pmderodat [Tue, 13 Aug 2019 08:07:18 +0000 (08:07 +0000)]
[Ada] Spurious error on nested instantiation
This fixes a spurious error given by the compiler for a call to a
subprogram which is the formal subprogram parameter of a generic
package, if the generic package is instantiated in the body of an
enclosing generic package with two formal types and two formal
subprogram parameter homonyms taking them, and this instantiation takes
one the two formal types as actual, and the enclosing generic package is
instantiated on the same actual type with a single actual subprogram
parameter, and the aforementioned call is overloaded.
In this case, the renaming generated for the actual subprogram parameter
in the nested instantiation is ambiguous and must be disambiguated using
the corresponding formal parameter of the enclosing instantiation,
otherwise a (sub)type mismatch is created and later subprogram
disambiguation is not really possible.
2019-08-13 Eric Botcazou <ebotcazou@adacore.com>
gcc/ada/
* sem_ch4.adb (Analyze_One_Call): Remove bypass for type
mismatch in nested instantiations.
* sem_ch8.adb (Find_Nearer_Entity): New function.
(Find_Renamed_Entity): Use it to disambiguate the candidates for
the renaming generated for an instantiation when it is
ambiguous.
gcc/testsuite/
* gnat.dg/generic_inst9.adb, gnat.dg/generic_inst9.ads,
gnat.dg/generic_inst9_pkg1-operator.ads,
gnat.dg/generic_inst9_pkg1.ads, gnat.dg/generic_inst9_pkg2.adb,
gnat.dg/generic_inst9_pkg2.ads: New testcase.
pmderodat [Tue, 13 Aug 2019 08:07:08 +0000 (08:07 +0000)]
[Ada] Fix incorrect binding to MapViewOfFile in s-win32.ads
Despite the "dw" prefix on the name, the dwNumberOfBytesToMap argument
to MapViewOfFile was changed from DWORD to SIZE_T when 64bit Windows
came about.
This change adjusts the binding we have for it in System.Win32
accordingly. For consistency with established practice, an s-win32
specific version of size_t is introduced and g-sercom__mingw.adb is
adjusted to disambiguate between this new size_t and the one already
exposed in System.CRTL.
2019-08-13 Olivier Hainque <hainque@adacore.com>
gcc/ada/
* libgnat/s-win32.ads: Define size_t and fix the MapViewOfFile
binding to use it instead of DWORD for the dwNumberOfBytesToMap
argument.
* libgnat/g-sercom__mingw.adb (Read): State which definition of
size_t to fetch in call to Last_Index.
pmderodat [Tue, 13 Aug 2019 08:06:55 +0000 (08:06 +0000)]
[Ada] Avoid spurious errors on dimensionality checking in GNATprove
Complete the partial treatment that was started in r273405. Instead of
checking for the special case of nodes inside inlined bodies at the call
site, check for this condition inside the top-level procedures called
for dimensionality checking.
There is no impact on compilation.
2019-08-13 Yannick Moy <moy@adacore.com>
gcc/ada/
* sem_dim.adb (Analyze_Dimension,
Analyze_Dimension_Array_Aggregate, Analyze_Dimension_Call,
Analyze_Dimension_Extension_Or_Record_Aggregate): Return
immediately when inside an inlined body.
* sem_res.adb (Resolve_Call): Remove special checking now done
inside Analyze_Dimension_Call.
pmderodat [Tue, 13 Aug 2019 08:06:50 +0000 (08:06 +0000)]
[Ada] Disable anonymous allocator warning for library-level objects
This patch modifies the behavior of anonymous allocator warnings so that
they no longer get triggered in the case of an object declaration at
library-level.
2019-08-13 Justin Squirek <squirek@adacore.com>
gcc/ada/
* exp_ch4.adb (Expand_N_Allocator): Add condition to detect
library-level object declarations
gcc/testsuite/
* gnat.dg/anon3.adb, gnat.dg/anon3.ads: New testcase.
pmderodat [Tue, 13 Aug 2019 08:06:39 +0000 (08:06 +0000)]
[Ada] Fix spurious error on nested instantiation with inlining
This prevents the compiler from issuing a spurious error in a convoluted
case where a child generic package declared in an enclosing parent
generic package, which contains a second child generic package, contains
an inlined subprogram and the second child generic package contains an
instantiation of the first, and the enclosing parent generic package is
instantiated with inlining across units enabled (-gnatn[12]).
The problem is that the compiler attempts to instantiate the body of the
first child generic package in the context of the enclosing parent
generic package, instead of doing it in the context of the instantiation
of the parent generic package, because of the presence of the inlined
subprogram.
2019-08-13 Eric Botcazou <ebotcazou@adacore.com>
gcc/ada/
* exp_ch6.adb (Expand_Call_Helper): If back-end inlining is
enabled, also instantiate the body of a generic unit containing
a subprogram subject to aspect/pragma Inline_Always at
optimization level zero.
* sem_ch12.adb (Might_Inline_Subp): Minor tweak.
(Analyze_Package_Instantiation): Do not instantiate the package
body because of inlining considerations if the instantiation is
done in a generic unit. Move around similar condition involving
the main unit. Add test on Back_End_Inlining to processing for
front-end inlining.
gcc/testsuite/
* gnat.dg/generic_inst8.adb, gnat.dg/generic_inst8.ads,
gnat.dg/generic_inst8_g.adb, gnat.dg/generic_inst8_g.ads: New
testcase.
pmderodat [Tue, 13 Aug 2019 08:06:34 +0000 (08:06 +0000)]
[Ada] Wrong initialization of Offset_To_Top in secondary DT
The compiler does not initialize well the runtime information required
to perform at runtime interface conversions on derivations of tagged
types that implement interfaces and have variable size components.
2019-08-13 Javier Miranda <miranda@adacore.com>
gcc/ada/
* exp_disp.adb (Make_Secondary_DT): Handle record type
derivations that have interface components located at fixed
positions and interface components located at variable offset.
The offset of components located at fixed positions is computed
using the dummy object (similar to the case where all the
interface components are located at fixed positions).
(Make_DT): Build the dummy object for all tagged types that
implement interface types (that is, build it also for types with
variable size components), and use the dummy object to compute
the offset of all tag components located at fixed positions when
initializing the Interface_Table object.
gcc/testsuite/
* gnat.dg/tag2.adb, gnat.dg/tag2_pkg.ads: New testcase.
pmderodat [Tue, 13 Aug 2019 08:06:29 +0000 (08:06 +0000)]
[Ada] Add ability to list compiler switches with gnatcmd
This patch adds a new switch (--help-ada) to gnatcmd for displaying all
the availible build flags as well as fixing some minor differences in
output between implicit and explicit help for gnatcmd (e.g. "gnat" vs
"gnat --help").
$ gnat
GNAT Pro 20.0w (19940713)
Copyright 1996-2019, Free Software Foundation, Inc.
To list Ada build switches use --help-ada
List of available commands
gnat bind gnatbind
gnat chop gnatchop
gnat clean gnatclean
gnat compile gnatmake -f -u -c
gnat check gnatcheck
gnat elim gnatelim
gnat find gnatfind
gnat krunch gnatkr
gnat link gnatlink
gnat list gnatls
gnat make gnatmake
gnat metric gnatmetric
gnat name gnatname
gnat preprocess gnatprep
gnat pretty gnatpp
gnat stack gnatstack
gnat stub gnatstub
gnat test gnattest
gnat xref gnatxref
Report bugs to report@adacore.com
$ gnat --help
GNAT Pro 20.0w (19940713)
Copyright 1996-2019, Free Software Foundation, Inc.
