PR c++/84463
* typeck.c (cp_build_addr_expr_1): Move handling of offsetof-like
tricks from here to ...
* cp-gimplify.c (cp_fold) <case ADDR_EXPR>: ... here. Only use it
if INDIRECT_REF's operand is INTEGER_CST cast to pointer type.
* g++.dg/cpp0x/constexpr-nullptr-1.C: Add -O1 to dg-options.
* g++.dg/cpp0x/constexpr-nullptr-2.C: Expect different diagnostics
in two cases. Uncomment two other tests and add expected dg-error for
them.
* g++.dg/init/struct2.C: Cast to int rather than long to avoid
-Wnarrowing diagnostics on some targets for c++11.
* g++.dg/parse/array-size2.C: Remove xfail.
* g++.dg/cpp0x/constexpr-84463.C: New test.
From-SVN: r259458
tinst_level objects are created during template instantiation, and
they're most often quite short-lived, but since there's no intervening
garbage collection, they accumulate throughout the pass while most by
far could be recycled after a short while. The original testcase in
PR80290, for example, creates almost 55 million tinst_level objects,
all but 10 thousand of them without intervening GC, but we don't need
more than 284 of them live at a time.
Furthermore, in many cases, TREE_LIST objects are created to stand for
the decl in tinst_level. In most cases, those can be recycled as soon
as the tinst_level object is recycled; in some relatively common
cases, they are modified and reused in up to 3 tinst_level objects.
In the same testcase, TREE_LISTs are used in all but 3 thousand of the
tinst_level objects, and we don't need more than 89 tinst_level
objects with TREE_LISTs live at a time. Furthermore, all but 2
thousand of those are created without intervening GC.
This patch arranges for tinst_level objects to be refcounted (I've
seen as many as 20 live references to a single tinst_level object in
my testing), and for pending_template, tinst_level and TREE_LIST
objects that can be recycled to be added to freelists; that's much
faster than ggc_free()ing them and reallocating them shortly
thereafter. In fact, the introduction of such freelists is what makes
this entire patch lighter-weight, when it comes to memory use, and
faster. With refcounting alone, the total memory footprint was still
about the same, and we spent more time.
In order to further reduce memory use, I've arranged for us to create
TREE_LISTs lazily, only at points that really need them (when printing
error messages). This grows tinst_level by an additional pointer, but
since a TREE_LIST holds a lot more than an extra pointer, and
tinst_levels with TREE_LISTs used to be allocated tens of thousands of
times more often than plain decl ones, we still save memory overall.
I was still not quite happy about growing memory use in cases that
used template classes but not template overload resolution, so I
changed the type of the errors field to unsigned short, from int.
With that change, in_system_header_p and refcount move into the same
word or half-word that used to hold errors, releasing a full word,
bringing tinst_level back to its original size, now without any
padding.
The errors field is only used to test whether we've had any errors
since the expansion of some template, to skip the expansion of further
templates. If we get 2^16 errors or more, it is probably reasonable
to refrain from expanding further templates, even if we would expand
them before this change.
With these changes, compile time for the original testcase at -O0,
with default checking enabled, is cut down by some 3.7%, total GCable
memory allocation is cut down by almost 20%, and total memory use (as
reported by GNU time as maxresident) is cut down by slightly over 15%.
for gcc/cp/ChangeLog
PR c++/80290
* cp-tree.h (struct tinst_level): Split decl into tldcl and
targs. Add private split_list_p, tree_list_p, and not_list_p
inline const predicates; to_list private member function
declaration; free public member function declaration; list_p,
get_node and maybe_get_node accessors, and refcount data
member. Narrow errors to unsigned short.
* error.c (print_instantiation_full_context): Use new
accessors.
(print_instantiation_partial_context_line): Likewise. Drop
const from tinst_level-typed parameter.
* mangle.c (mangle_decl_string): Likewise.
* pt.c (freelist): New template class.
(tree_list_freelist_head): New var.
(tree_list_freelist): New fn, along with specializations.
(tinst_level_freelist_head): New var.
(pending_template_freelist_head): Likewise.
(tinst_level_freelist, pending_template_freelist): New fns.
