Marek Polacek [Tue, 15 Dec 2020 22:54:45 +0000 (17:54 -0500)]
c++: Remove dg-ice in constexpr-52830.C [PR52830]
It turned out that r11-5942 fixed this old PR, and it was detected
by one of the few dg-ice tests -- exactly the point of them! Now
the PR won't be opened until someone notices that it'd been fixed.
The patch failed to remove the dg-ice though, so now it XPASSes. Fixing
this now.
Jakub Jelinek [Tue, 15 Dec 2020 21:43:46 +0000 (22:43 +0100)]
match.pd: Optimize X / bool_range_Y to X [PR96094]
When the divisor is bool or has [0, 1] range, as division by
0 is UB, the only remaining option in valid programs is division by 1,
so we can optimize X / bool_range_Y into X.
Iain Buclaw [Tue, 15 Dec 2020 09:36:00 +0000 (10:36 +0100)]
d: Fix ICE in gimplify_expr, at gimplify.c (PR98277)
The DMD front-end shouldn't, but can sometimes leak manifest constants
in the AST passed to the code generator. To prevent this being an
issue, the setting of DECL_INITIAL has been moved to the point where the
CONST_DECL is used, rather than in the declaration handler.
gcc/d/ChangeLog:
PR d/98277
* decl.cc (DeclVisitor::visit (VarDeclaration *)): Move setting of
DECL_INITIAL for manifest constants to ...
(get_symbol_decl): ... here.
Ian Lance Taylor [Tue, 15 Dec 2020 04:41:30 +0000 (20:41 -0800)]
compiler: discard global sink variables with static initializers
This is specifically for the test fixedbugs/issue23781.go, which
builds a large static array. The code does compile and work without
this change, but it takes a long time and generates a large object
file. Discarding the unnecessary static initializer makes this test
much faster.
Ian Lance Taylor [Tue, 15 Dec 2020 05:26:33 +0000 (21:26 -0800)]
compiler: better error for unexpected digit
A digit character is not invalid in general, but it's not permitted
at the start of an identifier. Report a better error message.
The test case is issue11359.go in the source repo.
Keith Packard [Tue, 15 Dec 2020 17:39:24 +0000 (17:39 +0000)]
libstdc++: Support libc with stdio-only I/O in libstdc++
The current libstdc++ basic_file_stdio.cc code assumes a POSIX API
underneath the stdio implementation provided by the host libc. This
means that the host must provide a fairly broad POSIX file API,
including read, write, open, close, lseek and ioctl.
This patch changes basic_file_stdio.cc to only use basic ANSI-C stdio
functions, allowing it to be used with libc implementations like
picolibc which may not have a POSIX operating system underneath.
This is enabled by a new --enable-cstdio=stdio_pure configure option.
Aided-by: Jonathan Wakely <jwakely@redhat.com> Signed-off-by: Keith Packard <keithp@keithp.com>
libstdc++-v3/ChangeLog:
* acinclude.m4 (GLIBCXX_ENABLE_CSTDIO): Allow "stdio_pure"
option and define _GLIBCXX_USE_PURE_STDIO when it is used. Also
add "stdio_posix" option as an alias for "stdio".
* config/io/basic_file_stdio.cc [_GLIBCXX_USE_PURE_STDIO]: Only
use defined stdio entry points for all I/O operations, without
direct calls to underlying POSIX functions.
* config.h.in: Regenerate.
* configure: Regenerate.
Patrick Palka [Tue, 15 Dec 2020 17:10:26 +0000 (12:10 -0500)]
c++: Fix return type deduction during satisfaction
During satisfaction that's entered through the three-parameter version
of satisfy_declaration_constraints, current_function_decl gets set to
the dependent DECL_TEMPLATE_RESULT for sake of access checking. This
makes the predicate in_template_function return true during satisfaction
from this entrypoint, which in turn makes calls to mark_used exit early
before it does its full processing. This leads to us accepting the
invalid testcase below due to mark_used never attempting to deduce the
return type of A::foo() and detecting failure thereof.
It seems wrong for in_template_function to be true during instantiation
or during satisfaction, so this patch makes in_template_function inspect
the less volatile cfun->decl instead of current_function_decl.
gcc/cp/ChangeLog:
* pt.c (in_template_function): Inspect cfun->decl instead of
current_function_decl.
Nathan Sidwell [Mon, 14 Dec 2020 19:40:44 +0000 (11:40 -0800)]
c++: Modules Is Landing
This is the C++ modules implementation. Dropping in the real
module.cc file into the module-shaped hole I carved out. I include
some sanity tests for now. I'll commit the remaining tests when the
initial round of failures seems abated.
