* config/i386/i386.md (mov<mode>cc): Macroize expander from movsfcc,
movdfcc and movxfcc using X87MODEF mode iterator and SSE_FLOAT_MODE_P
as insn constraint.
* config/i386/sse.md (<sse>_movup<ssemodesuffixf2c>): Macroize insn
from sse_movups adn sse2_movupd using SSEMODEF2P mode iterator and
SSE_VEC_FLOAT_MODE_P as insn constraint.
(<sse>_movmskp<ssemodesuffixf2c>): Ditto from similar patterns.
(sse4a_movnt<mode>): Macroize insn from sse4a_movntsf and
sse4a_movntdf using MODEF mode iterator.
(sse4a_vmmovnt<mode>): Macroize insn form sse4a_vmmovntv2df and
sse4a_vmmovntv4sf using SSEMODEF2P mode iterator.
(sse4_1_blendp<ssemodesuffixf2c>): Ditto from similar patterns.
(sse4_1_blendvp<ssemodesuffixf2c>): Ditto.
(sse4_1_dpp<ssemodesuffixf2c>): Ditto.
(sse4_1_roundp<ssemodesuffixf2c>): Ditto.
(sse4_1_rounds<ssemodesuffixf2c>): Ditto.
From-SVN: r132518
2008-02-20 Jerry DeLisle <jvdelisle@gcc.gnu.org>
PR libfortran/35132
* io/transfer.c (next_record_w): Truncate after the last record for
STREAM I/O.
PR libfortran/34954
* io/transfer.c (data_transfer_init): Initialize dtp->rec if writing.
PR libfortran/34974
* io/transfer.c (formatted_transfer_scalar): Flush the buffer if skips
is less than zero. (next_record_w): Use sseek to position the file to
the max position reached.
From-SVN: r132512
2008-02-20 Jerry DeLisle <jvdelisle@gcc.gnu.org>
PR libfortran/35036
* write_float.def (output_float): Add error checks for zero digits
after decimal point in E and D format specifiers.
From-SVN: r132510
* config/h8300/h8300.md (insv): Force source operand to be a register.
* config/h8300/h8300.c (h8300_expand_epilogue): Emit return insn
as a jump, not as a plain insn.
From-SVN: r132498
2008-02-20 Richard Guenther <rguenther@suse.de>
PR middle-end/35265
* builtins.c (validate_arg): If we want an INTEGER_TYPE,
be happy with INTEGRAL_TYPE_P.
* gcc.dg/builtins-66.c: New testcase.
From-SVN: r132485
2008-02-20 Richard Guenther <rguenther@suse.de>
* fold-const.c (split_tree): Associate floatig-point expressions
if flag_associative_math is set.
* gcc.dg/fold-reassoc-1.c: New testcase.
* gcc.dg/tree-ssa/recip-3.c: Adjust to not compute d/d.
From-SVN: r132481
2008-02-20 Richard Guenther <rguenther@suse.de>
* tree.h (fold_real_zero_addition_p): Declare.
* fold-const.c (fold_real_zero_addition_p): Export.
* tree-ssa-reassoc.c (eliminate_using_constants): Also handle
floating-point operations with zero and one.
* gcc.dg/tree-ssa/reassoc-13.c: New testcase.
From-SVN: r132480
2008-02-20 Paolo Bonzini <bonzini@gnu.org>
PR bootstrap/32009
PR bootstrap/32161
* configure.ac (CFLAGS_FOR_TARGET, CXXFLAGS_FOR_TARGET): Compute here.
* configure: Regenerate.
* Makefile.def: Define stage_libcflags for all bootstrap stages.
* Makefile.tpl (STAGE1_LIBCFLAGS, STAGE2_LIBCFLAGS, STAGE3_LIBCFLAGS,
STAGE4_LIBCFLAGS): New.
