This is the mail archive of the
gcc-regression@gcc.gnu.org
mailing list for the GCC project.
new FAILs on HEAD
- From: Michael Ritzert <gcc at ds217-115-141-84 dot dedicated dot hosteurope dot de>
- To: gcc-regression at gcc dot gnu dot org, bkoz at redhat dot com, dberlin at dberlin dot org, echristo at redhat dot com, ghazi at caip dot rutgers dot edu, kazu at cs dot umass dot edu, paul at codesourcery dot com, pcarlini at suse dot de, pop at cri dot ensmp dot fr, sje at cup dot hp dot com, tobias dot schlueter at physik dot uni-muenchen dot de, uweigand at de dot ibm dot com, zadeck at naturalbridge dot com
- Date: Thu, 7 Oct 2004 07:20:01 +0200
- Subject: new FAILs on HEAD
- Reply-to: gcc-regression at gcc dot gnu dot org
If you directly receive this mail, your name is tagged to one of the ChangeLog
entries. A copy is also mailed to gcc-regressions.
The build system:
debian GNU/Linux i686 unstable
i686-pc-linux-gnu
configure flags: --enable-threads=posix --prefix=/opt/gcc-HEAD --enable-__cxa_atexit --with-system-zlib --enable-languages=c,c++,f95,java,objc
These new FAILs appeared since the last run:
FAIL: gfortran.dg/empty_format_1.f90 -O0 (test for excess errors)
FAIL: gfortran.dg/empty_format_1.f90 -O1 (test for excess errors)
FAIL: gfortran.dg/empty_format_1.f90 -O2 (test for excess errors)
FAIL: gfortran.dg/empty_format_1.f90 -O3 -fomit-frame-pointer (test for excess errors)
FAIL: gfortran.dg/empty_format_1.f90 -O3 -fomit-frame-pointer -funroll-loops (test for excess errors)
FAIL: gfortran.dg/empty_format_1.f90 -O3 -fomit-frame-pointer -funroll-all-loops -finline-functions (test for excess errors)
FAIL: gfortran.dg/empty_format_1.f90 -O3 -g (test for excess errors)
FAIL: gfortran.dg/empty_format_1.f90 -Os (test for excess errors)
The complete testsuite logs can be found at
http://www.mailsys.de/~gcc/logs-HEAD-i686-1097096787.tar.bz2
These are the new ChangeLog entries since the last run:
2004-10-06 Eric Christopher <echristo@redhat.com>
* dwarf2.h: Sync with include/elf/dwarf2.h
2004-10-06 Daniel Berlin <dberlin@dberlin.org>
* tree-pretty-print.c (dump_generic_node): Fix printing of BINFO
and TREE_VEC nodes.
2004-10-06 Daniel Berlin <dberlin@dberlin.org>
* lambda-code.c (lambda_loopnest_to_gcc_loopnest): Convert
to use FOR_EACH_SSA_USE_OPERAND iterator, and propagate_value.
2004-10-06 Daniel Berlin <dberlin@dberlin.org>
* lambda-code.c (compute_nest_using_fourier_motzkin): New
function.
(lambda_compute_auxillary_space): Split from here.
2004-10-06 Daniel Berlin <dberlin@dberlin.org>
* tree-ssa-loop-ivopts.c (expr_invariant_in_loop): Make non-static.
* tree-flow.h: Add prototype.
* lambda-code.c (invariant_in_loop_and_outer_loops): Use
expr_invariant_in_loop.
2004-10-06 Kazu Hirata <kazu@cs.umass.edu>
* tree-cfg.c (thread_jumps): Remove a duplicate check for
EXIT_BLOCK_PTR.
2004-10-06 Ulrich Weigand <uweigand@de.ibm.com>
* config/s390/s390-protos.h (s_imm_operand): Remove.
(s390_pool_operand): Add prototype.
* config/s390/s390.c (general_s_operand): Remove.
(s_imm_operand): Remove.
(s_operand): Merge contents of general_s_operand.
(s390_pool_operand): New function.
* config/s390/s390.h (PREDICATE_CODES): Remove s_imm_operand.
* config/s390/s390.md ("*cmpdi_cct", "*cmpsi_cct"): New insns.
("*cmpdi_ccu"): Merge Q->Q alternative.
("*cmpsi_ccu", "*cmphi_ccu"): Likewise.
("*cmpqi_ccu"): Merge Q->Q, n->Q, n->S alternatives.
("*cli"): Remove.
("*cmpdi_ccu_mem", "*cmpsi_ccu_mem"): Likewise.
("*cmphi_ccu_mem", "*cmpqi_ccu_mem"): Likewise.
("*movstricthi"): Use memory_operand instead of s_imm_operand.
2004-10-06 Steve Ellcey <sje@cup.hp.com>
* config/ia64/ia64.h (FUNCTION_ARG_BOUNDARY): Change macro to
call ia64_function_arg_boundary.
* config/ia64/ia64-protos.h (ia64_function_arg_boundary): New.
* config/ia64/ia64.c (ia64_function_arg_boundary): New.
(ia64_function_arg_advance): Do not put 128 bit floats into
FP registers.
2004-10-06 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
* builtins.c (expand_builtin_strcpy): Delete duplicate code.
Accept an expression instead of an arglist.
(expand_builtin_stpcpy): Accept an expression instead of an
arglist.