Nathan Sidwell [Mon, 14 Dec 2020 18:22:21 +0000 (10:22 -0800)]
c++: module mapper
To avoid always requiring an active mapper to connect to, we provide a
default in-process mapper with similar functionality to the sample
server. This is that code. Also included is the client-side
connection mechanism, which determines what server to use and how to
connect to it.
Nathan Sidwell [Mon, 14 Dec 2020 17:30:00 +0000 (09:30 -0800)]
Add c++tools
Part of our module implementation adds a sample mapper server, the
guts of which are used by the default in-process mapping of cc1plus.
Rather than add another executable to gcc/cp/, this creates a new
c++tools directory where this and any other c++ tools might live.
The toplevel changes are a subsequent commit, because ... git.
Jonathan Wakely [Tue, 15 Dec 2020 15:39:58 +0000 (15:39 +0000)]
libstdc++: Do not define constexpr assertions for C++11
There's no point even checking is_constant_evaluated() in C++11 mode,
because the 'if' statement used for the assertion wouldn't be valid in a
C++11 constexpr function anyway.
libstdc++-v3/ChangeLog:
* include/bits/c++config (__glibcxx_assert_1): Define as empty
for C++11.
Nathan Sidwell [Mon, 14 Dec 2020 16:10:27 +0000 (08:10 -0800)]
Add libcody
In order to separate compiler from build system, C++ Modules, as
implemented in GCC introduces a communication channel between those
two entities. This is implemented by libcody. It is anticipated that
other implementations will also implement this protocol, or use
libcody to provide it.
The testsuite for libstdc++ aims to skips test cases for which not all
required locales are installed. This patch adds missing directives about
required locales to one test case to avoid false positive test failures
on systems that have a partial set of locales installed.
Verified by test suite runs that this patch changes the test case from
FAIL to UNSUPPORTED when not all required locales are available and that
the test case will run and PASS when the necessary locales have been
added.
libstdc++-v3/ChangeLog:
* testsuite/22_locale/locale/cons/5.cc: Add missing directives
for required locales.
Jonathan Wakely [Tue, 15 Dec 2020 11:40:06 +0000 (11:40 +0000)]
libstdc++: Use init_priority attribute for Init object [PR 98108]
This causes the global objects that run the <iostream> initialization
code to be constructed earlier, which avoids some bugs in user code due
to incorrectly relying on static initialization order.
Pascal Obry [Sat, 7 Nov 2020 09:31:43 +0000 (10:31 +0100)]
[Ada] Fix serial port baud rate setting on GNU/Linux
gcc/ada/
* libgnat/g-sercom__linux.adb (Set): Use cfsetospeed and
cfsetispeed to set the baud rate. Clear non-blocking serial port
status when blocking is requested.
Eric Botcazou [Wed, 25 Nov 2020 18:32:15 +0000 (19:32 +0100)]
[Ada] Small adjustment to parameterization of System.Value_R
gcc/ada/
* doc/gnat_ugn/gnat_and_program_execution.rst: Minor fix.
* gnat_ugn.texi: Regenerate.
* libgnat/s-valuer.ads (Precision_Limit): New formal parameter.
* libgnat/s-valuer.adb (Precision_Limit): Remove.
(Scan_Decimal_Digits): Robustify overflow check.
(Scan_Integral_Digits): Likewise.
* libgnat/s-valrea.adb: Add assertion on the size of the unsigned
type and instantiate System.Value_R with the mantissa limit.
(Integer_to_Real): Add Extra parameter and take it into account.
(Scan_Real): Pass Extra to Integer_to_Real.
(Value_Real): Likewise.
* libgnat/s-valued.adb: Add assertion on the size of the unsigned
type and instantiate System.Value_R with the mantissa limit.
* libgnat/s-valuef.adb: Likewise.
Justin Squirek [Mon, 16 Nov 2020 14:08:51 +0000 (09:08 -0500)]
[Ada] Postcondition checks performed before finalization
gcc/ada/
* contracts.adb, contracts.ads (Build_Postconditions_Procedure):
Add declarations for Postcond_Enabled,
Result_Object_For_Postcondition, and
Return_Success_For_Postcond, and place all postconditions within
an if statement to control their execution for interactions when
cleanup actions get generated.
(Get_Postcond_Enabled): Created to fetch object declared to
handle new expansion of postconditions.
(Get_Result_Object_For_Postcond): Created to fetch object
declared to handle new expansion of postconditions.
(Get_Return_Success_For_Postcond): Created to fetch object
declared to handle new expansion of postconditions.