To list Ada build switches use --help-ada
List of available commands
gnat bind gnatbind
gnat chop gnatchop
gnat clean gnatclean
gnat compile gnatmake -f -u -c
gnat check gnatcheck
gnat elim gnatelim
gnat find gnatfind
gnat krunch gnatkr
gnat link gnatlink
gnat list gnatls
gnat make gnatmake
gnat metric gnatmetric
gnat name gnatname
gnat preprocess gnatprep
gnat pretty gnatpp
gnat stack gnatstack
gnat stub gnatstub
gnat test gnattest
gnat xref gnatxref
Report bugs to report@adacore.com
$ gnat --help-ada
Usage: gnat switches sfile
sfile Source file name
-g Generate debugging information
-Idir Specify source files search path
-I- Do not look for sources in current directory
-O[0123] Control the optimization level
-gnata Assertions enabled. Pragma Assert/Debug to be activated
-gnatA Avoid processing gnat.adc, if present file will be ignored
-gnatb Generate brief messages to stderr even if verbose mode set
-gnatB Assume no bad (invalid) values except in 'Valid attribute
-gnatc Check syntax and semantics only (no code generation)
-gnatC Generate CodePeer intermediate format (no code generation)
-gnatd? Compiler debug option ? ([.]a-z,A-Z,0-9), see debug.adb
-gnatD Debug expanded generated code (max line length = 72)
-gnatDnn Debug expanded generated code (max line length = nn)
-gnateA Aliasing checks on subprogram parameters
-gnatec=? Specify configuration pragmas file, e.g. -gnatec=/x/f.adc
-gnateC Generate CodePeer messages (ignored without -gnatcC)
-gnated Disable synchronization of atomic variables
-gnateD? Define or redefine preprocessing symbol, e.g. -gnateDsym=val
-gnateE Generate extra information in exception messages
-gnatef Full source path in brief error messages
-gnateF Check overflow on predefined Float types
-gnateG Generate preprocessed source
-gnateinn Set maximum number of instantiations to nn
-gnateInn Index in multi-unit source, e.g. -gnateI2
-gnatel Turn on info messages on generated Elaborate[_All] pragmas
-gnateL Turn off info messages on generated Elaborate[_All] pragmas
-gnatem=? Specify mapping file, e.g. -gnatem=mapping
-gnatep=? Specify preprocessing data file, e.g. -gnatep=prep.data
-gnateP Pure/Prelaborate errors generate warnings rather than errors
-gnateS Generate SCO (Source Coverage Obligation) information
-gnatet=? Write target dependent information file ?, e.g. gnatet=tdf
-gnateT=? Read target dependent information file ?, e.g. gnateT=tdf
-gnateu Ignore unrecognized style/validity/warning switches
-gnateV Validity checks on subprogram parameters
-gnateY Ignore all Style_Checks pragmas in source
-gnatE Dynamic elaboration checking mode enabled
-gnatf Full errors. Verbose details, all undefined references
-gnatF Force all import/export external names to all uppercase
-gnatg GNAT implementation mode (used for compiling GNAT units)
-gnatG Output generated expanded code (max line length = 72)
-gnatGnn Output generated expanded code (max line length = nn)
-gnath Output this usage (help) information
-gnatH Legacy elaboration checking mode enabled
-gnati? Identifier char set (?=1/2/3/4/5/8/9/p/f/n/w)
-gnatI Ignore all representation clauses
-gnatjnn Format error and warning messages to fit nn character lines
-gnatJ Relaxed elaboration checking mode enabled
-gnatk Limit file names to nn characters (k = krunch)
-gnatl Output full source listing with embedded error messages
-gnatl=f Output full source listing to specified file
-gnatL List corresponding source text in -gnatG or -gnatD output
-gnatmnn Limit number of detected errors/warnings to nn (1-999999)
-gnatn[?] Enable pragma Inline across units (?=1/2 for moderate/full)
-gnato0 Disable overflow checking
-gnato Enable overflow checking in STRICT (-gnato1) mode (default)
-gnato? Enable overflow checks in STRICT/MINIMIZED/ELIMINATED (1/2/3) mode
-gnato?? Set mode for general/assertion expressions separately
-gnatp Suppress all checks
-gnatP Generate periodic calls to System.Polling.Poll
-gnatq Don't quit, try semantics, even if parse errors
-gnatQ Don't quit, write ali/tree file even if compile errors
-gnatr Treat pragma Restrictions as Restriction_Warnings
-gnatR? List rep info (?=0/1/2/3/4/e/m for none/types/all/sym/cg/ext/mech)
-gnatR?j List rep info in the JSON data interchange format
-gnatR?s List rep info to file.rep instead of standard output
-gnats Syntax check only
-gnatS Print listing of package Standard
-gnatt Tree output file to be generated
-gnatTnn All compiler tables start at nn times usual starting size
-gnatu List units for this compilation
-gnatU Enable unique tag for error messages
-gnatv Verbose mode. Full error output with source lines to stdout
-gnatVxx Enable selected validity checking mode, xx = list of parameters:
a turn on all validity checking options
c turn on checking for copies
C turn off checking for copies
d turn on default (RM) checking
D turn off default (RM) checking
e turn on checking for elementary components
E turn off checking for elementary components
f turn on checking for floating-point
F turn off checking for floating-point
i turn on checking for in params
I turn off checking for in params
m turn on checking for in out params
M turn off checking for in out params
n turn off all validity checks (including RM)
o turn on checking for operators/attributes
O turn off checking for operators/attributes
p turn on checking for parameters
P turn off checking for parameters
r turn on checking for returns
R turn off checking for returns
s turn on checking for subscripts
S turn off checking for subscripts
t turn on checking for tests
T turn off checking for tests
-gnatwxx Enable selected warning modes, xx = list of parameters:
* indicates default setting
+ indicates warning flag included in -gnatwa
a turn on all info/warnings marked below with +
A turn off all optional info/warnings
.a*+ turn on warnings for failing assertion
.A turn off warnings for failing assertion
_a*+ turn on warnings for anonymous allocators
_A turn off warnings for anonymous allocators
b+ turn on warnings for bad fixed value (not multiple of small)
B* turn off warnings for bad fixed value (not multiple of small)
.b*+ turn on warnings for biased representation
.B turn off warnings for biased representation
c+ turn on warnings for constant conditional
C* turn off warnings for constant conditional
.c+ turn on warnings for unrepped components
.C* turn off warnings for unrepped components
d turn on warnings for implicit dereference
D* turn off warnings for implicit dereference
.d turn on tagging of warnings with -gnatw switch
.D* turn off tagging of warnings with -gnatw switch
e treat all warnings (but not info) as errors
.e turn on every optional info/warning (no exceptions)
E treat all run-time warnings as errors
f+ turn on warnings for unreferenced formal
F* turn off warnings for unreferenced formal
.f turn on warnings for suspicious Subp'Access
.F* turn off warnings for suspicious Subp'Access
g*+ turn on warnings for unrecognized pragma
G turn off warnings for unrecognized pragma
.g turn on GNAT warnings
h turn on warnings for hiding declarations
H* turn off warnings for hiding declarations
.h turn on warnings for holes in records
.H* turn off warnings for holes in records
i*+ turn on warnings for implementation unit
I turn off warnings for implementation unit
.i*+ turn on warnings for overlapping actuals
.I turn off warnings for overlapping actuals
j+ turn on warnings for obsolescent (annex J) feature
J* turn off warnings for obsolescent (annex J) feature
.j+ turn on warnings for late dispatching primitives
.J* turn off warnings for late dispatching primitives
k+ turn on warnings on constant variable
K* turn off warnings on constant variable
