2016-05-11 Richard Biener <rguenther@suse.de>
PR middle-end/71002
* alias.c (reference_alias_ptr_type): Preserve alias-set zero
if the langhook insists on it.
* fold-const.c (make_bit_field_ref): Add arg for the original
reference and preserve its alias-set.
(decode_field_reference): Take exp by reference and adjust it
to the original memory reference.
(optimize_bit_field_compare): Adjust callers.
(fold_truth_andor_1): Likewise.
* gimplify.c (gimplify_expr): Adjust in-SSA form test.
* g++.dg/torture/pr71002.C: New testcase.
From-SVN: r236117
Revision 235794 regressed compat/scalar-by-value-6 for powerpc-linux
-m32 due to accidentally changing the ABI. By another historical
accident, complex long double is stupidly passed in gprs for -m32.
* config/rs6000/rs6000.c (is_complex_IBM_long_double,
abi_v4_pass_in_fpr): New functions.
(rs6000_function_arg_boundary): Exclude complex IBM long double
from 64-bit alignment when ABI_V4.
(rs6000_function_arg, rs6000_function_arg_advance_1,
rs6000_gimplify_va_arg): Use abi_v4_pass_in_fpr.
From-SVN: r236111
If we have a conditional jump that has only a return in both the branch
path and the fallthrough path, and the return on the branch path can not
be made a conditional return, we will try to make a conditional return
from the fallthrough path, and that does not work because we then try
to redirect the (new) jump in the fallthrough block to the original
dest in the branch path, which is the exit block.
For the testcase on ARM we end up in this situation because before the
jump2 pass there are some other insns in the return blocks as well, but
the same insns in both, so those are moved above the conditional jump.
Only later (in the ce3 pass) are the conditional jump and two returns
melded into one return, so we need to handle this strange case here.
PR rtl-optimization/71028
* cfgcleanup.c (try_optimize_cfg): Do not flip a conditional
jump with just a return in the fallthrough block if the branch
block contains just a returns as well.
From-SVN: r236106
read-md.c and read-rtl.c repeatedly use this pattern:
c = read_skip_spaces ();
if (c != ')')
fatal_expected_char (')', c);
Simplify them by introduce a helper function to do this.
gcc/ChangeLog:
* read-md.c (require_char_ws): New function.
(read_string): Simplify using require_char_ws.
(handle_constants): Likewise.
(handle_enum): Likewise.
(handle_file): Likewise.
* read-md.h (require_char_ws): New declaration.
* read-rtl.c (read_conditions): Simplify using require_char_ws.
(read_mapping): Likewise.
(read_rtx_code): Likewise.
(read_nested_rtx): Likewise.
From-SVN: r236101
* config/i386/i386.c (legitimize_pic_address): Merge 64-bit and 32-bit
gotoff_operand code paths. Use copy_to_suggested_regs and
expand_simple_binop where appropriate. Cleanup.
From-SVN: r236096
When the DWARF back-end generates DW_OP_neg operations in DWARF
procedures, we get an ICE because of inconsistent stack usage
computation for the embedding expression. This is because
resolve_args_picking_1 thinks DW_OP_neg is a binary operation (pops 2
stack slots, pushes 1) whereas it really is an unary one (one pop, one
push).
This change fixes resolve_args_picking_1 and adds a regression testcase
(which crashes with the current trunk). Bootstrapped and regtested
without regression on x86_64-linux.
gcc/
* dwarf2out.c (resolve_args_picking_1): Consider DW_OP_neg as an
unary operation, not a binary one.
gcc/testsuite/
* gnat.dg/debug6.adb, gnat.dg/debug6_pkg.ads: New testcase.
From-SVN: r236087
* include/experimental/bits/fs_dir.h (begin, end): Add noexcept.
* testsuite/experimental/filesystem/iterators/directory_iterator.cc:
Test begin and end functions.
* testsuite/experimental/filesystem/iterators/
recursive_directory_iterator.cc: Likewise.
