Jakub Jelinek [Sun, 31 May 2020 08:45:21 +0000 (10:45 +0200)]
expr: Fix fallout from optimize store_expr from STRING_CST [PR95052]
> Can't hurt, and debugging the assert tripping is likely a hell of a lot easier
> than debugging the resultant incorrect code. So if it passes, then I'd say go
> for it.
Testing passed, so I've committed it with those asserts (and thankfully I've
added them!) but it apparently broke Linux kernel build on arm.
The problem is that if the STRING_CST is very short, while the full object
has BLKmode, the short string could very well have
QImode/HImode/SImode/DImode and in that case it wouldn't take the path that
copies the string and then clears the remaining space, but different paths
in which it will ICE because of those asserts and without those it would
just emit wrong-code.
The following patch fixes it by enforcing BLKmode for the string MEM, even
if it is short, so that we copy it and memset the rest.
2020-05-31 Jakub Jelinek <jakub@redhat.com>
PR middle-end/95052
* expr.c (store_expr): For shortedned_string_cst, ensure temp has
BLKmode.
Arnaud Charlet [Sat, 30 May 2020 18:40:02 +0000 (14:40 -0400)]
Ability to build the GNAT runtime with project files
This change add project files to provide the ability to rebuild the
runtime with gprbuild after setup-rts is called.
gcc/ada/
* Makefile.rtl (ADA_INCLUDE_SRCS): Replace Makefile.adalib by
libada.gpr and associated project files.
(g-debpoo.o): Add missing rule to ensure subprograms are not reordered.
(setup-rts): Add generation of libgnat/libgnarl.lst.
(LIBGNAT_SRCS): Remove thread.c which is part of libgnarl.
* tracebak.c, tb-gcc.c: Merged the two files to simplify dependencies.
* libgnarl/libgnarl.gpr, libgnat/libada.gpr,
libgnat/libgnat.gpr, libgnat/libgnat_common.gpr: New files.
* doc/gnat_ugn/the_gnat_compilation_model.rst: Makefile.adalib
replaced by libada.gpr.
* libgnat/system-mingw.ads: Remove obsolete comment.
* gcc-interface/Makefile.in: Remove dependency on tb-gcc.c.
Harald Anlauf [Sat, 30 May 2020 18:59:41 +0000 (20:59 +0200)]
PR fortran/95373 - ICE in build_reference_type, at tree.c:7942
The use of KIND, LEN, RE, and IM inquiry references for applicable intrinsic
types is valid only for suffienctly new Fortran standards. Add appropriate
checks in the appropriate place.
2020-05-30 Harald Anlauf <anlauf@gmx.de>
gcc/fortran/
PR fortran/95373
* primary.c (is_inquiry_ref): Move validity check of inquiry
references against selected Fortran standard from here...
(gfc_match_varspec) ...to here.
This is necessary as libmsvcrt.a is not a pure import library, but
also contains some functions that invoke others in KERNEL32.DLL.
gcc/
* config/i386/mingw32.h (REAL_LIBGCC_SPEC): Insert -lkernel32
after -lmsvcrt. This is necessary as libmsvcrt.a is not a pure
import library, but also contains some functions that invoke
others in KERNEL32.DLL.
Signed-off-by: Liu Hao <lh_mouse@126.com> Signed-off-by: Jonathan Yong <10walls@gmail.com>
Patrick Palka [Sat, 30 May 2020 01:12:21 +0000 (21:12 -0400)]
c++: satisfaction value of type typedef to bool [PR95386]
In the testcase below, the satisfaction value of fn1<int>'s constraint
is INTEGER_CST '1' of type BOOLEAN_TYPE value_type, which is a typedef
to the standard boolean_type_node. But satisfaction_value expects to
see exactly boolean_true_node or integer_one_node, which this value is
neither, causing us to trip over the assert therein.
This patch changes satisfaction_value to accept INTEGER_CST of any
boolean type.
gcc/cp/ChangeLog:
PR c++/95386
* constraint.cc (satisfaction_value): Accept INTEGER_CST of any
boolean type.
gcc/testsuite/ChangeLog:
PR c++/95386
* g++.dg/concepts/pr95386.C: New test.
rs6000: Prefer VSX insns over VMX ones (part 1: perm and mrg)
There are various VSX insns that do the same job as (older) AltiVec
insns, just with a wider range of possible registers. Many patterns
for such insns have the "v" alternative before the "wa" alternative,
which makes the output less readable than possible (since vs32 is v0,
and most insns before or after this insn will be VSX as well).
This changes the define_insns for the mrg and perm machine instructions
to prefer the VSX form. No behaviour change. Only one testcase needed
a little adjustment as well.
Patrick Palka [Fri, 22 May 2020 14:28:19 +0000 (10:28 -0400)]
c++: P0848R3 and member function templates [PR95181]
When comparing two special member function templates to see if one hides
the other (as per P0848R3), we need to check satisfaction which we can't
do on templates. So this patch makes add_method skip the eligibility
test on member function templates and just lets them coexist.
gcc/cp/ChangeLog:
PR c++/95181
* class.c (add_method): Let special member function templates
coexist if they are not equivalently constrained, or in a class
template.
gcc/testsuite/ChangeLog:
PR c++/95181
* g++.dg/concepts/pr95181.C: New test.
* g++.dg/concepts/pr95181-2.C: New test.
