Andreas Schwab [Thu, 16 Jan 2014 10:14:29 +0000 (10:14 +0000)]
* gcc.c-torture/execute/pr59747.c (fn1): Return a value.
From-SVN: r206659
Uros Bizjak [Thu, 16 Jan 2014 08:23:33 +0000 (09:23 +0100)]
i386.c (ix86_hard_regno_mode_ok): Use VALID_AVX256_REG_OR_OI_MODE.
* config/i386/i386.c (ix86_hard_regno_mode_ok): Use
VALID_AVX256_REG_OR_OI_MODE.
From-SVN: r206658
Kirill Yukhin [Thu, 16 Jan 2014 04:25:38 +0000 (04:25 +0000)]
Fix testsuite/ChangeLog entry for PR59808.
From-SVN: r206657
GCC Administrator [Thu, 16 Jan 2014 00:17:01 +0000 (00:17 +0000)]
Daily bump.
From-SVN: r206655
Pat Haugen [Wed, 15 Jan 2014 23:48:12 +0000 (23:48 +0000)]
rs6000.c (rs6000_output_function_prologue): Check if current procedure should be profiled.
* config/rs6000/rs6000.c (rs6000_output_function_prologue): Check if
current procedure should be profiled.
From-SVN: r206650
Andrew Pinski [Wed, 15 Jan 2014 23:30:16 +0000 (23:30 +0000)]
aarch64.c (aarch64_register_move_cost): Correct cost of moving from/to the STACK_REG register class.
2014-01-15 Andrew Pinski <apinski@cavium.com>
* config/aarch64/aarch64.c (aarch64_register_move_cost): Correct cost
of moving from/to the STACK_REG register class.
From-SVN: r206649
Marek Polacek [Wed, 15 Jan 2014 22:02:53 +0000 (22:02 +0000)]
Add missing CL entry.
From-SVN: r206648
Richard Henderson [Wed, 15 Jan 2014 21:41:03 +0000 (13:41 -0800)]
re PR debug/54694 (internal compiler error: in dwarf2out_frame_debug_expr, at dwarf2out.c:2387)
PR debug/54694
Diagnose frame_pointer_required vs fixed hfp
From-SVN: r206647
Teresa Johnson [Wed, 15 Jan 2014 21:21:32 +0000 (21:21 +0000)]
tree-ssa-sccvn.c (visit_reference_op_call): Handle NULL vdef.
2014-01-15 Teresa Johnson <tejohnson@google.com>
* tree-ssa-sccvn.c (visit_reference_op_call): Handle NULL vdef.
From-SVN: r206646
Laurent Alfonsi [Wed, 15 Jan 2014 19:37:35 +0000 (19:37 +0000)]
re PR c++/49718 (please allow no_instrument_function attribute in class member definition/declaration)
PR c++/49718
* c-common.c (handle_no_instrument_function_attribute): Allow
no_instrument_function attribute in class member
definition/declaration.
PR c++/49718
* g++.dg/pr49718.C: New test.
From-SVN: r206643
Richard Sandiford [Wed, 15 Jan 2014 19:31:46 +0000 (19:31 +0000)]
umips-branch-4.c: Add addressing=absolute.
gcc/testsuite/
* gcc.target/mips/umips-branch-4.c: Add addressing=absolute.
From-SVN: r206642
Bill Schmidt [Wed, 15 Jan 2014 19:16:53 +0000 (19:16 +0000)]
altivec.md (mulv8hi3): Explicitly generate vmulesh and vmulosh rather than call gen_vec_widen_smult_*.
gcc:
2014-01-15 Bill Schmidt <wschmidt@vnet.linux.ibm.com>
* config/rs6000/altivec.md (mulv8hi3): Explicitly generate vmulesh
and vmulosh rather than call gen_vec_widen_smult_*.
(vec_widen_umult_even_v16qi): Test VECTOR_ELT_ORDER_BIG rather
than BYTES_BIG_ENDIAN to determine use of even or odd instruction.
(vec_widen_smult_even_v16qi): Likewise.
(vec_widen_umult_even_v8hi): Likewise.
(vec_widen_smult_even_v8hi): Likewise.
(vec_widen_umult_odd_v16qi): Likewise.
(vec_widen_smult_odd_v16qi): Likewise.
(vec_widen_umult_odd_v8hi): Likewise.
(vec_widen_smult_odd_v8hi): Likewise.
(vec_widen_umult_hi_v16qi): Explicitly generate vmuleub and
vmuloub rather than call gen_vec_widen_umult_*.
(vec_widen_umult_lo_v16qi): Likewise.
(vec_widen_smult_hi_v16qi): Explicitly generate vmulesb and
vmulosb rather than call gen_vec_widen_smult_*.
(vec_widen_smult_lo_v16qi): Likewise.
(vec_widen_umult_hi_v8hi): Explicitly generate vmuleuh and vmulouh
rather than call gen_vec_widen_umult_*.
(vec_widen_umult_lo_v8hi): Likewise.
(vec_widen_smult_hi_v8hi): Explicitly gnerate vmulesh and vmulosh
rather than call gen_vec_widen_smult_*.
(vec_widen_smult_lo_v8hi): Likewise.
gcc/testsuite:
2014-01-15 Bill Schmidt <wschmidt@vnet.linux.ibm.com>
* gcc.dg/vmx/mult-even-odd.c: New.
* gcc.dg/vmx/mult-even-odd-be-order.c: New.
From-SVN: r206641
Balaji V. Iyer [Wed, 15 Jan 2014 19:11:16 +0000 (19:11 +0000)]
target-supports.exp (check_libcilkrts_available): Added an extern "C" if we are using C++ along with a function prototype.
2014-01-15 Balaji V. Iyer <balaji.v.iyer@intel.com>
* lib/target-supports.exp (check_libcilkrts_available): Added an
extern "C" if we are using C++ along with a function prototype.
From-SVN: r206640
Jason Merrill [Wed, 15 Jan 2014 19:10:09 +0000 (14:10 -0500)]
re PR c++/59659 (large zero-initialized std::array compile time excessive)
PR c++/59659
* typeck2.c (massage_init_elt): New.
(process_init_constructor_record)
(process_init_constructor_union): Use it.
(process_init_constructor_array): Use it. Use RANGE_EXPR.
(split_nonconstant_init_1): Handle it.
* semantics.c (cxx_eval_vec_init_1): Use force_rvalue.
From-SVN: r206639
Jeff Law [Wed, 15 Jan 2014 18:13:52 +0000 (11:13 -0700)]
re PR tree-optimization/59747 (wrong code at -Os and above on x86_64-linux-gnu in 64-bit mode)
PR tree-optimization/59747
* ree.c (find_and_remove_re): Properly handle case where a second
eliminated extension requires widening a copy created for elimination
of a prior extension.
(combine_set_extension): Ensure that the number of hard regs needed
for a destination register does not change when we widen it.
