Michael Meissner [Mon, 22 May 2017 22:44:45 +0000 (22:44 +0000)]
re PR target/80718 (GCC generates slow code for offsettable vec_duplicate)
[gcc]
2017-05-22 Michael Meissner <meissner@linux.vnet.ibm.com>
PR target/80718
* config/rs6000/vsx.md (vsx_splat_<mode>, VSX_D iterator): Split
V2DF/V2DI splat into two separate patterns, one that handles
registers, and the other that only handles memory. Drop support
for splatting from a GPR on ISA 2.07 and then splitting the
splat into direct move and splat.
(vsx_splat_<mode>_reg): Likewise.
(vsx_splat_<mode>_mem): Likewise.
[gcc/testsuite]
2017-05-22 Michael Meissner <meissner@linux.vnet.ibm.com>
PR target/80718
* gcc.target/powerpc/pr80718.c: New test.
The function bb_is_just_return finds if the BB it is asked about does
just a return and nothing else. It currently does not allow clobbers
in the block either, which we of course can allow just fine.
Jakub Jelinek [Mon, 22 May 2017 18:55:29 +0000 (20:55 +0200)]
c-fold.c (c_fully_fold_internal): Save the c_fully_fold_internal result for SAVE_EXPR operand and set...
* c-fold.c (c_fully_fold_internal): Save the c_fully_fold_internal
result for SAVE_EXPR operand and set SAVE_EXPR_FOLDED_P even if
it returned invariant. Call tree_invariant_p unconditionally
afterwards to decide whether to return expr or op0.
Jakub Jelinek [Mon, 22 May 2017 18:54:54 +0000 (20:54 +0200)]
re PR middle-end/80809 (Multi-free error for variable size array used within OpenMP task)
PR middle-end/80809
* omp-low.c (finish_taskreg_remap): New function.
(finish_taskreg_scan): If unit size of ctx->record_type
is non-constant, unshare the size expression and replace
decls in it with possible outer var refs.
* testsuite/libgomp.c/pr80809-2.c: New test.
* testsuite/libgomp.c/pr80809-3.c: New test.
Jakub Jelinek [Mon, 22 May 2017 18:54:05 +0000 (20:54 +0200)]
re PR middle-end/80809 (Multi-free error for variable size array used within OpenMP task)
PR middle-end/80809
* gimplify.c (omp_add_variable): For GOVD_DEBUG_PRIVATE use
GOVD_SHARED rather than GOVD_PRIVATE with it.
(gimplify_adjust_omp_clauses_1, gimplify_adjust_omp_clauses): Expect
GOVD_SHARED rather than GOVD_PRIVATE with GOVD_DEBUG_PRIVATE.
Jakub Jelinek [Mon, 22 May 2017 18:51:54 +0000 (20:51 +0200)]
re PR middle-end/80853 (OpenMP ICE in build_outer_var_ref with array reduction)
PR middle-end/80853
* omp-low.c (lower_reduction_clauses): Pass OMP_CLAUSE_PRIVATE
as last argument to build_outer_var_ref for pointer bases of array
section reductions.
Nathan Sidwell [Mon, 22 May 2017 11:16:49 +0000 (11:16 +0000)]
name-lookup.c (pushdecl_top_level) pushdecl_top_level_and_finish): Move after namespace pushing and popping functions.
* name-lookup.c (pushdecl_top_level)
pushdecl_top_level_and_finish): Move after namespace pushing and
popping functions.
(push_to_top_level): Rename to ...
(do_push_to_top_level): ... here. Remove timing code.
(pop_from_top_level_1): Rename to ...
(do_pop_from_top_level): ... here.
(do_push_nested_namespace, do_pop_nested_namespace)
(push_to_top_level): New wrapper for do_push_to_top_level.
(pop_from_top_level): Adjust.
(push_nested_namepace, pop_nested_namespace): Wrappers for workers.
(--This line, and those below, will be ignored--
Jan Hubicka [Mon, 22 May 2017 10:39:46 +0000 (12:39 +0200)]
ipa-inline-analysis.c (inline_summary::reset): Do not reset self_time.
* ipa-inline-analysis.c (inline_summary::reset): Do not reset
self_time.