Jason Merrill [Fri, 29 May 2020 20:55:52 +0000 (16:55 -0400)]
c++: Template template parameter in constraint [PR95371]
any_template_parm_r was assuming that the DECL_TEMPLATE_RESULT of a template
will have a suitable TEMPLATE_INFO from which we can look at the generic
arguments for that template. But that wasn't true for a template template
parameter; this patch makes it so.
gcc/cp/ChangeLog:
PR c++/95371
* pt.c (process_template_parm): Set DECL_TEMPLATE_INFO
on the DECL_TEMPLATE_RESULT.
gcc/testsuite/ChangeLog:
PR c++/95371
* g++.dg/cpp2a/concepts-ttp1.C: New test.
Jan Hubicka [Fri, 29 May 2020 20:41:11 +0000 (22:41 +0200)]
Simplify tree streaming.
this patch cleans up tree streaming. The code is prepared to stream nested
trees, but we only handle flat trees. As a result we have quite heavy function
to stream in/out tree reference which is used many times and shows up in
profile.
This patch adds stream_write_tree_ref/stream_read_tree_ref which is used to
stream references to trees that are required to exist in the cache or be
indexable.
The actual implementation is just a first cut. I would like to make it more
compact. We used to stream 2 byte tag (as UHWI) + UHWI representing the index.
Now we stream one UHWI that represent type of reference + index for references
to cache, but still two integers for references to global stream. This is
becaue the abstraction is not very helpful here and I want to clean this up
incrementally.
I would also like to get rid of the ref_p parameters which seems unnecessary for
flat streams.
This reduces around 7% of global stream, 3% when compressed.
More reduction will happen once the format is sanitized a bit.
from
[WPA] read 4597161 unshared trees
[WPA] read 2937414 mergeable SCCs of average size 1.364280
[WPA] 8604617 tree bodies read in total
[WPA] tree SCC table: size 524287, 247507 elements, collision ratio: 0.377468
[WPA] tree SCC max chain length 2 (size 1)
[WPA] Compared 2689907 SCCs, 184 collisions (0.000068)
[WPA] Merged 2689890 SCCs
[WPA] Merged 3722677 tree bodies
[WPA] Merged 632040 types
...
[WPA] Compression: 88124141 input bytes, 234906430 uncompressed bytes (ratio: 2.665631)
[WPA] Size of mmap'd section decls: 88124141 bytes
...
[WPA] Compression: 113758813 input bytes, 316149514 uncompressed bytes (ratio: 2.779121)
[WPA] Size of mmap'd section decls: 88124141 bytes
[WPA] Size of mmap'd section function_body: 14485721 bytes
to
[WPA] read 4597174 unshared trees
[WPA] read 2937413 mergeable SCCs of average size 1.364280
[WPA] 8604629 tree bodies read in total
[WPA] tree SCC table: size 524287, 247509 elements, collision ratio: 0.377458
[WPA] tree SCC max chain length 2 (size 1)
[WPA] Compared 2689904 SCCs, 183 collisions (0.000068)
[WPA] Merged 2689888 SCCs
[WPA] Merged 3722675 tree bodies
[WPA] Merged 632041 types
....
[WPA] Size of mmap'd section decls: 86177293 bytes
[WPA] Compression: 86177293 input bytes, 217625095 uncompressed bytes (ratio: 2.525318)
....
[WPA] Compression: 111682269 input bytes, 297228756 uncompressed bytes (ratio: 2.661378)
[WPA] Size of mmap'd section decls: 86177293 bytes
[WPA] Size of mmap'd section function_body: 14349032 bytes
Harald Anlauf [Fri, 29 May 2020 19:19:31 +0000 (21:19 +0200)]
PR fortran/95090 - ICE: identifier overflow
The initial fix for this PR uncovered several latent issues with further
too small string buffers which showed up only when testing on i686.
Provide sufficiently large temporaries.
2020-05-29 Harald Anlauf <anlauf@gmx.de>
gcc/fortran/
PR fortran/95090
* class.c (get_unique_type_string): Enlarge temporary for
name-mangling. Use strncpy to prevent buffer overrun.
(get_unique_hashed_string): Enlarge temporary.
(gfc_hash_value): Enlarge temporary for name-mangling.
Marek Polacek [Tue, 26 May 2020 23:59:26 +0000 (19:59 -0400)]
c++: Fix bogus -Wparentheses warning [PR95344]
Since r267272, which added location wrappers, cp_fold loses
TREE_NO_WARNING on a MODIFY_EXPR that finish_parenthesized_expr set, and
that results in a bogus -Wparentheses warning.
I.e., previously we had "b = 1" but now we have "VIEW_CONVERT_EXPR<bool>(b) = 1"
and cp_fold_maybe_rvalue folds away the location wrapper and so we do
2718 x = fold_build2_loc (loc, code, TREE_TYPE (x), op0, op1);
in cp_fold and the flag is lost.
PR c++/95344
* cp-gimplify.c (cp_fold) <case MODIFY_EXPR>: Don't set
TREE_THIS_VOLATILE here.
(cp_fold): Set it here along with TREE_NO_WARNING.
Patrick Palka [Fri, 29 May 2020 13:40:40 +0000 (09:40 -0400)]
c++: lambdas inside constraints [PR92652]
When parsing a constraint-expression, a requires-clause or a
requires-expression, we temporarily increment processing_template_decl
so that we always obtain template trees which we could later reduce via
substitution even when not inside a template.