From-SVN: r236085
PR target/70927
* config/i386/sse.md (<sse>_andnot<mode>3<mask_name>),
*<code><mode>3<mask_name>): For !TARGET_AVX512DQ and EVEX encoding,
use vp*[dq] instead of v*p[sd] instructions and adjust mode attribute
accordingly.
* gcc.target/i386/avx512vl-logic-1.c: New test.
* gcc.target/i386/avx512vl-logic-2.c: New test.
* gcc.target/i386/avx512dq-logic-2.c: New test.
From-SVN: r236083
When debugging diagnostic-show-locus.c, it's invaluable to have a
"ruler" showing column numbers.
This patch adds in support via a new "show_ruler_p" flag within
the diagnostic_context. There's no direct way for end-users to enable
this, but plugins can enable it by setting the flag, so the
plugin that tests the diagnostic subsystem uses this to verify that
the ruler is correctly printed.
gcc/ChangeLog:
* diagnostic-show-locus.c (layout::layout): Call show_ruler
if show_ruler_p was set on the context.
(layout::show_ruler): New method.
* diagnostic.h (struct diagnostic_context): Add field
"show_ruler_p".
gcc/testsuite/ChangeLog:
* gcc.dg/plugin/diagnostic-test-show-locus-bw.c
(test_very_wide_line): Add ruler to expected output.
* gcc.dg/plugin/diagnostic-test-show-locus-color.c
(test_very_wide_line): Likewise.
* gcc.dg/plugin/diagnostic_plugin_test_show_locus.c
(test_show_locus): Within the handling of "test_very_wide_line",
enable show_ruler_p on the diagnostic context.
From-SVN: r236080
2016-05-10 Richard Biener <rguenther@suse.de>
PR tree-optimization/71039
* tree-ssa-phiprop.c: Include tree-ssa-loop.h.
(chk_uses): New function.
(propagate_with_phi): Verify we can safely replicate the lhs of an
aggregate assignment on all incoming edges.
* gcc.dg/torture/pr71039.c: New testcase.
From-SVN: r236079
Add rudimentary support for atomics on RX. It is implemented by flipping
interrupts off/on around the atomic sequences.
gcc/
* config/rx/rx-protos.h (is_interrupt_func, is_fast_interrupt_func):
Forward declare.
(rx_atomic_sequence): New class.
* config/rx/rx.c (rx_print_operand): Use symbolic names for PSW bits.
(is_interrupt_func, is_fast_interrupt_func): Make non-static and
non-inline.
(rx_atomic_sequence::rx_atomic_sequence,
rx_atomic_sequence::~rx_atomic_sequence): New functions.
* config/rx/rx.md (CTRLREG_PSW, CTRLREG_USP, CTRLREG_FPSW, CTRLREG_CPEN,
CTRLREG_BPSW, CTRLREG_BPC, CTRLREG_ISP, CTRLREG_FINTV,
CTRLREG_INTB): New constants.
(FETCHOP): New code iterator.
(fethcop_name, fetchop_name2): New iterator code attributes.
(QIHI): New mode iterator.
(atomic_exchange<mode>, atomic_exchangesi, xchg_mem<mode>,
atomic_fetch_<fetchop_name>si, atomic_fetch_nandsi,
atomic_<fetchop_name>_fetchsi, atomic_nand_fetchsi): New patterns.
From-SVN: r236075
PR c/70255
* c-decl.c (diagnose_mismatched_decls): Warn for optimize attribute
on a declaration following the definition.
* gcc.dg/attr-opt-1.c: New test.
From-SVN: r236071
Although the scalar variants of the vector instructions aren't
actually vector instructions they are still executed in the vector
facility and therefore need to be disabled when disabling the facility
with -mno-vx.
Fixed with the attached patch. Committed to head, GCC 6, and GCC 5
branches.
gcc/ChangeLog:
2016-05-10 Andreas Krebbel <krebbel@linux.vnet.ibm.com>
* config/s390/s390.md ("*vec_cmp<insn_cmp>df_cconly")
("*fixuns_truncdfdi2_z13")
("*fixuns_trunc<FP:mode><GPR:mode>2_z196")
("*fix_truncdfdi2_bfp_z13", "*floatunsdidf2_z13")
("*extendsfdf2_z13"): Replace TARGET_Z13 with TARGET_VX.
