1 2011-11-14 Uros Bizjak <ubizjak@gmail.com>
3 * config/i386/sse.md (round<mode>2): Use register_operand for
5 (round<mode>2_sfix): Ditto.
6 (round<mode>2_vec_pack_sfix): Ditto for operand 1 and operand 2.
8 2011-11-14 Nathan Sidwell <nathan@acm.org>
10 * gcov.c (canonicalize_name): Protect use of S_ISLNK.
12 2011-11-14 Jan Hubicka <jh@suse.cz>
15 * config/i386/i386.c (core cost model): Correct pasto.
17 2011-11-14 Uros Bizjak <ubizjak@gmail.com>
19 * config/i386/sse.md (round<mode>2_sfix): New expander.
20 (round<mode>2_vec_pack_sfix): Ditto.
21 (<sse4_1>_round<ssemodesuffix>_sfix<avxsizesuffix>): Ditto.
22 (<sse4_1>_round<ssemodesuffix>_vec_pack_sfix<avxsizesuffix>): Ditto.
23 * config/i386/builtin-types.def (V4SI_FTYPE_V4SF_ROUND,
24 V8SI_FTYPE_V8SF_ROUND, V4SI_FTYPE_V2DF_V2DF_ROUND,
25 V8SI_FTYPE_V4DF_V4DF_ROUND): New builtin types.
26 * config/i386/i386.c (ix86_builtins): Add
27 IX86_BUILTIN_{FLOORPD,CEILPD,ROUNDPD_AZ}_VEC_PACK_SFIX{,256} and
28 IX86_BUILTIN_{FLOORPS,CEILPS,ROUNDPS_AZ}_SFIX{,256} defines.
29 (bdesc_args): Add __builtin_ia32_{floorpd,ceilpd}_vec_pack_sfix{,256},
30 __builtin_ia32_roundpd_az_vec_pack_sfix{,256},
31 __builtin_ia32_{floorps,ceilps}_sfix{,256}and
32 __builtin_ia32_roundps_az_sfix{,256} descriptions.
33 (ix86_expand_sse_round_vec_pack_sfix): New.
34 (ix86_expand_args_builtin): Handle V4SI_FTYPE_V4SF_ROUND,
35 V8SI_FTYPE_V8SF_ROUND, V4SI_FTYPE_V2DF_V2DF_ROUND and
36 V8SI_FTYPE_V4DF_V4DF_ROUND types. Check last argument of
37 CODE_FOR_sse4_1_roundpd_vec_pack_sfix, CODE_FOR_sse4_1_roundps_sfix,
38 CODE_FOR_avx_roundpd_vec_pack_sfix256 and CODE_FOR_avx_roundps_sfix256.
39 (ix86_builtin_vectorized_function): Handle
40 BUILT_IN_{I,L,LL}FLOOR{,F}, BUILT_IN_{I,L,LL}CEIL{,F} and
41 BUILT_IN_{I,L,LL}ROUND{,F}
43 2011-11-14 Jan Hubicka <jh@suse.cz>
46 * cgraph.h (varpool_externally_visible_p): Declare.
47 (varpool_add_new_variable): Declare.
48 * tree-emultls.c (get_emutls_init_templ_addr,
49 new_emutls_decl): Use varpool_add_new_variable.
50 * ipa.c (varpool_externally_visible_p): Export.
51 * varpool.c (varpool_add_new_variable): New function.
53 2011-11-14 Zolotukhin Michael <michael.v.zolotukhin@gmail.com>
54 Jan Hubicka <jh@suse.cz>
56 * config/i386/i386.h (processor_costs): Add second dimension to
58 * config/i386/i386.c (cost models): Initialize second dimension of
60 (core_cost): New costs based on generic64 costs with updated stringop
62 (promote_duplicated_reg): Add support for vector modes, add
64 (promote_duplicated_reg_to_size): Likewise.
65 (processor_target): Set core costs for core variants.
66 (expand_set_or_movmem_via_loop_with_iter): New function.
67 (expand_set_or_movmem_via_loop): Enable reuse of the same iters in
68 different loops, produced by this function.
69 (emit_strset): New function.
70 (expand_movmem_epilogue): Add epilogue generation for bigger sizes,
71 use SSE-moves where possible.
72 (expand_setmem_epilogue): Likewise.
73 (expand_movmem_prologue): Likewise for prologue.
74 (expand_setmem_prologue): Likewise.
75 (expand_constant_movmem_prologue): Likewise.
76 (expand_constant_setmem_prologue): Likewise.
77 (decide_alg): Add new argument align_unknown. Fix algorithm of
78 strategy selection if TARGET_INLINE_ALL_STRINGOPS is set; Skip sse_loop
79 (decide_alignment): Update desired alignment according to chosen move
81 (ix86_expand_movmem): Change unrolled_loop strategy to use SSE-moves.
82 (ix86_expand_setmem): Likewise.
83 (ix86_slow_unaligned_access): Implementation of new hook
84 slow_unaligned_access.
85 * config/i386/i386.md (strset): Enable half-SSE moves.
86 * config/i386/sse.md (vec_dupv4si): Add expand for vec_dupv4si.
87 (vec_dupv2di): Add expand for vec_dupv2di.
89 2011-11-14 Dimitrios Apostolou <jimis@gmx.net>
93 * config/elfos.h (TARGET_ASM_INTERNAL_LABEL): Remove.
94 * varasm.c (default_elf_internal_label): Remove.
96 2011-11-14 Kai Tietz <ktietz@redhat.com>
98 * gcov.c (generate_results): Add missing semicolon and correct indent.
100 2011-11-14 Ira Rosen <ira.rosen@linaro.org>
103 * tree-vect-stmts.c (vectorizable_condition): Initialize comp_vectype.
105 2011-11-14 Mingjie Xing <mingjie.xing@gmail.com>
107 * doc/invoke.texi (Wunused-result): Change @pxref{Variable Attributes}
108 to @pxref{Function Attributes}.
110 2011-11-13 Oleg Endo <oleg.endo@t-online.de>
113 * config/sh/sh.h (IS_LITTLE_ENDIAN_OPTION, UNSUPPORTED_SH2A):
115 (DRIVER_SELF_SPECS): Use new macros to filter out unsupported options
116 taking the default configuration into account.
118 2011-11-13 Jonathan Wakely <jwakely.gcc@gmail.com>
120 * doc/install.texi (Specific): Fix anchor for
121 x86_64-*-solaris2.1[0-9]*.
123 2011-11-13 Nathan Sidwell <nathan@acm.org>
125 * gcov.c (source_prefix, source_length): New globals.
126 (flag_relative_only): Likewise.
127 (print_usage, options, process_args): Update.
128 (generate_results): Use coverage.name, check
129 flag_relative_only. Adjust messages.
130 (find_source): Check source_prefix.
131 (output_lines): Use coverage.name, adjust messages.
132 * doc/gcov.texi (Invoking Gcov): Document new options.
134 2011-11-12 Jason Merrill <jason@redhat.com>
137 * gimplify.c (gimplify_target_expr): Add a clobber to the cleanup.
138 (gimplify_modify_expr): Don't try to simplify it.
140 2011-11-12 Dimitrios Apostolou <jimis@gmx.net>
143 * config/elfos.h (STRING_ASM_OP): Define.
144 * varasm.c (default_elf_asm_output_limited_string): Use it.
147 * gcc/configure.ac: Check for stpcpy declaration.
148 * system.h: Declare it if not.
149 * config.in, configure: Regenerate.
151 2011-11-12 Richard Henderson <rth@redhat.com>
153 * config/rs6000/rs6000.md (fix_trunc<SFDF>si2_stfiwx): Use
154 nonimmediate_operand for the destination.
155 (fixuns_trunc<SFDF>si2_stfiwx): Likewise.
157 * builtins.c (expand_builtin_compare_and_swap): If target is const0,
158 don't pass the target to expand_atomic_compare_and_swap.
159 (expand_builtin_atomic_compare_exchange): Likewise.
161 * config/sh/linux.h (TARGET_INIT_LIBFUNCS): New.
162 * config/sh/sh.c (sh_init_sync_libfuncs): New.
164 * config/pa/pa-linux.h (TARGET_SYNC_LIBCALL): New.
165 * config/pa/pa.h (TARGET_SYNC_LIBCALL): New default.
166 * config/pa/pa.c (pa_init_libfuncs): Use init_sync_libfuncs.
168 2011-11-12 Iain Sandoe <iains@gcc.gnu.org>
171 * config/rs6000/rs6000.c (rs6000_legitimize_reload_address):
172 Only expand a symbol ref. into an access when the entity is defined
175 2011-11-12 Jan Hubicka <jh@suse.cz>
178 * cgraphunit.c (output_weakrefs): Output really only weakrefs.
180 2011-11-12 Nathan Sidwell <nathan@acm.org>
182 * gcov.c (struct name_map): New.
183 (names, n_names, a_names): New global vars.
184 (print_usage): Adjust usage.
185 (generate_results): Canonicalize main file name.
186 (release_structures): Adjust.
187 (name_search, name_sort): New callbacks.
188 (find_source): Look for and create a canonical name.
189 (canonicalize_name): New.
190 (make_gcov_file_name): Reimplement and fix mangling.
192 * doc/gcov.texi: Update documentation about path preservation.
194 2011-11-11 David S. Miller <davem@davemloft.net>
196 * config/sparc/linux.h (ASM_GENERATE_INTERNAL_LABEL): Delete.
197 * config/sparc/linux64.h (ASM_GENERATE_INTERNAL_LABEL): Delete.
199 2011-11-11 Jakub Jelinek <jakub@redhat.com>
201 * config/i386/i386-protos.h (ix86_maybe_emit_epilogue_vzeroupper):
203 * config/i386/i386.c (ix86_maybe_emit_epilogue_vzeroupper): New
205 (ix86_expand_epilogue): Use it.
206 * config/i386/i386.md (return, simple_return): Call it in the
209 PR tree-optimization/51091
210 * tree-stdarg.c (execute_optimize_stdarg): Ignore TREE_CLOBBER_P
211 rhs also in the va_list_simple_ptr case.
213 PR tree-optimization/51058
214 * tree-vect-slp.c (vect_remove_slp_scalar_calls): New function.
215 (vect_schedule_slp): Call it.
216 * tree-vect-stmts.c (vectorizable_call): If slp_node != NULL,
217 don't replace scalar calls with setting of their lhs to zero here.
219 PR tree-optimization/51074
220 * fold-const.c (vec_cst_ctor_to_array, fold_vec_perm): New functions.
221 (fold_binary_loc): Handle VEC_EXTRACT_EVEN_EXPR,
222 VEC_EXTRACT_ODD_EXPR, VEC_INTERLEAVE_HIGH_EXPR and
223 VEC_INTERLEAVE_LOW_EXPR with VECTOR_CST or CONSTRUCTOR operands.
224 (fold_ternary_loc): Handle VEC_PERM_EXPR with VECTOR_CST or
225 CONSTRUCTOR operands.
226 * tree-ssa-propagate.c (valid_gimple_rhs_p): Handle ternary
228 * tree-vect-generic.c (lower_vec_perm): Mask sel_int elements
229 to 0 .. 2 * elements - 1.
231 2011-11-11 Richard Henderson <rth@redhat.com>
233 * gimple-pretty-print.c (dump_gimple_omp_atomic_load): Dump needed.
234 (dump_gimple_omp_atomic_store): Likewise.
235 * optabs.c (can_atomic_exchange_p): New.
236 * optabs.h (can_atomic_exchange_p): Declare.
237 * omp-low.c (expand_omp_atomic_load): Implement.
238 (expand_omp_atomic_store): Likewise.
239 (expand_omp_atomic): Update for new arguments to load/store.
241 2011-11-11 Eric Botcazou <ebotcazou@adacore.com>
243 * cprop.c: Adjust outdated comments throughout.
244 (hash_scan_set): Rename PAT parameter into SET.
245 (cprop_pavloc): Rename into...
246 (cprop_avloc): ...this.
247 (cprop_absaltered): Rename into...
248 (cprop_kill): ...this.
249 (alloc_cprop_mem): Adjust for above renaming.
250 (free_cprop_mem): Likewise.
251 (compute_cprop_data): Likewise.
252 (compute_local_properties): Rename TRANSP parameter into KILL and
253 adjust throughout. Rework comments.
254 (try_replace_reg): Fix long line.
255 (cprop_jump): Likewise.
256 (constprop_register): Fix prototype and take INSN last.
257 (cprop_insn): Adjust calls to above function. Fix long lines.
258 (bypass_block): Likewise.
259 (one_cprop_pass): Likewise.
261 2011-11-11 Uros Bizjak <ubizjak@gmail.com>
263 * config/alpha/elf.h (ELF_ASCII_ESCAPES): Rename from ESCAPES.
264 (ELF_STRING_LIMIT): Rename from STRING_LIMIT.
266 2011-11-11 Jason Merrill <jason@redhat.com>
268 * varasm.c (default_elf_asm_output_limited_string): #ifdef
270 (default_elf_asm_output_ascii): Likewise.
272 2011-11-11 Uros Bizjak <ubizjak@gmail.com>
275 * config/i386/constraints.md (j): New address constraint.
276 * config/i386/predicates.md (lea_address_operand): Redefine as
278 * config/i386/i386.md (*lea_3_zext): Use "j" constraint for operand 1.
279 (*lea_4_zext): Ditto.
281 2011-11-11 David S. Miller <davem@davemloft.net>
284 2011-11-05 David S. Miller <davem@davemloft.net>
286 * config/sparc/sparc.md (UNSPEC_SHORT_LOAD): New unspec.
287 (zero-extend_v8qi_vis, zero_extend_v4hi_vis): New expanders.
288 (*zero_extend_v8qi_<P:mode>_insn,
289 *zero_extend_v4hi_<P:mode>_insn): New insns.
290 * config/sparc/sparc.c (vector_init_move_words,
291 vector_init_prepare_elts, sparc_expand_vector_init_vis2,
292 sparc_expand_vector_init_vis1): New functions.
293 (vector_init_bshuffle): Rewrite to handle more cases and make use
294 of locs[] array prepared by vector_init_prepare_elts.
295 (vector_init_fpmerge, vector_init_faligndata): Delete.
296 (sparc_expand_vector_init): Rewrite using new infrastructure.
298 2011-11-11 Jakub Jelinek <jakub@redhat.com>
300 * opts-common.c (generate_canonical_option): Free opt_text
301 it it has been allocated here and not stored anywhere.
303 * tree-vect-slp.c (vect_free_slp_tree): Also free SLP_TREE_CHILDREN
305 (vect_create_new_slp_node): Don't allocate node before checking stmt
307 (vect_free_oprnd_info): Remove FREE_DEF_STMTS argument, always
308 free def_stmts vectors and additionally free oprnd_info.
309 (vect_build_slp_tree): Adjust callers. Call it even if
310 stop_recursion. If vect_create_new_slp_node or
311 vect_build_slp_tree fails, properly handle freeing memory.
312 If it succeeded, clear def_stmts in oprnd_info.
314 2011-11-11 Dimitrios Apostolou <jimis@gmx.net>
316 * defaults.h (ASM_OUTPUT_LABEL, ASM_OUTPUT_INTERNAL_LABEL): Expand on
317 multiple lines for readability.
318 * output.h (fprint_whex, fprint_w, fprint_ul, sprint_ul): Declare.
319 (ASM_OUTPUT_LABELREF): Call fputs instead of asm_fprintf.
320 * dwarf2asm.c (dw2_assemble_integer, dw2_asm_output_data,
321 dw2_asm_data_uleb128, dw2_asm_delta_uleb128,
322 dw2_asm_delta_sleb128): Convert fprintf calls to the new functions.
323 * dwarf2out.c (dwarf2out_source_line): Call fprint_ul instead of
325 * final.c (fprint_whex, fprint_w, fprint_ul, sprint_ul): New functions
326 serving as fast replacements for fprintf integer to string conversions.
327 (sprint_ul_rev): Internal helper for the above.
328 (output_addr_const) <CONST_INT>: Don't use fprintf.
329 * varasm.c: Fix some whitespace inconsistencies.
330 (default_elf_asm_output_limited_string): New.
331 (default_elf_asm_output_ascii): Likewise.
332 (default_elf_internal_label): Likewise.
333 * config/elfos.h (ASM_GENERATE_INTERNAL_LABEL): Use sprint_ul and
334 stpcpy instead of sprintf.
335 (TARGET_ASM_INTERNAL_LABEL): Define as default_elf_internal_label.
336 (ELF_ASCII_ESCAPES, ELF_STRING_LIMIT): Rename from old ESCAPES and
338 (ASM_OUTPUT_LIMITED_STRING, ASM_OUTPUT_ASCII): Call functions that
339 provide the same functionality.
340 * config/i386/i386.c (print_reg): Call fprint_ul instead of fprintf.
342 2011-11-10 Andrew MacLeod <amacleod@redhat.com>
344 * doc/extend.texi: Document __atomic_test_and_set and __atomic_clear.
346 2011-11-10 Andrew MacLeod <amacleod@redhat.com>
349 * builtins.c (expand_builtin_atomic_clear): New. Expand atomic_clear.
350 (expand_builtin_atomic_test_and_set): New. Expand atomic test_and_set.
351 (expand_builtin): Add cases for test_and_set and clear.
352 * sync-builtins.def (BUILT_IN_ATOMIC_TEST_AND_SET): New.
353 (BUILT_IN_ATOMIC_CLEAR): New.
355 2011-11-10 Roberto Agostino Vitillo <ravitillo@lbl.gov>
358 * dwarf2out.c (set_cur_line_info_table): Restore the last is_stmt
359 value in the current line table.
361 2011-11-10 Nathan Sidwell <nathan@acm.org>
363 * gcov.c (struct function_info): Make src an index, not a pointer.
364 (struct source_info): Remove index and next source fields.
365 (fn_end): New static var.
366 (sources_index): Remove.
367 (sources): Now a pointer to an array, not a list.
368 (n_sources, a_sources): New.
369 (process_file): Adjust for changes to read_graph_file. Insert
370 functions into source lists and check line numbers here.
371 (generate_results): Only allocate lines for sources with
372 contents. Adjust for source array.
373 (release_structures): Likewise.
374 (find_source): Return source index, adjust for source array.
375 (read_graph_file): Return function list. Don't insert into source
377 (read_count_file): Take list of functions.
378 (solve_flow_graph): Reverse the arc lists here.
379 (add_line_counts): Adjust for source array.
381 2011-11-10 Jakub Jelinek <jakub@redhat.com>
384 * tree-object-size.c (addr_object_size): Check TREE_CODE of
385 MEM_REF's operand rather than code of the MEM_REF itself.
387 2011-11-10 Andrew MacLeod <amacleod@redhat.com>
389 PR rtl-optimization/51040
390 * optabs.c (expand_atomic_fetch_op): Patchup code for NAND should
391 be AND followed by NOT.
392 * builtins.c (expand_builtin_atomic_fetch_op): Patchup code for NAND
393 should be AND followed by NOT.
395 2011-11-10 Jakub Jelinek <jakub@redhat.com>
397 * vec.h (VEC_BASE): If base is at offset 0 in the structure,
398 use &(P)->base even if P is NULL.
400 PR rtl-optimization/51023
401 * combine.c (simplify_comparison) <case SIGN_EXTEND>: Don't use
402 val_signbit_known_clear_p for signed comparison narrowing
403 optimization. Don't check for non-VOIDmode, use
404 HWI_COMPUTABLE_MODE_P macro.
405 <case ZERO_EXTEND>: Don't check for non-VOIDmode.
406 Optimize even when const_op is equal to GET_MODE_MASK (mode),
407 don't optimize if const_op is negative.
409 2011-11-10 Richard Guenther <rguenther@suse.de>
411 PR tree-optimization/51042
412 * tree-ssa-pre.c (phi_translate_1): Avoid recursing on
413 self-referential expressions. Refactor code to avoid duplication.
415 2011-11-10 Richard Guenther <rguenther@suse.de>
417 PR tree-optimization/51070
418 * tree-loop-distribution.c (generate_builtin): Do not replace
419 the loop with a builtin if the partition contains statements which
420 results are used outside of the loop.
421 (pass_loop_distribution): Verify and collect.
423 2011-11-10 Richard Guenther <rguenther@suse.de>
425 PR tree-optimization/51030
426 * tree-ssa-phiopt.c (jump_function_from_stmt): New function.
427 (value_replacement): Use it to handle trivial non-empty
430 2011-11-10 Richard Guenther <rguenther@suse.de>
433 * gimple.c (gimple_has_side_effects): Remove checking code
434 that doesn't belong here.
436 2011-11-10 Ira Rosen <ira.rosen@linaro.org>
438 PR tree-optimization/51058
439 * tree-vect-slp.c (vect_get_constant_vectors): Handle CALL_EXPR.
441 2011-11-10 Jakub Jelinek <jakub@redhat.com>
443 PR tree-optimization/51000
444 * tree-vect-patterns.c (vect_recog_bool_pattern): If adding
445 a pattern stmt for a bool store, adjust DR_STMT too.
446 Don't handle bool conversions to single bit precision lhs.
447 * tree-vect-stmts.c (vect_remove_stores): If next is a pattern
448 stmt, remove its related stmt and free its stmt_vinfo.
449 (free_stmt_vec_info): Free also pattern stmt's vinfo and
450 pattern def stmt's vinfo.
451 * tree-vect-loop.c (destroy_loop_vec_info): Don't try to
452 free pattern stmt's vinfo here.
453 (vect_transform_loop): When calling vect_remove_stores,
454 do gsi_next first and don't call gsi_remove. If not strided
455 store, free stmt vinfo for gsi_stmt (si) rather than stmt.
457 2011-11-09 Jakub Jelinek <jakub@redhat.com>
460 * config/i386/i386.c (expand_vec_perm_interleave2): If d->vmode is
461 V4SImode, !TARGET_SSE2 and punpck[lh]* is needed, change dremap.vmode
464 * function.h (requires_stack_frame_p): New prototype.
465 * function.c (requires_stack_frame_p): No longer static.
466 * config/i386/i386.c (ix86_finalize_stack_realign_flags): If
467 stack_realign_fp was just a conservative guess for a function
468 which doesn't use sp/fp/argp at all, clear frame_pointer_needed
469 and stack realignment.
471 2011-11-09 Paolo Carlini <paolo.carlini@oracle.com>
473 PR preprocessor/51061
474 * incpath.c (merge_include_chains): Make sure to not pass null
475 pointers to free_path.
477 2011-11-09 Paolo Carlini <paolo.carlini@oracle.com>
479 * doc/invoke.texi ([Option Summary, C++ Language Options]):
480 Add -Wzero-as-null-pointer-constant.
482 2011-11-07 Matthew Gretton-Dann <matthew.gretton-dann@arm.com>
484 * config/arm/arm-cores.def: Add -mcpu=cortex-a7.
485 * config/arm/arm-tables.opt: Regenerate.
486 * config/arm/arm-tune.md: Likewise.
487 * config/arm/bpabi.h (BE8_LINK_SPEC): Add Cortex A-7.
488 * doc/invoke.texi: Document -mcpu=cortex-a7.
490 2011-11-09 Richard Guenther <rguenther@suse.de>
493 2011-11-09 Richard Guenther <rguenther@suse.de>
495 * tree-cfg.c (verify_gimple_call): Verify that
496 gimple_call_cannot_inline_p is returning a conservative
497 correct result according to gimple_check_call_matching_types.
499 2011-11-09 Richard Guenther <rguenther@suse.de>
501 PR tree-optimization/51039
502 * tree-cfg.c (verify_gimple_call): Verify that
503 gimple_call_cannot_inline_p is returning a conservative
504 correct result according to gimple_check_call_matching_types.
505 * ipa-inline-analysis.c (estimate_function_body_sizes): Remove
506 code dealing with un-inlinablility.
507 * gimple-streamer-in.c (input_gimple_stmt): Update the
510 2011-11-09 Richard Guenther <rguenther@suse.de>
512 PR tree-optimization/51039
513 * tree-inline.c (setup_one_parameter): Always perform a
514 valid gimple type change.
515 (declare_return_variable): Likewise.
517 2011-11-09 Jakub Jelinek <jakub@redhat.com>
519 * config/rs6000/vector.md (vcondv4sfv4si, vcondv4siv4sf,
520 vconduv4sfv4si): New patterns.
521 * config/rs6000/rs6000.c (rs6000_emit_vector_cond_expr): Handle
522 different dest_mode from comparison mode.
524 2011-11-09 Richard Guenther <rguenther@suse.de>
526 * gimple-fold.c (canonicalize_constructor_val): Mark
527 address-taken variables addressable.
529 2011-11-09 Ira Rosen <ira.rosen@linaro.org>
531 PR tree-optimization/51015
532 * tree-vect-loop.c (vect_determine_vectorization_factor): Expect
533 vectype to be set for pattern def stmts.
535 2011-11-09 Alan Modra <amodra@gmail.com>
537 * function.c (bb_active_p): Delete.
538 (dup_block_and_redirect, active_insn_between): New functions.