But incrementing processing_template_decl when we're already inside a
template has the unintended side effect of shifting up the template
parameter levels of a lambda defined inside one of these constructs,
which leads to confusion later during substitution into the lambda.
This patch fixes this issue by incrementing processing_template_decl
during parsing of these constructs only if it is 0.
Passes 'make check-c++', and also tested by building cmcstl2, does this
look OK to commit after a full bootstrap/regtest?
gcc/cp/ChangeLog:
PR c++/92652
PR c++/93698
PR c++/94128
* parser.c (cp_parser_requires_clause_expression): Temporarily
increment processing_template_decl only if it is 0.
(cp_parser_constraint_expression): Likewise.
(cp_parser_requires_expression): Likewise.
gcc/testsuite/ChangeLog:
PR c++/92652
PR c++/93698
PR c++/94128
* g++.dg/cpp2a/concepts-lambda8.C: New test.
* g++.dg/cpp2a/concepts-lambda9.C: New test.
* g++.dg/cpp2a/concepts-lambda10.C: New test.
Patrick Palka [Fri, 29 May 2020 13:44:09 +0000 (09:44 -0400)]
c++: constexpr ctor with RANGE_EXPR index [PR95241]
In the testcase below, the CONSTRUCTOR for 'field' contains a RANGE_EXPR
index:
{{aggr_init_expr<...>, [1...2]={.off=1}}}
but get_or_insert_ctor_field isn't prepared to handle looking up a
RANGE_EXPR index.
This patch adds limited support to get_or_insert_ctor_field for looking
up a RANGE_EXPR index. The limited scope of this patch should make it
more suitable for backporting, and more extensive support would be
needed only to handle self-modifying CONSTRUCTORs that contain a
RANGE_EXPR index, but I haven't yet been able to come up with a testcase
that actually creates such a CONSTRUCTOR.
gcc/cp/ChangeLog:
PR c++/95241
* constexpr.c (get_or_insert_ctor_field): Add limited support
for RANGE_EXPR index lookups.
gcc/testsuite/ChangeLog:
PR c++/95241
* g++.dg/cpp0x/constexpr-array25.C: New test.
This adds SLP_TREE_REPRESENTATIVE - a representative stmt-info that
is used by SLP analysis and code generation. This avoids the need
for the hack in vect_slp_rearrange_stmts which previously avoided
to re-arrange stmts that might not have been isomorphic because
of operand swapping. It also plays nice with future directions of SLP
and for the forseeable future is easier than replicating more and
more info in the SLP node as long as non-SLP is in-tree.
2020-05-29 Richard Biener <rguenther@suse.de>
PR tree-optimization/95272
* tree-vectorizer.h (_slp_tree::representative): Add.
(SLP_TREE_REPRESENTATIVE): Likewise.
* tree-vect-loop.c (vectorizable_reduction): Adjust SLP
node gathering.
(vectorizable_live_operation): Use the representative to
attach the reduction info to.
* tree-vect-slp.c (_slp_tree::_slp_tree): Initialize
SLP_TREE_REPRESENTATIVE.
(vect_create_new_slp_node): Likewise.
(slp_copy_subtree): Copy it.
(vect_slp_rearrange_stmts): Re-arrange even COND_EXPR stmts.
(vect_slp_analyze_node_operations_1): Pass the representative
to vect_analyze_stmt.
(vect_schedule_slp_instance): Pass the representative to
vect_transform_stmt.
Richard Biener [Fri, 29 May 2020 08:46:06 +0000 (10:46 +0200)]
tree-optimization/95356 - more vectorizable_shift massaging
The previous fix clashed with the rewrite to emit SLP invariants
during the SLP walk. Thus the following adjusts the SLP tree
hacking vectorizable_shift does appropriately.
Still resisting the attempt of a rewrite of vectorizable_shift ...
2020-05-29 Richard Biener <rguenther@suse.de>
PR tree-optimization/95356
* tree-vect-stmts.c (vectorizable_shift): Do in-place SLP
node hacking during analysis.
Jan Hubicka [Fri, 29 May 2020 10:25:48 +0000 (12:25 +0200)]
Fix streamer desynchornization caused by streamer debugging patch
it turns out I lost one hunk in the patch disabling extra streaming
which causes streamer to go out of sync in the case non-trivial scc
containing the node being streamed appears in local stream (which seems
quite rare since it does not happen during bootstrap).
Martin Liska [Fri, 29 May 2020 09:29:25 +0000 (11:29 +0200)]
Fix various limitations of git-backport.py.
I've just tested the script and I'm going to install the patch
to all active branches.
contrib/ChangeLog:
* git-backport.py: The script did 'git co HEAD~' when
there was no modified ChangeLog file in a successful
git cherry pick.
Run cherry-pick --continue without editor.
Jakub Jelinek [Fri, 29 May 2020 08:42:50 +0000 (10:42 +0200)]
expander: Optimize store_expr from STRING_CST [PR95052]
In the following testcase, store_expr of e.g. 97 bytes long string literal
into 1MB long array is implemented by copying the 97 bytes from .rodata
section, followed by clearing the remaining bytes. But, as the STRING_CST
has type char[1024*1024], we actually allocate whole 1MB in .rodata section
for it, even when we only use the first 97 bytes from that.
