Background: since gccgo does not currently merge identical types at link time,
the reflect function canonicalize() exists to choose a canonical specimen
for each set of identical types.
In this way, user code has the guarantee that identical types
will always compare as ==
Change: arrange reflect functions MapOf(), SliceOf(), StructOf() etc.
to call canonicalize() on the types they create, before storing the types
in internal lookup caches and returning them.
This fixes known cases where canonicalize() is needed but was missing.
Supersedes https://golang.org/cl/112575 and mostly fixes issue 25284.
Updates golang/go#25284
Reviewed-on: https://go-review.googlesource.com/115577
From-SVN: r261216
* config/i386/i386.md (simple_return_indirect_internal): New expander.
(*simple_return_indirect_internal<mode>): Rename from
simple_return_indirect_internal. Use W mode iterator.
(rstorssp): New expander.
(*rstorssp<mode>): Rename from rstorssp. Use P mode iterator.
(clrssbsy): New expander.
(*clrssbsy<mode>): Rename from clrssbsy. Use P mode iterator.
From-SVN: r261208
2018-06-05 Martin Liska <mliska@suse.cz>
PR gcov-profile/47618
* doc/invoke.texi: Document how -fprofile-dir format
is extended.
2018-06-05 Martin Liska <mliska@suse.cz>
PR gcov-profile/47618
* libgcov-driver-system.c (replace_filename_variables): New
function.
(gcov_exit_open_gcda_file): Use it.
From-SVN: r261199
2018-06-05 Richard Biener <rguenther@suse.de>
* tree-cfgcleanup.c (cleanup_control_flow_pre): For edge
removal pretend DOM info isn't available so we do not update
it and only remove edges, not dominated blocks. Actually free
DOM info in case we removed something. Remove unreachable blocks.
(mfb_keep_latches): Work with either DOM info or marked backedges.
(cleanup_tree_cfg_noloop): Do not remove unreachable blocks
first. Mark backedges if DOM info isn't available.
(Re-)compute DOM info after cleanup_control_flow_pre.
From-SVN: r261195
2018-06-05 Richard Biener <rguenther@suse.de>
* tree-cfg.c (struct locus_discrim_map): Store line, not location.
(locus_discrim_hasher::hash): Adjust.
(locus_discrim_hasher::equal): Likewise.
(next_discriminator_for_locus): Work on line directly.
(same_line_p): Pass in expanded locus1 as well.
(assign_discriminators): Avoid redundant location expansions.
From-SVN: r261194
2018-06-05 Richard Biener <rguenther@suse.de>
PR tree-optimization/86047
* tree-ssa-loop.c (for_each_index): Glob handling of all
decls and constants and really handle all of them.
* gcc.dg/pr86047.c: New testcase.
From-SVN: r261192
When trying to compile something with arm_acle.h using G++ we get a number of nasty errors:
arm_acle.h:48:49: error: invalid conversion from ‘const void*’ to ‘const int*’ [-fpermissive]
return __builtin_arm_ldc (__coproc, __CRd, __p);
This is because the intrinsics that are supposed to be void return the "result" of their builtin,
which is void. C lets that slide but C++ complains.
After fixing that we run into further errors:
arm_acle.h:48:46: error: invalid conversion from 'const void*' to 'const int*' [-fpermissive]
return __builtin_arm_ldc (__coproc, __CRd, __p);
^~~
Because the pointer arguments in these intrinsics are void pointers but the builtin
expects int pointers. So this patch introduces new qualifiers for void pointers and their
const-qualified versions and uses that in the specification of these intrinsics.
This gives us the opportunity of creating an arm subdirectory in g++.dg and inaugurates it
with the first arm-specific C++ tests (in that directory).
PR target/81497
* config/arm/arm-builtins.c (arm_type_qualifiers): Add
qualifier_void_pointer and qualifier_const_void_pointer.
(arm_ldc_qualifiers, arm_stc_qualifiers): Use the above.
(arm_init_builtins): Handle the above.
* config/arm/arm_acle.h (__arm_cdp, __arm_ldc, __arm_ldcl, __arm_stc,
__arm_stcl, __arm_mcr, __arm_cdp2, __arm_ldc2, __arm_ldcl2, __arm_stc2,
__arm_stcl2,__arm_mcr2, __arm_mcrr, __arm_mcrr2): Remove return for
void intrinsics.
* g++.target/arm/arm.exp: New file.
* g++.target/arm/pr81497.C: Likewise.
From-SVN: r261191
2018-06-05 Martin Liska <mliska@suse.cz>
* auto-profile.c (read_autofdo_file): Do not use
gcov_ctr_summary struct.
(afdo_callsite_hot_enough_for_early_inline): Likewise.
* coverage.c (struct counts_entry): Likewise.
(read_counts_file): Read just single summary entry.