* einfo.adb, einfo.ads: Modify flag Stores_Attribute_Old_Prefix
to apply to constants, variables, and types.
* exp_ch6.adb (Add_Return): Add assignment to
Return_Success_For_Postcond.
(Expand_Non_Function_Return): Add assignment to
Return_Success_For_Postcond
(Expand_Simple_Function_Return): Add assignment to
Result_Object_For_Postcond and Return_Success_For_Postcond.
* exp_ch7.adb (Build_Finalization_Master): Mark finalization
masters which finalize types created store 'Old objects as
storing 'Old objects.
(Build_Finalizer): Created to generated a unified and special
expansion for finalization when postconditions are present.
(Build_Finalizer_Helper): Renamed Build_Finalizer and added
parameter to facilitate the creation of separate finalization
routines for 'Old objects and general objects.
(Create_Finalizer): Add condition for the insertion of the
finalizer spec to avoid malformed trees.
(Expand_Cleanup_Actions): Move _postconditions and related
declarations to the new declarative section. Fix the loop to
properly stop at the subprogram declaration for the
postconditions procedure and exclude its body from being moved
to the new list of declarations to avoid freezing issues.
* exp_prag.adb (Expand_Attributes): Mark temporary created to
store 'Old objects as storing a 'Old attribute.
* sem_ch6.adb (Find_What_Applies_To): Remove strange exception
to postconditions when traversing the scope stack.
* sem_prag.adb (Find_Related_Declaration_Or_Body): Use the newly
created Enclosing_HSS function to find the HSS for a potentially
nested statement.
* sem_util.adb, sem_util.ads (Declare_Indirect_Temp): Mark types
created to store 'Old objects as storing 'Old attributes.
(Enclosing_HSS): Created to find the enclosing handled sequence
of statements for a given statement.
* snames.ads-tmpl: Add multiple names to aid in the expansion of
finalization and to control the evaluation of postconditions.
Including _finalization_controller, a new routine to centralize
finalization actions and postcondition evaluation.
Eric Botcazou [Mon, 23 Nov 2020 14:42:08 +0000 (15:42 +0100)]
[Ada] Implement tiered support for floating-point input operations
gcc/ada/
* Makefile.rtl (GNATRTL_NONTASKING_OBJS): Remove s-fatsfl$(objext)
and add s-valflt$(objext), s-vallfl$(objext), s-valllf$(objext).
* exp_attr.adb (Find_Fat_Info): Merge Short_Float and Float cases.
* exp_imgv.adb (Expand_Value_Attribute): Replace RE_Value_Real with
RE_Value_Long_Long_Float for fixed-point types and use appropriate
base type for floating-point types.
* rtsfind.ads (RTU_Id): Remove System_Fat_IEEE_Long_Float,
System_Fat_IEEE_Short_Float and System_Val_Real, add System_Val_Flt,
System_Val_LFlt and System_Val_LLF.
(RE_Id): Remove RE_Attr_IEEE_Long, RE_Fat_IEEE_Long,
RE_Attr_IEEE_Short, RE_Fat_IEEE_Short, RE_Attr_Short_Float, add
RE_Value_Float, RE_Value_Long_Float, RE_Value_Long_Long_Float,
(RE_Unit_Table): Likewise.
* libgnat/a-ticoau.ads: Add with clause for Float_Aux and make the
package generic.
(Get): Change parameter types to Num.
(Put): Likewise.
(Gets): Likewise.
(Puts): Likewise.
* libgnat/a-ticoau.adb: Remove clause and renaming for Float_Aux.
(Get): Change parameter types to Num.
(Gets): Likewise.
(Put): Likewise.
(Puts): Likewise. Add conversion to Long_Long_Float.
* libgnat/a-ticoio.adb: Remove with clause for Ada.Text_IO, add with
clause for Float_Aux, add with and use clauses for System.Val_Flt,
System.Val_LFlt and System.Val_LLF. Instantiate Float_Aux and
Complex_Aux on Float, Long_Float, and Long_Long_Float.
(OK_Float): New boolean constant.
(OK_Long_Float): Likewise.
(Get): Call appropriate Get routine from auxiliary package.
(Get): Call appropriate Gets routine from auxiliary package.
(Put): Call appropriate Put routine from auxiliary package.
(Put): Call appropriate Puts routine from auxiliary package.
* libgnat/a-tideau.adb: Remove with and use clause for Float_Aux.
* libgnat/a-tifiau.adb: Likewise.
* libgnat/a-tifiio.adb: Add with and use clause for System.Val_LLF.
Instantiate Float_Aux on Long_Long_Float.
(Get): Adjust call to Get routine from auxiliary package.