The following patch tweaks it so that if we are going to initialize only the
small part from it, we don't emit all the zeros that we never use after it.
2020-05-29 Jakub Jelinek <jakub@redhat.com>
PR middle-end/95052
* expr.c (store_expr): If expr_size is constant and significantly
larger than TREE_STRING_LENGTH, set temp to just the
TREE_STRING_LENGTH portion of the STRING_CST.
Richard Biener [Fri, 29 May 2020 07:25:53 +0000 (09:25 +0200)]
tree-optimization/95393 - fold MIN/MAX_EXPR generated by phiopt
This makes sure to fold generated stmts so they do not survive
until RTL expansion and cause awkward code generation.
2020-05-29 Richard Biener <rguenther@suse.de>
PR tree-optimization/95393
* tree-ssa-phiopt.c (minmax_replacement): Use gimple_build
to build the min/max expression so we simplify cases like
MAX(0, s) immediately.
* gcc.dg/tree-ssa/phi-opt-21.c: New testcase.
* g++.dg/vect/slp-pr87105.cc: Adjust.
Joe Ramsay [Fri, 29 May 2020 07:44:37 +0000 (08:44 +0100)]
aarch64: add support for unpacked EOR, ORR and AND
Extended patterns for these instructions to support unpacked vectors.
BIC will have to wait, as there is not currently support for unpacked
NOT.
2020-05-29 Joe Ramsay <joe.ramsay@arm.com>
gcc/
* config/aarch64/aarch64-sve.md (<LOGICAL:optab><mode>3): Add support
for unpacked EOR, ORR, AND.
gcc/testsuite/
* gcc.target/aarch64/sve/load_const_offset_2.c: Force using packed
vectors.
* gcc.target/aarch64/sve/logical_unpacked_and_1.c: New test.
* gcc.target/aarch64/sve/logical_unpacked_and_2.c: New test.
* gcc.target/aarch64/sve/logical_unpacked_and_3.c: New test.
* gcc.target/aarch64/sve/logical_unpacked_and_4.c: New test.
* gcc.target/aarch64/sve/logical_unpacked_and_5.c: New test.
* gcc.target/aarch64/sve/logical_unpacked_and_6.c: New test.
* gcc.target/aarch64/sve/logical_unpacked_and_7.c: New test.
* gcc.target/aarch64/sve/logical_unpacked_eor_1.c: New test.
* gcc.target/aarch64/sve/logical_unpacked_eor_2.c: New test.
* gcc.target/aarch64/sve/logical_unpacked_eor_3.c: New test.
* gcc.target/aarch64/sve/logical_unpacked_eor_4.c: New test.
* gcc.target/aarch64/sve/logical_unpacked_eor_5.c: New test.
* gcc.target/aarch64/sve/logical_unpacked_eor_6.c: New test.
* gcc.target/aarch64/sve/logical_unpacked_eor_7.c: New test.
* gcc.target/aarch64/sve/logical_unpacked_orr_1.c: New test.
* gcc.target/aarch64/sve/logical_unpacked_orr_2.c: New test.
* gcc.target/aarch64/sve/logical_unpacked_orr_3.c: New test.
* gcc.target/aarch64/sve/logical_unpacked_orr_4.c: New test.
* gcc.target/aarch64/sve/logical_unpacked_orr_5.c: New test.
* gcc.target/aarch64/sve/logical_unpacked_orr_6.c: New test.
* gcc.target/aarch64/sve/logical_unpacked_orr_7.c: New test.
* gcc.target/aarch64/sve/scatter_store_6.c: Force using packed vectors.
* gcc.target/aarch64/sve/scatter_store_7.c: Force using packed vectors.
* gcc.target/aarch64/sve/strided_load_3.c: Force using packed vectors.
* gcc.target/aarch64/sve/strided_store_3.c: Force using packed vectors.
* gcc.target/aarch64/sve/unpack_signed_1.c: Force using packed vectors.
Jakub Jelinek [Thu, 28 May 2020 21:40:54 +0000 (23:40 +0200)]
c++: Try to complete decomp types [PR95328]
Two years ago Paolo has added the
else if (processing_template_decl && !COMPLETE_TYPE_P (type))
pedwarn (...);
lines into cp_finish_decomp. For type dependent decl we punt much earlier,
but even for types which aren't type dependent COMPLETE_TYPE_P might be
false as this testcase shows, so this patch tries to complete_type first
(the reason for writing it that way is that it is then followed by another
else if and if complete_type returns error_mark_node, we shouldn't report
anything, as a bug should have been reported already.
2020-05-28 Jakub Jelinek <jakub@redhat.com>
PR c++/95328
* decl.c (cp_finish_decomp): Call complete_type before checking
COMPLETE_TYPE_P.
Harald Anlauf [Thu, 28 May 2020 20:28:08 +0000 (22:28 +0200)]
PR fortran/95373 - ICE in build_reference_type, at tree.c:7942
The use of KIND, LEN, RE, and IM inquiry references for applicable intrinsic
types is valid only for suffienctly new Fortran standards. Add appropriate
check.
2020-05-28 Harald Anlauf <anlauf@gmx.de>
gcc/fortran/
PR fortran/95373
* primary.c (is_inquiry_ref): Check validity of inquiry
references against selected Fortran standard.
gcc/testsuite/
PR fortran/95373
* gfortran.dg/pr95373_1.f90: New test.