PR tree-optimization/59747
* gcc.c-torture/execute/pr59747.c: New test.
From-SVN: r206638
Sebastian Huber [Wed, 15 Jan 2014 17:44:33 +0000 (17:44 +0000)]
config.gcc (*-*-rtems*): Add t-rtems to tmake_file.
2014-01-15 Sebastian Huber <sebastian.huber@embedded-brains.de>
* config.gcc (*-*-rtems*): Add t-rtems to tmake_file.
(arm*-*-uclinux*eabi*): Do not override an existing tmake_file.
(arm*-*-eabi* | arm*-*-symbianelf* | arm*-*-rtems*): Likwise.
(arm*-*-rtems*): Use t-rtems from existing tmake_file.
(avr-*-rtems*): Likewise.
(bfin*-rtems*): Likewise.
(moxie-*-rtems*): Likewise.
(h8300-*-rtems*): Likewise.
(i[34567]86-*-rtems*): Likewise.
(lm32-*-rtems*): Likewise.
(m32r-*-rtems*): Likewise.
(m68k-*-rtems*): Likewise.
(microblaze*-*-rtems*): Likewise.
(mips*-*-rtems*): Likewise.
(powerpc-*-rtems*): Likewise.
(sh-*-rtems*): Likewise.
(sparc-*-rtems*): Likewise.
(sparc64-*-rtems*): Likewise.
(v850-*-rtems*): Likewise.
(m32c-*-rtems*): Likewise.
From-SVN: r206637
Vladimir Makarov [Wed, 15 Jan 2014 17:32:47 +0000 (17:32 +0000)]
re PR rtl-optimization/59511 (FAIL: gcc.target/i386/pr36222-1.c scan-assembler-not movdqa with -mtune=corei7)
2014-01-15 Vladimir Makarov <vmakarov@redhat.com>
PR rtl-optimization/59511
* ira.c (ira_init_register_move_cost): Use memory costs for some
cases of register move cost calculations.
* lra-constraints.c (lra_constraints): Use REG_FREQ_FROM_BB
instead of BB frequency.
* lra-coalesce.c (move_freq_compare_func, lra_coalesce): Ditto.
* lra-assigns.c (find_hard_regno_for): Ditto.
From-SVN: r206636
H.J. Lu [Wed, 15 Jan 2014 17:08:38 +0000 (17:08 +0000)]
Silence vector ABI change warnings for x86
PR target/59794
* c-c++-common/convert-vec-1.c: Also prune ABI change for
Linux/x86.
* g++.dg/cpp0x/constexpr-53094-2.C: Likewise.
* g++.dg/ext/attribute-test-1.C: Likewise.
* g++.dg/ext/attribute-test-2.C: Likewise.
* g++.dg/ext/attribute-test-3.C: Likewise.
* g++.dg/ext/attribute-test-4.C: Likewise.
* g++.dg/ext/pr56790-1.C: Likewise.
* g++.dg/torture/pr38565.C: Likewise.
* gcc.dg/pr53060.c: Likewise.
* c-c++-common/scal-to-vec2.c: Add -msse2 for x86.
* c-c++-common/vector-compare-2.c: Likewise.
* gcc.dg/Wstrict-aliasing-bogus-ref-all-2.c: Likewise.
* g++.dg/conversion/simd1.C: Add -msse2 for x86. Adjust
dg-message line number.
From-SVN: r206634
François Dumont [Wed, 15 Jan 2014 16:55:34 +0000 (16:55 +0000)]
Add missing PR reference in ChangeLog.
From-SVN: r206633
François Dumont [Wed, 15 Jan 2014 16:53:02 +0000 (16:53 +0000)]
hashtable_policy.h: Fix some long lines.
2014-01-15 François Dumont <fdumont@gcc.gnu.org>
* include/bits/hashtable_policy.h: Fix some long lines.
* include/bits/hashtable.h (__hash_code_base_access): Define and
use it to check its _M_bucket_index noexcept qualification. Use
also in place of...
(__access_protected_ctor): ...this.
* testsuite/23_containers/unordered_set/instantiation_neg.cc:
Adapt line number.
* testsuite/23_containers/unordered_set/
not_default_constructible_hash_neg.cc: Likewise.
From-SVN: r206632
Matthias Klose [Wed, 15 Jan 2014 15:31:55 +0000 (15:31 +0000)]
go-test.exp: Set goarch for aarch64*-*-*.
2014-01-15 Matthias Klose <doko@ubuntu.com>
* go.test/go-test.exp: Set goarch for aarch64*-*-*.
From-SVN: r206631
Richard Biener [Wed, 15 Jan 2014 15:13:08 +0000 (15:13 +0000)]
re PR tree-optimization/59822 (ice in compute_live_loop_exits with -O3)
2014-01-15 Richard Biener <rguenther@suse.de>
PR tree-optimization/59822
* tree-vect-stmts.c (hoist_defs_of_uses): New function.
(vectorizable_load): Use it to hoist defs of uses of invariant
loads out of the loop.
* g++.dg/torture/pr59822.C: New testcase.
From-SVN: r206630
Kirill Yukhin [Wed, 15 Jan 2014 12:01:08 +0000 (12:01 +0000)]
re PR testsuite/59808 (r206596 caused: FAIL: gcc.target/i386/sse-14.c (test for excess errors))
PR target/59808
* gcc.target/i386/sse-12.c: Add `-mavx512[cd, er, pf]' options.
* gcc.target/i386/sse-14.c: Fix PR59808.
From-SVN: r206629
Matthew Gretton-Dann [Wed, 15 Jan 2014 10:27:55 +0000 (10:27 +0000)]
re PR target/59695 (bad code generation on aarch64 in aarch64_output_mi_thunk)
2014-01-15 Matthew Gretton-Dann <matthew.gretton-dann@linaro.org>
Kugan Vivekanandarajah <kuganv@linaro.org>
gcc/
PR target/59695
* config/aarch64/aarch64.c (aarch64_build_constant): Fix incorrect
truncation.
gcc/testsuite/
PR target/59695
* g++.dg/pr59695.C: New testcase.
Co-Authored-By: Kugan Vivekanandarajah <kuganv@linaro.org>
From-SVN: r206628
Jakub Jelinek [Wed, 15 Jan 2014 09:39:11 +0000 (10:39 +0100)]
re PR target/59803 (s390x -march=z10 reload ICE)
Add
PR target/59803
reference to ChangeLog entries.
From-SVN: r206625
Richard Biener [Wed, 15 Jan 2014 08:47:30 +0000 (08:47 +0000)]
re PR rtl-optimization/59802 (excessive compile time in RTL optimizers (loop unswitching, CPROP))
2014-01-15 Richard Biener <rguenther@suse.de>
PR rtl-optimization/59802
* lcm.c (compute_available): Use inverted postorder to seed
the initial worklist.