539 (convert_jumps_to_returns, emit_return_for_exit): New functions,
541 (thread_prologue_and_epilogue_insns): ..here. Delete
542 shadowing variables. Don't do prologue register clobber tests
543 when shrink wrapping already failed. Delete all last_bb_active
544 code. Instead compute tail block candidates for duplicating
545 exit path. Remove these from antic set. Duplicate tails when
546 reached from both blocks needing a prologue/epilogue and
547 blocks not needing such.
548 * ifcvt.c (dead_or_predicable): Test both flag_shrink_wrap and
550 * bb-reorder.c (get_uncond_jump_length): Make global.
551 * bb-reorder.h (get_uncond_jump_length): Declare.
552 * cfgrtl.c (rtl_create_basic_block): Comment typo fix.
553 (rtl_split_edge): Likewise. Warning fix.
554 (rtl_duplicate_bb): New function.
555 (rtl_cfg_hooks): Enable can_duplicate_block_p and duplicate_block.
556 * Makefile.in (function.o): Update dependencies.
558 2011-11-08 Ulrich Weigand <Ulrich.Weigand@de.ibm.com>
559 Georg-Johann Lay <avr@gjlay.de>
561 * doc/tm.texi.in (MODE_CODE_BASE_REG_CLASS): Add address space
563 (REGNO_MODE_CODE_OK_FOR_BASE_P): Likewise.
564 * doc/tm.texi: Regenerate.
566 * config/cris/cris.h (MODE_CODE_BASE_REG_CLASS): Add address
568 (REGNO_MODE_CODE_OK_FOR_BASE_P): Likewise.
569 * config/bfin/bfin.h (MODE_CODE_BASE_REG_CLASS): Likewise.
570 (REGNO_MODE_CODE_OK_FOR_BASE_P): Likewise.
571 * config/avr/avr.h (MODE_CODE_BASE_REG_CLASS): Add address space
573 (REGNO_MODE_CODE_OK_FOR_BASE_P): Ditto.
574 * config/avr/avr-protos.h (avr_mode_code_base_reg_class): Ditto.
575 (avr_regno_mode_code_ok_for_base_p): Ditto.
576 * config/avr/avr.c (avr_mode_code_base_reg_class): Ditto.
577 (avr_regno_mode_code_ok_for_base_p): Ditto.
578 (avr_reg_ok_for_addr_p): Pass AS down to
579 avr_regno_mode_code_ok_for_base_p.
581 * addresses.h (base_reg_class): Add address space argument.
582 Pass to MODE_CODE_BASE_REG_CLASS.
583 (ok_for_base_p_1): Add address space argument. Pass to
584 REGNO_MODE_CODE_OK_FOR_BASE_P.
585 (regno_ok_for_base_p): Add address space argument. Pass to
588 * regrename.c (scan_rtx_address): Add address space argument.
589 Pass address space to regno_ok_for_base_p and base_reg_class.
590 Update recursive calls.
591 (scan_rtx): Pass address space to scan_rtx_address.
592 (build_def_use): Likewise.
593 * regcprop.c (replace_oldest_value_addr): Add address space
594 argument. Pass to regno_ok_for_base_p and base_reg_class.
595 Update recursive calls.
596 (replace_oldest_value_mem): Pass address space to
597 replace_oldest_value_addr.
598 (copyprop_hardreg_forward_1): Likewise.
600 * reload.c (find_reloads_address_1): Add address space argument.
601 Pass address space to base_reg_class and regno_ok_for_base_p.
602 Update recursive calls.
603 (find_reloads_address): Pass address space to base_reg_class,
604 regno_ok_for_base_p, and find_reloads_address_1.
605 (find_reloads): Pass address space to base_reg_class.
606 (find_reloads_subreg_address): Likewise.
608 * ira-costs.c (record_reg_classes): Update calls to base_reg_class.
609 (ok_for_base_p_nonstrict): Add address space argument. Pass to
611 (record_address_regs): Add address space argument. Pass to
612 base_reg_class and ok_for_base_p_nonstrict. Update recursive calls.
613 (record_operand_costs): Pass address space to record_address_regs.
614 (scan_one_insn): Likewise.
616 * caller-save.c (init_caller_save): Update call to base_reg_class.
617 * ira-conflicts.c (ira_build_conflicts): Likewise.
618 * reload1.c (maybe_fix_stack_asms): Likewise.
620 2011-11-08 Michael Matz <matz@suse.de>
622 * gengtype.c (write_field_root): Avoid out-of-scope access of newv.
624 * tree-stdarg.c (execute_optimize_stdarg): Accept clobbers.
626 * tree.h (TREE_CLOBBER_P): New macro.
627 * gimple.h (gimple_clobber_p): New inline function.
628 * gimplify.c (gimplify_bind_expr): Add clobbers for all variables
629 that go out of scope and live in memory.
630 * tree-ssa-operands.c (get_expr_operands): Transfer volatility also
632 * cfgexpand.c (decl_to_stack_part): New static variable.
633 (add_stack_var): Allocate it, and remember mapping.
634 (fini_vars_expansion): Deallocate it.
635 (stack_var_conflict_p): Add early outs.
636 (visit_op, visit_conflict, add_scope_conflicts_1,
637 add_scope_conflicts): New static functions.
638 (expand_used_vars_for_block): Don't call add_stack_var_conflict, tidy.
639 (expand_used_vars): Add scope conflicts.
640 (expand_gimple_stmt_1): Expand clobbers to nothing.
641 (expand_debug_expr): Ditto.
643 * tree-pretty-print.c (dump_generic_node): Dump clobbers nicely.
644 * tree-ssa-live.c (remove_unused_locals): Remove clobbers that
645 refer to otherwise unused locals.
646 * tree-sra.c (build_accesses_from_assign): Ignore clobbers.
647 * tree-ssa-dce.c (mark_stmt_if_obviously_necessary): Clobbers of
648 SSA names aren't necessary.
649 (propagate_necessity): Accept and ignore constructors on the rhs, tidy.
650 * gimple.c (walk_gimple_op): Accept constructors like mem_rhs.
651 * tree-ssa-structalias.c (find_func_aliases): Clobbers don't store
653 * tree-ssa-sccvn.c (vn_reference_lookup_3): Ditto, in particular they
654 don't zero-initialize something.
655 * tree-ssa-phiopt.c (cond_if_else_store_replacement_1): Ignore
656 clobber RHS, we don't want PHI nodes with those.
658 2011-11-08 Jakub Jelinek <jakub@redhat.com>
660 * config/i386/i386.c (ix86_preferred_simd_mode): Even for TARGET_AVX
661 if not TARGET_PREFER_AVX128 return 32-byte integer vectors.
663 2011-11-08 Richard Guenther <rguenther@suse.de>
665 PR tree-optimization/51012
666 * ipa-prop.c (update_indirect_edges_after_inlining): Fixup
668 * cgraph.c (cgraph_make_edge_direct): Likewise.
670 2011-11-08 Eric Botcazou <ebotcazou@adacore.com>
672 PR rtl-optimization/47698
673 * ifcvt.c (noce_operand_ok): Move around comment.
675 2011-11-08 Richard Guenther <rguenther@suse.de>
678 * lto-opts.c (append_to_collect_gcc_options): Split out from...
679 (lto_write_options): ... here. Prepend frontend specific flags.
681 2011-11-08 Jakub Jelinek <jakub@redhat.com>
683 * config/i386/sse.md (*avx_vperm2f128<mode>3_nozero): Emit mask
684 0x12 and 0x20 as vinsert[fi]128 instead of vperm2[fi]128.
686 * tree-vect-stmts.c (vectorizable_call): Add SLP_NODE argument.
687 Handle vectorization of SLP calls.
688 (vect_analyze_stmt): Adjust caller, add call to it for SLP too.
689 (vect_transform_stmt): Adjust vectorizable_call caller, remove
691 * tree-vect-slp.c (vect_get_and_check_slp_defs): For calls start
693 (vect_build_slp_tree): Allow CALL_EXPR.
695 2011-11-08 Richard Guenther <rguenther@suse.de>
697 * gimple-fold.c (canonicalize_constructor_val): Make sure
698 we have referenced vars setup before adding to them.
700 2011-11-07 Richard Henderson <rth@redhat.com>
701 Aldy Hernandez <aldyh@redhat.com>
702 Andrew MacLeod <amacleod@redhat.com>
703 Torvald Riegel <triegel@redhat.com>
705 Merged from transactional-memory.
707 * gtm-builtins.def: New file.
708 * trans-mem.c: New file.
709 * trans-mem.h: New file.
711 * opts.c (finish_options): Error out when using -flto and -fgnu-tm.
713 * config/i386/i386.c: Define TARGET_VECTORIZE* transactional variants.
714 (ix86_handle_tm_regparm_attribute, struct bdesc_tm,
715 ix86_builtin_tm_load, ix86_builtin_tm_store,
716 ix86_init_tm_builtins): New.
717 (ix86_init_builtins): Initialize TM builtins.
718 (struct ix86_attribute_table): Add "*tm regparm".
719 * config/i386/i386-builtin-types.def (PV2SI): Define.
721 Define V2SI_FTYPE_PCV2SI.
722 Define V4SF_FTYPE_PCV4SF.
723 Define V8SF_FTYPE_PCV8SF.
724 Define VOID_PV2SI_V2SI.
726 * doc/invoke.texi (C Dialect Options): Document -fgnu-tm and
727 tm-max-aggregate-size.
728 * doc/tm.texi.in: Add TARGET_VECTORIZE_BUILTIN_TM_LOAD and
729 TARGET_VECTORIZE_BUILTIN_TM_STORE hooks.
730 * doc/tm.texi: Regenerate.
732 * attribs.c (apply_tm_attr): New.
733 (init_attributes): Allow '*' prefix for overrides.
734 (register_attribute): Likewise.
735 * builtin-attrs.def (ATTR_TM_TMPURE, ATTR_TM_REGPARM): New.
736 (ATTR_TM_NOTHROW_LIST, ATTR_TM_TMPURE_NOTHROW_LIST,
737 ATTR_TM_PURE_TMPURE_NOTHROW_LIST, ATTR_TM_NORETURN_NOTHROW_LIST,
738 ATTR_TM_CONST_NOTHROW_LIST, ATTR_TMPURE_MALLOC_NOTHROW_LIST,
739 ATTR_TMPURE_NOTHROW_LIST): New.
740 * builtin-types.def (BT_FN_I[1248]_VPTR, BT_FN_FLOAT_VPTR,
741 BT_FN_DOUBLE_VPTR, BT_FN_LDOUBLE_VPTR, BT_FN_VOID_VPTR_I[1248],
742 BT_FN_VOID_VPTR_FLOAT, BT_FN_VOID_VPTR_DOUBLE,
743 BT_FN_VOID_VPTR_LDOUBLE, BT_FN_VOID_VPTR_SIZE): New.
744 * builtins.def: Include gtm-builtins.def. Add comments regarding
745 transactional memory synchronization.
746 (DEF_TM_BUILTIN): New.
747 * c-parser.c (struct c_parser): Add in_transaction.
748 (c_parser_transaction, c_parser_transaction_expression,
749 c_parser_transaction_cancel, c_parser_transaction_attributes): New.
750 (c_parser_attribute_any_word): Split out from c_parser_attributes.
751 (c_parser_statement_after_labels): Handle RID_TRANSACTION*.
752 (c_parser_unary_expression): Same.
753 * c-tree.h (c_finish_transaction): Declare.
754 * c-typeck.c (c_finish_transaction): New.
755 (build_function_call_vec): Call tm_malloc_replacement.
756 * calls.c (is_tm_builtin): New.
757 (flags_from_decl_or_type): Add ECF_TM_BUILTIN and ECF_TM_PURE.
758 * cfgbuild.c (make_edges): Add edges for REG_TM notes.
759 * cfgexpand.c (expand_call_stmt): Call
760 mark_transaction_restart_calls.
761 (gimple_expand_cfg): Free the tm_restart map.
762 (mark_transaction_restart_calls): New.
763 * cfgrtl.c (purge_dead_edges): Look for REG_TM notes.
764 * cgraph.c (dump_cgraph_node): Handle tm_clone.
765 * cgraph.h (struct cgraph_node): Add tm_clone field.
766 (decl_is_tm_clone): New.
767 (struct cgraph_local_info): Add tm_may_enter_irr.
768 (cgraph_copy_node_for_versioning): Declare.
769 * cgraphunit.c (cgraph_copy_node_for_versioning): Export;
770 copy analyzed from old version.
771 * combine.c (distribute_notes): Handle REG_TM notes.
772 * common.opt: Add -fgnu-tm.
773 * crtstuff.c (__TMC_LIST__, __TMC_END__): New.
774 (__do_global_dtors_aux): Deregister clone table.
775 (frame_dummy): Register clone table.
776 * emit-rtl.c (try_split): Handle REG_TM. Early return if no function
778 * gimple-low.c (lower_stmt): Handle GIMPLE_EH_ELSE and
780 (gimple_stmt_may_fallthru): Handle GIMPLE_EH_ELSE.
781 * gimple-pretty-print.c: Include trans-mem.h.
782 (dump_gimple_fmt): Add %x.
783 (dump_gimple_call): Dump arguments for calls to _ITM_beginTransaction.
784 (dump_gimple_eh_else, dump_gimple_transaction): New.
785 (dump_gimple_stmt): Handle GIMPLE_EH_ELSE and GIMPLE_TRANSACTION.
786 * gimple.c (gimple_build_eh_else, gimple_build_transaction): New.
787 (walk_gimple_seq): Honor removed_stmt. Document usage of removed_stmt
789 (walk_gimple_op): Handle GIMPLE_TRANSACTION.
790 (walk_gimple_stmt): Initialize and honor removed_stmt.
791 Handle GIMPLE_EH_ELSE and GIMPLE_TRANSACTION.
792 (gimple_copy): Handle GIMPLE_EH_ELSE and GIMPLE_TRANSACTION.
793 * gimple.def (GIMPLE_TRANSACTION, GIMPLE_EH_ELSE): New.
794 * gimple.h (struct gimple_statement_eh_else,
795 gimple_statement_transaction, GTMA_*): New.
796 (gimple_statement_d): Add gimple_statement_eh_else and
798 (gimple_build_eh_else, gimple_build_transaction,
799 gimple_fold_call, diagnose_tm_safe_errors): Declare.
800 (get_call_expr_in): Remove prototype.
801 (gimple_has_substatements): Add GIMPLE_EH_ELSE and GIMPLE_TRANSACTION.
802 (gimple_eh_else_n_body, gimple_eh_else_e_body,
803 gimple_eh_else_set_n_body, gimple_eh_else_set_e_body,
804 gimple_transaction_body, gimple_transaction_label,
805 gimple_transaction_label_ptr, gimple_transaction_subcode,
806 gimple_transaction_set_body, gimple_transaction_set_label,
807 gimple_transaction_set_subcode): New.
808 (struct walk_stmt_info): Use BOOL_BITFIELD; add removed_stmt.
809 * gimplify.c (create_tmp_var_name): Use clean_symbol_name.
810 (voidify_wrapper_expr): Handle TRANSACTION_EXPR.
811 (gimplify_transaction): New.
812 (gimplify_expr): Handle TRANSACTION_EXPR.
813 * gsstruct.def (GSS_EH_ELSE, GSS_TRANSACTION): New.
814 * ipa-inline.c (can_inline_edge_p): Do not inline TM safe calling
816 * Makefile.in: Add trans-mem.o and dependencies.
817 (BUILTINS_DEF): Add gtm-builtins.def.
818 (gimple-pretty-print.o): Depend on TRANS_MEM_H.
819 (GTFILES): Add trans-mem.c.
820 * omp-low.c (WALK_SUBSTMTS): Add GIMPLE_TRANSACTION.
821 * output.h (record_tm_clone_pair, finish_tm_clone_pairs,
822 get_tm_clone_pair): Declare.
823 * params.def (PARAM_TM_MAX_AGGREGATE_SIZE): New.
824 * passes.c (init_optimization_passes): Place transactional memory
826 * print-tree.c (print_node): Dump tm-clone.
827 * recog.c (peep2_attempt): Handle REG_TM.
828 * reg-notes.def (TM): New.
829 * rtlanal.c (alloc_reg_note): Handle REG_TM.
830 * target.def (builtin_tm_load, builtin_tm_store): New.
831 * targhooks.c (default_builtin_tm_load_store): New.
832 * targhooks.h (default_builtin_tm_load_store): Declare.
833 * timevar.def (TV_TRANS_MEM): New.
834 * toplev.c (compile_file): Call finish_tm_clone_pairs.
835 * tree-cfg.c (make_edges): Handle GIMPLE_TRANSACTION.
836 (cleanup_dead_labels): Handle GIMPLE_TRANSACTION. Avoid unnecessary
837 writes into the statements to update labels.
838 (is_ctrl_altering_stmt): Add TM ending statements. Handle
840 (verify_gimple_transaction): New.
841 (verify_gimple_stmt): Handle GIMPLE_TRANSACTION.
842 (verify_gimple_in_seq_2): Handle GIMPLE_EH_ELSE and GIMPLE_TRANSACTION.
843 (gimple_redirect_edge_and_branch): Handle TM_TRANSACTION.
844 (dump_function_to_file): Display [tm-clone] if applicable.
845 * tree-eh.c (struct_ptr_eq): Make inline and move to tree.h.
846 (struct_ptr_hash): Same.
847 (collect_finally_tree): Handle GIMPLE_EH_ELSE.
848 (replace_goto_queue_1): Likewise.
850 (honor_protect_cleanup_actions): Handle GIMPLE_EH_ELSE.
851 (lower_try_finally_nofallthru): Likewise.
852 (lower_try_finally_onedest): Likewise.
853 (lower_try_finally_copy): Likewise.
854 (lower_try_finally_switch): Likewise.
855 (lower_try_finally): Likewise.
856 (decide_copy_try_finally): Likewise.
857 (lower_eh_constructs_2): Likewise.
858 (refactor_eh_r): Likewise.
859 * tree-flow.h (struct gimple_df): Add tm_restart field.
860 Define tm_restart_node.
861 * tree-inline.c (remap_gimple_stmt): Handle GIMPLE_TRANSACTION.
862 (estimate_num_insns): Likewise.
863 (init_inline_once): Init tm_cost.
864 * tree-inline.h (struct eni_weights_d): Add tm_cost.
865 * tree-pass.h (pass_diagnose_tm_blocks, pass_lower_tm, pass_tm_init,
866 pass_tm_mark, pass_tm_memopt, pass_tm_edges, pass_ipa_tm): Declare.
867 * tree-pretty-print.c (dump_generic_node): Handle TRANSACTION_EXPR.
868 * tree-ssa-alias.c (ref_maybe_used_by_call_p_1): Handle
869 BUILT_IN_TM_MEMSET, BUILT_IN_TM_MEMCPY, BUILT_IN_TM_MEMMOVE.
870 Add support for TM vector loads. Add support for TM logging builtins.
871 (call_may_clobber_ref_p_1): Add support for vector stores.
872 * tree-ssa-structalias.c (find_func_aliases): Add support for TM
873 vector stores and loads. Handle BUILT_IN_TM_MEMSET,
874 BUILT_IN_TM_MEMCPY, BUILT_IN_TM_MEMMOVE.
875 * tree.c (strip_invariant_refs): Moved from gimple.c to here.
876 (local_define_builtin): Handle ECF_TM_PURE.
877 (build_common_builtin_nodes): Set __builtin_eh_pointer to ECF_TM_PURE.
878 * tree.def (TRANSACTION_EXPR): New.
879 * tree.h (strip_invariant_refs): Moved from gimple.h to here.
880 (TRANSACTION_EXPR_BODY, TRANSACTION_EXPR_CHECK,
881 TRANSACTION_EXPR_OUTER, TRANSACTION_EXPR_RELAXED,
882 BUILTIN_TM_LOAD_STORE_P, BUILTIN_TM_LOAD_P, BUILTIN_TM_STORE_P,
883 CASE_BUILT_IN_TM_LOAD, CASE_BUILT_IN_TM_STORE): New.
884 (ECF_TM_PURE, ECF_TM_BUILTIN): New.
885 (struct tree_function_decl): Add tm_clone_flag.
886 (struct_ptr_eq, struct_ptr_hash): New.
887 (apply_tm_attr): Declare.
888 (is_tm_safe_or_pure): New.
889 (build_tm_abort_call, is_tm_safe, is_tm_pure,
890 is_tm_may_cancel_outer, is_tm_ending_fndecl, record_tm_replacement,
891 tm_malloc_replacement): Declare.
892 * varasm.c (tm_clone_hash): New.
893 (record_tm_clone_pair, finish_tm_clone_pairs, get_tm_clone_pair,
894 dump_tm_clone_to_vec, dump_tm_clone_pairs, tm_alias_pair_cmp): New.
895 (struct tm_alias_pair): New. Declare VEC types for object.
897 2011-11-07 Richard Henderson <rth@redhat.com>
899 * optabs.h (OTI_sync_compare_and_swap, OTI_sync_lock_test_and_set,
900 OTI_sync_old_add, OTI_sync_old_sub, OTI_sync_old_ior,
901 OTI_sync_old_and, OTI_sync_old_xor, OTI_sync_old_nand,
902 OTI_sync_new_add, OTI_sync_new_sub, OTI_sync_new_ior,
903 OTI_sync_new_and, OTI_sync_new_xor, OTI_sync_new_nand): Move and
904 rename from the direct_optab_index enum.
905 (sync_compare_and_swap_optab, sync_lock_test_and_set_optab,
906 sync_old_add_optab, sync_old_sub_optab, sync_old_ior_optab,
907 sync_old_and_optab, sync_old_xor_optab, sync_old_nand_optab,
908 sync_new_add_optab, sync_new_sub_optab, sync_new_ior_optab,
909 sync_new_and_optab, sync_new_xor_optab, sync_new_nand_optab): Read
910 from the optab_table, not the direct_optab_table.
911 (init_sync_libfuncs): Declare.
912 (can_compare_and_swap_p): Update parameters.
913 * optabs.c (init_sync_libfuncs_1, init_sync_libfuncs): New.
914 (can_compare_and_swap_p): Add allow_libcall parameter; if true,
915 test for the legacy compare-and-swap libcall.
916 (expand_atomic_exchange): Use the legacy test-and-set libcall.
917 (expand_atomic_compare_and_swap): Use the legacy CAS libcall.
918 (struct atomic_op_functions): Update for optab type changes.
919 (maybe_emit_op): Likewise.
920 (expand_atomic_fetch_op): Use the legacy fetch-op libcalls.
921 * builtins.c (fold_builtin_atomic_always_lock_free): Update call
922 to can_compare_and_swap_p.
923 * omp-low.c (expand_omp_atomic_fetch_op): Likewise.
924 (expand_omp_atomic_pipeline): Likewise.
925 * genopinit.c (optabs): Make sync_old_*_optab, sync_new_*_optab,
926 sync_compare_and_swap_optab, sync_lock_test_and_set_optab regular
928 * doc/md.texi (sync_compare_and_swap): Update docs for libcalls.
930 2011-11-07 Jakub Jelinek <jakub@redhat.com>
932 * config/i386/i386-bultin-types.def (V8SI_FTYPE_V4DF_V4DF): Add.
933 * config/i386/i386.c (enum ix86_builtins): Add
934 IX86_BUILTIN_VEC_PACK_SFIX256.
935 (bdesc_args): Add __builtin_ia32_vec_pack_sfix256.
936 (ix86_expand_args_builtin): Handle V8SI_FTYPE_V4DF_V4DF.
937 (ix86_builtin_vectorized_function): Also vectorize lrint using
938 256-bit vectors for -mavx.
940 2011-11-07 Anatoly Sokolov <aesok@post.ru>
942 * config/cris/constraints.md: New file.
943 * config/cris/cris.h (REG_CLASS_FROM_LETTER, CONSTRAINT_LEN,
944 CRIS_CONST_OK_FOR_LETTER_P, CONST_OK_FOR_CONSTRAINT_P,
945 CONST_DOUBLE_OK_FOR_LETTER_P, EXTRA_MEMORY_CONSTRAINT,
946 EXTRA_CONSTRAINT, EXTRA_CONSTRAINT_Q, EXTRA_CONSTRAINT_R,
947 EXTRA_CONSTRAINT_T, EXTRA_CONSTRAINT_S, EXTRA_CONSTRAINT_U): Remove.
948 * config/cris/cris.c: Incule "tm-constrs.h".
949 (cris_print_operand): Use satisfies_constraint_O.
950 (cris_normal_notice_update_cc, cris_rtx_costs): Use
951 satisfies_constraint_I.
952 (cris_address_cost): Use satisfies_constraint_L.
953 * config/cris/cris.md: Include "constraints.md".
954 (*mov_side<mode>, *mov_sidesisf, *mov_side<mode>_mem,
955 *mov_sidesisf_mem, *clear_side<mode>, *ext_sideqihi,
956 *ext_side<mode>si, *op_side<mode>, *op_swap_side<mode>,
957 *extopqihi_side, *extop<mode>si_side, *extopqihi_swap_side,
958 *extop<mode>si_swap_side): Use satisfies_constraint_N and
959 satisfies_constraint_J.
