Bill Schmidt [Sun, 31 Jul 2011 18:58:06 +0000 (18:58 +0000)]
re PR tree-optimization/49749 (Reassociation rank algorithm does not include all non-NULL operands)
2011-07-29 Bill Schmidt <wschmidt@linux.vnet.ibm.com>
PR tree-optimization/49749
* tree-ssa-reassoc.c (get_rank): New forward declaration.
(PHI_LOOP_BIAS): New macro.
(phi_rank): New function.
(loop_carried_phi): Likewise.
(propagate_rank): Likewise.
(get_rank): Add calls to phi_rank and propagate_rank.
re PR target/49920 (unable to find a register to spill in class ‘DIREG’)
PR target/49920
* config/i386/i386.md (strset): Do not expand strset_singleop
when %eax or $edi are fixed.
(*strsetdi_rex_1): Disable when %eax or %edi are fixed.
(*strsetsi_1): Ditto.
(*strsethi_1): Ditto.
(*strsetqi_1): Ditto.
(*rep_stosdi_rex64): Disable when %eax, %ecx or %edi are fixed.
(*rep_stossi): Ditto.
(*rep_stosqi): Ditto.
(cmpstrnsi): Also fail when %ecx is fixed.
(*cmpstrnqi_nz_1): Disable when %ecx, %esi or %edi are fixed.
(*cmpstrnqi_1): Ditto.
(*strlenqi_1): Ditto.
(*strmovdi_rex_1): Disable when %esi or %edi are fixed.
(*strmovsi_1): Ditto.
(*strmovhi_1): Ditto.
(*strmovqi_1): Ditto.
(*rep_movdi_rex64): Disable when %ecx, %esi or %edi are fixed.
(*rep_movsi): Ditto.
(*rep_movqi): Ditto.
testsuite/ChangeLog:
PR target/49920
* gcc.target/i386/pr49920.c: New test.
Janus Weil [Sun, 31 Jul 2011 10:25:07 +0000 (12:25 +0200)]
re PR fortran/49112 ([OOP] Missing type-bound procedure, "duplicate save" warnings and internal compiler error)
2011-07-31 Janus Weil <janus@gcc.gnu.org>
PR fortran/49112
* resolve.c (resolve_structure_cons): Don't do the full dt resolution,
only call 'resolve_fl_derived0'.
(resolve_typebound_procedures): Resolve typebound procedures of
parent type.
(resolve_fl_derived0): New function, which does a part of the work
for 'resolve_fl_derived'.
(resolve_fl_derived): Call 'resolve_fl_derived0' and do some additional
things.
Benjamin Kosnik [Fri, 29 Jul 2011 22:31:30 +0000 (22:31 +0000)]
Docbook conversion of existing ext/pb_ds documentation.
2011-07-29 Benjamin Kosnik <bkoz@redhat.com>
Docbook conversion of existing ext/pb_ds documentation.
* doc/Makefile.am (xml_sources_manual): Add
policy_data_structures.xml and
test_policy_data_structures.xml.
(stamp-html-copy): Remove special-case for ext/pb_ds directory.
(XSLTPROC_FLAGS): Split into XSLT_FLAGS and XSLT_PARAM, use.
* doc/Makefile.in: Regenerate.
* doc/xml/manual/policy_data_structures.xml: New, adapted from
previous html-only instance in doc/html/ext/pb_ds.
* doc/xml/manual/test_policy_data_structures.xml: New, same as above.
driver-native.c [__sgi__]: Include <invent.h>, <sys/sbd.h>.
* config/mips/driver-native.c [__sgi__]: Include <invent.h>,
<sys/sbd.h>.
(cpu_types): New array.
(cputype): New function.
(host_detect_local_cpu): Only define buf, f if !__sgi__.
Use scaninvent instead of /proc/cpuinfo if __sgi__.
* config.host: Also use driver-native.o, mips/x-native on
mips-sgi-irix*.
* config/mips/iris6.h [__mips__] (host_detect_local_cpu):
Declare.
(EXTRA_SPEC_FUNCTIONS, MARCH_MTUNE_NATIVE_SPECS): Define.
(DRIVER_SELF_SPECS): Add MARCH_MTUNE_NATIVE_SPECS.
Jakub Jelinek [Fri, 29 Jul 2011 17:45:42 +0000 (19:45 +0200)]
re PR middle-end/49897 (nesting lastprivate gives incorrect result)
PR middle-end/49897
PR middle-end/49898
* omp-low.c (use_pointer_for_field): If disallowing copy-in/out
in nested parallel and outer is a gimple_reg, mark it as addressable
and set its bit in task_shared_vars bitmap too.