.k turn on warnings for standard redefinition
.K* turn off warnings for standard redefinition
l turn on warnings for elaboration problems
L* turn off warnings for elaboration problems
.l turn on info messages for inherited aspects
.L* turn off info messages for inherited aspects
m+ turn on warnings for variable assigned but not read
M* turn off warnings for variable assigned but not read
.m*+ turn on warnings for suspicious modulus value
.M turn off warnings for suspicious modulus value
n* normal warning mode (cancels -gnatws/-gnatwe)
.n turn on info messages for atomic synchronization
.N* turn off info messages for atomic synchronization
o* turn on warnings for address clause overlay
O turn off warnings for address clause overlay
.o turn on warnings for out parameters assigned but not read
.O* turn off warnings for out parameters assigned but not read
p+ turn on warnings for ineffective pragma Inline in frontend
P* turn off warnings for ineffective pragma Inline in frontend
.p+ turn on warnings for suspicious parameter order
.P* turn off warnings for suspicious parameter order
q*+ turn on warnings for questionable missing parenthesis
Q turn off warnings for questionable missing parenthesis
.q+ turn on warnings for questionable layout of record types
.Q* turn off warnings for questionable layout of record types
r+ turn on warnings for redundant construct
R* turn off warnings for redundant construct
.r+ turn on warnings for object renaming function
.R* turn off warnings for object renaming function
s suppress all info/warnings
.s turn on warnings for overridden size clause
.S* turn off warnings for overridden size clause
t turn on warnings for tracking deleted code
T* turn off warnings for tracking deleted code
.t*+ turn on warnings for suspicious contract
.T turn off warnings for suspicious contract
u+ turn on warnings for unused entity
U* turn off warnings for unused entity
.u turn on warnings for unordered enumeration
.U* turn off warnings for unordered enumeration
v*+ turn on warnings for unassigned variable
V turn off warnings for unassigned variable
.v*+ turn on info messages for reverse bit order
.V turn off info messages for reverse bit order
w*+ turn on warnings for wrong low bound assumption
W turn off warnings for wrong low bound assumption
.w turn on warnings on pragma Warnings Off
.W* turn off warnings on pragma Warnings Off
x*+ turn on warnings for export/import
X turn off warnings for export/import
.x+ turn on warnings for non-local exception
.X* turn off warnings for non-local exception
y*+ turn on warnings for Ada compatibility issues
Y turn off warnings for Ada compatibility issues
.y turn on info messages for why pkg body needed
.Y* turn off info messages for why pkg body needed
z*+ turn on warnings for suspicious unchecked conversion
Z turn off warnings for suspicious unchecked conversion
.z*+ turn on warnings for record size not a multiple of alignment
.Z turn off warnings for record size not a multiple of alignment
-gnatW? Wide character encoding method (?=h/u/s/e/8/b)
-gnatx Suppress output of cross-reference information
-gnatX Language extensions permitted
-gnaty Enable default style checks (same as -gnaty3abcefhiklmnprst)
-gnatyxx Enable selected style checks xx = list of parameters:
1-9 check indentation
a check attribute casing
A check array attribute indexes
b check no blanks at end of lines
B check no use of AND/OR for boolean expressions
c check comment format (two spaces)
C check comment format (one space)
d check no DOS line terminators
e check end/exit labels present
f check no form feeds/vertical tabs in source
g check standard GNAT style rules, same as ydISux
h check no horizontal tabs in source
i check if-then layout
I check mode in
k check casing rules for keywords
l check reference manual layout
Lnn check max nest level < nn
m check line length <= 79 characters
Mnn check line length <= nn characters
n check casing of package Standard identifiers
N turn off all checks
o check subprogram bodies in alphabetical order
O check overriding indicators
p check pragma casing
r check casing for identifier references
s check separate subprogram specs present
S check separate lines after THEN or ELSE
t check token separation rules
u check no unnecessary blank lines
x check extra parentheses around conditionals
y turn on default style checks
- subtract (turn off) subsequent checks
+ add (turn on) subsequent checks
-gnatyN Cancel all previously set style checks
-gnatzc Distribution stub generation for caller stubs
-gnatzr Distribution stub generation for receiver stubs
-gnat83 Ada 83 mode
-gnat95 Ada 95 mode
-gnat2005 Ada 2005 mode
-gnat2012 Ada 2012 mode (default)
-gnat-p Cancel effect of previous -gnatp switch
2019-08-13 Justin Squirek <squirek@adacore.com>
gcc/ada/
* gnatcmd.adb (GNATCmd): Add constant for new compiler switch
--help-ada, and include usage subprogram. Add line to usage help
explaining the new flag.
(GNATCmd_Usage): Rename from locally declared Usage so as not to
confuse with the newly imported version. Add new argument case
for --help-ada and add bug report email to implicit display of
help without the --help flag so as to unify output between the
two cases.
pmderodat [Tue, 13 Aug 2019 08:06:18 +0000 (08:06 +0000)]
[Ada] Complete the more extended AST traversal used in GNATprove
Following the work on the Ada 202X feature of iterated aggregates
(AI12-061), Loop_Actions should have been listed as a semantic field of
nodes of kind N_Iterated_Component_Association. Fix this as well as the
new extended AST traversal which should traverse these fields.
There is no impact on compilation.
2019-08-13 Yannick Moy <moy@adacore.com>
gcc/ada/
* sem_util.adb (Traverse_More_Func): Take into account
Loop_Actions inside N_Iterated_Component_Association nodes.
* sinfo.ads: Document correctly Loop_Actions as a field of nodes
of kind N_Iterated_Component_Association.
pmderodat [Tue, 13 Aug 2019 08:06:12 +0000 (08:06 +0000)]
[Ada] Extend range type in search primitives of formal vectors
2019-08-13 Claire Dross <dross@adacore.com>
gcc/ada/
* libgnat/a-cfinve.adb, libgnat/a-cofove.adb (Find_Index,
Reverse_Find_Index): Use bigger type to avoid range check
failure at the last loop iteration.
tkoenig [Mon, 12 Aug 2019 20:21:37 +0000 (20:21 +0000)]
2019-08-12 Thomas Koenig <tkoenig@gcc.gnu.org>
PR fortran/91424
* frontend-passes.c (do_subscript): Do not warn for an
expression a second time. Do not warn about a zero-trip loop.
(doloop_warn): Also look at contained namespaces.
2019-08-12 Thomas Koenig <tkoenig@gcc.gnu.org>
PR fortran/91424
* gfortran.dg/do_subscript_3.f90: New test.
* gfortran.dg/do_subscript_4.f90: New test.
* gfortran.dg/pr70754.f90: Use indices that to not overflow.
redi [Mon, 12 Aug 2019 16:41:27 +0000 (16:41 +0000)]
PR libstdc++/90361 add missing macro definition
The src/c++17/string-inst.cc file needs to override the default string
ABI so that it still contains the expected symbols even when the library
is configured with --with-default-libstdcxx-abi=gcc4-compatible.
PR libstdc++/90361
* src/c++17/string-inst.cc: Use _GLIBCXX_USE_CXX11_ABI=1 by default.
redi [Mon, 12 Aug 2019 14:54:12 +0000 (14:54 +0000)]
Add noexcept-specifier to std::apply and std::make_from_tuple
When unpacking a std::tuple we know that the std::get calls are
noexcept, so only the invocation (for std::apply) and construction (for
std::make_from_tuple) can throw.
We also know the std::get calls won't throw for a std::array, but this
patch doesn't specialize the variable template for std::array. For an
arbitrary tuple-like type we don't know if the std::get calls will
throw, and so just use a potentially-throwing noexcept-specifier.