* Makefile.in (tree-ssa-loop-niter.o): Depends on tree-data-ref.h.
* cfgloop.c (initialize_loops_parallel_p): New.
(flow_loops_find): Initialize the parallel_p field to true for all
the loops.
* tree-ssa-loop-niter.c: Include "tree-data-ref.h".
(estimate_numbers_of_iterations_loop): Infers the loop bounds from
the size of the data accessed in the loop.
(struct nb_iter_bound): Moved...
* cfgloop.h (struct nb_iter_bound): ... here.
(estimated_nb_iterations, parallel_p): New fields in struct loop.
(record_estimate): Declare extern here.
* tree-chrec.c: Fix comments.
(nb_vars_in_chrec): New function.
* tree-chrec.h (nb_vars_in_chrec): Declared here.
* tree-data-ref.c: Don't include lambda.h, that is already included
in tree-data-ref.h.
(tree_fold_divides_p): Don't check for integer_onep.
(tree_fold_bezout): Removed.
(gcd): New static duplicated function.
(int_divides_p, dump_subscript): New.
(dump_data_dependence_relation): Use dump_subscript.
(dump_dist_dir_vectors, dump_ddrs, compute_estimated_nb_iterations,
estimate_niter_from_size_of_data): New.
(analyze_array_indexes, analyze_array): Call
estimate_niter_from_size_of_data during the detection of array
references. Pass in a pointer to the statement that contains the
array reference.
(all_chrecs_equal_p): New.
(compute_distance_vector): Renamed compute_subscript_distance.
Deal with multivariate conflict functions.
(initialize_data_dependence_relation): Initialize DDR_AFFINE_P,
DDR_SIZE_VECT, DDR_DIST_VECT, and DDR_DIR_VECT.
(non_affine_dependence_relation): New.
(analyze_ziv_subscript, analyze_siv_subscript_cst_affine,
analyze_siv_subscript, analyze_miv_subscript,
analyze_overlapping_iterations, subscript_dependence_tester):
Initialize and return last_conflicts function.
(initialize_matrix_A, FLOOR, compute_overlap_steps_for_affine_univar,
compute_overlap_steps_for_affine_1_2): New.
(analyze_siv_subscript_affine_cst): Removed.
(analyze_subscript_affine_affine): Disprove dependences based on the
iteration domains. Solve the univariate dependence case as before,
but use lambda_matrix_right_hermite instead of tree_fold_bezout.
Implement the multivariate case of 2 versus 1 variables.
(undetermined_conflicts_p, no_conflicts_p): New.
(build_classic_dist_vector, build_classic_dir_vector): Implement some
unhandled cases.
(find_data_references_in_loop): Compute and initialize
loop->estimated_nb_iterations and loop->parallel_p.
(analyze_all_data_dependences): Modify the debug dump order.
* tree-data-ref.h (SUB_LAST_CONFLICT_IN_A, SUB_LAST_CONFLICT_IN_B,
subscript->last_conflict_in_a, subscript->last_conflict_in_b): Removed.
(SUB_LAST_CONFLICT, subscript->last_conflict,
data_dependence_relation->affine_p, data_dependence_relation->size_vect,
DDR_AFFINE_P, DDR_SIZE_VECT): New.
(find_data_references_in_loop, initialize_data_dependence_relation,
dump_subscript, dump_ddrs, dump_dist_dir_vectors): Declared here.
2004-10-04 Sebastian Pop <pop@cri.ensmp.fr>
2004-10-06 Tobias Schlueter <tobias.schlueter@physik.uni-muenchen.de>
PR fortran/17568
* simplify.c (twos_complement): New function.
(gfc_simplify_ishft, gfc_simplify_ishftc): Revise.
2004-10-06 Paul Brook <paul@codesourcery.com>
* trans-stmt.c (gfc_trans_simple_do): New function.
(gfc_trans_do): Use it. Evaluate iteration bounds before entering
loop. Update comments.
2004-10-06 Ulrich Weigand <uweigand@de.ibm.com>
* parse.y (issue_warning_error_from_context): Use va_list *
instead of va_list parameter.
(parse_error_context): Update call.
(parse_warning_context): Likewise.
2004-10-06 Kenneth Zadeck <zadeck@naturalbridge.com>
gcc.dg/20011119-1.c : Added memory attribute to asm sequence.
2004-10-06 Paul Brook <paul@codesourcery.com>
PR libfortran/17709
* gfortran.dg/empty_format_1.f90: New test.
2004-10-06 Tobias Schlueter <tobias.schlueter@physik.uni-muenchen.de>
PR fortran/17568
* gfortran.dg/ishft.f90: New test.
PR fortran/17283
* gfortran.fortran-torture/execute/intrinsic_pack.f90: Enable new
tests.
2004-10-06 Paul Brook <paul@codesourcery.com>
* gfortran.dg/do_1.f90: New test.
2004-10-06 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
* gcc.c-torture/execute/builtins/lib/strcpy.c: Don't abort when
optimizing for size.
2004-10-06 Benjamin Kosnik <bkoz@redhat.com>
Paolo Carlini <pcarlini@suse.de>
PR libstdc++/17780
* src/mt_allocator.cc (__pool<true>::_M_reserve_block): Revert
to old locking order.
2004-10-06 Paul Brook <paul@codesourcery.com>
PR libfortran/17709
* io/transfer.c (data_transfer_init): Reset sf_seen_eor.