* gfortran.dg/pr95373_2.f90: New test.
Jason Merrill [Thu, 28 May 2020 04:35:56 +0000 (00:35 -0400)]
c++: Immediately deduce auto member [PR94926].
In r9-297 I was trying to be more flexible and treat static data members of
class templates more like variable templates, where the type need not be
determined until the variable is instantiated, but I suppose that in a class
the types of all the non-template members need to be determined at the time
of class instantiation.
gcc/cp/ChangeLog:
PR c++/94926
* decl.c (cp_finish_decl): Revert r9-297 change.
(check_static_variable_definition): Likewise.
* constexpr.c (ensure_literal_type_for_constexpr_object): Likewise.
* pt.c (instantiate_decl): Return early on type error.
Nicolás Bértolo [Fri, 22 May 2020 20:54:41 +0000 (17:54 -0300)]
jit: port libgccjit to Windows
2020-05-28 Nicolas Bértolo <nicolasbertolo@gmail.com>
/ChangeLog:
* configure.ac: Don't require --enable-host-shared when building
for Mingw.
* configure: Regenerate.
2020-05-28 Nicolas Bértolo <nicolasbertolo@gmail.com>
gcc/ChangeLog:
* Makefile.in: don't look for libiberty in the "pic" subdirectory
when building for Mingw. Add dependency on xgcc with the proper
extension.
2020-05-28 Nicolas Bértolo <nicolasbertolo@gmail.com>
gcc/c/ChangeLog:
* Make-lang.in: Remove extra slash.
2020-05-28 Nicolas Bértolo <nicolasbertolo@gmail.com>
gcc/jit/ChangeLog:
* Make-lang.in: Remove extra slash. Build libgccjit.dll and its
import library in Windows.
* config-lang.in: Update comment about --enable-host-shared.
* jit-w32.h: New file.
* jit-w32.c: New file.
(print_last_error): New function that prints the error
string corresponding to GetLastError().
(get_TOKEN_USER_current_user): Helper function used for getting
the SID belonging to the current user.
(create_directory_for_current_user): Helper function to create
a directory with permissions such that only the current user can
access it.
(win_mkdtemp): Create a temporary directory using Windows APIs.
* jit-playback.c: Do not chmod files in Windows. Use LoadLibrary,
FreeLibrary and GetProcAddress instead of libdl.
* jit-result.h, jit-result.c: Introduce result::handle_t to
abstract over the types used for dynamic library handles.
* jit-tempdir.c: Do not use mkdtemp() in Windows, use
win_mkdtemp().
Jeff Law [Thu, 28 May 2020 18:28:56 +0000 (12:28 -0600)]
Fix incorrect code generation with bit insns on H8/SX.
* config/h8300/logical.md (HImode H8/SX bit-and splitter): Don't
make a nonzero adjustment to the memory offset.
(b<ior,xor>hi_msx): Turn into a splitter.
Based on a patch in the comments of the PR. That patch fixed this
problem but caused the test cases for PR93484 to fail. It has been
changed to reduce initialisation expressions if the expression is
not EXPR_VARIABLE and not EXPR_CONSTANT.
2020-05-28 Steven G. Kargl <kargl@gcc.gnu.org>
Mark Eggleston <markeggleston@gcc.gnu.org>
gcc/fortran/
PR fortran/94397
* match.c (gfc_match_type_spec): New variable ok initialised
to true. Set ok with the return value of gfc_reduce_init_expr
called only if the expression is not EXPR_CONSTANT and is not
EXPR_VARIABLE. Add !ok to the check for type not being integer
or the rank being greater than zero.
2020-05-28 Mark Eggleston <markeggleston@gcc.gnu.org>
gcc/testsuite/
PR fortran/94397
* gfortran.dg/pr94397.F90: New test.
wb_candidate1 and wb_candidate2 exist for two overlapping cases:
when we use an STR or STP with writeback to allocate the frame,
and when we set up a frame chain record (either using writeback
allocation or not).
However, aarch64_layout_frame was leaving these fields with
legitimate register numbers even if we decided to do neither
of those things. This prevented those registers from being
shrink-wrapped, even though we were otherwise treating them
as normal saves and restores.
The case this patch handles isn't the common case, so it might
not be worth going out of our way to optimise it. But I think
the patch actually makes the output of aarch64_layout_frame more
consistent.
2020-05-28 Richard Sandiford <richard.sandiford@arm.com>
gcc/
* config/aarch64/aarch64.h (aarch64_frame): Add a comment above
wb_candidate1 and wb_candidate2.
* config/aarch64/aarch64.c (aarch64_layout_frame): Invalidate
wb_candidate1 and wb_candidate2 if we decided not to use them.
gcc/testsuite/
* gcc.target/aarch64/shrink_wrap_1.c: New test.
aarch64: Fix segfault in aarch64_expand_epilogue [PR95361]
The stack frame for the function in the testcase consisted of two
SVE save slots. Both saves had been shrink-wrapped, but for different
blocks, meaning that the stack allocation and deallocation were
separate from the saves themselves. Before emitting the deallocation,
we tried to attach a REG_CFA_DEF_CFA note to the preceding instruction,
to redefine the CFA in terms of the stack pointer. But in this case
there was no preceding instruction.