* include/std/tuple (__unpack_std_tuple): New variable template and
partial specializations.
(apply, make_from_tuple): Add noexcept-specifier.
* testsuite/20_util/tuple/apply/2.cc: New test.
* testsuite/20_util/tuple/make_from_tuple/2.cc: New test.
pmderodat [Mon, 12 Aug 2019 09:01:58 +0000 (09:01 +0000)]
[Ada] Fix IPv6 numeric address detection
IPv6 numeric address can't have less than 2 colons. It fixes the error
when Get_Host_By_Name called with hostname composed by only hexadecimal
symbols.
pmderodat [Mon, 12 Aug 2019 09:01:53 +0000 (09:01 +0000)]
[Ada] New parameter Quiet for procedure GNAT.Command_Line.Getopt
Getopt procedure is parsing the command line or set of strings. If the
command line contains unknown switch than the Getopt prints error
message to the console and raises the exception Invalid_Switch. The
printing can be inappropriate in some cases. The new parameter Quiet
allows avoiding console output.
* libgnat/g-comlin.ads, libgnat/g-comlin.adb (Getopt): Add
parameter Quiet. Need to do not output error messages to
console. Invalid_Switch exception generation surrounded by an
error message.
pmderodat [Mon, 12 Aug 2019 09:01:48 +0000 (09:01 +0000)]
[Ada] Missing check on outbound parameter of a non-null access type
This patch adds code to generate proper post-call checks when an actual
for an in-out or out parameter has a non-null access type. No
constraints are applied to an inbound access parameter, but on exit a
not-null check must be performed if the type of the actual requires it.
2019-08-12 Ed Schonberg <schonberg@adacore.com>
gcc/ada/
* exp_ch6.adb (Expand_Actuals. Add_Call_By_Copy_Code): Add code
to generate proper checks when an actual for an in-out or out
parameter has a non-null access type. No constraints are
applied to an inbound access parameter, but on exit a not-null
check must be performed if the type of the actual requires it.
pmderodat [Mon, 12 Aug 2019 09:01:43 +0000 (09:01 +0000)]
[Ada] Crash on illegal left-hand side in assignment of renamed variable
This patch fixes a crash on an assignment where the left-hand side is a
renaming of a function call that does not involve ceiling priorities.
This avoids a compiler crash in some cases, and prevents a useless
retrieval and compilation of run-time packages.
2019-08-12 Ed Schonberg <schonberg@adacore.com>
gcc/ada/
* sem_util.adb (Is_Expaned_Priority_Attribute): Check whether
call comes from a rewritten attribute before comparing name with
Get_Ceiling run-time subprogram.
pmderodat [Mon, 12 Aug 2019 09:01:33 +0000 (09:01 +0000)]
[Ada] Fix internal error on comparison of unaligned slices
This fixes an internal error in the code generator when it is trying to
take the address of a slice which does not start on a byte boundary, in
order to generate a comparison between slices with a dynamic length.
This case is not supported by the code generator and comes from an
explicit representation clause on a record type, so it must be detected
and handled by the front-end by expanding the comparison on an
element-by-element basis.
2019-08-12 Eric Botcazou <ebotcazou@adacore.com>
gcc/ada/
* exp_ch4.adb (Expand_N_Op_Eq): Expand the array equality if
either operand is a possibly unaligned slice.
* exp_ch6.adb (Expand_Simple_Function_Return): Do not generate a
copy for a possibly unaligned object if it is represented as a
scalar.
* exp_util.adb (Is_Possibly_Unaligned_Slice): Do not always
return false if the target doesn't have strict alignment.
pmderodat [Mon, 12 Aug 2019 09:01:25 +0000 (09:01 +0000)]
[Ada] Do not suppress checks in instances of internal generics
This patch removes suppression of checks in nested instances of internal
packages. No test.
This was inconsistent: only for packages, not for subprograms. Only for
nested instantiations, not library level ones. Not for GNAT units.
Furthermore, the user should have control via pragma Suppress or
switches.
Furthermore, without this change, there could be missing tampering
checks in Ada.Containers.
2019-08-12 Bob Duff <duff@adacore.com>
gcc/ada/
* sem_ch12.adb (Instantiate_Package_Body): Remove suppression of
checks in instances of internal units.
* sem_ch6.adb (Analyze_Function_Return): Do not generate a
constraint check on an extended_return_statement if the subtype
of the return object in the statement is identical to the return
subtype of the function.
pmderodat [Mon, 12 Aug 2019 09:01:20 +0000 (09:01 +0000)]
[Ada] Prevent crash in Is_Reachable
This patch fixes a bug in Is_Reachable, which causes a crash when checks
are on.
2019-08-12 Bob Duff <duff@adacore.com>
gcc/ada/
* libgnat/a-cbmutr.adb (Is_Reachable): Declare Idx to be of the
base subtype. Clearly it makes no sense to loop "while Idx >=
0", if Idx is of a nonnegative subtype.
pmderodat [Mon, 12 Aug 2019 09:01:09 +0000 (09:01 +0000)]
[Ada] Remove doc for language version switches
Remove documentation for Ada language version switches, and note that
they are no longer needed. These tools now silently ignore such
switches, and process the file correctly no matter what version of Ada
is used.
2019-08-12 Bob Duff <duff@adacore.com>
gcc/ada/
* doc/gnat_ugn/gnat_utility_programs.rst (gnatmetric, gnatpp,
gnatstub): Remove documentation for Ada language version
switches, and note that they are no longer needed.
pmderodat [Mon, 12 Aug 2019 09:01:04 +0000 (09:01 +0000)]
[Ada] Hang on loop in generic with subtype indication specifying a range
The compiler may hang when a for loop expanded in a generic
instantiation has a range specified by a subtype indication with an
explicit range that has a bound that is an attribute applied to a
discriminant-dependent array component. The Parent field of the bound
may not be set, which can lead to endless looping when an actual subtype
created for the array component is passed to Insert_Actions. This is
fixed by setting the Parent fields of the copied bounds before
Preanalyze is called on them.
2019-08-12 Gary Dismukes <dismukes@adacore.com>
gcc/ada/
* sem_ch5.adb (Prepare_Param_Spec_Loop): Set the parents of the
copied low and high bounds in the case where the loop range is
given by a discrete_subtype_indication, to prevent hanging (or
Assert_Failure) in Insert_Actions.
gcc/testsuite/
* gnat.dg/generic_inst7.adb, gnat.dg/generic_inst7_pkg.adb,
gnat.dg/generic_inst7_pkg.ads, gnat.dg/generic_inst7_types.ads:
New testcase.
pmderodat [Mon, 12 Aug 2019 09:00:59 +0000 (09:00 +0000)]
[Ada] Improper error message on equality op with different operand types
2019-08-12 Ed Schonberg <schonberg@adacore.com>
gcc/ada/
* sem_ch6.adb (heck_Untagged_Equality): Verify that user-defined
equality has the same profile as the predefined equality before
applying legality rule in RM 4.5.2 (9.8).
gcc/testsuite/
* gnat.dg/equal10.adb, gnat.dg/equal10.ads: New testcase.
pmderodat [Mon, 12 Aug 2019 09:00:39 +0000 (09:00 +0000)]
[Ada] Minor cleanups in exception handling
No change in behavior, so no test.
2019-08-12 Bob Duff <duff@adacore.com>
gcc/ada/
* libgnat/a-except.ads: Update obsolete comment, still making
clear that this is a variant. Add explicit default for Id
component of Exception_Occurrence, because that value is used.