* testsuite/libgomp.c/pr49897-1.c: New test.
* testsuite/libgomp.c/pr49897-2.c: New test.
* testsuite/libgomp.c/pr49898-1.c: New test.
* testsuite/libgomp.c/pr49898-2.c: New test.
re PR target/49313 (Inefficient libgcc implementations for avr)
PR target/49313
* config/avr/libgcc.S (__ffshi2): Don't skip 2-word instruction.
(__ctzsi2): Result for 0 may be undefined.
(__ctzhi2): Result for 0 may be undefined.
(__popcounthi2): Don't clobber r30. Use __popcounthi2_tail.
(__popcountsi2): Ditto. And don't clobber r26.
(__popcountdi2): Ditto. And don't clobber r27.
* config/avr/avr.md (UNSPEC_COPYSIGN): New c_enum.
(parityhi2): New expand.
(paritysi2): New expand.
(popcounthi2): New expand.
(popcountsi2): New expand.
(clzhi2): New expand.
(clzsi2): New expand.
(ctzhi2): New expand.
(ctzsi2): New expand.
(ffshi2): New expand.
(ffssi2): New expand.
(copysignsf3): New insn.
(bswapsi2): New expand.
(*parityhi2.libgcc): New insn.
(*parityqihi2.libgcc): New insn.
(*paritysihi2.libgcc): New insn.
(*popcounthi2.libgcc): New insn.
(*popcountsi2.libgcc): New insn.
(*popcountqi2.libgcc): New insn.
(*popcountqihi2.libgcc): New insn-and-split.
(*clzhi2.libgcc): New insn.
(*clzsihi2.libgcc): New insn.
(*ctzhi2.libgcc): New insn.
(*ctzsihi2.libgcc): New insn.
(*ffshi2.libgcc): New insn.
(*ffssihi2.libgcc): New insn.
(*bswapsi2.libgcc): New insn.
Richard Guenther [Fri, 29 Jul 2011 09:24:23 +0000 (09:24 +0000)]
tree-vrp.c (get_value_range): Only set parameter default definitions to varying, leave others at undefined.
2011-07-29 Richard Guenther <rguenther@suse.de>
* tree-vrp.c (get_value_range): Only set parameter default
definitions to varying, leave others at undefined.
(extract_range_from_binary_expr): Fix undefined handling.
(vrp_visit_phi_node): Handle merged undefined state.
* gcc.dg/uninit-suppress.c: Also disable VRP.
* gcc.dg/uninit-suppress_2.c: Likewise.
H.J. Lu [Fri, 29 Jul 2011 04:09:17 +0000 (04:09 +0000)]
Add x32 support to config/i386/morestack.S.
2011-07-28 H.J. Lu <hongjiu.lu@intel.com>
* config/i386/morestack.S (X86_64_SAVE_NEW_STACK_BOUNDARY): New.
Use X86_64_SAVE_NEW_STACK_BOUNDARY to save the new stack boundary
for x86-64. Properly check __x86_64__ and __LP64__.
Sebastian Pop [Thu, 28 Jul 2011 21:57:10 +0000 (21:57 +0000)]
Fix PR48648: Handle CLAST assignments.
The CLAST produced by CLooG-ISL contains an assignment and GCC chokes
on it. The exact CLAST contains an assignment followed by an if:
scat_1 = max(0,ceild(T_4-7,8));
if (scat_1 <= min(1,floord(T_4-1,8))) {
S7(scat_1);
}
This is equivalent to a loop that iterates only once, and so CLooG
generates an assignment followed by an if instead of a loop. This is
an important optimization that was improved in ISL, that allows
if-conversion: imagine GCC having to figure out that a loop like the
following actually iterates only once, and can be converted to an if:
for (scat_1 = max(0,ceild(T_4-7,8)); scat_1 <= min(1,floord(T_4-1,8)); scat_1++)
S7(scat_1);
This patch implements the translation of CLAST assignments.
Bootstrapped and tested on amd64-linux.
Sebastian Pop [Thu, 28 Jul 2011 21:57:00 +0000 (21:57 +0000)]
Fix PR49876: Continue code generation with integer_zero_node on gloog_error
When setting gloog_error, graphite should continue code generation
without early returns, as otherwise the SSA representation would not
be complete. So set the new expression to integer_zero_node, that
would not require more SSA updates, and continue code generation as
nothing happened.
Regstrapped on amd64-linux.