(CFLAGS_FOR_TARGET, CXXFLAGS_FOR_TARGET): Subst from autoconf, without
$(SYSROOT_CFLAGS_FOR_TARGET) and $(DEBUG_PREFIX_CFLAGS_FOR_TARGET).
(BASE_TARGET_EXPORTS): Append them here to C{,XX}FLAGS.
(EXTRA_TARGET_FLAGS): Append them here to {LIB,}C{,XX}FLAGS.
(configure-stage[+id+]-[+prefix+][+module+]): Pass stage_libcflags
for target modules. Don't export LIBCFLAGS.
(all-stage[+id+]-[+prefix+][+module+]): Pass stage_libcflags; pass
$(BASE_FLAGS_TO_PASS) where [+args+] was passed, and [+args+] after
the overridden CFLAGS_FOR_TARGET and CXXFLAGS_FOR_TARGET.
(invocations of `all'): Replace $(TARGET_FLAGS_TO_PASS) with
$(EXTRA_TARGET_FLAGS), $(FLAGS_TO_PASS) with $(EXTRA_HOST_FLAGS).
* Makefile.in: Regenerate.
2008-02-20 Paolo Bonzini <bonzini@gnu.org>
PR bootstrap/32009
* mh-ppc-darwin (BOOT_CFLAGS): Reenable.
2008-02-20 Paolo Bonzini <bonzini@gnu.org>
* doc/install.texi: Correct references to CFLAGS, replacing them
with BOOT_CFLAGS. Document flags used during bootstrap for
target libraries.
From-SVN: r132479
* config/i386/i386.h (SSE_VEC_FLOAT_MODE_P): New define.
* config/i386/i386.md (*sse_setcc<mode>): Macroize from *sse_setccsf
and *sse_setccdf using MODEF mode iterator and SSE_FLOAT_MODE_P as
insn constraint.
(smin<mode>3): Ditto from similar patterns.
(smax<mode>3): Ditto.
(*ieee_smin<mode>3): Ditto.
(*ieee_smax<mode>3): Ditto.
* config/i386/sse.md (sse): New mode attribute.
(mov<mode>): Macroize expander from movv4sf and movv2df using
SSEMODEF2P mode iterator.
(<sse>_movnt<mode>): Ditto from similar patterns. Use
SSE_VEC_FLOAT_MODE_P as insn constraint.
(storent<mode>): Ditto.
(storent<mode>): Macroize expander from storentsf and storentdf using
MODEF mode iterator.
(neg<mode>2): Macroize from negv4sf2 and negv2df2 using SSEMODEF2P
mode iterator and SSE_VEC_FLOAT_MODE_P as insn constraint.
(abs<mode>2): Ditto from similar patterns.
(add<mode>3, *add<mode>3, <sse>_vmadd<mode>3): Ditto.
(sub<mode>3, *sub<mode>3, <sse>_vmsub<mode>3): Ditto.
(<sse>_div<mode>3, <sse>_vmdiv<mode>3): Ditto.
(<sse>_vmsqrt<mode>2): Ditto.
(smin<mode>3, *smin<mode>3_finite, *smin<mode>3)
(<sse>_vmsmin<mode>3, *ieee_smin<mode>3): Ditto.
(smax<mode>3, *smax<mode>3_finite, *smax<mode>3)
(<sse>_vmsmax<mode>3, *ieee_smax<mode>3): Ditto.
(<sse>_maskcmp<mode>3): Macroize from sse_maskcmpv4sf3,
sse_maskcmpsf3, sse2_maskcmpv2df3 and sse2_maskcmpdf3 using SSEMODEF4
mode iterator. Use SSE_FLOAT_MODE_P with SSE_VEC_FLOAT_MODE_P as
insn constraint.
(<sse>_comi): Macroize from sse_comi and sse2_comi using MODEF mode
iterator and SSE_FLOAT_MODE_P as insn constraint.
(<sse>_ucomi): Ditto from similar patterns.