(dump_inline_summary): Do not print self_time.
(estimate_function_body_sizes): Do not set self_time.
(compute_inline_parameters): Likewise.
(inline_read_section, inline_write_summary): Do not stream self_time.
* ipa-inline.h (inline_summary): Drop self_time.
Eric Botcazou [Mon, 22 May 2017 09:38:09 +0000 (09:38 +0000)]
decl.c (gnat_to_gnu_entity): Skip regular processing for Itypes that are E_Access_Subtype.
* gcc-interface/decl.c (gnat_to_gnu_entity): Skip regular processing
for Itypes that are E_Access_Subtype.
<E_Access_Subtype>: Use the DECL of the base type directly.
Eric Botcazou [Mon, 22 May 2017 09:24:24 +0000 (09:24 +0000)]
sem_ch4.adb (Analyze_Call): In Ada2012 an incomplete type from a limited view may appear in the profile of...
* sem_ch4.adb (Analyze_Call): In Ada2012 an incomplete type from a
limited view may appear in the profile of a function, and a call to
that function in another unit in which the full view is available must
use this full view to spurious type errors at the point of call.
* inline.adb (Analyze_Inlined_Bodies): Remove restriction on loading
of parent body with a with clause for the main unit.
* gcc-interface/decl.c (defer_limited_with_list): Document new usage.
(gnat_to_gnu_entity) <E_Access_Type>: Handle completed Taft Amendment
types declared in external units like types from limited with clauses.
Adjust final processing of defer_limited_with_list accordingly.
* gcc-interface/trans.c (gnat_to_gnu) < N_Selected_Component>: Try
again to translate the prefix after the field if it is incomplete.
Paolo Carlini [Sun, 21 May 2017 22:20:05 +0000 (22:20 +0000)]
re PR c++/70265 (ICE on code with constexpr on x86_64-linux-gnu in tree check: expected statement_list, have nop_expr in tsi_start, at tree-iterator.h:42)
2017-05-21 Paolo Carlini <paolo.carlini@oracle.com>
Jerry DeLisle [Fri, 19 May 2017 15:48:35 +0000 (15:48 +0000)]
[multiple changes]
2017-05-19 Paul Thomas <pault@gcc.gnu.org>
PR fortran/80333
* trans-io.c (nml_get_addr_expr): If we are dealing with class
type data set tmp tree to get that address.
(transfer_namelist_element): Set the array spec to point to the
the class data.
2017-05-19 Paul Thomas <pault@gcc.gnu.org>
Jerry DeLisle <jvdelisle@gcc.gnu.org>
PR fortran/80333
* list_read.c (nml_read_obj): Compute pointer into class/type
arrays from the nl->dim information. Update it for each iteration
of the loop for the given object.
2017-05-19 Jerry DeLisle <jvdelisle@gcc.gnu.org>
PR libgfortran/80333
* gfortran.dg/dtio_30.f03: New test.
Thomas Schwinge [Fri, 19 May 2017 13:32:19 +0000 (15:32 +0200)]
OpenACC default clause maintenance
gcc/cp/
* parser.c (cp_parser_omp_clause_default): Avoid printing more
than one syntax error message.
gcc/testsuite/
* c-c++-common/goacc/default-1.c: New file.
* c-c++-common/goacc/default-2.c: Likewise.
* c-c++-common/goacc/data-default-1.c: Remove file, including its
test cases in...
* c-c++-common/goacc/default-3.c: ... this new file, and...
* c-c++-common/goacc/default-4.c: ... this new file.
* gfortran.dg/goacc/default.f95: Remove file, including its test
cases in...
* gfortran.dg/goacc/default-1.f95: ... this new file, and...
* gfortran.dg/goacc/default-2.f: ... this new file.
* gfortran.dg/goacc/default-3.f95: New file.
* gfortran.dg/goacc/default-4.f: Likewise.
Martin Liska [Fri, 19 May 2017 13:31:01 +0000 (15:31 +0200)]
Enable by default -Werror just for bootstrap-debug.mk and default build.