2011-07-28 Sebastian Pop <sebastian.pop@amd.com>
PR tree-optimization/49876
* sese.c (rename_uses): Do not return false on gloog_error: set
the new_expr to integer_zero_node and continue code generation.
(graphite_copy_stmts_from_block): Remove early exit on gloog_error.
Jakub Jelinek [Thu, 28 Jul 2011 20:56:50 +0000 (22:56 +0200)]
re PR fortran/31067 (MINLOC should sometimes be inlined (gas_dyn is sooooo sloooow))
PR fortran/31067
* frontend-passes.c (optimize_minmaxloc): New function.
(optimize_expr): Call it.
* gfortran.dg/maxloc_2.f90: New test.
* gfortran.dg/maxloc_3.f90: New test.
* gfortran.dg/minloc_1.f90: New test.
* gfortran.dg/minloc_2.f90: New test.
* gfortran.dg/minloc_3.f90: New test.
* gfortran.dg/minmaxloc_7.f90: New test.
Jakub Jelinek [Thu, 28 Jul 2011 20:35:12 +0000 (22:35 +0200)]
re PR debug/49846 (entryval: missing DW_TAG_GNU_call_site_parameter for stack-passed `double' parameter)
PR debug/49846
* var-tracking.c (prepare_call_arguments): For non-MODE_INT stack
arguments also check if they aren't initialized with a MODE_INT
mode of the same size.
* rtlanal.c (tablejump_p): False for returns.
* reorg.c (first_active_target_insn): New static function.
(find_end_label): Set JUMP_LABEL for a new returnjump.
(optimize_skip, get_jump_flags, rare_destination,
mostly_true_jump, get_branch_condition,
steal_delay_list_from_target, own_thread_p,
fill_simple_delay_slots, follow_jumps, fill_slots_from_thread,
fill_eager_delay_slots, relax_delay_slots, make_return_insns,
dbr_schedule): Adjust to handle ret_rtx in JUMP_LABELs.
* jump.c (delete_related_insns): Likewise.
(jump_to_label_p): New function.
(redirect_target): New static function.
(redirect_exp_1): Use it. Adjust to handle ret_rtx in JUMP_LABELS.
(redirect_jump_1): Assert that the new label is nonnull.
(redirect_jump): Likewise.
(redirect_jump_2): Check for ANY_RETURN_P rather than NULL labels.
* ifcvt.c (find_if_case_1): Take care when redirecting jumps to the
exit block.
(dead_or_predicable): Change NEW_DEST arg to DEST_EDGE. All callers
changed. Ensure that the right label is passed to redirect_jump.
* function.c (emit_return_into_block,
thread_prologue_and_epilogue_insns): Ensure new returnjumps have
ret_rtx in their JUMP_LABEL.
* print-rtl.c (print_rtx): Handle ret_rtx in a JUMP_LABEL.
* emit-rtl.c (skip_consecutive_labels): Allow the caller to
pass ret_rtx as label.
* cfglayout.c (fixup_reorder_chain): Use
force_nonfallthru_and_redirect rather than force_nonfallthru.
(duplicate_insn_chain): Copy JUMP_LABELs for returns.
* rtl.h (ANY_RETURN_P): New macro.
(jump_to_label_p): Declare.
* resource.c (find_dead_or_set_registers): Handle ret_rtx in
JUMP_LABELs.
(mark_target_live_regs): Likewise.
* basic-block.h (force_nonfallthru_and_redirect): Declare.
* cfgrtl.c (force_nonfallthru_and_redirect): No longer static.
* config/alpha/alpha.c (alpha_tablejump_addr_vec,
alpha_tablejump_best_label): Remove functions.
* config/alpha/alpha-protos.c (alpha_tablejump_addr_vec,
alpha_tablejump_best_label): Remove declarations.
* config/sh/sh.c (barrier_align, split_branches): Adjust for
ret_rtx in JUMP_LABELs.
* config/arm/arm.c (is_jump_table): Likewise.
Jakub Jelinek [Thu, 28 Jul 2011 16:21:08 +0000 (18:21 +0200)]
re PR debug/49871 (-gdwarf-3 creates invalid DWARF3 with DW_AT_data_member_location attribute)
PR debug/49871
* dwarf2out.c (size_of_die, value_format, output_die): Use
DW_FORM_udata instead of DW_FORM_data[48] for
dw_val_class_unsigned_const DW_AT_data_member_location for DWARF 3.
Alan Modra [Thu, 28 Jul 2011 07:44:24 +0000 (17:14 +0930)]
linux-unwind.h (frob_update_context <__powerpc64__>): Leave r2 REG_UNSAVED if stopped on the instruction that saves r2 in a plt call stub.
