]> gcc.gnu.org Git - gcc.git/blame - gcc/ChangeLog
Update Copyright year.
[gcc.git] / gcc / ChangeLog
CommitLineData
4017e262
JJ
12009-01-09 Jakub Jelinek <jakub@redhat.com>
2
3 PR middle-end/38771
4 * fold-const.c (fold_unary): For COMPOUND_EXPR and COND_EXPR,
5 fold_convert arg0 operands to TREE_TYPE (op0) first.
6
7cd06356
VM
72009-01-08 Vladimir Makarov <vmakarov@redhat.com>
8
9 * params.def (ira-max-conflict-table-size): Decrease default value
10 to 1000.
11
96867bbd
JJ
122009-01-08 Jakub Jelinek <jakub@redhat.com>
13
14 PR tree-optimization/37031
15 * lambda-code.c (lambda_collect_parameters): Call pointer_set_destroy
16 on parameter_set.
17 (build_access_matrix): Reserve correct size for AM_MATRIX vector,
18 allocate it using gc instead of heap, use VEC_quick_push instead of
19 VEC_safe_push.
20 * graphite.c (build_access_matrix): Allocate AM_MATRIX vector using gc
21 instead of heap, use VEC_quick_push instead of VEC_safe_push.
22 * tree-data-ref.h (struct access_matrix): Change matrix to gc
23 allocated vector from heap allocated.
24 * lambda.h: Add DEF_VEC_ALLOC_P for gc allocated lambda_vector.
25 * tree-loop-linear.c (linear_transform_loops): Allocate nest
26 vector only after perfect_loop_nest_depth call.
27
68f61c3d
SP
282009-01-08 Sebastian Pop <sebastian.pop@amd.com>
29 Jan Sjodin <jan.sjodin@amd.com>
30
31 PR tree-optimization/38559
32 * graphite.c (debug_value, copy_constraint,
33 swap_constraint_variables, scale_constraint_variable, ): New.
34 (get_lower_bound, get_upper_bound): Removed.
35 (graphite_trans_bb_strip_mine): Clean up this code that works
36 only for constant number of iterations. Fully copy upper and
37 lower bound constraints, not only the constant part of them.
38 * graphite.h (debug_value): Declared.
39
82567e33
IR
402009-01-08 Ira Rosen <irar@il.ibm.com>
41
42 PR tree-optimization/37194
43 * tree-vect-transform.c (vect_estimate_min_profitable_iters):
44 Don't add the cost of cost model guard in prologue to scalar
45 outside cost in case of known number of iterations.
46
7da13f1d
NF
472009-01-07 Nathan Froyd <froydnj@codesourcery.com>
48 Alan Modra <amodra@bigpond.net.au>
49
50 * config/rs6000/rs6000.c (rs6000_legitimize_address): Check for
51 non-word-aligned REG+CONST addressing.
52
c0f64575
UB
532009-01-07 Uros Bizjak <ubizjak@gmail.com>
54
55 PR target/38706
56 * config/alpha/alpha.c (alpha_end_function): For TARGET_ABI_OSF, call
57 free_after_compilation when outputting a thunk.
58 (alpha_output_mi_thunk_osf): Assert that we are processing a thunk.
59 Do not call free_after_compilation here.
60
301b6a4e
UB
612009-01-07 Uros Bizjak <ubizjak@gmail.com>
62
63 * config/i386/i386.c (ix86_target_string): Use ARRAY_SIZE.
301b6a4e
UB
64 (ix86_valid_target_attribute_inner_p): Ditto.
65
6a114766
JS
662009-01-07 Jan Sjodin <jan.sjodin@amd.com>
67
68 PR tree-optimization/38492
69 PR tree-optimization/38498
70 * tree-check.c (operator_is_linear, scev_is_linear_expression): New.
71 * tree-chrec.h (scev_is_linear_expression): Declared.
72 * graphite.c (graphite_cannot_represent_loop_niter): New.
73 (scopdet_basic_block_info): Call graphite_cannot_represent_loop_niter.
74 (graphite_loop_normal_form): Use gcc_assert.
75 (scan_tree_for_params): Use CASE_CONVERT.