This in practice only happens for SVE because:
(a) We don't try to shrink-wrap wb_candidate* registers even when
we've decided to treat them as normal saves and restores.
I have a fix for that.
(b) Even with (a) fixed, we're (almost?) guaranteed to emit
a stack tie for frames that are 64k or larger, so we end
up hanging the REG_CFA_DEF_CFA note on that instead.
We should only need to redefine the CFA if it was previously
defined in terms of the frame pointer. In other cases the CFA
should already be defined in terms of the stack pointer,
so redefining it is unnecessary but usually harmless.
2020-05-28 Richard Sandiford <richard.sandiford@arm.com>
gcc/
PR testsuite/95361
* config/aarch64/aarch64.c (aarch64_expand_epilogue): Assert that
we have at least some CFI operations when using a frame pointer.
Only redefine the CFA if we have CFI operations.
gcc/testsuite/
PR testsuite/95361
* gcc.target/aarch64/sve/pr95361.c: New test.
Richard Biener [Wed, 27 May 2020 13:38:20 +0000 (15:38 +0200)]
Code generate externals/invariants during the SLP graph walk
This generates vector defs for externals and invariants during the SLP
walk rather than as part of getting vectorized defs when vectorizing
the users. This is a requirement to make sharing of external/invariant
nodes be reflected in actual code generation.
This temporarily adds a SLP_TREE_VEC_DEFS vector alongside the
SLP_TREE_VEC_STMTS one. Eventually the latter can go away.
2020-05-27 Richard Biener <rguenther@suse.de>
* tree-vectorizer.h (_slp_tree::vec_defs): Add.
(SLP_TREE_VEC_DEFS): Likewise.
* tree-vect-slp.c (_slp_tree::_slp_tree): Adjust.
(_slp_tree::~_slp_tree): Likewise.
(vect_mask_constant_operand_p): Remove unused function.
(vect_get_constant_vectors): Rename to...
(vect_create_constant_vectors): ... this. Take the
invariant node as argument and code generate it. Remove
dead code, remove temporary asserts. Pass a NULL stmt_info
to vect_init_vector.
(vect_get_slp_defs): Simplify.
(vect_schedule_slp_instance): Code-generate externals and
invariants using vect_create_constant_vectors.
Richard Biener [Mon, 18 May 2020 14:01:08 +0000 (16:01 +0200)]
make vect_finish_stmt_generation work w/o stmt_vec_info
This makes the call chain below vec_init_vector happy with a NULL
stmt_vec_info which is used as "context".
2020-05-27 Richard Biener <rguenther@suse.de>
* tree-vect-stmts.c (vect_finish_stmt_generation_1):
Conditionalize stmt_info use, assert the new stmt cannot throw
when not specified.
(vect_finish_stmt_generation): Adjust assert.
Richard Biener [Thu, 28 May 2020 08:27:48 +0000 (10:27 +0200)]
tree-optimization/95273 - more vectorizable_shift massaging
Covering all bases in vectorizable_shift is hard - this makes sure
to appropriately handle the case of PR95356 without breaking others.
2020-05-28 Richard Biener <rguenther@suse.de>
PR tree-optimization/95273
PR tree-optimization/95356
* tree-vect-stmts.c (vectorizable_shift): Adjust when and to
what we set the vector type of the shift operand SLP node
again.
So far, we expect from a commit that renames a file to contain a
changelog entry only for the new name. For example, after the following
commit:
$ git move foo bar
$ git commit
We expect the following changelog:
* bar: Renamed from foo.
Git does not keep track of renamings, only file deletions and additions.
The display of patches then uses heuristics (with config-dependent
parameters) to try to match deleted and added files in the same commit.
It is thus brittle to rely on this information.
This commit modifies changelog processing so that renames are considered
as a deletion of a file plus an addition of another file. The following
changelog is now expected for the above example:
* foo: Move...
* bar: Here.
contrib/
* gcc-changelog/git_email.py (GitEmail.__init__): Interpret file
renamings as a file deletion plus a file addition.
* gcc-changelog/git_repository.py (parse_git_revisions):
Likewise.
* gcc-changelog/test_email.py: New testcase.
* gcc-changelog/test_patches.txt: New testcase.
liuhongt [Mon, 25 May 2020 08:10:06 +0000 (16:10 +0800)]
Fix nonconforming memory_operand for vpmovq{d,w,b}/vpmovd{w,b}/vpmovwb.
According to Intel SDM, VPMOVQB xmm1/m16 {k1}{z}, xmm2 has 16-bit
memory_operand instead of 128-bit one which existed in current
implementation. Also for other vpmov instructions which have
memory_operand narrower than 128bits.
Alexandre Oliva [Thu, 28 May 2020 02:16:52 +0000 (23:16 -0300)]
outputs.exp: no lto, linker default output, cdtor temps, empty args
This patch fixes various issues in the testsuite that came up after
the dump/aux output revamp, namely:
- many outputs.exp tests used -flto without checking that LTO was
supported, getting lots of failures. With this patch, we test for LTO
support, and skip -flto tests on platforms that do not support it.
- some linkers error out if an output file is not named, and the
a.{out,exe} construct that we used throughout outputs.exp to match the
default linker output would trigger a bug in tcl globbing. With this
patch, we detect the default linker output early. If none is found,
we arrange to pass -o a.out explicitly in tests that used to test the
default linker output. We now look for the detected default, or for
explicitly-specified output.
