[Bug ada/101970] New: [11 regression] ICE on 'Enum_Rep for the enum that filled its representation range
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101970 Bug ID: 101970 Summary: [11 regression] ICE on 'Enum_Rep for the enum that filled its representation range Product: gcc Version: 11.2.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada Assignee: unassigned at gcc dot gnu.org Reporter: demoonlit at panathenaia dot halfmoon.jp Target Milestone: --- Created attachment 51322 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=51322=edit full test code Perhaps this is a regression by fixing the bug 101094. 'Enum_Rep causes GNAT bug Box, with the enum type containing the number of elements to fill just its (signed?) representation range. Thanks. test code Here is the abbreviated test code. The full code is attached. with Ada.Integer_Text_IO; with Ada.Text_IO; procedure test_enumrep2 is type T is (E80, ..., E00, ..., E7F); -- 256 elements for T use (E80 => -16#80#, ..., E00 => 16#00#, ..., E7F => 16#7F#); -- -128 to +127 for T'Size use 8; procedure P (X : T); pragma No_Inline (P); procedure P (X : T) is begin Ada.Integer_Text_IO.Put (T'Pos (X)); Ada.Integer_Text_IO.Put (T'Enum_Rep (X)); -- GNAT BUG DETECTED Ada.Text_IO.New_Line; end P; begin P (T'First); P (T'Last); end test_enumrep2; gcc-10.2.0 % /opt/ytomino/gcc-10.2.0/bin/gnatmake --version GNATMAKE 10.2.0 Copyright (C) 1995-2020, Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. % /opt/ytomino/gcc-10.2.0/bin/gnatmake test_enumrep2.adb gcc -c test_enumrep2.adb gnatbind -x test_enumrep2.ali gnatlink test_enumrep2.ali % ./test_enumrep2 0 -128 255127 It's fine. gcc-11.1.0 % /opt/ytomino/gcc-11.1.0/bin/gnatmake --version GNATMAKE 11.1.0 Copyright (C) 1995-2021, Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. % /opt/ytomino/gcc-11.1.0/bin/gnatmake test_enumrep2.adb gcc -c test_enumrep2.adb gnatbind -x test_enumrep2.ali gnatlink test_enumrep2.ali % ./test_enumrep2 0128 255127 Note, T'Enum_Rep(T'First) returns the unsigned value in gcc-11.1 because bug 101094. gcc-11.2.0 % /opt/ytomino/gcc-11.2.0/bin/gnatmake --version GNATMAKE 11.2.0 Copyright (C) 1995-2021, Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. % /opt/ytomino/gcc-11.2.0/bin/gnatmake test_enumrep2 gcc -c test_enumrep2.adb +===GNAT BUG DETECTED==+ | 11.2.0 (x86_64-linux-gnu) GCC error: | | in build_binary_op, at ada/gcc-interface/utils2.c:1164 | | Error detected at test_enumrep2.adb:107:44 | | Please submit a bug report; see https://gcc.gnu.org/bugs/ . | | Use a subject line meaningful to you and us to track the bug.| | Include the entire contents of this bug box in the report. | | Include the exact command that you entered. | | Also include sources listed below. | +==+ Please include these source files with error report Note that list may not be accurate in some cases, so please double check that the problem can still be reproduced with the set of files listed. Consider also -gnatd.n switch (see debug.adb). test_enumrep2.adb compilation abandoned gnatmake: "test_enumrep2.adb" compilation error
[Bug ada/101094] New: [11 regression] 'Enum_Rep returns a unsigned value for the negative representation in gcc-11
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101094
Bug ID: 101094
Summary: [11 regression] 'Enum_Rep returns a unsigned value for
the negative representation in gcc-11
Product: gcc
Version: 11.1.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: ada
Assignee: unassigned at gcc dot gnu.org
Reporter: demoonlit at panathenaia dot halfmoon.jp
Target Milestone: ---
'Enum_Rep returns a unsigned value for the run-time value having the negative
representation in gcc-11.
It also returns just a representation value for the compile-time value.
In gcc-10 or older, it always returns just a representation value.
Is this intentional changed or not?
Test code
-
with Ada.Text_IO;
with Ada.Integer_Text_IO;
procedure main is
type T is (Nega, Zero, Posi);
for T use (Nega => -1, Zero => 0, Posi => 1);
begin
-- literal
Ada.Text_IO.Put ("Lite:");
Ada.Integer_Text_IO.Put (T'Enum_Rep (Nega)); -- "-1"
Ada.Text_IO.New_Line;
-- compile-time constant
declare
Cons : constant T := Nega;
begin
Ada.Text_IO.Put ("Cons:");
Ada.Integer_Text_IO.Put (T'Enum_Rep (Cons)); -- "-1"
Ada.Text_IO.New_Line;
end;
-- variable
declare
Vari : T := T'Value ("Nega");
begin
Ada.Text_IO.Put ("Vari:");
Ada.Integer_Text_IO.Put (T'Enum_Rep (Vari)); -- "255"
Ada.Text_IO.New_Line;
end;
-- parameter
declare
procedure Sub1 (Para : T) is
begin
Ada.Text_IO.Put ("Para:");
Ada.Integer_Text_IO.Put (T'Enum_Rep (Para)); -- "255"
Ada.Text_IO.New_Line;
end Sub1;
begin
Sub1 (Nega);
end;
end main;
With gcc-10
---
$ gnatmake --version
GNATMAKE 10.2.0
Copyright (C) 1995-2020, Free Software Foundation, Inc.
This is free software; see the source for copying conditions.
There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR
PURPOSE.
$ gnatmake main
gcc -c main.adb
gnatbind -x main.ali
gnatlink main.ali
$ ./main
Lite: -1
Cons: -1
Vari: -1
Para: -1
With gcc-11
---
$ gnatmake --version
GNATMAKE 11.1.0
Copyright (C) 1995-2021, Free Software Foundation, Inc.
This is free software; see the source for copying conditions.
There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR
PURPOSE.
$ gnatmake main
gcc -c main.adb
gnatbind -x main.ali
gnatlink main.ali
% ./main
Lite: -1
Cons: -1
Vari:255
Para:255
Thanks.
[Bug ada/70867] [9/10/11/12 regression] access discriminant in return aggregate wrongly detected as dangling
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70867 --- Comment #13 from yuta tomino --- Sorry, perhaps my example is illegal in Ada 2012. "aliased" is required to return a reference of a part of the parameter. Randy explained the change in https://groups.google.com/g/comp.lang.ada/c/7dw7Oqi8lIk/m/5NUBhFN8DQAJ . (I wonder if it was legal before Ada 2005?? Because aliased parameter is added in Ada 2012 and the relevant text has been changed by AI05-0234-1.) I think Nicolas's example is always legal. Thanks.
[Bug ada/96799] New: C and Ada frontends have a different interpretation of double constants
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96799 Bug ID: 96799 Summary: C and Ada frontends have a different interpretation of double constants Product: gcc Version: 10.1.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada Assignee: unassigned at gcc dot gnu.org Reporter: demoonlit at panathenaia dot halfmoon.jp Target Milestone: --- Hello. Each of the C and Ada frontends has a different interpretation of the some kind of double constants. Reproducing: Write the same small double constants for each language. double1.c const double d_min = 0x0.3eP-1072; double2.ads package double2 is d_min : constant Long_Float := 16#0.3e#e-268; end double2; (Note: These constants are the hexadecimal of __DBL_DENORM_MIN__ and wanted to be rounded to the minimal value of double. The precise value is 0x0.4P-1072, but __DBL_DENORM_MIN__ is a written as decimal, so these constants are got from it by the conversion. This value is unimportant because perhaps this bug can be reproduced with other values.) Compile them. $ gcc -W -Wall -S double1.c $ gcc -W -Wall -S double2.ads double2.ads:2:35: warning: floating-point value underflows to 0.0 Ada frontend reports the underflow. C frontend does not. And compare double1.s and double2.s. double1.s .text .globl _d_min .const .align 3 _d_min: .long 1 # <- It has been rounded. .long 0 .ident "GCC: (GNU) 10.1.0" .subsections_via_symbols double2.s .text .globl _double2_E .data .align 1 _double2_E: .space 2 .globl _double2__d_min .const .align 3 _double2__d_min: .space 8# <- It has been truncated. .ident "GCC: (GNU) 10.1.0" .subsections_via_symbols I first noticed it in the FreeBSD 32bit. FreeBSD 32bit has the lesser precision of long double than other environments. However, then I tried it in FreeBSD 64bit, Linux, and MacOSX and it is reproducible too.