(get_coverage_counts): Use gcov_summary struct.
* coverage.h (get_coverage_counts): Likewise.
* gcov-dump.c (dump_working_sets): Likewise.
(tag_summary): Dump just single summary.
* gcov-io.c (gcov_write_summary): Write just histogram
summary.
(gcov_read_summary): Read just single summary.
(compute_working_sets): Use gcov_summary struct.
* gcov-io.h (GCOV_TAG_SUMMARY_LENGTH): Remove usage
of GCOV_COUNTERS_SUMMABLE.
(GCOV_COUNTERS_SUMMABLE): Remove.
(GCOV_FIRST_VALUE_COUNTER): Replace with
GCOV_COUNTER_V_INTERVAL.
(struct gcov_ctr_summary): Remove.
(struct gcov_summary): Directly use fields of former
gcov_ctr_summary.
(compute_working_sets): Use gcov_summary struct.
* gcov.c (read_count_file): Do not use ctrs fields.
* lto-cgraph.c (merge_profile_summaries): Use gcov_summary
struct.
* lto-streamer.h (struct GTY): Make profile_info gcov_summary
struct.
* profile.c: Likewise.
* profile.h: Likewise.
2018-06-05 Martin Liska <mliska@suse.cz>
* libgcov-driver.c (gcov_compute_histogram): Remove usage
of gcov_ctr_summary.
(compute_summary): Do it just for a single summary.
(merge_one_data): Likewise.
(merge_summary): Simplify as we read just single summary.
(dump_one_gcov): Pass proper argument.
* libgcov-util.c (compute_one_gcov): Simplify as we have just
single summary.
(gcov_info_count_all_cold): Likewise.
(calculate_overlap): Likewise.
From-SVN: r261189
2018-06-05 Martin Liska <mliska@suse.cz>
PR gcov-profile/84846
* gcov.c (output_lines): Print working directory only
in intermediate format.
From-SVN: r261188
Symlinks are not supported on mingw.
* testsuite/27_io/filesystem/operations/read_symlink.cc: XFAIL for
mingw* targets.
* testsuite/27_io/filesystem/operations/symlink_status.cc: Likewise.
* testsuite/experimental/filesystem/operations/read_symlink.cc:
Likewise.
From-SVN: r261185
The builtin was accidentally defined to have an integer return value.
Fixed with the attached patch.
gcc/ChangeLog:
2018-06-05 Andreas Krebbel <krebbel@linux.ibm.com>
* config/s390/s390-builtin-types.def: Add void function type.
* config/s390/s390-builtins.def: Use the function type for the
tbeginc builtin.
gcc/testsuite/ChangeLog:
2018-06-05 Andreas Krebbel <krebbel@linux.ibm.com>
* gcc.target/s390/htm-builtins-compile-4.c: New test.
From-SVN: r261183
gcc/
* config/riscv/riscv-protos.h (riscv_expand_epilogue): Change bool arg
to int.
* config/riscv/riscv.c (riscv_for_each_saved_reg): New args epilogue
and maybe_eh_return. Change regno to unsigned int. Use new args to
handle EH_RETURN_DATA_REGNO registers properly.
(riscv_expand_prologue): Pass new args to riscv_for_each_saved_reg.
(riscv_expand_epilogue): Update comment. Change argument name and
type. Update code to use new name and type. Pass new args to
riscv_for_each_saved_reg. Only use EH_RETURN_STACKADJ_RTX when
EXCEPTION_RETURN.
* config/riscv/riscv.md (NORMAL_RETURN): New.
(SIBCALL_RETURN, EXCEPTION_RETURN): New.
(epilogue, sibcall_epilogue): Update riscv_expand_epilogue arg.
(eh_return): Call gen_eh_return_internal and emit barrier.
(eh_return_internal): Call riscv_expand_epilogue.
From-SVN: r261176
* gimple-ssa-store-merging.c (struct merged_store_group): Move up
bit_insertion field and declare can_be_merged_into method.
(merged_store_group::can_be_merged_into): New method.
(imm_store_chain_info::coalesce_immediate): Call it to decide whether
consecutive non-overlapping stores can be merged. Turn MEM_REF stores
into BIT_INSERT_EXPR stores if the group contains a non-MEM_REF store.
From-SVN: r261173
This patch fixes type traversal to avoid compiler crashes for test
cases where a type T includes an expression that refers back to the
type without actually explicitly mentioning T. Examples include
var x [uintptr(unsafe.Sizeof(&x))]byte
var a [len(a)]int
The fix involves expanding the set of types that the traversal code
"remembers" (to avoid cycles) to include array types, and introducing an
additional guard in Builtin_call_expression::do_is_constant to catch
cyclic type constructs.
