Jakub Jelinek [Thu, 27 Dec 2007 08:31:54 +0000 (09:31 +0100)]
re PR target/34281 (ARM: varargs broken with 64bit parameters.)
PR target/34281
* config/arm/arm.c (arm_setup_incoming_varargs): If last named
argument needs double word alignment and cum->nregs is odd, account
for the inserted padding.
Paolo Carlini [Tue, 25 Dec 2007 13:49:54 +0000 (13:49 +0000)]
hashtable_policy.h (__lower_bound): Add.
2007-12-25 Paolo Carlini <pcarlini@suse.de>
* include/tr1_impl/hashtable_policy.h (__lower_bound): Add.
(_Prime_rehash_policy::_M_next_bkt, _M_bkt_for_elements,
_M_need_rehash): Use __lower_bound.
* include/std/unordered_map: Do not include the whole <algorithm>,
include <bits/stl_algobase.h>.
* include/std/unordered_set: Likewise.
* include/tr1/unordered_map: Likewise.
* include/tr1/unordered_set: Likewise.
Daniel Franke [Tue, 25 Dec 2007 10:41:44 +0000 (05:41 -0500)]
re PR fortran/34533 (DTIME returns total process time and not since last invocation)
gcc/fortran:
2007-12-25 Daniel Franke <franke.daniel@gmail.com>
PR fortran/34533
* intrinsic.h (gfc_check_etime): Renamed to ...
(gfc_check_dtime_etime): ... this.
(gfc_check_etime_sub): Renamed to ...
(gfc_check_dtime_etime_sub): ... this.
(gfc_resolve_dtime_sub): New prototype.
* check.c (gfc_check_etime): Renamed to ...
(gfc_check_dtime_etime): ... this.
(gfc_check_etime_sub): Renamed to ...
(gfc_check_dtime_etime_sub): ... this.
* iresolve.c (gfc_resolve_dtime_sub): New implementation.
* intrinsic.c (add_functions): Removed alias from ETIME to DTIME,
added stand-alone intrinsic DTIME.
(add_subroutines): Adjusted check and resolve function names for
DTIME and ETIME.
* trans-intrinsic.c (gfc_conv_intrinsic_function): Added DTIME
to known functions in switch.
* intrinsic.texi (DTIME): Added paragraph about thread-safety,
fixed return value section.
(CPU_TIME): Clarified intent and added implementation notes.
libgfortran:
2007-12-25 Daniel Franke <franke.daniel@gmail.com>
PR fortran/34533
* intrinsics/cpu_time.c: Moved code commonly usable for CPU_TIME,
DTIME and ETIME to ...
* intrinsics/time_1.h: ... here.
* intrinsics/dtime.c: New file.
* intrinsics/etime.c: Newly implemented using the common
time-aquisition function from time_1.h.
* gfortran.map (_gfortran_dtime, _gfortran_dtime_sub): New.
* Makefile.am: Added new file.
* Makefile.in: Regenerated.
* configure: Regenerated.
Geoffrey Keating [Sun, 23 Dec 2007 00:57:19 +0000 (00:57 +0000)]
widget.html: Mark as HTML 4.01, which it now is, rather than XHTML, which it never was.
* GCC_Regression_Tester.wdgt/widget.html: Mark as HTML 4.01,
which it now is, rather than XHTML, which it never was.
(<head>): Mark as UTF-8. Add title. Move CSS to top of
document, specify type. Mark script as 'defer'.
(updateContents): If loaded from HTTP, look for status in same
place as widget.
(gotContents): Use DOM methods to change text rather than innerHTML.
(<body>): Eliminate unnecessary DIV element.
* GCC_Regression_Tester.wdgt/Info.plist: Update version, copyright
notice.
Eric Botcazou [Sat, 22 Dec 2007 23:05:57 +0000 (23:05 +0000)]
trans.c (call_to_gnu): Make the temporary for non-addressable In parameters passed by reference.
* trans.c (call_to_gnu): Make the temporary for non-addressable
In parameters passed by reference.