From-SVN: r206624
Andreas Krebbel [Wed, 15 Jan 2014 08:36:44 +0000 (08:36 +0000)]
s390.c (s390_preferred_reload_class): Don't return ADDR_REGS for invalid symrefs in non-PIC code.
2014-01-15 Andreas Krebbel <Andreas.Krebbel@de.ibm.com>
* config/s390/s390.c (s390_preferred_reload_class): Don't return
ADDR_REGS for invalid symrefs in non-PIC code.
2014-01-15 Andreas Krebbel <Andreas.Krebbel@de.ibm.com>
* gcc.c-torture/compile/pr59803.c: New testcase.
From-SVN: r206623
Jakub Jelinek [Wed, 15 Jan 2014 08:08:17 +0000 (09:08 +0100)]
re PR other/58712 (issues found by --enable-checking=valgrind)
PR other/58712
* builtins.c (determine_block_size): Initialize *probable_max_size
even if len_rtx is CONST_INT.
From-SVN: r206621
Jakub Jelinek [Wed, 15 Jan 2014 08:03:53 +0000 (09:03 +0100)]
re PR c/58943 (wrong calculation of indirect structure member arithmetic via function call)
PR c/58943
* c-typeck.c (build_modify_expr): For lhs op= rhs, if rhs has side
effects, preevaluate rhs using SAVE_EXPR first.
* c-omp.c (c_finish_omp_atomic): Set in_late_binary_op around
build_modify_expr with non-NOP_EXPR opcode. Handle return from it
being COMPOUND_EXPR.
(c_finish_omp_for): Handle incr being COMPOUND_EXPR with first
operand a SAVE_EXPR and second MODIFY_EXPR.
* gcc.c-torture/execute/pr58943.c: New test.
* gcc.dg/tree-ssa/ssa-fre-33.c (main): Avoid using += in the test.
From-SVN: r206620
Chris Manghane [Wed, 15 Jan 2014 00:17:22 +0000 (00:17 +0000)]
compiler: Use backend interface for interface info and field expressions.
* go-gcc.cc (Gcc_backend::compound_expression): New function.
(Gcc_backend::conditional_expression): New function.
From-SVN: r206615
GCC Administrator [Wed, 15 Jan 2014 00:16:39 +0000 (00:16 +0000)]
Daily bump.
From-SVN: r206614
Ian Lance Taylor [Tue, 14 Jan 2014 23:21:16 +0000 (23:21 +0000)]
compiler: Add Backend Function type.
From-SVN: r206611
Andrew Pinski [Tue, 14 Jan 2014 19:55:59 +0000 (19:55 +0000)]
aarch64-protos.h (tune_params): Add issue_rate.
2014-01-14 Andrew Pinski <apinski@cavium.com>
* config/aarch64/aarch64-protos.h (tune_params): Add issue_rate.
* config/aarch64/aarch64.c (generic_tunings): Add issue rate of 2.
(cortexa53_tunings): Likewise.
(aarch64_sched_issue_rate): New function.
(TARGET_SCHED_ISSUE_RATE): Define.
From-SVN: r206609
Vladimir Makarov [Tue, 14 Jan 2014 19:08:32 +0000 (19:08 +0000)]
ira-costs.c (find_costs_and_classes): Add missed ira_init_register_move_cost_if_necessary.
2014-01-14 Vladimir Makarov <vmakarov@redhat.com>
* ira-costs.c (find_costs_and_classes): Add missed
ira_init_register_move_cost_if_necessary.
From-SVN: r206606
Vladimir Makarov [Tue, 14 Jan 2014 19:07:01 +0000 (19:07 +0000)]
re PR target/59787 ([ARM] mmx-2.c causes ICE when GCC is configured for cortex-a5/vfpv3-d16-fp16)
2014-01-14 Vladimir Makarov <vmakarov@redhat.com>
PR target/59787
* config/arm/arm.c (arm_coproc_mem_operand): Add lra_in_progress.
From-SVN: r206605
H.J. Lu [Tue, 14 Jan 2014 16:41:10 +0000 (08:41 -0800)]
Consolidate ABI warning into type_natural_mode
gcc/
PR target/59794
* config/i386/i386.c (type_natural_mode): Add a bool parameter
to indicate if type is used for function return value. Warn
ABI change if the vector mode isn't available for function
return value.
(ix86_function_arg_advance): Pass false to type_natural_mode.
(ix86_function_arg): Likewise.
(ix86_gimplify_va_arg): Likewise.
(function_arg_32): Don't warn ABI change.
(ix86_function_value): Pass true to type_natural_mode.
(ix86_return_in_memory): Likewise.
(ix86_struct_value_rtx): Removed.
(TARGET_STRUCT_VALUE_RTX): Likewise.
gcc/testsuite/
PR target/59794
* g++.dg/ext/vector23.C: Also prune ABI change for Linux/x86.
* gcc.target/i386/pr39162.c (y): New __m256i variable.
(bar): Change return type to void. Set y to x.
* gcc.target/i386/pr59794-1.c: New testcase.
* gcc.target/i386/pr59794-2.c: Likewise.
* gcc.target/i386/pr59794-3.c: Likewise.
* gcc.target/i386/pr59794-4.c: Likewise.
* gcc.target/i386/pr59794-5.c: Likewise.
* gcc.target/i386/pr59794-6.c: Likewise.
* gcc.target/i386/pr59794-7.c: Likewise.
From-SVN: r206603
Richard Sandiford [Tue, 14 Jan 2014 14:45:07 +0000 (14:45 +0000)]
jump.c (redirect_jump_2): Remove REG_CROSSING_JUMP notes when converting a conditional jump into a...
gcc/
* jump.c (redirect_jump_2): Remove REG_CROSSING_JUMP notes when
converting a conditional jump into a conditional return.
From-SVN: r206600
Richard Biener [Tue, 14 Jan 2014 09:04:50 +0000 (09:04 +0000)]
re PR tree-optimization/58921 (ICE with segfault on valid code at -O3 on x86_64-linux-gnu)
2014-01-14 Richard Biener <rguenther@suse.de>
PR tree-optimization/58921
PR tree-optimization/59006
* tree-vect-loop-manip.c (vect_loop_versioning): Remove code
hoisting invariant stmts.
* tree-vect-stmts.c (vectorizable_load): Insert the splat of
invariant loads on the preheader edge if possible.
* gcc.dg/torture/pr58921.c: New testcase.
* gcc.dg/torture/pr59006.c: Likewise.
* gcc.dg/vect/pr58508.c: XFAIL no longer handled cases.
From-SVN: r206599
Jakub Jelinek [Tue, 14 Jan 2014 09:00:30 +0000 (10:00 +0100)]
re PR testsuite/59494 (FAIL: gfortran.dg/vect/fast-math-mgrid-resid.f scan-tree-dump-times optimized "vect_[^\\n]*\\+ " 13)
PR testsuite/59494
* gfortran.dg/vect/fast-math-mgrid-resid.f: Change
-fdump-tree-optimized to -fdump-tree-pcom-details in dg-options and
cleanup-tree-dump from optimized to pcom. Remove scan-tree-dump-times
for vect_\[^\\n\]*\\+, add scan-tree-dump-times for no suitable chains and
Executing predictive commoning without unrolling.
