1 2002-05-31 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
3 * config/tc-mips.c (OPTION_NO_M7000_HILO_FIX): Rename to
4 OPTION_MNO_7000_HILO_FIX. Add alternate "mno-fix7000"
5 command line switch conforming to gcc conventions.
6 * doc/c-mips.texi: Document -mno-fix7000 instead of no-mfix-7000.
8 2002-05-31 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
10 * config/tc-mips.c (macro_build_jalr): New Function.
11 (md_begin): NewABI uses big GOTs.
12 (macro_build): Recognize BFD_RELOC_MIPS_GOT_DISP,
13 BFD_RELOC_MIPS_GOT_PAGE, BFD_RELOC_MIPS_GOT_OFST as valid.
14 (load_address): Add some NewABI PIC support.
16 (md_apply_fix): Special handling for BFD_RELOC_MIPS_JALR.
17 (tc_gen_reloc): Don't encode NewABI vtables in REL relocations.
19 2002-05-31 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
21 * config/tc-mips.c (load_address): Use mips_gp_register instead
22 of hardcoded value. Remove dbl parameter, use HAVE_32BIT_ADDRESSES
24 (macro): Use mips_gp_register instead of hardcoded value.
26 2002-05-30 Richard Henderson <rth@redhat.com>
28 * expr.h (operatorT): Add O_md17..O_md32.
29 * config/tc-alpha.c (O_lituse_tlsgd, O_lituse_tlsldm, O_tlsgd,
30 O_tlsldm, O_gotdtprel, O_dtprelhi, O_dtprello, O_dtprel, O_gottprel,
31 O_tprelhi, O_tprello, O_tprel): New.
32 (USER_RELOC_P, alpha_reloc_op_tag, debug_exp): Include them.
33 (DUMMY_RELOC_LITUSE_TLSGD, DUMMY_RELOC_LITUSE_TLSLDM): New.
34 (LITUSE_TLSGD, LITUSE_TLSLDM): New.
35 (struct alpha_reloc_tag): Add master, saw_tlsgd, saw_tlsld,
36 saw_lu_tlsgd, saw_lu_tlsldm. Make multi_section_p a bit field.
37 (md_apply_fix3): Handle TLS relocations.
38 (alpha_force_relocation, alpha_fix_adjustable): Likewise.
39 (alpha_adjust_symtab_relocs): Sort LITERAL relocs after the
40 associated TLS reloc. Check lituse_tls relocs match up.
41 (emit_insn): Handle TLS relocations.
44 * doc/c-alpha.texi: Add docs for tls relocations.
46 2002-05-30 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
48 * config/tc-mips.c (mips_gprel_offset): New variable.
51 2002-05-30 Diego Novillo <dnovillo@redhat.com>
53 * gas/config/tc-d10v.c (check_resource_conflict): Only check
54 write-write conflicts.
55 (md_assemble): Reformat introductory comment.
56 (parallel_ok): Prevent packing only if the first
57 instruction cannot be packed.
59 2002-05-30 Jason Eckhardt <jle@redhat.com>
60 Tom Rix <trix@redhat.com>
62 * config/tc-d10v.c (build_insn): Check for immediates.
64 2002-05-28 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
66 * config/tc-mips.c: Replace GP in comments by $gp.
67 (mips_big_got): Initialize.
68 (mips_trap): Initialize.
69 (load_address): Use mips_gp_register instead of hardcoded value.
70 Remove dbl parameter, use HAVE_32BIT_ADDRESSES instead.
71 (macro): Use mips_gp_register instead of hardcoded value.
72 (macro2): Change load_address calls.
73 (md_pcrel_from): Comment formatting.
74 (s_cpload): Use mips_gp_register instead of hardcoded value.
75 (s_cprestore): Likewise. Comment formatting.
76 (s_gpword): Fix data type.
77 (s_cpadd): Use mips_gp_register instead of hardcoded value.
78 (nopic_need_relax): Replace GP in comments by $gp.
79 (mips_elf_final_processing): Better comment.
81 2002-05-28 Kuang Hwa Lin <kuang@sbcglobal.net>
83 * configure.in: Add DLX configuraton
84 * Makefile.am: Add DLX configuraton
85 * configure: Regenerate.
86 * Makefile.in: Regenerate.
87 * config/tc-dlx.c: New file.
88 * config/tc-dlx.h: New files.
89 * NEWS: Mention new support.
91 2002-05-27 Nick Clifton <nickc@cambridge.redhat.com>
93 * config/obj-coff.c (write_object_file): Add missing semicolon.
95 2002-05-26 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
97 * config/tc-mips.c (mips_emit_delays): Replace magic constant for RA
98 by the define. Remove superfluous check of mips_opts.mips16.
99 (append_insn): Likewise. Canonicalize variable increments.
100 (macro_build): Canonicalize variable increments.
101 (macro_build_lui): Likewise.
102 (load_register): Likewise.
103 (load_address): Move pointer initialization.
104 (macro): Move pointer to a more local scope. Canonicalize variable
105 increments. Better comments. Replace magic constant for RA by the
107 (macro2): Replace magic constant for RA by the define. Canonicalize
109 (mips_ip): Canonicalize variable increments.
110 (mips16_ip): Replace magic constant for RA by the define.
111 (my_getSmallParser): Canonicalize variable increments/decrements.
112 (my_getPercentOp): Likewise.
113 (my_getSmallExpression): Likewise.
115 (s_mipsset): Likewise.
116 (s_cpsetup): Likewise.
117 (s_insn): Remove superfluous check of mips_opts.mips16.
118 (s_mips_stab): Likewise.
119 (mips_handle_align): Canonicalize variable increments.
120 (s_mips_ent): Likewise.
121 (s_mips_end): Add comment.
123 2002-05-26 Jason Thorpe <thorpej@wasabisystems.com>
125 * configure.in: Use ns32k-*-netbsd* instead of ns32k-pc532-netbsd*.
126 * configure: Regenerate.
128 2002-05-25 Alan Modra <amodra@bigpond.net.au>
130 * Makefile.am (OBJS): Depend on ansidecl.h and fopen-same.h.
131 * Makefile.in: Regenerate.
132 * dep-in.sed: Reorder to match OBJS in Makefile.am.
133 * configure.in (ALL_OBJ_DEPS): Add symcat.h when need_bfd.
134 * configure: Regenerate.
135 * as.h: Use #include "" instead of <> for local header files.
136 * flonum-konst.c: Likewise.
137 * flonum-mult.c: Likewise.
139 * listing.c: Likewise.
140 * config/tc-ia64.h: Likewise.
141 * config/tc-v850.h: Likewise.
143 2002-05-24 TAMURA Kent <kent@netbsd.org>
145 * configure.in: Add a target for i386-netbsdpe.
146 * configure: Regenerate.
148 2002-05-23 Jakub Jelinek <jakub@redhat.com>
150 * config/obj-elf.c (elf_common): Renamed from obj_elf_common.
151 (obj_elf_common): Call elf_common.
152 (obj_elf_tls_common): New function.
153 (elf_pseudo_tab): Support .tls_common.
154 (special_sections): Add .tdata and .tbss.
155 (obj_elf_change_section): Set SEC_THREAD_LOCAL for SHF_TLS
157 (obj_elf_parse_section_letters): Support T in section flags (SHF_TLS).
158 (obj_elf_parse_section_letters): Include T in error message.
159 * config/tc-ppc.c (ppc_section_letter): Likewise.
160 * config/tc-alpha.c (alpha_elf_section_letter): Likewise.
161 (tc_gen_reloc): Handle SEC_THREAD_LOCAL the same way as
163 * config/tc-sparc.c (md_apply_fix3): Likewise.
164 * config/tc-i386.c (tc_i386_fix_adjustable): Add TLS relocs.
165 Define them if not BFD_ASSEMBLER.
166 (lex_got): Support @TLSGD, @TLSLDM, @GOTTPOFF, @TPOFF, @DTPOFF
168 (md_apply_fix3): Add TLS relocs.
169 * config/tc-ia64.c (enum reloc_func): Add FUNC_DTP_MODULE,
170 FUNC_DTP_RELATIVE, FUNC_TP_RELATIVE, FUNC_LT_DTP_MODULE,
171 FUNC_LT_DTP_RELATIVE, FUNC_LT_TP_RELATIVE.
172 (pseudo_func): Support @dtpmod(), @dtprel() and @tprel().
173 (ia64_elf_section_letter): Include T in error message.
174 (md_begin): Support TLS operators.
175 (md_operand): Likewise.
176 (ia64_gen_real_reloc_type): Support TLS relocs.
177 * testsuite/gas/i386/tlspic.s: New file.
178 * testsuite/gas/i386/tlsd.s: New file.
179 * testsuite/gas/i386/tlsnopic.s: New file.
180 * testsuite/gas/i386/tlsd.d: New file.
181 * testsuite/gas/i386/tlsnopic.d: New file.
182 * testsuite/gas/i386/tlspic.d: New file.
183 * testsuite/gas/i386/i386.exp: Add tlsd, tlsnopic and tlspic tests.
184 * testsuite/gas/ia64/tls.s: New file.
185 * testsuite/gas/ia64/tls.d: New file.
186 * testsuite/gas/ia64/ia64.exp: Add tls test.
187 * write.c (adjust_reloc_syms): Don't change symbols in
188 SEC_THREAD_LOCAL sections to STT_SECTION + addend.
190 2002-05-23 Nick Clifton <nickc@cambridge.redhat.com>
192 * config/tc-arm.c (md_apply_fix3): For the Thumb BLX reloc
193 round the relocation up rather than down.
195 2002-05-23 Nick Clifton <nickc@cambridge.redhat.com>
197 * config/obj-coff.c (obj_coff_section): Silently ignore an 'a'
199 * doc/as.texinfo: Document that the COFF version of .section
200 ignores the 'a' flag.
202 2002-05-23 Alan Modra <amodra@bigpond.net.au>
204 * config/tc-alpha.c (assemble_tokens): Protect use of
205 ALPHA_RELOC_TABLE with #ifdef RELOC_OP_P.
207 * write.c (size_seg): Check adjustment to last frag.
208 (SUB_SEGMENT_ALIGN): If HANDLE_ALIGN defined, pad out last frag to
210 * config/obj-coff.c (SUB_SEGMENT_ALIGN): Likewise.
211 * config/obj-ieee.c (SUB_SEGMENT_ALIGN): Likewise.
212 (write_object_file): Invoke md_do_align if available, and use
213 frag_align_code on text sections.
214 * config/obj-vms.h (SUB_SEGMENT_ALIGN): Now two args.
215 * config/tc-m88k.h (SUB_SEGMENT_ALIGN): Likewise.
216 * config/tc-ppc.h (SUB_SEGMENT_ALIGN): Likewise.
217 * config/tc-sh.h (SUB_SEGMENT_ALIGN): Likewise.
218 * config/tc-i386.h (SUB_SEGMENT_ALIGN): Likewise. Define for
221 2002-05-22 H.J. Lu <hjl@gnu.org>
223 * dwarf2dbg.c (dwarf2_directive_loc): Call listing_source_file
226 2002-05-22 Nick Clifton <nickc@cambridge.redhat.com>
228 * config/tc-arm.c (arm_s_section): Enable for COFF builds as well
231 2002-05-22 H.J. Lu <hjl@gnu.org>
233 * dwarf2dbg.c (dwarf2_emit_insn): Emit only one line symbol
234 for one .loc for compiler.
236 2002-05-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
238 * config/tc-mips.c (macro): Relax warning, it's toot strict for
241 2002-05-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
243 * config/tc-mips.c (macro2): Add 64 bit drol, dror macros.
244 Optimize the rotate by zero case.
246 2002-05-21 Nick Clifton <nickc@cambridge.redhat.com>
248 * configure.in: Remove accidental enabling of bfd_gas=yes for
250 * configure: Regenerate.
252 2002-05-18 Kazu Hirata <kazu@cs.umass.edu>
254 * app.c: Fix formatting.
258 * input-file.c: Likewise.
259 * listing.c: Likewise.
262 * symbols.c: Likewise.
264 2002-05-17 Alan Modra <amodra@bigpond.net.au>
266 * doc/internals.texi (md_apply_fix3): Expand.
267 (TC_FORCE_RELOCATION, TC_FORCE_RELOCATION_ABS,
268 TC_FORCE_RELOCATION_LOCAL, TC_FORCE_RELOCATION_SUB_SAME,
269 TC_FORCE_RELOCATION_SUB_ABS, TC_FORCE_RELOCATION_SUB_LOCAL,
270 TC_FIX_ADJUSTABLE): Document.
271 * symbols.c (S_FORCE_RELOC): New function.