(<sse>_vmmaskcmp<mode>3): Macroize from sse_vmmaskcmpv4sf3 and
sse2_vmmaskcmpv2df3 using SSEMODEF2P mode iterator and
SSE_VEC_FLOAT_MODE_P as insn constraint.
(vcond<mode>): Ditto from similar patterns.
(and<mode>3, *and<mode>3): Ditto.
(<sse>_nand<mode>3): Ditto.
(ior<mode>3, *ior<mode>3): Ditto.
(xor<mode>3, *xor<mode>3): Ditto.
(*and<mode>3): Macroize from *andsf3 and *anddf3 using MODEF mode
iterator and SSE_FLOAT_MODE_P as insn constraint.
(*nand<mode>3): Ditto from similar patterns.
(*ior<mode>3): Ditto.
(*xor<mode>3): Ditto.
From-SVN: r132478
* g++.dg/compat/struct-layout-1_generate.c (DG_OPTIONS): New define.
(switchfiles): Use DG_OPTIONS. Explicitly name arguments in fprintf
templates to reduce number of passed arguments.
From-SVN: r132471
2007-02-19 Andreas Tobler <a.tobler@schweiz.org>
* dyn_load.c (GC_dyld_image_add): Remove ifdef clause and use the macro
GC_GETSECTBYNAME instead.
* include/private/gc_priv.h: Define GC_GETSECTBYNAME according to the
architecture (Darwin).
From-SVN: r132452
PR bootstrap/35218
* Makefile.in (build_file_translate): New.
(gcc-vers.texi): Use it for translating $(abs_srcdir).
* config.build (build_file_translate): Set to `CMD //c' on MinGW.
* configure.ac (build_file_translate): Substitute it.
* configure: Regenerate.
From-SVN: r132444
2008-02-19 Daniel Franke <franke.daniel@gmail.com>
PR fortran/35030
* expr.c (gfc_check_pointer_assign): Add type and kind information
to type-mismatch message.
(gfc_check_assign): Unify error messages.
From-SVN: r132442
2008-02-19 H.J. Lu <hongjiu.lu@intel.com>
PR Ada/35186
* config/i386/i386-modes.def: Revert the last DI alignment
change until Ada people can look into it.
From-SVN: r132433
2008-02-19 Richard Guenther <rguenther@suse.de>
PR tree-optimization/34989
* tree-ssa-forwprop.c (forward_propagate_addr_expr_1): Re-structure.
Allow propagation to INDIRECT_REF if we can simplify only.
* gcc.dg/pr34989-1.c: New testcase.
* gcc.dg/pr34989-2.c: Likewise.
From-SVN: r132429
PR other/31349
* opts.c (undocumented_msg): Leave blank unless checking is
enabled.
(handle_options): Fix indentation.
(print_filtered_help): If no language-specific options were
displayed tell the user how to list all the options supported by
the language's front-end.
(print_specific_help): Fix indentation and remove duplicate line.
(common_handle_option): Handle the -v option.
For --help enable the display of undocumented options if the -v
switch has been included on the command line.
For --help= check for overlaps in the arguments between the option
classes and the language names and issue a warning when they
cannot be disambiguated.
* c.opt (v): Pass on to the common option handler.
From-SVN: r132427
PR c++/34964
PR c++/35244
* semantics.c (finish_omp_threadprivate): Do nothing for error_operand_p
vars. Afterwards ensure v is VAR_DECL.
* gcc.dg/gomp/pr34964.c: New test.
* g++.dg/gomp/pr34964.C: New test.
* gcc.dg/gomp/pr35244.c: New test.
* g++.dg/gomp/pr35244.C: New test.
From-SVN: r132425
PR c++/35078
* parser.c (cp_parser_omp_for_loop): If DECL has REFERENCE_TYPE, don't
call cp_finish_decl.
* semantics.c (finish_omp_for): Fail if DECL doesn't have integral type
early.
* g++.dg/gomp/pr35078.C: New test.
From-SVN: r132424