Define Null_Occurrence less redundantly.
* libgnat/a-einuoc.adb: Minor simplification of code.
This patch corrects certain behaviors within Ada.Directories to better
conform to conformance tests and implements the package
Ada.Directories.Hierarchical_File_Names outlined in AI05-0049-1.
Only partial test sources are included.
------------
-- Source --
------------
-- main.ads
with Ada.Directories.Hierarchical_File_Names;
use Ada.Directories.Hierarchical_File_Names;
with Ada.Exceptions; use Ada.Exceptions;
with Ada.Text_IO; use Ada.Text_IO;
procedure Main is
FULL_PATH_A : constant String := "/export/work/user/bug";
FULL_PATH_B : constant String := "/export/work/user";
* libgnat/a-dhfina.adb, libgnat/a-dhfina.ads (Is_Simple_Name,
Is_Root_Directory, Is_Parent_Directory,
Is_Current_Directory_Name, Is_Relative_Name, Initial_Directory,
Relative_Name, Compose): Add implementation and documentation.
* libgnat/a-direct.adb (Containing_Directory): Modify routine to
use routines from Ada.Directories.Hierarchical_File_Names and
remove incorrect special case for parent directories.
(Fetch_Next_Entry): Add check for current directory and parent
directory and ignore them under certain circumstances.
(Simple_Nmae): Add check for null result from
Simple_Name_Internal and raise Name_Error.
(Simple_Name_Internal): Add explicit check for root directories,
sanitize trailing directory separators, and modify behavior so
that current and parent directories are considered valid
results.
* Makefile.rtl: Add entry to GNATRTL_NONTASKING_OBJS.
pmderodat [Mon, 12 Aug 2019 08:59:58 +0000 (08:59 +0000)]
[Ada] Suppress_Initialization not respected for private subtypes
The compiler fails to suppress initialization on a variable of a subtype
of a private type (such as System.Address) even though the subtype has
aspect Suppress_Initialization. This can lead to errors on object
declarations specified with Thread_Local_Storage when Initialize_Scalars
is applied (as well as leading to default initialization when it
shouldn't).
2019-08-12 Gary Dismukes <dismukes@adacore.com>
gcc/ada/
* sem_prag.adb (Analyze_Pragma, Pragma_Suppress_Initialization):
For private types, set the Suppress_Initialization flag on the
Full_View of the entity rather than the entity's base type.
gcc/testsuite/
* gnat.dg/suppress_initialization2.adb,
gnat.dg/suppress_initialization2.ads: New testcase.
pmderodat [Mon, 12 Aug 2019 08:59:53 +0000 (08:59 +0000)]
[Ada] New aspect/pragma No_Caching for analysis of volatile data
A new aspect/pragma can be attached to volatile variables to indicate
that such a variable is not used for interactions with the external
world, but only that accesses to that variable should not be optimized
by the compiler. This is in particular useful for guarding against fault
injection. SPARK Reference manual has been updated to allow this use of
volatile data, see section 7.1.2, so that GNATprove can analyze such
variables as not volatile.
2019-08-12 Yannick Moy <moy@adacore.com>
gcc/ada/
* aspects.adb, aspects.ads (Aspect_No_Caching): New aspect.
* contracts.adb, contracts.ads (Add_Contract_Item): Add handling
of No_Caching.
(Analyze_Object_Contract): Add handling of No_Caching.
* einfo.adb, einfo.ads
(Get_Pragma): Add handling of No_Caching.
* doc/gnat_rm/implementation_defined_aspects.rst,
doc/gnat_rm/implementation_defined_pragmas.rst: Document new
aspect/pragma.
* gnat_rm.texi: Regenerate.
* par-prag.adb (Prag): New pragma Pragma_No_Caching.
* sem_ch13.adb (Analyze_Aspect_Specifications,
Check_Aspect_At_Freeze_Point): Add handling of No_Caching.
* sem_prag.adb (Analyze_Pragma): Deal with pragma No_Caching.
* sem_prag.ads (Analyze_External_Property_In_Decl_Part): Now
applies to No_Caching.
* sem_util.adb, sem_util.ads (Is_Effectively_Volatile): Add
handling of No_Caching.
(No_Caching_Enabled): New query function.
* snames.ads-tmpl: New names for pragma.
gcc/testsuite/
* gnat.dg/no_caching.adb, gnat.dg/no_caching.ads: New testcase.
pmderodat [Mon, 12 Aug 2019 08:59:42 +0000 (08:59 +0000)]
[Ada] More precise handling of Size/Object_Size in GNATprove
GNATprove does a partial expansion which did not allow getting the
most precise value for attributes Size/Object_Size. Now fixed.
There is no impact on compilation.
2019-08-12 Yannick Moy <moy@adacore.com>
gcc/ada/
* exp_attr.adb, exp_attr.ads (Expand_Size_Attribute): New
procedure to share part of the attribute expansion with
GNATprove mode.
(Expand_N_Attribute_Reference): Extract part of the
Size/Object_Size expansion in the new procedure
Expand_Size_Attribute.
* exp_spark.adb (Expand_SPARK_N_Attribute_Reference): Expand
Size/Object_Size attributes using the new procedure
Expand_Size_Attribute.
pmderodat [Mon, 12 Aug 2019 08:59:33 +0000 (08:59 +0000)]
[Ada] Inconsistent compile time Constraint_Error warning
This patch corrects several bugs within the compiler which led to
inconsistent handling of compile time Constraint_Errors. Notibly,
subtype out of range checks which are only out of range of the subtype
must be warnings while out of range checks where the value is out of
range of the base type must be an error. Also, type conversions and
qualified expressions on literals constitute errors on any out of range
value. The compiler needed many of these cases clarified.
------------
-- Source --
------------
-- main.ads
with System;
package Main is
type T_Enum is (Enum_1, Enum_2, Unknown)
with Default_Value => Unknown;
subtype T_Valid_Enum is T_Enum range Enum_1 .. Enum_2;
Value : T_Valid_Enum; -- WARNING
generic
type T_Element is (<>);
Init : T_Element;
package Generic_Test is
Value : T_Element := Init;
end;
package titi is new Generic_Test (T_Valid_Enum, Unknown); -- WARNING
$ gnatmake -q -gnatw_a main.adb
main.ads:9:12: warning: value not in range of type "T_Valid_Enum" defined at
line 7
main.ads:9:12: warning: "Constraint_Error" will be raised at run time
main.ads:18:52: warning: value not in range of type "T_Element" defined at
line 12, instance at line 18
main.ads:18:52: warning: "Constraint_Error" will be raised at run time
main.ads:25:23: value not in range of type "My_Float" defined at line 20
main.ads:25:23: static expression fails Constraint_Check
main.ads:27:19: value not in range of type "Standard.Positive"
main.ads:27:19: static expression fails Constraint_Check
main.ads:28:19: value not in range of type "Standard.Positive"
main.ads:28:19: static expression fails Constraint_Check
main.ads:29:19: warning: value not in range of type "Standard.Positive"
main.ads:29:19: warning: "Constraint_Error" will be raised at run time
main.ads:30:19: value not in range of type "Standard.Positive"
main.ads:30:19: static expression fails Constraint_Check
main.ads:31:27: value not in range of type "Standard.Positive"
main.ads:31:27: static expression fails Constraint_Check
main.ads:32:27: value not in range of type "Standard.Positive"
main.ads:32:27: static expression fails Constraint_Check
gnatmake: "main.ads" compilation error
2019-08-12 Justin Squirek <squirek@adacore.com>
gcc/ada/
* sem_eval.adb (Check_Non_Static_Context): Add a condition to
determine if a range violation constitues a warning or an error.