960 (moversideqi movemsideqi mover2side peephole2): Use
961 satisfies_constraint_N and satisfies_constraint_J.
962 (andu peephole2): Use satisfies_constraint_I and
963 satisfies_constraint_O.
965 2011-11-07 Uros Bizjak <ubizjak@gmail.com>
967 * config/i386/i386.c (ix86_builtin_vectorized_function): Handle
968 BUILT_IN_IRINT, BUILT_IN_IRINTF, BUILT_IN_LLRINT and BUILT_IN_LLRINTF.
970 2011-11-07 Andrew MacLeod <amacleod@redhat.com>
972 * optabs.c (get_atomic_op_for_code): Fill in optab table at runtime so
973 SWITCHABLE_TARGET can change the values during compilation.
974 (expand_atomic_fetch_op): Handle parameter change ripples for
975 get_atomic_op_for_code call.
977 2011-11-07 Andrew MacLeod <amacleod@redhat.com>
979 * doc/extend.texi: Docuemnt behaviour change for __atomic_exchange and
981 * optabs.c (expand_atomic_exchange): Expand to __sync_lock_test_and_set
982 only when originated from that builtin.
983 (expand_atomic_store): Expand to __sync_lock_release when originated
985 * builtins.c (expand_builtin_sync_lock_test_and_set): Add flag that
986 expand_atomic_exchange call originated from here.
987 (expand_builtin_sync_lock_release): Add flag that expand_atomic_store
988 call originated from here.
989 (expand_builtin_atomic_exchange): Add origination flag.
990 (expand_builtin_atomic_store): Add origination flag.
991 * expr.h (expand_atomic_exchange, expand_atomic_store): Add boolean
992 parameters to indicate implementation fall back options.
994 2011-11-07 Georg-Johann Lay <avr@gjlay.de>
996 * config/avr/avr.c (output_reload_in_const): Can handle CONSTANT_P
997 now, not only CONST_INT and CONST_DOUBLE.
998 (output_movqi): Use output_reload_in_const.
999 (output_reload_inhi): Ditto.
1000 (output_reload_insisf): Move assertion to output_reload_in_const.
1001 (avr_out_reload_inpsi): Ditto.
1003 2011-11-07 Nathan Sidwell <nathan@acm.org>
1005 * gcov.c (object_summary): Replace with ...
1006 (object_runs): ... this.
1007 (process_file): Remove functions with no data.
1008 (generate_results): Ignore files with no lines.
1009 (release_function): New helper, broken out of ...
1010 (release_structures): ... here. Use it.
1011 (read_count_file): Adjust for new data file format.
1012 (output_lines): Use object_runs.
1013 * gcov-io.h (GCOV_TAG_OBJECT_SUMMARY): Obsolete.
1014 (struct gcov_ctr_info): Move definition.
1015 (struct gcov_fn_info): Add key field, use gcov_ctr_info for
1017 (struct gcov_info): Add merge function array, remove mask and
1018 counts. Trailing array is array of pointers to function info.
1019 * coverage.c (struct function_list): Replace counter numbers with
1020 counter arrays. Add fndecl field. GTYify.
1021 (counts_entry): Remove chain workspace.
1022 (functions_head): GTYify.
1023 (prg_n_ctrs): Remove.
1025 (tree_ctr_tables): Remove.
1026 (read_counts_file): Cope with blank entries and expect program
1027 summaries before functions. Don't warn on missing entries.
1028 (coverage_counter_alloc): Allocate individual function arrays.
1029 (tree_coverage_counter_ref, tree_coverage_counter_addr): Adjust
1030 for individual function arrays.
1031 (coverage_end_function): GTYify function list object. Finalize
1032 function's counter arrays.
1033 (build_var): New. Create a counter-related variable with
1034 appropriate linkage.
1035 (build_fn_info_type): Adjust for new runtime structure.
1036 (build_fn_info_value): Rename to ...
1037 (build_fn_info): ... here. Build new format data.
1038 (build_ctr_info_type, build_ctr_info_value): Remove.
1039 (build_info_type): New. Build new format data structure.
1040 (build_info): Adjust for new format data.
1041 (create_coverage): Likewise.
1042 * gcov-dump.c (tag_function): Recognize placeholders.
1044 2011-11-07 Georg-Johann Lay <avr@gjlay.de>
1046 * config/avr/constraints.md (Cm2): New constraint for int -2.
1047 * config/avr/avr.md (addqi3): Use it. New alternatives for +/-2.
1048 (*negqihi2): New insn.
1050 2011-11-07 H.J. Lu <hongjiu.lu@intel.com>
1052 * dwarf2cfi.c (dwarf2out_frame_debug_expr): Check
1053 HARD_FRAME_POINTER_REGNUM instead of hard_frame_pointer_rtx in Rule 18.
1055 2011-11-07 Jakub Jelinek <jakub@redhat.com>
1057 * config/i386/i386.c (ix86_expand_builtin): If gather mask
1058 argument is known to have all high bits set, pass pc_rtx as
1059 second argument to the expander instead of op0.
1060 * config/i386/sse.md (*avx2_gathersi<mode>_2,
1061 *avx2_gatherdi<mode>_2): New patterns.
1062 * config/i386/avx2intrin.h (_mm256_i32gather_pd,
1063 _mm256_i64gather_pd, _mm256_i32gather_ps): Set mask using
1064 _mm256_cmp_pd with zero vector arguments and _CMP_EQ_OQ instead of
1067 PR tree-optimization/50789
1068 * tree-vect-stmts.c (process_use): Add force argument, avoid
1069 exist_non_indexing_operands_for_use_p check if true.
1070 (vect_mark_stmts_to_be_vectorized): Adjust callers. Handle
1071 STMT_VINFO_GATHER_P.
1072 (gen_perm_mask): New function.
1073 (perm_mask_for_reverse): Use it.
1074 (reverse_vec_element): Rename to...
1075 (permute_vec_elements): ... this. Add Y and MASK_VEC arguments,
1076 generalize for any permutations.
1077 (vectorizable_load): Adjust caller. Handle STMT_VINFO_GATHER_P.
1078 * target.def (TARGET_VECTORIZE_BUILTIN_GATHER): New hook.
1079 * doc/tm.texi.in (TARGET_VECTORIZE_BUILTIN_GATHER): Document it.
1080 * doc/tm.texi: Regenerate.
1081 * tree-data-ref.c (initialize_data_dependence_relation,
1082 compute_self_dependence): No longer static.
1083 * tree-data-ref.h (initialize_data_dependence_relation,
1084 compute_self_dependence): New prototypes.
1085 * tree-vect-data-refs.c (vect_check_gather): New function.
1086 (vect_analyze_data_refs): Detect possible gather load data refs.
1087 * tree-vectorizer.h (struct _stmt_vec_info): Add gather_p field.
1088 (STMT_VINFO_GATHER_P): Define.
1089 (vect_check_gather): New prototype.
1090 * config/i386/i386-builtin-types.def: Add types for alternate
1092 * config/i386/sse.md (AVXMODE48P_DI): Remove.
1093 (VEC_GATHER_MODE): Rename mode_attr to...
1094 (VEC_GATHER_IDXSI): ... this.
1095 (VEC_GATHER_IDXDI, VEC_GATHER_SRCDI): New mode_attrs.
1096 (avx2_gathersi<mode>, *avx2_gathersi<mode>): Use <VEC_GATHER_IDXSI>
1097 instead of <VEC_GATHER_MODE>.
1098 (avx2_gatherdi<mode>): Use <VEC_GATHER_IDXDI> instead of
1099 <AVXMODE48P_DI> and <VEC_GATHER_SRCDI> instead of VEC_GATHER_MODE
1100 on src and mask operands.
1101 (*avx2_gatherdi<mode>): Likewise. Use VEC_GATHER_MODE iterator
1102 instead of AVXMODE48P_DI.
1103 (avx2_gatherdi<mode>256, *avx2_gatherdi<mode>256): Removed.
1104 * config/i386/i386.c (enum ix86_builtins): Add
1105 IX86_BUILTIN_GATHERALTSIV4DF, IX86_BUILTIN_GATHERALTDIV8SF,
1106 IX86_BUILTIN_GATHERALTSIV4DI and IX86_BUILTIN_GATHERALTDIV8SI.
1107 (ix86_init_mmx_sse_builtins): Create those builtins.
1108 (ix86_expand_builtin): Handle those builtins and adjust expansions
1109 of other gather builtins.
1110 (ix86_vectorize_builtin_gather): New function.
1111 (TARGET_VECTORIZE_BUILTIN_GATHER): Define.
1113 2011-11-07 Uros Bizjak <ubizjak@gmail.com>
1115 * config/i386/f16cintrin.h: Remove extra _X86INTRIN_H_INCLUDED check.
1117 2011-11-07 Tristan Gingold <gingold@adacore.com>
1119 * config.gcc (*-*-*vms*): Set c_target_objs, cxx_target_objs.
1120 * config/vms/t-vms: Add vms-c.o rule.
1121 * config/vms/vms-c.c: New file.
1122 * config/vms/vms-protos.h (vms_c_register_pragma): New prototype.
1123 * config/vms/vms.h (REGISTER_TARGET_PRAGMAS): Define.
1125 2011-11-07 Tristan Gingold <gingold@adacore.com>
1127 * config/alpha/vms.h (TARGET_OBJECT_SUFFIX,
1128 TARGET_EXECUTABLE_SUFFIX, TARGET_OS_CPP_BUILTINS,
1129 TARGET_ABI_OPEN_VMS, LONG_TYPE_SIZE, ADA_LONG_TYPE_SIZE,
1130 POINTER_SIZE, POINTERS_EXTEND_UNSIGNED): Move to config/vms/vms.h
1131 (SUBTARGET_OS_CPP_BUILTINS): Define.
1132 (TARGET_DEFAULT): Tune according to POINTER_SIZE.
1133 (MASK_RETURN_ADDR): Define in 64 bit mode.
1134 * config/ia64/vms.h: Likewise.
1135 * config/vms/vms.h: New file.
1136 * config/vms/vms64.h: New file.
1137 * config/alpha/vms64.h: Removed.
1138 * config/ia64/vms64.h: Removed.
1139 * config.gcc (alpha64-dec-*vms*, alpha*-dec-*vms*, ia64-hp-*vms*):
1140 Adjust for above change.
1142 2011-11-07 Enkovich Ilya <ilya.enkovich@intel.com>
1145 * config/i386/i386-protos.h (ix86_use_lea_for_mov): New.
1146 * config/i386/i386.c (ix86_use_lea_for_mov): Likewise.
1147 * config/i386/i386.md (movsi_internal): Emit lea if profitable.
1148 (movdi_internal_rex64): Likewise.
1150 2011-11-07 Sergey Ostanevich <sergos.gnu@gmail.com>
1152 PR rtl-optimization/47698
1153 * ifcvt.c (noce_operand_ok): Return false for mems with side effects.
1155 2011-11-07 Tristan Gingold <gingold@adacore.com>
1157 * common/config/alpha/alpha-common.c (alpha_option_init_struct):
1159 (TARGET_OPTION_INIT_STRUCT): Define.
1160 * config/alpha/alpha.c (TARGET_MIN_ANCHOR_OFFSET)
1161 (TARGET_MAX_ANCHOR_OFFSET)
1162 (TARGET_USE_BLOCKS_FOR_CONSTANT_P): Redefine.
1164 2011-11-06 Quentin Neill <quentin.neill@amd.com>
1166 Fix r180999, update ChangeLog
1167 * config.gcc: Add f16cintrin.h.
1168 * config/i386/f16cintrin.h: Add missing endif.
1170 2011-11-06 Sebastian Huber <sebastian.huber@embedded-brains.de>
1172 * config.gcc (arm*-*-rtemseabi*): New target.
1173 * config/arm/rtems-eabi.h: New.
1174 * config/arm/t-rtems-eabi: New.
1176 2011-11-06 David S. Miller <davem@davemloft.net>
1178 * config/sparc/sparc.md (UNSPEC_SHORT_LOAD): Delete.
1179 (zero_extend_v8qi_vis, zero_extend_v4hi_vis,
1180 *zero_extend_v8qi_<P:mode>_insn,
1181 *zero_extend_v4hi_<P:mode>_insn): Express using vec_merge
1182 and vec_duplicate instead of using an UNSPEC.
1184 2011-11-07 Alan Modra <amodra@gmail.com>
1187 * config/rs6000/rs6000.c (rs6000_emit_stack_reset): Always emit
1188 blockage for ABI_V4.
1190 2011-11-06 Dave Korn <dave.korn.cygwin@gmail.com>
1192 * config/i386/cygwin.h (LIBGCJ_SONAME): Updated to match recent
1193 libgcj version bump.
1194 * config/i386/mingw32.h (LIBGCJ_SONAME): Likewise.
1196 2011-11-06 Eric Botcazou <ebotcazou@adacore.com>
1198 * gcse.c: Adjust outdated comments throughout.
1199 (struct mem_conflict_info): New structure.
1200 (mems_conflict_for_gcse_p): Use it to communicate with caller.
1201 (load_killed_in_block_p): Pass it to note_stores.
1202 (hash_expr): Remove superfluous line break.
1203 (hash_scan_set): Rename PAT parameter into SET.
1204 (hash_scan_insn): Reorder cases.
1205 (canon_list_insert): Fix long line.
1206 (edge_list): Delete.
1207 (prune_expressions): Rename E local variable into EXPR.
1208 (compute_pre_data): Return struct edge_list * object.
1209 (pre_expr_reaches_here_p_work): Fix formatting.
1210 (process_insert_insn): Move around comment.
1211 (pre_edge_insert): Fix long line.
1212 (pre_insert_copies): Likewise.
1213 (gcse_emit_move_after): Swap SRC and DEST parameters.
1214 (pre_delete): Adjust call to gcse_emit_move_after.
1215 (pre_gcse): Take struct edge_list * parameter. Fix long line.
1216 (one_pre_gcse_pass): Use flag_gcse_lm condition for all routines.
1217 Use a local list of edges.
1218 (hoist_code): Fix long line. Adjust call to gcse_emit_move_after.
1219 (pre_ldst_expr_hash): Fix long line.
1220 (free_ldst_mems): Rename into...
1221 (free_ld_motion_mems): ...this.
1222 (first_ls_expr): Delete.
1223 (next_ls_expr): Likewise.
1224 (print_ldst_list): Do not use above two functions.
1225 (simple_mem): Adjust interface.
1226 (compute_ld_motion_mems): Fix formatting.
1227 (update_ld_motion_stores): Reuse local variable.
1229 2011-11-06 Joseph Myers <joseph@codesourcery.com>
1231 * c-decl.c (shadow_tag_warned, grokdeclarator): Handle _Alignas
1233 (build_null_declspecs): Initialize align_log and alignas_p fields.
1234 (declspecs_add_alignas): New.
1235 * c-parser.c (c_token_starts_declspecs): Handle RID_ALIGNAS.
1236 (c_parser_declspecs): Handle _Alignas specifiers.
1237 (c_parser_alignas_specifier): New.
1238 (c_parser_alignof_expression): Diagnose alignof use for non-C1X.
1239 Diagnose _Alignof (expression).
1240 * c-tree.h (struct c_declspecs): Add align_log and alignas_p fields.
1241 (declspecs_add_alignas): Declare.
1242 * ginclude/stddef.h (max_align_t): Define for C1X and C++11.
1243 * ginclude/stdalign.h: New.
1244 * Makefile.in (USER_H): Add stdalign.h.
1246 2011-11-06 Joern Rennecke <joern.rennecke@embecosm.com>
1248 * regset.h (fixed_reg_set_regset): Declare.
1249 * dse.c: Include regset.h .
1250 (struct insn_info): Add member fixed_regs_live.
1251 (note_add_store_info): New typedef.
1252 (note_add_store): New function.
1253 (emit_inc_dec_insn_before): Expect arg to be of type insn_info_t.
1254 Use gen_add3_insn / gen_move_insn.
1255 Check new insn for unwanted clobbers before emitting it.
1256 (check_for_inc_dec): Rename to...
1257 (check_for_inc_dec_1:) ... this. Return bool. Take insn_info
1258 parameter. Changed all callers in file.
1259 (check_for_inc_dec, copy_fixed_regs): New functions.
1260 (scan_insn): Set fixed_regs_live field of insn_info.
1261 * rtl.h (check_for_inc_dec): Update prototype.
1262 * postreload.c (reload_cse_simplify): Take new signature of
1263 check_ind_dec into account.
1264 * reginfo.c (fixed_reg_set_regset): New variable.
1265 (init_reg_sets_1): Initialize it.
1267 2011-11-06 Jakub Jelinek <jakub@redhat.com>
1269 * tree-cfg.c (gimple_can_merge_blocks_p): For -O0 don't remove
1272 2011-11-06 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1274 * config/pa/pa.c (pa_hpux_init_libfuncs): Rename to pa_init_libfuncs.
1275 Remove dependence of declaration and target define on definition of
1276 HPUX_LONG_DOUBLE_LIBRARY. Update implementation.
1278 2011-11-06 Andrew Macleod <amacleod@redhat.com>
1279 Richard Henderson <rth@redhat.com>
1280 Aldy Hernandez <aldyh@redhat.com>
1282 Merged from cxx-mem-model.
1284 * cppbuiltin.c (define__GNUC__): Define __ATOMIC memory models
1285 * coretypes.h (enum memmodel): New. enumerated memory model type.
1286 * Makefile.in (cppbuiltin.o) Add missing dependency on $(TREE_H)
1287 * genopinit,c (optabs): Add atomic direct optab handlers.
1288 * sync-builtins.def (BUILT_IN_ATOMIC_*): New builtins.
1289 * builtin-types.def (BT_CONST_VOLATILE_PTR,
1290 BT_FN_I{1,2,4,8,16}_CONST_VPTR_INT, BT_FN_VOID_VPTR_INT,
1291 BT_FN_BOOL_VPTR_INT, BT_FN_BOOL_SIZE_CONST_VPTR,
1292 BT_FN_I{1,2,4,8,16}_VPTR_I{1,2,4,8,16}_INT,
1293 BT_FN_VOID_VPTR_I{1,2,4,8,16}_INT, BT_FN_VOID_SIZE_VPTR_PTR_INT,
1294 BT_FN_VOID_SIZE_CONST_VPTR_PTR_INT, BT_FN_VOID_SIZE_VPTR_PTR_PTR_INT,
1295 BT_FN_BOOL_VPTR_PTR_I{1,2,4,8,16}_BOOL_INT_INT): New builtin types.
1296 * expr.h (expand_atomic_*): Add prototypes.
1297 (expand_{bool,val}_compare_and_swap): Remove prototypes.
1298 * c-typeck.c (build_function_call_vec): Don't reprocess __atomic
1300 * common.opt (Winvalid-memory-model): New warning flag.
1301 (finline-atomics): New. Flag to disable atomic inlining.
1302 * params.h (ALLOW_LOAD_DATA_RACES): New.
1303 (ALLOW_PACKED_LOAD_DATA_RACES): New.
1304 (ALLOW_PACKED_STORE_DATA_RACES): New.
1305 * params.def (PARAM_ALLOW_LOAD_DATA_RACES): New.
1306 (PARAM_ALLOW_PACKED_LOAD_DATA_RACES): New.
1307 (PARAM_ALLOW_PACKED_STORE_DATA_RACES): New.
1308 * builtins.c (is_builtin_name): Handle __atomic.
1309 (get_memmodel): New. Extract memory model.
1310 (expand_expr_force_mode): New. Factor out common code for ensuring an
1311 integer argument is in the proper mode.
1312 (expand_builtin_sync_operation): Remove ignore param. Always call
1313 expand_atomic_fetch_op instead of the old expanders.
1314 (expand_builtin_compare_and_swap,
1315 expand_builtin_sync_lock_test_and_set): Use expand_expr_force_mode,
1316 call atomic expanders instead of sync expanders.
1317 (expand_builtin_sync_lock_release): Call atomic_store expander.
1318 (expand_builtin_atomic_compare_exchange, expand_builtin_atomic_load,
1319 expand_builtin_atomic_store, expand_builtin_atomic_fetch_op): New.
1320 (expand_builtin_atomic_exchange): New.
1321 (fold_builtin_atomic_always_lock_free,
1322 expand_builtin_atomic_always_lock_free,
1323 fold_builtin_atomic_is_lock_free, expand_builtin_atomic_is_lock_free):
1325 (expand_builtin_mem_thread_fence, expand_builtin_atomic_thread_fence,
1326 expand_builtin_atomic_signal_fence): New.
1327 (expand_builtin_mem_signal_fence): New.
1328 (expand_builtin): Add cases for BUILT_IN_ATOMIC_*.
1329 (fold_builtin_2): Add cases for BUILT_IN_ATOMIC_{IS,ALWAYS}_LOCK_FREE.
1330 * optabs.h (DOI_atomic_*): Define new atomics.
1331 (atomic_*_optab): Define.
1332 (can_compare_and_swap_p, expand_atomic_compare_and_swap): New
1334 * optabs.c (expand_sync_operation, expand_sync_fetch_operation): Remove.
1335 (expand_sync_lock_test_and_set): Remove.
1336 (expand_atomic_load, expand_atomic_store): New.
1337 (expand_atomic_exchange): New.
1338 (expand_atomic_compare_and_swap): New. Implements
1339 atomic_compare_exchange via compare and swap.
1340 (struct atomic_op_functions): Opcode table struct for fetch ops.
1341 (get_atomic_op_for_code): New. Return an opcode table entry.
1342 (maybe_emit_op): New. Try to emit a fetch op.
1343 (expand_atomic_fetch_op): New.
1344 (expand_val_compare_and_swap_1): Remove.
1345 (expand_val_compare_and_swap, expand_bool_compare_and_swap): Remove.
1346 (expand_atomic_compare_and_swap): Rename from
1347 expand_atomic_compare_exchange. Rewrite to return both success and
1348 oldval return values; expand via both atomic and sync optabs.
1349 (can_compare_and_swap_p): New.
1350 (expand_compare_and_swap_loop): Use expand_atomic_compare_and_swap.
1351 (maybe_gen_insn): Handle 7 and 8 operands.
1352 * omp-low.c (expand_omp_atomic_fetch_op): Don't test individual
1353 fetch_op optabs, only test can_compare_and_swap_p. Use __atomic
1354 builtins instead of __sync builtins.
1355 (expand_omp_atomic_pipeline): Use can_compare_and_swap_p.
1356 * doc/extend.texi: Document __atomic built-in functions.
1357 * doc/invoke.texi: Document data race parameters.
1358 * doc/md.texi: Document atomic patterns.
1359 * config/i386/i386.md (UNSPEC_MOVA): New.
1360 (UNSPECV_CMPXCHG): Split into ...
1361 (UNSPECV_CMPXCHG_1, UNSPECV_CMPXCHG_2,
1362 UNSPECV_CMPXCHG_3, UNSPECV_CMPXCHG_4): New.
1363 * config/i386/sync.md (ATOMIC): New mode iterator.
1364 (atomic_load<ATOMIC>, atomic_store<ATOMIC>): New.
1365 (atomic_loaddi_fpu, atomic_storedi_fpu, movdi_via_fpu): New.
1366 (mem_thread_fence): Rename from memory_barrier.
1367 Handle the added memory model parameter.
1368 (mfence_nosse): Rename from memory_barrier_nosse.
1369 (sync_compare_and_swap<CASMODE>): Split into ...
1370 (atomic_compare_and_swap<SWI124>): this and ...
1371 (atomic_compare_and_swap<CASMODE>): this. Handle the new parameters.
1372 (atomic_compare_and_swap_single<SWI>): Rename from
1373 sync_compare_and_swap<SWI>; rewrite to use split unspecs.
1374 (atomic_compare_and_swap_double<DCASMODE>): Rename from
1375 sync_double_compare_and_swap<DCASMODE>; rewrite to use split unspecs.
1376 (*atomic_compare_and_swap_doubledi_pic): Rename from
1377 sync_double_compare_and_swapdi_pic; rewrite to use split unspecs.
1378 (atomic_fetch_add<SWI>): Rename from sync_old_add<SWI>; add memory
1380 (*atomic_fetch_add_cmp<SWI>): Similarly.
1381 (atomic_add<SWI>, atomic<any_logic><SWI>): Similarly.
1382 (atomic_sub<SWI>): Similarly. Use x86_maybe_negate_const_int.
1383 (sync_lock_test_and_set<SWI>): Merge with ...
1384 (atomic_exchange<SWI>): ... this.
1386 2011-11-6 Richard Guenther <rguenther@suse.de>
1388 * ipa-prop.c (ipa_modify_call_arguments): Re-compute inlinable flag.
1390 2011-11-06 Ira Rosen <ira.rosen@linaro.org>
1392 * tree-vectorizer.h (vectorizable_condition): Add argument.