* config/rs6000/linux-unwind.h (frob_update_context <__powerpc64__>):
Leave r2 REG_UNSAVED if stopped on the instruction that saves r2
in a plt call stub. Do restore r2 if stopped on bctrl.
Propagate the source location of a template's function_decl to the template_decl built out of it.
Propagate the source location of a template's function_decl to the
template_decl built out of it. Without this, specializations
generated before the template was defined get the location of the
closing ')' instead of the location of the template's name, which can
produce inconsistent locations between translation units, which makes
gold's ODR checker unhappy.
* gcc/cp/pt.c (build_template_decl): Copy the function_decl's source
location to the new template_decl.
* gcc/testsuite/g++.old-deja/g++.pt/crash60.C: Updated.
* libstdc++-v3/testsuite/20_util/weak_ptr/comparison/cmp_neg.cc: Updated.
Sebastian Pop [Wed, 27 Jul 2011 16:53:09 +0000 (16:53 +0000)]
Fix PR49471: canonicalize_loop_ivs should not generate unsigned types.
2011-07-27 Sebastian Pop <sebastian.pop@amd.com>
PR tree-optimization/49471
* tree-ssa-loop-manip.c (canonicalize_loop_ivs): Build an unsigned
iv only when the largest type is unsigned. Do not call
lang_hooks.types.type_for_size.
Sebastian Pop [Wed, 27 Jul 2011 16:52:52 +0000 (16:52 +0000)]
Fix PR47691: do not abort compilation when code generation fails
2011-07-27 Sebastian Pop <sebastian.pop@amd.com>
PR middle-end/47691
* graphite-clast-to-gimple.c (translate_clast_user): Update use of
copy_bb_and_scalar_dependences.
* sese.c (rename_uses): Do not call gcc_assert. Set gloog_error.
(graphite_copy_stmts_from_block): Update call to rename_uses.
(copy_bb_and_scalar_dependences): Update call to
graphite_copy_stmts_from_block.
* sese.h (copy_bb_and_scalar_dependences): Update declaration.
re PR target/49313 (Inefficient libgcc implementations for avr)
PR target/49313
* config/avr/libgcc.S (__ffshi2): Don't skip 2-word instruction.
(__ctzsi2): Result for 0 may be undefined.
(__ctzhi2): Result for 0 may be undefined.
(__popcounthi2): Don't clobber r30. Use __popcounthi2_tail.
(__popcountsi2): Ditto. And don't clobber r26.
(__popcountdi2): Ditto. And don't clobber r27.
* config/avr/avr.md (UNSPEC_COPYSIGN): New c_enum.
(parityhi2): New expand.
(paritysi2): New expand.
(popcounthi2): New expand.
(popcountsi2): New expand.
(clzhi2): New expand.
(clzsi2): New expand.
(ctzhi2): New expand.
(ctzsi2): New expand.
(ffshi2): New expand.
(ffssi2): New expand.
(copysignsf3): New insn.
(bswapsi2): New expand.
(*parityhi2.libgcc): New insn.
(*parityqihi2.libgcc): New insn.
(*paritysihi2.libgcc): New insn.
(*popcounthi2.libgcc): New insn.
(*popcountsi2.libgcc): New insn.
(*popcountqi2.libgcc): New insn.
(*popcountqihi2.libgcc): New insn-and-split.
(*clzhi2.libgcc): New insn.
(*clzsihi2.libgcc): New insn.
(*ctzhi2.libgcc): New insn.
(*ctzsihi2.libgcc): New insn.
(*ffshi2.libgcc): New insn.
(*ffssihi2.libgcc): New insn.
(*bswapsi2.libgcc): New insn.
* gcc.target/i386/avx-os-support.h: New.
* gcc.target/i386/avx-check.h: Include avx-os-support.h
(main): Check avx_os_support before the test is run.
* gcc.target/i386/aes-avx-check.h: Ditto.
* gcc.target/i386/pclmul-avx-check.h: Ditto.
Daniel Carrera [Wed, 27 Jul 2011 10:10:06 +0000 (10:10 +0000)]
re PR fortran/49755 (ALLOCATE with STAT= produces invalid code for already allocated vars)
2011-07-26 Daniel Carrera <dcarrera@gmail.com>
PR fortran/49755
* trans.c (gfc_allocate_using_malloc): Change function signature.
Return nothing. New parameter "pointer". Eliminate temorary variables.