(Get): Adjust call to Gets routine from auxiliary package.
(Put): Adjust call to Put routine from auxiliary package.
(Put): Adjust call to Puts routine from auxiliary package.
* libgnat/a-tifiio__128.adb: Likewise.
(Get): Likewise.
(Get): Likewise.
(Put): Likewise.
(Put): Likewise.
* libgnat/a-tiflau.ads: Make the package generic.
(Get): Change parameter type to Num.
(Put): Likewise.
(Gets): Likewise.
(Puts): Likewise.
* libgnat/a-tiflau.adb: Remove clauses for System.Val_Real.
(Get): Change parameter type to Num and call Scan routine.
(Gets): Likewise.
(Load_Real): Move to...
(Put): Change parameter type and add conversion to Long_Long_Float.
(Puts): Likewise.
* libgnat/a-tiflio.adb: Add with and use clauses for System.Val_Flt,
System.Val_LFlt and System.Val_LLF. Instantiate Float_Aux on Float,
Long_Float and Long_Long_Float.
(OK_Float): New boolean constant.
(OK_Long_Float): Likewise.
(Get): Call appropriate Get routine from auxiliary package.
(Get): Call previous variant.
(Get): Call appropriate Gets routine from auxiliary package.
(Put): Call appropriate Put routine from auxiliary package.
(Put): Call previous variant.
(Put): Call appropriate Puts routine from auxiliary package.
* libgnat/a-tigeau.ads (Load_Real): New procedure.
* libgnat/a-tigeau.adb (Load_Real): ...here.
* libgnat/a-wtcoau.ads: Add with clause for Float_Aux and make the
package generic.
(Get): Change parameter types to Num.
(Put): Likewise.
(Gets): Likewise.
(Puts): Likewise.
* libgnat/a-wtcoau.adb: Remove clause and renaming for Float_Aux.
(Get): Change parameter types to Num.
(Gets): Likewise.
(Put): Likewise.
(Puts): Likewise. Add conversion to Long_Long_Float.
* libgnat/a-wtcoio.ads: Remove use clause for Complex_Types and use
qualified names throughout accordingly.
* libgnat/a-wtcoio.adb: Remove clause for Ada.Unchecked_Conversion,
add with clause for Float_Aux, add clauses for System.Val_Flt,
System.Val_LFlt and System.Val_LLF. Add clause for Complex_Types.
Instantiate Float_Aux and Complex_Aux on Float, Long_Float, and
Long_Long_Float. Remove LLF subtype and TFT instantiation.
(OK_Float): New boolean constant.
(OK_Long_Float): Likewise.
(Get): Call appropriate Get routine from auxiliary package.
(Get): Call appropriate Gets routine from auxiliary package.
(Put): Call appropriate Put routine from auxiliary package.
(Put): Call appropriate Puts routine from auxiliary package.
* libgnat/a-wtdeau.adb: Remove with and use clause for Float_Aux.
* libgnat/a-wtfiau.adb: Likewise.
* libgnat/a-wtfiio.adb: Add with and use clause for System.Val_LLF.
Instantiate Float_Aux on Long_Long_Float.
(Get): Adjust call to Get routine from auxiliary package.
(Get): Adjust call to Gets routine from auxiliary package.
(Put): Adjust call to Put routine from auxiliary package.
(Put): Adjust call to Puts routine from auxiliary package.
* libgnat/a-wtfiio__128.adb: Likewise.
(Get): Likewise.
(Get): Likewise.
(Put): Likewise.
(Put): Likewise.
* libgnat/a-wtflau.ads: Make the package generic.
(Get): Change parameter type to Num.
(Put): Likewise.
(Gets): Likewise.
(Puts): Likewise.
* libgnat/a-wtflau.adb: Remove clauses for System.Val_Real.
(Get): Change parameter type to Num and call Scan routine. Set
Ptr parameter lazily.
(Gets): Likewise.
(Load_Real): Move to...
(Put): Change parameter type and add conversion to Long_Long_Float.
Bump buffer length to Max_Real_Image_Length.
(Puts): Likewise.
* libgnat/a-wtflio.adb: Add with and use clauses for System.Val_Flt,
System.Val_LFlt and System.Val_LLF. Instantiate Float_Aux on Float,
Long_Float and Long_Long_Float.
(OK_Float): New boolean constant.
(OK_Long_Float): Likewise.
(Get): Call appropriate Get routine from auxiliary package. Add
pragma Unsuppress (Range_Check) and manual validity check.
(Get): Call appropriate Gets routine from auxiliary package. Add
pragma Unsuppress (Range_Check) and manual validity check.
(Put): Call appropriate Put routine from auxiliary package.