1393 * tree-vect-loop.c (vectorizable_reduction): Fail for condition
1394 in SLP. Update calls to vectorizable_condition.
1395 * tree-vect-stmts.c (vect_is_simple_cond): Add basic block info to
1396 the arguments. Pass it to vect_is_simple_use_1.
1397 (vectorizable_condition): Add slp_node to the arguments. Support
1398 vectorization of basic blocks. Fail for reduction in SLP. Update
1399 calls to vect_is_simple_cond and vect_is_simple_use. Support SLP:
1400 call vect_get_slp_defs to get vector operands.
1401 (vect_analyze_stmt): Update calls to vectorizable_condition.
1402 (vect_transform_stmt): Likewise.
1403 * tree-vect-slp.c (vect_create_new_slp_node): Handle COND_EXPR.
1404 (vect_get_and_check_slp_defs): Handle COND_EXPR. Allow pattern
1406 (vect_build_slp_tree): Handle COND_EXPR.
1407 (vect_analyze_slp_instance): Push pattern statements to root node.
1408 (vect_get_constant_vectors): Fix comments. Handle COND_EXPR.
1410 2011-11-05 David S. Miller <davem@davemloft.net>
1412 * config/sparc/sparc.md (UNSPEC_SHORT_LOAD): New unspec.
1413 (zero-extend_v8qi_vis, zero_extend_v4hi_vis): New expanders.
1414 (*zero_extend_v8qi_<P:mode>_insn,
1415 *zero_extend_v4hi_<P:mode>_insn): New insns.
1416 * config/sparc/sparc.c (vector_init_move_words,
1417 vector_init_prepare_elts, sparc_expand_vector_init_vis2,
1418 sparc_expand_vector_init_vis1): New functions.
1419 (vector_init_bshuffle): Rewrite to handle more cases and make use
1420 of locs[] array prepared by vector_init_prepare_elts.
1421 (vector_init_fpmerge, vector_init_faligndata): Delete.
1422 (sparc_expand_vector_init): Rewrite using new infrastructure.
1424 2011-11-05 Joern Rennecke <joern.rennecke@embecosm.com>
1426 * config.gcc (epiphany-*-*): New architecture.
1427 (epiphany-*-elf): New configuration.
1428 * config/epiphany, common/config/epiphany : New directories.
1429 * doc/extend.texi (disinterrupt attribute): Add Epiphany.
1430 (interrupt attribute): Add Epiphany.
1431 (long_call, short_call attribute): Add Epiphany.
1432 * doc/invoke.texi (Options): Add Epiphany options.
1433 * doc/md.texi (Machine Constraints): Add Epiphany constraints.
1434 * doc/install.texi (Options specification):
1435 Add --with-stack-offset=@var{num} description.
1436 (host/target specific issues): Add epiphany-*-elf.
1437 * doc/contrib.texi (Contributors): Mention Epiphany port.
1439 2011-11-05 Jakub Jelinek <jakub@redhat.com>
1441 PR tree-optimization/50693
1442 * tree-cfg.c (gimple_can_merge_blocks_p): Allow merging with
1443 non-forced user labels.
1444 (gimple_merge_blocks): Turn non-forced user labels into
1445 debug bind stmt with the label as first operand and reset value.
1446 (gimple_duplicate_bb): Don't duplicate label debug stmts.
1447 * dwarf2out.c (gen_label_die): Handle NOTE_INSN_DELETED_DEBUG_LABEL.
1448 * final.c (final_scan_insn): Likewise.
1449 (rest_of_clean_state): Don't dump NOTE_INSN_DELETED_DEBUG_LABEL.
1450 * var-tracking.c (debug_label_num): New variable.
1451 (delete_debug_insns): Don't delete DEBUG_INSNs for LABEL_DECLs,
1452 instead turn them into NOTE_INSN_DELETED_DEBUG_LABEL notes.
1453 * cfglayout.c (skip_insns_after_block, duplicate_insn_chain): Handle
1454 NOTE_INSN_DELETED_DEBUG_LABEL.
1455 (duplicate_insn_chain): Don't duplicate LABEL_DECL DEBUG_INSNs.
1456 * insn-notes.def (DELETED_DEBUG_LABEL): New note kind.
1457 * print-rtl.c (print_rtx): Handle NOTE_INSN_DELETED_DEBUG_LABEL.
1458 * gengtype.c (adjust_field_rtx_def): Likewise.
1459 * config/i386/i386.c (ix86_output_function_epilogue): For MachO
1460 clear CODE_LABEL_NUMBER of NOTE_INSN_DELETED_DEBUG_LABEL
1461 if their are at the end of function and nop hasn't been emitted.
1462 * config/rs6000/rs6000.c (rs6000_output_function_epilogue): Likewise.
1464 2011-11-05 Georg-Johann Lay <avr@gjlay.de>
1466 PR rtl-optimization/50448
1467 * cprop.c (try_replace_reg): Also try to replace uses of FROM that
1470 2011-11-05 Peter Dufault <dufault@hda.com>,
1471 Sebastian Huber <sebastian.huber@embedded-brains.de>
1473 * config/rs6000/rtems.h (SUBSUBTARGET_OVERRIDE_OPTIONS):
1474 Extend rs6000_spe handling.
1476 2011-11-05 Ralf Corsépius <ralf.corsepius@rtems.org>
1478 * config/rs6000/t-rtems: Add -mcpu=8540/-mfloat-gprs=double multilib.
1479 Remove -mcpu=601 multilib.
1480 Remove -Dmpc8260 multilib.
1481 * config/rs6000/rtems.h: Allow --float-gprs=... to override grps
1484 2011-11-05 Quentin Neill <quentin.neill@amd.com>
1486 Piledriver f16cintrin.h fix.
1487 * config/i386/f16cintrin.h: Contents moved from immintrin.h.
1488 * config/i386/immintrin.h: Include f16cintrin.h.
1489 * config.gcc (i[34567]86-*-*, x86_64-*-*): Add f16cintrin.h.
1492 2011-11-04 Eric Botcazou <ebotcazou@adacore.com>
1495 * c-parser.c (c_parser_postfix_expression) <RID_OFFSETOF>: Adjust call
1497 * c-typeck.c (build_unary_op) <ADDR_EXPR>: Call fold_offsetof_1.
1499 2011-11-04 Alan Modra <amodra@gmail.com>
1501 * reload1.c (gen_reload): Don't use REGNO on SUBREGs.
1502 * print-rtl.c (print_rtx): Don't segfault on negative regno.
1504 2011-11-04 David S. Miller <davem@davemloft.net>
1507 * config/sparc/sparc.c (sparc_expand_conditional_move): Handle the
1508 fact that sparc_emit_float_lib_cmp modifies the comparison in
1511 2011-11-04 Ralf Corsépius <ralf.corsepius@rtems.org>
1513 * config/lm32/t-rtems: New.
1514 * config.gcc (lm32-*-rtems*): Add t-rtems.
1516 2011-11-04 Eric Botcazou <ebotcazou@adacore.com>
1519 * config/sparc/sparc.h (ASM_CPU_SPEC): Pass -Av8 if -mcpu=v8.
1521 2011-11-04 Jiangning Liu <jiangning.liu@arm.com>
1523 PR rtl-optimization/38644
1524 * config/arm/arm.c (thumb1_expand_epilogue): Add memory barrier
1525 for epilogue having stack adjustment.
1527 2011-11-04 Georg-Johann Lay <avr@gjlay.de>
1530 * config/avr/avr-modes.def: New file defining PSImode.
1531 * config/avr/avr-c.c (__INT24_MAX__, __INT24_MIN__,
1532 __UINT24_MAX__): New built-in defines.
1533 * config/avr/avr.md (adjust_len): Add tstpsi, mov24, reload_in24,
1534 ashlpsi, ashrpsi, lshrpsi.
1535 (QISI, QIDI, HISI, HIDI, MPUSH, rotx, rotsmode): Add PSI.
1536 (MOVMODE): New mode iterator.
1537 (movpsi): New expander.
1538 (movqi, movhi, movsi, movsf, movpsi): Write as one using MOVMODE.
1539 (*reload_inpsi, *movpsi): New insns.
1540 (*reload_inpsi): New RTL peephole.
1541 (addpsi3, *addpsi3_zero_extend.qi, *addpsi3_zero_extend.hi,
1542 *addpsi3_sign_extend.hi): New insns.
1543 (subpsi3, *subpsi3_zero_extend.qi, *subpsi3_zero_extend.hi,
1544 *subpsi3_sign_extend.hi): New insns.
1545 (divmodpsi4, udivmodpsi4): New define insn-and-split.
1546 (*divmodpsi4_call, *udivmodpsi4_call): New insns.
1547 (andpsi3, iorpsi3, xorpsi3): New insns.
1548 (*rotlpsi2.1, *rotlpsi2.23): New insns.
1549 (*rotw<mode>): Insn condition only allow even-sized modes.
1550 (*rotb<mode>): Insn condition allows odd-sized modes.
1551 (ashlpsi3, ashrpsi3, lshrpsi3, *addpsi3.lt0): New insns.
1552 (negpsi2, one_cmplpsi2): New insns.
1553 (extendqipsi2, extendhipsi2, extendpsisi2): New insns.
1554 (zero_extendqipsi2, zero_extendhipsi2, zero_extendpsisi2): New
1556 (*cmppsi, *negated_tstpsi, *reversed_tstpsi): New insns.
1557 (cbranchpsi4): New expander.
1558 * config/avr/constraints.md (Ca3, Co3, Cx3): New constraints.
1559 * config/avr/avr-protos.h (avr_out_tstpsi, avr_out_movpsi,
1560 avr_out_ashlpsi3, avr_out_ashrpsi3, avr_out_lshrpsi3,
1561 avr_out_reload_inpsi): New prototypes.
1563 * config/avr/avr.c (TARGET_SCALAR_MODE_SUPPORTED_P): Define to...
1564 (avr_scalar_mode_supported_p): ...this new static function.
1565 (avr_asm_len): Always return "".
1566 (avr_out_load_psi, avr_out_store_psi): New static functions.
1567 (avr_out_movpsi, avr_out_reload_inpsi): New functions.
1568 (avr_out_tstpsi): New function.
1569 (avr_out_ashlpsi3, avr_out_ashrpsi3, avr_out_lshrpsi3): New functions.
1570 (avr_out_plus_1, output_reload_in_const): Handle 3-byte types.
1571 (avr_simplify_comparison_p): Ditto.
1572 (adjust_insn_length): Handle ADJUST_LEN_RELOAD_IN24,
1573 ADJUST_LEN_MOV24, ADJUST_LEN_TSTPSI, ADJUST_LEN_ASHLPSI,
1574 ADJUST_LEN_ASHRPSI, ADJUST_LEN_LSHRPSI.
1575 (avr_rtx_costs_1): Report PSI costs.
1576 (avr_libcall_value): Handle odd-sized parameters.
1577 (avr_init_builtin_int24): New static function to define built-in
1578 24-bit types __int24 and __uint24.
1579 (avr_init_builtins): Use it.
1581 2011-11-04 Thomas Doerfler <thomas.doerfler@embedded-brains.de>
1584 * config/arm/rtems-elf.h, config/arm/t-rtems: Add optional
1585 support for VFP floating point model.
1587 2011-11-04 Tristan Gingold <gingold@adacore.com>
1589 * config/alpha/vms.h (ASM_OUTPUT_DEF): Do not switch section.
1591 2011-11-04 Ira Rosen <ira.rosen@linaro.org>
1594 2011-10-24 Ira Rosen <ira.rosen@linaro.org>
1596 PR tree-optimization/50730
1597 * tree-vect-data-refs.c (vect_analyze_data_refs): Stop basic block
1598 analysis if encountered unsupported data-ref.
1600 2011-11-04 Jakub Jelinek <jakub@redhat.com>
1602 * config/i386/i386.c (ix86_expand_vector_convert_uns_vsivsf): New
1604 * config/i386/i386-protos.h (ix86_expand_vector_convert_uns_vsivsf):
1606 * config/i386/sse.md (floatuns<sseintvecmodelower><mode>2): Use it.
1607 For floatunsv8siv8sf2 require TARGET_AVX2.
1609 * config/i386/i386.c (ix86_expand_adjust_ufix_to_sfix_si): Add
1610 XORP argument. Subtract 0x1p31 instead of 0x1p32. Use normal
1611 signalling comparison instead of non-signalling. Store into
1612 *XORP pseudo holding 0x80000000 integers if 0x1p31 has been
1613 subtracted and 0 otherwise.
1614 * config/i386/i386-protos.h (ix86_expand_adjust_ufix_to_sfix_si):
1616 * config/i386/sse.md (fixuns_trunc<mode><sseintvecmodelower>2): Enable
1617 already for TARGET_SSE2. Xor in vector initialized by
1618 ix86_expand_adjust_ufix_to_sfix_si at the end.
1619 (vec_pack_ufix_trunc_<mode>): Likewise.
1621 * tree-vect-stmts.c (vectorizable_conversion): Rewritten to handle
1622 not just FLOAT_EXPR and FIX_TRUNC_EXPR, but also CONVERT_EXPR_CODE_P,
1623 WIDEN_MULT_EXPR and WIDEN_LSHIFT_EXPR to handle what
1624 vectorizable_type_demotion and vectorizable_type_promotion did.
1625 Additionally handle FLOAT_EXPR and FIX_TRUNC_EXPR where the integer
1626 is {,un}signed {char,short}.
1627 (vect_create_vectorized_demotion_stmts): Fix comment typo. For
1628 recursive calls unconditionally use VEC_PACK_TRUNC_EXPR.
1629 Push vec_dest back to the vec_dsts vector at the end.
1630 (vect_create_vectorized_promotion_stmts): Don't recurse, do just
1631 one step. Removed multi_step_cvt, vec_dsts, slp_node and
1632 prev_stmt_info arguments, add vec_dest argument. Push always
1633 into vec_tmp, not just when multi_step_cvt != 0, replace *vec_oprdn0
1634 with vec_tmp at the end after freeing old *vec_oprnd0 vector.
1635 (vectorizable_type_demotion, vectorizable_type_promotion): Removed.
1636 (vect_analyze_stmt): Don't call vectorizable_type_demotion and
1637 vectorizable_type_promotion. Call vectorizable_conversion even
1638 for SLP bb vectorization.
1639 (vect_transform_stmt): Call vectorizable_conversion instead of
1640 vectorizable_type_demotion and vectorizable_type_promotion.
1641 (supportable_widening_operation): Clear *multi_step_cvt first,
1642 simplify c1/c2 computation, free *interm_types vector on failure.
1643 (supportable_narrowing_operation): Clear *multi_step_cvt first,
1644 free *interm_types vector on failure, handle multi-step
1647 2011-11-04 Tristan Gingold <gingold@adacore.com>
1649 * config/alpha/alpha.c (alpha_write_linkage): Remove fundecl
1650 argument. Conditionally generate crash debug info. Adjust
1651 for alpha_funcs_tree removal.
1652 (machine_function): Add links field.
1653 (alpha_start_function): Conditionally generate crash debug info.
1654 (alpha_end_function): Adjust call to alpha_write_linkage.
1655 (alpha_funcs): Remove.
1656 (links_kind): Remove.
1657 (alpha_links): Remove num, target and lkind field. Add func field.
1658 (alpha_links_tree): Remove.
1659 (alpha_funcs_tree): Remove.
1660 (alpha_need_linkage): Remove.
1661 (alpha_use_linkage): Change prototype. Adjust.
1662 (alpha_write_one_linkage): Use ASM_OUTPUT_INTERNAL_LABEL.
1663 Use SYMBOL_REF_EXTERNAL_P and SYMBOL_REF_LOCAL_P macro.
1664 * config/alpha/alpha-protos.h (alpha_use_linkage): Update.
1665 (alpha_need_linkage): Remove.
1666 * config/alpha/alpha.md: Update calls to alpha_use_linkage.
1667 Adjust calls to alpha_need_linkage.
1669 2011-11-03 Uros Bizjak <ubizjak@gmail.com>
1671 * sched-vis.c (print_value): Handle STRICT_LOW_PART.
1673 2011-11-03 Uros Bizjak <ubizjak@gmail.com>
1675 * config/i386/i386.md (lround<X87MODEF:mode><SWI248x:mode>2,
1676 rint<mode>2, floor<mode>2, lfloor<MODEF:mode><SWI48:mode>2,
1677 btrunc<mode>2, lwp_lwpval<mode>3): Use operands[N] instead of operandN.
1679 2011-11-03 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
1682 * config/arm/t-bpabi: New file.
1683 * config.gcc (arm*-*-linux*): Add arm/t-bpabi to tmake_file for
1685 (arm*-*-uclinux*): Add arm/t-bpabi to tmake_file for
1686 arm*-*-uclinux*eabi.
1687 (arm*-*-eabi*, arm*-*-symbianelf*): Add arm/t-bpabi to tmake_file
1690 2011-11-03 Michael Matz <matz@suse.de>
1693 * configure.ac: Check for -fno-exceptions -fno-rtti.
1694 * configure: Regenerate.
1695 * Makefile.in (NOEXCEPTION_FLAGS): New flag.
1696 (ALL_CXXFLAGS): Use it.
1698 2011-11-03 Uros Bizjak <ubizjak@gmail.com>
1700 * config/i386/i386.md: Use {} for multi-line preparation statements.
1702 2011-11-03 Eric Botcazou <ebotcazou@adacore.com>
1704 * config/sparc/sparc.md (movtf_insn_sp32_no_fpu): Consolidate into...
1705 (movtf_insn_sp32): ...this.
1706 (movtf_insn_sp64_no_fpu): Consolidate into...
1707 (movtf_insn_sp64): ...this.
1708 (movtf_insn_sp64_hq): Do not test TARGET_FPU.
1709 * config/sparc/sparc.c (sparc_legitimate_address_p): Likewise.
1711 2011-11-03 Tristan Gingold <gingold@adacore.com>
1713 * config/vms/vms.c (vms_patch_builtins): Fix typo.
1715 2011-11-03 Richard Guenther <rguenther@suse.de>
1718 * lto-opts.c: Re-implement.
1719 * lto-streamer.h (lto_register_user_option): Remove.
1720 (lto_read_file_options): Likewise.
1721 (lto_reissue_options): Likewise.
1722 (lto_clear_user_options): Likewise.
1723 (lto_clear_file_options): Likewise.
1724 * opts-global.c (post_handling_callback): Remove.
1725 (set_default_handlers): Do not set post_handling_callback.
1726 (decode_options): Remove LTO specific code.
1727 * lto-wrapper.c (merge_and_complain): New function.
1728 (run_gcc): Read all input file options and
1729 prepend a merged set before the linker driver options.
1730 * gcc.c (driver_post_handling_callback): Remove.
1731 (set_option_handlers): Do not set post_handling_callback.
1732 * opts-common.c (handle_option): Do not call post_handling_callback.
1733 * opts.h (struct cl_option_handlers): Remove post_handling_callback.
1735 2011-11-03 Richard Guenther <rguenther@suse.de>
1737 * collect2.c (main): Guard object_nbr variable with TARGET_AIX_VERSION.
1739 2011-11-03 Martin Jambor <mjambor@suse.cz>
1741 * ipa-prop.c (type_change_info): New fields offset, object,
1742 known_current_type and multiple_types_encountered.
1743 (extr_type_from_vtbl_ptr_store): New function.
1744 (check_stmt_for_type_change): Use it, set multiple_types_encountered if
1745 the result is different from the previous one.
1746 (detect_type_change): Renamed to detect_type_change_1. New parameter
1747 comp_type. Set up new fields in tci, build known type jump
1748 functions if the new type can be identified.
1749 (detect_type_change): New function.
1750 * tree.h (DECL_CONTEXT): Comment new use.
1752 2011-11-03 Richard Guenther <rguenther@suse.de>
1755 * lto-wrapper.c (get_options_from_collect_gcc_options): Properly
1756 decode an encoded literal '.
1758 2011-11-03 Tristan Gingold <gingold@adacore.com>
1760 * collect2.c (main): Add support of -f (response file) on AIX.
1762 2011-11-03 Ira Rosen <ira.rosen@linaro.org>
1764 PR tree-optimization/50912
1765 * tree-vectorizer.h (slp_void_p): New.
1766 (struct _slp_tree): Replace left and right with children. Update
1768 (struct _slp_oprnd_info): New.
1769 (vect_get_vec_defs): Declare.
1770 (vect_get_slp_defs): Update arguments.
1771 * tree-vect-loop.c (vect_create_epilog_for_reduction): Call
1772 vect_get_vec_defs instead of vect_get_slp_defs.
1773 (vectorizable_reduction): Likewise.
1774 * tree-vect-stmts.c (vect_get_vec_defs): Remove static, add argument.
1775 Update call to vect_get_slp_defs.
1776 (vectorizable_conversion): Update call to vect_get_vec_defs.
1777 (vectorizable_assignment, vectorizable_shift,
1778 vectorizable_operation): Likewise.
1779 (vectorizable_type_demotion): Call vect_get_vec_defs instead of
1781 (vectorizable_type_promotion, vectorizable_store): Likewise.
1782 (vect_analyze_stmt): Fix typo.
1783 * tree-vect-slp.c (vect_free_slp_tree): Update SLP tree traversal.
1784 (vect_print_slp_tree, vect_mark_slp_stmts,
1785 vect_mark_slp_stmts_relevant, vect_slp_rearrange_stmts,
1786 vect_detect_hybrid_slp_stmts, vect_slp_analyze_node_operations,
1787 vect_schedule_slp_instance): Likewise.
1788 (vect_create_new_slp_node): New.
1789 (vect_create_oprnd_info, vect_free_oprnd_info): Likewise.
1790 (vect_get_and_check_slp_defs): Pass information about defs using
1791 oprnds_info, allow any number of operands.
1792 (vect_build_slp_tree): Likewise. Update calls to
1793 vect_get_and_check_slp_defs. Fix comments.
1794 (vect_analyze_slp_instance): Move node creation to
1795 vect_create_new_slp_node.
1796 (vect_get_slp_defs): Allow any number of operands.
1798 2011-11-02 Peter Bergner <bergner@vnet.ibm.com>
1799 Iain Sandoe <iains@gcc.gnu.org>
1801 * config/rs6000/rs6000.c (USE_HIDDEN_LINKONCE): New define.
1802 (get_ppc476_thunk_name): Use it.
1803 (rs6000_code_end): Likewise.
1804 (macho_branch_islands): Fix typo.
1806 2011-11-02 Paolo Carlini <paolo.carlini@oracle.com>
1807 Jason Merrill <jason@redhat.com>
1810 * configure.ac: Add -Wno-narrowing to warning options.
1811 * doc/invoke.texi ([-Wnarrowing], [-Wc++0x-compat]): Update.
1813 2011-11-02 Eric Botcazou <ebotcazou@adacore.com>
1816 * config/sparc/sparc.md (movsf_insn): Reindent constraints.
1817 (movdf_insn_sp32): Likewise. Remove redundant G constraint.
1818 (movdf_insn_sp64): Likewise.
1819 (DFmode splitter): Do not test TARGET_FPU.
1820 (movtf_insn_sp32): Reindent constraints.
1821 (movtf_insn_sp32_no_fpu): Likewise.
1822 (movtf_insn_sp64): Likewise.
1823 (movtf_insn_sp64_hq): Likewise.
1824 (movtf_insn_sp64_no_fpu): Likewise.
1826 2011-11-02 Paolo Carlini <paolo.carlini@oracle.com>
1829 * builtins.c (fold_builtin_memchr): Fix cast.
1831 2011-11-02 Teresa Johnson <tejohnson@google.com>
1833 * config/i386/predicates.md (promotable_binary_operator): Add minus
1834 to the list of promotable operators.
1836 2011-11-02 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
1838 * gthr-single.h, gthr.h: Move to ../libgcc.
1839 * gthr-aix.h: Move to ../libgcc/config/rs6000.
1840 * gthr-dce.h: Move to ../libgcc/config/pa.
1841 * gthr-lynx.h: Move to ../libgcc/config.
1842 * gthr-mipssde.h: Move to ../libgcc/config/mips.
1843 * gthr-posix.h: Move to ../libgcc/config.
1844 * gthr-rtems.h: Likewise.
1845 * gthr-tpf.h: Move to ../libgcc/config/s390.
1846 * gthr-vxworks.h: Move to ../libgcc/config.
1847 * gthr-win32.h: Move to ../libgcc/config/i386.
1848 * configure.ac (gthread_flags): Remove
1849 (gthr-default.h): Don't create.
1850 (thread_file): Don't substitute.
1851 * configure: Regenerate.
1852 * Makefile.in (GCC_THREAD_FILE): Remove.
1853 (GTHREAD_FLAGS): Remove.
1854 (libgcc.mvars): Remove GTHREAD_FLAGS.
1855 * config/t-vxworks (EXTRA_HEADERS): Remove.
1857 2011-11-02 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
1858 Paolo Bonzini <bonzini@gnu.org>
1860 * configure.ac (libgcc_tm_file_list, libgcc_tm_include_list): Remove.
1861 * configure: Regenerate.