(gfc_allocate_using_lib): Ditto.
(gfc_allocate_allocatable): Ditto. Update call to gfc_allocate_using_lib
and gfc_allocate_using_malloc. Do not free and then reallocate a
variable that is already allocated.
(gfc_likely): New function. Basedon gfc_unlikely.
* trans-array.c (gfc_array_init_size): New parameter "descriptor_block".
Instructions to modify the array descriptor are stored in this block
while other instructions continue to be stored in "pblock".
(gfc_array_allocate): Update call to gfc_array_init_size. Move the
descriptor_block so that the array descriptor is only updated if
the array was allocated successfully.
Update calls to gfc_allocate_allocatable and gfc_allocate_using_malloc.
* trans.h (gfc_allocate_allocatable): Change function signature.
Function now returns void.
(gfc_allocate_using_lib): Ditto, and new function parameter.
(gfc_allocate_using_malloc): Ditto.
* trans-openmp.c (gfc_omp_clause_default_ctor,
gfc_omp_clause_copy_ctor,gfc_trans_omp_array_reduction): Replace a call
to gfc_allocate_allocatable with gfc_allocate_using_malloc.
* trans-stmt.c (gfc_trans_allocate): Update function calls for
gfc_allocate_allocatable and gfc_allocate_using_malloc.
2011-07-26 Daniel Carrera <dcarrera@gmail.com>
PR fortran/49755
* gfortran.dg/multiple_allocation_1.f90: Fix test. Allocating an
allocated array should *not* change its size.
* gfortran.dg/multiple_allocation_3.f90: New test.
Jakub Jelinek [Wed, 27 Jul 2011 09:36:25 +0000 (11:36 +0200)]
dwarf2out.c (output_macinfo_op): Ensure fd->filename points to GC allocated copy of the string.
* dwarf2out.c (output_macinfo_op): Ensure fd->filename points
to GC allocated copy of the string.
(dwarf2out_finish): Emit .debug_macinfo or .debug_macro sections
before .debug_line, not after it.
H.J. Lu [Tue, 26 Jul 2011 21:43:57 +0000 (21:43 +0000)]
Call simplify_gen_subreg for PIC with mode of x.
2011-07-26 H.J. Lu <hongjiu.lu@intel.com>
PR target/47372
* config/i386/i386.c (ix86_delegitimize_address): Call
simplify_gen_subreg for PIC with mode of x only if modes of
x and orig_x are different.
Jakub Jelinek [Tue, 26 Jul 2011 21:12:39 +0000 (23:12 +0200)]
dwarf2out.c (output_macinfo_op): Ensure fd->filename points to GC allocated copy of the string.
* dwarf2out.c (output_macinfo_op): Ensure fd->filename points
to GC allocated copy of the string.
(dwarf2out_finish): Emit .debug_macinfo or .debug_macro sections
before .debug_line, not after it.
"Bug 47046 - gcc.target/i386/sse4_1-movntdqa.c ICEs with -fgraphite-identity"
The problem here is that we are left with the following code to be
translated in the new representation following the transform that
Graphite has chosen:
In this particular case we have a nonlinear expression "i * i" for
which we have to generate code following the new graphite_iv variables.
The patch fixes the function that detects whether we are passing non
linear stuff to graphite: evolution_function_is_affine_p. It seems
like for the moment evolution_function_is_affine_p is testing whether
an evolution function is affine only in the innermost loop, without
looking recursively at what happens in outer loops.
The chrec for this case is: {0, +, {0, +, {1, +, 2}_1}_1}_2 and we are
testing whether the evolution is affine only for the loop_2, which is
true as we have {0, +, blah}_2 with blah invariant in loop_2.
The patch adds the recursive call to evolution_function_is_affine_p.
Bootstrapped and tested on amd64-linux.
2011-07-26 Sebastian Pop <sebastian.pop@amd.com>
PR middle-end/47046
* tree-chrec.h (evolution_function_is_affine_p): Recursively call
evolution_function_is_affine_p on CHREC_RIGHT.
re PR target/47369 ([x32] internal compiler error: in extract_insn, at recog.c:2109)
PR target/47369
PR target/49853
* config/i386/i386.c (ix86_expand_move): Call convert_to_mode
if legitimize_tls_address returned operand in wrong mode. Allow
SImode and DImode symbolic operand for PIC. Call convert_to_mode
if legitimize_pic_address returned operand in wrong mode.
Co-Authored-By: H.J. Lu <hongjiu.lu@intel.com>
From-SVN: r176798
projects / thirdparty / gcc.git / log