(Put): Call appropriate Puts routine from auxiliary package.
* libgnat/a-wtgeau.ads (Load_Real): New procedure.
* libgnat/a-wtgeau.adb (Load_Real): ...here.
* libgnat/a-ztcoau.ads: Add with clause for Float_Aux and make the
package generic.
(Get): Change parameter types to Num.
(Put): Likewise.
(Gets): Likewise.
(Puts): Likewise.
* libgnat/a-ztcoau.adb: Remove clause and renaming for Float_Aux.
(Get): Change parameter types to Num.
(Gets): Likewise.
(Put): Likewise.
(Puts): Likewise. Add conversion to Long_Long_Float.
* libgnat/a-ztcoio.ads: Remove use clause for Complex_Types and use
qualified names throughout accordingly.
* libgnat/a-ztcoio.adb: Remove clause for Ada.Unchecked_Conversion,
add with clause for Float_Aux, add clauses for System.Val_Flt,
System.Val_LFlt and System.Val_LLF. Add clause for Complex_Types.
Instantiate Float_Aux and Complex_Aux on Float, Long_Float, and
Long_Long_Float. Remove LLF subtype and TFT instantiation.
(OK_Float): New boolean constant.
(OK_Long_Float): Likewise.
(Get): Call appropriate Get routine from auxiliary package.
(Get): Call appropriate Gets routine from auxiliary package.
(Put): Call appropriate Put routine from auxiliary package.
(Put): Call appropriate Puts routine from auxiliary package.
* libgnat/a-ztdeau.adb: Remove with and use clause for Float_Aux.
* libgnat/a-ztfiau.adb: Likewise.
* libgnat/a-ztfiio.adb: Add with and use clause for System.Val_LLF.
Instantiate Float_Aux on Long_Long_Float.
(Get): Adjust call to Get routine from auxiliary package.
(Get): Adjust call to Gets routine from auxiliary package.
(Put): Adjust call to Put routine from auxiliary package.
(Put): Adjust call to Puts routine from auxiliary package.
* libgnat/a-ztfiio__128.adb: Likewise.
(Get): Likewise.
(Get): Likewise.
(Put): Likewise.
(Put): Likewise.
* libgnat/a-ztflau.ads: Make the package generic.
(Get): Change parameter type to Num.
(Put): Likewise.
(Gets): Likewise.
(Puts): Likewise.
* libgnat/a-ztflau.adb: Remove clauses for System.Val_Real.
(Get): Change parameter type to Num and call Scan routine. Set
Ptr parameter lazily.
(Gets): Likewise.
(Load_Real): Move to...
(Put): Change parameter type and add conversion to Long_Long_Float.
Bump buffer length to Max_Real_Image_Length.
(Puts): Likewise.
* libgnat/a-ztflio.adb: Add with and use clauses for System.Val_Flt,
System.Val_LFlt and System.Val_LLF. Instantiate Float_Aux on Float,
Long_Float and Long_Long_Float.
(OK_Float): New boolean constant.
(OK_Long_Float): Likewise.
(Get): Call appropriate Get routine from auxiliary package. Add
pragma Unsuppress (Range_Check) and manual validity check.
(Get): Call appropriate Gets routine from auxiliary package. Add
pragma Unsuppress (Range_Check) and manual validity check.
(Put): Call appropriate Put routine from auxiliary package.
(Put): Call appropriate Puts routine from auxiliary package.
* libgnat/a-ztgeau.ads (Load_Real): New procedure.
* libgnat/a-ztgeau.adb (Load_Real): ...here.
* libgnat/s-fatsfl.ads: Delete.
* libgnat/s-valflt.ads: New package.
* libgnat/s-vallfl.ads: Likewise.
* libgnat/s-valllf.ads: Likewise.
* libgnat/s-valrea.ads: Make generic. Add assertions, defensive
code and clarify intent.
(Scan_Real): Change parameter type to Num.
(Value_Real): Likewise.
* libgnat/s-valrea.adb: Instantiate Value_R on Uns.
(Integer_to_Real): Change parameter and result to Num.
Call Float_Control.Reset only if the mantissa is 64 bits. Use
a divide to compute the final value if the scale is negative.
(Scan_Real): Change result to Num.
(Value_Real): Likewise.
* libgnat/s-valuer.adb: Add assertions, defensive code and
clarify intent.
(F_Limit): Delete.
(I_Limit): Likewise.
(Precision_Limit): Always use the integer limit.
* libgnat/s-fatgen.adb: Add pragma Annotate.