1862 * Makefile.in (libgcc_tm_file_list, libgcc_tm_include_list): Remove.
1863 (TM_H): Remove libgcc_tm.h, $(libgcc_tm_file_list).
1864 (libgcc_tm.h, cs-libgcc_tm.h): Remove.
1865 (clean): Remove libgcc_tm.h
1866 * mkconfig.sh: Don't include libgcc_tm.h in tm.h.
1867 * config.gcc (libgcc_tm_file): Remove.
1868 (arm*-*-linux*): Remove libgcc_tm_file for arm*-*-linux-*eabi.
1869 (arm*-*-uclinux*): Remove libgcc_tm_file for arm*-*-uclinux*eabi.
1870 (arm*-*-eabi*, arm*-*-symbianelf*): Remove libgcc_tm_file.
1871 (avr-*-rtems*): Likewise.
1872 (avr-*-*): Likewise.
1873 (frv-*-elf): Likewise.
1874 (frv-*-*linux*): Likewise.
1875 (h8300-*-rtems*): Likewise.
1876 (h8300-*-elf*): Likewise.
1877 (i[34567]86-*-darwin*): Likewise.
1878 (x86_64-*-darwin*): Likewise.
1879 (rx-*-elf*): Likewise.
1880 (tic6x-*-elf): Likewise.
1881 (tic6x-*-uclinux): Likewise.
1882 (i[34567]86-*-linux*, x86_64-*-linux*): Likewise.
1884 2011-11-02 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
1886 * Makefile.in (LIBGCC2_DEBUG_CFLAGS LIBGCC2_CFLAGS)
1887 (LIBGCC2_INCLUDES, TARGET_LIBGCC2_CFLAGS, LIB2FUNCS_EXTRA)
1888 (LIB2FUNCS_STATIC_EXTRA, LIB2FUNCS_EXCLUDE, T, T_TARGET)
1889 (INCLUDES_FOR_TARGET): Remove.
1890 (LIBGCC2_CFLAGS): Don't export.
1891 (LIB2FUNCS_ST, LIB2_DIVMOD_FUNCS, LIB2ADD, LIB2ADD_ST, srcdirify):
1893 (libgcc-support): Remove $(LIB2ADD), $(LIB2ADD_ST) dependencies.
1894 (libgcc.mvars): Likewise.
1895 Don't emit LIB2FUNCS_ST, LIB2FUNCS_EXCLUDE, LIB2ADD, LIB2ADD_ST,
1896 LIB2_SIDITI_CONV_FUNCS, LIB2_DIVMOD_FUNCS, LIBGCC2_CFLAGS,
1897 TARGET_LIBGCC2_CFLAGS.
1899 * libgcc2.c, libgcc2.h, gbl-ctors.h, longlong.h: Move to ../libgcc.
1900 * config/darwin-64.c: Move to ../libgcc/config.
1901 * config/divmod.c, config/floatunsidf.c, config/floatunsisf.c,
1902 config/floatunsitf.c, config/floatunsixf.c, config/udivmod.c,
1903 config/udivmodsi4.c: Move to ../libgcc/config.
1904 * config/gthr-posix.c: Move to ../libgcc/config/alpha.
1905 * config/memcmp.c, config/memcpy.c, config/memmove.c,
1906 config/memset.c: Move to ../libgcc/config.
1907 * config/t-darwin (TARGET_LIBGCC2_CFLAGS): Remove.
1908 * config/t-freebsd: Remove.
1909 * config/t-freebsd-thread: Move to ../libgcc/config.
1910 * config/t-libgcc-pic: Move to ../libgcc/config.
1911 * config/t-libunwind (TARGET_LIBGCC2_CFLAGS): Remove.
1912 * config/t-linux: Remove.
1913 * config/t-lynx (TARGET_LIBGCC2_CFLAGS, LIBGCC, INSTALL_LIBGCC):
1915 * config/t-openbsd-thread: Move to ../libgcc/config.
1916 * config/t-rtems (LIBGCC2_INCLUDES): Remove.
1917 * config/t-sol2 (TARGET_LIBGCC2_CFLAGS): Remove.
1918 * config/t-svr4: Remove.
1919 * config/t-vxworks (LIBGCC, INSTALL_LIBGCC, TARGET_LIBGCC2_CFLAGS)
1920 (LIBGCC2_DEBUG_CFLAGS, LIB2FUNCS_EXTRA, LIBGCC2_INCLUDES): Remove.
1921 * config/vxlib.c, config/vxlib-tls.c: Move to ../libgcc/config.
1922 * config/alpha/qrnnd.asm: Move to ../libgcc/config/alpha/qrnnd.S.
1923 * config/alpha/t-alpha, config/alpha/t-ieee: Remove.
1924 * config/alpha/t-vms (LIB2FUNCS_EXTRA, LIBGCC, INSTALL_LIBGCC): Remove.
1925 * config/alpha/vms-gcc_shell_handler.c: Move to ../libgcc/config/alpha.
1926 * config/arm/bpabi.c, config/arm/unaligned-funcs.c,
1927 config/arm/fp16.c, config/arm/linux-atomic.c,
1928 config/arm/linux-atomic-64bit.c: Move to ../libgcc/config/arm.
1929 * config/arm/t-arm-elf (LIBGCC, INSTALL_LIBGCC)
1930 (TARGET_LIBGCC2_CFLAGS): Remove.
1931 * config/arm/t-bpabi, config/arm/t-linux: Remove.
1932 * config/arm/t-linux-eabi (TARGET_LIBGCC2_CFLAGS)
1933 (LIB2FUNCS_STATIC_EXTRA): Remove.
1934 * config/arm/t-netbsd: Remove.
1935 * config/arm/t-strongarm-elf (LIBGCC, INSTALL_LIBGCC)
1936 (TARGET_LIBGCC2_CFLAGS): Remove.
1937 * config/arm/t-symbian (LIB2FUNCS_STATIC_EXTRA): Remove.
1938 * config/arm/t-wince-pe (LIBGCC, INSTALL_LIBGCC)
1939 (TARGET_LIBGCC2_CFLAGS): Remove.
1940 * config/avr/t-avr (LIB2FUNCS_EXCLUDE, TARGET_LIBGCC2_CFLAGS)
1941 (LIBGCC, INSTALL_LIBGCC): Remove.
1942 * config/bfin/t-bfin-elf (TARGET_LIBGCC2_CFLAGS): Remove.
1943 * config/bfin/t-bfin-linux: Likewise.
1944 * config/bfin/t-bfin-uclinux: Likewise.
1945 * config/c6x/eqd.c, config/c6x/eqf.c, config/c6x/ged.c,
1946 config/c6x/gef.c, config/c6x/gtd.c, config/c6x/gtf.c,
1947 config/c6x/led.c, config/c6x/lef.c, config/c6x/ltd.c,
1948 config/c6x/ltf.c: Move to ../libgcc/config/c6x.
1949 * config/c6x/t-c6x-elf (LIB2FUNCS_EXCLUDE, LIB2FUNCS_EXTRA): Remove.
1950 * config/c6x/t-c6x-uclinux (TARGET_LIBGCC2_CFLAGS): Remove.
1951 * config/cris/arit.c: Move to ../libgcc/config/cris.
1952 * config/cris/cris_abi_symbol.c: Remove.
1953 * config/cris/cris.h: Remove obsolete comment.
1954 * config/cris/mulsi3.asm: Move to ../libgcc/config/cris/mulsi3.S.
1955 * config/cris/t-cris (LIB2FUNCS_EXTRA, CRIS_LIB1CSRC)
1956 ($(LIB2FUNCS_EXTRA)): Remove.
1957 * config/cris/t-elfmulti (LIB2FUNCS_STATIC_EXTRA, INSTALL_LIBGCC)
1959 * config/cris/t-linux (TARGET_LIBGCC2_CFLAGS): Remove.
1960 * config/fr30/t-fr30: Remove.
1961 * config/frv/cmovd.c, config/frv/cmovh.c, config/frv/cmovw.c,
1962 config/frv/modi.c, config/frv/uitod.c, config/frv/uitof.c,
1963 config/frv/ulltod.c, config/frv/ulltof.c, config/frv/umodi.c: Move
1964 to ../libgcc/config/frv.
1965 * config/frv/t-frv (LIB2FUNCS_EXTRA, TARGET_LIBGCC2_CFLAGS)
1966 (cmovh.c, cmovw.c, cmovd.c, modi.c, umodi.c, uitof.c, uitod.c)
1967 (ulltof.c, LIBGCC, INSTALL_LIBGCC): Remove.
1968 * config/frv/t-linux (TARGET_LIBGCC2_CFLAGS): Remove.
1969 * config/h8300/clzhi2.c, config/h8300/ctzhi2.c,
1970 config/h8300/fixunssfsi.c, config/h8300/parityhi2.c,
1971 config/h8300/popcounthi2.c: Move to ../libgcc/config/h8300.
1972 * config/h8300/t-h8300 (LIB2FUNCS_EXTRA, TARGET_LIBGCC2_CFLAGS)
1973 (LIBGCC, INSTALL_LIBGCC): Remove.
1974 * config/i386/gthr-win32.c: Move to ../libgcc/config/i386.
1975 * config/i386/t-cygming (LIBGCC2_INCLUDES): Remove.
1976 * config/i386/t-cygwin: Remove.
1977 * config/i386/t-darwin (LIB2_SIDITI_CONV_FUNCS, LIB2FUNCS_EXTRA)
1978 (LIB2FUNCS_EXCLUDE): Remove.
1979 * config/i386/t-darwin64 (LIB2_SIDITI_CONV_FUNCS, LIB2FUNCS_EXTRA)
1980 (LIBGCC, INSTALL_LIBGCC): Remove.
1981 * config/i386/t-gthr-win32: Move to ../libgcc/config/i386.
1982 * config/i386/t-linux64 (LIBGCC, INSTALL_LIBGCC): Remove.
1983 * config/i386/t-mingw-w32: Likewise.
1984 * config/i386/t-mingw-w64: Likewise.
1985 * config/i386/t-openbsd: Likewise.
1986 * config/i386/t-nto: Remove.
1987 * config/ia64/quadlib.c: Move to ../libgcc/config/ia64.
1988 * config/ia64/t-hpux (LIBGCC, INSTALL_LIBGCC, LIB2FUNCS_EXTRA)
1989 (quadlib.c): Remove.
1990 * config/ia64/t-ia64: Remove comment.
1991 * config/iq2000/lib2extra-funcs.c: Move to
1992 ../libgcc/config/iq2000/lib2funcs.c.
1993 * config/iq2000/t-iq2000: Remove.
1994 * config/m32c/m32c-lib2.c: Move to ../libgcc/config/m32c/lib2funcs.c.
1995 * config/m32c/m32c-lib2-trapv.c: Move to ../libgcc/config/m32c/trapv.c.
1996 * config/m32r/t-linux (TARGET_LIBGCC2_CFLAGS): Remove.
1997 * config/m32c/t-m32c (LIB2FUNCS_EXTRA): Remove.
1998 * config/m32r/t-m32r (TARGET_LIBGCC2_CFLAGS, LIBGCC)
1999 (INSTALL_LIBGCC): Remove.
2000 * config/m68k/fpgnulib.c: Move to ../libgcc/config/m68k.
2001 * config/m68k/t-floatlib: Remove.
2002 * config/m68k/t-mlibs (LIBGCC, INSTALL_LIBGCC): Remove.
2003 * config/mcore/t-mcore (TARGET_LIBGCC2_CFLAGS): Remove.
2005 (LIBGCC, INSTALL_LIBGCC): Remove.
2006 * config/mep/mep-lib2.c: Move to ../libgcc/config/mep/lib2funcs.c.
2007 * config/mep/mep-tramp.c: Move to ../libgcc/config/mep/tramp.c.
2008 * config/mep/t-mep (LIB2FUNCS_EXTRA): Remove.
2009 * config/mips/t-elf (TARGET_LIBGCC2_CFLAGS, LIBGCC)
2010 (INSTALL_LIBGCC): Remove.
2011 * config/mips/t-isa3264: Likewise.
2012 * config/mips/t-mips (LIB2_SIDITI_CONV_FUNCS): Remove.
2013 * config/mips/t-r3900 (TARGET_LIBGCC2_CFLAGS, LIBGCC)
2014 (INSTALL_LIBGCC): Remove.
2015 * config/mips/t-sde (LIBGCC, INSTALL_LIBGCC): Remove.
2016 * config/mips/t-sr71k (TARGET_LIBGCC2_CFLAGS, LIBGCC)
2017 (INSTALL_LIBGCC): Remove.
2018 * config/mips/t-vr (TARGET_LIBGCC2_CFLAGS)
2019 (LIB2FUNCS_STATIC_EXTRA): Remove.
2020 * config/mips/vr4120-div.S: Move to ../libgcc/config/mips.
2021 * config/mmix/t-mmix (TARGET_LIBGCC2_CFLAGS): Remove.
2022 * config/mn10300/t-mn10300 (LIBGCC, INSTALL_LIBGCC): Remove.
2023 * config/pa/fptr.c, config/pa/linux-atomic.c: Move to
2024 ../libgcc/config/pa.
2025 * config/pa/lib2funcs.asm: Move to ../libgcc/config/pa/lib2funcs.S.
2026 * config/pa/quadlib.c: Move to ../libgcc/config/pa.
2027 * config/pa/t-dce-thr (LIBGCC, INSTALL_LIBGCC): Remove.
2028 * config/pa/t-linux, config/pa/t-linux64: Remove.
2029 * config/pa/t-pa-hpux, config/pa/t-pa-hpux10,
2030 config/pa/t-pa-hpux11, config/pa/t-pa64: Remove.
2031 * config/pdp11/t-pdp11 (TARGET_LIBGCC2_CFLAGS, LIB2FUNCS_EXTRA):
2033 * config/picochip/libgccExtras: Move to ../libgcc/config/picochip.
2034 * config/picochip/t-picochip (LIB2FUNCS_EXTRA, RANLIB_FOR_TARGET)
2035 (TARGET_LIBGCC2_CFLAGS, LIBGCC2_DEBUG_CFLAGS): Remove.
2036 * config/rs6000/crtresfpr.asm: Move to
2037 ../libgcc/config/rs6000/crtresfpr.S.
2038 * config/rs6000/crtresgpr.asm: Move to
2039 ../libgcc/config/rs6000/crtresgpr.S.
2040 * config/rs6000/crtresxfpr.asm: Move to
2041 ../libgcc/config/rs6000/crtresxfpr.S.
2042 * config/rs6000/crtresxgpr.asm: Move to
2043 ../libgcc/config/rs6000/crtresxgpr.S.
2044 * config/rs6000/crtsavfpr.asm: Move to
2045 ../libgcc/config/rs6000/crtsavfpr.S.
2046 * config/rs6000/crtsavgpr.asm: Move to
2047 ../libgcc/config/rs6000/crtsavgpr.S.
2048 * config/rs6000/darwin-asm.h: Move to ../libgcc/config/rs6000.
2049 * config/rs6000/darwin-fpsave.asm: Move to
2050 ../libgcc/config/rs6000/darwin-fpsave.S.
2051 * config/rs6000/darwin-gpsave.asm: Move to
2052 ../libgcc/config/rs6000/darwin-gpsave.S.
2053 * config/rs6000/darwin-tramp.asm: Move to
2054 ../libgcc/config/rs6000/darwin-tramp.S.
2055 * config/rs6000/darwin-vecsave.asm: Move to
2056 ../libgcc/config/rs6000/darwin-vecsave.S.
2057 * config/rs6000/darwin-world.asm: Move to
2058 ../libgcc/config/rs6000/darwin-world.S.
2059 * config/rs6000/e500crtres32gpr.asm: Move to
2060 ../libgcc/config/rs6000/e500crtres32gpr.S.
2061 * config/rs6000/e500crtres64gpr.asm: Move to
2062 ../libgcc/config/rs6000/e500crtres64gpr.S.
2063 * config/rs6000/e500crtres64gprctr.asm: Move to
2064 ../libgcc/config/rs6000/e500crtres64gprctr.S.
2065 * config/rs6000/e500crtrest32gpr.asm: Move to
2066 ../libgcc/config/rs6000/e500crtrest32gpr.S.
2067 * config/rs6000/e500crtrest64gpr.asm: Move to
2068 ../libgcc/config/rs6000/e500crtrest64gpr.S.
2069 * config/rs6000/e500crtresx32gpr.asm: Move to
2070 ../libgcc/config/rs6000/e500crtresx32gpr.S.
2071 * config/rs6000/e500crtresx64gpr.asm: Move to
2072 ../libgcc/config/rs6000/e500crtresx64gpr.S.
2073 * config/rs6000/e500crtsav32gpr.asm: Move to
2074 ../libgcc/config/rs6000/e500crtsav32gpr.S.
2075 * config/rs6000/e500crtsav64gpr.asm: Move to
2076 ../libgcc/config/rs6000/e500crtsav64gpr.S.
2077 * config/rs6000/e500crtsav64gprctr.asm: Move to
2078 ../libgcc/config/rs6000/e500crtsav64gprctr.S.
2079 * config/rs6000/e500crtsavg32gpr.asm: Move to
2080 ../libgcc/config/rs6000/e500crtsavg32gpr.S.
2081 * config/rs6000/e500crtsavg64gpr.asm: Move to
2082 ../libgcc/config/rs6000/e500crtsavg64gpr.S.
2083 * config/rs6000/e500crtsavg64gprctr.asm: Move to
2084 ../libgcc/config/rs6000/e500crtsavg64gprctr.S.
2085 * config/rs6000/eabi.asm: Move to ../libgcc/config/rs6000/eabi.S.
2086 * config/rs6000/t-aix43 (LIBGCC, INSTALL_LIBGCC, LIB2FUNCS_EXTRA)
2087 (TARGET_LIBGCC2_CFLAGS): Remove.
2088 * config/rs6000/t-aix52: Likewise.
2089 * config/rs6000/t-darwin: Remove.
2090 * config/rs6000/t-darwin64 (LIB2_SIDITI_CONV_FUNCS)
2091 (LIB2FUNCS_EXTRA): Remove.
2092 * config/rs6000/t-fprules (LIBGCC, INSTALL_LIBGCC): Remove.
2093 * config/rs6000/t-linux64 (TARGET_LIBGCC2_CFLAGS): Remove.
2094 * config/rs6000/t-lynx (LIB2FUNCS_EXTRA, tramp.S, LIBGCC)
2095 (INSTALL_LIBGCC): Remove.
2096 * config/rs6000/t-netbsd (LIB2FUNCS_EXTRA)
2097 (LIB2FUNCS_STATIC_EXTRA, tramp.S, crtsavfpr.S, crtresfpr.S)
2098 (crtsavgpr.S, crtresgpr.S, crtresxfpr.S, crtresxgpr.S, LIBGCC)
2099 (INSTALL_LIBGCC, $(T)crtsavfpr$(objext), $(T)crtresfpr$(objext))
2100 ($(T)crtsavgpr$(objext), $(T)crtresgpr$(objext))
2101 ($(T)crtresxfpr$(objext), $(T)crtresxgpr$(objext)): Remove.
2102 * config/rs6000/t-ppccomm (LIB2FUNCS_EXTRA)
2103 (LIB2FUNCS_STATIC_EXTRA, eabi.S, tramp.S): Remove.
2104 * config/rs6000/t-spe (LIBGCC, INSTALL_LIBGCC): Remove.
2105 * config/rs6000/t-vxworks: Remove comment.
2106 * config/rs6000/tramp.asm: Move to ../libgcc/config/rs6000/tramp.S.
2107 * config/rx/t-rx (LIBGCC, INSTALL_LIBGCC): Remove.
2108 * config/sh/linux-atomic.asm: Move to
2109 ../libgcc/config/sh/linux-atomic.S.
2110 * config/sh/t-linux (LIB2FUNCS_EXTRA): Remove.
2111 * config/sh/t-netbsd: Remove.
2112 * config/sh/t-sh (TARGET_LIBGCC2_CFLAGS, LIBGCC, INSTALL_LIBGCC):
2114 * config/sparc/t-elf (LIBGCC, INSTALL_LIBGCC): Remove.
2115 * config/sparc/t-leon: Likewise.
2116 * config/sparc/t-leon3: Likewise.
2117 * config/sparc/t-linux64: Likewise.
2118 * config/sparc/t-netbsd64: Fix typo.
2120 * config/spu/divmodti4.c, config/spu/divv2df3.c,
2121 config/spu/float_disf.c, config/spu/float_unsdidf.c,
2122 config/spu/float_unsdisf.c, config/spu/float_unssidf.c,
2123 config/spu/mfc_multi_tag_release.c,
2124 config/spu/mfc_multi_tag_reserve.c, config/spu/mfc_tag_release.c,
2125 config/spu/mfc_tag_reserve.c, config/spu/mfc_tag_table.c,
2126 config/spu/multi3.c: Move to ../libgcc/config/spu.
2127 * config/spu/t-spu-elf (TARGET_LIBGCC2_CFLAGS, LIB2FUNCS_EXCLUDE)
2128 (LIB2FUNCS_STATIC_EXTRA, LIB2_SIDITI_CONV_FUNCS, LIBGCC)
2129 (INSTALL_LIBGCC): Remove.
2130 * config/stormy16/stormy16-lib2.c: Move to
2131 ../libgcc/config/stormy16/lib2.c.
2132 * config/stormy16/stormy16-lib2-ashlsi3.c: Move to
2133 ../libgcc/config/stormy16/ashlsi3.c.
2134 * config/stormy16/stormy16-lib2-ashrsi3.c: Move to
2135 ../libgcc/config/stormy16/ashrsi3.c.
2136 * config/stormy16/stormy16-lib2-clzhi2.c: Move to
2137 ../libgcc/config/stormy16/clzhi2.c.
2138 * config/stormy16/stormy16-lib2-cmpsi2.c: Move to
2139 ../libgcc/config/stormy16/cmpsi2.c.
2140 * config/stormy16/stormy16-lib2-ctzhi2.c: Move to
2141 ../libgcc/config/stormy16/ctzhi2.c.
2142 * config/stormy16/stormy16-lib2-divsi3.c: Move to
2143 ../libgcc/config/stormy16/divsi3.c.
2144 * config/stormy16/stormy16-lib2-ffshi2.c: Move to
2145 ../libgcc/config/stormy16/ffshi2.c.
2146 * config/stormy16/stormy16-lib2-lshrsi3.c: Move to
2147 ../libgcc/config/stormy16/lshrsi3.c.
2148 * config/stormy16/stormy16-lib2-modsi3.c: Move to
2149 ../libgcc/config/stormy16/modsi3.c.
2150 * config/stormy16/stormy16-lib2-parityhi2.c: Move to
2151 ../libgcc/config/stormy16/parityhi2.c.
2152 * config/stormy16/stormy16-lib2-popcounthi2.c: Move to
2153 ../libgcc/config/stormy16/popcounthi2.c.
2154 * config/stormy16/stormy16-lib2-ucmpsi2.c: Move to
2155 ../libgcc/config/stormy16/ucmpsi2.c.
2156 * config/stormy16/stormy16-lib2-udivmodsi4.c: Move to
2157 ../libgcc/config/stormy16/udivmodsi4.c.
2158 * config/stormy16/stormy16-lib2-udivsi3.c: Move to
2159 ../libgcc/config/stormy16/udivsi3.c.
2160 * config/stormy16/stormy16-lib2-umodsi3.c: Move to
2161 ../libgcc/config/stormy16/umodsi3.c.
2162 * config/stormy16/t-stormy16: Move to ../libgcc/config/t-stormy16.
2163 * config/v850/t-v850 (INSTALL_LIBGCC): Remove.
2164 * config/xtensa/lib2funcs.S: Move to ../libgcc/config/xtensa.
2165 * config/xtensa/t-elf: Remove.
2166 * config/xtensa/t-xtensa (LIB2FUNCS_EXTRA): Remove.
2167 * config.gcc (*-*-freebsd*): Remove t-freebsd, t-freebsd-thread
2169 (*-*-linux*, frv-*-*linux*, *-*-kfreebsd*-gnu, *-*-knetbsd*-gnu,
2170 *-*-gnu*, *-*-kopensolaris*-gnu): Remove t-linux from tmake_file.
2171 (*-*-netbsd*): Remove t-libgcc-pic from tmake_file.
2172 (*-*-openbsd*): Likewise.
2173 Remove t-openbsd-thread for posix threads.
2174 (alpha*-*-linux*): Remove alpha/t-alpha, alpha/t-ieee from tmake_file.
2175 (alpha*-*-freebsd*): Likewise.
2176 (alpha*-*-netbsd*): Likewise.
2177 (alpha*-*-openbsd*): Likewise.
2178 (alpha64-dec-*vms*): Likewise.
2179 (alpha*-dec-*vms*): Likewise.
2180 (arm*-*-netbsdelf*): Remove arm/t-netbsd from tmake_file.
2181 (arm*-*-linux*): Remove t-linux from tmake_file.