76 (phi_node_is_iv, bb_contains_non_iv_scalar_phi_nodes): New.
77 (build_scop_conditions_1): Call bb_contains_non_iv_scalar_phi_nodes.
78 Use gcc_assert. Discard scops that contain unhandled cases.
79 (build_scop_conditions): Return a boolean status for unhandled cases.
80 (strip_mine_profitable_p): Print the loop number, not its depth.
81 (is_interchange_valid): Pass the depth of the loop nest, don't
82 recompute it wrongly.
83 (graphite_trans_bb_block): Same.
84 (graphite_trans_bb_block): Print tentative of loop blocking.
85 (graphite_trans_scop_block): Do not print that the loop has been
86 blocked.
87 (graphite_transform_loops): Do not handle scops that contain condition
88 scalar phi nodes.
89
65b82caa
L
902009-01-07 H.J. Lu <hongjiu.lu@intel.com>
91
92 AVX Programming Reference (December, 2008)
93 * config/i386/avxintrin.h (_mm256_stream_si256): New.
94 (_mm256_stream_pd): Likewise.
95 (_mm256_stream_ps): Likewise.
96
97 * config/i386/i386.c (ix86_builtins): Add IX86_BUILTIN_MOVNTDQ256,
98 IX86_BUILTIN_MOVNTPD256 and IX86_BUILTIN_MOVNTPS256.
99 (ix86_special_builtin_type): Add VOID_FTYPE_PV4DI_V4DI.
100 (bdesc_special_args): Add __builtin_ia32_movntdq256,
101 __builtin_ia32_movntpd256 and __builtin_ia32_movntps256.
102 (ix86_init_mmx_sse_builtins): Handle VOID_FTYPE_PV4DI_V4DI.
103 (ix86_expand_special_args_builtin): Likewise.
104
105 * config/i386/sse.md (AVXMODEDI): New.
106 (avx_movnt<mode>): Likewise.
107 (avx_movnt<mode>): Likewise.
108 (<sse>_movnt<mode>): Remove AVX support.
109 (sse2_movntv2di): Likewise.
110
cef158f9
RG
1112009-01-07 Richard Guenther <rguenther@suse.de>
112
113 PR middle-end/38751
114 * fold-const.c (extract_muldiv): Remove obsolete comment.
115 (fold_plusminus_mult_expr): Undo MINUS_EXPR
116 to PLUS_EXPR canonicalization for the canonicalization.
117
246cb5be
GP
1182009-01-07 Gerald Pfeifer <gerald@pfeifer.com>
119
120 * doc/install.texi (alpha*-dec-osf*): Remove note on 32-bit
121 hosted cross-compilers generating less efficient code.
122
84b9b872
RS
1232009-01-06 Richard Sandiford <rdsandiford@googlemail.com>
124
125 * function.h (rtl_data): Add a dbr_scheduled_p field.
126 * reorg.c (dbr_schedule): Set it.
127 (gate_handle_delay_slots): Check it.
128 * config/mips/mips.c (mips_base_delayed_branch): Delete.
129 (mips_reorg): Check flag_delayed_branch instead of
130 mips_base_delayed_branch.
131 (mips_override_options): Don't set mips_base_delayed_branch
132 or flag_delayed_branch.
133
ace984c8
RS
1342009-01-06 Richard Sandiford <rdsandiford@googlemail.com>
135
136 PR rtl-optimization/38426.
137 * ira.c (ira): Set current_function_is_leaf earlier.
138
7ad7809b
JJ
1392009-01-06 Jakub Jelinek <jakub@redhat.com>
140
141 PR rtl-optimization/38722
142 * combine.c (try_combine): Don't modify PATTERN (i3) and notes
143 too early, only set a flag and modify after last possible
144 undo_all point.
145
c52ec948
JJ
1462009-01-06 Janis Johnson <janis187@us.ibm.com>
147
148 PR c/34252
149 * ginclude/float.h: Rename DECnn_DEN to DECnn_SUBNORMAL_MIN.
150 * real.c (decimal_single_format): Correct values of emin and emax.
151 (decimal_double_format): Ditto.