272 * symbols.h (S_FORCE_RELOC): Declare. s/CONST/const/.
273 * config/obj-aout.h (S_FORCE_RELOC): Define.
274 * config/obj-bout.h (S_FORCE_RELOC): Define.
275 * config/obj-coff.h (S_FORCE_RELOC): Define. Formatting fixes.
276 * config/obj-ieee.h (S_FORCE_RELOC): Define. Formatting fixes.
277 * config/obj-vms.h (S_FORCE_RELOC): Define.
278 * expr.c (clean_up_expression <O_subtract>): Use S_FORCE_RELOC
279 instead of SEG_NORMAL. Allow diff != 0.
280 (expr <O_subtract>): Use S_FORCE_RELOC instead of SEG_NORMAL.
281 * read.c (pseudo_set): Likewise.
282 * write.c (TC_FORCE_RELOCATION_ABS): Define.
283 (TC_FORCE_RELOCATION_LOCAL): Define, replaces..
284 (TC_RELOC_RTSYM_LOC_FIXUP): ..this. Remove.
285 (TC_FORCE_RELOCATION_SUB_SAME): Define, replaces..
286 (TC_FORCE_RELOCATION_SECTION): ..this. Remove.
287 (TC_FORCE_RELOCATION_SUB_ABS): Define.
288 (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
289 (RELOC_ENUM): Define.
290 (fix_new_internal): Use RELOC_ENUM.
292 (fix_new_exp): Likewise.
293 (adjust_reloc_syms): Split out code setting used_in_reloc to..
294 (mark_reloc_syms): ..here. New function. Use S_FORCE_RELOC.
295 (write_object_file): Call mark_reloc_syms.
296 (write_relocs): Formatting. Avoid symbol loops in
297 RELOC_EXPANSION_POSSIBLE case too. Report bfd_reloc_outofrange
298 errors, and error number in other cases.
299 (fixup_segment): Rewrite.
300 * config/tc-alpha.h (TC_FORCE_RELOCATION_LOCAL): Define, replaces..
301 (TC_RELOC_RTSYM_LOC_FIXUP): Delete.
302 * config/tc-arm.h: Likewise.
303 * config/tc-cris.h: Likewise.
304 * config/tc-i386.h: Likewise.
305 * config/tc-i960.h: Likewise.
306 * config/tc-ia64.h: Likewise.
307 * config/tc-m68k.h: Likewise.
308 * config/tc-ppc.h: Likewise.
309 * config/tc-s390.h: Likewise.
310 * config/tc-sh.h: Likewise.
311 * config/tc-sh64.h: Likewise.
312 * config/tc-sparc.h: Likewise.
313 * config/tc-hppa.h (TC_FORCE_RELOCATION_SUB_SAME): Define, replaces..
314 (TC_FORCE_RELOCATION_SECTION): Delete.
315 * config/tc-i386.c (tc_i386_force_relocation): Remove.
316 (BFD_RELOC_8, BFD_RELOC_16): Define for non BFD_ASSEMBLER.
317 (md_apply_fix3): Formatting. Hack for bfd_install_relocation when
318 fx_pcrel, not when fx_addsy.
319 * config/tc-i386.h (TC_FORCE_RELOCATION): Define using code moved
320 from tc_i386_force_relocation.
321 (TC_FIX_ADJUSTABLE): Remove redundant tc_fix_adjustable condition.
322 * config/tc-s390.h (TC_FIX_ADJUSTABLE): Likewise.
323 * config/tc-sh.h (TC_FIX_ADJUSTABLE): Likewise.
324 * config/tc-i960.c (reloc_callj): Remove declaration. Return false.
325 * config/tc-i960.h (TC_FORCE_RELOCATION_SUB): Define.
326 (TC_FORCE_RELOCATION_ABS): Define.
327 * config/tc-ppc.c (md_apply_fix3): Rename arg to match comments.
328 Don't undo adjustments done by write.c:fixup_segment.
329 * config/tc-ppc.h (TC_FIX_ADJUSTABLE): Instead arrange for them
330 to not occur in the first place.
331 (TC_FORCE_RELOCATION_SECTION): Delete.
332 * config/tc-s390.c (md_apply_fix3): Use TC_FORCE_RELOCATION_LOCAL
333 instead of TC_RELOC_RTSYM_LOC_FIXUP.
334 * config/tc-sh.c (sh_fix_adjustable): Likewise.
335 * config/tc-sh.h (obj_fix_adjustable): Delete, instead..
336 (tc_fix_adjustable): Define.
338 2002-05-17 Alan Modra <amodra@bigpond.net.au>
340 * config/obj-generic.c: Delete file.
341 * config/obj-generic.h: Likewise.
343 2002-05-16 Marek Michalkiewicz <marekm@amelek.gda.pl>
345 * config/tc-avr.c (mcu_types): Update for new devices.
347 2002-05-15 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
349 * config/tc-mips.c (macro): Warn about wrong la/dla use.
351 2002-05-15 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
353 * config/tc_mips.c (s_cpsetup): Fix completely bogus code which had
354 worked sometimes by accident. Fix copy&paste comment.
356 2002-05-15 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
358 * config/tc-mips.c (md_begin): Fix .reginfo and .MIPS.option section
359 alignment for NewABI. Let n32 use .reginfo. Remove useless casts.
360 (mips_elf_final_processing): Let n32 use .reginfo.
362 2002-05-15 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
364 * config/tc-mips.c (append_insn): Fix too small range of variable.
366 2002-05-14 Nick Clifton <nickc@cambridge.redhat.com>
368 * config/tc-arm.c (arm_cleanup): Remove redundant call to
371 2002-05-13 Nick Clifton <nickc@cambridge.redhat.com>
373 * config/tc-arm.c (md_assemble): Remove redundant call to
376 * dwarf2dbg.c (dwarf2_emit_insn): Do not reset
379 * stabs.c (s_stab_generic): Fix grammatical error in warning
382 2002-05-13 Alan Modra <amodra@bigpond.net.au>
384 * write.c (subsegs_finish): Don't specially align last subseg.
386 2002-05-11 Nick Clifton <nickc@cambridge.redhat.com>
388 * stabs.c (s_stab_generic): Warn about a description field that is
391 * config/obj-coff.c: Fix compile time warnings when compiling
392 without BFD_ASSEMBLER defined.
395 * config/tc-sh.c (md_pcrel_from): Define for use with sh-hms
397 (md_pcrel_from_section): Use md_pcrel_from().
399 2002-05-11 Bruno Haible <bruno@clisp.org>
401 * dwarf2dbg.c (dwarf2_emit_insn): Use the 'current' struct filled
402 by dwarf2_directive_loc, instead of calling dwarf2_where.
404 2002-05-11 Kazu Hirata <kazu@cs.umass.edu>
406 * config/obj-coff.h: Fix formatting.
407 * config/tc-mcore.c: Likewise.
408 * config/tc-mn10300.c: Likewise.
409 * config/tc-openrisc.c: Likewise.
410 * config/tc-or32.c: Likewise.
411 * config/tc-pdp11.c: Likewise.
412 * config/tc-ppc.c: Likewise.
413 * config/tc-ppc.h: Likewise.
414 * config/tc-sh64.c: Likewise.
415 * config/tc-sh.c: Likewise.
416 * config/tc-tic54x.c: Likewise.
417 * config/tc-xstormy16.c: Likewise.
418 * config/tc-xstormy16.h: Likewise.
420 2002-05-09 Kazu Hirata <kazu@cs.umass.edu>
422 * config/obj-coff.c: Fix formatting.
423 * config/obj-elf.c: Likewise.
424 * config/tc-alpha.c: Likewise.
425 * config/tc-arm.c: Likewise.
426 * config/tc-d10v.c: Likewise.
427 * config/tc-d30v.c: Likewise.
428 * config/tc-h8300.c: Likewise.
429 * config/tc-hppa.c: Likewise.
431 2002-05-09 Alan Modra <amodra@bigpond.net.au>
433 * config/tc-i386.c (md_estimate_size_before_relax) Don't lose
434 reloc when no_cond_jump_promotion.
436 2002-05-08 Jim Wilson <wilson@redhat.com>
438 * config/tc-i960.c (md_estimate_size_before_relax): Return size of
439 current variable part of frag.
441 2002-05-08 Kazu Hirata <kazu@cs.umass.edu>
443 * config/tc-mmix.c: Fix formatting.
444 * config/tc-mmix.h: Likewise.
446 2002-05-08 Alan Modra <amodra@bigpond.net.au>
448 * configure: Regenerate.
450 2002-05-07 Kazu Hirata <kazu@cs.umass.edu>
452 * config/tc-m68k.c: Fix formatting.
454 2002-05-07 Federico G. Schwindt <fgsch@olimpo.com.br>
456 * Makefile.am: Honour DESTDIR.
457 * Makefile.in: Regenerate.
459 2002-05-06 Kazu Hirata <kazu@cs.umass.edu>
461 * config/tc-ia64.c: Fix formatting.
462 * config/tc-ia64.h: Likewise.
464 2002-05-04 Kazu Hirata <kazu@cs.umass.edu>
466 * config/tc-mips.c: Fix formatting.
467 * config/tc-s390.c: Likewise.
468 * config/tc-s390.h: Likewise.
470 2002-05-03 Alexandre Oliva <aoliva@redhat.com>
472 * config/tc-s390.c (md_gather_operands): Emit dwarf2 line-number
473 information for instructions.
475 2002-05-02 Kazu Hirata <kazu@cs.umass.edu>
477 * as.h: Fix formatting.
480 * dwarf2dbg.c: Likewise.
486 * symbols.c: Likewise.
488 2002-05-02 Alan Modra <amodra@bigpond.net.au>
490 * app.c (mri_pseudo): Only declare for TC_M68K.
492 * config/tc-ppc.c (mapping): Map sectoff to BFD_RELOC_16_BASEREL.
493 (ppc_elf_validate_fix): Replace BFD_RELOC_32_BASEREL with
494 BFD_RELOC_16_BASEREL.
495 (md_assemble): Likewise.
496 (md_apply_fix3): Likewise.
498 2002-05-02 Nick Clifton <nickc@cambridge.redhat.com>
500 * config/tc-arm.c (thumb_add_sub): Do not convert a subtract of
501 zero into an add of zero - it is not the same.
503 2002-05-01 Arati Dikey <aratid@kpit.com>
505 * tc-sh.c (get_specific): Generate warning if the same
506 destination register is used in parallel instructions.
508 2002-05-01 Andrew Macleod <amacleod@cygnus.com>
510 * config/tc-i386.c (extra_symbol_chars): Add '[' to the list.
512 2002-05-01 Alan Modra <amodra@bigpond.net.au>
514 * write.c (cvt_frag_to_fill): Set fr_offset to zero on .org
515 backwards to prevent cascading errors.
517 2002-04-30 Mark Mitchell <mark@codesourcery.com>
519 * configure.in: Add support for powerpc-*-windiss.
520 * configure: Regenerated.
522 2002-04-28 Alan Modra <amodra@bigpond.net.au>
524 * config/tc-s390.c (md_parse_option): Formatting.
526 * config/tc-i386.c: Formatting fixes, add missing space in error
529 2002-04-24 Christian Groessler <chris@groessler.org>
531 * config/tc-z8k.c (build_bytes): Add support for new cases:
532 CLASS_IGNORE and ARG_NIM4.
533 (md_assemble): Prevent destruction of input_line_pointer if
534 get_operands returns failure.
536 2002-04-24 Chris G. Demetriou <cgd@broadcom.com>
538 * config/tc-mips.c (macro_build): Do _not_ allow MIPS-3D
539 instructions to be generated by macros.
541 2002-04-24 Andreas Schwab <schwab@suse.de>
543 * config/tc-i386.c (output_jump, output_disp)
544 (md_estimate_size_before_relax): Don't set fx_pcrel_adjust any
546 (md_apply_fix3): Remember addend value for rela relocations.
547 (tc_gen_reloc): Correctly compute pc-relative relocation addend.
549 2002-04-22 Chris Demetriou <cgd@broadcom.com>
551 * config/tc-mips.c (macro_build): Add close-parenthesis missing
552 from previous change.