Arnaud Charlet [Mon, 16 Nov 2020 17:10:25 +0000 (18:10 +0100)]
[Ada] Transform_Function_Array issues
gcc/ada/
* exp_ch6.adb (Build_Procedure_Body_Form): Adjust, the
declaration of the procedure form is now insert before the
original function body rather than after.
(Expand_N_Subprogram_Declaration): Deal with private types whose
full views are arrays.
* exp_unst.adb (Unnest_Subprogram): Deal with private types.
(Needs_Fat_Pointer): Code cleanup.
* freeze.adb (Freeze_Subprogram): Ditto.
* exp_util.adb (Build_Procedure_Form): Insert the procedure form
decl before and not after.
* sem_ch6.adb (Analyze_Subprogram_Body_Helper): Build missing
spec when needed for Transform_Function_Array.
* sem_util.adb (Get_Fullest_View): Deal with null entity.
Piotr Trojanek [Mon, 23 Nov 2020 23:31:49 +0000 (00:31 +0100)]
[Ada] Refactor repeated code for Callable and Terminated attributes
gcc/ada/
* sem_attr.adb (Analyze_Attribute): Merge identical code for
Callable and Terminated attributes; refactor calls to Set_Etype
occurring in both THEN and ELSE branches of an IF statement for
attribute Storage_Size.
Piotr Trojanek [Mon, 23 Nov 2020 23:00:17 +0000 (00:00 +0100)]
[Ada] Refactor repeated code for fixed point attributes
gcc/ada/
* sem_attr.adb (Analyze_Attribute): Consistently call
Check_Fixed_Point_Type before checking the number of attribute
expressions (like it is done for floating point types); reuse
Check_Fixed_Point_Type_0.
Jakub Jelinek [Tue, 15 Dec 2020 09:16:08 +0000 (10:16 +0100)]
i386: Fix up -march=x86-64-v[234] vs. target attribute [PR98274]
The following testcase fails to compile. The problem is that
when ix86_option_override_internal is called the first time for command
line, it sees -mtune= wasn't present on the command line and so as fallback
sets ix86_tune_string to ix86_arch_string value ("x86-64-v2"), but
ix86_tune_specified is false, so we don't find the tuning in the table
but don't error on it.
When processing the target attribute, ix86_tune_string is what
it was earlier left with, but this time ix86_tune_specified is true and
so we error on it.
The following patch does what is done already e.g. for "x86-64" march,
in particular default the tuning to "generic".
2020-12-15 Jakub Jelinek <jakub@redhat.com>
PR target/98274
* config/i386/i386-options.c (ix86_option_override_internal): Set
ix86_tune_string to "generic" even when it wasn't specified and
ix86_arch_string is "x86-64-v2", "x86-64-v3" or "x86-64-v4".
Remove useless {}s around a single statement.
Jakub Jelinek [Tue, 15 Dec 2020 08:51:28 +0000 (09:51 +0100)]
i386: Make -march=x86-64-v[234] behave more like other -march= options
If somebody has -march=x86-64-v2 (or -v3 or -v4) in $CFLAGS, $CXXFLAGS etc.,
then -m32 or -mabi=ms stops working.
What is worse, if one configures gcc --with-arch-64=x86-64-v2 (or -v3 or -v4),
then -mabi=ms stops working.
I think that is a nightmare user experience. It is ok that x86-64-v[234]
behave slightly different from other -march= options (in that they imply
unless overridden -mtune=generic rather then -mtune= equal to the -march
argument), but the error when one mixes it with -mabi=ms, or -m32 doesn't
improve anything.
It is true that the exact option set is only defined in the x86-64 psABI
(IMHO that is a mistake too, we should copy that into the GCC documentation
like we document it for any other -march= option), but there is no reason
why that exact set of CPU features can't be used for other ABIs, it is just
a set of CPU features. If we add micro-architecture levels to the 32-bit
ABI (I doubt anyone wants to do that, but just hypothetically), then those
micro-architecture levels wouldn't certainly be called x86-64-v* but perhaps
i386-v*.
In the tests, __GCC_HAVE_SYNC_COMPARE_AND_SWAP_16 can't be expected on -m32
not because the CPU feature wouldn't be set, but because the instruction
is 64-bit only and 32-bit code doesn't have __int128 etc. support.
2020-12-15 Jakub Jelinek <jakub@redhat.com>
* config/i386/i386-options.c (ix86_option_override_internal): Don't
error on -march=x86-64-v[234] with -m32 or -mabi=ms.
* config.gcc: Don't reject --with-arch=x86-64-v[234] or
--with-arch_32=x86-64-v[234].
* doc/invoke.texi (-march=x86-64-v[234]): Document what the option
does for other ABIs.
