David S. Miller [Mon, 7 Nov 2011 02:22:40 +0000 (02:22 +0000)]
Get rid of sparc's UNSPEC_SHORT_LOAD.
* config/sparc/sparc.md (UNSPEC_SHORT_LOAD): Delete.
(zero_extend_v8qi_vis, zero_extend_v4hi_vis,
*zero_extend_v8qi_<P:mode>_insn,
*zero_extend_v4hi_<P:mode>_insn): Express using vec_merge
and vec_duplicate instead of using an UNSPEC.
Eric Botcazou [Mon, 7 Nov 2011 00:22:18 +0000 (00:22 +0000)]
gcse.c: Adjust outdated comments throughout.
* gcse.c: Adjust outdated comments throughout.
(struct mem_conflict_info): New structure.
(mems_conflict_for_gcse_p): Use it to communicate with caller.
(load_killed_in_block_p): Pass it to note_stores.
(hash_expr): Remove superfluous line break.
(hash_scan_set): Rename PAT parameter into SET.
(hash_scan_insn): Reorder cases.
(canon_list_insert): Fix long line.
(edge_list): Delete.
(prune_expressions): Rename E local variable into EXPR.
(compute_pre_data): Return struct edge_list * object.
(pre_expr_reaches_here_p_work): Fix formatting.
(process_insert_insn): Move around comment.
(pre_edge_insert): Fix long line.
(pre_insert_copies): Likewise.
(gcse_emit_move_after): Swap SRC and DEST parameters.
(pre_delete): Adjust call to gcse_emit_move_after.
(pre_gcse): Take struct edge_list * parameter. Fix long line.
(one_pre_gcse_pass): Use flag_gcse_lm condition for all routines.
Use a local list of edges.
(hoist_code): Fix long line. Adjust call to gcse_emit_move_after.
(pre_ldst_expr_hash): Fix long line.
(free_ldst_mems): Rename into...
(free_ld_motion_mems): ...this.
(first_ls_expr): Delete.
(next_ls_expr): Likewise.
(print_ldst_list): Do not use above two functions.
(simple_mem): Adjust interface.
(compute_ld_motion_mems): Fix formatting.
(update_ld_motion_stores): Reuse local variable.
c-family:
* c-common.c (c_common_reswords): Add _Alignas and _Alignof.
(c_sizeof_or_alignof_type): Diagnose alignof applied to a function
type.
(check_user_alignment): New. Split out of
handle_aligned_attribute. Disallow integer constants with
noninteger types. Conditionally allow zero.
(handle_aligned_attribute): Use check_user_alignment.
* c-common.h (RID_ALIGNAS, check_user_alignment): New.
Jonathan Wakely [Sun, 6 Nov 2011 23:25:25 +0000 (23:25 +0000)]
backwards_compatibility.xml: Fix autoconf tests for C++11 compiler features and library headers.
* doc/xml/manual/backwards_compatibility.xml: Fix autoconf tests for
C++11 compiler features and library headers. Add stable id
attributes. Use <filename> element for headers and surround in angle
brackets. Use <classname> for classes.
* doc/html/*: Regenerate.
Joern Rennecke [Sun, 6 Nov 2011 22:58:54 +0000 (22:58 +0000)]
regset.h (fixed_reg_set_regset): Declare.
* regset.h (fixed_reg_set_regset): Declare.
* dse.c: Include regset.h .
(struct insn_info): Add member fixed_regs_live.
(note_add_store_info): New typedef.
(note_add_store): New function.
(emit_inc_dec_insn_before): Expect arg to be of type insn_info_t .
Use gen_add3_insn / gen_move_insn.
Check new insn for unwanted clobbers before emitting it.
(check_for_inc_dec): Rename to...
(check_for_inc_dec_1:) ... this. Return bool. Take insn_info
parameter. Changed all callers in file.
(check_for_inc_dec, copy_fixed_regs): New functions.
(scan_insn): Set fixed_regs_live field of insn_info.
* rtl.h (check_for_inc_dec): Update prototype.
* postreload.c (reload_cse_simplify): Take new signature of
check_ind_dec into account.