- collect2 will leave <execname>.cdtor.* files behind in -save-temps
tests. Ignore them.
- The prepending of -Wl, to file names in ldflags et al was done in a
way that introduced empty arguments when consecutive blanks appeared
in these board configuration knobs. Skip the empty strings between
consecutive blanks to avoid this problem.
gcc/testsuite/ChangeLog:
* lib/gcc-defs.exp: Avoid introducing empty arguments between
consecutive blanks in board linking options.
* gcc.misc-tests/outputs.exp: Likewise. Document
-gsplit-dwarf testing, skip LTO tests if -flto is not
supported, detect the default linker output name, cope with
the need for an explicit executable output.
Max Filippov [Mon, 25 May 2020 02:41:13 +0000 (19:41 -0700)]
gcc: xtensa: delegitimize UNSPEC_PLT
This fixes 'non-delegitimized UNSPEC 3 found in variable location' notes
issued when building libraries which interferes with running tests.
2020-05-27 Max Filippov <jcmvbkbc@gmail.com>
gcc/
* config/xtensa/xtensa.c (xtensa_delegitimize_address): New
function.
(TARGET_DELEGITIMIZE_ADDRESS): New macro.
Eric Botcazou [Wed, 27 May 2020 22:31:15 +0000 (00:31 +0200)]
Add support for __builtin_bswap128
This patch introduces a new builtin named __builtin_bswap128 on targets
where TImode is supported, i.e. 64-bit targets only in practice. The
implementation simply reuses the existing double word path in optab, so
no routine is added to libgcc (which means that you get two calls to
_bswapdi2 in the worst case).
gcc/ChangeLog:
* builtin-types.def (BT_UINT128): New primitive type.
(BT_FN_UINT128_UINT128): New function type.
* builtins.def (BUILT_IN_BSWAP128): New GCC builtin.
* doc/extend.texi (__builtin_bswap128): Document it.
* builtins.c (expand_builtin): Deal with BUILT_IN_BSWAP128.
(is_inexpensive_builtin): Likewise.
* fold-const-call.c (fold_const_call_ss): Likewise.
* fold-const.c (tree_call_nonnegative_warnv_p): Likewise.
* tree-ssa-ccp.c (evaluate_stmt): Likewise.
* tree-vect-stmts.c (vect_get_data_ptr_increment): Likewise.
(vectorizable_call): Likewise.
* optabs.c (expand_unop): Always use the double word path for it.
* tree-core.h (enum tree_index): Add TI_UINT128_TYPE.
* tree.h (uint128_type_node): New global type.
* tree.c (build_common_tree_nodes): Build it if TImode is supported.
Jonathan Wakely [Wed, 27 May 2020 21:55:21 +0000 (22:55 +0100)]
libstdc++: Fix atomic<FP>::load (PR 95282)
PR libstdc++/95282
* include/bits/atomic_base.h (__atomic_impl::load): Add
cv-qualifiers to parameter so that _Tp is deduced as the
unqualified type.
* testsuite/29_atomics/atomic_float/95282.cc: New test.
Jonathan Wakely [Wed, 27 May 2020 21:08:15 +0000 (22:08 +0100)]
libstdc++: Fix view adaptors for mixed-const sentinels and iterators (PR 95322)
The bug report is that transform_view's sentinel<false> cannot be
compared to its iterator<true>. The comparison is supposed to use
operator==(iterator<Const>, sentinel<Const>) after converting
sentinel<false> to sentinel<true>. However, the operator== is a hidden
friend so is not a candidate when comparing iterator<true> with
sentinel<false>. The required conversion would only happen if we'd found
the operator, but we can't find the operator until after the conversion
happens.
A new LWG issue has been reported, but not yet assigned a number. The
solution suggested by Casey Carter is to make the hidden friends of the
sentinel types work with iterators of any const-ness, so that no
conversions are required.
Patrick Palka observed that join_view has a similar problem and a
similar fix is used for its sentinel.
PR libstdc++/95322
* include/std/ranges (transform_view::_Sentinel): Allow hidden
friends to work with _Iterator<true> and _Iterator<false>.
(join_view::_Sentinel): Likewise.
* testsuite/std/ranges/adaptors/95322.cc: New test.
The std::reverse_iterator comparisons have always been implemented only
in terms of equality and less than. In C++98 that made no difference for
reasonable code, because when the underlying operators are the same type
they are required to support all comparisons anyway.
But since LWG 280 it's possible to compare reverse_iterator<X> and
reverse_iterator<Y>, and comparisons between X and Y might not support
the full set of equality and relational operators. This means that it
matters whether we implement operator!= as x.base() != y.base() or
!(x.base() == y.base()), and the current implementation is
non-conforming.
This was already fixed in GCC 10.1 for C++20, this change also fixes it
for all other -std modes.
PR libstdc++/94354
* include/bits/stl_iterator.h (reverse_iterator): Fix comparison
operators to use the correct operations on the underlying
iterators.
* testsuite/24_iterators/reverse_iterator/rel_ops.cc: New test.