152 (decimal_quad_format): Ditto.
153 * c-cppbuiltin.c (builtin_define_decimal_float_constants): Adjust
154 computation of DECnn_MIN and DECnn_MAX for corrected values of
155 emin and emax. Define __DECnn_SUBNORMAL_MIN__ instead of
156 __DECnn_MIN__, and adjust its computation for the corrected value
157 of emin.
158
300240e1
JH
1592009-01-06 Jan Hubicka <jh@suse.cz>
160
161 PR target/38744
301b6a4e 162 * i386.c (ix86_expand_call): Use ARRAY_SIZE.
300240e1 163
271c069e
GP
1642009-01-06 Gerald Pfeifer <gerald@pfeifer.com>
165
166 * doc/contrib.texi (Contributors): Slightly adjust the end note.
167 Add Robert Clark to the list of testers.
168
64e4c301
JH
1692009-01-06 Jan Hubicka <jh@suse.cz>
170 Kai Tietz <kai.tietz@onevision.com>
171
172 * i386.md (*msabi_syvabi): Add SSE regs clobbers.
173 * i386.c (ix86_expand_call): Add clobbers.
174
7650b83e
JH
1752009-01-06 Jan Hubicka <jh@suse.cz>
176 Kai Tietz <kai.tietz@onevision.com>
177
178 * i386.h (CONDITIONAL_CALL_USAGE): SSE regs are not used for w64 ABI.
179 * i386.c (struct ix86_frame): Add padding0 and nsseregs.
180 (ix86_nsaved_regs): Count only general purpose regs.
181 (ix86_nsaved_sseregs): New.
182 (ix86_compute_frame_layout): Update nsseregs; set preferred alignment
183 to 16 for w64; compute padding and size of sse reg save area.
301b6a4e
UB
184 (ix86_emit_save_regs, ix86_emit_save_regs_using_mov): Save only
185 general purpose regs.
7650b83e
JH
186 (ix86_emit_save_sse_regs_using_mov): New.
187 (ix86_expand_prologue): Save SSE regs if needed.
188 (ix86_emit_restore_regs_using_mov): Use only general purpose regs.
189 (ix86_emit_restore_sse_regs_using_mov): New.
190 (ix86_expand_epilogue): Save SSE regs if needed.
191
9aa5c1b2
JH
1922009-01-06 Jan Hubicka <jh@suse.cz>
193 Kai Tietz <kai.tietz@onevision.com>
194
195 * i386.h (ACCUMULATE_OUTGOING_ARGS): Enable for MSABI
301b6a4e
UB
196 * i386.c (init_cumulative_args): Disallow calls of MSABI functions
197 when accumulate outgoing args is off.
9aa5c1b2 198
f7556aae
L
1992009-01-06 H.J. Lu <hongjiu.lu@intel.com>
200
201 PR bootstrap/38742
202 * ira-color.c (ira_reuse_stack_slot): Check ENABLE_IRA_CHECKING
203 before using pseudos_have_intersected_live_ranges_p.
204
205 * ira-int.h (ira_assert): Always define.
206
e47b7d04
L
2072009-01-06 H.J. Lu <hongjiu.lu@intel.com>
208
209 AVX Programming Reference (December, 2008)
210 * config/i386/avxintrin.h (_mm_permute2_pd): Removed.
211 (_mm256_permute2_pd): Likewise.
212 (_mm_permute2_ps): Likewise.
213 (_mm256_permute2_ps): Likewise.
214 * config/i386/i386.md (UNSPEC_VPERMIL2): Likewise.
215 * config/i386/sse.md (avx_vpermil2<mode>3): Likewise.
216
217 * config/i386/i386.c (ix86_builtins): Remove
218 IX86_BUILTIN_VPERMIL2PD, IX86_BUILTIN_VPERMIL2PS,
219 IX86_BUILTIN_VPERMIL2PD256 and IX86_BUILTIN_VPERMIL2PS256.
220 (ix86_builtin_type): Remove V8SF_FTYPE_V8SF_V8SF_V8SI_INT,
221 V4DF_FTYPE_V4DF_V4DF_V4DI_INT, V4SF_FTYPE_V4SF_V4SF_V4SI_INT
222 and V2DF_FTYPE_V2DF_V2DF_V2DI_INT.