[Bug ada/65696] ASAN reports global-buffer-overrun for local tagged types
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65696 --- Comment #4 from yuta tomino --- I'm trying the released gcc-10.1, and confirmed that this is fixed. Thank you.
[Bug ada/81956] [7/8/9 regression] calling a null procedure is not skipped
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81956 --- Comment #3 from yuta tomino --- I'm trying gcc-8.3. This is probably fixed in 8.x. Thanks.
[Bug ada/83002] New: Missing finalization of generic package body
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83002 Bug ID: 83002 Summary: Missing finalization of generic package body Product: gcc Version: 7.2.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada Assignee: unassigned at gcc dot gnu.org Reporter: demoonlit at panathenaia dot halfmoon.jp Target Milestone: --- Created attachment 42613 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=42613=edit bug triggering source code Hello. The finalizer of package body (postfixed "__finalize_body") for controlled object is not generated when it is generic package (and its spec does not need finalizer ?). See the attached file and run. I've expected for getting "OK" outputted by the finalizer, but did not. And it works correctly if the controlled object declaration is moved to its spec. This bug also appeared with GNAT-GPL 2017. with x; generic package y is pragma elaborate_body; end y; package body y is object : x.t; -- this controlled object would not be finalized end y;
[Bug ada/82746] New: strange error on BIP function returning private type with Disable_Controlled
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82746
Bug ID: 82746
Summary: strange error on BIP function returning private type
with Disable_Controlled
Product: gcc
Version: 7.2.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: ada
Assignee: unassigned at gcc dot gnu.org
Reporter: demoonlit at panathenaia dot halfmoon.jp
Target Milestone: ---
Created attachment 42488
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=42488=edit
bug triggering source code
The compiler outputs strange error message ("" is undefined) on a
build-in-place function returning a type derived from a private type with
Disable_Controlled => True.
Compile the attached file:
% gcc -c pkg.adb
pkg.adb:17:07: "" is undefined
function BIP_2 return T is
begin
return Result : T do -- pkg.adb:17:07: "" is undefined
null;
end return;
end BIP_2;
[Bug ada/81961] [7 regression] an imported unsized C array in the auto-translated binding raises Storage_Error
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81961 --- Comment #1 from yuta tomino --- Note, -gnatp or pragma Suppress can not suppress this exception.
[Bug ada/81961] New: [7 regression] an imported unsized C array in the auto-translated binding raises Storage_Error
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81961 Bug ID: 81961 Summary: [7 regression] an imported unsized C array in the auto-translated binding raises Storage_Error Product: gcc Version: 7.2.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada Assignee: unassigned at gcc dot gnu.org Reporter: demoonlit at panathenaia dot halfmoon.jp Target Milestone: --- For example, translate a C header like following x.h with -fdump-ada-spec. x.h extern int unsized[]; = $ gcc -c -fdump-ada-spec x.h $ cat x_h.ads pragma Ada_2005; pragma Style_Checks (Off); with Interfaces.C; use Interfaces.C; package x_h is unsized : aliased array (size_t) of aliased int; -- x.h:1 pragma Import (C, unsized, "unsized"); end x_h; And prepare a main routine, compile and execute it. main.adb with x_h; procedure main is begin null; end main; == $ gnatmake main gcc -c main.adb gnatbind -x main.ali gnatlink main.ali $ ./main raised STORAGE_ERROR : x_h.ads:8 object too large This exception has been not appeared with gcc-6.
[Bug ada/81956] New: [7 regression] calling a null procedure is not skipped
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81956 Bug ID: 81956 Summary: [7 regression] calling a null procedure is not skipped Product: gcc Version: 7.2.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada Assignee: unassigned at gcc dot gnu.org Reporter: demoonlit at panathenaia dot halfmoon.jp Target Milestone: --- Hello. Has the behavior of null procedures been changed from gcc-7? package p is procedure Null_Proc is null; end p; with p; procedure main is begin p.Null_Proc; end main; With gcc-7.2, the calling Null_Proc is generated into the object code. $ gcc -c -O1 main.adb $ nm main.o 0010 s EH_frame1 S __ada_main U _p__null_proc With gcc-6.2, this calling Null_Proc has been skipped. I think this behavior is desirable. $ gcc -c -O1 main.adb $ nm main.o 0008 s EH_frame1 S __ada_main
[Bug ada/65696] ASAN reports global-buffer-overrun for local tagged types
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65696 --- Comment #2 from yuta tomino --- Here is a proposing fix. I has tried this patch with gcc from 4.8 to 7, and it seems to work fine with all versions. diff --git a/gcc/ada/exp_atag.adb b/gcc/ada/exp_atag.adb index 587432c..4313446 100644 --- a/gcc/ada/exp_atag.adb +++ b/gcc/ada/exp_atag.adb @@ -744,7 +744,8 @@ package body Exp_Atag is function Build_Inherit_Predefined_Prims (Loc : Source_Ptr; Old_Tag_Node : Node_Id; - New_Tag_Node : Node_Id) return Node_Id + New_Tag_Node : Node_Id; + Num_Predef_Prims : Int) return Node_Id is begin return @@ -759,7 +760,7 @@ package body Exp_Atag is New_Tag_Node, Discrete_Range => Make_Range (Loc, Make_Integer_Literal (Loc, Uint_1), -New_Occurrence_Of (RTE (RE_Max_Predef_Prims), Loc))), +Make_Integer_Literal (Loc, Num_Predef_Prims))), Expression => Make_Slice (Loc, @@ -772,7 +773,7 @@ package body Exp_Atag is Discrete_Range => Make_Range (Loc, Make_Integer_Literal (Loc, 1), - New_Occurrence_Of (RTE (RE_Max_Predef_Prims), Loc; + Make_Integer_Literal (Loc, Num_Predef_Prims; end Build_Inherit_Predefined_Prims; - diff --git a/gcc/ada/exp_atag.ads b/gcc/ada/exp_atag.ads index d53466f..eb4be55 100644 --- a/gcc/ada/exp_atag.ads +++ b/gcc/ada/exp_atag.ads @@ -111,7 +111,8 @@ package Exp_Atag is function Build_Inherit_Predefined_Prims (Loc : Source_Ptr; Old_Tag_Node : Node_Id; - New_Tag_Node : Node_Id) return Node_Id; + New_Tag_Node : Node_Id; + Num_Predef_Prims : Int) return Node_Id; -- Build code that inherits the predefined primitives of the parent. -- -- Generates: Predefined_DT (New_T).D (All_Predefined_Prims) := diff --git a/gcc/ada/exp_disp.adb b/gcc/ada/exp_disp.adb index 2b63377..ef7440f 100644 --- a/gcc/ada/exp_disp.adb +++ b/gcc/ada/exp_disp.adb @@ -4524,6 +4524,7 @@ package body Exp_Disp is Iface_Table_Node : Node_Id; Name_ITable: Name_Id; Nb_Predef_Prims: Nat := 0; + Cp_Predef_Prims: Nat := 0; Nb_Prim: Nat := 0; New_Node : Node_Id; Num_Ifaces : Nat := 0; @@ -5690,10 +5691,6 @@ package body Exp_Disp