Fixesgolang/go#25299Fixesgolang/go#25679Fixesgolang/go#25315Fixesgolang/go#25680
Reviewed-on: https://go-review.googlesource.com/115796
From-SVN: r261168
2018-06-04 Richard Biener <rguenther@suse.de>
PR tree-optimization/85955
* builtins.c (fold_builtin_sincos): Convert pointers to
destination to appropriate type before dereferencing.
* gcc.dg/pr85955.c: New testcase.
From-SVN: r261165
Without this patch absif2 always FAILs. There is no testcase for
that, nor do we see it during bootstrap, but it is obvious.
* config/rs6000/rs6000.md (abs<mode>2 for FLOAT128): Handle IFmode.
From-SVN: r261158
The handling of bitfield references in expand_expr_real_1 includes:
machine_mode ext_mode = mode;
if (ext_mode == BLKmode
&& ! (target != 0 && MEM_P (op0)
&& MEM_P (target)
&& multiple_p (bitpos, BITS_PER_UNIT)))
ext_mode = int_mode_for_size (bitsize, 1).else_blk ();
if (ext_mode == BLKmode)
{
[...]
gcc_assert (MEM_P (op0)
Here "mode" is the TYPE_MODE of the result, so when mode == BLKmode,
the target must be a MEM if nonnull, since no other rtl objects can
have BLKmode. But there's no guarantee that the source value op0 is also
BLKmode and thus also a MEM: we can reach the assert for any source if
the bitsize being extracted is larger than the largest integer mode
(or larger than MAX_FIXED_MODE_SIZE).
This triggered for SVE with -msve-vector-bits=512, where we could
sometimes try to extract a BLKmode value from a 512-bit vector,
and where int_mode_for_size would rightly fail for large bitsizes.
The patch reuses the existing:
/* Otherwise, if this is a constant or the object is not in memory
and need be, put it there. */
else if (CONSTANT_P (op0) || (!MEM_P (op0) && must_force_mem))
{
memloc = assign_temp (TREE_TYPE (tem), 1, 1);
emit_move_insn (memloc, op0);
op0 = memloc;
clear_mem_expr = true;
}
to handle this case.
2018-05-29 Richard Sandiford <richard.sandiford@linaro.org>
gcc/
* expr.c (expand_expr_real_1): Force the operand into memory if
its TYPE_MODE is BLKmode and if there is no integer mode for
the number of bits being extracted.
gcc/testsuite/
* gcc.target/aarch64/sve/extract_5.c: New test.
From-SVN: r261150
2018-06-04 Richard Biener <rguenther@suse.de>
PR tree-optimization/86038
* tracer.c (find_best_successor): Check probability for
being initialized, bail out if not.
* gcc.dg/pr86038.c: New testcase.
From-SVN: r261142
The XScale cpu configuration in GCC has always been somewhat
non-conforming. Although XScale isn't an architecture (it's simply an
implementation of ARMv5te), we do by tradition emit a specific
pre-define for it. We achieve this effect by adding an additional
feature bit to the xscale CPU definition that isn't part of the base
architecture.
When I restructured the options last year I overlooked this oddity and
the result, of course, is that this configuration now fails to build
as intended.
What happens is that the driver (correctly) constructs an architecture
for the xscale cpu name (as armv5te) and passes it in addition to the
CPU name. The backend code, on finding both a cpu and an architecture
specifies attempts to correlate the two and finds a difference due to
the additional feature bit and reports an inconsistency (fatally if
-werror is specified).
I think the best fix to this is to treat the xscale feature bit using
the same mechanism that we use for other 'quirks' in CPU
implementations and simply filter it out before comparing the
capabilities. It has the additional benefit that it's also the
simplest fix.
PR target/86003
* config/arm/arm-cpus.in (ALL_QUIRKS): Add xscale feature to the list
of bits to ignore when comparing architectures.
From-SVN: r261140
PR tree-optimization/69615
* fold-const.c (merge_ranges): If range1 is - [x, x] and x is the
maximum or minimum of the type, try to merge it also as if
range1 is + [-, x - 1] or + [x + 1, -].
* gcc.dg/pr69615.c: New test.
From-SVN: r261139
PR tree-optimization/86034
* gimple-ssa-store-merging.c (output_merged_store): Convert the RHS to
the unsigned bitfield type in a bit insertion sequence if it does not
have a larger precision than the bitfield size.
(process_store): Also bypass widening conversions for BIT_INSERT_EXPR.
From-SVN: r261128
2018-06-03 Paul Thomas <pault@gcc.gnu.org>
PR fortran/36497
* decl.c (variable_decl): Use gfc_add_type for cray pointees.
2018-06-03 Paul Thomas <pault@gcc.gnu.org>
PR fortran/36497
* gfortran.dg/cray_pointer_12.f90: New test.
From-SVN: r261127