* reginfo.c (fixed_reg_set_regset): New variable.
(init_reg_sets_1): Initialize it.
Co-Authored-By: Eric Botcazou <ebotcazou@adacore.com>
From-SVN: r181046
pa.c (pa_hpux_init_libfuncs): Rename to pa_init_libfuncs.
* config/pa/pa.c (pa_hpux_init_libfuncs): Rename to pa_init_libfuncs.
Remove dependence of declaration and target define on definition of
HPUX_LONG_DOUBLE_LIBRARY. Update implementation.
François Dumont [Sun, 6 Nov 2011 17:16:00 +0000 (17:16 +0000)]
41975.cc: Add tests to check performance with or without cache of hash code and with...
2011-11-06 François Dumont <fdumont@gcc.gnu.org>
* testsuite/performance/23_containers/insert_erase/41975.cc: Add
tests to check performance with or without cache of hash code and with
string type that has a costlier hash functor than int type.
* tree-vectorizer.h (vectorizable_condition): Add argument.
* tree-vect-loop.c (vectorizable_reduction): Fail for condition
in SLP. Update calls to vectorizable_condition.
* tree-vect-stmts.c (vect_is_simple_cond): Add basic block info
to the arguments. Pass it to vect_is_simple_use_1.
(vectorizable_condition): Add slp_node to the arguments.
Support vectorization of basic blocks. Fail for reduction in
SLP. Update calls to vect_is_simple_cond and
vect_is_simple_use. Support SLP: call vect_get_slp_defs to get
vector operands.
(vect_analyze_stmt): Update calls to vectorizable_condition.
(vect_transform_stmt): Likewise.
* tree-vect-slp.c (vect_create_new_slp_node): Handle COND_EXPR.
(vect_get_and_check_slp_defs): Handle COND_EXPR. Allow pattern
def stmts.
(vect_build_slp_tree): Handle COND_EXPR.
(vect_analyze_slp_instance): Push pattern statements to root
node.
(vect_get_constant_vectors): Fix comments. Handle COND_EXPR.
David S. Miller [Sun, 6 Nov 2011 02:39:03 +0000 (19:39 -0700)]
More improvements to sparc VIS vec_init code generation.
gcc/
* config/sparc/sparc.md (UNSPEC_SHORT_LOAD): New unspec.
(zero-extend_v8qi_vis, zero_extend_v4hi_vis): New expanders.
(*zero_extend_v8qi_<P:mode>_insn,
*zero_extend_v4hi_<P:mode>_insn): New insns.
* config/sparc/sparc.c (vector_init_move_words,
vector_init_prepare_elts, sparc_expand_vector_init_vis2,
sparc_expand_vector_init_vis1): New functions.
(vector_init_bshuffle): Rewrite to handle more cases and make use
of locs[] array prepared by vector_init_prepare_elts.
(vector_init_fpmerge, vector_init_faligndata): Delete.
(sparc_expand_vector_init): Rewrite using new infrastructure.
gcc/testsuite/
* lib/test-supports.exp
(check_effective_target_ultrasparc_vis2_hw): New proc.
(check_effective_target_ultrasparc_vis3_hw): New proc.
* gcc.target/sparc/vec-init-1.inc: New vector init common code.
* gcc.target/sparc/vec-init-2.inc: Likewise.
* gcc.target/sparc/vec-init-3.inc: Likewise.
* gcc.target/sparc/vec-init-1-vis1.c: New test.
* gcc.target/sparc/vec-init-1-vis2.c: New test.
* gcc.target/sparc/vec-init-1-vis3.c: New test.
* gcc.target/sparc/vec-init-2-vis1.c: New test.
* gcc.target/sparc/vec-init-2-vis2.c: New test.
* gcc.target/sparc/vec-init-2-vis3.c: New test.
* gcc.target/sparc/vec-init-3-vis1.c: New test.
* gcc.target/sparc/vec-init-3-vis2.c: New test.
* gcc.target/sparc/vec-init-3-vis3.c: New test.