is Pos : Nat; begin -if not Building_Static_DT (Typ) then - Nb_Predef_Prims := Max_Predef_Prims; - -else Prim_Elmt := First_Elmt (Primitive_Operations (Typ)); while Present (Prim_Elmt) loop Prim := Node (Prim_Elmt); @@ -5703,13 +5700,18 @@ package body Exp_Disp is then Pos := UI_To_Int (DT_Position (Prim)); - if Pos > Nb_Predef_Prims then -Nb_Predef_Prims := Pos; + if Pos > Cp_Predef_Prims then +Cp_Predef_Prims := Pos; end if; end if; Next_Elmt (Prim_Elmt); end loop; + +if not Building_Static_DT (Typ) then + Nb_Predef_Prims := Max_Predef_Prims; +else + Nb_Predef_Prims := Cp_Predef_Prims; end if; declare @@ -6054,7 +6056,8 @@ package body Exp_Disp is (Node (Next_Elmt (First_Elmt -(Access_Disp_Table (Typ, Loc))); +(Access_Disp_Table (Typ, Loc), + Num_Predef_Prims => Cp_Predef_Prims)); if Nb_Prims /= 0 then Append_To (Elab_Code, @@ -6143,7 +6146,8 @@ package body Exp_Disp is Unchecked_Convert_To (RTE (RE_Tag), New_Occurrence_Of (Node (Next_Elmt (Sec_DT_Typ)), - Loc; + Loc)), +Num_Predef_Prims => Cp_Predef_Prims)); if Num_Prims /= 0 then Append_To (Elab_Code, @@ -6189,7 +6193,8 @@ package body Exp_Disp is Unchecked_Convert_To (RTE (RE_Tag), New_Occurrence_Of (Node (Next_Elmt (Sec_DT_Typ)), - Loc; + Loc)), +Num_Predef_Prims => Cp_Predef_Prims));
[Bug ada/70900] New: a library-level class-wide subprogram is not generated into .o/.s
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70900 Bug ID: 70900 Summary: a library-level class-wide subprogram is not generated into .o/.s Product: gcc Version: 6.1.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada Assignee: unassigned at gcc dot gnu.org Reporter: demoonlit at panathenaia dot halfmoon.jp Target Milestone: --- gcc-6.1.0 does not generate machine-code for some kind of library-level class-wide subprograms into .o/.s. Minor difference of nest-level, profile, or body of it affects possible to reproduce. I have not found the detailed condition, but found one reproducible small code, below: package p is type t is tagged null record; end p; with p; function p.liblevelcwsubp (obj : p.t'class) return p.t'class is begin return obj; end p.liblevelcwsubp; % x86_64-apple-darwin10-gcc -c p-liblevelcwsubp.adb % nm p-liblevelcwsubp.o nm: no name list % x86_64-pc-linux-gnu-gcc -c p-liblevelcwsubp.adb % x86_64-pc-linux-gnu-nm p-liblevelcwsubp.o (empty output) % x86_64-apple-darwin10-gcc -S p-liblevelcwsubp.adb % cat p-liblevelcwsubp.s .subsections_via_symbols (end of file) % x86_64-pc-linux-gnu-gcc -S p-liblevelcwsubp.adb % cat p-liblevelcwsubp.s .file "p-liblevelcwsubp.adb" .ident "GCC: (GNU) 6.1.0" .section.note.GNU-stack,"",@progbits (end of file) % x86_64-apple-darwin10-gcc --version x86_64-apple-darwin10-gcc (GCC) 6.1.0 Copyright (C) 2016 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. % x86_64-pc-linux-gnu-gcc --version x86_64-pc-linux-gnu-gcc (GCC) 6.1.0 Copyright (C) 2016 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. -- gcc-5 works: % gcc -c p-liblevelcwsubp.adb % nm p-liblevelcwsubp.o 0230 s EH_frame1 U ___gnat_rcheck_CE_Access_Check U ___gnat_rcheck_PE_Accessibility_Check T __ada_p__liblevelcwsubp U _memcpy U _system__secondary_stack__ss_allocate % gcc -S p-liblevelcwsubp.adb % cat p-liblevelcwsubp.s .const LC0: .ascii "p-liblevelcwsubp.adb" .space 1 .text .globl __ada_p__liblevelcwsubp __ada_p__liblevelcwsubp: LFB2: pushq %rbp LCFI0: movq%rsp, %rbp (...more contents) % gcc --version gcc (GCC) 5.3.0 Copyright (C) 2015 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
[Bug ada/70867] access discriminant in return aggregate detected as dangling if set from an implicit dereference in generic package
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70867 yuta tomino changed: What|Removed |Added CC||demoonlit at panathenaia dot halfm ||oon.jp --- Comment #1 from yuta tomino --- Hello. I have encountered a similar case, too. package p is type T (D : access constant Integer) is null record; type R is limited record Component : aliased Integer; end record; function F (Object : R) return T is (D => Object.Component'Access); end p; p.ads:7:12: access discriminant in return aggregate would be a dangling reference Inserting "aliased" between "Object :" and "R" resolves it. However, "Object : R" is already passed by reference without aliased, because in RM 6.2(7/3), an explicitly limited record type is a by-reference type unless aliased. So, it does not make a dangling reference.
[Bug ada/70645] New: [4.9/5/6 Regression] -fguess-branch-probability breaks debug-information, only in Ada
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70645
Bug ID: 70645
Summary: [4.9/5/6 Regression] -fguess-branch-probability breaks
debug-information, only in Ada
Product: gcc
Version: 6.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: ada
Assignee: unassigned at gcc dot gnu.org
Reporter: demoonlit at panathenaia dot halfmoon.jp
Target Milestone: ---
I can not debug some executable files generated by recent gcc Ada frontend,
with all optimization-levels including -fguess-branch-probability.
(I'm troubled with -Og. I've understood that -Og is intended to debug.)
gcc-4.9, gcc-5, and gcc-6 have failed.
However, gcc-4.8 works well.
C/C++ and Fortran frontends work well on all versions.
Note, I tried gdb from 7.9 to 7.11 as debugger.
% cat hello.adb
with Ada.Text_IO;
procedure Hello is
begin
Ada.Text_IO.Put_Line ("Hello.");
end Hello;
% gcc -v
gcc (GCC) 6.0.0 20160401 (experimental)
Copyright (C) 2016 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
% gnatmake -g -Og hello
gcc -c -g -Og hello.adb
gnatbind -x hello.ali
gnatlink hello.ali -g -Og
% gdb hello
GNU gdb (GDB) 7.11
Copyright (C) 2016 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law. Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-apple-darwin10.8.0".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from hello...Reading symbols from
hello.dSYM/Contents/Resources/DWARF/hello...done.
done.
(gdb) start
Temporary breakpoint 1 at 0x114f1
Starting program: hello
Temporary breakpoint 1, 0x000114f1 in hello ()
(gdb) s
Single stepping until exit from function _ada_hello,
which has no line number information. # <<<< Please look here
Hello.
main (argc=, argv=, envp=)
at b~hello.adb:197
197 adafinal;
(gdb) quit
A debugging session is active.
Inferior 1 [process 45663] will be killed.
Quit anyway? (y or n) y
% gnatmake -f -g -Og -fno-guess-branch-probability hello
gcc -c -g -Og -fno-guess-branch-probability hello.adb
gnatbind -x hello.ali
gnatlink hello.ali -g -Og -fno-guess-branch-probability
% gdb hello
GNU gdb (GDB) 7.11
Copyright (C) 2016 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law. Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-apple-darwin10.8.0".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from hello...Reading symbols from
hello.dSYM/Contents/Resources/DWARF/hello...done.
done.