554 2002-04-22 Eric Christopher <echristo@redhat.com>
556 * config/tc-mips.c (macro_build): Add warning if macro instructions
557 are expanded into a branch delay slot.
559 2002-04-17 Geoffrey Keating <geoffk@redhat.com>
561 * dwarf2dbg.c (dwarf2_gen_line_info): Do emit duplicate line
562 numbers, gdb relies on them to detect the start of the prologue.
564 2002-04-17 Martin Schwidefsky <schwidefsky@de.ibm.com>
566 * config/tc-s390.c (tc_s390_fix_adjustable): Prevent adjustments to
567 symbols in merge sections.
569 2002-04-16 Alan Modra <amodra@bigpond.net.au>
571 * as.c (main): Don't reference _bfd_chunksize.
573 2002-04-15 Tom Rix <trix@redhat.com>
575 * config/tc-d10v.c (d10v_fix_adjustable): Prevent adjustments to
576 symbols in merge sections.
578 2002-04-11 Richard Sandiford <rsandifo@redhat.com>
580 * doc/invoke.texi (TC_LARGEST_EXPONENT_IS_NORMAL): Document.
581 * config/atof-ieee.c (TC_LARGEST_EXPONENT_IS_NORMAL): Add an
582 argument for the precision.
583 (gen_to_words): Update accordingly.
585 2002-04-10 Alan Modra <amodra@bigpond.net.au>
587 * as.c (parse_args <OPTION_VERSION>): Use VERSION is
588 BFD_VERSION_STRING unavailable.
589 * config/tc-i386.c (INLINE): Define (for non-BFD assembler).
591 Tue Apr 9 16:45:48 2002 J"orn Rennecke <joern.rennecke@superh.com>
593 * config/tc-sh.h (TC_FIX_ADJUSTABLE): Disable adjusting if
594 symbol_used_in_reloc_p is true.
596 * config/tc-sh.c (md_apply_fix3): Don't zero relocations on big
599 2002-04-04 Alan Modra <amodra@bigpond.net.au>
601 * dep-in.sed: Cope with absolute paths.
602 * Makefile.am (dep.sed): Subst TOPDIR.
604 * Makefile.in: Regenerate.
606 2002-04-04 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
608 * config/tc-mips.c (mips16_macro_build): Cast type mismatch.
609 (mips_ip): Remove unused variable.
610 (md_apply_fix3): Cast signed/unsignes mismatches. Replace
611 unsigned char with bfd_byte.
612 (s_file): Remove unused variable.
613 (s_mips_ent): Likewise.
615 2002-04-03 Tom Rix <trix@redhat.com>
617 * config/tc-d10v.c (d10v_insert_operand): Fix warning in as_bad_where.
619 (find_opcode): Fix warning in as_warn.
620 * config/tc-d10v.h: Update Copyright.
622 2002-04-03 Alan Matsuoka <alanm@redhat.com>
623 Tom Rix <trix@redhat.com>
625 From Jeff Knaggs <jknaggs@redhat.com>
626 * config/tc-d10v.c (check_resource_conflict): New function to
627 check for resource conflicts.
629 From Jason Eckhardt <jle@redhat.com>
630 * config/tc-d10v.c (build_insn): Check for unresolved imm4 or
632 * config/tc-d10v.c (find_opcode): Emit a warning if one of the
633 reserved control registers is used.
634 * config/tc-d10v.c (build_insn): Check for unresolved imm4 or
637 From 2001-03-28 Diego Novillo <dnovillo@redhat.com>
638 * tc-d10v.c (parallel_ok): Prevent packing only if the first
639 instruction cannot be packed.
641 From 2001-03-30 Diego Novillo <dnovillo@redhat.com>
642 * gas/config/tc-d10v.c (check_resource_conflict): Only check
643 write-write conflicts.
644 (md_assemble): Reformat introductory comment.
645 * opcodes/d10v-opc.c (d10v_opcodes): `btsti' does not modify its
648 2002-04-03 Alan Modra <amodra@bigpond.net.au>
650 * symbols.c (resolve_symbol_value <O_uminus, O_bit_not,
651 O_logical_not>): Derive final_seg from add_symbol.
652 <O_multiply..O_logical_or>: More final_seg twiddles.
654 2002-04-01 Jessica Han <jessica@cup.hp.com>
656 * config/tc-ia64.c (ia64_cons_fix_new): Handle 8 byte iplt reloc
659 2002-03-27 Andreas Schwab <schwab@suse.de>
661 * config/tc-i386.c (output_jump): Set fx_pcrel_adjust to size of
662 field for pc-relative fixups.
663 (output_disp): Likewise.
664 (md_estimate_size_before_relax): Likewise.
665 (tc_gen_reloc): Subtract fx_pcrel_adjust instead of fx_size for
666 pc-relative fixups in 64bit mode.
668 2002-03-22 Alan Modra <amodra@bigpond.net.au>
670 * config/te-aix5.h: Typo fix.
672 2002-03-21 Alan Modra <amodra@bigpond.net.au>
674 * Makefile.am: Run "make dep-am".
675 * Makefile.in: Regenerate.
676 * doc/Makefile.in: Regenerate.
678 2002-03-20 Albert Chin-A-Young <china@thewrittenword.com>
680 * config/tc-arm.c (vfp_dp_reg_required_here): Fix typo
681 (vfp_sp_reg_pos -> vfp_dp_reg_pos).
683 2002-03-18 Alexandre Oliva <aoliva@redhat.com>
685 * config/tc-mips.c (md_estimate_size_before_relax): Do not modify
686 the EXTENDED bit here; report the estimate according to the
689 2002-03-18 Tom Rix <trix@redhat.com>
691 * configure.in: Add AIX 5 support.
692 * config/tc-ppc.c (ppc_target_format): Add AIX 5 64 bit target.
693 * config/te-aix5.h: New file. AIX 5 support.
694 * configure: Regenerate.
696 2002-03-18 Nick Clifton <nickc@cambridge.redhat.com>
698 * po/fr.po: Updated version.
700 2002-03-16 Andreas Jaeger <aj@suse.de>
702 * doc/c-mips.texi (Machine Dependencies): Add new node.
704 2002-03-15 Chris G. Demetriou <cgd@broadcom.com>
706 * config/tc-mips.c (mips_set_options): New "ase_mips3d" member.
707 (mips_opts): Initialize "ase_mips3d" member.
708 (file_ase_mips3d): New variable.
709 (CPU_HAS_MIPS3D): New macro.
710 (md_begin): Initialize mips_opts.ase_mips3d and file_ase_mips3d
711 based on command line options and configuration defaults.
712 (macro_build, mips_ip): Accept MIPS-3D instructions if
713 mips_opts.ase_mips3d is set.
714 (OPTION_MIPS3D, OPTION_NO_MIPS3D, md_longopts, md_parse_option):
715 Add support for "-mips3d" and "-no-mips3d" options.
716 (OPTION_ELF_BASE): Move to accomodate new options.
717 (s_mipsset): Support ".set mips3d" and ".set nomips3d".
718 (mips_elf_final_processing): Add a comment indicating that a
719 MIPS-3D ASE ELF header flag should be set, when one exists.
720 * doc/as.texinfo: Document -mips3d and -no-mips3d options.
721 * doc/c-mips.texi: Likewise, and document ".set mips3d" and ".set
722 nomips3d" directives.
724 2002-03-14 Hans-Peter Nilsson <hp@bitrange.com>
726 * config/tc-mmix.c (md_estimate_size_before_relax): Don't consider
727 a weak symbol in same section to be within reach.
729 2002-03-13 Nick Clifton <nickc@cambridge.redhat.com>
731 * po/fr.po: Updated version.
733 2002-03-12 Andreas Schwab <schwab@suse.de>
735 * config/tc-ia64.c (fixup_unw_records): Clear region when seeing a
736 body record so that an error is given for misplaced .save
739 2002-03-09 Alan Modra <amodra@bigpond.net.au>
741 * config/tc-i386.h (REX_OPCODE): Define.
742 (REX_MODE64, REX_EXTX, REX_EXTY, REX_EXTZ): Define.
743 (rex_byte): typedef to int.
744 * config/tc-i386.c: Group prototypes and vars together.
745 Formatting fixes. Remove occurrences of "register" keyword.
748 (mode_from_disp_size): Add INLINE keyword to prototype.
749 (fits_in_signed_byte): Likewise.
750 (fits_in_unsigned_byte): Likewise.
751 (fits_in_unsigned_word): Likewise.
752 (fits_in_signed_word): Likewise.
753 (fits_in_unsigned_long): Likewise.
754 (fits_in_signed_long): Likewise.
755 (type_names): Constify.
756 (intel_float_operand): Constify param.
757 (add_prefix): Use REX_OPCODE.
758 (md_assemble): Likewise. Modify for changed rex_byte.
759 (parse_insn): Split out of md_assemble.
760 (parse_operands): Likewise.
761 (swap_operands): Likewise.
762 (optimize_imm): Likewise.
763 (optimize_disp): Likewise.
764 (match_template): Likewise.
765 (check_string): Likewise.
766 (process_suffix): Likewise.
767 (check_byte_reg): Likewise.
768 (check_long_reg): Likewise.
769 (check_qword_reg): Likewise.
770 (check_word_reg): Likewise.
771 (finalize_imm): Likewise.
772 (process_operands): Likewise.
773 (build_modrm_byte): Likewise.
774 (output_insn): Likewise.
775 (output_branch): Likewise.
776 (output_jump): Likewise.
777 (output_interseg_jump): Likewise.
778 (output_disp): Likewise.
779 (output_imm): Likewise.
781 2002-03-07 Daniel Jacobowitz <drow@mvista.com>
783 * doc/as.texinfo: Wrap @menu in @ifnottex, not @ifinfo.
785 2002-03-06 Alan Modra <amodra@bigpond.net.au>
787 * config/tc-i386.c (tc_gen_reloc): Don't attempt to handle 8 byte
788 relocs except when BFD64.
790 * write.c (number_to_chars_bigendian): Don't abort when N is
791 larger than sizeof (VAL).
792 (number_to_chars_littleendian): Likewise.
794 2002-03-05 John David Anglin <dave@hiauly1.hia.nrc.ca>
796 * config/tc-hppa.c (md_apply_fix3): Add cast.
797 (hppa_fix_adjustable): Adjust list of selectors using e_lrsel and
800 2002-03-05 Paul Koning <pkoning@equallogic.com>
802 * tc-pdp11.c: Use VAX float format support for PDP-11 target.
803 (parse_ac5): New function for parsing float regs in float operand.
804 (parse_expression): Remove attempt to make literals be octal.
805 (parse_op_no_deferred): Support float literals.
806 (parse_op): Reject attempts to refer to float regs.
807 (parse_fop): New function, like parse_op but for float operand.
808 (md_assemble): Add cases to parse float operands. Also fix
809 IMM3, IMM6, IMM8 cases to pick up the operand from the right spot.
811 2002-03-04 H.J. Lu <hjl@gnu.org>
813 * config/obj-elf.c (special_section): Add .init_array,
814 .fini_array and .preinit_array.
816 * config/tc-ia64.h (ELF_TC_SPECIAL_SECTIONS): Remove
817 .init_array and .fini_array.
819 2002-03-01 Jakub Jelinek <jakub@redhat.com>
821 * config/obj-elf.c (elf_copy_symbol_attributes): Don't copy
823 (obj_frob_symbol): Copy visibility.
825 2002-02-28 Jakub Jelinek <jakub@redhat.com>
827 * config/tc-alpha.c (s_alpha_text): Use obj_elf_text for OBJ_ELF, not
829 (s_alpha_data): Use obj_elf_data for OBJ_ELF, not s_data.
831 2002-02-27 Nick Clifton <nickc@cambridge.redhat.com>
835 2002-02-26 Chris Demetriou <cgd@broadcom.com>
837 * config/tc-mips.c (mips_need_elf_addend_fixup): For embedded-PIC
838 only, undo the changes made on 2001-06-08, with the
839 effect being that common or extern symbols are
840 adjusted for embedded-PIC, but weak symbols are not.
841 (md_estimate_size_before_relax: Likewise, with the effect
842 that extern symbols are treated the same as weak symbols
843 only if not embedded-PIC.
844 (mips_fix_adjustable) Likewise, with the effect that
845 weak or extern symbols are not adjusted for embedded-PIC.
846 (md_apply_fix3): Tweak so that the case where value is zero
847 is handled more correctly for embedded-PIC code.