223 (bdesc_args): Remove __builtin_ia32_vpermil2pd,
224 __builtin_ia32_vpermil2ps, __builtin_ia32_vpermil2pd256 and
225 __builtin_ia32_vpermil2ps256.
226 (ix86_init_mmx_sse_builtins): Updated.
227 (ix86_expand_args_builtin): Likewise.
228
44b86471
JDA
2292009-01-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
230
e47b7d04
L
231 * pa.c (output_call): Relocate non-jump insns in the delay slot of
232 long absolute calls when generating PA 2.0 code.
44b86471 233
311aab06
VM
2342009-01-05 Vladimir Makarov <vmakarov@redhat.com>
235
236 PR rtl-optimization/38583
237 * params.h (IRA_MAX_CONFLICT_TABLE_SIZE): New macro.
238
239 * params.def (ira-max-conflict-table-size): New.
240
241 * doc/invoke.texi (ira-max-conflict-table-size): Decribe.
242
243 * ira.h (ira_conflicts_p): New external definition.
244
245 * ira-conflicts.c (build_conflict_bit_table): Do not build too big
246 table. Report this. Return result of building.
247 (ira_build_conflicts): Use ira_conflicts_p. Check result of
248 building conflict table.
249
301b6a4e 250 * ira-color.c (fast_allocation): Use num instead of ira_allocnos_num.
311aab06
VM
251 (ira_color): Use ira_conflicts_p.
252
253 * global.c: Include ira.h.
254 (pseudo_for_reload_consideration_p, build_insn_chain): Use
255 ira_conflicts_p.
256
257 * Makefile.in (global.o): Add ira.h.
258
259 * ira-build.c (mark_all_loops_for_removal,
260 propagate_some_info_from_allocno): New.
261 (remove_unnecessary_allocnos): Call
262 propagate_some_info_from_allocno.
263 (remove_low_level_allocnos): New.
264 (remove_unnecessary_regions): Add parameter. Call
265 mark_all_loops_for_removal and remove_low_level_allocnos. Pass
266 parameter to remove_unnecessary_regions.
267 (ira_build): Remove all regions but root if the conflict table was
268 not built. Update conflict hard regs for allocnos crossing calls.
269
270 * ira.c (ira_conflicts_p): New global.
271 (ira): Define and use ira_conflicts_p.
272
273 * reload1.c (compute_use_by_pseudos, reload, count_pseudo,
274 count_spilled_pseudo, find_reg, alter_reg, finish_spills,
301b6a4e 275 emit_input_reload_insns, delete_output_reload): Use ira_conflicts_p.
311aab06 276
5641785f
BE
2772009-01-06 Ben Elliston <bje@au.ibm.com>
278
279 * gengtype-lex.l (YY_NO_INPUT): Define.
280
e4e5261f
AP
2812009-01-05 Andrew Pinski <andrew_pinski@playstation.sony.com>
282
283 PR c/34911
5641785f
BE
284 * c-common.c (handle_vector_size_attribute): Also reject
285 BOOLEAN_TYPE types.
e4e5261f 286
7b10257f
SP
2872009-01-05 Sebastian Pop <sebastian.pop@amd.com>
288
289 PR tree-optimization/38492
290 * graphite.c (rename_map_elt, debug_rename_elt,
301b6a4e
UB
291 debug_rename_map_1, debug_rename_map, new_rename_map_elt,
292 rename_map_elt_info, eq_rename_map_elts,
293 get_new_name_from_old_name, bb_in_sese_p): Moved around.
7b10257f
SP
294 (sese_find_uses_to_rename_use): Renamed sese_build_livein_liveouts_use.
295 (sese_find_uses_to_rename_bb): Renamed sese_build_livein_liveouts_bb.
296 (sese_build_livein_liveouts): New.
297 (new_sese, free_sese): New.
298 (new_scop): Call new_sese.
299 (free_scop): Call free_sese.
300 (rename_variables_from_edge, rename_phis_end_scop): Removed.