Uros Bizjak [Wed, 27 May 2020 19:46:49 +0000 (21:46 +0200)]
i386: Fix V2SF horizontal add/subtract insns
PFPNACC insn is incorrectly modelled to perform addition and subtraction
of two operands, but in reality it performs horizontal addition and
subtraction:
Harald Anlauf [Wed, 27 May 2020 19:20:24 +0000 (21:20 +0200)]
PR fortran/95090 - ICE: identifier overflow
For long module name, derive type and component name, the generated
name-mangled symbol did not fit into a buffer when coarrays were
enabled. Provide sufficiently large temporary.
2020-05-27 Harald Anlauf <anlauf@gmx.de>
gcc/fortran/
PR fortran/95090
* iresolve.c (gfc_get_string): Enlarge temporary for
name-mangling.
gcc/testsuite/
PR fortran/95090
* gfortran.dg/pr95090.f90: New test.
David Malcolm [Wed, 27 May 2020 13:44:07 +0000 (09:44 -0400)]
jit: use deep unsharing of trees [PR 95314]
PR jit/95314 reports a internal error inside verify_gimple, which
turned out to be due to reusing the result of
gcc_jit_lvalue_get_address in several functions, leading to tree nodes
shared between multiple function bodies.
This patch fixes the issue by adopting the "Deep unsharing" strategy
described in the comment in gimplify.c preceding mostly_copy_tree_r:
to mark all of the jit "frontend"'s expression tree nodes with
TREE_VISITED, and to set LANG_HOOKS_DEEP_UNSHARING, so that "they are
unshared on the first reference within functions when the regular
unsharing algorithm runs".
gcc/jit/ChangeLog:
PR jit/95314
* dummy-frontend.c (LANG_HOOKS_DEEP_UNSHARING): Define to be true.
* jit-playback.h (gcc::jit::playback::rvalue): Mark tree node with
TREE_VISITED.
Here, when considering the two 'insert' overloads, we look for aggregate
conversions from the same initializer-list to B<3> or
initializer_list<B<3>>. But since my fix for reshape_init overhead on the
PR14179 testcase we reshaped the initializer-list directly, leading to an
error when we then tried to reshape it differently for the second overload.
gcc/cp/ChangeLog:
PR c++/95319
* decl.c (reshape_init_array_1): Don't reuse in overload context.
gcc/testsuite/ChangeLog:
PR c++/95319
* g++.dg/cpp0x/initlist-array12.C: New test.
Mark Eggleston [Thu, 7 May 2020 07:15:26 +0000 (08:15 +0100)]
Fortran : ICE in gfc_trans_label_assign PR50392
A function may contain an assigned goto. If the the return variable
is an integer a statement can be assigned to it. Prior to this fix
this resulted in an ICE.
PR fortran/50392
* trans-decl.c (gfc_get_symbol_decl): Remove unnecessary block
delimiters. Add auxiliary variables if a label is assigned to
a return variable. (gfc_gat_fake_result): If the symbol has an
assign attribute set declaration from the symbol's backend
declaration.
2020-05-27 Mark Eggleston <markeggleston@gnu.gcc.org>
gcc/testsuite/
PR fortran/50392
* gfortran.dg/pr50392.f: New test.
Jason Merrill [Tue, 26 May 2020 21:27:55 +0000 (17:27 -0400)]
c++: operator<=> and -Wzero-as-null-pointer-constant [PR95242]
In C++20, if there is no viable operator< available, lhs < rhs gets
rewritten to (lhs <=> rhs) < 0, where operator< for the comparison
categories is intended to accept literal 0 on the RHS but not other
integers. We don't want this to produce a warning from
-Wzero-as-null-pointer-constant.
gcc/cp/ChangeLog:
* call.c (build_new_op_1): Suppress
warn_zero_as_null_pointer_constant across comparison of <=> result
to 0.
Jason Merrill [Mon, 25 May 2020 23:04:05 +0000 (19:04 -0400)]
c++: Fix stdcall attribute in template. [PR95222]
Another case that breaks with my fix for PR90750: we shouldn't move type
attributes in TYPENAME context either, as there's no decl for them to move
to.
Jonathan Wakely [Wed, 27 May 2020 12:13:19 +0000 (13:13 +0100)]
libstdc++: Add new testcase for comparison category types
Comparing a comparison category type to anything except a literal 0 is
undefined. This verifies that at least some misuses are diagnosed at
compile time.
* testsuite/18_support/comparisons/categories/zero_neg.cc: New test.
Richard Biener [Mon, 25 May 2020 14:10:12 +0000 (16:10 +0200)]
Add debug (slp_tree) and dump infrastructure for this
This adds an alternate debug_dump_context similar to the one for
selftests but for interactive debugging routines. This allows
to share code between user-visible dumping via the dump_* API
and those debugging routines. The primary driver was SLP node
dumping which wasn't accessible from inside a gdb session up to
now.
2020-05-27 Richard Biener <rguenther@suse.de>
* dump-context.h (debug_dump_context): New class.
(dump_context): Make it friend.
* dumpfile.c (debug_dump_context::debug_dump_context):
Implement.
(debug_dump_context::~debug_dump_context): Likewise.
* tree-vect-slp.c: Include dump-context.h.
(vect_print_slp_tree): Dump a single SLP node.
(debug): New overload for slp_tree.
(vect_print_slp_graph): Rename from vect_print_slp_tree and
use that.
(vect_analyze_slp_instance): Adjust.