From-SVN: r206598
Joey Ye [Tue, 14 Jan 2014 07:29:25 +0000 (07:29 +0000)]
plugin.texi (Building GCC plugins): Update to C++.
2014-01-14 Joey Ye <joey.ye@arm.com>
* doc/plugin.texi (Building GCC plugins): Update to C++.
From-SVN: r206597
Kirill Yukhin [Tue, 14 Jan 2014 06:21:06 +0000 (06:21 +0000)]
avx512erintrin.h (_mm_rcp28_round_sd): New.
gcc/
* config/i386/avx512erintrin.h (_mm_rcp28_round_sd): New.
(_mm_rcp28_round_ss): Ditto.
(_mm_rsqrt28_round_sd): Ditto.
(_mm_rsqrt28_round_ss): Ditto.
(_mm_rcp28_sd): Ditto.
(_mm_rcp28_ss): Ditto.
(_mm_rsqrt28_sd): Ditto.
(_mm_rsqrt28_ss): Ditto.
* config/i386/avx512fintrin.h (_mm512_stream_load_si512): Ditto.
* config/i386/i386-builtin-types.def (V8DI_FTYPE_PV8DI): Ditto.
* config/i386/i386.c (IX86_BUILTIN_MOVNTDQA512): Ditto.
(IX86_BUILTIN_RCP28SD): Ditto.
(IX86_BUILTIN_RCP28SS): Ditto.
(IX86_BUILTIN_RSQRT28SD): Ditto.
(IX86_BUILTIN_RSQRT28SS): Ditto.
(bdesc_special_args): Define __builtin_ia32_movntdqa512,
__builtin_ia32_rcp28sd_round, __builtin_ia32_rcp28ss_round,
__builtin_ia32_rsqrt28sd_round, __builtin_ia32_rsqrt28ss_round.
(ix86_expand_special_args_builtin): Expand new FTYPE.
* config/i386/sse.md (define_mode_attr "sse4_1_avx2"): Expand to V8DI.
(srcp14<mode>): Make insn unary.
(avx512f_vmscalef<mode><round_name>): Use substed predicate.
(avx512f_sgetexp<mode><round_saeonly_name>): Ditto.
(avx512f_rndscale<mode><round_saeonly_name>): Ditto.
(<sse4_1_avx2>_movntdqa): Extend to 512 bits.
(avx512er_exp2<mode><mask_name><round_saeonly_name>):
Fix rounding: make it SAE only.
(<mask_codefor>avx512er_rcp28<mode><mask_name><round_saeonly_name>): Ditto.
(<mask_codefor>avx512er_rsqrt28<mode><mask_name><round_saeonly_name>): Ditto.
(avx512er_vmrcp28<mode><round_saeonly_name>): Ditto.
(avx512er_vmrsqrt28<mode><round_saeonly_name>): Ditto.
(avx512f_getmant<mode><mask_name><round_saeonly_name>): Ditto.
* config/i386/subst.md (round_saeonly_mask_scalar_operand3): Remove.
(round_saeonly_mask_scalar_operand4): Ditto.
(round_saeonly_mask_scalar_op3): Ditto.
(round_saeonly_mask_scalar_op4): Ditto.
gcc/testsuite/
* gcc.target/i386/avx-1.c: Fix __builtin_ia32_exp2ps_mask,
__builtin_ia32_exp2pd_mask, __builtin_ia32_rcp28ps_mask,
__builtin_ia32_rcp28pd_mask, __builtin_ia32_rsqrt28ps_mask,
__builtin_ia32_rsqrt28pd_mask. Add __builtin_ia32_rcp28ss_round,
__builtin_ia32_rcp28sd_round, __builtin_ia32_rsqrt28ss_round,
__builtin_ia32_rsqrt28sd_round.
* gcc.target/i386/avx512er-vexp2pd-1.c: Fix rounding mode.
* gcc.target/i386/avx512er-vexp2ps-1.c: Ditto.
* gcc.target/i386/avx512er-vrcp28pd-1.c: Ditto.
* gcc.target/i386/avx512er-vrcp28ps-1.c: Ditto.
* gcc.target/i386/avx512er-vrsqrt28pd-1.c: Ditto.
* gcc.target/i386/avx512er-vrsqrt28ps-1.c: Ditto.
* gcc.target/i386/avx512er-vrcp28sd-1.c: New.
* gcc.target/i386/avx512er-vrcp28sd-2.c: Ditto.
* gcc.target/i386/avx512er-vrcp28ss-1.c: Ditto.
* gcc.target/i386/avx512er-vrcp28ss-2.c: Ditto.
* gcc.target/i386/avx512er-vrsqrt28sd-1.c: Ditto.
* gcc.target/i386/avx512er-vrsqrt28sd-2.c: Ditto.
* gcc.target/i386/avx512er-vrsqrt28ss-1.c: Ditto.
* gcc.target/i386/avx512er-vrsqrt28ss-2.c: Ditto.
* gcc.target/i386/avx512f-vmovntdqa-1.c: Ditto.
* gcc.target/i386/avx512f-vmovntdqa-2.c: Ditto.
* gcc.target/i386/avx512f-vrcp14sd-2.c: Fix.
* gcc.target/i386/avx512f-vrcp14ss-2.c: Ditto.
* gcc.target/i386/sse-22.c: Extend with new built-ins,
fix wrong rounding mode (see above).
* gcc.target/i386/sse-23.c: Ditto.
From-SVN: r206596
GCC Administrator [Tue, 14 Jan 2014 00:16:45 +0000 (00:16 +0000)]
Daily bump.
From-SVN: r206593
Bill Schmidt [Mon, 13 Jan 2014 23:18:05 +0000 (23:18 +0000)]
rs6000-c.c (altivec_resolve_overloaded_builtin): Implement -maltivec=be for vec_insert and vec_extract.
gcc:
2014-01-13 Bill Schmidt <wschmidt@linux.vnet.ibm.com>
* config/rs6000/rs6000-c.c (altivec_resolve_overloaded_builtin):
Implement -maltivec=be for vec_insert and vec_extract.
gcc/testsuite:
2014-01-13 Bill Schmidt <wschmidt@linux.vnet.ibm.com>
* gcc.dg/vmx/insert.c: New.
* gcc.dg/vmx/insert-be-order.c: New.
* gcc.dg/vmx/extract.c: New.
* gcc.dg/vmx/extract-be-order.c: New.
From-SVN: r206590
François Dumont [Mon, 13 Jan 2014 20:07:22 +0000 (20:07 +0000)]
set.h (set): Implement C++11 allocator-aware container requirements.