Jakub Jelinek [Sat, 5 Nov 2011 19:58:37 +0000 (20:58 +0100)]
re PR tree-optimization/50693 (Loop optimization restricted by GOTOs)
PR tree-optimization/50693
* tree-cfg.c (gimple_can_merge_blocks_p): Allow merging with
non-forced user labels.
(gimple_merge_blocks): Turn non-forced user labels into
debug bind stmt with the label as first operand and reset value.
(gimple_duplicate_bb): Don't duplicate label debug stmts.
* dwarf2out.c (gen_label_die): Handle NOTE_INSN_DELETED_DEBUG_LABEL.
* final.c (final_scan_insn): Likewise.
(rest_of_clean_state): Don't dump NOTE_INSN_DELETED_DEBUG_LABEL.
* var-tracking.c (debug_label_num): New variable.
(delete_debug_insns): Don't delete DEBUG_INSNs for LABEL_DECLs,
instead turn them into NOTE_INSN_DELETED_DEBUG_LABEL notes.
* cfglayout.c (skip_insns_after_block, duplicate_insn_chain): Handle
NOTE_INSN_DELETED_DEBUG_LABEL.
(duplicate_insn_chain): Don't duplicate LABEL_DECL DEBUG_INSNs.
* insn-notes.def (DELETED_DEBUG_LABEL): New note kind.
* print-rtl.c (print_rtx): Handle NOTE_INSN_DELETED_DEBUG_LABEL.
* gengtype.c (adjust_field_rtx_def): Likewise.
* config/i386/i386.c (ix86_output_function_epilogue): For MachO
clear CODE_LABEL_NUMBER of NOTE_INSN_DELETED_DEBUG_LABEL
if their are at the end of function and nop hasn't been emitted.
* config/rs6000/rs6000.c (rs6000_output_function_epilogue): Likewise.
David S. Miller [Fri, 4 Nov 2011 20:25:59 +0000 (20:25 +0000)]
Fix sparc regression due to recent movcc pattern changes.
PR target/49965
* config/sparc/sparc.c (sparc_expand_conditional_move): Handle the
fact that sparc_emit_float_lib_cmp modifies the comparison in
operands[1].
gcc-dg.exp (gcc_force_conventional_output): New global variable, default empty, -ffat-lto-objects for effective_target_lto.
* lib/gcc-dg.exp (gcc_force_conventional_output): New global
variable, default empty, -ffat-lto-objects for effective_target_lto.
(gcc-dg-test-1): Add options from dg-final methods.
* lib/scanasm.exp (scan-assembler_required_options)
(scan-assembler-not_required_options): New procs.
Arnaud Charlet [Fri, 4 Nov 2011 14:09:05 +0000 (15:09 +0100)]
[multiple changes]
2011-11-04 Robert Dewar <dewar@adacore.com>
* sprint.adb (Sprint_Node_Actual, case Qualified_Expression):
Avoid junk semicolon after argument of machine code Asm operand.
2011-11-04 Robert Dewar <dewar@adacore.com>
* exp_ch11.adb (Generate_Push_Pop): Inhibit push/pop nodes in
CodePeer mode or if restriction No_Exception_Handlers is present.
* exp_ch6.adb (Expand_N_Subprogram_Body): (Inhibit push/pop
nodes in CodePeer mode or if restriction No_Exception_Handlers
is present.
2011-11-04 Robert Dewar <dewar@adacore.com>
* s-tassta.adb, atree.ads, errout.adb, sinput.adb: Minor reformatting.
2011-11-04 Ed Schonberg <schonberg@adacore.com>
* sem_ch6.adb (Analyze_Subprogram_Specification): The
specification is legal if it is a function that returns an
abstract type, if it comes from an attribute renaming of a stream
attribute of an abstract type.
2011-11-04 Gary Dismukes <dismukes@adacore.com>
* exp_util.adb (Is_Possibly_Unaligned_Object): In case of indexed
components, check whether recursively check whether the prefix
denotes an unaligned object.
2011-11-04 Ed Schonberg <schonberg@adacore.com>
* sem_prag.adb (Analyze_Pragma, case Debug): The argument of
the pragma is legal if it is an expanded name that denotes a
procedure that be can called without parameters.