(addressable_p): Return true for STRING_CST and CALL_EXPR.
Jerry DeLisle [Sat, 22 Dec 2007 01:57:07 +0000 (01:57 +0000)]
re PR libfortran/34540 (cshift, eoshift, kind=1 and kind=2 arguments...)
2007-12-21 Jerry DeLisle <jvdelisle@gcc.gnu.org>
PR fortran/34540
* iresolve.c (gfc_resolve_cshift): Take optional dim path
only if the argument is an optional itself.
* iresolve.c (gfc_resolve_eoshift): Same.
Paul Thomas [Fri, 21 Dec 2007 21:20:38 +0000 (21:20 +0000)]
re PR fortran/34438 (gfortran not compliant w.r.t default initialization of derived type component and implicit SAVE attribute)
2007-12-21 Paul Thomas <pault@gcc.gnu.org>
PR fortran/34438
* trans-decl.c (gfc_finish_var_decl): Do not mark derived types
with default initializers as TREE_STATIC unless they are in the
main program scope.
(gfc_get_symbol_decl): Pass derived types with a default
initializer to gfc_defer_symbol_init.
(init_default_dt): Apply default initializer to a derived type.
(init_intent_out_dt): Call init_default_dt.
(gfc_trans_deferred_vars): Ditto.
* module.c (read_module): Check sym->module is there before
using it in a string comparison.
2007-12-21 Paul Thomas <pault@gcc.gnu.org>
PR fortran/34438
* gfortran.dg/default_initialization_3.f90: New test.
mips.c (mips_emit_loadgp): Replace gen_* calls with separate gen_*_si and gen_*_di calls.
gcc/
* config/mips/mips.c (mips_emit_loadgp): Replace gen_* calls with
separate gen_*_si and gen_*_di calls. Pass pic_offset_table_rtx
as the first argument.
* config/mips/mips.md (loadgp_newabi, loadgp_absolute)
(loadgp_rtp): Rename to...
(loadgp_newabi_<mode>, loadgp_absolute<mode>, loadgp_rtp<mode>):
...these. Add modes to all operands. Add the target register
as an operand. Combine loadgp_rtp<mode> with its splitter.
Bob Wilson [Thu, 20 Dec 2007 22:35:59 +0000 (22:35 +0000)]
xtensa.md (fix_return_addr): Remove.
* config/xtensa/xtensa.md (fix_return_addr): Remove.
* config/xtensa/xtensa-protos.h (xtensa_initialize_trampoline): New.
(xtensa_trampoline_template): New.
* config/xtensa/xtensa.c (MIN_FRAME_SIZE): Moved here from xtensa.h.
(xtensa_return_addr): Expand to standard Xtensa insns instead of
fix_return_addr. Get high bits from a local label.
(xtensa_trampoline_template): New function with code moved from
TRAMPOLINE_TEMPLATE in xtensa.h. Use L32R instead of CALL0 except
when using CONST16 or absolute-mode literals.
(xtensa_initialize_trampoline): New function with code moved from
INITIALIZE_TRAMPOLINE in xtensa.h. Use different offsets depending
on which trampoline version is used.
* config/xtensa/lib2funcs.S (TRAMPOLINE_SIZE): Add comment.
* config/xtensa/xtensa.h (TARGET_ABSOLUTE_LITERALS): Define.
(MIN_FRAME_SIZE): Moved to xtensa.c.
(TRAMPOLINE_TEMPLATE): Use xtensa_trampoline_template.
(TRAMPOLINE_SIZE): Two versions of the trampoline have different sizes.
(INITIALIZE_TRAMPOLINE): Use xtensa_initialize_trampoline.
* config/xtensa/ieee754-df.S (XCHAL_NO_MUL): Define.
(__muldf3): Use CALL12 instead of CALL0 to invoke .Lmul_mulsi3
helper when not using the CALL0 ABI. Change .Lmul_mulsi3 to match.
* config/xtensa/lib1funcs.asm (__umulsidi3): Likewise.