(tinst_level::to_list, tinst_level::free): Define.
(inc_refcount_use, dec_refcount_use): New fns for tinst_level.
(set_refcount_ptr): New template fn.
(add_pending_template): Adjust for refcounting, freelists and
new accessors.
(neglectable_inst_p): Take a NULL d as a non-DECL.
(limit_bad_template_recursion): Use new accessors.
(push_tinst_level): New overload to create the list.
(push_tinst_level_loc): Make it static, split decl into two
args, adjust tests and initialization to cope with split
lists, use freelist, adjust for refcounting.
(push_tinst_level_loc): New wrapper with the old interface.
(pop_tinst_level): Adjust for refcounting.
(record_last_problematic_instantiation): Likewise.
(reopen_tinst_level): Likewise. Use new accessors.
(instantiate_alias_template): Adjust for split list.
(fn_type_unification): Likewise.
(get_partial_spec_bindings): Likewise.
(instantiate_pending_templates): Use new accessors. Adjust
for refcount. Release pending_template to freelist.
(instantiating_current_function_p): Use new accessors.
From-SVN: r259457
Bring in https://golang.org/cl/98616 from gc tip.
Original CL description:
This change modifies Go to disable loading of users' shell history for
TestTerminalSignal tests. TestTerminalSignal, as part of its workload,
will execute a new interactive bash shell. Bash will attempt to load the
user's history from the file pointed to by the HISTFILE environment
variable. For users with large histories that may take up to several
seconds, pushing the whole test past the 5 second timeout and causing
it to fail.
Reviewed-on: https://go-review.googlesource.com/107624
From-SVN: r259452
PR debug/84637
* dbxout.c (dbxout_int): Perform negation in unsigned int type.
(stabstr_D): Change type of unum from unsigned int to
unsigned HOST_WIDE_INT. Perform negation in unsigned HOST_WIDE_INT
type.
From-SVN: r259451
gcc/
PR 84856
* config/riscv/riscv.c (riscv_compute_frame_info): Add calls to
RISCV_STACK_ALIGN when using outgoing_args_size and pretend_args_size.
Set arg_pointer_offset after using pretend_args_size.
From-SVN: r259449
PR sanitizer/85230
* asan.c (handle_builtin_stack_restore): Adjust comment. Emit
__asan_allocas_unpoison call and last_alloca_addr = new_sp before
__builtin_stack_restore rather than after it.
* builtins.c (expand_asan_emit_allocas_unpoison): Pass
arg1 + (virtual_dynamic_stack_rtx - stack_pointer_rtx) as second
argument instead of virtual_dynamic_stack_rtx.
From-SVN: r259446
2018-04-17 Michael Meissner <meissner@linux.vnet.ibm.com>
PR target/85424
* config/rs6000/rs6000.md (pack<mode>): Do not try handle a pack
where the inputs overlap with the output.
From-SVN: r259441
* config/i386/sse.md (vec_extract_lo_<mode><mask_name>): Add
(=v, v) alternative and explicit "memory" attribute.
(vec_extract_lo_<mode><mask_name>): Likewise. Also add
"type", "prefix", "prefix_extra", "length_immediate" and "mode"
attributes.
(vec_extract_lo_<mode><mask_name>): Add (=v, v) alternative and use
"sselog1" type instead of "sselog".
(vec_extract_hi_<mode><mask_name>): Use "sselog1" type instead of
"sselog". Remove explicit "memory" attribute.
(vec_extract_lo_v32hi): Add (=v, v) alternative and explicit "memory",
"type", "prefix", "prefix_extra", "length_immediate" and "mode"
attributes.
(vec_extract_hi_v32hi): Merge all alternatives into one, use
"sselog1" type instead of "sselog". Remove explicit "memory"
attribute.
(vec_extract_hi_v16hi): Merge each pair of alternatives into one,
use "sselog1" type instead of "sselog". Remove explicit "memory"
attribute.
(vec_extract_lo_v64qi): Add (=v, v) alternative and explicit "memory",
"type", "prefix", "prefix_extra", "length_immediate" and "mode"
attributes.