849 2002-02-26 Nick Clifton <nickc@cambridge.redhat.com>
851 * doc/as.texinfo (Overview): Add missing @ifset IA64
853 * configure.in (LINGUAS): Add es.po.
854 * configure: Regenerate.
855 * po/es.po: New file.
857 2002-02-25 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
859 * config/tc-mips.c (set_at): Fix handling of 64bit register loads.
860 (macro): Likewise. Fix la/dla address expansions for EMBEDDED_PIC
861 and NO_PIC cases. Code cleanup.
862 (macro2): Fix handling of 64bit register loads.
864 2002-02-25 David Mosberger <davidm@hpl.hp.com>
866 * doc/as.texinfo: Add entry for IA64.
867 * doc/c-ia64.texi: New file.
869 2002-02-25 Alan Modra <amodra@bigpond.net.au>
871 * config/tc-hppa.c: Update copyright date.
873 * doc/c-ppc.texi (PowerPC-Opts): Add -mpower4, -maltivec and -m7400
874 Remove references to chip manufacturers.
875 * config/tc-ppc.c (md_parse_option): Handle -mpower4 option.
877 (md_show_usage): Remove references to chip manufacturers. Mention
879 (md_begin): Test power4 opcode flag bits.
881 2002-02-22 David Mosberger <davidm@hpl.hp.com>
883 * config/tc-ia64.c (dot_restore): Issue error message of epilogue
884 count exceeds prologue count.
885 (md_show_usage): Describe -mconstant-gp and -mauto-pic.
886 (unwind.label_prologue_count): New member.
888 Based on a patch by Hans Boehm <hboehm@hpl.hp.com>:
890 (get_saved_prologue_count): New function.
891 (save_prologue_count): New function.
892 (free_saved_prologue_count): New function.
893 (dot_label_state): Record state label by calling save_prologue_count().
894 (dot_copy_state): Restore prologue count by calling
895 get_saved_prologue_count().
896 (generate_unwind_image): Free up list of saved prologue
897 counts by calling free_saved_prologue_counts().
899 2002-02-22 Nick Clifton <nickc@cambridge.redhat.com>
901 * config/tc-tic54x.c: Add missing prototypes and remove ANSI style
902 function declarations.
904 2002-02-21 Nick Clifton <nickc@cambridge.redhat.com>
906 * NEWS: Note that GASP is now deprecated.
907 * Makefile.am: Do not build gasp-new by default.
908 * Makefile.in: Regenerate.
909 * doc/Makefile.am: Do not install gasp.info.
910 * doc/Makefile.in: Regenerate.
911 * gas/gasp.texi: Note that gasp is now deprecated.
913 2002-02-20 Nick Clifton <nickc@cambridge.redhat.com>
915 * NEWS: Mark 2.12 branch.
917 2002-02-19 Tom Tromey <tromey@redhat.com>
919 * config/tc-xstormy16.h (DWARF2_LINE_MIN_INSN_LENGTH): Define.
921 2002-02-19 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
923 * config/tc-mips.c (md_parse_option): Complain about invalid -mabi
926 2002-02-19 Martin Schwidefsky <schwidefsky@de.ibm.com>
928 * config/tc-s390.c (md_parse_option): Add switches -m31 and -m64.
929 Make bit size independent of architecture switch.
930 (md_begin): Add warning for -m64 with -Aesa.
931 (s390_md_end): Use renamed architecture defines.
933 2002-02-18 Daniel Jacobowitz <drow@mvista.com>
935 * config/obj-coff.h: Check !target_big_endian, not shl, for coff-sh.
937 2002-02-16 Nick Clifton <nickc@cambridge.redhat.com>
939 * doc/as.texinfo (Machine Dependencies): Fix typo: MMIX used
942 2002-02-15 Chris Demetriou <cgd@broadcom.com>
944 * config/tc-mips.c (md_estimate_size_before_relax): Really
945 make sure we treat weak like extern only for ELF. (Fixes
946 patch from 2001-07-25.)
948 2002-02-15 Ben Elliston <bje@redhat.com>
950 * doc/as.texinfo: Add duplicate directory entry so that "info gas"
953 2002-02-15 Hans-Peter Nilsson <hp@bitrange.com>
955 * NEWS: Mention support for MMIX.
957 2002-02-13 Chris Demetriou <cgd@broadcom.com>
959 * config/tc-mips.c (mips_need_elf_addend_fixup): Restructure into
960 a sequence of indpendent 'if' statements for easier debugging
961 and future modification.
963 2002-02-13 Matt Fredette <fredette@netbsd.org>
965 * config/tc-m68k.c (md_show_usage): No longer display a
966 hard-coded "68020" for the default CPU, instead display the
967 canonical name of the true, configured default CPU.
968 (m68k_elf_final_processing): Mark objects for sub-68020
969 CPUs with the new EF_M68000 flag.
971 2002-02-13 Andreas Schwab <schwab@suse.de>
973 * config/tc-i386.c (tc_i386_fix_adjustable): Don't adjust
974 pc-relative relocations to merge sections in 64-bit mode.
976 2002-02-13 Ben Elliston <bje@redhat.com>
978 * NEWS: Document floating point number handling in gasp.
979 * gasp.c: Include <assert.h> and "xregex.h".
980 (is_flonum): New function.
981 (chew_flownum): Likewise.
982 (change_base): Consume flonums from the input, where possible.
983 * doc/gasp.texi (Constants): Document floating point numbers.
985 2002-02-12 John David Anglin <dave@hiauly1.hia.nrc.ca>
987 * config/tc-hppa.c (hppa_fix_adjustable): Don't adjust final types
988 that implicitly use LR and RR selectors.
990 2002-02-12 Alexandre Oliva <aoliva@redhat.com>
992 * config/tc-mn10300.c (other_registers): Added `epsw'. Mark `pc'
993 and `epsw' as available on AM33 and above only.
994 (other_register_name): Add logic to handle machine type encoded in
997 2002-02-11 Tom Rix <trix@redhat.com>
999 * config/tc-ppc.c (md_apply_fix3): Check for insert fop invalid for
1002 Mon Feb 11 11:23:38 2002 Alexandre Oliva <aoliva@redhat.com>
1004 * config/tc-sparc.c (U0x80000000, U0xffffffff): New constants.
1007 Mon Feb 11 12:59:29 CET 2002 Jan Hubicka <jh@suse.cz>
1009 * i386.c (md_assemble): Support 32bit address prefix.
1010 (i386_displacement): Likewise.
1011 (i386_index_check): Accept 32bit addressing in 64bit mode.
1013 2002-02-11 Alexandre Oliva <aoliva@redhat.com>
1015 * config/tc-sh.c (dot): Removed unused function.
1017 2002-02-11 Alan Modra <amodra@bigpond.net.au>
1019 * Makefile.am: "make dep-am".
1020 * Makefile.in: Regenerate.
1021 * aclocal.m4: Regenerate.
1022 * config.in: Regenerate.
1023 * configure: Regenerate.
1024 * doc/Makefile.in: Regenerate.
1026 2002-02-10 Richard Henderson <rth@redhat.com>
1028 * doc/c-alpha.texi: New file.
1029 * doc/Makefile.am (CPU_DOCS): Add it.
1030 * doc/all.texi, doc/as.texinfo: Add hooks for Alpha.
1032 2002-02-09 Richard Henderson <rth@redhat.com>
1034 * config/tc-alpha.c (O_samegp): New.
1035 (USER_RELOC_P): Include it.
1036 (alpha_reloc_op_tag, debug_exp, find_macro_match): Add it.
1037 (md_apply_fix3): Handle BFD_RELOC_ALPHA_BRSGP.
1038 (alpha_force_relocation, alpha_fix_adjustable): Likewise.
1039 (alpha_validate_fix): New.
1040 * config/tc-alpha.h (TC_VALIDATE_FIX): New.
1042 2002-02-09 Hans-Peter Nilsson <hp@axis.com>
1044 * doc/c-cris.texi: New.
1045 * doc/all.texi: @set CRIS.
1046 * doc/as.texinfo: Ditto. Add CRIS gas manpage option overview.
1047 Include c-cris.texi.
1048 * doc/Makefile.am (CPU_DOCS): Add c-cris.texi
1049 * doc/Makefile.in: Regenerate.
1051 2002-02-08 Chris Demetriou <cgd@broadcom.com>
1053 * config/tc-mips.c (IS_SEXT_32BIT_NUM): New macro to
1054 determine if a number is a sign-extended 32-bit number.
1055 (load_register): Use IS_SEXT_32BIT_NUM.
1056 (macro): Check if load/store macro handling is using a
1057 constant 32-bit address on 64-bit address systems, and if
1058 so optimize the generation of that address.
1060 2002-02-08 Richard Henderson <rth@redhat.com>
1062 * config/tc-alpha.c (alpha_force_relocation): Don't assert that
1063 we've eliminated all foreign relocation types yet.
1064 (alpha_fix_adjustable): Likewise.
1066 2002-02-08 Alexandre Oliva <aoliva@redhat.com>
1068 Contribute sh64-elf.
1069 2002-02-08 Alexandre Oliva <aoliva@redhat.com>
1070 Stephen Clarke <Stephen.Clarke@st.com>
1071 * doc/c-sh64.texi: Fix citation of SH64 architecture manual.
1072 2002-01-31 Alexandre Oliva <aoliva@redhat.com>
1073 * config/tc-sh.c (md_relax_table): Added default sizes for
1074 non-PC-relative UNDEF_MOVI, and relaxation sequences for
1075 MOVI_16, MOVI_32 and MOVI_48.
1076 * config/tc-sh64.c (shmedia_md_apply_fix3): Fix warning.
1077 (shmedia_md_convert_frag): Handle non-PC-relative UNDEF_MOVI
1079 (shmedia_md_estimate_size_before_relax): Remove redundant
1080 blocks. Set fragP->fr_var even if relaxation type unchanged.
1081 Retain UNDEF_MOVI until expression decays to number.
1082 2002-01-24 Alexandre Oliva <aoliva@redhat.com>
1083 * config/tc-sh64.c (shmedia_init_reloc): Handle new SHmedia PIC
1084 relocation types. Take fixP->fx_addnumber into account too.
1085 (shmedia_md_apply_fix): Likewise.
1086 (shmedia_md_convert_frag): Likewise.
1087 (shmedia_build_Mytes): Likewise.
1088 (sh64_consume_datalabel): Complain about nested datalabel.
1089 Support PIC relocs. Call sh_parse_name.
1090 * config/tc-sh64.h (TC_RELOC_RTSYM_LOC_FIXUP): Extend definition
1091 in tc-sh.h to SHmedia reloc types.
1092 * config/tc-sh.c (SH64PCRELPLT, MOVI_PLT, MOVI_GOTOFF,
1093 MOVI_GOTPC): New relaxation constants.
1094 (md_relax_table): Introduce relaxation directives for PIC-related
1096 (sh_PIC_related_p): Handle datalabel.
1097 (sh_check_fixup): Choose SH5 PIC relocations.
1098 (sh_cons_fix_new): Added BDF_RELOC_64.
1099 (md_apply_fix3, sh_parse_name): Handle GOTPLT.
1100 2002-01-18 Alexandre Oliva <aoliva@redhat.com>
1101 * config/tc-sh64.c (sh64_max_mem_for_rs_align_code): If the
1102 current ISA is SHmedia, get 7 bytes.
1103 2001-11-28 Nick Clifton <nickc@cambridge.redhat.com>
1104 * config/tc-sh.c (md_apply_fix3): Treat shmedia_md_apply_fix3 as a
1106 * config/tc-sh64.c (shmedia_apply_fix): Rename to
1107 shmedia_apply_fix3 and make void.
1108 2001-05-17 Alexandre Oliva <aoliva@redhat.com>
1109 * config/tc-sh64.c (s_sh64_abi): Remove unused arguments passed to
1111 2001-04-12 Alexandre Oliva <aoliva@redhat.com>
1112 * config/tc-sh64.h (md_parse_name): Take &c as argument.
1113 2001-03-14 DJ Delorie <dj@redhat.com>
1114 * doc/Makefile.am (CPU_DOCS): Added c-sh64.texi
1115 * doc/Makefile.in(CPU_DOCS): Ditto.
1116 * doc/c-sh64.texi: New file.
1117 * doc/as.texinfo: Add SH64 support.
1118 2001-03-13 DJ Delorie <dj@redhat.com>
1119 * config/tc-sh64.c (shmedia_get_operands): Rename A_RESV_Fx to
1120 A_REUSE_PREV so that its purpose is more obvious.
1121 (shmedia_build_Mytes): Ditto.