* config/xtensa/ieee754-sf.S (__mulsf3): Likewise.
Jakub Jelinek [Thu, 20 Dec 2007 14:40:33 +0000 (15:40 +0100)]
re PR c++/34459 (incorrect code when compiled with optimization (-O1))
PR c++/34459
* tree-ssa-dse.c (dse_optimize_stmt): Don't eliminate store if
USE_STMT not only stores into the same object as STMT, but might
read it too.
Robert Dewar [Wed, 19 Dec 2007 16:26:20 +0000 (17:26 +0100)]
gnat_rm.texi, [...]: Update documentation of -gnatw.o Fix name of Wide_Wide_Latin_1/9 file names Add...
2007-12-19 Robert Dewar <dewar@adacore.com>
* gnat_rm.texi, gnat_ugn.texi: Update documentation of -gnatw.o
Fix name of Wide_Wide_Latin_1/9 file names
Add documentation for Ada.Exceptions.Last_Chance_Handler (a-elchha.ads)
Add missing documentation for Ada.Wide_[Wide_]Characters.Unicode
Add missing documentation for Ada.Command_Line.Response_File
Update list of warning letters for Warnings pragma
Add documentation for pragma Unmodified
Gary Dismukes [Wed, 19 Dec 2007 16:25:18 +0000 (17:25 +0100)]
re PR ada/34149 (GNAT crash - deeply inrerited function)
2007-12-19 Gary Dismukes <dismukes@adacore.com>
PR ada/34149
* sem_disp.adb (Check_Dispatching_Call): Augment existing test for
presence of a statically tagged operand (Present (Static_Tag)) with
test for Indeterm_Ancestor_Call when determining whether to propagate
the static tag to tag-indeterminate operands (which forces dispatching
on such calls).
(Check_Controlling_Formals): Ada2005, access parameters can have
defaults.
(Add_Dispatching_Operation, Check_Operation_From_Private_View): do
not insert subprogram in list of primitive operations if already there.
Ed Schonberg [Wed, 19 Dec 2007 16:24:55 +0000 (17:24 +0100)]
sem_ch8.adb (Analyze_Subprogram_Renaming): Diagnose illegal renamings whose renamed entity is a subprogram that requires...
2007-12-19 Ed Schonberg <schonberg@adacore.com>
Gary Dismukes <dismukes@adacore.com>
* sem_ch8.adb (Analyze_Subprogram_Renaming): Diagnose illegal renamings
whose renamed entity is a subprogram that requires overriding.
(Premature_Usage): Test for the case of N_Full_Type_Declaration when
issuing an error for premature usage and issue a message that says
'type' rather than 'object'.
Bob Duff [Wed, 19 Dec 2007 16:24:44 +0000 (17:24 +0100)]
sem_ch4.adb (Analyze_Concatenation_Rest): New procedure.
2007-12-19 Bob Duff <duff@adacore.com>
* sem_ch4.adb (Analyze_Concatenation_Rest): New procedure.
(Analyze_Concatenation): Use iteration instead of recursion in order
to avoid running out of stack space for deeply nested concatenations.
Ed Schonberg [Wed, 19 Dec 2007 16:24:34 +0000 (17:24 +0100)]
PR ada/15803, ada/15805
2007-12-19 Ed Schonberg <schonberg@adacore.com>
Gary Dismukes <dismukes@adacore.com>
PR ada/15803, ada/15805
* sem_ch6.adb, sem_ch3.adb (Constrain_Access): In Ada2005, diagnose
illegal access subtypes when there is a constrained partial view.
(Check_For_Premature_Usage): New procedure inside
Access_Subprogram_Declaration for checking that an access-to-subprogram
type doesn't reference its own name within any formal parameters or
result type (including within nested anonymous access types).
(Access_Subprogram_Declaration): Add call to Check_For_Premature_Usage.
(Sem_Ch3.Analyze_Object_Declaration, Sem_ch6.Process_Formals): if the
context is an access_to_variable, the expression cannot be an
access_to_constant.