(vec_extract_hi_v64qi): Merge all alternatives into one, use
"sselog1" type instead of "sselog". Remove explicit "memory"
attribute.
(vec_extract_hi_v32qi): Merge each pair of alternatives into one,
use "sselog1" type instead of "sselog". Remove explicit "memory"
attribute.
From-SVN: r259438
2018-04-17 Martin Liska <mliska@suse.cz>
PR lto/85405
* ipa-devirt.c (odr_types_equivalent_p): Remove trailing
in message, remote space in between '_G' and '('.
2018-04-17 Martin Liska <mliska@suse.cz>
PR lto/85405
* g++.dg/lto/pr85405b_0.C: New test.
* g++.dg/lto/pr85405b_1.C: New test.
From-SVN: r259431
PR target/85281
* config/i386/sse.md (reduces<mode><mask_scalar_name>,
avx512f_vmcmp<mode>3<round_saeonly_name>,
avx512f_vmcmp<mode>3_mask<round_saeonly_name>,
avx512f_sgetexp<mode><mask_scalar_name><round_saeonly_scalar_name>,
avx512f_rndscale<mode><round_saeonly_name>,
avx512dq_ranges<mode><mask_scalar_name><round_saeonly_scalar_name>,
avx512f_vgetmant<mode><mask_scalar_name><round_saeonly_scalar_name>):
Use %<iptr>2 instead of %2 for -masm=intel.
(avx512f_vcvtss2usi<round_name>, avx512f_vcvtss2usiq<round_name>,
avx512f_vcvttss2usi<round_saeonly_name>,
avx512f_vcvttss2usiq<round_saeonly_name>): Use %k1 instead of %1 for
-masm=intel.
(avx512f_vcvtsd2usi<round_name>, avx512f_vcvtsd2usiq<round_name>,
avx512f_vcvttsd2usi<round_saeonly_name>,
avx512f_vcvttsd2usiq<round_saeonly_name>, ufloatv2siv2df2<mask_name>):
Use %q1 instead of %1 for -masm=intel.
(avx512f_sfixupimm<mode><sd_maskz_name><round_saeonly_name>,
avx512f_sfixupimm<mode>_mask<round_saeonly_name>): Use %<iptr>3 instead
of %3 for -masm=intel.
(sse2_shufpd_v2df_mask): Fix a typo, change %{6%} to %{%6%} for
-masm=intel.
(*avx512vl_<code>v2div2qi2_store): Use %w0 instead of %0 for
-masm=intel.
(*avx512vl_<code><mode>v4qi2_store): Use %k0 instead of %0 for
-masm=intel.
(avx512vl_<code><mode>v4qi2_mask_store): Use a single pattern with
%k0 and %1 for -masm=intel rather than two patterns, one with %0 and
%g1.
(*avx512vl_<code><mode>v8qi2_store): Use %q0 instead of %0 for
-masm=intel.
(avx512vl_<code><mode>v8qi2_mask_store): Use a single pattern with
%q0 and %1 for -masm=intel rather than two patterns, one with %0 and
%g1 and one with %0 and %1.
(avx512er_vmrcp28<mode><round_saeonly_name>,
avx512er_vmrsqrt28<mode><round_saeonly_name>): Use %<iptr>1 instead of
%1 for -masm=intel.
(avx5124fmaddps_4fmaddps_mask, avx5124fmaddps_4fmaddss_mask,
avx5124fmaddps_4fnmaddps_mask, avx5124fmaddps_4fnmaddss_mask,
avx5124vnniw_vp4dpwssd_mask, avx5124vnniw_vp4dpwssds_mask): Swap order
of %0 and %{%4%} for -masm=intel.
(avx5124fmaddps_4fmaddps_maskz, avx5124fmaddps_4fmaddss_maskz,
avx5124fmaddps_4fnmaddps_maskz, avx5124fmaddps_4fnmaddss_maskz,
avx5124vnniw_vp4dpwssd_maskz, avx5124vnniw_vp4dpwssds_maskz): Swap
order of %0 and %{%5%}%{z%} for -masm=intel.