2182 Remove arm/t-bpabi from tmake_file for arm*-*-linux-*eabi.
2183 (arm*-*-uclinux*): Remove arm/t-bpabi from tmake_file for
2184 arm*-*-uclinux*eabi.
2185 (arm*-*-eabi*, arm*-*-symbianelf* ): Remove arm/t-bpabi from
2186 tmake_file for arm*-*-eabi*.
2187 (fr30-*-elf): Remove tmake_file.
2188 (hppa*64*-*-linux*): Remove tmake_file.
2189 (hppa*-*-linux*): Likewise.
2190 (hppa[12]*-*-hpux10*): Remove pa/t-pa-hpux10, pa/t-pa-hpux from
2192 (hppa*64*-*-hpux11*): Remove pa/t-pa64, pa/t-pa-hpux from tmake_file.
2193 (hppa[12]*-*-hpux11*): Remove pa/t-pa-hpux11, pa/t-pa-hpux from
2195 (i[34567]86-*-elf*): Remove tmake_file.
2196 (x86_64-*-elf*): Likewise.
2197 (i[34567]86-*-nto-qnx*): Likewise.
2198 (i[34567]86-*-cygwin*): Remove i386/t-cygwin from tmake_file.
2199 (i[34567]86-*-mingw*, x86_64-*-mingw*): Remove i386/t-gthr-win32
2200 from tmake_file if using win32 threads.
2201 (iq2000*-*-elf*): Remove tmake-file.
2202 (microblaze*-linux*): Likewise.
2203 (sh-*-elf*, sh[12346l]*-*-elf*, sh-*-linux*)
2204 (sh[2346lbe]*-*-linux*, sh-*-netbsdelf*, shl*-*-netbsdelf*)
2205 (sh5-*-netbsd*, sh5l*-*-netbsd*, sh64-*-netbsd*)
2206 (sh64l*-*-netbsd*): Remove sh/t-netbsd from tmake_file for
2207 sh5*-*-netbsd*, sh64*-netbsd*, *-*-netbsd.
2208 (xtensa*-*-elf*): Remove tmake_file.
2210 2011-11-02 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
2212 * Makefile.in (LIB1ASMSRC): Don't export.
2213 (libgcc.mvars): Don't emit LIB1ASMFUNCS, LIB1ASMSRC.
2214 * config/arm/arm.c: Update lib1funcs.asm filename.
2215 * config/arm/linux-eabi.h: Likewise.
2216 * config/arm/bpabi-v6m.S, config/arm/bpabi.S,
2217 config/arm/ieee754-df.S, config/arm/ieee754-sf.S: Move to
2218 ../libgcc/config/arm.
2219 * config/arm/lib1funcs.asm: Move to ../libgcc/config/arm/lib1funcs.S.
2220 * config/arm/t-arm (LIB1ASMSRC, LIB1ASMFUNCS): Remove.
2221 * config/arm/t-arm-elf (LIB1ASMFUNCS): Remove.
2222 * config/arm/t-bpabi: Likewise.
2223 * config/arm/t-linux (LIB1ASMSRC, LIB1ASMFUNCS): Remove.
2224 * config/arm/t-linux-eabi (LIB1ASMFUNCS): Remove.
2225 * config/arm/t-strongarm-elf: Likewise.
2226 * config/arm/t-symbian: Likewise.
2227 * config/arm/t-vxworks: Likewise.
2228 * config/arm/t-wince-pe: Likewise.
2229 * config/avr/libgcc.S: Move to ../libgcc/config/avr.
2230 * config/avr/t-avr (LIB1ASMSRC, LIB1ASMFUNCS): Remove.
2231 * config/bfin/lib1funcs.asm: Move to ../libgcc/config/bfin/lib1funcs.S.
2232 * config/bfin/t-bfin: Remove.
2233 * config/bfin/t-bfin-elf (LIB1ASMSRC, LIB1ASMFUNCS): Remove.
2234 * config/bfin/t-bfin-linux: Likewise.
2235 * config/bfin/t-bfin-uclinux: Likewise.
2236 * config/c6x/lib1funcs.asm: Move to ../libgcc/config/c6x/lib1funcs.S.
2237 * config/c6x/t-c6x-elf (LIB1ASMSRC, LIB1ASMFUNCS): Remove.
2238 * config/fr30/lib1funcs.asm: Move to ../libgcc/config/fr30/lib1funcs.S.
2239 * config/fr30/t-fr30 (LIB1ASMSRC, LIB1ASMFUNCS): Remove.
2240 * config/frv/lib1funcs.asm: Move to ../libgcc/config/frv/lib1funcs.S.
2241 * config/frv/t-frv (CROSS_LIBGCC1, LIB1ASMSRC, LIB1ASMFUNCS): Remove.
2242 * config/h8300/fixunssfsi.c: Update lib1funcs.asm filename.
2243 * config/h8300/lib1funcs.asm: Move to
2244 ../libgcc/config/h8300/lib1funcs.S.
2245 * config/h8300/t-h8300 (LIB1ASMSRC, LIB1ASMFUNCS): Remove.
2246 * config/i386/cygwin.asm: Move to ../libgcc/config/i386/cygwin.S.
2247 * config/i386/t-cygming (LIB1ASMSRC, LIB1ASMFUNCS): Remove.
2248 * config/i386/t-interix: Likewise.
2249 * config/ia64/lib1funcs.asm: Move to ../libgcc/config/ia64/lib1funcs.S.
2250 * config/ia64/t-hpux (LIB1ASMFUNCS, LIBGCC1_TEST): Remove.
2251 * config/ia64/t-ia64 (LIB1ASMSRC, LIB1ASMFUNCS): Remove.
2252 * config/iq2000/t-iq2000 (LIBGCC1, CROSS_LIBGCC1): Remove.
2253 * config/m32c/m32c.c: Update m32c-lib1.S filename.
2254 * config/m32c/m32c-lib1.S: Move to ../libgcc/config/m32c/lib1funcs.S.
2255 * config/m32c/t-m32c (LIB1ASMSRC, LIB1ASMFUNCS): Remove.
2256 * config/m32r/t-linux (CROSS_LIBGCC1, LIBGCC1, LIBGCC1_TEST): Remove.
2257 * config/m68k/lb1sf68.asm: Move to ../libgcc/config/m68k/lb1sf68.S.
2258 * config/m68k/t-floatlib (LIB1ASMSRC, LIB1ASMFUNCS): New file.
2259 * config/mcore/lib1.asm: Move to ../libgcc/config/mcore/lib1funcs.S.
2260 * config/mcore/t-mcore (LIB1ASMSRC, LIB1ASMFUNCS): Remove.
2261 * config/mep/mep-lib1.asm: Move to ../libgcc/config/mep/lib1funcs.S.
2262 * config/mep/t-mep (LIB1ASMSRC, LIB1ASMFUNCS): Remove.
2263 * config/mips/mips16.S: Move to ../libgcc/config/mips.
2264 * config/mips/t-libgcc-mips16: Remove.
2265 * config/mips/t-sr71k (LIBGCC1, CROSS_LIBGCC1): Remove.
2266 * config/pa/milli64.S: Move to ../libgcc/config/pa.
2267 * config/pa/t-linux (LIB1ASMFUNCS, LIB1ASMSRC): Remove.
2268 * config/pa/t-linux64: Likewise.
2269 * config/picochip/libgccExtras/fake_libgcc.asm: Move to
2270 ../libgcc/config/picochip/lib1funcs.S.
2271 * config/picochip/t-picochip (LIB1ASMFUNCS, LIB1ASMSRC): Remove.
2272 * config/sh/lib1funcs.asm: Move to ../libgcc/config/sh/lib1funcs.S.
2273 * config/sh/lib1funcs.h: Move to ../libgcc/config/sh.
2274 * config/sh/sh.h: Update lib1funcs.asm filename.
2275 * config/sh/t-linux (LIB1ASMFUNCS_CACHE): Remove.
2276 * config/sh/t-netbsd: Likewise.
2277 * config/sh/t-sh (LIB1ASMSRC, LIB1ASMFUNCS, LIB1ASMFUNCS_CACHE):
2279 * config/sh/t-sh64 (LIB1ASMFUNCS): Remove.
2280 * config/sparc/lb1spc.asm: Move to ../libgcc/config/sparc/lb1spc.S.
2281 * config/sparc/lb1spl.asm: Remove.
2282 * config/sparc/t-elf (LIB1ASMSRC, LIB1ASMFUNCS): Remove.
2283 * config/sparc/t-leon: Likewise.
2284 * config/spu/t-spu-elf (LIBGCC1, CROSS_LIBGCC1): Remove.
2285 * config/v850/lib1funcs.asm: Move to ../libgcc/config/v850/lib1funcs.S.
2286 * config/v850/t-v850 (LIB1ASMSRC, LIB1ASMFUNCS): Remove
2287 * config/vax/lib1funcs.asm: Move to ../libgcc/config/vax/lib1funcs.S.
2288 * config/vax/t-linux: Remove.
2289 * config/xtensa/ieee754-df.S, config/xtensa/ieee754-sf.S: Move to
2290 ../libgcc/config/xtensa.
2291 * config/xtensa/lib1funcs.asm: Move to
2292 ../libgcc/config/xtensa/lib1funcs.S.
2293 * config/xtensa/t-xtensa (LIB1ASMSRC, LIB1ASMFUNCS): Remove.
2294 * config.gcc (bfin*-rtems*): Remove bfin/t-bfin from tmake_file.
2295 (bfin*-*): Likewise.
2296 (mips64*-*-linux*, mipsisa64*-*-linux*): Remove
2297 mips/t-libgcc-mips16 from tmake_file.
2298 (mips*-*-linux*): Likewise.
2299 (mips*-sde-elf*): Likewise.
2300 (mipsisa32-*-elf*, mipsisa32el-*-elf*, mipsisa32r2-*-elf*)
2301 (mipsisa32r2el-*-elf*, mipsisa64-*-elf*, mipsisa64el-*-elf*)
2302 (mipsisa64r2-*-elf*, mipsisa64r2el-*-elf*): Likewise.
2303 (mipsisa64sb1-*-elf*, mipsisa64sb1el-*-elf*): Likewise.
2304 (mips-*-elf*, mipsel-*-elf*): Likewise.
2305 (mips64-*-elf*, mips64el-*-elf*): Likewise.
2306 (mips64orion-*-elf*, mips64orionel-*-elf*): Likewise.
2307 (mips*-*-rtems*): Likewise.
2308 (mipstx39-*-elf*, mipstx39el-*-elf*): Likewise.
2309 (vax-*-linux*): Remove vax/t-linux from tmake_file.
2311 2011-11-02 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
2313 * config.gcc (extra_parts): Remove.
2314 (*-*-freebsd*): Remove extra_parts.
2315 (*-*-linux*, frv-*-*linux*, *-*-kfreebsd*-gnu, *-*-knetbsd*-gnu,
2316 *-*-gnu*, *-*-kopensolaris*-gnu): Likewise.
2317 (*-*-netbsd*): Remove t-libc-ok, t-netbsd from tmake_file.
2318 Remove extra_parts for *-*-netbsd*1.[7-9]*, *-*-netbsd[2-9]*,
2319 *-*-netbsdelf[2-9]*.
2320 (*-*-openbsd*): Remove t-libc-ok from tmake_file.
2321 (alpha*-*-linux*): Remove extra_parts.
2322 (alpha*-*-freebsd*): Likewise.
2323 (bfin*-linux-uclibc*): Likewise.
2324 (fr30-*-elf): Likewise.
2325 (moxie-*-elf): Likewise.
2326 (moxie-*-uclinux*): Likewise.
2327 (h8300-*-rtems*): Remove h8300/t-elf from tmake_file.
2328 (h8300-*-elf*): Likewise.
2329 (hppa*64*-*-hpux11*): Remove extra_parts.
2330 (i[34567]86-*-elf*): Remove i386/t-i386elf, i386/t-crtstuff from
2332 (x86_64-*-elf*): Likewise.
2333 (i[34567]86-*-freebsd*): Remove tmake_file.
2334 (x86_64-*-freebsd*): Likewise.
2335 (x86_64-*-netbsd*): Likewise.
2336 (i[34567]86-*-openbsd2.*, i[34567]86-*openbsd3.[0123]): Remove
2337 t-libc-ok from tmake_file.
2338 (i[34567]86-*-linux*, i[34567]86-*-kfreebsd*-gnu,
2339 i[34567]86-*-knetbsd*-gnu, i[34567]86-*-gnu*,
2340 i[34567]86-*-kopensolaris*-gnu): Remove i386/t-crtstuff from
2343 (x86_64-*-linux*, x86_64-*-kfreebsd*-gnu, x86_64-*-knetbsd*-gnu):
2344 Remove i386/t-crtstuff from tmake_file.
2345 (i[34567]86-*-lynxos*): Likewise.
2347 (ia64*-*-elf*): Remove extra_parts.
2348 (ia64*-*-freebsd*): Likewise.
2349 (ia64*-*-linux*): Likewise.
2350 (ia64-hp-*vms*): Remove ia64/t-vms from tmake_file.
2351 (m32r-*-elf*): Remove extra_parts.
2352 (m32rle-*-elf*): Likewise.
2353 (m32r-*-rtems*): Likewise.
2354 (m68k-*-elf*, fido-*-elf*): Likewise.
2355 (m68k*-*-openbsd*): Remove t-libc-ok from tmake_file.
2356 (m68k-*-rtems*): Remove extra_parts.
2357 (mep-*-*): Likewise.
2358 (microblaze*-linux*): Likewise.
2359 (mips64*-*-linux*, mipsisa64*-*-linux*): Likewise.
2360 (mips*-*-linux*): Likewise.
2361 (powerpc-*-lynxos*): Likewise.
2362 (s390x-ibm-tpf*): Likewise.
2363 (score-*-elf): Likewise.
2365 (sh-*-elf*, sh[12346l]*-*-elf*, sh-*-linux*, sh[2346lbe]*-*-linux*,
2366 sh-*-netbsdelf*, shl*-*-netbsdelf*, sh5-*-netbsd*, sh5l*-*-netbsd*,
2367 sh64-*-netbsd*, sh64l*-*-netbsd*): Remove sh/t-elf from tmake_file.
2368 Remove sh/t-superh from tmake_file for sh*-superh-elf.
2369 Remove sh/t-linux64 from tmake_file for sh64*-*-linux*.
2370 (sh-*-rtems*): Remove sh/t-elf from tmake_file.
2371 (sh-wrs-vxworks): Likewise.
2372 (sparc-*-linux*): Remove extra_parts.
2373 (sparc64-*-linux*): Likewise.
2374 (sparc64-*-freebsd*, ultrasparc-*-freebsd*): Likewise.
2375 (xstormy16-*-elf): Likewise.
2376 (xtensa*-*-linux*): Remove xtensa/t-linux from tmake_file.
2377 (am33_2.0-*-linux*): Remove extra_parts.
2378 * configure.ac (extra_parts): Don't substitute.
2379 * configure: Regenerate.
2380 * crtstuff.c: Move to ../libgcc.
2381 * Makefile.in (CRTSTUFF_CFLAGS): Remove.
2382 (EXTRA_PARTS): Remove.
2383 (CRTSTUFF_T_CFLAGS): Remove.
2384 (MOSTLYCLEANFILES): Remove $(EXTRA_PARTS).
2385 (GCC_EXTRA_PARTS): Remove.
2386 (libgcc.mvars): Remove GCC_EXTRA_PARTS, CRTSTUFF_CFLAGS,
2387 CRTSTUFF_T_CFLAGS, CRTSTUFF_T_CFLAGS_S.
2388 Emit GCC_CFLAGS, INHIBIT_LIBC_CFLAGS.
2389 ($(T)crtbegin.o, $(T)crtend.o, $(T)crtbeginS.o, $(T)crtendS.o)
2390 ($(T)crtbeginT.o): Remove.
2391 * config/alpha/t-vms (EXTRA_PARTS): Remove.
2392 ($(T)vms-dwarf2.o, $(T)vms-dwarf2eh.o): Remove.
2393 * config/alpha/vms-dwarf2.asm: Move to
2394 ../libgcc/config/alpha/vms-dwarf2.S.
2395 * config/alpha/vms-dwarf2eh.asm: Move to
2396 ../libgcc/config/alpha/vms-dwarf2eh.S.
2397 * config/arm/crti.asm: Move to ../libgcc/config/arm/crti.S.
2398 * config/arm/crtn.asm: Move to ../libgcc/config/arm/crtn.S.
2399 * config/arm/t-arm-elf (EXTRA_MULTILIB_PARTS): Remove.
2400 ($(T)crti.o, $(T)crtn.o): Remove.
2401 * config/arm/t-linux: Remove comment.
2402 * config/arm/t-linux-eabi (EXTRA_MULTILIB_PARTS): Remove.
2403 * config/arm/t-strongarm-elf (EXTRA_MULTILIB_PARTS): Remove.
2404 ($(T)crti.o, $(T)crtn.o): Remove.
2405 * config/arm/t-symbian (EXTRA_MULTILIB_PARTS): Remove.
2406 * config/bfin/crti.s: Move to ../libgcc/config/bfin/crti.S.
2407 * config/bfin/crtn.s: Move to ../libgcc/config/bfin/crtn.S.
2408 * config/bfin/crtlibid.s: Move to ../libgcc/config/bfin/crtlibid.S.
2409 * config/bfin/t-bfin (EXTRA_PARTS): Remove.
2410 ($(T)crti.o, $(T)crtn.o): Remove.
2411 * config/bfin/t-bfin-elf (CRTSTUFF_T_CFLAGS): Remove.
2412 ($(T)crti.o, $(T)crtn.o, $(T)crtlibid.o): Remove
2413 (EXTRA_MULTILIB_PARTS): Remove.
2414 * config/bfin/t-bfin-linux (CRTSTUFF_T_CFLAGS,
2415 EXTRA_MULTILIB_PARTS): Remove.
2416 * config/bfin/t-bfin-uclinux (CRTSTUFF_T_CFLAGS): Remove.
2417 ($(T)crtlibid.o): Remove.
2418 (EXTRA_MULTILIB_PARTS): Remove.
2419 * config/c6x/crti.s: Move to ../libgcc/config/c6x/crti.S.
2420 * config/c6x/crtn.s: Move to ../libgcc/config/c6x/crtn.S.
2421 * config/c6x/t-c6x-elf ($(T)crti.o, $(T)crtn.o): Remove.
2422 (EXTRA_MULTILIB_PARTS): Remove.
2423 (CRTSTUFF_T_CFLAGS, CRTSTUFF_T_CFLAGS_S): Remove.
2424 * config/c6x/t-c6x-uclinux (CRTSTUFF_T_CFLAGS,
2425 CRTSTUFF_T_CFLAGS_S): Remove.
2426 * config/cris/t-elfmulti (CRTSTUFF_T_CFLAGS): Remove.
2427 * config/cris/t-linux (CRTSTUFF_T_CFLAGS_S): Remove.
2428 * config/fr30/crti.asm: Move to ../libgcc/config/fr30/crti.S.
2429 * config/fr30/crtn.asm: Move to ../libgcc/config/fr30/crtn.S.
2430 * config/fr30/t-fr30 ($(T)crti.o, $(T)crtn.o): Remove.
2431 * config/frv/frvbegin.c, config/frv/frvend.c: Move to
2432 ../libgcc/config/frv.
2433 * config/frv/t-frv (EXTRA_MULTILIB_PARTS): Remove.
2434 (FRVSTUFF_CFLAGS, $(T)frvbegin$(objext), $(T)frvend$(objext)): Remove.
2435 * config/frv/t-linux (EXTRA_MULTILIB_PARTS): Remove.
2436 (CRTSTUFF_T_CFLAGS): Remove.
2437 * config/h8300/crti.asm: Move to ../libgcc/config/h8300/crti.S.
2438 * config/h8300/crtn.asm: Move to ../libgcc/config/h8300/crtn.S.
2439 * config/h8300/t-elf: Remove.
2440 * config/i386/cygming-crtbegin.c, config/i386/cygming-crtend.c:
2441 Move to ../libgcc/config/i386.
2442 * config/i386/t-crtstuff: Remove.
2443 * config/i386/t-i386elf: Remove.
2444 * config/i386/t-linux64 (EXTRA_MULTILIB_PARTS): Remove.
2445 * config/i386/t-nto (CRTSTUFF_T_CFLAGS, EXTRA_PARTS): Remove.
2446 * config/ia64/crtbegin.asm: Move to ../libgcc/config/ia64/crtbegin.S.
2447 * config/ia64/crtend.asm: Move to ../libgcc/config/ia64/crtend.S.
2448 * config/ia64/crti.asm: Move to ../libgcc/config/ia64/crti.S.
2449 * config/ia64/crtn.asm: Move to ../libgcc/config/ia64/crtn.S.
2450 * config/ia64/t-vms: Remove.
2451 * config/ia64/vms-crtinit.asm: Move to
2452 ../libgcc/config/ia64/vms-crtinit.S.
2453 * config/m32c/t-m32c (EXTRA_MULTILIB_PARTS): Remove.
2454 * config/m32r/initfini.c: Move to ../libgcc/config/m32r.
2455 * config/m32r/t-linux (CRTSTUFF_T_CFLAGS_S): Remove.
2456 * config/m32r/t-m32r (CRTSTUFF_T_CFLAGS): Remove.
2457 ($(T)crtinit.o, $(T)crtfini.o): Remove.
2458 (m32rx, m32r2): Remove.
2459 (EXTRA_MULTILIB_PARTS): Remove.
2460 * config/m68k/crti.s: Move to ../libgcc/config/m68k/crti.S.
2461 * config/m68k/crtn.s: Move to ../libgcc/config/m68k/crtn.S.
2462 * config/m68k/t-crtstuff: Remove.
2463 * config/m68k/t-linux (EXTRA_MULTILIB_PARTS): Remove.
2464 * config/m68k/t-m68kelf: Remove.
2465 * config/m68k/t-uclinux (EXTRA_MULTILIB_PARTS): Remove.
2466 * config/mcore/crti.asm: Move to ../libgcc/config/mcore/crti.S.
2467 * config/mcore/crtn.asm: Move to ../libgcc/config/mcore/crtn.S.
2468 * config/mcore/t-mcore ($(T)crti.o, $(T)crtn.o): Remove.
2469 (EXTRA_PARTS, EXTRA_MULTILIB_PARTS): Remove.
2470 * config/mep/t-mep (CRTSTUFF_CFLAGS): Remove.
2471 (EXTRA_MULTILIB_PARTS): Remove.
2472 * config/microblaze/crti.s: Move to ../libgcc/config/microblaze/crti.S.
2473 * config/microblaze/crtn.s: Move to ../libgcc/config/microblaze/crtn.S.
2474 * config/microblaze/t-microblaze (EXTRA_MULTILIB_PARTS,
2475 EXTRA_PARTS): Remove.
2476 ($(T)crti$(objext), $(T)crtn$(objext)): Remove.
2477 * config/mips/crti.asm: Move to ../libgcc/config/mips/crti.S.
2478 * config/mips/crtn.asm: Move to ../libgcc/config/mips/crtn.S.
2479 * config/mips/t-elf (CRTSTUFF_T_CFLAGS): Remove.
2480 ($(T)crti.o, $(T)crtn.o): Remove.
2481 (EXTRA_MULTILIB_PARTS): Remove.
2482 * config/mips/t-isa3264: Likewise.
2483 * config/mips/t-linux64 (EXTRA_MULTILIB_PARTS): Remove.
2484 * config/mips/t-r3900 (EXTRA_MULTILIB_PARTS): Remove.
2485 (CRTSTUFF_T_CFLAGS): Remove.
2486 * config/mips/t-sde (CRTSTUFF_T_CFLAGS): Remove.
2487 ($(T)crti.o, $(T)crtn.o): Remove.
2488 (EXTRA_MULTILIB_PARTS): Remove.
2489 * config/mips/t-sr71k (EXTRA_MULTILIB_PARTS, CRTSTUFF_T_CFLAGS):
2491 ($(T)crti.o, $(T)crtn.o): Remove.
2492 * config/mips/t-st (EXTRA_MULTILIB_PARTS): Remove.
2493 * config/mips/t-vr (CRTSTUFF_T_CFLAGS): Remove.
2494 (EXTRA_MULTILIB_PARTS): Remove.
2495 ($(T)crti.o, $(T)crtn.o): Remove.
2496 * config/mmix/crti.asm: Move to ../libgcc/config/crti.S.
2497 * config/mmix/crtn.asm: Move to ../libgcc/config/crtn.S.
2498 * config/mmix/t-mmix (CRTSTUFF_T_CFLAGS): Remove.
2499 * config/moxie/crti.asm, config/moxie/crtn.asm: Remove.
2500 * config/pa/stublib.c: Move to libgcc/config/pa.
2501 * config/pa/t-linux (CRTSTUFF_T_CFLAGS_S): Remove.
2502 * config/pa/t-linux64 (CRTSTUFF_T_CFLAGS_S): Remove.
2503 * config/pa/t-pa-hpux11 (LIBGCCSTUB_OBJS, stublib.c): Remove.
