops_1, do_c_binops_2, do_c_binops_3): Delete.
3261 (do_c_binops_1[a-o], do_c_binops_2[a-c], do_c_binops_3[a-d]): New
3263 (do_c_triple_4, do_c_triple_5): Delete.
3264 (do_c_triple_4[ab], do_c_triple_5[a-h]): New functions.
3265 (do_c_quad_6): Delete.
3266 (do_c_quad_6[ab]): New functions.
3267 (do_c_binops, do_c_triple, do_c_quad, do_c_shift, do_c_ldst): Rework
3268 arguments to use new register parsing methods.
3269 (cirrus_reg_required_here): Likewise.
3270 (insns): Reclassify cirrus maverick worker functions.
3271 (cirrus_valid_reg): Delete.
3273 2002-01-07 Jason Thorpe <thorpej@wasabisystems.com>
3275 * configure.in (sh*le): Set cpu_type=sh and endian=little.
3276 (sh*-*-netbsdelf*): New target.
3277 * configure: Regenerate.
3278 * tc-sh.h: Update copyright years.
3279 (TARGET_FORMAT): Add version for TE_NetBSD.
3281 2002-01-07 Nick Clifton <nickc@cambridge.redhat.com>
3283 * read.c (emit_expr): Do not allow 'size' or 'nbytes' to go
3286 2002-01-06 Alan Modra <amodra@bigpond.net.au>
3288 * config/tc-m68k.h (md_prepare_relax_scan): Rewrite.
3289 * config/tc-m68k.c (md_relax_table): Add md_prepare_relax_scan comment.
3291 2002-01-05 Daniel Jacobowitz <drow@mvista.com>
3293 * tc-mips.c (mips_cprestore_valid): New flag.
3294 (mips_frame_reg_valid): New flag.
3295 (macro) [M_JAL_2]: Check both flags.
3296 [M_JAL_A]: Likewise.
3297 (s_cprestore): Set mips_cprestore_valid.
3298 (tc_get_register): If setting mips_frame_reg, set
3299 mips_frame_reg_valid and clear mips_cprestore_valid.
3300 (s_mips_ent): Clear both flags.
3301 (s_mips_end): Clear both flags.
3303 2002-01-05 Alan Modra <amodra@bigpond.net.au>
3305 * write.c (write_object_file): Make use of bfd_section_list_remove.
3306 * config/obj-ecoff.c (ecoff_frob_file): Likewise.
3307 * config/tc-mmix.c (mmix_frob_file): Likewise.
3309 2002-01-04 Jason Thorpe <thorpej@wasabisystems.com>
3311 * configure.in (i386-*-netbsdelf*): Collapse target into...
3312 (i386-*-netbsd*): ...this. Add support for x86-64.
3313 * configure: Regenerated.
3315 2002-01-03 matthew green <mrg@redhat.com>
3317 * config/tc-ppc.c (md_parse_option): BookE is not Motorola specific.
3319 2002-01-02 Nick Clifton <nickc@cambridge.redhat.com>
3321 * config/tc-m68k.c (md_estimate_size_before_relax): Test for a
3324 For older changes see ChangeLog-0001
3330 version-control: never