From-SVN: r259430
2018-04-17 Martin Liska <mliska@suse.cz>
PR ipa/85329
* multiple_target.c (create_dispatcher_calls): Set apostrophes
for target_clone error message. Make default implementation
clone to be a local declaration.
(separate_attrs): Add new argument and check for an empty
string.
(expand_target_clones): Handle it.
(ipa_target_clone): Make redirection just for target_clones
functions.
2018-04-17 Martin Liska <mliska@suse.cz>
PR ipa/85329
* g++.dg/ext/pr85329-2.C: New test.
* g++.dg/ext/pr85329.C: New test.
* gcc.target/i386/mvc12.c: New test.
From-SVN: r259428
Types defined within a __builtin_offsetof argument don't always get
properly recorded as members of their context types, so if they're
anonymous, we may fail to assign them an anon type index for mangling
and ICE.
We shouldn't allow types to be introduced in __builtin_offsetof, I
think, and Jason says the std committee agrees, so I've arranged for
us to reject them.
Even then, we still parse the definitions and attempt to assign
mangled names to its member functions, so the ICE remains. Since
we've already reported an error, we might as well complete the name
assignment with an arbitrary index, thus avoiding the ICE.
We used to have a test that expected to be able to define types in
__builtin_offsetof; this patch removes that specific test.
for gcc/cp/ChangeLog
PR c++/85039
* parser.c (cp_parser_builtin_offset): Reject type definitions.
* mangle.c (nested_anon_class_index): Avoid crash returning -1
if we've seen errors.
for gcc/testsuite/ChangeLog
PR c++/85039
* g++.dg/pr85039-1.C: New.
* g++.dg/pr85039-2.C: New.
* g++.dg/parse/semicolon3.C: Remove test_offsetof.
From-SVN: r259423
2018-04-16 Cesar Philippidis <cesar@codesourcery.com>
Tom de Vries <tom@codesourcery.com>
PR middle-end/84955
* omp-expand.c (expand_oacc_for): Add dummy false branch for
tiled basic blocks without omp continue statements.
* testsuite/libgomp.oacc-c-c++-common/pr84955.c: New test.
* testsuite/libgomp.oacc-fortran/pr84955.f90: New test.
Co-Authored-By: Tom de Vries <tom@codesourcery.com>
From-SVN: r259406
2018-04-16 Aaron Sawdey <acsawdey@linux.ibm.com>
PR target/83660
* config/rs6000/rs6000-c.c (altivec_resolve_overloaded_builtin): Mark
vec_extract expression as having side effects to make sure it gets
a cleanup point.
2018-04-16 Aaron Sawdey <acsawdey@linux.ibm.com>
PR target/83660
* gcc.target/powerpc/pr83660.C: New test.
From-SVN: r259403
PR target/84945
* config/i386/cpuinfo.c (set_feature): Wrap into do while (0) to avoid
-Wdangling-else warnings. Mask shift counts to avoid
-Wshift-count-negative and -Wshift-count-overflow false positives.
From-SVN: r259398
2018-04-16 Paolo Carlini <paolo.carlini@oracle.com>
Jason Merrill <jason@redhat.com>
PR c++/85112
* convert.c (convert_to_integer_1): Use direct recursion for
enumeral types and types with a precision less than the number
of bits in their mode.
/testsuite
2018-04-16 Paolo Carlini <paolo.carlini@oracle.com>
Jason Merrill <jason@redhat.com>
PR c++/85112
* g++.dg/cpp0x/pr85112.C: New.
Co-Authored-By: Jason Merrill <jason@redhat.com>
From-SVN: r259397
The -mno-direct-move option causes a lot of problems, since it forces
us to be able to generate code for p8 and up with some crucial
instructions missing. This patch removes the -m[no-]direct-move
options so that the user cannot put us into this unexpected situation
anymore. Internally we still have all the same flags, and they are
automatically set based on -mcpu; getting rid of that is a lot more
work and will have to wait for GCC 9 (in some places the flag is used
to see if we are compiling for a p8 _at all_).
PR target/85293
* config/rs6000/rs6000.opt (mdirect-move): Make deprecated.