1122 2001-03-07 DJ Delorie <dj@redhat.com>
1123 * config/tc-sh64.c (sh64_vtable_entry): New, strip datalabels
1125 (sh64_vtable_inherit): Ditto.
1126 (strip_datalabels): New, strip "datalabel" from given line.
1127 * config/tc-sh.c (md_pseudo_table): Add sh64-specific vtable
1129 2001-03-06 Hans-Peter Nilsson <hpn@cygnus.com>
1130 * config/tc-sh64.c (shmedia_md_assemble): Move dwarf2_emit_insn
1132 (shmedia_build_Mytes): ... to here.
1133 2001-03-06 DJ Delorie <dj@redhat.com>
1134 * config/tc-sh.c: Remove sh64-specific uaquad now that there
1136 2001-01-21 Hans-Peter Nilsson <hpn@cygnus.com>
1137 * config/tc-sh64.h (DWARF2_LINE_MIN_INSN_LENGTH): Override.
1138 * config/tc-sh64.c (shmedia_md_assemble): Offset recorded insn
1139 address by one in call to dwarf2_emit_insn.
1140 2001-01-13 Hans-Peter Nilsson <hpn@cygnus.com>
1141 Implement ".abi" pseudo and correct .cranges descriptors. Correct
1142 alignment handling broken by imported changes.
1143 * config/tc-sh64.h (HANDLE_ALIGN): Override definition in tc-sh.h.
1144 (sh64_handle_align): Declare.
1145 (MAX_MEM_FOR_RS_ALIGN_CODE): Override definition in tc-sh.h.
1146 (sh64_max_mem_for_rs_align_code): Declare.
1147 (enum sh64_isa_values): Moved here from tc-sh64.c.
1148 (md_do_align): Define.
1149 (sh64_do_align): Declare.
1150 (struct sh64_tc_frag_data): New.
1151 (TC_FRAG_TYPE): Change to struct sh64_tc_frag_data. Users
1153 (TC_FRAG_INIT): Change to set new datatype.
1154 (struct sh64_segment_info_type): Rename member
1155 last_flushed_location to last_contents_mark. All users changed.
1156 (md_elf_section_change_hook, TC_CONS_FIX_NEW): Do not define.
1157 (shmedia_elf_new_section, sh64_tc_cons_fix_new): Do not prototype.
1158 * config/tc-sh.c (md_pseudo_table): Add ".abi".
1159 (sh_elf_cons) [HAVE_SH64]: Call sh64_update_contents_mark instead
1160 of unsetting seen_insn.
1161 (md_assemble) [HAVE_SH64] <before new SHcompact sequence>: Also
1162 call sh64_update_contents_mark.
1163 (sh_handle_align): Remove HAVE_SH64-conditioned code.
1164 * config/tc-sh64.c (sh64_isa_mode): Correct type from boolean to
1165 enum sh64_isa_values.
1166 (sh64_set_contents_type): Drop segT parameter. All callers changed.
1167 (emitting_crange): Boolean guard moved to file scope from function
1168 scope in sh64_set_contents_type.
1170 (sh64_update_contents_mark): New; most split out from
1171 sh64_flush_pending_output.
1172 (shmedia_md_end): Call sh64_update_contents_mark. Set
1173 sh64_isa_mode to sh64_isa_sh5_guard unless sh64_isa_unspecified.
1174 (sh64_do_align): New function.
1175 (sh64_max_mem_for_rs_align_code): New function.
1176 (sh64_handle_align): Rename from shmedia_do_align. Make
1177 non-static. Add head comment. Emit zero bytes for n bytes modulo
1178 four. Change return-type to void.
1179 (shmedia_elf_new_section): Remove.
1180 (shmedia_md_assemble): Call sh64_update_contents_mark.
1181 (s_sh64_mode): Ditto. Do not call md_flush_pending_output. Make
1182 new frag. Call sh64_update_contents_mark after making the new
1184 (sh64_flush_pending_output): Just call sh64_update_contents_mark
1185 and sh_flush_pending_output.
1186 (sh64_flag_output): Also call md_flush_pending_output, but add
1187 condition on not emitting_crange.
1188 (sh64_tc_cons_fix_new): Remove.
1189 2001-01-12 Nick Clifton <nickc@redhat.com>
1190 * config/tc-sh64.c (shmedia_do_align): Fix to work with new
1191 alignment handling scheme imported from sourceware.
1192 2001-01-12 Hans-Peter Nilsson <hpn@cygnus.com>
1193 * config/tc-sh64.h (TARGET_FORMAT): Define.
1194 (sh64_target_format): Prototype.
1195 * config/tc-sh64.c (sh64_target_mach): New function.
1196 2001-01-07 Hans-Peter Nilsson <hpn@cygnus.com>
1197 * config/tc-sh64.c (shmedia_md_end): When equating a symbol, use
1198 zero_address_frag instead of copying the frag of the symbol.
1199 (shmedia_frob_file_before_adjust): Ditto.
1200 (shmedia_md_apply_fix) <case BFD_RELOC_SH_IMM_MEDLOW16>: Cast mask
1201 to valueT to remove signedness.
1202 (shmedia_md_convert_frag): Add parameter final. Rename parameter
1203 headers to output_bfd. Do not evaluate symbols if final is false;
1205 (shmedia_md_estimate_size_before_relax) <case C (MOVI_IMM_32,
1206 UNDEF_MOVI) et al>: If symbol cannot be modified to be PC-relative
1207 to the current frag, call shmedia_md_convert_frag to emit fixups
1208 and make frag_wane neutralize the frag. Update comments.
1209 * config/tc-sh.c (md_convert_frag): Change caller of
1210 shmedia_md_convert_frag.
1211 2001-01-06 Hans-Peter Nilsson <hpn@cygnus.com>
1212 * config/tc-sh64.h: Tweak comments and correct formatting.
1213 * config/tc-sh64.c: Ditto.
1214 (shmedia_md_convert_frag) <PT/PTA/PTB 32, 48 and 64 bit
1215 expansion, MOVI pcrel expansion>: Fix thinko calculating offset
1216 for the no-relocation case.
1217 (shmedia_check_limits): Fix range check being off-by-one for PTA.
1218 * config/tc-sh.c: Ditto. Add proper comments to #ifdef/#ifndef
1220 (SH64PCREL16_F): Increment for proper max-PTA handling. Update
1222 (SH64PCREL16_M, MOVI_16_M): Correct range thinko.
1223 (SH64PCREL48_M, MOVI_48_M): Similar; don't count in length of
1225 (SH64PCREL32_M, MOVI_32_M): Ditto; handle overflowing expression.
1227 2001-01-05 Hans-Peter Nilsson <hpn@cygnus.com>
1228 * config/tc-sh64.c (shmedia_md_apply_fix) <second switch, case
1229 BFD_RELOC_SH_PT_16>: Set lowest bit in field to be relocated to 1.
1230 (shmedia_md_convert_frag) <case C (SH64PCREL16_32, SH64PCREL16) et
1231 al>: Set lowest bit of field to relocate to 1 and rest to empty,
1232 if reloc is emitted.
1233 2000-12-31 Hans-Peter Nilsson <hpn@cygnus.com>
1234 New options plus bugfixes.
1235 * config/tc-sh.c (md_longopts): New options "-no-expand" and
1237 (md_parse_option): Handle new options.
1238 (md_show_usage): Add blurb for new options.
1239 * config/tc-sh64.c (SHMEDIA_BFD_RELOC_PT): New macro.
1240 (sh64_expand, sh64_pt32): New variables.
1241 (shmedia_init_reloc): Handle BFD_RELOC_SH_PT_16.
1242 (shmedia_md_apply_fix): Hold original fixP->fx_r_type in
1243 orig_fx_r_type. Change SHMEDIA_BFD_RELOC_PT into
1244 BFD_RELOC_SH_PT_16. Handle BFD_RELOC_SH_PT_16 as pc-relative.
1245 <resolved previously-pc-relative relocs>: Handle
1246 SHMEDIA_BFD_RELOC_PT and BFD_RELOC_SH_PT_16.
1247 (shmedia_md_convert_frag) <case C (SH64PCREL16PT_64, SH64PCREL16),
1248 case C (SH64PCREL16PT_32, SH64PCREL16)>: Modify to PTB if operand
1249 points to SHcompact code.
1250 <case C (SH64PCREL16_32, SH64PCREL16), case C (SH64PCREL16_64,
1251 SH64PCREL16)>: Check that ISA of what operand points at and
1252 PTA/PTB matches, or emit error.
1253 (shmedia_check_limits): Handle BFD_RELOC_SH_PT_16 and
1254 SHMEDIA_BFD_RELOC_PT.
1255 (shmedia_immediate_op): If pcrel, emit fixup also for constant
1257 (shmedia_build_Mytes) <case A_IMMS16>: Also check sh64_expand in
1258 condition for MOVI expansion.
1259 <case A_PCIMMS16BY4>: Handle expansion to 32 bits only, if
1260 sh64_pt32. Emit only a BFD_RELOC_SH_PT_16 fixup if not
1262 <case A_PCIMMS16BY4_PT>: Likewise, but emit a SHMEDIA_BFD_RELOC_PT
1264 (sh64_target_format): Error-check setting of sh64_pt32 and
1265 sh64_expand. Fix typo in check for sh64_shcompact_const_crange.
1266 (shmedia_md_pcrel_from_section): Handle BFD_RELOC_SH_PT_16 and
1267 SHMEDIA_BFD_RELOC_PT as coming from SHmedia code.
1268 2000-12-31 Hans-Peter Nilsson <hpn@cygnus.com>
1269 * config/tc-sh64.c: Improve comments.
1270 (shmedia_md_convert_frag): Remove inactive is_pt_variant code. Do
1271 not say the linker will check correctness of PTA/PTB expansion.
1272 (shmedia_md_end): Make non-static.
1273 * config/tc-sh64.h (md_end): Define to shmedia_md_end. Add
1275 * config/tc-sh.c (sh_finalize): Remove.
1276 * config/tc-sh.h (md_end): Do not define.
1277 Remove prototype for sh_finalize.
1278 2000-12-30 Hans-Peter Nilsson <hpn@cygnus.com>
1279 * config/tc-sh64.c (shmedia_frob_section_type): Use a struct
1280 sh64_section_data container when storing section type in tdata
1281 field in elf_section_data.
1282 * config/tc-sh.c (sh_elf_final_processing): Change from EF_SH64 to
1284 * Makefile.am: Update dependencies.
1285 * Makefile.in: Regenerate.
1286 2000-12-22 Hans-Peter Nilsson <hpn@cygnus.com>
1287 * config/tc-sh64.c (shmedia_md_assemble): Don't protect
1288 dwarf2_emit_insn call with test on debug_type.
1289 2000-12-19 Hans-Peter Nilsson <hpn@cygnus.com>
1290 * config/tc-sh64.c (sh64_set_contents_type): Make contents-type
1291 CRT_SH5_ISA32 sticky for 64-bit.
1292 2000-12-18 Hans-Peter Nilsson <hpn@cygnus.com>
1293 Generate .crange sections when switching ISA mode or emitting
1294 constants in same section as code.
1295 * config/tc-sh64.c: Reformat structure definitions.
1296 (sh64_end_of_assembly, sh64_mix, sh64_shcompact_const_crange): New
1298 (sh64_set_contents_type): Rename from sh64_init_section. Rewrite
1299 to emit a .cranges descriptor when contents type changes. Only
1300 emit error if changing contents type and -no-mix is in effect.
1301 (sh64_emit_crange, sh64_flush_last_crange, sh64_flag_output,
1302 sh64_flush_pending_output, sh64_tc_cons_fix_new): New functions.
1303 (shmedia_md_end): Set sh64_end_of_assembly. Pass
1304 sh64_flush_last_crange over sections.
1305 When checking main symbol of datalabel symbol, check for
1306 STO_SH5_ISA32, not ISA type of section in definition.
1307 (shmedia_frob_file_before_adjust): Check main symbol for
1308 STO_SH5_ISA32; don't check ISA type of section in definition.
1309 (shmedia_frob_section_type): Adjust for .cranges; set section flag
1310 to SHF_SH5_ISA32_MIXED or SHF_SH5_ISA32 according to whether
1311 .cranges entries have been output.
1312 (shmedia_elf_new_section): Just call md_flush_pending_output.