(gdb) start
Temporary breakpoint 1 at 0x1154e: file hello.adb, line 2.
Starting program: hello
Temporary breakpoint 1, _ada_hello () at hello.adb:2
2 procedure Hello is
(gdb) s
4 Ada.Text_IO.Put_Line ("Hello."); # <<<< It works fine.
(gdb) s
ada__text_io__put_line__2 (item=...) at a-textio.adb:1424
1424 procedure Put_Line (Item : String) is
(gdb) quit
A debugging session is active.
Inferior 1 [process 45704] will be killed.
Quit anyway? (y or n) y
# Example of gcc-4.8 from here:
% gcc --version
gcc (GCC) 4.8.1
Copyright (C) 2013 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
% gnatmake -f -g -Og hello
gcc -c -g -Og hello.adb
gnatbind -x hello.ali
gnatlink hello.ali -g -Og
% gdb hello
GNU gdb (GDB) 7.11
Copyright (C) 2016 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free so
[Bug ada/69219] [5/6 regression] failed to compile nested subprograms with Inline_Always and Intrinsic
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69219 --- Comment #5 from yuta tomino --- > "built-in" is because you use Intrinsic! In any case, the rule of thumb is to > never use pragma Inline_Always with anything except for very small > subprograms. Well... I believe above "IA" is a sufficiently small subprogram because it's one line! ...Instead of such a viewpoint, I think there are two sides of this problem. 1. The compiler rejects the code that seems syntactically legal, and this behavior is depending on the place where the subprogram is instantiated. There is no reason why it is accepted in the library-level and rejected in the nested scope, especially when the pragmas are written in the generic package practically. 2. The error message is confused. "" is one too. I wish that "Intrinsic" is reported if Intrinsic is the cause. And the other lines are also unclear. > : In function 'nestedinlinealways.ia': > nestedinlinealways.adb:3:14: error: subprogram 'nestedinlinealways.ni' > not marked Inline_Always > nestedinlinealways.adb:11:4: error: parent subprogram cannot be inlined At first, I got them wrong that the cause is at "NI". Next, it took time for me to understand which the "parent" is indicating. These message should be improved like your explanation, even if this pattern should be an error.
[Bug ada/69219] [5/6 regression] failed to compile nested subprograms with Inline_Always and Intrinsic
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69219 --- Comment #8 from yuta tomino --- I'm glad that you are understanding that. > pragma Inline_Always is not defined by the Ada language so you cannot > invoke legality rules when you use it. > It will issue an error when it cannot inline something and, yes, this > depends on the context. Sorry, I mixed and expected pragma Inline_Always as similar to pragma Inline. I will pay attention to use of Inline_Always. Thanks.
[Bug ada/69219] [5/6 regression] failed to compile nested subprograms with Inline_Always and Intrinsic
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69219 --- Comment #2 from yuta tomino --- > Yes, remove the convention Intrinsic, it doesn't make any sense here or add > pragma Inline_Always on the nested subprogram as indicated. I agree with your sense. The local subprograms would be optimized well without the pragmas. Of course, I did not write the code directly like described above. Practically, I found this bug from some instantiations of the generic packages. The real problem is that we can not use the pattern of Inline_Always and Intrinsic, in writing any generic package, to reduce the code size for instantiating it in the library-level, because of also allowing use of it in the nested scope.
[Bug ada/69219] [5/6 regression] failed to compile nested subprograms with Inline_Always and Intrinsic
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69219 yuta tomino changed: What|Removed |Added Status|RESOLVED|UNCONFIRMED Resolution|INVALID |--- --- Comment #3 from yuta tomino --- In addition, even if it should be an error, I think that the proper error message should be reported instead of the error.
[Bug ada/69219] New: [5/6 regression] failed to compile nested subprograms with Inline_Always and Intrinsic
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69219 Bug ID: 69219 Summary: [5/6 regression] failed to compile nested subprograms with Inline_Always and Intrinsic Product: gcc Version: 6.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada Assignee: unassigned at gcc dot gnu.org Reporter: demoonlit at panathenaia dot halfmoon.jp Target Milestone: --- Created attachment 37297 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=37297=edit minimal bug triggering source code gcc-5 and later have generated no machine-code for subprograms having the combination of Inline_Always and Intrinsic. However, this pattern causes error for nested subprograms. procedure nestedinlinealways is -- outer subprogram procedure NI; pragma No_Inline (NI); procedure IA; pragma Convention (Intrinsic, IA); pragma Inline_Always (IA); -- pragma Inline (IA); -- it works fine. procedure IA is begin NI; -- calling other subprogram here is one of the triggers. end IA; procedure NI is null; begin IA; end nestedinlinealways; Please try to compile above code (the attached file): % gcc -c nestedinlinealways.adb : In function 'nestedinlinealways.ia': nestedinlinealways.adb:3:14: error: subprogram 'nestedinlinealways.ni' not marked Inline_Always nestedinlinealways.adb:11:4: error: parent subprogram cannot be inlined It will become to be compiled if I replace pragma Inline_Always to normal Inline or remove pragma Convention.
[Bug ada/48002] internal error on calling inherited, overloaded and abstract subprograms with string literal
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=48002 yuta tomino changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #6 from yuta tomino --- I'm unsure about the rule of this Bugzilla, but allow me to close because this bug has already been fixed.
[Bug ada/48039] Legal program rejected, a formal function is not visible in generic
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=48039 yuta tomino changed: What|Removed |Added Version|4.5.2 |6.0 Known to fail||4.5.2 --- Comment #2 from yuta tomino --- It has appeared on 6.0.0-20150823.
[Bug ada/48013] generic instantiation breaks the restriction of No_Elaboration_Code
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=48013 yuta tomino changed: What|Removed |Added Version|4.5.2 |6.0 Known to fail||4.5.2 --- Comment #2 from yuta tomino --- It has appeared on 6.0.0-20150823.
[Bug ada/48002] internal error on calling inherited, overloaded and abstract subprograms with string literal
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=48002 --- Comment #5 from yuta tomino demoonlit at panathenaia dot halfmoon.jp --- This bug seems already being fixed as François-Xavier wrote. (it could be compiled with gcc-5.1) Thanks.
[Bug ada/65696] New: ASAN reports global-buffer-overrun for local tagged types
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65696 Bug ID: 65696 Summary: ASAN reports global-buffer-overrun for local tagged types Product: gcc Version: unknown Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada Assignee: unassigned at gcc dot gnu.org Reporter: demoonlit at panathenaia dot halfmoon.jp package pkg_a is type Root is tagged null record; end pkg_a; with pkg_a; procedure main is type Derived is new pkg_a.Root with null record; begin null; end main; % gnatmake main.adb -cargs -fsanitize=address -largs -lasan.1 gcc -c -fsanitize=address main.adb gcc -c -fsanitize=address pkg_a.ads gnatbind -x main.ali gnatlink main.ali -lasan.1 % ./main = ==78928==ERROR: AddressSanitizer: unknown-crash on address 0x0001000174c8 at pc 0x1303d bp 0x7fff5fbfed10 sp 0x7fff5fbfed08 READ of size 120 at 0x0001000174c8 thread T0 #0 0x1303c (main+0x1303c) #1 0x123d4 (main+0x123d4) #2 0x119d7 (main+0x119d7) #3 0x0 (main+0x0) 0x000100017510 is located 0 bytes to the right of global variable 'pkg_a__rootR38s___UNC' from 'pkg_a.ads' (0x1000174c0) of size 80 SUMMARY: AddressSanitizer: unknown-crash ??:0 ?? Shadow bytes around the buggy address: 0x100020002e40: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0x100020002e50: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0x100020002e60: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0x100020002e70: 00 00 00 00 00 00 00 00 00 f9 f9 f9 f9 f9 f9 f9 0x100020002e80: 00 f9 f9 f9 f9 f9 f9 f9 00 00 00 00 00 00 00 00 =0x100020002e90: 00 00 f9 f9 f9 f9 f9 f9 00[00]00 00 00 00 00 00 0x100020002ea0: 00 00 f9 f9 f9 f9 f9 f9 00 f9 f9 f9 f9 f9 f9 f9 0x100020002eb0: 00 00 00 00 00 f9 f9 f9 f9 f9 f9 f9 00 00 00 00 0x100020002ec0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0x100020002ed0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0x100020002ee0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 Shadow byte legend (one shadow byte represents 8 application bytes): Addressable: 00 Partially addressable: 01 02 03 04 05 06 07 Heap left redzone: fa Heap right redzone: fb Freed heap region: fd Stack left redzone: f1 Stack mid redzone: f2 Stack right redzone: f3 Stack partial redzone: f4 Stack after return: f5 Stack use after scope: f8 Global redzone: f9 Global init order: f6 Poisoned by user:f7 Contiguous container OOB:fc ASan internal: fe ==78928==ABORTING Perhaps, a cause is the implicit code for making a child type's dispatching table. The size of copying of a parent type's predefined primitives to a child's is Max_Predef_Prims (15), but the real number of a parent type's predefined primitives seems 9.