2011-11-04 Eric Botcazou <ebotcazou@adacore.com>
* gnat_ugn.texi (Performance Considerations) <Vectorization
of loops>: New sub-section. <Other Optimization Switches>:
Minor tweak.
Arnaud Charlet [Fri, 4 Nov 2011 14:00:29 +0000 (15:00 +0100)]
[multiple changes]
2011-11-04 Robert Dewar <dewar@adacore.com>
* gnat_rm.texi: Minor reformatting.
2011-11-04 Matthew Heaney <heaney@adacore.com>
* a-convec.adb, a-coinve.adb, a-cobove.adb (Merge): Raise PE
when Target and Source denote same non-empty object
* a-cdlili.adb, a-cidlli.adb, a-cbdlli.adb (Merge): Ditto
2011-11-04 Robert Dewar <dewar@adacore.com>
* exp_attr.adb: Minor reformatting.
2011-11-04 Ed Schonberg <schonberg@adacore.com>
* exp_ch5.adb (Expand_Assign_Record): Do not generate a
discriminant assignment within an initialization proc if the
record is an unchecked union, as it can only come from the
initialization of an unchecked union component.
2011-11-04 Robert Dewar <dewar@adacore.com>
* gnat_ugn.texi: Minor reformatting.
2011-11-04 Robert Dewar <dewar@adacore.com>
* par-labl.adb (Rewrite_As_Loop): Generate info msg rather than
warning message.
2011-11-04 Robert Dewar <dewar@adacore.com>
* exp_ch4.adb: Minor code reorganization (remove junk obsolete
var Save_Space).
* exp_alfa.adb: Add local constant
Disable_Processing_Of_Renamings;
(Expand_Alfa_N_Object_Renaming_Declaration): Disable
the name evaluation of object renamings for now.
(Expand_Potential_Renaming): Do not perform the substitution
for now.
* exp_util.adb (Remove_Side_Effects): Remove processing for
functions with side effects in Alfa mode.
Arnaud Charlet [Fri, 4 Nov 2011 13:55:34 +0000 (14:55 +0100)]
[multiple changes]
2011-11-04 Gary Dismukes <dismukes@adacore.com>
* bindgen.adb (Gen_Elab_Calls): In the case
of the AAMP target, set elaboration entities to 1 rather than
incrementing.
2011-11-04 Ed Schonberg <schonberg@adacore.com>
* sem_ch10.adb (Install_Limited_With_Unit): To establish the
proper entities on the ancestors of a child unit that appear
in a limited_with clause, follow the unit links because the
units are not analyzed and scope information is incomplete.
2011-11-04 Eric Botcazou <ebotcazou@adacore.com>
* exp_ch4.adb (Expand_N_Selected_Component): Refine code
setting the Atomic_Sync_Required flag to detect one more case.
* exp_util.adb (Activate_Atomic_Synchronization): Refine code
setting the Atomic_Sync_Required flag to exclude more cases,
depending on the parent of the node to be examined.
2011-11-04 Bob Duff <duff@adacore.com>
* g-excact.adb: Minor: use named notation.
2011-11-04 Ed Schonberg <schonberg@adacore.com>
* sem_ch5.adb: Improve error messages for illegal iterators.
* exp_alfa.adb: Add with and use clauses for Exp_Ch8 and
Sem_Util.
(Expand_Alfa): Alphabetize cases on first choice. Add
processing for object renaming declarations, identifiers and
expanded names.
(Expand_Alfa_N_In): Remove useless return.
(Expand_Alfa_N_Object_Renaming_Declaration): New routine.
(Expand_Potential_Renaming): New routine.
* exp_ch8.adb (Evaluate_Name): Moved to the top level.
(Expand_N_Object_Declaration): Alphabetize local variables. Move
Evaluate_Name out to the top level.
* exp_ch8.ads (Evaluate_Name): Moved from body to package spec.
* exp_util.adb (Remove_Side_Effects): Add processing for
functions with side effects in Alfa mode.