2504 (pthread_default_stacksize_np-stub.o, pthread_mutex_lock-stub.o)
2505 (pthread_mutex_unlock-stub.o, pthread_once-stub.o)
2506 ($(T)libgcc_stub.a): Remove.
2507 * config/pa/t-pa64 (LIBGCCSTUB_OBJS, stublib.c): Remove.
2508 (rfi-stub.o, dfi-stub.o, cxaf-stub.o, jvrc-stub.o)
2509 (pthread_default_stacksize_np-stub.o, pthread_mutex_lock-stub.o)
2510 (pthread_mutex_unlock-stub.o, pthread_once-stub.o)
2511 ($(T)libgcc_stub.a): Remove.
2512 * config/rs6000/eabi-cn.asm: Move to
2513 ../../../libgcc/config/rs6000/eabi-cn.S.
2514 * config/rs6000/eabi-ci.asm: Move to
2515 ../../../libgcc/config/rs6000/eabi-ci.S.
2516 * config/rs6000/sol-ci.asm: Move to
2517 ../../../libgcc/config/rs6000/sol-ci.S.
2518 * config/rs6000/sol-cn.asm: Move to
2519 ../../../libgcc/config/rs6000/sol-cn.S.
2520 * config/rs6000/t-lynx (EXTRA_MULTILIB_PARTS): Remove.
2521 (CRTSTUFF_T_CFLAGS, CRTSTUFF_T_CFLAGS_S): Remove.
2522 * config/rs6000/t-netbsd (CRTSTUFF_T_CFLAGS, CRTSTUFF_T_CFLAGS_S):
2524 (EXTRA_MULTILIB_PARTS): Remove.
2525 * config/rs6000/t-ppccomm (EXTRA_MULTILIB_PARTS): Remove.
2526 (ecrti.S, ecrtn.S, ncrti.S, ncrtn.S): Remove.
2527 ($(T)ecrti$(objext), $(T)ecrtn$(objext), $(T)ncrti$(objext),
2528 ($(T)ncrtn$(objext)): Remove.
2529 (CRTSTUFF_T_CFLAGS, CRTSTUFF_T_CFLAGS_S): Remove.
2530 * config/rs6000/t-vxworks (EXTRA_MULTILIB_PARTS): Remove.
2531 * config/rx/t-rx (EXTRA_MULTILIB_PARTS): Remove.
2532 * config/score/crti.asm: Move to ../libgcc/config/score/crti.S.
2533 * config/score/crtn.asm: Move to ../libgcc/config/score/crtn.S.
2534 * config/score/t-score-elf: Remove.
2535 * config/sh/crt1.asm: Move to ../libgcc/config/sh/crt1.S.
2536 * config/sh/crti.asm: Move to ../libgcc/config/sh/crti.S.
2537 * config/sh/crtn.asm: Move to ../libgcc/config/sh/crtn.S.
2538 * config/sh/lib1funcs-4-300.asm: Move to
2539 ../../../libgcc/config/sh/lib1funcs-4-300.S.
2540 * config/sh/lib1funcs-Os-4-200.asm: Move to
2541 ../libgcc/config/sh/lib1funcs-Os-4-200.S.
2542 * config/sh/t-elf: Remove.
2543 * config/sh/t-linux (EXTRA_MULTILIB_PARTS): Remove.
2544 * config/sh/t-linux64: Remove.
2545 * config/sh/t-netbsd (EXTRA_MULTILIB_PARTS): Remove.
2546 * config/sh/t-sh ($(T)crt1.o, $(T)crti.o, $(T)crtn.o): Remove.
2547 (IC_EXTRA_PARTS, OPT_EXTRA_PARTS, EXTRA_MULTILIB_PARTS): Remove.
2548 ($(T)ic_invalidate_array_4-100.o)
2549 ($(T)libic_invalidate_array_4-100.a)
2550 ($(T)ic_invalidate_array_4-200.o)
2551 ($(T)libic_invalidate_array_4-200.a, $(T)ic_invalidate_array_4a.o)
2552 ($(T)libic_invalidate_array_4a.a, $(T)sdivsi3_i4i-Os-4-200.o)
2553 ($(T)udivsi3_i4i-Os-4-200.o, $(T)unwind-dw2-Os-4-200.o)
2554 ($(T)libgcc-Os-4-200.a, $(T)div_table-4-300.o)
2555 ($(T)libgcc-4-300.a): Remove.
2556 * config/sh/t-superh: Remove.
2557 * config/sh/t-vxworks (EXTRA_MULTILIB_PARTS): Remove.
2558 * config/sparc/t-linux64 (CRTSTUFF_T_CFLAGS): Remove.
2559 * config/spu/cache.S: Move to ../libgcc/config/spu.
2560 * config/spu/cachemgr.c: Move to ../libgcc/config/spu.
2561 * config/spu/t-spu-elf (CRTSTUFF_T_CFLAGS): Remove.
2562 (EXTRA_MULTILIB_PARTS): Remove.
2563 ($(T)cachemgr.o, $(T)cachemgr_nonatomic.o, $(T)libgcc_%.a): Remove.
2564 ($(T)cache8k.o, $(T)cache16k.o, $(T)cache32k.o, $(T)cache32k.o)
2565 ($(T)cache64k.o, $(T)cache128k.o): Remove.
2566 * config/t-freebsd (CRTSTUFF_T_CFLAGS_S): Remove.
2567 * config/t-libc-ok: Remove.
2568 * config/t-linux (CRTSTUFF_T_CFLAGS_S): Remove.
2569 * config/t-lynx (CRTSTUFF_T_CFLAGS_S): Remove.
2570 * config/t-netbsd: Remove.
2571 * config/t-svr4 (CRTSTUFF_T_CFLAGS_S): Remove.
2572 * config/t-vxworks (EXTRA_MULTILIB_PARTS): Remove.
2573 * config/vms/t-vms (VMS_EXTRA_PARTS): Remove.
2574 ($(T)vcrt0.o, $(T)pcrt0.o): Remove.
2575 * config/vms/vms-ucrt0.c: Move to ../libgcc/config/vms.
2576 * config/xtensa/crti.asm: Move to ../libgcc/config/xtensa/crti.S.
2577 * config/xtensa/crtn.asm: Move to ../libgcc/config/xtensa/crtn.S.
2578 * config/xtensa/t-elf (CRTSTUFF_T_CFLAGS, CRTSTUFF_T_CFLAGS_S): Remove.
2579 (EXTRA_MULTILIB_PARTS): Remove.
2580 * config/xtensa/t-linux: Remove.
2581 * config/xtensa/t-xtensa ($(T)crti.o, $(T)crtn.o): Remove.
2583 2011-11-02 Uros Bizjak <ubizjak@gmail.com>
2585 * config/i386/i386.c (bdesc_args) [IX86_BUILTIN_CVTTPD2DQ256]: Use
2586 CODE_FOR_fix_truncv4dfv4si2, not CODE_FOR_fix_truncv4sfv4si2.
2588 2011-11-02 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
2590 PR translation/45116
2591 * Makefile.in (slibdir): Remove, don't export.
2592 (SHLIB_NM_FLAGS): Remove.
2593 (libgcc.mvars): Don't emit SHLIB_LINK, SHLIB_INSTALL,
2594 SHLIB_DLLDIR, SHLIB_EXT, SHLIB_MKMAP, SHLIB_MKMAP_OPTS,
2595 SHLIB_MAPFILES, SHLIB_NM_FLAGS.
2596 (DRIVER_DEFINES): Test SHLIB instead of SHLIB_LINK.
2597 (gcc.o): Pass SHLIB instead of SHLIB_LINK.
2598 (gccspec.o): Likewise.
2599 (installdirs): Don't create $(DESTDIR)$(slibdir).
2600 * configure.ac (slibdir): Remove.
2601 * configure: Regenerate.
2602 * libgcc-libsystem.ver: Move to ../libgcc/config.
2603 * mkmap-flat.awk, mkmap-symver.awk: Move to ../libgcc.
2604 * config/libgcc-glibc.ver: Move to ../libgcc/config.
2605 * config/t-libunwind (SHLIB_LC): Remove.
2606 * config/t-linux (SHLIB_MAPFILES): Remove.
2607 * config/t-slibgcc-dummy: Rename to config/t-slibgcc.
2608 * config/t-slibgcc-elf-ver: Remove.
2609 * config/t-slibgcc-libgcc, config/t-slibgcc-nolc-override: Move to
2611 * config/alpha/libgcc-alpha-ldbl.ver, config/alpha/t-linux: Move
2612 to ../libgcc/config/alpha.
2613 * config/alpha/t-vms (shlib_version, SHLIB_EXT, SHLIB_OBJS,
2614 SHLIB_NAME, SHLIB_MULTILIB, SHLIB_INSTALL, SHLIB_SYMVEC,
2615 SHLIB_SYMVECX2, SHLIB_LINK): Remove.
2616 * config/arm/libgcc-bpabi.ver: Move to ../libgcc/config/arm.
2617 * config/arm/t-bpabi (SHLIB_MAPFILES): Remove.
2618 * config/arm/t-netbsd (SHLIB_EXT, SHLIB_NAME, SHLIB_SONAME,
2619 SHLIB_OBJS, SHLIB_LINK, SHLIB_INSTALL): Remove.
2620 * config/arm/t-symbian (SHLIB_LC): Remove.
2621 * config/bfin/libgcc-bfin.ver: Move to
2622 ../libgcc/config/bfin/libgcc-glibc.ver.
2623 * config/bfin/t-bfin-linux (SHLIB_MAPFILES): Remove.
2624 * config/c6x/libgcc-c6xeabi.ver: Move to
2625 ../libgcc/config/c6x/libgcc-eabi.ver.
2626 * config/c6x/t-c6x-elf (SHLIB_MAPFILES): Remove.
2627 * config/cris/libgcc.ver: Move to
2628 ../libgcc/config/cris/libgcc-glibc.ver.
2629 * config/cris/t-linux (SHLIB_MAPFILES): Remove.
2630 * config/frv/libgcc-frv.ver: Move to ../libgcc/config/frv.
2631 * config/frv/t-linux (SHLIB_MAPFILES): Remove.
2632 * config/i386/darwin-libgcc.10.4.ver: Move to
2633 ../libgcc/config/i386/libgcc-darwin.10.4.ver.
2634 * config/i386/darwin-libgcc.10.5.ver: Move to
2635 ../libgcc/config/i386/libgcc-darwin.10.5.ver.
2636 * config/i386/libgcc-glibc.ver: Move to ../libgcc/config/i386.
2637 * config/i386/t-cygming (SHLIB_EXT, SHLIB_IMPLIB, SHLIB_SOVERSION,
2638 SHLIB_SONAME, SHLIB_MAP, SHLIB_OBJS, SHLIB_DIR, SHLIB_SLIBDIR_QUAL)
2639 SHLIB_PTHREAD_CFLAG, SHLIB_PTHREAD_LDFLAG, SHLIB_LINK,
2640 SHLIB_INSTALL, SHLIB_MKMAP, SHLIB_MKMAP_OPTS, SHLIB_MAPFILES): Remove.
2641 * config/i386/t-cygwin (SHLIB_LC, SHLIB_EH_EXTENSION,
2642 SHLIB_IMPLIB, SHLIB_SONAME, SHLIB_MKMAP_OPTS): Remove.
2643 * config/i386/t-dlldir, config/i386/t-dlldir-x: Move to
2644 ../libgcc/config/i386.
2645 * config/i386/t-dw2-eh, config/i386/t-sjlj-eh: Move to
2646 ../libgcc/config/i386.
2647 * config/i386/t-linux: Move to ../libgcc/config/i386.
2648 * config/i386/t-mingw-pthread: Move to ../libgcc/config/i386.
2649 * config/i386/t-mingw-w32 (SHLIB_LC): Remove.
2650 * config/i386/t-mingw-w64: Likewise.
2651 * config/i386/t-mingw32: Remove.
2652 * config/ia64/libgcc-glibc.ver, config/ia64/libgcc-ia64.ver: Move
2653 to ../libgcc/config/ia64.
2654 * config/ia64/t-glibc: Remove.
2655 * config/ia64/t-hpux (SHLIB_EXT, SHLIB_LINK, SHLIB_INSTALL): Remove.
2656 * config/ia64/t-ia64 (SHLIB_MAPFILES): Remove.
2657 * config/ia64/t-vms (shlib_version, SHLIB_EXT, SHLIB_OBJS,
2658 SHLIB_NAME, SHLIB_MULTILIB, SHLIB_INSTALL, SHLIB_LINK): Remove.
2659 * config/ia64/vms_symvec_libgcc_s.opt: Remove.
2660 * config/m32r/libgcc-glibc.ver: Move to ../libgcc/config/m32r.
2661 * config/m32r/t-linux (SHLIB_MAPFILES): Remove.
2662 * config/m68k/t-slibgcc-elf-ver: Move to ../libgcc/config/m68k.
2663 * config/mips/t-libgcc-mips16 (SHLIB_MAPFILES): Remove.
2664 * config/pa/t-hpux-shlib: Move to ../libgcc/config/pa/t-slibgcc-hpux.
2665 * config/pa/t-slibgcc-dwarf-ver, config/pa/t-slibgcc-sjsj-ver:
2666 Move to ../libgcc/config/pa.
2667 * config/rs6000/darwin-libgcc.10.4.ver: Move to
2668 ../libgcc/config/rs6000/libgcc-darwin.10.4.ver.
2669 * config/rs6000/darwin-libgcc.10.5.ver: Move to
2670 ../libgcc/config/rs6000/libgcc-darwin.10.5.ver.
2671 * config/rs6000/t-aix43 (SHLIB_EXT, SHLIB_LINK, SHLIB_INSTALL,
2672 SHLIB_LIBS, SHLIB_MKMAP, SHLIB_NM_FLAGS, AR_FLAGS_FOR_TARGET): Remove.
2673 * config/rs6000/t-aix52: Likewise.
2674 * config/sh/libgcc-excl.ver, config/sh/libgcc-glibc.ver: Move to
2675 ../libgcc/config/sh.
2676 * config/sparc/libgcc-sparc-glibc.ver: Move to
2677 ../libgcc/config/sparc/libgcc-glibc.ver.
2678 * config/sparc/t-linux: Move to ../libgcc/config/sparc.
2679 * config/xtensa/t-linux (SHLIB_MAPFILES): Remove.
2680 * config/xtensa/libgcc-xtensa.ver: Move to
2681 ../libgcc/config/xtensa/libgcc-glibc.ver.
2682 * config.gcc (*-*-freebsd*): Replace t-slibgcc-elf-ver with
2683 t-slibgcc in tmake_file.
2684 Remove t-slibgcc-nolc-override for *-*-freebsd[34],
2685 *-*-freebsd[34].* with pthreads.
2686 (*-*-linux*, frv-*-*linux*, *-*-kfreebsd*-gnu,
2687 *-*-knetbsd*-gnu, *-*-gnu*, *-*-kopensolaris*-gnu): Replace
2688 t-slibgcc-elf-ver with t-slibgcc in tmake_file.
2689 (*-*-netbsd*): Likewise.
2690 (*-*-solaris2*): Replace t-slibgcc-dummy with t-slibgcc in tmake_file.
2691 (*-*-*vms*): Add t-slibgcc to tmake_file.
2692 (alpha*-*-linux*): Remove alpha/t-linux from tmake_file.
2693 (alpha*-dec-osf5.1*): Replace t-slibgcc-dummy with t-slibgcc in
2695 (arm*-*-linux*): Remove t-slibgcc-libgcc from tmake_file for
2697 (bfin*-linux-uclibc*): Replace t-slibgcc-dummy with t-slibgcc in
2699 (crisv32-*-linux*, cris-*-linux*): Likewise.
2700 (hppa*-*-linux*): Remove t-slibgcc-libgcc, pa/t-slibgcc-sjlj-ver,
2701 pa/t-slibgcc-dwarf-ver from tmake_file.
2702 (hppa[12]*-*-hpux10*): Replace pa/t-hpux-shlib with t-slibgcc in
2704 Remove pa/t-slibgcc-sjlj-ver, pa/t-slibgcc-dwarf-ver from tmake_file.
2705 (hppa*64*-*-hpux11*): Likewise.
2706 (hppa[12]*-*-hpux11*): Likewise.
2707 (i[34567]86-*-darwin*): Replace t-slibgcc-dummy in t-slibgcc in
2709 (x86_64-*-darwin*): Likewise.
2710 (i[34567]86-*-cygwin*): Remove tmake_eh_file, tmake_dlldir_file.
2711 Add t-slibgcc to tmake_file.
2712 (i[34567]86-*-mingw*, x86_64-*-mingw*): Likewise.
2713 Remove i386/t-mingw32 from tmake_file unless x86_64-w64-*,
2715 Remove i386/t-mingw-pthread from tmake_file.
2716 (ia64*-*-linux*): Remove ia64/t-glibc from tmake_file.
2717 (ia64*-*-hpux*): Add t-slibgcc to tmake_file.
2718 (ia64-hp-*vms*): Likewise.
2719 (m32r-*-linux*): Replace t-slibgcc-elf-ver with t-slibgcc in
2721 (m32rle-*-linux*): Likewise.
2722 (m68k-*-linux*): Remove m68k/t-slibgcc-elf-ver from tmake_file.
2723 (microblaze*-linux*): Remove t-slibgcc-elf-ver,
2724 t-slibgcc-nolc-override from tmake_file.
2725 (mips-sgi-irix6.5*): Replace t-slibgcc-dummy with t-slibgcc in
2727 (powerpc-*-darwin*): Likewise.
2728 (powerpc64-*-darwin*): Likewise.
2729 (powerpc-*-freebsd*): Remove t-slibgcc-libgcc from tmake_file.
2730 (powerpc-*-linux*, powerpc64-*-linux*): Likewise.
2731 (rs6000-ibm-aix4.[3456789]*, powerpc-ibm-aix4.[3456789]*): Add
2732 t-slibgcc to tmake_file.
2733 (rs6000-ibm-aix5.1.*, powerpc-ibm-aix5.1.*): Likewise.
2734 (rs6000-ibm-aix5.2.*, powerpc-ibm-aix5.2.*): Likewise.
2735 (rs6000-ibm-aix5.3.*, powerpc-ibm-aix5.3.*): Likewise.
2736 (rs6000-ibm-aix[6789].*, powerpc-ibm-aix[6789].*): Likewise.
2737 (sparc-*-linux*): Remove sparc/t-linux from tmake_file.
2738 (sparc64-*-linux*): Likewise.
2739 (tic6x-*-uclinux): Replace t-slibgcc-elf-ver with t-slibgcc in
2741 (i[34567]86-*-linux*, x86_64-*-linux*, i[34567]86-*-kfreebsd*-gnu,
2742 x86_64-*-kfreebsd*-gnu, i[34567]86-*-gnu*): Remove i386/t-linux
2745 2011-11-02 Richard Guenther <rguenther@suse.de>
2747 PR tree-optimization/50902
2748 * tree-vect-stmts.c (vectorizable_load): Properly convert
2749 an invariant initializer element.
2751 2010-11-02 Richard Guenther <rguenther@suse.de>
2753 PR tree-optimization/50890
2754 * gimple.h (gimple_fold_call): Remove.
2755 * gimple-fold.c (fold_stmt_1): Move all call related code to ...
2756 (gimple_fold_call): ... here. Make static. Update the
2757 cannot-inline flag on direct calls.
2758 * ipa-inline.c (early_inliner): Copy the cannot-inline flag
2759 from the statements to the edges.
2761 2011-11-01 Ian Lance Taylor <iant@google.com>
2763 * godump.c (struct macro_hash_value): Define.
2764 (macro_hash_hashval): New static function.
2765 (macro_hash_eq, macro_hash_del): New static functions.
2766 (go_define): Use macro_hash_value to store values in macro_hash.
2767 Replace an old value on a redefinition. Don't print anything to
2769 (go_undef): Delete the entry from the hash table.
2770 (go_output_typedef): For an enum, use macro_hash_value, and don't
2771 print anything to go_dump_file.
2772 (go_print_macro): New static function.
2773 (go_finish): Traverse macro_hash with go_print_macro.
2774 (dump_go_spec_init): Update macro_hash creation for macro_hash_value.
2776 2011-11-02 Alan Modra <amodra@gmail.com>
2778 * config/rs6000/rs6000.c (rs6000_code_end): Declare ATTRIBUTE_UNUSED.
2780 2011-11-01 Paolo Carlini <paolo.carlini@oracle.com>
2783 * doc/invoke.texi: Document -Wzero-as-null-pointer-constant.
2785 2011-11-01 Andrew Stubbs <ams@codesourcery.com>
2787 * config/arm/bpabi.h (BE8_LINK_SPEC): Recognize generic-armv7 tuning.
2789 2011-11-01 Uros Bizjak <ubizjak@gmail.com>
2791 * config/i386/i386.md (splitters for int-float conversion): Use
2792 SUBREG_REG on SUBREGs in splitter constraints.
2794 2011-11-01 Jakub Jelinek <jakub@redhat.com>
2796 * config/i386/i386-protos.h (ix86_expand_adjust_ufix_to_sfix_si): New
2798 * config/i386/i386.c (ix86_expand_adjust_ufix_to_sfix_si): New
2800 * config/i386/sse.md (fixuns_trunc<mode><sseintvecmodelower>2): Use it.
2801 (ssepackfltmode): New mode attr.
2802 (vec_pack_ufix_trunc_<mode>): New expander.
2804 2011-11-01 Uros Bizjak <ubizjak@gmail.com>
2807 * config/i386/i386.md (floatsi<mode>2_vector_sse_with_temp splitter):
2808 Compare <ssevecmode>mode to V4SFmode, not V4SImode.
2810 2011-11-01 Peter Bergner <bergner@vnet.ibm.com>
2812 * config.gcc (powerpc*-*-linux*): Add powerpc*-*-linux*ppc476* variant.
2813 * config/rs6000/476.h: New file.
2814 * config/rs6000/476.opt: Likewise.
2815 * config/rs6000/rs6000.h (TARGET_LINK_STACK): New define.
2816 (SET_TARGET_LINK_STACK): Likewise.
2817 (TARGET_ASM_CODE_END): Define.
2818 * config/rs6000/rs6000.c (rs6000_option_override_internal): Enable
2819 TARGET_LINK_STACK for -mtune=476 and -mtune=476fp.
2820 (rs6000_legitimize_tls_address): Emit the link stack preserving GOT
2821 code if TARGET_LINK_STACK.
2822 (rs6000_emit_load_toc_table): Likewise.
2823 (output_function_profiler): Likewise
2824 (macho_branch_islands): Likewise
2825 (machopic_output_stub): Likewise
2826 (get_ppc476_thunk_name): New function.
2827 (rs6000_code_end): Likewise.
2828 * config/rs6000/rs6000.md (load_toc_v4_PIC_1, load_toc_v4_PIC_1b):
2829 Convert to a define_expand.
2830 (load_toc_v4_PIC_1_normal): New define_insn.
2831 (load_toc_v4_PIC_1_476): Likewise.
2832 (load_toc_v4_PIC_1b_normal): Likewise.
2833 (load_toc_v4_PIC_1b_476): Likewise.
2835 2011-11-01 Georg-Johann Lay <avr@gjlay.de>
2838 * config/avr/avr.opt (-mbranch-cost=): New option.
2839 * config/avr/avr.h (BRANCH_COST): Define to avr_branch_cost.
2840 * config/avr/avr.c (avr_rtx_costs_1): Adjust [U]DIV/[U]MOD costs.
2841 * config/avr/avr.md (*addqi3.lt0, *addhi3.lt0, *addsi3.lt0): New insns.
2842 (*addhi3_zero_extend1): Remove % in constraint of operand 1.
2843 (*addhi3.sign_extend1, *subhi3.sign_extend2): New insns.
2845 2011-11-01 Tom de Vries <tom@codesourcery.com>
2847 PR tree-optimization/50908
2848 * tree-ssa-tail-merge.c (update_vuses): Now that edges are removed
2849 before update_vuses, test for 1 predecessor rather than two.
2850 (delete_block_update_dominator_info): New function, part of it factored
2852 (replace_block_by): Use delete_block_update_dominator_info. Call
2853 update_vuses after deleting bb1 and updating dominator info, instead of
2856 2011-11-01 David S. Miller <davem@davemloft.net>
2858 * config/sparc/sparc.c (vector_init_faligndata): New function.
2859 (sparc_expand_vector_init): Use it for V4HImode on VIS1.
2861 * config/sparc/sparc.c (sparc_expand_vcond): New function.
2862 * config/sparc/sparc-protos.h (sparc_expand_vcond): Declare it.
2863 * config/sparc/sparc.md (vcond<mode><mode>): New VIS3 expander.
2864 (vconduv8qiv8qi): Likewise.
2866 2011-11-01 Alexandre Oliva <aoliva@redhat.com>
2869 * cselib.c (cfa_base_preserved_regno): Initialize.
2870 (cselib_expand_value_rtx_1): Don't expand it.
2871 * var-tracking.c (vt_expand_var_loc_chain): Initialize depth.