301 (register_old_new_names): Renamed register_old_and_new_names.
302 (register_scop_liveout_renames, add_loop_exit_phis,
303 insert_loop_close_phis, struct igp,
304 default_liveout_before_guard, add_guard_exit_phis,
305 insert_guard_phis, copy_renames): New.
306 (translate_clast): Call insert_loop_close_phis and insert_guard_phis.
307 (sese_add_exit_phis_edge): Renamed scop_add_exit_phis_edge.
308 (rewrite_into_sese_closed_ssa): Renamed scop_insert_phis_for_liveouts.
309 (scop_adjust_phis_for_liveouts): New.
310 (gloog): Call scop_adjust_phis_for_liveouts.
311
312 * graphite.h (struct sese): Documented. Added fields liveout,
301b6a4e 313 num_ver and livein.
7b10257f
SP
314 (SESE_LIVEOUT, SESE_LIVEIN, SESE_LIVEIN_VER, SESE_NUM_VER): New.
315 (new_sese, free_sese, sese_build_livein_liveouts): Declared.
316 (struct scop): Added field liveout_renames.
317 (SCOP_LIVEOUT_RENAMES): New.
318
9761fcc7
HJ
3192009-01-05 Harsha Jagasia <harsha.jagasia@amd.com>
320
321 PR tree-optimization/38510
322 * graphite.c (recompute_all_dominators): Call mark_irreducible_loops.
323 (translate_clast): Call recompute_all_dominators before
324 graphite_verify.
325 (gloog): Call recompute_all_dominators before graphite_verify.
326
3272009-01-05 Harsha Jagasia <harsha.jagasia@amd.com>
328 Jan Sjodin <jan.sjodin@amd.com>
329
330 PR tree-optimization/38500
331 * graphite.c (create_sese_edges): Call fix_loop_structure after
332 splitting blocks.
333
48a01864
JS
3342009-01-05 Joel Sherrill <joel.sherrill@oarcorp.com>
335
336 * config.gcc: Add m32r*-*-rtems*.
337 * config/m32r/rtems.h: New file.
338
fc0cd180
BE
3392009-01-05 Ben Elliston <bje@au.ibm.com>
340
341 * Makefile.in (.po.gmo): Use mkinstalldirs, not test -d || mkdir.
342 (.po.pox): Likewise.
343 (po/gcc.pot): Likewise.
344
3106d8e5
DM
3452009-01-04 David S. Miller <davem@davemloft.net>
346
347 * config/sparc/sparc.h (SECONDARY_MEMORY_NEEDED_RTX): Delete.
348 (STARTING_FRAME_OFFSET): Always set to zero.
349
0009b473
RS
3502009-01-04 Richard Sandiford <rdsandiford@googlemail.com>
351
352 * tree.def (LSHIFT_EXPR, RSHIFT_EXPR): Add commentary.
353 * tree-cfg.c (verify_gimple_assign_binary): Allow shifts of
354 fixed-point types, and vectors of the same.
355
4f24c6d6
RS
3562009-01-04 Richard Sandiford <rdsandiford@googlemail.com>
357
358 * config/mips/sync.md (*mb_barrier): Rename to...
359 (*memory_barrier): ...this.
360
1df48f5c
JW
3612009-01-04 Jonathan Wakely <jwakely.gcc@gmail.com>
362
363 * doc/extend.texi (Function Attributes): Move @cindex after @item
eefe5da0 364 for 'artificial' and 'flatten'. Fix grammar for 'externally_visible'
1df48f5c
JW
365 and put in alphabetical order. Fix 'target' name and put in order.
366 * doc/invoke.texi (-Wstrict-null-sentinel, -fipa-matrix-reorg): Fix
367 typos.
368
1a8c13b3
UB
3692009-01-04 Uros Bizjak <ubizjak@gmail.com>
370
371 * config/s390/s390.md (UNSPEC_MB): Rename from UNSPECV_MB.
372 (memory_barrier): Expand as unspec instead of unspec_volatile.
373 Remove mem:BLK from insn operands. Use Pmode scratch register.
374 (*memory_barrier): Define as unspec instead of unspec_volatile.