2014-01-13 François Dumont <fdumont@gcc.gnu.org>
* include/debug/set.h (set): Implement C++11 allocator-aware
container requirements.
* include/debug/map.h (map): Likewise.
* include/debug/multiset.h (multiset): Likewise.
* include/debug/multimap.h (multimap): Likewise.
* include/debug/set.h (set::operator=(set&&)): Add noexcept and
fix implementation regarding management of safe iterators.
* include/debug/map.h (map::operator=(map&&)): Likewise.
* include/debug/multiset.h (multiset::operator=(multiset&&)): Likewise.
* include/debug/multimap.h (multimap::operator=(multimap&&)):
Likewise.
* include/debug/set.h (set::operator=(std::initializer_list<>)):
Rely on the same operator from normal mode.
* include/debug/map.h (map::operator=(std::initializer_list<>)):
Likewise.
* include/debug/multiset.h
(multiset::operator=(std::initializer_list<>)): Likewise.
* include/debug/multimap.h
(multimap::operator=(std::initializer_list<>)): Likewise.
* include/debug/set.h (set::swap(set&)): Add noexcept
specification, add allocator equality check.
* include/debug/map.h (map::swap(map&)): Likewise.
* include/debug/multiset.h (multiset::swap(multiset&)): Likewise.
* include/debug/multimap.h (multimap::swap(multimap&)): Likewise.
From-SVN: r206588
H.J. Lu [Mon, 13 Jan 2014 19:36:17 +0000 (19:36 +0000)]
Make sure that -msse/-mavx are appended at the end
PR libitm/53113
* Makefile.am (x86_sse.lo): Append -msse to CXXFLAGS.
(x86_avx.lo): Append -mavx to CXXFLAGS.
* Makefile.in: Regenerate.
From-SVN: r206587
DJ Delorie [Mon, 13 Jan 2014 19:17:45 +0000 (14:17 -0500)]
msp430.md (call_internal): Don't allow memory references with SP as the base register.
* config/msp430/msp430.md (call_internal): Don't allow memory
references with SP as the base register.
(call_value_internal): Likewise.
* config/msp430/constraints.md (Yc): New. For memory references
that don't use SP as a base register.
* config/msp430/msp430.c (msp430_print_operand): Add 'J' to mean
"an integer without a # prefix"
* config/msp430/msp430.md (epilogue_helper): Use it.
From-SVN: r206586
Jakub Jelinek [Mon, 13 Jan 2014 19:16:30 +0000 (20:16 +0100)]
re PR tree-optimization/59617 ([vectorizer] ICE in vectorizable_mask_load_store with AVX-512F's gathers enabled.)
PR target/59617
* config/i386/i386.c (ix86_vectorize_builtin_gather): Uncomment
AVX512F gather builtins.
* tree-vect-stmts.c (vectorizable_mask_load_store): For now punt
on gather decls with INTEGER_TYPE masktype.
(vectorizable_load): For INTEGER_TYPE masktype, put the INTEGER_CST
directly into the builtin rather than hoisting it before loop.
From-SVN: r206585
Jakub Jelinek [Mon, 13 Jan 2014 19:14:03 +0000 (20:14 +0100)]
re PR tree-optimization/59387 (wrong code (hangs) at -Os on x86_64-linux-gnu)
PR tree-optimization/59387
* tree-scalar-evolution.c: Include gimple-fold.h and gimplify-me.h.
(scev_const_prop): If folded_casts and type has undefined overflow,
use force_gimple_operand instead of force_gimple_operand_gsi and
for each added stmt if it is assign with
arith_code_with_undefined_signed_overflow, call
rewrite_to_defined_overflow.
* tree-ssa-loop-im.c: Don't include gimplify-me.h, include
gimple-fold.h instead.
(arith_code_with_undefined_signed_overflow,
rewrite_to_defined_overflow): Moved to ...
* gimple-fold.c (arith_code_with_undefined_signed_overflow,
rewrite_to_defined_overflow): ... here. No longer static.
Include gimplify-me.h.
* gimple-fold.h (arith_code_with_undefined_signed_overflow,
rewrite_to_defined_overflow): New prototypes.
* gcc.c-torture/execute/pr59387.c: New test.
From-SVN: r206583
Kyrylo Tkachov [Mon, 13 Jan 2014 15:58:47 +0000 (15:58 +0000)]
arm.h (MAX_CONDITIONAL_EXECUTE): Fix typo in description.
2014-01-13 Kyrylo Tkachov <kyrylo.tkachov@arm.com>
* config/arm/arm.h (MAX_CONDITIONAL_EXECUTE): Fix typo in description.
From-SVN: r206580
Richard Biener [Mon, 13 Jan 2014 11:27:43 +0000 (11:27 +0000)]
lto.exp: Do check_effective_target_lto check before adjusting mathlib options.
2014-01-13 Richard Biener <rguenther@suse.de>
* g++.dg/lto/lto.exp: Do check_effective_target_lto check before
adjusting mathlib options.
* gfortran.dg/lto/lto.exp: Likewise.
From-SVN: r206577
Eric Botcazou [Mon, 13 Jan 2014 10:36:27 +0000 (10:36 +0000)]
builtins.c (get_object_alignment_2): Minor tweak.
* builtins.c (get_object_alignment_2): Minor tweak.
* tree-ssa-loop-ivopts.c (may_be_unaligned_p): Rewrite.
From-SVN: r206576
Christian Bruel [Mon, 13 Jan 2014 09:16:48 +0000 (10:16 +0100)]
sh-mem.cc (sh_expand_cmpnstr): Unroll small sizes and optimized non constant lengths.
2014-01-13 Christian Bruel <christian.bruel@st.com>
* config/sh/sh-mem.cc (sh_expand_cmpnstr): Unroll small sizes and
optimized non constant lengths.
From-SVN: r206575
Jakub Jelinek [Mon, 13 Jan 2014 09:13:15 +0000 (10:13 +0100)]
vect-simd-clone-10.c: Add dg-do run.
* gcc.dg/vect/vect-simd-clone-10.c: Add dg-do run.
* gcc.dg/vect/vect-simd-clone-12.c: Likewise.
From-SVN: r206574
Jakub Jelinek [Mon, 13 Jan 2014 07:56:40 +0000 (08:56 +0100)]
re PR libgomp/59194 (tsan detects race for real variables in an OMP reduction clause)
PR libgomp/59194
* omp-low.c (expand_omp_atomic_pipeline): Expand the initial
load as __atomic_load_N if possible.
From-SVN: r206572
GCC Administrator [Mon, 13 Jan 2014 00:17:02 +0000 (00:17 +0000)]
Daily bump.
From-SVN: r206571
Eric Botcazou [Sun, 12 Jan 2014 14:29:12 +0000 (14:29 +0000)]
re PR ada/59772 (floating-point constants are not correctly encoded)
PR ada/59772
* gcc-interface/cuintp.c (build_cst_from_int): Use 32-bit integer type
as intermediate type.