(Out_Of_Range): Add a condition to determine if a range
violation constitues a warning or an error.
pmderodat [Mon, 12 Aug 2019 08:59:28 +0000 (08:59 +0000)]
[Ada] Eliminate redundant range checks on conversions
This gets rid of redundant range checks generated in 5 out of the 9
cases of scalar conversions, i.e. (integer, fixed-point, floating-point)
converted to (integer, fixed-point, floating-point).
The problem is that the Real_Range_Check routine rewrites the conversion
node into a conversion to the base type so, when its parent node is
analyzed, a new conversion to the subtype may be introduced, depending
on the context, giving rise to a second range check against the subtype
bounds.
This change makes Real_Range_Check rewrite the expression of the
conversion node instead of the node, so that the type of the node is
preserved and no new conversion is introduced. As a matter of fact,
this is exactly what happens in the float-to-float case which goes to
the Generate_Range_Check circuit instead and does not suffer from the
duplication of range checks.
For the following procedure, the compiler must now generate exactly one
range check per nested function:
procedure P is
type I1 is new Integer range -100 .. 100;
type I2 is new Integer range -200 .. 200;
type D1 is delta 0.5 range -100.0 .. 100.0;
type D2 is delta 0.5 range -200.0 .. 200.0;
type F1 is new Long_Float range -100.0 .. 100.0;
type F2 is new Long_Float range -200.0 .. 200.0;
function Conv (A : I2) return I1 is
begin
return I1 (A);
end;
function Conv (A : D2) return I1 is
begin
return I1 (A);
end;
function Conv (A : F2) return I1 is
begin
return I1 (A);
end;
function Conv (A : I2) return D1 is
begin
return D1 (A);
end;
function Conv (A : D2) return D1 is
begin
return D1 (A);
end;
function Conv (A : F2) return D1 is
begin
return D1 (A);
end;
function Conv (A : I2) return F1 is
begin
return F1 (A);
end;
function Conv (A : D2) return F1 is
begin
return F1 (A);
end;
function Conv (A : F2) return F1 is
begin
return F1 (A);
end;
begin
null;
end;
2019-08-12 Eric Botcazou <ebotcazou@adacore.com>
gcc/ada/
* exp_ch4.adb (Real_Range_Check): Do not rewrite the conversion
node but its expression instead, after having fetched its
current value. Clear the Do_Range_Check flag on entry. Return
early for a rewritten float-to-float conversion. Remove
redundant local variable. Suppress all checks when inserting
the temporary and do not reanalyze the node.
pmderodat [Mon, 12 Aug 2019 08:59:18 +0000 (08:59 +0000)]
[Ada] Fix leak of Do_Range_Check flag in -gnatVa mode
This fixes a small glitch in Insert_Valid_Check, which needs to
propagate the Do_Range_Check flag onto the rewritten expression, but
uses its Original_Node as the source of the copy. Now Original_Node
does not necessarily point to the node that was just rewritten, but to
the ultimately original node, which is not the same node if the
expression was rewritten multiple times. The end result is that a
stalled Do_Range_Check flag can be wrongly resintated and leak to the
code generator.
2019-08-12 Eric Botcazou <ebotcazou@adacore.com>
gcc/ada/
* checks.adb (Insert_Valid_Check): Do not retrieve the
Do_Range_Check flag from the Original_Node but from the
Validated_Object. Remove useless bypass for floating-point
types.
pmderodat [Mon, 12 Aug 2019 08:59:13 +0000 (08:59 +0000)]
[Ada] Extended traversal subprograms for GNATprove
GNATprove needs traversal subprograms that do not simply traverse
syntactic nodes like Atree.Traverse_Func and Atree.Traverse_Proc, but
also traverse semantic nodes which are logically children of the nodes.
Now available through Sem_Util.Traverse_More_Func and
Sem_Util.Traverse_More_Proc.
There is no impact on compilation.
2019-08-12 Yannick Moy <moy@adacore.com>
gcc/ada/
* sem_util.adb, sem_util.ads (Traverse_More_Func,
Traverse_More_Proc): New traversal subprograms.
pmderodat [Mon, 12 Aug 2019 08:58:57 +0000 (08:58 +0000)]
[Ada] Add special bypass for obsolete code pattern
This change prevents the analysis phase of the front-end from setting
the Do_Range_Check flag in the very peculiar case of the source of a
conversion whose result is passed by reference to a "valued procedure",
because the expansion phase would not be able to generate the check.
This pattern appears in the ancient DEC Starlet package and it doesn't
seem to be useful at this point to change the expander to deal with it,
so instead the analysis phase is adjusted. Morever the compiler already
issues a warning in this case so this is probably good enough.
2019-08-12 Eric Botcazou <ebotcazou@adacore.com>
gcc/ada/
* sem_res.adb: Add with & use clause for Sem_Mech and
alphabetize.
(Resolve_Actuals): Do not apply a scalar range check for the
source of a conversion whose result is passed by reference to a
valued procedure.
pmderodat [Mon, 12 Aug 2019 08:58:52 +0000 (08:58 +0000)]
[Ada] Fix missing range check for In/Out parameter with -gnatVa
This plugs another small loophole in the front-end which fails to
generate a range check for a scalar In/Out parameter when -gnatVa is
specified. This also fixes a few more leaks of the Do_Range_Check flag
on actual parameters, both in regular and -gnatVa modes, as well as a
leak specific to expression function in -gnatp mode.
2019-08-12 Eric Botcazou <ebotcazou@adacore.com>
gcc/ada/
* checks.adb (Insert_Valid_Check): Reset the Do_Range_Check flag
on the validated object.
* exp_ch6.adb (Add_Call_By_Copy_Code): Reset the Do_Range_Check
flag on the actual here, as well as on the Expression if the
actual is a N_Type_Conversion node.
(Add_Validation_Call_By_Copy_Code): Generate the incoming range
check if needed and reset the Do_Range_Check flag on the
Expression if the actual is a N_Type_Conversion node.
(Expand_Actuals): Do not reset the Do_Range_Check flag here.
Generate the incoming range check for In parameters here instead
of...
(Expand_Call_Helper): ...here. Remove redudant condition.
* sem_res.adb (Resolve_Actuals): Use local variable A_Typ and
remove obsolete comments.
(Resolve_Type_Conversion): Do not force the Do_Range_Check flag
on the operand if range checks are suppressed.
pmderodat [Mon, 12 Aug 2019 08:58:46 +0000 (08:58 +0000)]
[Ada] Fix incorrect Do_Range_Check on type conversion
This gets rid of another leak of the Do_Range_Check flag to the back-end
which is specific to expression functions. No functional changes.
2019-08-12 Eric Botcazou <ebotcazou@adacore.com>
gcc/ada/
* checks.adb (Activate_Range_Check): Remove redundant argument.
(Generate_Range_Check): Likewise.
(Apply_Float_Conversion_Check): Reset the Do_Range_Check flag on
entry and remove redundant condition.
jb [Sun, 11 Aug 2019 09:42:41 +0000 (09:42 +0000)]
PR fortran/91413 Generate warning when making array static
When moving a local variable from the stack to static storage, the
procedure is no longer safe to be called recursively or concurrently
from multiple threads. Thus generate a warning when this is done.
Also double the default limit for switching from stack to static.
Regtested on x86_64-pc-linux-gnu.
gcc/fortran/ChangeLog:
2019-08-11 Janne Blomqvist <jb@gcc.gnu.org>
PR fortran/91413
* invoke.texi (-fmax-stack-var-size): Document increased default.