1313 (shmedia_md_assemble): Do not emit a BFD_RELOC_SH_SHMEDIA_CODE
1314 fix. Do not set tc_segment_info_data.in_code for section. Call
1315 sh64_set_contents_type for SHmedia code.
1316 (s_sh64_mode): Do not call sh64_init_section or set seen_insn to
1317 false. Call md_flush_pending_output.
1318 (sh64_target_format): Check that -no-mix and
1319 -shcompact-const-crange are used in sane combination with other
1321 (shmedia_md_pcrel_from_section): Check type of fix for how to
1323 (sh64_consume_datalabel): Check symbol for having STO_SH5_ISA32,
1324 not ISA type of section in definition.
1325 * config/tc-sh64.h (struct sh64_segment_info_type): Rewrite to
1326 hold contents-type state.
1327 (md_flush_pending_output): Redefine to sh64_flush_pending_output.
1328 (sh64_flush_pending_output): Declare.
1329 (TC_CONS_FIX_NEW): Define to sh64_tc_cons_fix_new.
1330 (sh64_tc_cons_fix_new): Declare.
1331 * config/tc-sh.c (sh_elf_cons) [HAVE_SH64]: Unset seen_insn and
1332 call sh64_flag_output.
1333 (md_assemble) [HAVE_SH64]: Do not emit BFD_RELOC_SH_CODE. Just
1334 call sh64_set_contents_type to mark SHcompact code and set
1336 (md_longopts): New options "-no-mix" and
1337 "-shcompact-const-crange".
1338 (md_parse_option): Handle new options.
1339 (md_show_usage): Add blurb for new options.
1340 (md_number_to_chars) [HAVE_SH64]: Call sh64_flag_output.
1341 2000-12-15 Hans-Peter Nilsson <hpn@cygnus.com>
1342 * config/tc-sh64.c: Delete investigated and obsolete fixme:s.
1343 (sh64_last_insn_frag): New.
1344 (shmedia_md_convert_frag): Use tc_frag_data field of incoming frag
1345 to get frag for insn opcode for generating fixups; do not assume it
1347 (shmedia_build_Mytes): Set sh64_last_insn_frag after growing frag
1349 * config/tc-sh64.h (ELF_TC_SPECIAL_SECTIONS): Define for .cranges
1351 (TC_FRAG_TYPE): Define as fragS *.
1352 (TC_FRAG_INIT): Define to set tc_frag_data to sh64_last_insn_frag.
1353 (sh64_last_insn_frag): Declare.
1354 (sh64_consume_datalabel): Fix typo; check for seginfo != NULL,
1355 not == NULL before dereferencing.
1356 2000-12-12 Hans-Peter Nilsson <hpn@cygnus.com>
1357 Get rid of BFD section flag and EF_SH64_ABI64.
1358 * config/tc-sh64.c (shmedia_frob_section_type): Use
1359 elf_section_data (sec)->tdata, not a specific BFD section flag, to
1360 communicate the section as containing SHmedia code. Describe why.
1361 * config/tc-sh.c (sh_elf_final_processing): Tweak comment. Set
1362 EF_SH64 regardless of ABI.
1363 * config/tc-sh64.c (shmedia_md_apply_fix): Decapitalize "invalid"
1364 in error message. Handle resolved expressions for
1365 BFD_RELOC_SH_IMMS10, BFD_RELOC_SH_IMMS10BY2,
1366 BFD_RELOC_SH_IMMS10BY4 and BFD_RELOC_64.
1367 (shmedia_check_limits): Handle BFD_RELOC_64.
1368 (sh64_adjust_symtab): Do not decrement the GAS symbol value for
1369 a STO_SH5_ISA32 symbol, only the BFD value.
1370 2000-12-11 Ben Elliston <bje@redhat.com>
1371 * config/tc-sh64.c: Call dwarf2_emit_insn, not the defunct
1372 dwarf2_generate_asm_lineno.
1373 2000-12-11 Hans-Peter Nilsson <hpn@cygnus.com>
1374 Handle PC-relative MOVI expansions with assembler relaxation.
1375 Generate PC-relative relocs from 16-bit PC-relative expressions.
1376 * config/tc-sh64.c (SHMEDIA_MD_PCREL_FROM_FIX): Break out from...
1377 (shmedia_md_pcrel_from_section): ...here.
1378 (shmedia_md_apply_fix): Handle fixups for 16-bit operands that has
1379 turned completely resolved. Adjust relocation type for 16-bit
1380 immediate operands that has turned PC-relative. Adjust back for
1381 MD_PCREL_FROM_SECTION being applied twice.
1382 (shmedia_md_convert_frag): Always emit reloc for expression with
1383 global or weak symbol. Handle relaxation result for PC-relative
1385 (shmedia_md_estimate_size_before_relax): An expression with a weak
1386 or global symbol can not be relaxed. Break out tests for
1387 relaxable symbol into variable sym_relaxable.
1388 <cases C (MOVI_IMM_64, UNDEF_MOVI) and C (MOVI_IMM_32,
1389 UNDEF_MOVI)>: Break out any PC-relative expression and change
1391 (shmedia_build_Mytes): CSE &operands->operands[j] into variable
1393 <case A_IMMS16>: Fix typo for initial minor relaxation type of
1394 MOVI expansion. If X_op_symbol of the immediate expression is
1395 set, make an expression symbol for the argument to frag_var.
1396 * config/tc-sh.c (MOVI_IMM_32_PCREL, MOVI_IMM_64_PCREL): New
1398 (END): Adjust for new relaxations.
1399 (md_relax_table): Add entries for new relaxations.
1400 2000-12-07 Ben Elliston <bje@redhat.com>
1401 * config/tc-sh64.c (shmedia_parse_reg): Initialize variable len.
1402 2000-12-07 Hans-Peter Nilsson <hpn@cygnus.com>
1403 * config/tc-sh64.c (shmedia_md_convert_frag): Correct all MOVI and
1404 SHORI operand offsets in PT/PTA/PTB expansions.
1405 2000-12-05 Hans-Peter Nilsson <hpn@cygnus.com>
1406 Implement DataLabel semantics.
1407 * config/tc-sh.c (sh_frob_file) [HAVE_SH64]: Call
1408 shmedia_frob_file_before_adjust.
1409 * config/tc-sh64.c [! OBJ_ELF]: Emit #error.
1410 (DATALABEL_SUFFIX): Define.
1411 (shmedia_md_end) <before adjusting STO_SH5_ISA32 symbols>: Walk
1412 symbol list to update "datalabel" symbols to their main symbol
1414 (shmedia_frob_file_before_adjust): New.
1415 (sh64_adjust_symtab): For remaining datalabel symbols, set to
1416 undefined and set STT_DATALABEL.
1417 (sh64_frob_label): Initialize TC symbol field.
1418 (sh64_consume_datalabel): Actually implement semantics. New
1419 parameter operandf, call it instead of expression.
1420 (sh64_exclude_symbol): New.
1421 * config/tc-sh64.h (md_parse_name): Pass on the function operand
1422 to sh64_consume_datalabel.
1423 (tc_symbol_new_hook): Define to tc_frob_symbol.
1424 (TC_SYMFIELD_TYPE): Define to symbolS *.
1425 (tc_frob_symbol): Define to call sh64_exclude_symbol.
1426 2000-12-01 Hans-Peter Nilsson <hpn@cygnus.com>
1427 * config/tc-sh64.c (shmedia_init_reloc): Tweak comment for default
1429 (shmedia_md_assemble): Call dwarf2_generate_asm_lineno if
1430 generating dwarf2 debug information.
1431 2000-11-30 Hans-Peter Nilsson <hpn@cygnus.com>
1432 * config/tc-sh64.c (sh64_target_format): Use elf64-sh64l and
1433 elf64-sh64 for the 64-bit ABI.
1434 * config/tc-sh.c (md_show_usage): Tweak usage output for -abi=*
1436 2000-11-29 Hans-Peter Nilsson <hpn@cygnus.com>
1437 * config/tc-sh.c: Remove conditionalizing on HAVE_SH64 for
1439 2000-11-27 Hans-Peter Nilsson <hpn@cygnus.com>
1440 * config/tc-sh64.c: Tweak comments, formatting and error messages.
1441 (enum sh64_abi_values): New type.
1442 (enum sh64_isa_values): New type.
1443 (sh64_isa_mode): Replace shmedia_mode. All referers changed.
1444 (seen_shcompact_mode, seen_shmedia_mode): Delete.
1445 (sh64_abi): Replace shmedia_64.
1446 (shmedia_md_convert_frag) <C (MOVI_IMM_64, MOVI_64),
1447 C (MOVI_IMM_32, MOVI_32)>: Correct register number handling.
1448 (s_sh64_mode): Check validity for this target.
1449 (sh64_target_format): Initialize defaults for ISA and ABI.
1450 Fallback to old object format if no SH64 ISA or ABI has been
1452 * config/tc-sh.c (md_parse_option): Check combinations for errors.
1453 (sh_elf_final_processing): Change to have EF_SH64_ABI64 for 64-bit
1454 ABI and EF_SH64 for 32-bit ABI, if SH64 options are specified.
1455 * config/tc-sh64.h: Fix typo in comment.
1456 2000-11-25 Hans-Peter Nilsson <hpn@cygnus.com>
1457 * config/tc-sh64.c (shmedia_md_estimate_size_before_relax)
1458 <PT fixups for absolute values>: Size will be longest, not
1460 (shmedia_md_convert_frag): Disable PTB-warning machinery. Correct
1461 all MOVI and SHORI operand offsets in PT/PTA/PTB expansions.
1462 * config/tc-sh.c (parse_reg) [HAVE_SH64]: Add local variables l0
1463 and l1 to hold lowercase of two first characters. Change all
1464 remaining TO_LOWER to tolower.
1465 * config/tc-sh64.c (TO_LOWER): Delete.
1466 (shmedia_find_cooked_opcode): Use tolower, not TO_LOWER.
1467 (md_parse_name): Define.
1468 (sh64_consume_datalabel): Declare.
1469 (DOLLAR_DOT): Define.
1470 * config/tc-sh64.c (shmedia_parse_exp): New.
1471 (sh64_consume_datalabel): New; just ignoring datalabel semantics.
1472 (shmedia_parse_reg): Remove const from src
1474 (shmedia_get_operands): Ditto for args parameter and ptr variable.
1475 (shmedia_md_assemble): Ditto for op_end variable.
1476 (shmedia_get_operand): Ditto for ptr parameter and src variable.
1477 Use shmedia_parse_exp, not parse_exp.
1478 * config/tc-sh64.c (shmedia_parse_reg): Add shmedia_arg_type
1479 parameter. All callers changed.
1480 (shmedia_get_operand): Add shmedia_arg_type parameter. All
1482 (shmedia_parse_reg): Put first two character in local variables.
1483 Use tolower, not TO_LOWER. If no register is found and argtype
1484 indicates a control register, scan shmedia_creg_table
1486 2000-11-24 Hans-Peter Nilsson <hpn@cygnus.com>
1487 * Makefile.am (CPU_TYPES): Add sh64.
1488 (TARGET_CPU_CFILES): Add config/tc-sh64.c.
1489 (TARGET_CPU_HFILES): Add config/tc-sh64.h.
1490 Regenerate dependencies.
1491 * Makefile.in: Regenerate.
1492 * configure.in: Add support for sh64-*-elf*.
1493 * configure: Regenerate.
1494 * config/tc-sh64.h: New.
1495 * config/tc-sh64.c: New.
1496 * config/tc-sh.c (md_pseudo_table) [HAVE_SH64]: New pseudos
1497 .mode, .isa and .uaquad.
1498 [HAVE_SH64] (SH64PCREL16_32, SH64PCREL16_64, SH64PCREL16PT_32,
1499 SH64PCREL16PT_64, MOVI_IMM_32, MOVI_IMM_64): Define.
1500 [HAVE_SH64] (END): Define as 10.
1501 [HAVE_SH64] (UNDEF_SH64PCREL, SH64PCREL16, SH64PCREL32,
1502 SH64PCREL48, SH64PCREL64, UNDEF_MOVI, MOVI_16, MOVI_32, MOVI_48,
1504 [HAVE_SH64] (SH64PCREL16_F, SH64PCREL16_M, SH64PCREL16_LENGTH,
1505 SH64PCREL32_F, SH64PCREL32_M, SH64PCREL32_LENGTH, SH64PCREL48_F,
1506 SH64PCREL48_M, SH64PCREL48_LENGTH, SH64PCREL64_LENGTH,
1507 MOVI_16_LENGTH, MOVI_32_LENGTH, MOVI_48_LENGTH, MOVI_64_LENGTH):
1509 (md_relax_table) [HAVE_SH64]: Provide relaxations for SHmedia.