(UI_To_gnu): Likewise.
From-SVN: r206565
Janus Weil [Sun, 12 Jan 2014 11:08:31 +0000 (12:08 +0100)]
re PR fortran/58026 ([F03] Bad error recovery for allocatable component of undeclared type)
2014-01-12 Janus Weil <janus@gcc.gnu.org>
PR fortran/58026
* decl.c (gfc_match_data_decl): Improve error recovery.
2014-01-12 Janus Weil <janus@gcc.gnu.org>
PR fortran/58026
* gfortran.dg/alloc_comp_basics_6.f90: New.
From-SVN: r206564
GCC Administrator [Sun, 12 Jan 2014 00:17:10 +0000 (00:17 +0000)]
Daily bump.
From-SVN: r206563
Steven G. Kargl [Sat, 11 Jan 2014 21:38:30 +0000 (21:38 +0000)]
re PR fortran/59700 (Misleading/buggy runtime error message: Bad integer for item 0 in list input)
2014-01-11 Steven G. Kargl <kargl@gcc.gnu.org>
PR fortran/59700
* gfortran.dg/pr59700.f90: New test.
From-SVN: r206559
Jerry DeLisle [Sat, 11 Jan 2014 21:36:57 +0000 (21:36 +0000)]
Missed file for previous commit.
From-SVN: r206558
Dominique d'Humieres [Sat, 11 Jan 2014 21:30:43 +0000 (22:30 +0100)]
re PR libfortran/59419 (Failing OPEN with FILE='xxx' and IOSTAT creates the file 'xxx' after revision 196783)
2014-01-11 Dominique d'Humieres <dominiq@lps.ens.fr>
* gfortran.dg/binding_label_tests_10_main.f03: Cleanup mod file.
* gfortran.dg/use_only_3.f90: Likewise.
* gfortran.dg/inquire_10.f90: Delete opened file.
* gfortran.dg/inquire_15.f90: Likewise.
* gfortran.dg/pr16597.f90: Likewise.
* gfortran.dg/open_negative_unit_1.f90: Likewise + test
for PR59419.
From-SVN: r206557
David Edelsohn [Sat, 11 Jan 2014 19:25:54 +0000 (19:25 +0000)]
rs6000.c (rs6000_expand_mtfsf_builtin): Remove target parameter.
* config/rs6000/rs6000.c (rs6000_expand_mtfsf_builtin): Remove
target parameter.
(rs6000_expand_builtin): Adjust call.
From-SVN: r206555
David Edelsohn [Sat, 11 Jan 2014 18:57:56 +0000 (18:57 +0000)]
re PR target/58115 (testcase gcc.target/i386/intrinsics_4.c failure)
PR target/58115
* config/rs6000/rs6000.h (SWITCHABLE_TARGET): Define.
* config/rs6000/rs6000.c: Include target-globals.h.
(rs6000_set_current_function): Instead of doing target_reinit
unconditionally, use save_target_globals_default_opts and
restore_target_globals.
* config/rs6000/rs6000-builtin.def (mffs, mtfsf): Add builtins for
FPSCR.
* config/rs6000/rs6000.c (rs6000_expand_mtfsf_builtin): New.
(rs6000_expand_builtin): Handle mffs and mtfsf.
(rs6000_init_builtins): Define mffs and mtfsf.
* config/rs6000/rs6000.md (UNSPECV_MFFS, UNSPECV_MTFSF): New.
(rs6000_mffs): New pattern.
(rs6000_mtfsf): New pattern.
From-SVN: r206554
Jerry DeLisle [Sat, 11 Jan 2014 18:57:20 +0000 (18:57 +0000)]
re PR fortran/59700 (Misleading/buggy runtime error message: Bad integer for item 0 in list input)
2014-01-11 Jerry DeLisle <jvdelisle@gcc.gnu>
Dominique d'Humieres <dominiq@lps.ens.fr>
Steven G. Kargl <kargl@gcc.gnu.org>
PR libfortran/59700
PR libfortran/59764
* io/io.h (struct st_parameter_dt): Assign expanded_read flag to
unused bit. Define new variable line_buffer_pos.
* io/list_read.c (free_saved, next_char, l_push_char,
read_logical, read_real): Replace use of item_count with
line_buffer_pos for line_buffer look ahead.
(read_logical, read_integer, parse_real, read_real, check_type):
Adjust location of free_line to after generating error messages
to retain the correct item count for the message.
Co-Authored-By: Dominique d'Humieres <dominiq@lps.ens.fr>
Co-Authored-By: Steven G. Kargl <kargl@gcc.gnu.org>
From-SVN: r206553
Bin Cheng [Sat, 11 Jan 2014 09:33:39 +0000 (09:33 +0000)]
tree-ssa-loop-ivopts.c (iv_ca_narrow): New parameter.
* tree-ssa-loop-ivopts.c (iv_ca_narrow): New parameter.
Start narrowing with START. Apply candidate-use pair
and check overall cost in narrowing.
(iv_ca_prune): Pass new argument.
From-SVN: r206552
GCC Administrator [Sat, 11 Jan 2014 00:16:58 +0000 (00:16 +0000)]
Daily bump.
From-SVN: r206548
Jeff Law [Fri, 10 Jan 2014 22:13:18 +0000 (15:13 -0700)]
re PR middle-end/59743 (ICE: Segmentation fault)
PR middle-end/59743
* ree.c (combine_reaching_defs): Ensure the defining statement
occurs before the extension when optimizing extensions with
different source and destination hard registers.
PR middle-end/59743
* gcc.c-torture/compile/pr59743.c: New test.
From-SVN: r206545
Jan Hubicka [Fri, 10 Jan 2014 21:34:37 +0000 (22:34 +0100)]
re PR middle-end/58585 (ICE in ipa with virtual inheritance)
PR ipa/58585
* ipa-devirt.c (build_type_inheritance_graph): Also add types of vtables
into the type inheritance graph.
* g++.dg/torture/pr58585.C: New testcase.
From-SVN: r206543
Jakub Jelinek [Fri, 10 Jan 2014 21:27:52 +0000 (22:27 +0100)]
re PR rtl-optimization/59754 ([ree.c] Incorrect merge while working with vector registers)
PR rtl-optimization/59754
* ree.c (combine_reaching_defs): Disallow !SCALAR_INT_MODE_P
modes in the REGNO != REGNO case.
From-SVN: r206542
Bill Schmidt [Fri, 10 Jan 2014 21:26:25 +0000 (21:26 +0000)]
rs6000-builtin.def: Fix pasto for VPKSDUS.
2014-01-10 Bill Schmidt <wschmidt@linux.vnet.ibm.com>
* config/rs6000/rs6000-builtin.def: Fix pasto for VPKSDUS.