* doc/invoke.texi (RS/6000 and PowerPC Options): Remove -mdirect-move
and -mno-direct-move.
gcc/testsuite/
PR target/85293
* gcc.target/powerpc/pr80098-2.c: Remove -mdirect-move. Remove the
corresponding dg-error clause.
* gcc.target/powerpc/pr80098-3.c: Ditto.
* gcc.target/powerpc/pr80103-1.c: Delete.
From-SVN: r259386
gcc/fortran/ChangeLog:
2018-04-14 Andre Vehreschild <vehre@gcc.gnu.org>
PR fortran/81773
PR fortran/83606
* dependency.c (gfc_dep_resolver): Coarray indexes are to be ignored
during dependency computation. They define no data dependency.
* trans-array.c (conv_array_index_offset): The stride can not be set
here, prevent fail.
* trans-intrinsic.c (conv_caf_send): Add creation of temporary array
for caf_get's result and copying to the array with vectorial
indexing.
gcc/testsuite/ChangeLog:
2018-04-14 Andre Vehreschild <vehre@gcc.gnu.org>
PR fortran/81773
PR fortran/83606
* gfortran.dg/coarray/get_to_indexed_array_1.f90: New test.
* gfortran.dg/coarray/get_to_indirect_array.f90: New test.
From-SVN: r259385
2018-04-14 Thomas Koenig <tkoenig@gcc.gnu.org>
PR fortran/85387
* frontend-passes.c (traverse_io_block): Check for start, end or
stride being defined by an outer implied DO loop.
2018-04-14 Thomas Koenig <tkoenig@gcc.gnu.org>
PR fortran/85387
* gfortran.dg/implied_do_io_5.f90: New test.
From-SVN: r259384
The powerpc versions of _mm_slli_epi32 and __mm_slli_epi64 in emmintrin.h
do not properly handle shift values between 16 and 31, inclusive.
These are setting up the shift with vec_splat_s32, which only accepts
*5 bit signed* shift values, or a range of -16 to 15. Values above 15
produce an error:
error: argument 1 must be a 5-bit signed literal
Fix is to effectively reduce the range for which vec_splat_s32 is used
to < 32 and use vec_splats otherwise.
Also, __mm_slli_epi{16,32,64}, when given a negative shift value,
should always return a vector of {0}.
PR target/83402
* config/rs6000/emmintrin.h (_mm_slli_epi{16,32,64}):
Ensure that vec_splat_s32 is only called with 0 <= shift < 16.
Ensure negative shifts result in {0}.
gcc/testsuite/
PR target/83402
* gcc.target/powerpc/sse2-psllw-1.c: Refactor and add tests for
several values: positive, negative, and zero.
* gcc.target/powerpc/sse2-pslld-1.c: Same.
* gcc.target/powerpc/sse2-psllq-1.c: Same.
From-SVN: r259380
2018-04-13 Vladimir Makarov <vmakarov@redhat.com>
PR rtl-optimization/79916
* config/rs6000/rs6000.c (rs6000_emit_move): Use assigned hard
regs (if any) to define how to gnerate SD moves when LRA is in
progress.
2018-04-13 Vladimir Makarov <vmakarov@redhat.com>
PR rtl-optimization/79916
* gcc.target/powerpc/pr79916.c: New.
From-SVN: r259379
PR rtl-optimization/85393
* except.h (expand_dw2_landing_pad_for_region): Remove declaration.
* except.c (expand_dw2_landing_pad_for_region): Make static.
* bb-reorder.c (fix_up_crossing_landing_pad): In new_bb emit just
a label and unconditional jump to old_bb, rather than
expand_dw2_landing_pad_for_region insn(s) and jump to single_succ
basic block.
* g++.dg/opt/pr85393.C: New test.
* g++.dg/opt/pr85393-aux.cc: New file.
From-SVN: r259378
PR rtl-optimization/85376
* simplify-rtx.c (simplify_const_unary_operation): For CLZ and CTZ and
zero op0, if C?Z_DEFINED_VALUE_AT_ZERO is false, return NULL_RTX
instead of a specific value.
* gcc.dg/pr85376.c: New test.
From-SVN: r259377