1510 (md_begin) [HAVE_SH64]: Call shmedia_md_begin.
1511 (parse_reg) [HAVE_SH64]: Parse register names case-insensitive.
1512 (md_assemble) [HAVE_SH64]: Call shmedia_md_assemble if assembling
1513 SHmedia instructions. Handle state-change after switching to
1515 (md_longopts) [HAVE_SH64]: New options --isa=* and --abi=*.
1516 (md_parse_option) [HAVE_SH64]: Parse new options.
1517 (md_show_usage) [HAVE_SH64]: Show usage of new options.
1518 (md_convert_frag) [HAVE_SH64] <default>: Call
1519 shmedia_md_convert_frag instead of abort.
1520 (sh_force_relocation) [HAVE_SH64]: Also force relocation for
1521 BFD_RELOC_SH_SHMEDIA_CODE.
1522 (sh_elf_final_processing) [HAVE_SH64]: Set flags identifying
1523 SHcompact or SHmedia code.
1524 (md_apply_fix) [HAVE_SH64] <default>: Return result from calling
1525 shmedia_md_apply_fix instead of abort.
1526 (md_estimate_size_before_relax) [HAVE_SH64] <default>: Return
1527 result from calling shmedia_md_estimate_size_before_relax instead
1529 (sh_do_align) [HAVE_SH64]: If shmedia_mode, let shmedia_do_align
1531 (tc_gen_reloc) [HAVE_SH64]: For unrecognized relocs, call
1532 shmedia_init_reloc and do nothing more if it returns non-zero.
1533 (sh_finalize) [HAVE_SH64]: Call shmedia_md_end.
1534 * po/POTFILES.in: Regenerate.
1535 * po/gas.pot: Regenerate.
1537 2002-02-06 Alexandre Oliva <aoliva@redhat.com>
1539 * config/tc-sh.c (parse_at): Install the correct version of
1542 * config/tc-sh.c (md_apply_fix3) <BFD_RELOC_32_PLT_PCREL>: Don't
1543 assume fixP->fx_subsy is non-NULL.
1545 2002-02-04 Alexandre Oliva <aoliva@redhat.com>
1547 * config/tc-sh.c (parse_at): Set arg type of @(expr,pc) to
1548 A_DISP_PC_ABS, and adjust it by -4.
1549 (get_specific): Accept A_DISP_PC_ABS where A_DISP_PC is
1551 (build_Mytes): Mark PCRELIMM fix-ups as pc-relative only if
1552 the operand type is not A_DISP_PC_ABS.
1554 2002-02-04 Hans-Peter Nilsson <hp@bitrange.com>
1556 * config/tc-mmix.c (tc_gen_reloc): Don't try and take the value of
1557 common and weak symbols. Handle common and weak symbols as
1558 undefined symbols with regards to GREG handling and merging.
1559 (mmix_frob_file): Ditto.
1561 2002-02-02 Jason Thorpe <thorpej@wasabisystems.com>
1563 * configure.in (hppa-*-netbsd*): New target.
1564 * configure: Regenerate.
1565 * config/tc-hppa.h: Also define WARN_COMMENTS if TE_NetBSD.
1567 2002-02-02 Alan Modra <amodra@bigpond.net.au>
1569 * config/tc-v850.c: Add missing prototypes amd use old-style
1570 function definitions.
1571 (AREA_ZDA, AREA_SDA, AREA_TDA): Delete.
1572 (sdata_section tdata_section, zdata_section, sbss_section,
1573 tbss_section, zbss_section, rosdata_section, rozdata_section,
1574 scommon_section, tcommon_section, zcommon_section,
1575 call_table_data_section, call_table_text_section): Delete.
1576 (v850_sdata, v850_tdata, v850_zdata, v850_sbss, v850_tbss,
1577 v850_zbss, v850_bss, v850_rosdata, v850_rozdata,
1578 v850_call_table_data, v850_call_table_text): Delete.
1579 (struct v850_seg_entry): New.
1580 (v850_seg_table): New.
1581 (SDATA_SECTION TDATA_SECTION, ZDATA_SECTION, SBSS_SECTION,
1582 TBSS_SECTION, ZBSS_SECTION, BSS_SECTION, ROSDATA_SECTION,
1583 ROZDATA_SECTION, SCOMMON_SECTION, TCOMMON_SECTION, ZCOMMON_SECTION,
1584 CALL_TABLE_DATA_SECTION, CALL_TABLE_TEXT_SECTION): Define.
1587 (v850_comm): Use do_v850_seg and v850_seg_table. Simplify
1588 recording of alignment.
1589 (md_pseudo_table): Use v850_seg.
1590 (md_begin): Don't init .call_table_data and .call_table_text here.
1591 Set v850_seg_table bss entry.
1593 2002-02-01 Hans-Peter Nilsson <hp@bitrange.com>
1595 Support on-demand global register allocation by passing on
1596 base-plus-offset relocs to the linker.
1597 * config/tc-mmix.c: Tweak and fix typos in comments.
1598 (allocate_undefined_gregs_in_linker): New variable.
1599 (OPTION_LINKER_ALLOCATED_GREGS): New option macro.
1600 (md_longopts): Add --linker-allocated-gregs.
1601 (md_parse_option) <case 'x'>: Imply --linker-allocated-gregs.
1602 <case OPTION_LINKER_ALLOCATED_GREGS>: New.
1603 (md_show_usage): Update text for -x. Add text for
1604 --linker-allocated-gregs.
1605 (tc_gen_reloc): Derive default value for addend from val and
1606 baddsy. Use addsec and bfd_is_abs_section in more places. Don't
1607 emit error for BFD_RELOC_MMIX_BASE_PLUS_OFFSET without suitable
1608 GREG if allocate_undefined_gregs_in_linker.
1609 * doc/as.texinfo (Overview) <Target MMIX options>: Add
1610 --linker-allocated-gregs.
1611 * doc/c-mmix.texi (MMIX-Opts): Add blurb about
1612 --linker-allocated-gregs. Mention that it's implied by -x.
1613 (MMIX-Pseudos) <GREG>: Mention when and how a GREG can be omitted.
1614 (MMIX-mmixal): Clarify dated comparison and location of MMIXware.
1616 * config/tc-mmix.h (md_parse_name): Use ISUPPER, not isupper.
1618 2002-02-01 Alan Modra <amodra@bigpond.net.au>
1620 * Makefile.am: Run "make dep-am"
1621 * Makefile.in: Regenerate.
1623 2002-01-31 Ivan Guzvinec <ivang@opencores.org>
1625 * config/tc-or32.c: New file.
1626 * config/tc-or32.h: New file.
1627 * configure.in: Add support for or32 targets.
1628 * configure: Regenerate.
1629 * config/obj-coff.c: Add support for or32 targets.
1630 * config/obj-coff.h: Add support for or32 targets.
1631 * Makefile.am: Add support for or32 targets.
1632 * Makefile.in: Regenerate.
1633 * NEWS: Mention support for OpenRISC.
1634 * doc/Makefile.in: Regenerate.
1635 * po/POTFILES.in: Regenerate.
1636 * po/gas.pot: Regenerate.
1638 2002-01-30 Richard Sandiford <rsandifo@redhat.com>
1640 * config/tc-sh.c (parse_reg): Fix end-of-word check for is, ix, iy
1643 2002-01-29 Chris Demetriou <cgd@broadcom.com>
1645 * config/tc-mips.c (tc_gen_reloc): Arrange for
1646 BFD_RELOC_PCREL_HI16_S relocations to be output relative to
1647 their LO16 parts, even for ELF.
1649 2002-01-29 Daniel Jacobowitz <drow@mvista.com>
1651 * config/tc-i386.c: Protect definitions of true and false
1654 2002-01-28 Jakub Jelinek <jakub@redhat.com>
1656 * config/obj-elf.c (elf_frob_file_before_adjust): Remove symbols
1657 made because of .weak, if they are neither defined nor used in any
1660 2002-01-27 Daniel Jacobowitz <drow@mvista.com>
1662 * configure: Regenerated.
1664 2002-01-26 Hans-Peter Nilsson <hp@bitrange.com>
1666 * doc/Makefile.am (install): Depend on install-info.
1667 * doc/Makefile.in: Regenerate.
1669 2002-01-26 Nick Clifton <nick@redhat.com>
1671 * po/fr.po: Updated version
1673 2002-01-24 Kazu Hirata <kazu@hxi.com>
1675 * config/tc-h8300.c (check_operand): Don't print a warning
1676 when a valid 24-bit address is given to a 16-bit address
1679 2002-01-24 Alexandre Oliva <aoliva@redhat.com>
1681 * config/tc-sh.c (sh_elf_suffix): Removed.
1682 (sh_PIC_related_p, sh_check_fixup, sh_cons_fix_new,
1683 sh_end_of_match, sh_parse_name): New functions.
1684 (sh_elf_cons): Simplify.
1685 (parse_exp): Reject misplaced PIC operands.
1686 (md_undefined_symbol): Simplify.
1687 (sh_fix_adjustable): Let @GOTOFF be adjusted.
1688 (md_apply_fix3): Write @PLT and @GOTOFF addends in place.
1689 (tc_gen_reloc): Move fixp subsy absolute value into addnumber.
1690 Complain if subsy remains at the end.
1691 * config/tc-sh.h (sh_parse_name, sh_cons_fix_new): Declare.
1692 (md_parse_name, TC_CONS_FIX_NEW, O_PIC_reloc): Define.
1694 2002-01-22 Alexandre Oliva <aoliva@redhat.com>
1696 * config/tc-mn10300.c (xr_registers): Move `pc'...
1697 (other_registers): ... here.
1699 2002-01-22 Alan Modra <amodra@bigpond.net.au>
1701 * Makefile.am: Run "make dep-am".
1702 * Makefile.in: Regenerate.
1703 * gas/po/POTFILES.in: Regenerate.
1705 2002-01-21 DJ Delorie <dj@redhat.com>
1707 * config/obj-coff.c (obj_coff_init_stab_section): Make the
1708 stabstr_name allocation permanent, as it will be referenced from
1711 2002-01-21 Jason Thorpe <thorpej@wasabisystems.com>
1713 * configure.in (ia64-*-netbsd*): New target.
1714 * configure: Regenerate.
1716 2002-01-21 Hans-Peter Nilsson <hp@bitrange.com>
1718 * doc/as.texinfo (Overview) <Target ARM options>: Add missing {}
1720 <Detailed description, ARM options>: Ditto.
1721 * doc/c-arm.texi (ARM Options): Ditto.
1723 2002-01-18 Richard Earnshaw <rearnsha@arm.com>
1725 * tc-arm.c (do_xsc_mia, do_xsc_mar, do_xsc_mra): Renamed from
1726 do_mia, do_mar and do_mra respectively.
1727 (do_mav_*): Renamed from do_c_*.
1728 (mav_reg_required_here, mav_parse_offset): Renamed from
1729 cirrus_reg_required_here and cirrus_parse_offset respectively.
1730 (MAV_MODE?): Renamed from CIRRUS_MODE?.
1732 2002-01-18 Richard Earnshaw <rearnsha@arm.com>
1733 Keith Walker <keith.walker@arm.com>
1735 * tc-arm.c (ARM_EXT_V5J, ARM_ARCH_V5TEJ): Define.
1736 (insns): Add pattern for bxj instruction.
1737 (do_bxj): New function.
1738 (arm_cpus): Add arm926ej.
1739 (arm_archs): Add armv5tej.
1741 2002-01-18 Richard Earnshaw <rearnsha@arm.com>
1743 * doc/c-arm.texi: Add new fpe options to list of supported flags.
1745 2002-01-19 Keith Walker <keith.walker@arm.com>
1747 * tc-arm.c (arm_fpus): Add fpe2 and fpe3.
1749 2002-01-18 Richard Earnshaw <rearnsha@arm.com>
1751 * NEWS: Mention new ARM command-line options and VFP support.
1753 * config/tc-arm.c (ARM_CEXT_XSCALE): Replaces ARM_EXT_XSCALE. All
1755 (ARM_CEXT_MAVERICK): Similarly.
1756 (ARM_ANY): Now means any core instruction.
1757 (CPU_DEFAULT): Default to ARM_ANY.