2017-05-19 Martin Liska <mliska@suse.cz>
* configure.ac: Add --enable-werror-always just for
bootstrap-debug (or none build config) and do not add it
when we are on a release branch.
* configure: Regenerated.
gcc/testsuite/
* gcc.target/arm/tls-disable-literal-pool.c: Change
require-effective-target to tls_native.
Move dg-error to return statement line and change to dg-message.
If we cannot load a constant into a register in one insn, and that
constant is a valid mask (for rotate instructions), we currently
prefer to load -1 and then mask it. This patch makes us not do that if
instead we could use two add or or instructions, since those are
sometimes faster on certain CPUs (and never are slower).
* config/rs6000/rs6000.md (splitter to load of -1 and mask): Don't
use this splitter if two add or or instructions would also work for
the constant we want to generate.
Richard Biener [Fri, 19 May 2017 08:23:37 +0000 (08:23 +0000)]
re PR middle-end/80764 (ICE at -O3 in both 32-bit and 64-bit modes on x86_64-linux-gnu (internal compiler error: in verify_loop_structure, at cfgloop.c:1644))
This patch implements separate shrink-wrapping for the non-volatile
floating point registers (FPR14..FPR31). It is pretty much exactly
analogous to the code for GPRs.
* config/rs6000/rs6000.c (struct machine_function): Add field
fpr_is_wrapped_separately.
(rs6000_get_separate_components): Use 64 components. Handle the
new FPR components.
(rs6000_components_for_bb): Handle the FPR components.
(rs6000_emit_prologue_components): Handle the FPR components.
(rs6000_emit_epilogue_components): Handle the FPR components.
(rs6000_set_handled_components): Handle the FPR components.
(rs6000_emit_prologue): Don't output prologue code for those FPRs
that are already separately shrink-wrapped.
(rs6000_emit_epilogue): Don't output epilogue code for those FPRs
that are already separately shrink-wrapped.
Michael Meissner [Thu, 18 May 2017 19:34:13 +0000 (19:34 +0000)]
re PR target/80510 (Optimize Power7/power8 Altivec load/stores)
[gcc]
2017-05-18 Michael Meissner <meissner@linux.vnet.ibm.com>
PR target/80510
* config/rs6000/predicates.md (simple_offsettable_mem_operand):
New predicate.
* config/rs6000/rs6000.md (ALTIVEC_DFORM): New iterator.
(define_peephole2 for Altivec d-form load): Add peepholes to catch
cases where the register allocator uses a move and an offsettable
memory operation to/from a FPR register on ISA 2.06/2.07.
(define_peephole2 for Altivec d-form store): Likewise.
[gcc/testsuite]
2017-05-18 Michael Meissner <meissner@linux.vnet.ibm.com>
PR target/80510
* gcc.target/powerpc/pr80510-1.c: New test.
* gcc.target/powerpc/pr80510-2.c: Likewise.
Matthias Klose [Thu, 18 May 2017 18:49:16 +0000 (18:49 +0000)]
gcc_release (build_gzip): Build xz tarball instead of bz2 tarball.
2017-05-18 Matthias Klose <doko@ubuntu.com>
* gcc_release (build_gzip): Build xz tarball instead of bz2 tarball.
(build_diffs): Handle building diffs from either bz2 or xz tarballs,
compress diffs using xz instead of bz2.
(build_diff): Likewise.
(upload_files): Check for *.xz files instead of *.bz2 files.
(announce_snapshot): Announce xz tarball instead of bz2 tarball.
(XZ): New definition.
(<toplevel>): Look for both bz2 and xz compressed old tarballs.
Ian Lance Taylor [Thu, 18 May 2017 18:04:29 +0000 (18:04 +0000)]
compiler: ignore struct field tags for type conversion
Go 1.8 includes a language change (https://golang.org/doc/go1.8#language):
in an explicit conversion from one struct type to another, any field
tags are ignored.
This CL implements this language change in the gofrontend. The tests
for this are in the gc testsuite, which will be copied into the gccgo
repository in due course.
Jonathan Wakely [Thu, 18 May 2017 17:32:06 +0000 (18:32 +0100)]
PR libstdc++/80478 make std::mem_fn work with noexcept functions
PR libstdc++/80478
* include/std/functional (_Mem_fn_traits_base): Add specializations
for noexcept member function types.