Arnaud Charlet [Fri, 4 Nov 2011 13:48:52 +0000 (14:48 +0100)]
[multiple changes]
2011-11-04 Yannick Moy <moy@adacore.com>
* sem_prag.adb: Minor refactoring (renaming of a parameter).
2011-11-04 Robert Dewar <dewar@adacore.com>
* atree.ads: Minor reformatting.
2011-11-04 Robert Dewar <dewar@adacore.com>
* checks.adb (Atomic_Synchronization_Disabled): Check -gnatd.d
and -gnatd.e here
* exp_ch2.adb (Expand_Entity_Reference): Use
Activate_Atomic_Synchronization
* exp_ch4.adb (Expand_N_Explicit_Dereference): Use
Activate_Atomic_Synchronization (Expand_N_Indexed_Compoonent):
Activate_Atomic_Synchronization (Expand_N_Selected_Component):
Use Activate_Atomic_Synchronization
* exp_util.ads, exp_util.adb (Activate_Atomic_Synchronization): New
procedure.
* sinfo.ads, sinfo.adb (Atomic_Sync_Required): Can now apply to
N_Selected_Component node
Arnaud Charlet [Fri, 4 Nov 2011 13:45:01 +0000 (14:45 +0100)]
[multiple changes]
2011-11-04 Yannick Moy <moy@adacore.com>
* atree.adb, atree.ads (Set_Original_Node): New set procedure.
* sem_ch13.adb (Analyze_Aspect_Specifications/Pre_Post_Aspects):
In ASIS mode, no splitting of aspects between conjuncts.
(Analyze_Aspect_Specifications/Aspect_Test_Case): Make pragma
expressions refer to the original aspect expressions through
the Original_Node link. This is used in semantic analysis for
ASIS mode, so that the original expression also gets analyzed.
* sem_prag.adb (Preanalyze_TC_Args,
Check_Precondition_Postcondition,
Analyze_Pragma/Pragma_Test_Case): In ASIS mode, for a pragma
generated from a source aspect, also analyze the original aspect
expression.
(Check_Expr_Is_Static_Expression): New procedure
similar to existing procedure Check_Arg_Is_Static_Expression,
except called on expression inside pragma.
2011-11-04 Tristan Gingold <gingold@adacore.com>
* prj-env.adb, prj-env.ads (Find_Name_In_Path): New function, from
Find_Project.Try_Path_Name.
(Find_Project): Use Find_Name_In_Path to implement Try_Path_Name.
* projects.texi: Add short description for qualifiers aggregate
and aggregate library.
2011-11-04 Matthew Heaney <heaney@adacore.com>
* Makefile.rtl, impunit.adb: Added a-cogeso.ad[sb]
* a-cgaaso.adb: Replaced implementation with instantiation
of Generic_Sort.
* a-cogeso.ad[sb] This is the new Ada 2012 unit
Ada.Containers.Generic_Sort
Arnaud Charlet [Fri, 4 Nov 2011 12:04:55 +0000 (13:04 +0100)]
[multiple changes]
2011-11-04 Robert Dewar <dewar@adacore.com>
* exp_ch2.adb (Expand_Entity_Reference): Do not set
Atomic_Sync_Required for the case of a prefix of an attribute.
* exp_ch4.adb (Expand_N_Explicit_Dereference): May require
atomic synchronization
(Expand_N_Indexed_Component): Ditto.
(Expand_B_Selected_Component): Ditto.
* sem_prag.adb (Process_Suppress_Unsuppress):
Disable/Enable_Atomic_Synchronization can now occur for array
types with pragma Atomic_Components.
* sinfo.ads, sinfo.adb (Atomic_Sync_Required): Can now occur on
N_Explicit_Dereference nodes and on N_Indexed_Component nodes.
2011-11-04 Gary Dismukes <dismukes@adacore.com>
* gnat_ugn.texi: Editorial corrections for gnattest section.
Robert Dewar [Fri, 4 Nov 2011 11:00:33 +0000 (11:00 +0000)]
sem_prag.adb: Minor reformatting.
2011-11-04 Robert Dewar <dewar@adacore.com>
* sem_prag.adb: Minor reformatting.