[Bug ada/47748] Legal program rejected, ARM 12.7(3/2): generic formal package with others = in the formal_package_actual_part
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=47748 --- Comment #1 from yuta tomino demoonlit at panathenaia dot halfmoon.jp --- It seems fixed with gcc-4.7.2, 4.8.1, 4.9.2 and 5-20150405.
[Bug ada/65683] New: [5 regression] access types across limited with breaks restriction of No_Elaboration_Code
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65683 Bug ID: 65683 Summary: [5 regression] access types across limited with breaks restriction of No_Elaboration_Code Product: gcc Version: unknown Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada Assignee: unassigned at gcc dot gnu.org Reporter: demoonlit at panathenaia dot halfmoon.jp package pkg_a is type T is null record; -- non-controlled type end pkg_a; pragma Restrictions (No_Elaboration_Code); limited with pkg_a; package pkg_b is type P is access all pkg_a.T; for P'Storage_Size use 0; end pkg_b; Compile it with gcc-4.9.2. % gcc -c pkg_b.ads (success with no messages) Compile it with gcc-5-20150405. % gcc -c pkg_b.ads pkg_b.ads:5:09: violation of restriction no_elaboration_code at line 2 See pkg_b.ads.dg by -gnatD, the unnecessary master is generated. pragma restrictions (no_elaboration_code); limited with pkg_a; with system.system__soft_links; with system.system__tasking; package pkg_b is pkg_b___master : constant system__tasking__master_id := system__soft_links__current_master.all; pkg_b__pM : system__tasking__master_id renames pkg_b___master; type pkg_b__p is access all pkg_a.t; for pkg_b__p'storage_size use 0; freeze pkg_b__p [] end pkg_b; [subtype T1s is String (1 .. 73);] freeze_generic T1s freeze_generic object_pointer This master has not appeared until gcc-4.9. And, if limited is removed, it disappears with gcc-5-20150405.
[Bug ada/62205] GNAT does not accept class types for Default_Iterator
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62205 --- Comment #3 from yuta tomino demoonlit at panathenaia dot halfmoon.jp --- It seems fixed with gcc-5-20150405.
[Bug ada/65524] New: gnatbind generates decrementing the unexisting elab-counter into finalize_library
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65524 Bug ID: 65524 Summary: gnatbind generates decrementing the unexisting elab-counter into finalize_library Product: gcc Version: unknown Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada Assignee: unassigned at gcc dot gnu.org Reporter: demoonlit at panathenaia dot halfmoon.jp I found some cases that gnatbind generates decrementing the unexisting elaboration counter of each package Exxx. E142 : Short_Integer; pragma Import (Ada, E142, system__exn_lli_E); ... some Exxx are declared ... procedure finalize_library is begin ... E132 := E132 - 1; -- * E132 is not declared in above * ... end finalize_library; A cause is a package having _finalize_spec/_finalize_body without _elabs/_elabb. I still have not been able to make the minimal example, but have made a patch. In bindgen.adb, Gen_Elab_Externals refers U.Set_Elab_Entity to generate Exxx. However, Gen_Finalize_Library does not refer it. --- a/gcc/ada/bindgen.adb +++ b/gcc/ada/bindgen.adb @@ -1434,7 +1434,9 @@ package body Bindgen is -- has a finalizer. In that case, this is where we decrement -- the elaboration entity. -if U.Utype = Is_Body and then Uspec.Has_Finalizer then +if U.Utype = Is_Body and then Uspec.Has_Finalizer + and then Uspec.Set_Elab_Entity +then if not Lib_Final_Built then Gen_Header; Lib_Final_Built := True; @@ -1548,7 +1550,9 @@ package body Bindgen is WBI ( begin); -if U.Utype /= Is_Spec then +if U.Utype /= Is_Spec + and then Uspec.Set_Elab_Entity +then Set_String ( E); Set_Unit_Number (Unum); Set_String ( := E);
[Bug ada/65524] gnatbind generates decrementing the unexisting elab-counter into finalize_library
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65524 --- Comment #2 from yuta tomino demoonlit at panathenaia dot halfmoon.jp --- Created attachment 35115 -- https://gcc.gnu.org/bugzilla/attachment.cgi?id=35115action=edit example I found the way of reproducing. A tiny change of a-tags.ads is necessary. Insert is null into Ada.Tags.Register_Tag to suppress the elaboration. (Is there a pragma or restriction for this purpose?) Note, Unregister_Tag is unrelated. And, make a package as Pure or Preelaborate. Add a tagged type. So the package would have _finalize_spec without _elabs. (The direct cause of this case is the asymmetry that _elabs is not generated if Register_Tag is null, against that, _fialize_spec is always generated even if Unregister_Tag is null. Register/Unregister_XXX are sometimes suppressed in parts of customized runtimes. My customized runtime satisfies the conditions about this time. However, the way is only example for reproducing. I think another approach is possibility because Build_Finalizer in exp_ch7.adb is very complex...) And then, prepare a main subprogram using the package. Compile and gnatbind. See b~main.adb. The log of compiling the attached example: % gnatmake -a -gnatp -g main gnatbind -x main.ali gnatlink main.ali -g b~main.adb:31:10: E81 is undefined (more references follow) b~main.adb:34:07: E05 is undefined (more references follow) gnatmake: *** link failed. By the way, sorry, I realized that my first patch is not proper fix. For multiple-elaboration, an elaboration-counter should be generated and incremented in adainit when a _finalize_spec/body is existing, even if _elabs is not existing. Thanks.
[Bug ada/64548] Lost visibility of package System after raise
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64548 --- Comment #5 from yuta tomino demoonlit at panathenaia dot halfmoon.jp --- Hello, Simon. I'm glad to look at this report. I have a similar experience like this. The trigger of mine was not private with but a nested package in the spec of Ada.Exceptions. I should think it's useless information, but I would write, just in case.