* gcc.target/i386/x86-64-v2.c: Don't expect
__GCC_HAVE_SYNC_COMPARE_AND_SWAP_16 to be defined with -m32.
* gcc.target/i386/x86-64-v2-other.c: New test.
* gcc.target/i386/x86-64-v2-msabi.c: New test.
* gcc.target/i386/x86-64-v3.c: Fix a comment pasto. Don't expect
__GCC_HAVE_SYNC_COMPARE_AND_SWAP_16 to be defined with -m32.
* gcc.target/i386/x86-64-v3-other.c: New test.
* gcc.target/i386/x86-64-v3-msabi.c: New test.
* gcc.target/i386/x86-64-v4.c:Don't expect
__GCC_HAVE_SYNC_COMPARE_AND_SWAP_16 to be defined with -m32.
* gcc.target/i386/x86-64-v4-other.c: New test.
* gcc.target/i386/x86-64-v4-msabi.c: New test.
Ilya Leoshkevich [Thu, 10 Dec 2020 23:54:49 +0000 (00:54 +0100)]
aix: Fixinclude updates [PR98208]
After 92648faa1cb2 ("aix: Fixinclude") make check-fixincludes began to
fail (at least on gcc121 machine). Fix by updating fixincludes/tests
and rerunning genfixes.
* fixincl.x: Rerun genfixes.
* inclhack.def(aix_physadr_t): Change test_text to something
that needs to be replaced.
* tests/base/sys/types.h(aix_physadr_t): Add expectation.
Resolves:
PR middle-end/98166 - bogus -Wmismatched-dealloc on user-defined allocator and inlining
PR c++/57111 - 57111 - Generalize -Wfree-nonheap-object to delete
PR middle-end/98160 - ICE in default_tree_printer at gcc/tree-diagnostic.c:270
gcc/ChangeLog:
PR middle-end/98166
PR c++/57111
PR middle-end/98160
* builtins.c (check_access): Call tree_inlined_location
fndecl_alloc_p): Handle BUILT_IN_ALIGNED_ALLOC and
BUILT_IN_GOMP_ALLOC.
call_dealloc_p): Remove unused function.
(new_delete_mismatch_p): Call valid_new_delete_pair_p and rework.
(matching_alloc_calls_p): Handle built-in deallocation functions.
(warn_dealloc_offset): Corrct the handling of user-defined operators
delete.
(maybe_emit_free_warning): Avoid assuming expression is a decl.
Simplify.
* doc/extend.texi (attribute malloc): Update.
* tree-ssa-dce.c (valid_new_delete_pair_p): Factor code out into
valid_new_delete_pair_p in tree.c.
* tree.c (tree_inlined_location): Define new function.
(valid_new_delete_pair_p): Define.
* tree.h (tree_inlined_location): Declare.
(valid_new_delete_pair_p): Declare.
gcc/c-family/ChangeLog:
PR middle-end/98166
PR c++/57111
PR middle-end/98160
* c-attribs.c (maybe_add_noinline): New function.
(handle_malloc_attribute): Call it. Use ATTR_FLAG_INTERNAL.
Implicitly add attribute noinline to functions not declared inline
and warn on those.
PR middle-end/98166
PR c++/57111
PR middle-end/98160
* g++.dg/warn/Wmismatched-dealloc-2.C: Adjust test of expected warning.
* g++.dg/warn/Wmismatched-new-delete.C: Same.
* gcc.dg/Wmismatched-dealloc.c: Same.
* c-c++-common/Wfree-nonheap-object-2.c: New test.
* c-c++-common/Wfree-nonheap-object-3.c: New test.
* c-c++-common/Wfree-nonheap-object.c: New test.
* c-c++-common/Wmismatched-dealloc.c: New test.
* g++.dg/warn/Wfree-nonheap-object-3.C: New test.
* g++.dg/warn/Wfree-nonheap-object-4.C: New test.
* g++.dg/warn/Wmismatched-dealloc-2.C: New test.
* g++.dg/warn/Wmismatched-new-delete-2.C: New test.
* g++.dg/warn/Wmismatched-new-delete.C: New test.
* gcc.dg/Wmismatched-dealloc-2.c: New test.
* gcc.dg/Wmismatched-dealloc-3.c: New test.
* gcc.dg/Wmismatched-dealloc.c: New test.
Wilco Dijkstra [Thu, 3 Dec 2020 18:40:34 +0000 (18:40 +0000)]
AArch64: Add support for --with-tune
Add support for --with-tune. Like --with-cpu and --with-arch, the argument is
validated and transformed into a -mtune option to be processed like any other
command-line option. --with-tune has no effect if a -mcpu or -mtune option
is used. The validating code didn't allow --with-cpu=native, so explicitly
allow that.