* gnat_rm.texi: Update documentation for pragma Warnings (Off,
"***") usage.
* exp_ch2.adb (Expand_Entity_Reference): Only set
Atomic_Sync_Required on entities that are variables. Doesn't
make any sense on anything else.
Robert Dewar [Fri, 4 Nov 2011 10:58:44 +0000 (10:58 +0000)]
exp_ch2.adb (Expand_Entity_Reference): Extend handling of atomic sync to type case.
2011-11-04 Robert Dewar <dewar@adacore.com>
* exp_ch2.adb (Expand_Entity_Reference): Extend handling of
atomic sync to type case.
* sem_prag.adb (Process_Suppress_Unsuppress): Atomic Sync can
apply to types.
Jakub Jelinek [Fri, 4 Nov 2011 10:39:25 +0000 (11:39 +0100)]
i386.c (ix86_expand_vector_convert_uns_vsivsf): New function.
* config/i386/i386.c (ix86_expand_vector_convert_uns_vsivsf): New
function.
* config/i386/i386-protos.h (ix86_expand_vector_convert_uns_vsivsf):
New prototype.
* config/i386/sse.md (floatuns<sseintvecmodelower><mode>2): Use it.
For floatunsv8siv8sf2 require TARGET_AVX2.
* gcc.dg/torture/vec-cvt-1.c: Enable commented out inttoflttestui
test.
* config/i386/i386.c (ix86_expand_adjust_ufix_to_sfix_si): Add
XORP argument. Subtract 0x1p31 instead of 0x1p32. Use normal
signalling comparison instead of non-signalling. Store into
*XORP pseudo holding 0x80000000 integers if 0x1p31 has been
subtracted and 0 otherwise.
* config/i386/i386-protos.h (ix86_expand_adjust_ufix_to_sfix_si):
Adjust prototype.
* config/i386/sse.md (fixuns_trunc<mode><sseintvecmodelower>2): Enable
already for TARGET_SSE2. Xor in vector initialized by
ix86_expand_adjust_ufix_to_sfix_si at the end.
(vec_pack_ufix_trunc_<mode>): Likewise.
Jakub Jelinek [Fri, 4 Nov 2011 10:35:44 +0000 (11:35 +0100)]
tree-vect-stmts.c (vectorizable_conversion): Rewritten to handle not just FLOAT_EXPR and FIX_TRUNC_EXPR...
* tree-vect-stmts.c (vectorizable_conversion): Rewritten to handle
not just FLOAT_EXPR and FIX_TRUNC_EXPR, but also CONVERT_EXPR_CODE_P,
WIDEN_MULT_EXPR and WIDEN_LSHIFT_EXPR to handle what
vectorizable_type_demotion and vectorizable_type_promotion did.
Additionally handle FLOAT_EXPR and FIX_TRUNC_EXPR where the integer
is {,un}signed {char,short}.
(vect_create_vectorized_demotion_stmts): Fix comment typo. For
recursive calls unconditionally use VEC_PACK_TRUNC_EXPR.
Push vec_dest back to the vec_dsts vector at the end.
(vect_create_vectorized_promotion_stmts): Don't recurse, do just
one step. Removed multi_step_cvt, vec_dsts, slp_node and
prev_stmt_info arguments, add vec_dest argument. Push always
into vec_tmp, not just when multi_step_cvt != 0, replace *vec_oprdn0
with vec_tmp at the end after freeing old *vec_oprnd0 vector.
(vectorizable_type_demotion, vectorizable_type_promotion): Removed.
(vect_analyze_stmt): Don't call vectorizable_type_demotion and
vectorizable_type_promotion. Call vectorizable_conversion even
for SLP bb vectorization.
(vect_transform_stmt): Call vectorizable_conversion instead of
vectorizable_type_demotion and vectorizable_type_promotion.
(supportable_widening_operation): Clear *multi_step_cvt first,
simplify c1/c2 computation, free *interm_types vector on failure.
(supportable_narrowing_operation): Clear *multi_step_cvt first,
free *interm_types vector on failure, handle multi-step
FIX_TRUNC_EXPR.