From-SVN: r206541
Jakub Jelinek [Fri, 10 Jan 2014 20:37:52 +0000 (21:37 +0100)]
re PR tree-optimization/59745 (internal compiler error: tree check: expected tree that contains 'typed' structure, have '<invalid tree code>' in operand_equal_p, at fold-const.c:2492)
PR tree-optimization/59745
* tree-predcom.c (tree_predictive_commoning_loop): Call
free_affine_expand_cache if giving up because components is NULL.
From-SVN: r206540
Jakub Jelinek [Fri, 10 Jan 2014 20:25:05 +0000 (21:25 +0100)]
target-globals.c (save_target_globals): Allocate < 4KB structs using GC in payload of target_globals struct instead of...
* target-globals.c (save_target_globals): Allocate < 4KB structs using
GC in payload of target_globals struct instead of allocating them on
the heap and the larger structs separately using GC.
* target-globals.h (struct target_globals): Make regs, hard_regs,
reload, expmed, ira, ira_int and lra_fields GTY((atomic)) instead
of GTY((skip)) and change type to void *.
(reset_target_globals): Cast loads from those fields to corresponding
types.
From-SVN: r206539
Hans-Peter Nilsson [Fri, 10 Jan 2014 19:18:34 +0000 (19:18 +0000)]
* gcc.dg/pr46309.c: Disable for cris*-*-*.
From-SVN: r206537
Steve Ellcey [Fri, 10 Jan 2014 17:54:10 +0000 (17:54 +0000)]
re PR plugins/59335 (Plugin doesn't build on trunk)
2014-01-10 Steve Ellcey <sellcey@mips.com>
PR plugins/59335
* Makefile.in (PLUGIN_HEADERS): Add gimplify.h, gimple-iterator.h,
gimple-ssa.h, fold-const.h, tree-cfg.h, tree-into-ssa.h,
tree-ssanames.h, print-tree.h, varasm.h, and context.h.
From-SVN: r206535
Eric Botcazou [Fri, 10 Jan 2014 16:10:37 +0000 (16:10 +0000)]
* gcc.target/arm/neon-nested-apcs.c: Tweak dg directives.
From-SVN: r206532
Richard Earnshaw [Fri, 10 Jan 2014 15:21:21 +0000 (15:21 +0000)]
re PR target/59744 (miscompilation of unsigned comparison on aarch64)
PR target/59744
Fix ChangeLog typos in previous commit (r206529).
From-SVN: r206530
Richard Earnshaw [Fri, 10 Jan 2014 15:12:03 +0000 (15:12 +0000)]
re PR c++/9744 (Explicit parameterization of templated return value fails within templated function)
PR target/9744
gcc:
* aarch64-modes.def (CC_Zmode): New flags mode.
* aarch64.c (aarch64_select_cc_mode): Only allow NEG when the condition
represents an equality.
(aarch64_get_condition_code0): Handle CC_Zmode.
* aarch64.md (compare_neg<mode>): Restrict to equality operations.
gcc/testsuite:
* gcc.target/aarch64/cmn-neg.c: Use equality comparisons.
* gcc.target/aarch64/cmn-neg2.c: New test.
From-SVN: r206529
Andreas Krebbel [Fri, 10 Jan 2014 14:59:41 +0000 (14:59 +0000)]
s390.c (s390_expand_tbegin): Remove jump over CC extraction in good case.
2014-01-10 Andreas Krebbel <Andreas.Krebbel@de.ibm.com>
* config/s390/s390.c (s390_expand_tbegin): Remove jump over CC
extraction in good case.
From-SVN: r206527
Jonathan Wakely [Fri, 10 Jan 2014 14:30:32 +0000 (14:30 +0000)]
re PR libstdc++/59699 (Effective C++ has moved)
PR libstdc++/59699
* doc/xml/manual/support.xml (std.support.types.null): Update links.
From-SVN: r206526
Jonathan Wakely [Fri, 10 Jan 2014 14:30:27 +0000 (14:30 +0000)]
re PR libstdc++/59687 (The description of ios::noreplace is hilarious)
PR libstdc++/59687
* doc/xml/manual/backwards_compatibility.xml
(backwards.third.nocreate_noreplace): Correct and expand.
From-SVN: r206525
Jonathan Wakely [Fri, 10 Jan 2014 14:30:21 +0000 (14:30 +0000)]
re PR libstdc++/59698 (The type of NULL is described elsewhere)
PR libstdc++/59698
* doc/xml/manual/status_cxx1998.xml (iso.1998.specific): Markup
and stylistic improvements.
* doc/xml/manual/codecvt.xml (std.localization.facet.codecvt): Likewise
and update for C++11.
* doc/xml/manual/ctype.xml (std.localization.facet.ctype): Likewise.
From-SVN: r206524
Richard Biener [Fri, 10 Jan 2014 13:07:42 +0000 (13:07 +0000)]
re PR tree-optimization/59374 (-ftree-slp-vectorize breaks unique_ptr's move constructor)
2014-01-10 Richard Biener <rguenther@suse.de>
PR tree-optimization/59374
* tree-vect-slp.c (vect_slp_analyze_bb_1): Move dependence
checking after SLP discovery. Mark stmts not participating
in any SLP instance properly.
* gcc.dg/torture/pr59374-3.c: New testcase.
From-SVN: r206523
Kyrylo Tkachov [Fri, 10 Jan 2014 11:48:37 +0000 (11:48 +0000)]
arm.c (arm_new_rtx_costs): Use destination mode when handling a SET rtx.
2014-01-10 Kyrylo Tkachov <kyrylo.tkachov@arm.com>
* config/arm/arm.c (arm_new_rtx_costs): Use destination mode
when handling a SET rtx.
From-SVN: r206521
Kyrylo Tkachov [Fri, 10 Jan 2014 11:47:04 +0000 (11:47 +0000)]
arm-cores.def (cortex-a53): Specify FL_CRC32.
2014-01-10 Kyrylo Tkachov <kyrylo.tkachov@arm.com>
* config/arm/arm-cores.def (cortex-a53): Specify FL_CRC32.
(cortex-a57): Likewise.
(cortex-a57.cortex-a53): Likewise. Remove redundant flags.
From-SVN: r206520
Kyrylo Tkachov [Fri, 10 Jan 2014 11:43:48 +0000 (11:43 +0000)]
[gcc/testsuite/]
2014-01-10 Kyrylo Tkachov <kyrylo.tkachov@arm.com>
* lib/target-supports.exp
(check_effective_target_arm_crypto_ok_nocache): New.
(check_effective_target_arm_crypto_ok): Use above procedure.
(add_options_for_arm_crypto): Use et_arm_crypto_flags.
From-SVN: r206519
Kyrylo Tkachov [Fri, 10 Jan 2014 11:42:03 +0000 (11:42 +0000)]
arm.c (arm_init_iwmmxt_builtins): Skip non-iwmmxt builtins.
