PR c/18079
c/
* c-decl.c (diagnose_mismatched_decls): Warn for mismatched
always_inline/noinline and hot/cold attributes.
c-family/
* c-common.c (handle_noinline_attribute): Warn if the attribute
conflicts with always_inline attribute.
(handle_always_inline_attribute): Warn if the attribute conflicts
with noinline attribute.
testsuite/
* gcc.dg/pr18079.c: New test.
* gcc.dg/pr18079-2.c: New test.
From-SVN: r209796
2014-04-25 Richard Biener <rguenther@suse.de>
PR ipa/60912
* tree-ssa-structalias.c (ipa_pta_execute): Compute direct
call stmt use/clobber sets during stmt walk instead of
walking the possibly incomplete set of caller edges.
* g++.dg/opt/pr60912.C: New testcase.
From-SVN: r209780
2014-04-25 Richard Biener <rguenther@suse.de>
PR ipa/60911
* passes.c (apply_ipa_transforms): Inline into only caller ...
(execute_one_pass): ... here. Properly bring in function
bodies for nodes we want to apply IPA transforms to.
* gcc.dg/lto/pr60911_0.c: New testcase.
From-SVN: r209779
PR go/60931
runtime: Fix garbage collector issue with non 4kB system page size
The go garbage collector tracks memory in terms of 4kB pages.
Most of the code checks getpagesize() at runtime and does the
right thing.
On a 64kB ppc64 box I see SEGVs in long running processes
which has been diagnosed as a bug in scavengelist.
scavengelist does a madvise(MADV_DONTNEED) without rounding
the arguments to the system page size. A strace of one of the
failures shows the problem:
madvise(0xc211030000, 4096, MADV_DONTNEED) = 0
The kernel rounds the length up to 64kB and we mark 60kB of
valid data as no longer needed.
Round start up to a system page and end down before calling
madvise.
From-SVN: r209777
2014-04-24 Cong Hou <congh@google.com>
PR tree-optimization/60896
* tree-vect-patterns.c (vect_recog_dot_prod_pattern): Pick up
all statements in PATTERN_DEF_SEQ in recognized widen-mult pattern.
(vect_mark_pattern_stmts): Set the def type of all statements in
PATTERN_DEF_SEQ as vect_internal_def.
2014-04-24 Cong Hou <congh@google.com>
PR tree-optimization/60896
* g++.dg/vect/pr60896.cc: New test.
From-SVN: r209773
2014-04-24 Vishnu K S <Vishnu.k_s@atmel.com>
* gcc/config/avr/avr.c: Add comment on why -fdelete-null-pointer-checks
is disabled.
2014-04-24 Vishnu K S <Vishnu.k_s@atmel.com>
* gcc/testsuite/gcc.dg/tree-ssa/isolate-1.c: Skip test if
keeps_null_pointer_checks.
* gcc/testsuite/gcc.dg/tree-ssa/isolate-2.c: Ditto
* gcc/testsuite/gcc.dg/tree-ssa/isolate-3.c: Ditto
* gcc/testsuite/gcc.dg/tree-ssa/isolate-4.c: Ditto
* gcc/testsuite/gcc.dg/tree-ssa/isolate-5.c: Ditto
From-SVN: r209767
gcc/c/
* c-parser.c (c_parser_omp_atomic): Allow seq_cst before
atomic-clause, allow comma in between atomic-clause and
seq_cst.
gcc/cp/
* parser.c (cp_parser_omp_atomic): Allow seq_cst before
atomic-clause, allow comma in between atomic-clause and
seq_cst.
gcc/testsuite/
* c-c++-common/gomp/atomic-16.c: Remove all dg-error directives.
Replace load with read and store with write.
libgomp/
* testsuite/libgomp.c++/atomic-14.C: Allow seq_cst and
atomic type clauses in any order and optional comma in between.
* testsuite/libgomp.c++/atomic-15.C: Likewise.
* testsuite/libgomp.c/atomic-17.c: Likewise.