* testsuite/20_util/function_objects/mem_fn/80478.cc: New test.
Wilco Dijkstra [Thu, 18 May 2017 15:58:33 +0000 (15:58 +0000)]
Add an assert to leaf_function_p to ensure it is not called from a prolog or...
Add an assert to leaf_function_p to ensure it is not called from a
prolog or epilog sequence (which would incorrectly return true in a
non-leaf function). There are several targets which still call
leaf_function_p, and while most appear safe or appear aware of the
issue, it is likely not all such calls are safe. This check enables
any such latent bugs to be found.
gcc/
* final.c (leaf_function_p): Check we are not in a sequence.
Marek Polacek [Thu, 18 May 2017 11:26:25 +0000 (11:26 +0000)]
c-common.c (c_common_type_for_size): Use NULL_TREE instead of 0.
* c-common.c (c_common_type_for_size): Use NULL_TREE instead of 0.
(c_common_fixed_point_type_for_size): Likewise.
(c_common_type_for_mode): Likewise.
(shorten_compare): Likewise.
(c_promoting_integer_type_p): Use false/true instead of 0/1.
* c-pragma.c (maybe_apply_renaming_pragma): Use NULL_TREE instead of 0.
Marek Polacek [Thu, 18 May 2017 10:55:59 +0000 (10:55 +0000)]
c-common.c (self_promoting_args_p): Change the return type to bool.
* c-common.c (self_promoting_args_p): Change the return type to bool.
Use false/true instead of 0/1.
* c-common.h (self_promoting_args_p): Update.
* c-decl.c (start_decl): Use false/true instead of 0/1.
(grokdeclarator): Likewise.
(finish_struct): Likewise.
(start_function): Change the return type to bool. Use false/true
instead of 0/1.
(declspecs_add_qual): Use UNKNOWN_LOCATION instead of 0.
* c-tree.h (start_function): Update.
* c-typeck.c (same_translation_unit_p): Change the return type to bool.
(set_designator): Change the return type to bool. Use false/true
instead of 0/1.
Sheldon Lobo [Thu, 18 May 2017 09:34:26 +0000 (09:34 +0000)]
Minor SPARC T4 and M7 fixes and additions.
* config/sparc/sparc.c (sparc_option_override): Set function
alignment for -mcpu=niagara7 to 64 to match the I$ line.
* config/sparc/sparc.h (BRANCH_COST): Set the SPARC M7 branch
latency to 1.
* config/sparc/sparc.h (BRANCH_COST): Set the SPARC T4 branch
latency to 2.
* config/sparc/sol2.h: Fix a ASM_CPU32_DEFAULT_SPEC typo.
* gcc.target/sparc/niagara7-align.c: New test.
Decimal float negative zero should compare equal to positive zero.
Decimal float zeroes are encoded as value class "normal" (in real.c);
they need to be handled specially, but in this one case that does not
yet happen. This fixes it.
PR middle-end/80692
* real.c (do_compare): Give decimal_do_compare preference over
comparing just the signs.
gcc/testsuite/
PR middle-end/80692
* gcc.c-torture/execute/pr80692.c: New testcase.
Andreas Tobler [Wed, 17 May 2017 20:54:39 +0000 (22:54 +0200)]
unwind-arm.h: Make _Unwind_GetIP...
2017-05-17 Andreas Tobler <andreast@gcc.gnu.org>
* config/arm/unwind-arm.h: Make _Unwind_GetIP, _Unwind_GetIPInfo and
_Unwind_SetIP available as functions for arm*-*-freebsd*.
* config/arm/unwind-arm.c: Implement the above.
Ian Lance Taylor [Wed, 17 May 2017 20:16:13 +0000 (20:16 +0000)]
libgo: add "vendor" to pkgpath for vendored standard packages
Ensure that the packages vendored into the standard library do not
have the same pkgpath as the actual packages. If we don't, attempts
to build and test the actual packages will get confused. The specific
error I was seeing was import loops, causing some of the packages to
fail to get initialized, causing an obscure run time crash.
projects / thirdparty / gcc.git / log