* options.c (gfc_post_options): Increase default stack var size to
65536 bytes.
* trans-decl.c (gfc_finish_var_decl): Generate warning when local
array moved to static storage.
ibuclaw [Sun, 11 Aug 2019 06:53:14 +0000 (06:53 +0000)]
d: Fix ICE: gimplification failed (gimplify.c at 13436)
The expression that caused the ICE
++(a += 1.0);
The D front-end rewrites and applies implicit type conversions so the
expression gets simplified as
(int)((double) a += 1.0) += 1
The codegen pass would subsequently generate the following invalid code
(int)(double) a = (int)((double) a + 1.0) + 1
The LHS expression `(int)(double) a', represented as a FIX_TRUNC_EXPR
being what trips as it is not a valid lvalue for assignment.
While LHS casts are stripped away, convert_expr adds a double cast
because it converts the expression to its original type before
converting it to its target type. There is no valid reason why this is
done, so it has been removed.
gcc/d/ChangeLog:
PR d/90601
* d-convert.cc (convert_expr): Don't convert an expression to its
original front-end type before converting to its target type.
kargl [Sat, 10 Aug 2019 18:26:13 +0000 (18:26 +0000)]
2019-08-10 Steven G. Kargl <kargl@gcc.gnu.org>
* decl.c (match_old_style_init): Use a clearer error message.
* expr.c (gfc_check_assign): Update BOZ checking to provide a stricter
adherence to the Fortran standard. Use gfc_invalid_boz () to
relax errors into warnings.
* gfortran.h (gfc_isym_id): Add new ids GFC_ISYM_DFLOAT,
GFC_ISYM_FLOAT, GFC_ISYM_REALPART, and GFC_ISYM_SNGL
* intrinsic.c (add_functions): Use new ids to split REAL generic into
REAL, FLOAT, DFLOAT, SNGL, and REALPART generics.
(gfc_intrinsic_func_interface): Allow new intrinsics in an
initialization expression
* resolve.c (resolve_operator): Deal with BOZ as operands.
Use gfc_invalid_boz to allow for errors or warnings via the
-fallow-invalid-boz option. A BOZ cannot be an operand to an
unary operator. Both operands of a binary operator cannot be BOZ.
For binary operators, convert a BOZ operand into the type and
kind of the other operand for REAL or INTEGER operand.
* trans-intrinsic.c: Use new ids to cause conversions to happen.
2019-08-10 Steven G. Kargl <kargl@gcc.gnu.org>
* gfortran.dg/boz_8.f90: Adjust error messages.
* gfortran.dg/nan_4.f90: Ditto.
* gfortran.dg/boz_1.f90: Add -fallow-invalid-boz to dg-options,
and test for warnings.
* gfortran.dg/boz_3.f90: Ditto.
* gfortran.dg/boz_4.f90: Ditto.
* gfortran.dg/dec_structure_6.f90: Ditto.
* gfortran.dg/ibits.f90: Ditto.
segher [Fri, 9 Aug 2019 17:33:25 +0000 (17:33 +0000)]
rs6000: vec-rotate-*.c fixes
This fixes two minor problems with the new testcases. The first is
that almost all other tests, including all vec* tests, for powerpc use
names with dashes, not underscores. The more important one is the the
vec-rotate-1.c and vec-rotate-3.c tests need the -maltivec flag.
rearnsha [Fri, 9 Aug 2019 16:14:59 +0000 (16:14 +0000)]
[aarch64] PR target/91386 Use copy_rtx to avoid modifying original insns in peep2 pattern
PR target/91386 is a situation where a peephole2 pattern substitution
is discarded late because the selected instructions contain
frame-related notes that we cannot redistribute (because the pattern
has more than one insn in the output). Unfortunately, the original
insns were being modified during the generation, so after the undo we
are left with corrupt RTL.
We avoid this by ensuring that the modifications are always made on a
copy, so that the original insns are never changed.
PR target/91386
* config/aarch64/aarch64.c (aarch64_gen_adjusted_ldpstp): Use copy_rtx
to preserve the contents of the original insns.
rearnsha [Fri, 9 Aug 2019 15:33:18 +0000 (15:33 +0000)]
[arm] Recognize thumb2 16-bit variants of the add and compare instructions
The addsi3_compare_op[12] patterns currently only have constraints to
pick the 32-bit variants of the instructions. Although the assembler
may sometimes opportunistically match a 16-bit t2 instruction, there's
no real control over that within the compiler. Consequently we might
emit a 32-bit adds instruction with a 16-bit subs instruction would
serve equally well. We do, of course still have to be careful about
the small number of boundary cases by controlling the order quite
carefully.
This patch adds the constraints and templates to match the t2 16-bit
variants of these instructions. Now, for example, we can generate
aoliva [Fri, 9 Aug 2019 09:20:58 +0000 (09:20 +0000)]
skip Cholesky decomposition in is>>n_mv_dist
normal_mv_distribution maintains the variance-covariance matrix param
in Cholesky-decomposed form. Existing param_type constructors, when
taking a full or lower-triangle varcov matrix, perform Cholesky
decomposition to convert it to the internal representation. This
internal representation is visible both in the varcov() result, and in
the streamed-out representation of a normal_mv_distribution object.
The problem is that when that representation is streamed back in, the
read-back decomposed varcov matrix is used as a lower-triangle
non-decomposed varcov matrix, and it undergoes Cholesky decomposition
again. So, each cycle of stream-out/stream-in changes the varcov
matrix to its "square root", instead of restoring the original
params.
This patch includes Corentin's changes that introduce verification in
testsuite/ext/random/normal_mv_distribution/operators/serialize.cc and
other similar tests that the object read back in compares equal to the
written-out object: the modified tests pass only if (u == v).
This patch also fixes the error exposed by his change, introducing an
alternate private constructor for param_type, used only by operator>>.
for libstdc++-v3/ChangeLog
* include/ext/random
(normal_mv_distribution::param_type::param_type): New private
ctor taking a decomposed varcov matrix, for use by...
(operator>>): ... this, befriended.
* include/ext/random.tcc (operator>>): Use it.
(normal_mv_distribution::param_type::_M_init_lower): Adjust
member function name in exception message.
for libstdc++-v3/ChangeLog
from Corentin Gay <gay@adacore.com>
* testsuite/ext/random/beta_distribution/operators/serialize.cc,
testsuite/ext/random/hypergeometric_distribution/operators/serialize.cc,
testsuite/ext/random/normal_mv_distribution/operators/serialize.cc,
testsuite/ext/random/triangular_distribution/operators/serialize.cc,
testsuite/ext/random/von_mises_distribution/operators/serialize.cc:
Add call to `VERIFY`.
marxin [Fri, 9 Aug 2019 08:33:05 +0000 (08:33 +0000)]
Add -flto=auto option value.
2019-08-09 Martin Liska <mliska@suse.cz>
* doc/invoke.texi: Document the option value.
* lto-wrapper.c (run_gcc): Set auto_parallel
only with -flto=auto.
2019-08-09 Martin Liska <mliska@suse.cz>
marxin [Fri, 9 Aug 2019 07:50:12 +0000 (07:50 +0000)]
Implement -fopt-info support for IPA ICF.
2019-08-09 Martin Liska <mliska@suse.cz>
* ipa-icf.c (sem_function::merge): Define AUTO_DUMP_SCOPE and
use dump_printf to report optimization.
(sem_variable::merge): Likwise.
(sem_item_optimizer::merge_classes): Use dump_printf to report
ICF hits.
2019-08-09 Martin Liska <mliska@suse.cz>
jakub [Fri, 9 Aug 2019 07:24:51 +0000 (07:24 +0000)]
c/
* c-parser.c (check_no_duplicate_clause): Simplify using
omp_find_clause.