Ed Schonberg [Wed, 19 Dec 2007 16:24:06 +0000 (17:24 +0100)]
par-ch3.adb (P_Record_Declaration): Guard against cascaded errors in mangled declaration
2007-12-19 Ed Schonberg <schonberg@adacore.com>
* par-ch3.adb (P_Record_Declaration): Guard against cascaded errors in
mangled declaration
(P_Type_Declaration): Diagnose misuse of "abstract" in untagged record
declarations.
(P_Variant_Part): Cleaner patch for parenthesized discriminant
Ed Schonberg [Wed, 19 Dec 2007 16:23:55 +0000 (17:23 +0100)]
freeze.adb (Freeze_Record_Type, [...]): Implement properly the Ada2005 rules concerning when...
2007-12-19 Ed Schonberg <schonberg@adacore.com>
* freeze.adb (Freeze_Record_Type, Check_Current_Instance): Implement
properly the Ada2005 rules concerning when the current instance of a
record type is aliased.
Robert Dewar [Wed, 19 Dec 2007 16:23:32 +0000 (17:23 +0100)]
exp_ch9.adb (Null_Statements): Moved to library level
2007-12-19 Robert Dewar <dewar@adacore.com>
* exp_ch9.adb (Null_Statements): Moved to library level
(Trivial_Accept_OK): New function
(Expand_Accept_Declaration): Use Trivial_Accept_OK
(Expand_N_Accept_Statement): Use Trivial_Accept_OK
Robert Dewar [Wed, 19 Dec 2007 16:23:09 +0000 (17:23 +0100)]
exp_attr.adb (Expand_N_Attribute_Reference, case Size): Fix error in handling compile time known size of record or array (case of front...
2007-12-19 Robert Dewar <dewar@adacore.com>
* exp_attr.adb (Expand_N_Attribute_Reference, case Size): Fix error in
handling compile time known size of record or array (case of front end
layout active, e.g. in GNAAMP).
Ed Schonberg [Wed, 19 Dec 2007 16:22:56 +0000 (17:22 +0100)]
exp_aggr.adb (Not_OK_For_Backend): A component of a private type with discriminants forces expansion of the...
2007-12-19 Ed Schonberg <schonberg@adacore.com>
* exp_aggr.adb (Not_OK_For_Backend): A component of a private type with
discriminants forces expansion of the aggregate into assignments.
(Init_Record_Controller): If the type of the aggregate is untagged and
is not inherently limited, the record controller is not limited either.
Bob Duff [Wed, 19 Dec 2007 16:22:40 +0000 (17:22 +0100)]
atree.ads, atree.adb (Traverse_Func): Walk Field2 last, and eliminate the resulting tail recursion by hand.
2007-12-19 Bob Duff <duff@adacore.com>
* atree.ads, atree.adb (Traverse_Func): Walk Field2 last, and eliminate
the resulting tail recursion by hand. This prevents running out of
memory on deeply nested concatenations, since Field2 is where the left
operand of concatenations is stored.
Fix bug (was returning OK_Orig in some cases). Fix return subtype to
clarify that it can only return OK or Abandon.
* sem_res.adb (Resolve_Op_Concat): Replace the recursion on the left
operand by iteration, in order to avoid running out of memory on
deeply-nested concatenations. Use the Parent pointer to get back up the
tree.
(Resolve_Op_Concat_Arg, Resolve_Op_Concat_First,
Resolve_Op_Concat_Rest): New procedures split out of
Resolve_Op_Concat, so the iterative algorithm in Resolve_Op_Concat is
clearer.
* checks.adb (Remove_Checks): Use Traverse_Proc instead of
Traverse_Func, because the former already takes care of discarding the
result.
* errout.adb (First_Node): Use Traverse_Proc instead of Traverse_Func,
because the former already takes care of discarding the result.
(Remove_Warning_Messages): Use appropriate subtype for Status and
Discard
projects / thirdparty / gcc.git / log