From-SVN: r236067
Track from which abstract lexical block concrete ones come from in DWARF
so that debuggers can inherit the former from the latter. This enables
debuggers to properly handle the following case:
* function Child2 is nested in a lexical block, itself nested in
function Child1;
* function Child1 is inlined into some call site;
* function Child2 is never inlined.
Here, Child2 is described in DWARF only in the abstract instance of
Child1. So when debuggers decode Child1's concrete instances, they need
to fetch the definition for Child2 in the corresponding abstract
instance: the DW_AT_abstract_origin link on the lexical block that
embeds Child1 enables them to do that.
Bootstrapped and regtested on x86_64-linux.
gcc/ChangeLog:
* dwarf2out.c (add_abstract_origin_attribute): Adjust
documentation comment. For BLOCK nodes, add a
DW_AT_abstract_origin attribute that points to the DIE generated
for the origin BLOCK.
(gen_lexical_block_die): Call add_abstract_origin_attribute for
blocks from inlined functions.
gcc/testsuite/Changelog:
* gcc.dg/debug/dwarf2/nested_fun.c: New testcase.
From-SVN: r236065
libgcc/
PR libgcc/70720
* config.host (moxie-*-rtems*): Merge this stanza with other moxie
targets so the same extra_parts are built. Also have tmake_file add
on to its value rather than override.
From-SVN: r236064
* config/rs6000/rs6000.c (rs6000_reassociation_width): Add
function for TARGET_SCHED_REASSOCIATION_WIDTH to enable
parallel reassociation for power8 and forward.
From-SVN: r236043
* config/i386/i386.md (absneg splitters with general regs): Use
general_reg_operand predicate.
(btsq peephole2): Use x86_64_immediate_operand to check if new
value is suitable for immediate operand. Generate emitted insn
using RTL expressions.
(btcq peephole2): Ditto.
(btrq peephole2): Ditto. Generate correct immediate operand
for AND masking.
testsuite/ChangeLog:
* gcc.target/i386/fabsneg-1.c New test.
From-SVN: r236042
expand_debug_expr handled negative bit positions using:
else if (bitpos < 0)
{
HOST_WIDE_INT units
= (-bitpos + BITS_PER_UNIT - 1) / BITS_PER_UNIT;
op0 = adjust_address_nv (op0, mode1, units);
bitpos += units * BITS_PER_UNIT;
}
Here "units" is the negative of the (negative) byte offset, so I think
we should be offsetting OP0 by -units instead. E.g. a bitpos of -17
would give units==3, so this code would move OP0 up by 3 bytes and set
bitpos to 7, giving a total bitpos of 31.
Just noticed by inspection. An assert triggered for:
gcc.target/i386/mpx/bitfields-1-lbv.c
gcc.target/i386/mpx/field-addr-7-lbv.c
gcc.target/i386/mpx/reference-3-lbv.cpp
gcc.target/i386/mpx/reference-4-lbv.cpp
at -m32 but otherwise this case doesn't seem to trigger during a
bootstrap and regtest.
Tested on x86_64-linux-gnu.
gcc/
* cfgexpand.c (expand_debug_expr): Fix address offset for negative
bitpos.
From-SVN: r236041
wide_int_constant_multiple_p used:
if (*mult_set && mult != 0)
return false;
to check whether we had previously seen a nonzero multiple, but "mult" is
a pointer to the previous value rather than the previous value itself.
Noticed by inspection while working on another patch, so I don't have a
testcase. I tried adding an assert for combinations that were wrongly
rejected before but it didn't trigger during a bootstrap and regtest.
Tested on x86_64-linux-gnu.
gcc/
* tree-affine.c (wide_int_constant_multiple_p): Add missing
pointer dereference.
From-SVN: r236040