(c_parser_omp_clause_if): Fix up printing of target {enter,exit} data
directive name modifiers.
(c_parser_omp_clause_proc_bind): Check for duplicate proc_bind clause.
cp/
* parser.c (check_no_duplicate_clause): Simplify using
omp_find_clause.
(cp_parser_omp_clause_if): Fix up printing of target {enter,exit} data
directive name modifiers.
testsuite/
* c-c++-common/gomp/if-4.c: New test.
* c-c++-common/gomp/clause-dups-1.c: New test.
jakub [Fri, 9 Aug 2019 07:23:03 +0000 (07:23 +0000)]
PR c/91401
c/
* c-parser.c (c_parser_omp_clause_dist_schedule): Fix up typos in the
check_no_duplicate_clause call. Comment it out, instead emit a
warning for duplicate dist_schedule clauses.
cp/
* parser.c (cp_parser_omp_clause_dist_schedule): Comment out the
check_no_duplicate_clause call, instead emit a warning for duplicate
dist_schedule clauses.
testsuite/
* c-c++-common/gomp/pr91401-1.c: New test.
* c-c++-common/gomp/pr91401-2.c: New test.
aoliva [Fri, 9 Aug 2019 03:48:53 +0000 (03:48 +0000)]
use rand instead of random
rand is in ISO C, whereas random is only in POSIX, so it makes sense
to use the more portable function everywhere instead of falling back
from one to the other on systems that miss the less portable one.
for gcc/testsuite/ChangeLog
* gcc.target/i386/sse2-mul-1.c: Use rand. Drop fallback.
* gcc.target/i386/sse4_1-blendps-2.c: Likewise.
* gcc.target/i386/sse4_1-blendps.c: Likewise.
* gcc.target/i386/xop-vshift-1.c: Likewise.
* gcc.target/powerpc/direct-move.h: Likewise.
segher [Thu, 8 Aug 2019 20:16:53 +0000 (20:16 +0000)]
rs6000: Use iterators in more DFP patterns
I noticed some patterns in dfp.md could use the D64_D128 iterator but
don't yet. This converts all remaining simple cases.
* config/rs6000/dfp.md (D64_D128): Move earlier in the file.
(dfp_suffix): Ditto.
(adddd3, addtd3): Merge to ...
(add<mode>3 for D64_D128): ... this.
(subdd3, subtd3): Merge to ...
(sub<mode>3 for D64_D128): ... this.
(muldd3, multd3): Merge to ...
(mul<mode>3 for D64_D128): ... this.
(divdd3, divtd3): Merge to ...
(div<mode>3 for D64_D128): ... this.
(*cmpdd_internal1, *cmptd_internal1): Merge to ...
(*cmp<mode>_internal1 for D64_D128): ... this.
(ftruncdd2, ftrunctd2): Merge to ...
(ftrunc<mode>2 for D64_D128): ... this.
(fixdddi2, fixtddi2): Merge to ...
(fix<mode>di2 for D64_D128): ... this.
wilson [Thu, 8 Aug 2019 19:04:56 +0000 (19:04 +0000)]
RISC-V: Fix C ABI for flattened struct with 0-length bitfield.
gcc/
PR target/91229
* config/riscv/riscv.c (riscv_flatten_aggregate_field): New arg
ignore_zero_width_bit_field_p. Skip zero size bitfields when true.
Pass into recursive call.
(riscv_flatten_aggregate_argument): New arg. Pass to
riscv_flatten_aggregate_field.
(riscv_pass_aggregate_in_fpr_pair_p): New local warned. Call
riscv_flatten_aggregate_argument twice, with false and true as last
arg. Process result twice. Compare results and warn if different.
(riscv_pass_aggregate_in_fpr_and_gpr_p): Likewise.
gcc/testsuite/
* gcc.target/riscv/flattened-struct-abi-1.c: New test.
* gcc.target/riscv/flattened-struct-abi-2.c: New test.
rsandifo [Thu, 8 Aug 2019 17:12:05 +0000 (17:12 +0000)]
[C] Fix bogus nested enum error message
For:
enum a { A };
enum a { B };
we emit a bogus error about nested definitions before the real error:
foo.c:2:6: error: nested redefinition of ‘enum a’
2 | enum a { B };
| ^
foo.c:2:6: error: redeclaration of ‘enum a’
foo.c:1:6: note: originally defined here
1 | enum a { A };
| ^
This is because we weren't clearing C_TYPE_BEING_DEFINED once the
definition was over.
I think it's OK to clear C_TYPE_BEING_DEFINED even for a definition
that actually is nested (and so whose outer definition is still open),
since we'll already have given an error by then. It means that second
and subsequent attempts to define a nested enum will usually get the
redeclaration error instead of the nested error, but that seems just
as accurate (nested_first and nested_second in the test). The only
exception is if the first nested enum was also invalid by being empty,
but then the enum as a whole has already produced two errors
(nested_empty in the test).
2019-08-08 Richard Sandiford <richard.sandiford@arm.com>
redi [Thu, 8 Aug 2019 10:18:53 +0000 (10:18 +0000)]
P0325R4 to_array from LFTS with updates
As an extension to what the standard requires, this also adds
conditional noexcept-specifiers to the std::to_array functions.
P0325R4 to_array from LFTS with updates
* include/experimental/array (to_array): Qualify call to __to_array.
* include/std/array (__cpp_lib_to_array, to_array): Define for C++20.
* include/std/version (__cpp_lib_to_array): Likewise.
* testsuite/23_containers/array/creation/1.cc: New test.
* testsuite/23_containers/array/creation/2.cc: New test.
* testsuite/23_containers/array/creation/3_neg.cc: New test.
* testsuite/23_containers/array/tuple_interface/tuple_element_neg.cc:
Use zero for dg-error line number.
jakub [Thu, 8 Aug 2019 06:39:02 +0000 (06:39 +0000)]
* gimplify.c (omp_add_variable): Use GOVD_PRIVATE | GOVD_EXPLICIT
for VLA helper variables on target data even if not GOVD_FIRSTPRIVATE.
(gimplify_scan_omp_clauses): For OMP_CLAUSE_USE_DEVICE_* use just
GOVD_EXPLICIT flags.
(gimplify_omp_workshare): For OMP_TARGET_DATA move all
OMP_CLAUSE_USE_DEVICE_* clauses to the end of clauses chain.
* omp-low.c (scan_sharing_clauses): For OMP_CLAUSE_USE_DEVICE_*
call install_var_field with mask 11 instead of 3.
(lower_omp_target): For OMP_CLAUSE_USE_DEVICE_* use pass
(splay_tree_key) &DECL_UID (var) to build_sender_ref instead of var.
gcc/c/
* c-typeck.c (c_finish_omp_clauses): For C_ORT_OMP
OMP_CLAUSE_USE_DEVICE_* clauses use oacc_reduction_head bitmap
instead of generic_head to track duplicates.
gcc/cp/
* semantics.c (finish_omp_clauses): For C_ORT_OMP
OMP_CLAUSE_USE_DEVICE_* clauses use oacc_reduction_head bitmap
instead of generic_head to track duplicates.
libgomp/
* target.c (gomp_map_vars_internal): For GOMP_MAP_USE_DEVICE_PTR
perform the lookup in the first loop only if !not_found_cnt, otherwise
perform lookups for it in the second loop guarded with
if (not_found_cnt || has_firstprivate).
* testsuite/libgomp.c/target-37.c: New test.
* testsuite/libgomp.c++/target-22.C: New test.