1758 (uses_apcs_26, atcps, support_interwork, uses_apcs_float)
1759 (pic_code): Declare for all object types. Make type int.
1760 (legacy_cpu, legacy_fpu, mcpu_cpu_opt, mcpu_fpu_opt, march_cpu_opt)
1761 (march_fpu_opt, mfpu_opt): Declare.
1762 (md_longopts): Tidy up conditional definitions.
1763 (arm_opts, arm_cpus, arm_archs, arm_fpus, arm_extensions)
1764 (arm_long_opts): New tables.
1765 (arm_parse_cpu, arm_parse_arch, arm_parse_fpu): New functions.
1766 (arm_parse_extension): New function.
1767 (md_parse_option): Rewrite using new table-driven system.
1768 (md_show_usage): Use new table-driven system.
1769 (md_begin): Calculate cpu_variant from command line option data.
1770 * doc/as.texinfo (ARM ISA options): Docuement new ARM-specific
1771 command-line options.
1772 * doc/c-arm.texi: Likewise.
1774 2002-01-18 Andreas Jaeger <aj@suse.de>
1776 * as.c (parse_args): Update year.
1778 2002-01-17 Timothy Wall <twall@alum.mit.edu>
1780 * config/tc-tic54x.c (encode_address): Add a more informative
1781 warning about incorrect syntax.
1783 2002-01-17 Nick Clifton <nickc@cambridge.redhat.com>
1785 * po/gas.pot: Regenerate.
1787 2002-01-17 Nick Clifton <nickc@cambridge.redhat.com>
1789 * macro.c (get_any_string): Add no-c-format comment to prevent
1790 confusion when translating string.
1791 * gasp.c (get_any_string): Add no-c-format comment to prevent
1792 confusion when translating string.
1794 2002-01-16 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1795 Johannes Stezenbach <js@convergence.de>
1796 * config/tc-mips.c (percent_op): Ensure longer percent_op's are
1797 matched before the shorter ones.
1798 (my_getSmallParser): Fix handling of nested parentheses in
1799 percent_op's. Code cleanup.
1800 (my_getPercentOp): New function, code from my_getSmallParser.
1801 (my_getSmallExpression): Fix handling of closing parentheses.
1802 Code cleanup. Better comments.
1804 2002-01-16 Nick Clifton <nickc@redhat.com>
1806 po/tr.po: New file: Turkish translation.
1807 configure.in (LINGUAS): Add "tr".
1808 configure: Regenerate.
1810 2002-01-15 Richard Earnshaw <rearnsha@arm.com>
1812 Support for VFP instructions
1813 * tc-arm.c (CP_WB_OK, CP_NO_WB): New defines.
1814 (cp_address_required_here): New argument wb_ok. When false, do not
1815 accept write-back forms of addressing. Change all callers.
1816 (FPU_VFP_EXT_NONE, FPU_VFP_EXT_V1xD, FPU_VFP_VFP_V1)
1817 (FPU_VFP_EXT_V2): Define.
1818 (FPU_ARCH_VFP, FPU_ARCH_VFP_V1xD, FPU_ARCH_VFP_V1, FPU_ARCH_VFP_V2):
1819 Define in terms of above.
1820 (vfp_dp_reg_pos, vfp_sp_reg_pos, vfp_ldstm_type): New enums.
1821 (vfp_reg): New struct.
1822 (vfp_regs): New array of registers.
1823 (insns): Add VFP instructions.
1824 (sn_table): New array of VFP single-precision register names.
1825 (dn_table): New array of VFP double-precision register names.
1826 (all_reg_maps): Add the new register tables.
1827 (arm_reg_type): Add new values for above. Increase RET_TYPE_MAX.
1828 (vfp_sp_reg_required_here, vfp_dp_reg_required_here, do_vfp_sp_monadic)
1829 (do_vfp_dp_monadic, do_vfp_sp_dyadic, do_vfp_dp_dyadic)
1830 (do_vfp_reg_from_sp, do_vfp_sp_reg2, do_vfp_sp_from_reg)
1831 (do_vfp_reg_from_dp, do_vfp_reg2_from_dp, do_vfp_dp_from_reg)
1832 (do_vfp_dp_from_reg2, vfp_psr_parse, vfp_psr_required_here)
1833 (do_vfp_reg_from_ctrl, do_vfp_ctrl_from_reg, do_vfp_sp_ldst)
1834 (do_vfp_dp_ldst, vfp_sp_reg_list, vfp_dp_reg_list, vfp_sp_ldstm)
1835 (vfp_dp_ldstm, do_vfp_sp_ldstmia, do_vfp_sp_ldstmdb, do_vfp_ldstmia)
1836 (do_vfp_dp_ldstmdb, do_vfp_xp_ldstmia, do_vfp_xp_ldstmdb)
1837 (do_vfp_sp_compare_z, do_vfp_dp_compare_z, do_vfp_dp_sp_cvt)
1838 (do_vfp_sp_dp_cvt): New functions.
1839 (md_begin): Set soft-float flag for appropriate VFP work.
1840 (md_atof): Handle VFP-format doubles.
1841 (md_parse_option): Handle VFP command-line options.
1842 (md_show_usage): Display VFP command-line options.
1844 2002-01-15 Richard Earnshaw <rearnsha@arm.com>
1846 * tc-arm.c (md_parse_option): Tidy up setting of cpu_variant for
1847 various command line options.
1849 2002-01-15 Nick Clifton <nickc@cambridge.redhat.com>
1851 * config/tc-xstormy16.c: (xstormy16_fix_adjustable): Do not fix
1853 (xstormy16_md_apply_fix3): Do not return a value.
1855 2002-01-14 Richard Earnshaw <rearnsha@arm.com>
1857 * tc-arm.c (md_longopts): On targets that aren't bi-endian, support
1858 the -EL/-EB option that matches the target's endianness.
1859 (md_parse_option): Likewise.
1861 2002-01-14 Richard Earnshaw <rearnsha@arm.com>
1863 * tc-arm.c (md_longopts): Fix misplaced #endif -- the -oabi option
1864 is not dependent on ARM_BI_ENDIAN.
1866 2002-01-14 Richard Earnshaw <rearnsha@arm.com>
1868 * tc-arm.c (all error messages): Normalize capitalization of messages.
1870 * tc-arm.h (md_operand): Delete define.
1871 * tc-arm.c (in_my_get_expression): New static variable.
1872 (my_get_expression): Set and clear it.
1873 (md_operand): New function. If called from my_get_expression
1874 put the error in inst.error.
1875 (output_inst): Now takes argument of instruction being assembled.
1876 Print it out with any error message.
1877 (do_ldst, do_ldstv4, thumb_load_store): Fault attempt to use a store
1879 (end_of_line): Don't update inst.error if it is already set.
1881 2002-01-11 Richard Earnshaw <rearnsha@arm.com>
1883 * tc-arm.c ((do_ldst): Fix handling an immediate expression pseudo
1884 op that can be translated into a mvn instruction.
1886 2002-01-11 Steve Ellcey <sje@cup.hp.com>
1888 * gas/config/tc-ia64.h (MD_FLAGS_DEFAULT): New Macro for
1889 setting default md.flags.
1890 (SHT_INIT_ARRAY): New elf special section used by HP-UX.
1891 (SHT_FINI_ARRAY): New elf special section used by HP-UX.
1892 * gas/config/tc-ia64.c (setup_unwind_header): Add support
1893 for 32 bit unwind info blocks.
1894 (generate_unwind_image): Add support for different types
1895 of unwind images (32 bits and/or big-endian).
1896 (ia64_init): Use MD_FLAGS_DEFAULT to set md.flags.
1897 (ia64_target_format): Add support for hpux target formats.
1898 (ia64_gen_real_reloc_type): Add support for FUNC_IPLT_RELOC.
1899 (ia64_elf_section_type): Add support for SHT_INIT_ARRAY and
1900 SHT_FINI_ARRAY elf section types.
1902 2002-01-10 Richard Earnshaw <rearnsha@arm.com>
1904 * tc-arm.c (struct reg_entry): Move before prototypes.
1905 (int_register, cp_register, fp_register): Delete.
1906 (reg_table): Delete. Replaced with ...
1907 (rn_table, cp_table, cn_table, fn_table, mav_mvf_table)
1908 (mav_mvd_table, mav_mvfx_table, mav_mvdx_table, mav_mvax_table)
1909 (mav_dspsc_table): ... one table per register set.
1910 (arm_reg_hsh): Delete.
1911 (struct reg_map): New structure.
1912 (all_reg_maps): New array.
1913 (enum arm_reg_type): New enums.
1914 (build_reg_hsh): New function.
1915 (insert_reg_alias): Use hash table passed by caller. Adjust all
1917 (create_register_alias): New function, split out from ...
1918 (md_assemble): ... here.
1919 (md_begin): Build new register hash tables.
1920 (arm_reg_parse): New argument for the hash table to search. Adjust all
1922 (arm_reg_parse_any): New function.
1923 (co_proc_number): Look up the processor number in the processor hash
1925 (cirrus_regtype): Delete.
1926 (cirrus_register, cirrus_mvf_register, cirrus_mvd_register)
1927 (cirrus_mvfx_register, cirrus_mvdx_register, cirrus_mvax_register)
1928 (ARM_EXT_MAVERICKsc_register): Delete.
1929 (do_c_binops_1, do_c_binops_2, do_c_binops_3): Delete.
1930 (do_c_binops_1[a-o], do_c_binops_2[a-c], do_c_binops_3[a-d]): New
1932 (do_c_triple_4, do_c_triple_5): Delete.
1933 (do_c_triple_4[ab], do_c_triple_5[a-h]): New functions.
1934 (do_c_quad_6): Delete.
1935 (do_c_quad_6[ab]): New functions.
1936 (do_c_binops, do_c_triple, do_c_quad, do_c_shift, do_c_ldst): Rework
1937 arguments to use new register parsing methods.
1938 (cirrus_reg_required_here): Likewise.
1939 (insns): Reclassify cirrus maverick worker functions.
1940 (cirrus_valid_reg): Delete.
1942 2002-01-07 Jason Thorpe <thorpej@wasabisystems.com>
1944 * configure.in (sh*le): Set cpu_type=sh and endian=little.
1945 (sh*-*-netbsdelf*): New target.
1946 * configure: Regenerate.
1947 * tc-sh.h: Update copyright years.
1948 (TARGET_FORMAT): Add version for TE_NetBSD.
1950 2002-01-07 Nick Clifton <nickc@cambridge.redhat.com>
1952 * read.c (emit_expr): Do not allow 'size' or 'nbytes' to go
1955 2002-01-06 Alan Modra <amodra@bigpond.net.au>
1957 * config/tc-m68k.h (md_prepare_relax_scan): Rewrite.
1958 * config/tc-m68k.c (md_relax_table): Add md_prepare_relax_scan comment.
1960 2002-01-05 Daniel Jacobowitz <drow@mvista.com>
1962 * tc-mips.c (mips_cprestore_valid): New flag.
1963 (mips_frame_reg_valid): New flag.
1964 (macro) [M_JAL_2]: Check both flags.
1965 [M_JAL_A]: Likewise.
1966 (s_cprestore): Set mips_cprestore_valid.
1967 (tc_get_register): If setting mips_frame_reg, set
1968 mips_frame_reg_valid and clear mips_cprestore_valid.
1969 (s_mips_ent): Clear both flags.
1970 (s_mips_end): Clear both flags.
1972 2002-01-05 Alan Modra <amodra@bigpond.net.au>
1974 * write.c (write_object_file): Make use of bfd_section_list_remove.
1975 * config/obj-ecoff.c (ecoff_frob_file): Likewise.
1976 * config/tc-mmix.c (mmix_frob_file): Likewise.
1978 2002-01-04 Jason Thorpe <thorpej@wasabisystems.com>
1980 * configure.in (i386-*-netbsdelf*): Collapse target into...
1981 (i386-*-netbsd*): ...this. Add support for x86-64.
1982 * configure: Regenerated.
1984 2002-01-03 matthew green <mrg@redhat.com>
1986 * config/tc-ppc.c (md_parse_option): BookE is not Motorola specific.
1988 2002-01-02 Nick Clifton <nickc@cambridge.redhat.com>
1990 * config/tc-m68k.c (md_estimate_size_before_relax): Test for a
1993 For older changes see ChangeLog-0001
1999 version-control: never
projects / deliverable / binutils-gdb.git / blob