2014-01-10 Kyrylo Tkachov <kyrylo.tkachov@arm.com>
* config/arm/arm.c (arm_init_iwmmxt_builtins): Skip
non-iwmmxt builtins.
From-SVN: r206518
Tom de Vries [Fri, 10 Jan 2014 10:37:22 +0000 (10:37 +0000)]
Note that pass_pre destroys PROP_no_crit_edges
2014-01-09 Tom de Vries <tom@codesourcery.com>
* tree-ssa-pre.c (pass_data_pre): Add comment about PROP_no_crit_edges
in properties_required. Add PROP_no_crit_edges to properties_destroyed.
* tree-ssa-sink.c (pass_data_sink_code): Remove PROP_no_crit_edges
from properties_required.
From-SVN: r206517
Jan Hubicka [Fri, 10 Jan 2014 09:33:24 +0000 (10:33 +0100)]
re PR c++/58252 (ice in gimple_get_virt_method_for_binfo with -O2)
PR ipa/58252
PR ipa/59226
* ipa-devirt.c record_target_from_binfo): Take as argument
stack of binfos and lookup matching one for virtual inheritance.
(possible_polymorphic_call_targets_1): Update.
* g++.dg/ipa/devirt-20.C: New testcase.
* g++.dg/torture/pr58252.C: Likewise.
* g++.dg/torture/pr59226.C: Likewise.
From-SVN: r206516
Max Ostapenko [Fri, 10 Jan 2014 09:16:39 +0000 (11:16 +0200)]
no-asan-stack.c: New test.
2014-01-10 Max Ostapenko <m.ostapenko@partner.samsung.com>
* c-c++-common/asan/no-asan-stack.c: New test.
From-SVN: r206515
Huacai Chen [Fri, 10 Jan 2014 08:51:40 +0000 (08:51 +0000)]
driver-native.c (host_detect_local_cpu): Handle new kernel strings for Loongson-2E/2F/3A.
gcc/
2014-01-10 Huacai Chen <chenhc@lemote.com>
* config/mips/driver-native.c (host_detect_local_cpu): Handle new
kernel strings for Loongson-2E/2F/3A.
From-SVN: r206512
Ian Lance Taylor [Fri, 10 Jan 2014 06:42:08 +0000 (06:42 +0000)]
compiler: Use backend interface for slice info.
From-SVN: r206509
GCC Administrator [Fri, 10 Jan 2014 00:16:59 +0000 (00:16 +0000)]
Daily bump.
From-SVN: r206507
Jonathan Wakely [Thu, 9 Jan 2014 23:39:31 +0000 (23:39 +0000)]
re PR libstdc++/59680 (Compile error in thread.cc)
PR libstdc++/59680
* src/c++11/thread.cc (__sleep_for): Fix call to ::sleep.
From-SVN: r206504
Jakub Jelinek [Thu, 9 Jan 2014 23:39:06 +0000 (00:39 +0100)]
re PR middle-end/59670 (ICE: expected gimple_call(error_mark), have gimple_assign(plus_expr) in gimple_call_internal_p, at gimple.h:2432)
PR middle-end/59670
* tree-vect-data-refs.c (vect_analyze_data_refs): Check
is_gimple_call before calling gimple_call_internal_p.
* gcc.dg/pr59670.c: New test.
From-SVN: r206503
Ian Lance Taylor [Thu, 9 Jan 2014 23:28:10 +0000 (23:28 +0000)]
compiler: Add flattening pass
From-SVN: r206502
Ian Lance Taylor [Thu, 9 Jan 2014 23:16:56 +0000 (23:16 +0000)]
runtime: fix 32-bit malloc for pointers >= 0x80000000
The spans array is allocated in runtime_mallocinit. On a
32-bit system the number of entries in the spans array is
MaxArena32 / PageSize, which (2U << 30) / (1 << 12) == (1 << 19).
So we are allocating an array that can hold 19 bits for an
index that can hold 20 bits. According to the comment in the
function, this is intentional: we only allocate enough spans
(and bitmaps) for a 2G arena, because allocating more would
probably be wasteful.
But since the span index is simply the upper 20 bits of the
memory address, this scheme only works if memory addresses are
limited to the low 2G of memory. That would be OK if we were
careful to enforce it, but we're not. What we are careful to
enforce, in functions like runtime_MHeap_SysAlloc, is that we
always return addresses between the heap's arena_start and
arena_start + MaxArena32.
We generally get away with it because we start allocating just
after the program end, so we only run into trouble with
programs that allocate a lot of memory, enough to get past
address 0x80000000.
This changes the code that computes a span index to subtract
arena_start on 32-bit systems just as we currently do on
64-bit systems.
From-SVN: r206501
Steve Ellcey [Thu, 9 Jan 2014 21:02:43 +0000 (21:02 +0000)]
Makefile.in (TREE_FLOW_H): Remove.
2014-01-09 Steve Ellcey <sellcey@mips.com>
* Makefile.in (TREE_FLOW_H): Remove.
(TREE_SSA_H): Add file names from tree-flow.h.
* doc/tree-ssa.texi (Annotations): Remove reference to tree-flow.h
* tree.h: Remove tree-flow.h reference.
* hash-table.h: Remove tree-flow.h reference.
* tree-ssa-loop-niter.c (dump_affine_iv): Replace tree-flow.h
reference with tree-ssa-loop.h.
From-SVN: r206496
Steve Ellcey [Thu, 9 Jan 2014 21:00:42 +0000 (21:00 +0000)]
delay-slot-1.c: Restrict -mabi=64 to 64 bit processors.
2014-01-09 Steve Ellcey <sellcey@mips.com>
* gcc.dg/delay-slot-1.c: Restrict -mabi=64 to 64 bit processors.
From-SVN: r206495
Bill Schmidt [Thu, 9 Jan 2014 20:30:50 +0000 (20:30 +0000)]
invoke.texi: Add -maltivec={be,le} options...
2014-01-09 Bill Schmidt <wschmidt@linux.vnet.ibm.com>
* doc/invoke.texi: Add -maltivec={be,le} options, and document
default element-order behavior for -maltivec.
* config/rs6000/rs6000.opt: Add -maltivec={be,le} options.
* config/rs6000/rs6000.c (rs6000_option_override_internal): Ensure
that -maltivec={le,be} implies -maltivec; disallow -maltivec=le
when targeting big endian, at least for now.
* config/rs6000/rs6000.h: Add #define of VECTOR_ELT_ORDER_BIG.
From-SVN: r206494
Jakub Jelinek [Thu, 9 Jan 2014 20:12:36 +0000 (21:12 +0100)]
re PR middle-end/47735 (Unnecessary adjustments to stack pointer)
PR middle-end/47735
* cfgexpand.c (expand_one_var): For SSA_NAMEs, if the underlying
var satisfies use_register_for_decl, just take into account type
alignment, rather than decl alignment.
* gcc.target/i386/pr47735.c: New test.
From-SVN: r206493