375 Use (match_dup 0) as input operand.
376
377 * config/sparc/sparc.md (UNSPEC_MEMBAR): Rename from UNSPECV_MEMBAR.
378 * config/sparc/sync.md (memory_barrier): Expand as unspec instead of
379 unspec_volatile. Remove mem:BLK from insn operands. Use Pmode
380 scratch register. Remove operand 1.
381 (*stbar): Define as unspec instead of unspec_volatile.
382 Use (match_dup 0) as input operand, remove (const_int 8).
383 (*membar): Define as unspec instead of unspec_volatile.
384 Use (match_dup 0) as input operand, remove input operand 2.
385
386 * config/xtensa/xtensa.md (UNSPEC_MEMW): Rename from UNSPECV_MEMW.
387 (memory_barrier): Expand as unspec instead of unspec_volatile.
388 Remove mem:BLK from insn operands. Use Pmode scratch register.
389 (*memory_barrier): Define as unspec instead of unspec_volatile.
390 Use (match_dup 0) as input operand.
391
392 * config/ia64/sync.md (memory_barrier): Redefine as expander pattern.
393 Remove mem:BLK from insn operands. Use Pmode scratch register.
394 Set volatile flag on operand 0.
395 (*memory_barrier): New insn pattern.
396
397 * config/rs6000/sync.md (memory_barrier): Remove mem:BLK from
398 insn operands.
399 (*memory_barrier): Use (match_dup 0) as input operand.
400
401 * config/mips/sync.md (memory_barrier): Redefine as expander pattern.
402 Remove mem:BLK from insn operands. Use Pmode scratch register.
403 Set volatile flag on operand 0.
301b6a4e 404 (*mb_internal): New insn pattern.
1a8c13b3
UB
405
406 * config/alpha/sync.md (*memory_barrier): Rename from *mb_internal.
407
fb0703f7
SB
4082009-01-04 Steven Bosscher <steven@gcc.gnu.org>
409
410 PR middle-end/38586
411 * function.c (struct temp_slot): Move to the section of the file
412 that deals with temp slots. Remove field 'address'.
413 (temp_slot_address_table): New hash table of address -> temp slot.
414 (struct temp_slot_address_entry): New struct, items for the table.
415 (temp_slot_address_compute_hash, temp_slot_address_hash,
416 temp_slot_address_eq, insert_temp_slot_address): Support functions
417 for the new table.
418 (find_temp_slot_from_address): Rewrite to use the new hash table.
419 (remove_unused_temp_slot_addresses): Remove addresses of temp
420 slots that have been made available.
421 (remove_unused_temp_slot_addresses_1): Call-back for htab_traverse,
422 worker function for remove_unused_temp_slot_addresses.
423 (assign_stack_temp_for_type): Don't clear the temp slot address list.
424 Add the temp slot address to the address -> temp slot map.
425 (update_temp_slot_address): Update via insert_temp_slot_address.
426 (free_temp_slots): Call remove_unused_temp_slot_addresses.
427 (pop_temp_slots): Likewise.
428 (init_temp_slots): Allocate the address -> temp slot map, or empty
429 the map if it is already allocated.
430 (prepare_function_start): Initialize temp slot processing.
431
b5a430f3
SB
4322009-01-04 Steven Bosscher <steven@gcc.gnu.org>
433
434 PR middle-end/38584
435 * cfgexpand.c (estimate_stack_frame_size): Simplify the estimate:
436 Calculate the size of all stack vars assuming no packing of stack
437 vars will happen, replacing a quadratic algorithm with a linear one.
438
5d560619
JJ
4392009-01-03 Jakub Jelinek <jakub@redhat.com>
440
441 PR target/38707
442 * expmed.c (store_bit_field_1): Don't modify op0 if movstrict insn
443 can't be used.
444
953a7caa
DN
4452009-01-03 Diego Novillo <dnovillo@google.com>
446
447 * doc/contrib.texi: Update contributions.