[Bug ada/62117] [4.9 regression] wrong code for passing small array argument'Address, in generic
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62117 --- Comment #3 from yuta tomino demoonlit at panathenaia dot halfmoon.jp --- I understood by trial and error that pragma Pure_Function is equivalent to __attribute__((const)). I'm sorry. The first case (in my first description) is not compiler's bug but my mistake. The parameters Left and Right were erased by optimization because compiler assumed that Pure_Function does not dereference, in my current understanding. Also, the case2 (in comment 2) is not reproduced by gcc-5. However, the reason is that System.Fat_Flt.Unaligned_Valid is removed on gcc-5. ...By the way, I found an another point that I'm concerned. In the document of pragma Pure_Function, a function having Address parameter is not considered as Pure_Function. https://gcc.gnu.org/onlinedocs/gnat_rm/Pragma-Pure_005fFunction.html One exception is any function that has at least one formal of type System.Address or a type derived from it. Such functions are not considered pure by default, But, with System; package case3_p is pragma Pure; function F (X : System.Address) return System.Address; end case3_p; package body case3_p is function F (X : System.Address) return System.Address is begin return X; end F; end case3_p; with case3_p; with System.Storage_Elements; use System.Storage_Elements; with GNAT.IO; procedure case3_m is begin GNAT.IO.Put (Integer (To_Integer (case3_p.F (System'To_Address (10); GNAT.IO.Put (Integer (To_Integer (case3_p.F (System'To_Address (10); GNAT.IO.Put (Integer (To_Integer (case3_p.F (System'To_Address (10); end case3_m; Compile this case3_m.adb with optimization. % gcc -S -O -gnatp case3_m.adb .text .globl __ada_case3_m __ada_case3_m: LFB1: pushq%rbx LCFI0: movl$10, %edi call_case3_p__f movq%rax, %rbx movl%eax, %edi call_gnat__io__put__2 movl%ebx, %edi call_gnat__io__put__2 movl%ebx, %edi call_gnat__io__put__2 popq%rbx LCFI1: ret _case3_p__f is called once on the machine code against that case3_p.F is called three times from case3_m. Perhaps, case3_p.F is treated as Pure_Function. Which of those is correct, the document or the behaviour of compiler?
[Bug ada/64869] New: use all type clause is ineffective
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64869 Bug ID: 64869 Summary: use all type clause is ineffective Product: gcc Version: 5.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada Assignee: unassigned at gcc dot gnu.org Reporter: demoonlit at panathenaia dot halfmoon.jp use all type clauses for nested packages from an outer subprogram are ineffective. procedure case1 is package P is type T is (X, Y, Z); procedure Proc (Obj : T) is null; end P; -- procedure Nested is --use all type P.T; -- effective in nested subprogram --Obj : P.T := X; -- begin --Proc (Obj); -- visible -- end Nested; use all type P.T; Obj : P.T := X; -- invisible begin Proc (Obj); -- invisible end case1; % gcc -c case1.adb case1.adb:2:09: missing body for P case1.adb:13:22: X is not visible case1.adb:13:22: non-visible declaration at line 3 case1.adb:15:09: Proc is not visible case1.adb:15:09: non-visible declaration at line 4 with Ada.Containers.Vectors; procedure case2 is package Vectors is new Ada.Containers.Vectors (Positive, Character); -- procedure Nested is --use all type Vectors.Vector; -- effective in nested subprogram --X : Vectors.Vector := To_Vector (0); -- visible -- begin --Append (X, 'A'); -- visible -- end Nested; use all type Vectors.Vector; X : Vectors.Vector := To_Vector (0); -- visible begin Append (X, 'A'); -- invisible end case2; % gcc -c case2.adb case2.adb:13:09: Append is not visible case2.adb:13:09: non-visible declaration at a-convec.ads:243, instance at line 3 case2.adb:13:09: non-visible declaration at a-convec.ads:239, instance at line 3 Note, use all type clauses from nested sibling subprograms (commented out subprograms Nested in above examples) are effective, and, strangely, above examples will become to be compiled with the subprograms Nested (??)
[Bug ada/64349] [5 Regression] Bootstrapping Ada fails on darwin(9|10).
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64349
yuta tomino demoonlit at panathenaia dot halfmoon.jp changed:
What|Removed |Added
CC||demoonlit at panathenaia dot
halfm
||oon.jp
--- Comment #10 from yuta tomino demoonlit at panathenaia dot halfmoon.jp ---
Please move #ifdef in the correct order.
--- a/gcc/ada/env.c
+++ b/gcc/ada/env.c
@@ -44,12 +44,6 @@
#include stdlib.h
#endif
-#if defined (__APPLE__) !defined (__arm__)
-/* On Darwin, _NSGetEnviron must be used for shared libraries; but it is not
- available on iOS. */
-#include crt_externs.h
-#endif
-
#if defined (__vxworks)
#if defined (__RTP__)
/* On VxWorks 6 Real-Time process mode, environ is defined in unistd.h.
*/
@@ -78,6 +72,12 @@
#include system.h
#endif /* IN_RTS */
+#if defined (__APPLE__) !defined (__arm__)
+/* On Darwin, _NSGetEnviron must be used for shared libraries; but it is not
+ available on iOS. */
+#include crt_externs.h
+#endif
+
#ifdef __cplusplus
extern C {
#endif
@@ -215,11 +215,11 @@ __gnat_environ (void)
#elif defined (sun)
extern char **_environ;
return _environ;
+#elif defined (__APPLE__) !defined (__arm__)
+ return *_NSGetEnviron ();
#elif ! (defined (__vxworks))
extern char **environ;
return environ;
-#elif defined (__APPLE__) !defined (__arm__)
- return *_NSGetEnviron ();
#else
return environ;
#endif
[Bug ada/63276] New: implicit access checking is inserted for not null access type against pragma Suppress (Access_Check)
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63276 Bug ID: 63276 Summary: implicit access checking is inserted for not null access type against pragma Suppress (Access_Check) Product: gcc Version: 4.9.1 Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada Assignee: unassigned at gcc dot gnu.org Reporter: demoonlit at panathenaia dot halfmoon.jp The compiler inserts implicit access checking for not null access types against pragma Suppress (Access_Check). example: global.ads package global is type ptr is access all Integer; pragma Suppress (Access_Check, ptr); var : aliased Integer := 0; ref : not null ptr := var'Access; -- this not null is the trigger pragma Suppress (Access_Check, ref); end global; main.ads with global; procedure main is begin global.ref.all := 1; end main; Compile main.ads and use nm to look for inserted implicit checking $ gcc -c -gnatB main.adb $ nm main.o 0048 s EH_frame1 U ___gnat_rcheck_CE_Access_Check # unexpected access checking is inserted! T __ada_main U _global__ref For confirmation that not null is the trigger, delete not null from global.ads ref : ptr := var'Access; And recompile main.adb, use nm $ gcc -c -gnatB main.adb $ nm build/main.o 0018 s EH_frame1 T __ada_main U _global__ref The access checking is removed as intended, this time.