Co-authored-by: Delia Burduv <delia.burduv@arm.com>
Bootstrap OK, regress pass, OK to commit?
Justin Squirek [Fri, 20 Nov 2020 13:11:12 +0000 (08:11 -0500)]
[Ada] Incorrect accessibility level on type in formal package
gcc/ada/
* sem_util.adb, sem_util.ads (In_Generic_Formal_Package):
Created to identify type declarations occurring within generic
formal packages.
* sem_res.adb (Resolve_Allocator): Add condition to avoid
emitting an error for allocators when the type being allocated
is class-wide and from a generic formal package.
Eric Botcazou [Sat, 21 Nov 2020 23:54:18 +0000 (00:54 +0100)]
[Ada] Adjust again previous change to System.Fat_Gen
gcc/ada/
* libgnat/s-fatgen.adb: Add with clause for Interfaces and use
type clause for Interfaces.Unsigned_64.
(Small): Comment out.
(Tiny): Likewise.
(Tiny16): New integer constant.
(Tiny32): Likewise.
(Tiny64): Likewise.
(Tiny80): New integer array constant.
(Pred): Declare a local overlay for Tiny.
(Succ): Likewise.
Eric Botcazou [Fri, 20 Nov 2020 20:29:13 +0000 (21:29 +0100)]
[Ada] Fix internal error on bit-packed array in Volatile_Full_Access record
gcc/ada/
* exp_pakd.adb (Expand_Bit_Packed_Element_Set): Fix again packed
array type in complex cases where array is Volatile.
* exp_util.adb (Remove_Side_Effects): Do not force a renaming to
be handled by the back-end.
Eric Botcazou [Fri, 20 Nov 2020 18:33:21 +0000 (19:33 +0100)]
[Ada] Adjust previous change to System.Fat_Gen
gcc/ada/
* libgnat/s-fatgen.adb: Remove use clause for
System.Unsigned_Types.
(Scaling): Add renaming of System.Unsigned_Types and use type
clause for Long_Long_Unsigned.
Gary Dismukes [Thu, 19 Nov 2020 20:18:39 +0000 (15:18 -0500)]
[Ada] Fix documentation of -gnatw.K switch (activates => disables)
gcc/ada/
* doc/gnat_ugn/building_executable_programs_with_gnat.rst:
Correct documentation of the -gnatw.K switch to say that it
disables rather than activates the warning.
* gnat_ugn.texi: Regenerate.
Gary Dismukes [Wed, 18 Nov 2020 23:06:14 +0000 (18:06 -0500)]
[Ada] Additional fixes for Default_Initial_Condition
gcc/ada/
* exp_aggr.adb (Build_Array_Aggr_Code.Gen_Assign): Move
generation of the call for DIC check past the optional
generation of calls to controlled Initialize procedures.
* exp_ch3.adb
(Build_Array_Init_Proc.Init_One_Dimension.Possible_DIC_Call):
Suppress generation of a DIC call when the array component type
is controlled. The call will now be generated later inside the
array's DI (Deep_Initialize) procedure.
* exp_ch7.adb
(Make_Deep_Array_Body.Build_Initialize_Statements): Generate a
DIC call (when needed by the array component type) after any
call to the component type's controlled Initialize procedure, or
generate the DIC call by itself if there's no Initialize to
call.
* sem_aggr.adb (Resolve_Record_Aggregate.Add_Association):
Simplify condition to only test Is_Box_Init_By_Default (previous
condition was overkill, as well as incorrect in some cases).
* sem_elab.adb (Active_Scenarios.Output_Call): For
Default_Initial_Condition, suppress call to
Output_Verification_Call when the subprogram is a partial DIC
procedure.
Eric Botcazou [Wed, 18 Nov 2020 20:42:18 +0000 (21:42 +0100)]
[Ada] Fix couple of bugs in the implementation of Round attribute
gcc/ada/
* exp_attr.adb (Expand_N_Attribute_Reference) <Attribute_Round>:
Adjust commentary and set the Rounded_Result flag on the type
conversion node when the node is needed.
* exp_ch4.adb (Expand_N_Type_Conversion): Minor tweak.
(Fixup_Universal_Fixed_Operation): Look through the type conversion
only when it is to Universal_Real.
* exp_fixd.adb: Remove with and use clauses for Snames.
(Build_Divide): Remove redundant test.
(Expand_Convert_Float_To_Fixed): Use Rounded_Result flag on the
node to set the truncation parameter.
projects / thirdparty / gcc.git / log