448
5f26a230
JJ
4492009-01-03 Jakub Jelinek <jakub@redhat.com>
450
493aa551
JJ
451 PR c++/38705
452 * builtins.c (fold_builtin_memory_op): Give up if either operand
453 is volatile. Set srctype or desttype to non-qualified version
454 of the other type.
455
5f26a230
JJ
456 PR c/38700
457 * builtins.c (fold_builtin_expect): Only check DECL_WEAK for VAR_DECLs
458 and FUNCTION_DECLs.
459
fafe34f9
KZ
4602009-01-02 Kenneth Zadeck <zadeck@naturalbridge.com>
461
462 PR rtl-optimization/35805
463 * df-problems.c (df_lr_finalize): Add recursive call to resolve lr
464 problem if fast dce is able to remove any instructions.
465 * dce.c (dce_process_block): Fix dump message.
466
0b355794
MM
4672009-01-02 Mark Mitchell <mark@codesourcery.com>
468
469 PR 33649
470 * tree-ssa-pre.c (compute_antic): Correct loop bounds.
471
0cf0d02b
JJ
4722009-01-02 Jakub Jelinek <jakub@redhat.com>
473
474 PR middle-end/38690
475 * tree-flow.h (op_code_prio, op_prio): New prototypes.
476 * tree-pretty-print.c (op_code_prio): New function.
477 (op_prio): No longer static. Use op_code_prio.
478 * gimple-pretty-print.c (dump_unary_rhs, dump_binary_rhs):
479 Use op_prio and op_code_prio to determine if () should be
480 printed around operand(s) or not.
481
482 * gimple-pretty-print.c (dump_unary_rhs, dump_binary_rhs,
483 dump_gimple_call, dump_gimple_switch, dump_gimple_cond,
484 dump_gimple_label, dump_gimple_try, dump_symbols, dump_gimple_phi,
485 dump_gimple_mem_ops, dump_bb_header, dump_bb_end, pp_cfg_jump): Use
486 pp_character instead of pp_string for single letter printing.
487
e3606f3b
RS
4882009-01-02 Richard Sandiford <rdsandiford@googlemail.com>
489
490 * doc/extend.texi: Fix '#pragma GCC option' typo.
491
00f39bd5
RG
4922009-01-02 Richard Guenther <rguenther@suse.de>
493
a5f061dc
RG
494 * doc/install.texi (--enable-checking): Mention different
495 default for stage1.
496 (--enable-stage1-checking): Document.
00f39bd5 497
26de0bcb
AP
4982009-01-01 Andrew Pinski <pinskia@gmail.com>
499
500 PR middle-end/30142
501 * tree-cfg.c (verify_expr): Add INDIRECT_REF case. Change MODIFY_EXPR
502 case to be an error.
503
7a7d8fab
BE
5042009-01-02 Ben Elliston <bje@au.ibm.com>
505
506 * config/fp-bit.h (pack_d): Constify argument.
507 * config/fp-bit.c (makenan): Constify return type. Remove casts.
508 (isnan): Constify argument.
509 (isinf): Likewise.
510 (iszero): Likewise.
511 (pack_d): Likewise.
512 (_fpadd_parts): Constify return type.
513 (_fpmul_parts): Likewise.
514 (_fpdiv_parts): Likewise.
515
b295aee2
JJ
5162009-01-01 Jakub Jelinek <jakub@redhat.com>
517
518 PR c/36489
519 * c-typeck.c (add_pending_init): Add IMPLICIT argument. Only
520 warn about overwriting initializer with side-effects or
521 -Woverride-init if !IMPLICIT.
522 (output_init_element): Likewise. Pass IMPLICIT down to
523 add_pending_init.
524 (process_init_element): Add IMPLICIT argument. Pass it down
525 to output_init_element.
526 (push_init_element, pop_init_level, set_designator): Adjust
527 process_init_element callers.
528 (set_nonincremental_init, set_nonincremental_init_from_string):
529 Adjust add_pending_init callers.
530 (output_pending_init_elements): Adjust output_init_element callers.
531 * c-tree.h (process_init_element): Adjust prototype.
532 * c-parser.c (c_parser_initelt, c_parser_initval): Adjust
533 process_init_element callers.
This page took 0.175014 seconds and 5 git commands to generate.