[Bug ada/62117] [4.9 regression] wrong code for passing small array argument'Address, in generic
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62117 yuta tomino demoonlit at panathenaia dot halfmoon.jp changed: What|Removed |Added Status|RESOLVED|UNCONFIRMED Resolution|INVALID |--- --- Comment #2 from yuta tomino demoonlit at panathenaia dot halfmoon.jp --- Thank you for looking, Eric. Surely, as you say, memcmp is unsuitable for presenting an example. It's my mistake. But I had not found other case at the time. I found another case since then. Pure is not a trigger in this case. Please look this, too. In the new case, it uses Float'Valid attribute calling System.Fat_Flt.Unaligned_Valid implicitly. Unaligned_Valid takes an Address argument. But it's not marked as pure. And, it's a detail of compiler implementation. I think this behavior should not affect application code. It needs the type U for calling Unaligned_Valid. If this type is removed (and some other method like Unchecked_Conversion is used), System.Fat_Flt.Valid may be called instead of Unaligned_Valid. System.Fat_Flt.Valid works correctly. I apologize for taking your time. -- case2.ads package case2 is -- not pure package function Packed_Unaligned_Valid (Item : Long_Long_Integer) return Boolean; -- not pure function end case2; -- case2.adb with system.storage_elements; package body case2 is use type System.Storage_Elements.Storage_Offset; -- A Float value is packed into the argument Item, at unaligned position. type U is record C : Character; F : Float; end record; pragma Pack (U); function Packed_Unaligned_Valid (Item : Long_Long_Integer) return Boolean is X : U; for X'Address use Item'Address; begin return X.F'Valid; -- implicit calling System.Fat_Flt.Unaligned_Valid end Packed_Unaligned_Valid; end case2; gcc-4.9 _case2__packed_unaligned_valid: LFB3: subq$24, %rsp LCFI0: leaq9(%rsp), %rdi call_system__fat_flt__attr_float__unaligned_valid addq$24, %rsp LCFI1: ret === gcc-4.8 _case2__packed_unaligned_valid: LFB3: subq$24, %rsp LCFI0: movq%rdi, 8(%rsp) # this operation is missing in gcc-4.9 leaq9(%rsp), %rdi call_system__fat_flt__attr_float__unaligned_valid addq$24, %rsp LCFI1: ret
[Bug ada/62230] New: Legal program rejected, RM 4.3.2(3, 5/3), ancestor_part can not be expression against error message by compiler itself
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62230 Bug ID: 62230 Summary: Legal program rejected, RM 4.3.2(3, 5/3), ancestor_part can not be expression against error message by compiler itself Product: gcc Version: unknown Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada Assignee: unassigned at gcc dot gnu.org Reporter: demoonlit at panathenaia dot halfmoon.jp Created attachment 33380 -- https://gcc.gnu.org/bugzilla/attachment.cgi?id=33380action=edit minimal bug triggering source code GNAT rejects any expressions implying aggregate for ancestor_part. For example: (Please see the attachment file, also) type T1 is limited new Ada.Finalization.Limited_Controlled with null record; type T2 is new T1 with null record; X : T2 := (T1 with null record); -- OK Z : T2 := (T1'(Ada.Finalization.Limited_Controlled with null record) with null record); -- ERROR GNAT outputs: main.adb:23:09: limited ancestor part must be aggregate or function call It seems for me that aggregate expression is allowed into ancestor_part according to the error message outputted by GNAT itself. But, it is not. And, RM 4.3.2(3, 5/3) allows it explicitly.
[Bug ada/62117] New: [4.9 regression] wrong code for passing small array argument'Address, in generic
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62117 Bug ID: 62117 Summary: [4.9 regression] wrong code for passing small array argument'Address, in generic Product: gcc Version: 4.9.1 Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada Assignee: unassigned at gcc dot gnu.org Reporter: demoonlit at panathenaia dot halfmoon.jp Created attachment 33306 -- https://gcc.gnu.org/bugzilla/attachment.cgi?id=33306action=edit minimal bug triggering source code When instantiating a generic function like below, with a small array type placed in register, according to x86_64 System V calling convension, generic type Key_Type is limited private; function Compare (Left, Right : Key_Type) return Integer; 'Address attributes of arguments point wrong place. The compiler probably has forgotten to store the arguments to memory. For example, if the body calls memcmp, function Compare (Left, Right : Key_Type) return Integer is begin return memcmp (Left'Address, Right'Address, Key_Type'Size / 8); end Compare; The generated code compiled with -S -Og: gcc-4.9 _inst__compare: LFB1: subq$40, %rsp LCFI0: movq%rsp, %rsi # Left and Right are passed by %rdi and %rsi leaq16(%rsp), %rdi # but this generated code overwrites them movl$4, %edx call_memcmp addq$40, %rsp LCFI1: ret This bug has appeared from gcc-4.9. gcc-4.8 is ok. === gcc-4.8 _inst__compare: LFB1: subq$40, %rsp LCFI0: movl%edi, 16(%rsp) # Left and Right are stored on the stack movl%esi, (%rsp) movq%rsp, %rsi leaq16(%rsp), %rdi movl$4, %edx call_memcmp addq$40, %rsp LCFI1: ret
[Bug ada/59234] New: Legal program rejected, a generic package having intricate formal package parameters could not be instantiated
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59234 Bug ID: 59234 Summary: Legal program rejected, a generic package having intricate formal package parameters could not be instantiated Product: gcc Version: 4.8.1 Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada Assignee: unassigned at gcc dot gnu.org Reporter: demoonlit at panathenaia dot halfmoon.jp Created attachment 31263 -- http://gcc.gnu.org/bugzilla/attachment.cgi?id=31263action=edit minimal bug triggering source code It could not compile a generic package having intricate formal package parameters. The generic package (X) having two formal package parameters, another generic package (Y) and its child generic package (Y.C). The 2nd generic package (Y) also having a formal package parameter that is 4th generic package (Z). In this case, The generic package X could not be instantiated. (If Y.C or Z is removed, X would be able to be instantiated.) An example is in the attached file: $ gnatmake main.adb gcc -c main.adb main.adb:8:30: actual parameter must be instance of nested main.adb:8:30: instantiation abandoned gnatmake: main.adb compilation error This bug has also been discussed on comp.lang.ada. Some people suggested me to report. https://groups.google.com/d/msg/comp.lang.ada/3UyNvf6KXRE/sjJB1zcmg7kJ Regards.
[Bug ada/55725] New: Anonymous access returned from library-level function is finalized too early, RM 3.10.2(10/2 or 10.3/3)
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55725 Bug #: 55725 Summary: Anonymous access returned from library-level function is finalized too early, RM 3.10.2(10/2 or 10.3/3) Classification: Unclassified Product: gcc Version: 4.7.2 Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada AssignedTo: unassig...@gcc.gnu.org ReportedBy: demoon...@panathenaia.halfmoon.jp Created attachment 28989 -- http://gcc.gnu.org/bugzilla/attachment.cgi?id=28989 minimal bug triggering source code An allocated object via anonymous access type returned from a library-level function like F may be finalized inside of the function. with Something; function F return access T; function F return access T is begin return new T; -- It returns new allocated value -- The allocated object may be finalized here end F; If a function is declared in a package, it behaves right. This bug has also been discussed on comp.lang.ada.
[Bug ada/48293] New: Legal program rejected, empty discriminant blocks object allocation with built-in-place
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48293 Summary: Legal program rejected, empty discriminant blocks object allocation with built-in-place Product: gcc Version: 4.6.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada AssignedTo: unassig...@gcc.gnu.org ReportedBy: demoon...@panathenaia.halfmoon.jp Created attachment 23774 -- http://gcc.gnu.org/bugzilla/attachment.cgi?id=23774 minimal bug triggering source code Compiler rejects allocating limited-type with function call (built-in-place) if the limited type has *empty* discriminant. (simple workaround is available, adding *dummy* discriminant) with Ada.Finalization; procedure min is -- bug sample package P1 is type T () is limited private; function F return T; private type T is new Ada.Finalization.Limited_Controlled with null record; end P1; package body P1 is function F return T is begin return (Ada.Finalization.Limited_Controlled with null record); end F; end P1; X1 : P1.T := P1.T'(P1.F); -- OK -- X2 : access P1.T := new P1.T; -- this code is illegal, it should be error. X3 : access P1.T := new P1.T'(P1.F); -- this code is legal, but error -- workaround package P2 is type T () is limited private; function F return T; private type T (Dummy : Integer) is new Ada.Finalization.Limited_Controlled with null record; -- add dummy discriminants end P2; package body P2 is function F return T is begin return (Ada.Finalization.Limited_Controlled with Dummy = 0); end F; end P2; X4 : access P2.T := new P2.T'(P2.F); -- OK begin null; end min;
[Bug ada/48293] Legal program rejected, empty discriminant blocks object allocation with built-in-place
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48293 --- Comment #1 from yuta tomino demoonlit at panathenaia dot halfmoon.jp 2011-03-26 07:31:35 UTC --- I'm sorry about wrong attachments. I deleted limited of P1.T by mistake, please add limited or look the code in the report.