From-SVN: r209762
* tree.h (OMP_CLAUSE_LINEAR_GIMPLE_SEQ): Define.
* gimplify.c (omp_is_private): Change last argument's type to int.
Only diagnose lastprivate if the simd argument is 1, only diagnose
linear if the simd argument is 2.
(gimplify_omp_for): Adjust omp_is_private callers. When adding
lastprivate or private, add the clause to OMP_FOR_CLAUSES. Pass
GOVD_EXPLICIT to omp_add_variable. For simd with collapse == 1
create OMP_CLAUSE_LINEAR rather than OMP_CLAUSE_PRIVATE for var.
If var != decl and decl is in OMP_CLAUSE_LINEAR, gimplify decl
increment to OMP_CLAUSE_LINEAR_GIMPLE_SEQ.
* omp-low.c (scan_sharing_clauses, lower_lastprivate_clauses): Handle
OMP_CLAUSE_LINEAR_GIMPLE_SEQ.
* tree-nested.c (convert_nonlocal_omp_clauses,
convert_local_omp_clauses): Handle OMP_CLAUSE_LINEAR.
* testsuite/libgomp.c/simd-7.c: New test.
* testsuite/libgomp.c/simd-8.c: New test.
* testsuite/libgomp.c/simd-9.c: New test.
* testsuite/libgomp.c/loop-16.c: New test.
From-SVN: r209760
When making pass::execute functions take a function * argument I missed
the sh_treg_combine pass, so fix that here.
gcc/
* config/sh/sh_treg_combine.c (sh_treg_combine::execute): Take
function * argument.
From-SVN: r209751
gcc/ChangeLog:
2014-04-11 Alan Lawrence <alan.lawrence@arm.com>
* config/aarch64/aarch64.c (aarch64_evpc_tbl): Enable for bigendian.
gcc/testsuite/ChangeLog:
2014-04-11 Alan Lawrence <alan.lawrence@arm.com>
* lib/target-supports.exp (check_effective_target_vect_perm): Return
true for aarch64_be.
From-SVN: r209749
2014-04-24 Radovan Obradovic <robradovic@mips.com>
Tom de Vries <tom@codesourcery.com>
* gcc.dg/fuse-caller-save.c: New test.
Co-Authored-By: Tom de Vries <tom@codesourcery.com>
From-SVN: r209745
2014-04-24 Radovan Obradovic <robradovic@mips.com>
Tom de Vries <tom@codesourcery.com>
* common.opt (fuse-caller-save): New option.
Co-Authored-By: Tom de Vries <tom@codesourcery.com>
From-SVN: r209743
2014-04-24 Tejas Belagod <tejas.belagod@arm.com>
* config/aarch64/aarch64.c (aarch64_evpc_tbl): Reverse order of elements
for big-endian.
From-SVN: r209742
2014-04-24 Richard Biener <rguenther@suse.de>
* expr.c (expand_expr_real_1): Avoid gimple_assign_rhs_to_tree
during TER and instead use the sepops interface for expanding
non-GIMPLE_SINGLE_RHS.
From-SVN: r209741
* config/aarch64/aarch64.h (TARGET_CPU_CPP_BUILTINS): Check
TARGET_SIMD rather than TARGET_GENERAL_REGS_ONLY.
(TARGET_SIMD): Take AARCH64_ISA_SIMD into account.
(TARGET_FLOAT): Take AARCH64_ISA_FP into account.
(TARGET_CRYPTO): Take TARGET_SIMD into account.
From-SVN: r209737
2014-04-24 Terry Guo <terry.guo@arm.com>
* config/arm/arm.h (machine_function): Define variable
after_arm_reorg here.
* config/arm/arm.c (after_arm_reorg): Remove the definition.
(arm_split_constant): Update the way to access variable
after_arm_reorg.
(arm_reorg): Ditto.
(arm_output_function_epilogue): Remove the reset of after_arm_reorg.
From-SVN: r209735