2872 Check it's only zero if result is NULL.
2874 2011-11-01 Jakub Jelinek <jakub@redhat.com>
2876 * config/i386/sse.md (fixuns_trunc<mode><sseintvecmodelower>2): New
2879 * config/i386/sse.md (sseintvecmode): Remove duplicate modes.
2880 (sseintvecmodelower): New mode iterator.
2881 (floatv8siv8sf2, floatunsv4siv4sf2): Macroize into...
2882 (float<sseintvecmodelower><mode>2): ... this using VF1 iterator.
2883 (floatunsv4siv4sf2): Macroize into...
2884 (floatuns<sseintvecmodelower><mode>2): ... this using VF1 iterator.
2886 2011-10-31 David S. Miller <davem@davemloft.net>
2888 * config/sparc/sparc.md (cmask patterns): Allow zero operand.
2890 * dwarf2out.c (cached_next_real_insn): New.
2891 (dwarf2out_end_epilogue): Set it to NULL_RTX.
2892 (dwarf2out_var_location): Remove cached_next_real_insn local static.
2894 2011-10-31 Richard Henderson <rth@redhat.com>
2896 * config/i386/sse.md (floatv8siv8sf2): Rename from avx_cvtdq2ps256.
2897 (floatv4siv4sf2): Rename from sse2_cvtdq2ps.
2898 (floatunsv4siv4sf2): Rename from sse2_cvtudq2ps.
2899 (fix_truncv8sfv8si2): Rename from avx_cvttps2dq256.
2900 (fix_truncv4sfv4si2): Rename from sse2_cvttps2dq.
2901 (floatv4siv4df2): Rename from avx_cvtdq2pd256.
2902 (fix_truncv4dfv4si2): Rename from avx_cvttpd2dq256.
2903 (vec_unpacku_float_hi_v8si): Update for insn pattern name changes.
2904 * config/i386/i386.md (splitters for int-float conversion): Likewise.
2905 * config/i386/i386.c (ix86_split_convert_uns_si_sse): Likewise.
2906 (bdesc_args): Likewise.
2907 (enum ix86_builtins) [IX86_BUILTIN_CVTUDQ2PS]: Remove.
2908 (ix86_vectorize_builtin_conversion): Remove.
2909 (TARGET_VECTORIZE_BUILTIN_CONVERSION): Remove.
2911 2011-10-31 Martin Jambor <mjambor@suse.cz>
2913 * ipa-prop.c (mark_modified): Moved up in the file.
2914 (is_parm_modified_before_call): Renamed to
2915 is_parm_modified_before_stmt, moved up in the file.
2916 (load_from_unmodified_param): New function.
2917 (compute_complex_assign_jump_func): Also attempt to create pass
2918 through jump functions for values loaded from (addressable)
2921 2011-10-31 Jakub Jelinek <jakub@redhat.com>
2923 * tree-vect-stmts.c (vectorizable_shift): If op1 is vect_external_def
2924 in a loop and has different type from op0, cast it to op0's type
2925 before the loop first. For slp give up. Don't crash if op1_vectype
2928 2011-10-31 Paul Brook <paul@codesourcery.com>
2930 * cgraphunit.c: Don't mark clones as static constructors.
2932 2011-10-31 David Edelsohn <dje.gcc@gmail.com>
2934 * gcc-ar: Do not include stdio.h.
2936 2011-10-31 Diego Novillo <dnovillo@google.com>
2938 * tree-streamer-out.c (pack_ts_base_value_fields): Emit
2940 * tree-streamer-in.c (unpack_ts_base_value_fields): Read
2943 2011-10-30 David S. Miller <davem@davemloft.net>
2945 * config/sparc/sparc.c (vector_init_bshuffle): New function.
2946 (vector_init_fpmerge): New function.
2947 (sparc_expand_vector_init): Use them to improve non-const cases.
2949 * dwarf2out.c (dwarf2out_var_location): When processing several
2950 consecutive location notes, cache the result of next_real_insn().
2952 2011-10-30 Uros Bizjak <ubizjak@gmail.com>
2954 * config/i386/i386.md (avx2_vec_dup<mode>): Macroize insn from
2955 avx2_vec_dup{v8sf,v4sf} using VF1 mode iterator.
2956 (vec_dupv4sf): Remove expander.
2957 (vec_dupv4sf): Merge from *vec_dupv4sf and *vec_dupv4sf_avx.
2958 (vec_dupv2df): Remove expander.
2959 (vec_dupv2df): Merge from *vec_dupv2df and *vec_dupv2df_sse3.
2960 (*vec_concatv2df): Merge *vec_concatv2df_sse3.
2961 (*vec_dupv4si): Merge *vec_dupv4si_avx.
2962 (*vec_dupv2di): Merge *vec_dupv2di_sse3.
2964 2011-10-30 Dmitry Plotnikov <dplotnikov@ispras.ru>
2966 * tree-cfg.c (verify_gimple_assign_unary): Allow vector conversions.
2967 * optabs.c (supportable_convert_operation): New function.
2968 * optabs.h (supportable_convert_operation): New prototype.
2969 * tree-vect-stmts.c (vectorizable_conversion): Change condition and
2970 behavior for NONE modifier case.
2971 * tree.h (VECTOR_INTEGER_TYPE_P): New macro.
2973 2011-10-30 Jakub Jelinek <jakub@redhat.com>
2975 * tree-vectorizer.h (NUM_PATTERNS): Bump to 9.
2976 * tree-vect-patterns.c (vect_recog_vector_vector_shift_pattern): New
2978 (vect_vect_recog_func_ptrs): Add it.
2980 2011-10-30 David S. Miller <davem@davemloft.net>
2982 * reorg.c (label_before_next_insn): New function.
2983 (relax_delay_slots): Use it instead of prev_label.
2984 * rtl.h (prev_label): Delete declaration.
2985 * emit-rtl.c (prev_label): Remove.
2987 2011-10-30 Revital Eres <revital.eres@linaro.org>
2989 * modulo-sched.c (generate_prolog_epilog): Mark prolog and epilog
2990 as BB_DISABLE_SCHEDULE.
2991 (mark_loop_unsched): New function.
2992 (sms_schedule): Call it.
2994 2011-10-29 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2997 * config/pa/protos.h (attr_length_save_restore_dltp): Delete.
2998 (cmpib_comparison_operator): Likewise.
2999 (following_cal, output_and, output_ior, output_move_double,
3000 output_fp_move_double, output_block_move, output_block_clear,
3001 output_cbranch, output_lbranch, output_bb, output_bvb, output_dbra,
3002 output_movb, output_parallel_movb, output_parallel_addb, output_call,
3003 output_indirect_call, output_millicode_call, output_mul_insn,
3004 output_div_insn, output_mod_insn, singlemove_string,
3005 output_arg_descriptor, output_global_address, print_operand,
3006 legitimize_pic_address, hppa_encode_label, symbolic_expression_p,
3007 fmpyaddoperands, fmpysuboperands, emit_bcond_fp, emit_move_sequence,
3008 emit_hpdiv_const, is_function_label_plus_const, jump_in_call_delay,
3009 hppa_fpstore_bypass_p, attr_length_millicode_call, attr_length_call,
3010 attr_length_indirect_call, return_addr_rtx, function_arg_padding,
3011 insn_refs_are_delayed, get_deferred_plabel, ldil_cint_p, zdepi_cint_p,
3012 output_ascii, compute_frame_size, and_mask_p, cint_ok_for_move,
3013 hppa_expand_prologue, hppa_expand_epilogue, ior_mask_p,
3014 compute_zdepdi_operands, output_64bit_and, output_64bit_ior,
3015 reloc_needed, magic_milli, shadd_constant_p): Consistently prefix
3016 exported functions and variables with "pa_".
3017 * config/pa/predicates.md: Likewise.
3018 * config/pa/pa64-hpux.h: likewise.
3019 * config/pa/som.h: Likewise.
3020 * config/pa/elf.h: Likewise.
3021 * config/pa/pa64-linux.h: Likewise.
3022 * config/pa/pa.md: Likewise.
3023 * config/pa/pa.c: Likewise.
3024 * config/pa/pa-linux.h: Likewise.
3025 * config/pa/pa.h: Likewise.
3026 * config/pa/constraints.md: Likewise.
3028 2011-10-29 Uros Bizjak <ubizjak@gmail.com>
3030 * config/i386/i386.md (xop_sha<mode>3): Rename from xop_ashl<mode>3.
3032 (xop_shl<mode>3): Rename from xop_lshl<mode>3. Update all uses.
3033 * config/i386/i386.c: Update all uses.
3035 2011-10-29 Uros Bizjak <ubizjak@gmail.com>
3037 * config/i386/i386.md (lshlv16qi3): Remove expander.
3038 (lshrv16qi3): New expander.
3039 (<shift_insn>v16qi3): Macroize expander from ashrv16qi3 and lshrv16qi3
3040 using any_shiftrt code iterator. Cleanup.
3041 (ashlv16qi3): Cleanup.
3044 2011-10-29 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
3047 * config/pa/pa.c (emit_move_sequence): Legitimize TLS symbol
3049 (pa_legitimate_constant_p): Return false for TLS_MODEL_GLOBAL_DYNAMIC
3050 and TLS_MODEL_LOCAL_DYNAMIC symbol references.
3052 2011-10-29 Georg-Johann Lay <avr@gjlay.de>
3055 * config/avr/avr.opt (-maccumulate-args): New option.
3056 * config/avr/avr.h (STARTING_FRAME_OFFSET): Redefine to
3057 avr_starting_frame_offset.
3058 (ACCUMULATE_OUTGOING_ARGS): Define to avr_accumulate_outgoing_args.
3059 * config/avr/avr.md (UNSPECV_WRITE_SP_IRQ_ON): Remove.
3060 (UNSPECV_WRITE_SP_IRQ_OFF): Remove.
3061 (UNSPECV_WRITE_SP): New constant.
3062 (*addhi3_sp_R): Rewrite to...
3063 (*addhi3_sp): ...this new insn.
3064 (movhi_sp_r_irq_off, movhi_sp_r_irq_on): Combine to...
3065 (movhi_sp_r): ...this new insn.
3066 * config/avr/avr-protos.h (avr_accumulate_outgoing_args): New.
3067 (avr_starting_frame_offset): New.
3068 * config/avr/avr.c (avr_accumulate_outgoing_args): New function.
3069 (avr_starting_frame_offset): New function.
3070 (avr_outgoing_args_size): New static function.
3071 (avr_initial_elimination_offset): Use it.
3072 (avr_simple_epilogue): Use it.
3073 (avr_asm_function_end_prologue): Use it.
3074 (expand_epilogue): Use it.
3075 (expand_prologue): Use it. Break out code to...
3076 (avr_prologue_setup_frame): ...this new static function.
3077 (avr_can_eliminate): Allow eliminating to frame pointer if there
3079 (avr_frame_pointer_required_p): Use frame pointer if target has a
3081 * config/avr/constraints.md (R): Remove.
3082 (Csp): New constraint.
3083 * config/avr/predicates.md (avr_sp_immediate_operand): Use it.
3085 2011-10-29 Andi Kleen <ak@linux.intel.com>
3087 * gcc-ar.c (target_machine): Add.
3089 2011-10-29 Anatoly Sokolov <aesok@post.ru>
3091 * config/cris/cris.c (reg_ok_for_base_p, reg_ok_for_index_p,
3092 cris_constant_index_p, cris_base_p, cris_index_p,
3093 cris_base_or_autoincr_p, cris_bdap_index_p, cris_biap_index_p,
3094 cris_legitimate_address_p): New functions.
3095 (TARGET_LEGITIMATE_ADDRESS_P): Define.
3096 (cris_pic_symbol_type, cris_valid_pic_const): Change arguments type
3097 from rtx to const_rtx.
3098 (cris_print_operand_address, cris_address_cost,
3099 cris_side_effect_mode_ok): Use
3100 cris_constant_index_p, cris_base_p, cris_base_or_autoincr_p,
3101 cris_biap_index_p and cris_bdap_index_p.
3102 * config/cris/cris.h (CONSTANT_INDEX_P, BASE_P, BASE_OR_AUTOINCR_P,
3103 BDAP_INDEX_P, BIAP_INDEX_P, GO_IF_LEGITIMATE_ADDRESS,
3104 REG_OK_FOR_BASE_P, REG_OK_FOR_INDEX_P): Remove.
3105 (EXTRA_CONSTRAINT_Q, EXTRA_CONSTRAINT_R, EXTRA_CONSTRAINT_T): Use
3106 cris_constant_index_p, cris_base_p, cris_base_or_autoincr_p,
3107 cris_biap_index_p and cris_bdap_index_p.
3108 * config/cris/cris.md (moversideqi movemsideqi peephole2): Use
3110 * config/cris/cris-protos.h (cris_constant_index_p, cris_base_p,
3111 cris_base_or_autoincr_p, cris_bdap_index_p, cris_biap_index_p): New
3113 (cris_pic_symbol_type, cris_valid_pic_const): Update prototype.
3115 2011-10-21 Andi Kleen <ak@linux.intel.com>
3117 * ggc-page (PAGE_ALIGN): Add.
3118 (alloc_page, ggc_pch_total_size, ggc_pch_this_base, ggc_pch_read):
3119 Replace ROUND_UP with PAGE_ALIGN.
3121 2011-10-20 Andi Kleen <ak@linux.intel.com>
3123 * ggc-page (alloc_anon): Add check argument.
3124 (alloc_page): Add fallback to 1 page allocation.
3125 Adjust alloc_anon calls to new argument.
3127 2011-10-18 Andi Kleen <ak@linux.intel.com>
3129 * ggc-page (release_pages): First free large continuous
3130 chunks in the madvise path.
3132 2011-10-18 Andi Kleen <ak@linux.intel.com>
3134 * ggc-page.c (alloc_pages): Always round up entry_size.
3136 2011-10-19 Andi Kleen <ak@linux.intel.com>
3138 * Makefile.in (MOSTLYCLEANFILES): Add gcc-ar/nm/ranlib.
3139 (native): Add gcc-ar, gcc-nm, gcc-ranlib.
3140 (AR_LIBS, gcc-ar, gcc-ar.o, gcc-ranlib, gcc-ranlib.o,
3141 gcc-nm, gcc-nm.o, gcc-ranlib.c, gcc-nm.c): Add.
3142 (install): Depend on install-gcc-ar.
3143 (install-gcc-ar): Add.
3144 (uninstall): Uninstall gcc-ar, gcc-nm, gcc-ranlib.
3145 * gcc-ar.c: Add new file.
3147 2011-10-28 Pat Haugen <pthaugen@us.ibm.com>
3149 * config/rs6000/rs6000.md (define_attr "type"): Add vecdouble.
3150 * config/rs6000/vsx.md (VStype_simple, VStype_mul): Use vecdouble
3152 (VStype_div): Use vector types for V2DF/V4SF.
3153 (VStype_sqrt): Use *sqrt types.
3154 (VS_spdp_type): Change type to vecdouble.
3155 (*vsx_fmav2df4, *vsx_nfmsv2df4, vsx_xvcvdpsxws, vsx_xvcvdpuxws,
3156 vsx_xvcvuxdsp, vsx_xvcvsxwdp, vsx_xvcvuxwdp, vsx_xvcvspsxds,
3157 vsx_xvcvspuxds): Likewise.
3158 (*vsx_fms<mode>4): Set type via <VStype_mul>.
3159 (*vsx_eq_<mode>_p, *vsx_gt_<mode>_p, *vsx_ge_<mode>_p): Set type via
3161 * config/rs6000/power7.md (power7-vecstore): Correct VSU pipe.
3162 (power7-fpcompare, power7-sdiv, power7-ddiv, power7-sqrt,
3163 power7-dsqrt): Correct insn latency.
3164 (power7-vecsimple): Add veccmp type and correct dispatch/VSU values.
3165 (power7-veccmp): Delete.
3166 (power7-vecfloat): Correct latency/dispatch/VSU values.
3167 (define_bypass "power7-vecfloat"): Correct latency and types.
3168 (power7-veccomplex, power7-vecperm): Correct dispatch/VSU values.
3169 (power7-vecdouble, power7-vecfdiv, power7-vecdiv): New.
3171 2011-10-28 Uros Bizjak <ubizjak@gmail.com>
3173 * config/i386/i386.md (shift_insn): Rename code attribute from
3174 shiftrt_insn. Also handle ashift RTX.
3175 (shift): Rename code attribute from shiftrt. Also handle ashift RTX.
3176 (vshift): New code attribute.
3177 (<shift_insn>*): Rename from <shiftrt_insn>*. Update asm templates.
3178 (any_lshift): Move and rename code iterator from ...
3179 * config/i386/sse.md (lshift): ... here.
3180 (lshift_insn): Remove code attribute.
3181 (lshift): Remove code attribute.
3182 (vlshr<mode>3): Use lshiftrt RTX.
3183 (vashr<mode>3, ashrv16qi3, ashrv2di3): Use ashiftrt RTX.
3184 (vashl<mode>3, ashlv16qi3): Use ashift RTX.
3185 (avx2_<lshift>v<mode>): Rename from avx2_<shift_insn>v<mode>. Use
3186 any_lshift code iterator. Update asm template.
3187 (<shift_insn><mode>3): Macroize insn from lshr<mode>3 and ashl<mode>3
3188 usign any_lshift code iterator.
3189 * config/i386/mmx.md (mmx_<shift_insn><mode>3): Macroize insn from
3190 mmx_lshr<mode>3 and mmx_ashl<mode>3 usign any_lshift code iterator.
3191 * config/i386/i386.c (bdesc_args) <__builtin_ia32_psll>: Update.
3193 2011-10-28 Georg-Johann Lay <avr@gjlay.de>
3196 * config/avr/avr.md (parityhi2): Expand allowing pseudos.
3197 (*parityhi2): New pre-reload insn-and-split to map 16-bit parity
3199 (*parityqihi2): Same for 8-bit parity.
3201 2011-10-28 Julian Brown <julian@codesourcery.com>
3203 PR rtl-optimization/47918
3204 * reload1.c (set_initial_label_offsets): Use initial offsets
3205 for labels on the nonlocal_goto_handler_labels chain.
3207 2011-10-28 Iain Sandoe <iains@gcc.gnu.org>
3209 * config/rs6000/t-darwin (LIB2FUNCS_STATIC_EXTRA):
3210 Move darwin-fpsave.asm from here to ... LIB2FUNCS_EXTRA.
3211 (LIB2FUNCS_EXTRA): Add darwin-gpsave.asm.
3212 (TARGET_LIBGCC2_CFLAGS): Ensure that fPIC and -pipe are inherited from
3214 * config/rs6000/darwin.h (FP_SAVE_INLINE): Adjust to enable.
3215 (GP_SAVE_INLINE): Likewise.
3216 (SAVE_FP_PREFIX, SAVE_FP_SUFFIX, RESTORE_FP_PREFIX,
3217 RESTORE_FP_SUFFIX): Set to empty strings.
3218 * config/rs6000/rs6000.c (rs6000_savres_strategy): Implement for Darwin.
3219 (debug_stack_info): Print savres_strategy.
3220 (rs6000_savres_routine_name): Implement for Darwin.
3221 (rs6000_make_savres_rtx): Adjust used register for Darwin.
3222 (rs6000_emit_prologue): Implement out-of-line saves for Darwin.
3223 (rs6000_output_function_prologue): Don't emit .extern for Mach-O.
3224 (rs6000_emit_epilogue): Implement out-of-line saves for Darwin.
3225 * config/rs6000/darwin-gpsave.asm: New file.
3227 2011-10-28 Jakub Jelinek <jakub@redhat.com>
3229 * config/i386/sse.md (VI4SD_AVX2): Removed.
3230 (VI48_AVX2, VI128_128, VI48_128, VI48_256): New mode iterators.
3231 (vashl<mode>3): Use VI12_128 iterator instead of VI124_128.
3232 Add another expander using VI48_128 iterator for
3233 TARGET_AVX2 || TARGET_XOP and another using VI48_256 iterator
3235 (vlshr<mode>3): Likewise. Change register_operand predicate to
3236 nonimmediate_operand on last operand in the VI12_128 expander.
3237 (vashr<mode>3): Use VI128_128 iterator instead of VI124_128.
3238 (vashrv4si3, vashrv8si3): New expanders.
3239 (avx2_ashrvv8si, avx2_ashrvv4si, avx2_<lshift>vv8si,
3240 avx2_<lshift>vv2di): Removed.
3241 (avx2_ashrv<mode>): New insn with VI4_AVX2 iterator.
3242 (avx2_<lshift>v<mode>): Macroize using VI48_AVX2
3243 iterator. Simplify pattern.
3245 2010-10-28 Richard Guenther <rguenther@suse.de>
3248 * lto-wrapper.c (get_options_from_collect_gcc_options):
3249 Properly count arguments.
3250 (run_gcc): Use an obstack to collect argv, properly separate
3251 switches and their arguments.
3253 2011-10-28 Jakub Jelinek <jakub@redhat.com>
3255 * tree-vect-stmts.c (vectorizable_shift): Give up if op1 has different
3256 vector mode from vectype's mode.
3258 2011-10-28 Chung-Lin Tang <cltang@codesourcery.com>
3260 PR rtl-optimization/49720
3261 * simplify-rtx.c (simplify_relational_operation_1): Detect
3262 infinite recursion condition in "(eq/ne (plus x cst1) cst2)
3263 simplifies to (eq/ne x (cst2 - cst1))" case.
3265 2011-10-27 David S. Miller <davem@davemloft.net>
3267 * config/sparc/sparc.md (snedi_special): Only match when not VIS3.
3268 (*snedi_zero): Likewise.
3269 (*snedi_zero_trunc): Likewise.
3270 (snedi_special_vis3): New expander.
3271 (*snedi_zero_vis3): New insn.
3272 (*snedi_zero_trunc_vis3): Likewise.
3273 (*sltu_insn_vis3): Likewise.
3274 (*sltu_insn_vis3_trunc): Likewise.
3276 (*addxc_trunc_sp64_vis3): Likewise.
3277 * config/sparc/sparc.c (emit_scc_insn): When VIS3 use the
3278 gen_snedi_special_vis3 expander, and try GTU/LTU addx based
3279 sequences on DImode values.
3281 * config/sparc/sparc.md (64-bit vector moves): Use 'e' not 'f'
3284 * regcprop.c (copyprop_hardreg_forward_1): Reject the
3285 transformation when we narrow the mode on big endian.
3287 2011-10-27 Jakub Jelinek <jakub@redhat.com>
3289 * config/i386/sse.md (avx_cvtpd2dq256_2, avx_cvttpd2dq256_2,
3290 vec_pack_sfix_trunc_v4df, vec_pack_sfix_v4df): New expanders.
3291 (*avx_cvtpd2dq256_2, *avx_cvttpd2dq256_2): New insns.
3293 * config/i386/i386.c (ix86_print_operand): Handle 'q' and 'x'
3294 overrides for -masm=intel memory.
3295 * config/i386/sse.md (sse2_cvtdq2pd, sse2_cvtps2pd,
3296 sse4_1_<code>v8qiv8hi2, avx2_<code>v8qiv8si2,
3297 sse4_1_<code>v4hiv4si2, avx2_<code>v4hiv4di2,
3298 sse4_1_<code>v2siv2di2): Use %q1 instead of %1 for -masm=intel.
3299 (sse4_1_<code>v4qiv4si2, avx2_<code>v4qiv4di2,
3300 sse4_1_<code>v2hiv2di2): Use %k1 instead of %1 for -masm=intel.
3301 (sse4_1_<code>v2qiv2di2): Use %w1 instead of %1 for -masm=intel.
3303 2011-10-27 Martin Jambor <mjambor@suse.cz>
3305 * ipa-prop.c (compute_pass_through_member_ptrs): Rename parm_info
3307 (ipa_compute_jump_functions_for_edge): Likewise.
3308 (ipa_compute_jump_functions): Likewise.
3309 (ipa_analyze_indirect_call_uses): Likewise.
3310 (ipa_analyze_call_uses): Likewise.
3311 (ipa_analyze_params_uses): Likewise.
3312 (ipa_analyze_node): Likewise.
3314 2011-10-27 Uros Bizjak <ubizjak@gmail.com>
3317 * config/i386/sse.md (*avx_unpcklpd256): Remove extra insn
3318 constraints. Change alternative 1 to "x,m,1".
3320 2011-10-27 Jakub Jelinek <jakub@redhat.com>
3322 * Makefile.in (build/gencheck.o): Depend on tree.def and
3323 c-family/c-common.def.
3325 * tree-ssa-strlen.c: Include expr.h.
3326 (get_stridx): Don't use c_strlen, instead use string_constant
3327 and compute string length from it.
3328 * Makefile.in (tree-ssa-strlen.o): Depend on $(EXPR_H).
3330 2011-10-27 Eric Botcazou <ebotcazou@adacore.com>
3332 PR rtl-optimization/46603
3334 * reload.c (push_reload): In the out case, restore previous behavior
3335 for subregs that don't have word mode.