[Bug ada/48039] New: Legal program rejected, a formal function is not visible in generic
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48039 Summary: Legal program rejected, a formal function is not visible in generic Product: gcc Version: 4.5.2 Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada AssignedTo: unassig...@gcc.gnu.org ReportedBy: demoon...@panathenaia.halfmoon.jp Created attachment 23589 -- http://gcc.gnu.org/bugzilla/attachment.cgi?id=23589 minimal bug triggering source code Name lookup is abnormal in complex generic code with gcc-4.5.1/4.5.2. -- p.ads package p is end p; -- p-gh.ads generic function p.gh return Integer; -- p-gh.adb function p.gh return Integer is begin return 0; end p.gh; -- p-gu.ads generic package p.gu is generic with function h return Integer is ; -- with function fh return Integer is ; -- ok function gh return Integer; end p.gu; -- p-gu.adb package body p.gu is function gh return Integer is begin return h; -- error !! -- return fh; -- ok end gh; end p.gu; -- p-h.ads with p.gh; function p.h is new gh; -- p-u.ads with p.gu; package p.u is new gu; -- p-r.ads with p.u; package p.r is end p.r; -- p-r-h.ads with p.h; function p.r.h is new u.gh (h); % gnatmake p-r-h.ads gcc -c p-r-h.ads p-r-h.ads:4:01: instantiation error at p-gu.adb:6 p-r-h.ads:4:01: h is not visible p-r-h.ads:4:01: instantiation error at p-gu.adb:6 p-r-h.ads:4:01: non-visible declaration at line 4 p-r-h.ads:4:01: instantiation error at p-gu.adb:6 p-r-h.ads:4:01: non-visible declaration at p-h.ads:4 gnatmake: p-r-h.ads compilation error This code will be compiled If we rename formal function h to another name (for example, fh).
[Bug ada/48013] New: generic instantiation breaks the restriction of No_Elaboration_Code
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48013 Summary: generic instantiation breaks the restriction of No_Elaboration_Code Product: gcc Version: 4.5.2 Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada AssignedTo: unassig...@gcc.gnu.org ReportedBy: demoon...@panathenaia.halfmoon.jp Created attachment 23565 -- http://gcc.gnu.org/bugzilla/attachment.cgi?id=23565 minimal bug triggering source code Unnecessary elaboration code is generated by generic instantiation with gcc-4.5.1/4.5.2/4.6.0(experimental at 20110225). 1. instantiate any generic package/subprogram in the library-level package. 2. gnatbind an application including this package. pragma Restrictions (No_Elaboration_Code) and pragma No_Run_Time are not effective in this problem. I expected that elaboration code will be removed or reported warning/error by Restrictions (No_Elaboration_Code). -- t_m.adb pragma Restrictions (No_Elaboration_Code); pragma No_Run_Time; with t_i; procedure t_m is begin null; end t_m; -- t_g.ads pragma Restrictions (No_Elaboration_Code); pragma No_Run_Time; generic procedure t_g; pragma Pure (t_g); -- t_g.adb pragma Restrictions (No_Elaboration_Code); pragma No_Run_Time; procedure t_g is begin null; end t_g; -- t_i.adb pragma Restrictions (No_Elaboration_Code); pragma No_Run_Time; with t_g; package t_i is pragma Pure (t_i); procedure nested is new t_g; -- *1 end t_i; % gnatmake -g t_m gcc -c -g t_m.adb gcc -c -g t_i.ads gcc -c -g t_g.adb gnatbind -x t_m.ali gnatlink t_m.ali -g Look adainit in b~t_m.adb. procedure adainit is E2 : Boolean; pragma Import (Ada, E2, t_i_E); begin null; t_i'elab_spec; -- it violates No_Elaboration_Code ! E2 := True; end adainit; % nm t_i.o 0261 D _t_i_E 0008 T _t_i___elabs T _t_i__nested (Remove generic instantiation (*1), and re-compile these, then, elaboration code disappeared from adainit in b~t_m.adb and __elabs procedure disappeared from t_i.o too.)
[Bug ada/48002] New: internal error on calling inherited, overloaded and abstract subprograms with string literal
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48002 Summary: internal error on calling inherited, overloaded and abstract subprograms with string literal Product: gcc Version: 4.5.2 Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada AssignedTo: unassig...@gcc.gnu.org ReportedBy: demoon...@panathenaia.halfmoon.jp Created attachment 23560 -- http://gcc.gnu.org/bugzilla/attachment.cgi?id=23560 minimal bug triggering source code gcc-4.5.1 and 4.5.2 crash with internal error. 1. declare a type (named A.T) and primitive subprogram (named A.F) having a String parameter. 2. declare new type (named B.T) derived from A.T, this type has inherited subprogram B.F from A.F. 3. re-declare B.F as abstract to hide. 4. overload B.F having a Wide_String parameter. 5. call Wide_String version of B.F with string literal. minimal source: procedure overload_str is package A is type T is new Integer; procedure F (X : T; Y : String) is null; end A; package B is type T is new A.T; procedure F (X : T; Y : Wide_String) is null; procedure F (X : T; Y : String) is abstract; -- hide inherited F end B; begin B.F (B.T'(0), ABC); -- compiler may crash B.F (B.T'(0), Wide_String'(ABC)); -- OK end overload_str; % gnatmake overload_str gcc -c overload_str.adb +===GNAT BUG DETECTED==+ | 4.5.2 (i686-apple-darwin9) Assert_Failure einfo.adb:1698 | | Error detected at overload_str.adb:12:10 | | Please submit a bug report; see http://gcc.gnu.org/bugs.html.| | Use a subject line meaningful to you and us to track the bug.| | Include the entire contents of this bug box in the report. | | Include the exact gcc or gnatmake command that you entered. | | Also include sources listed below in gnatchop format | | (concatenated together with no headers between files). | +==+ Please include these source files with error report Note that list may not be accurate in some cases, so please double check that the problem can still be reproduced with the set of files listed. Consider also -gnatd.n switch (see debug.adb). overload_str.adb compilation abandoned gnatmake: overload_str.adb compilation error
[Bug ada/48012] New: The executable file hangs up by combination of controlled-type, generic, renames and Inline_Always
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48012 Summary: The executable file hangs up by combination of controlled-type, generic, renames and Inline_Always Product: gcc Version: 4.5.2 Status: UNCONFIRMED Severity: normal Priority: P3 Component: ada AssignedTo: unassig...@gcc.gnu.org ReportedBy: demoon...@panathenaia.halfmoon.jp Created attachment 23564 -- http://gcc.gnu.org/bugzilla/attachment.cgi?id=23564 minimal bug triggering source code My application hangs up conpilied with gcc-4.5.2. 1. declare a controlled-type (named T) as private. 2. declare a generic function (named G) return T. 3. instantiate G. 4. declare a function (named F) renames it, and write pramga Inline_Always. 5. call F, it may hang up! Probably, an object of T is attached to a finalization list, twice. -- iacm.adb with iact; with iacp; procedure iacm is X : iact.T := iacp.F; -- hang up !! begin null; end iacm; -- iact.ads with Ada.Finalization; package iact is type T is private; -- hidden controll type generic function G return T; private type T is new Ada.Finalization.Controlled with null record; end iact; -- iact.adb package body iact is function G return T is begin return (Ada.Finalization.Controlled with null record); end G; end iact; -- iacp.ads with iact; package iacp is function F return iact.T; pragma Inline_Always (F); -- this pragma causes hang up private function G is new iact.G; function F return iact.T renames G; end iacp; % gnatmake iacm gcc -c iacm.adb gcc -c iacp.ads gcc -c iact.adb gnatbind -x iacm.ali gnatlink iacm.ali % ./iacm ( hang up !! )
