[Bug c/64509] New: _Generic throws error in unselected generic association

[Bug c/64509] New: _Generic throws error in unselected generic association

[maurits.de.jong at ericsson dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64509

            Bug ID: 64509
           Summary: _Generic throws error in unselected generic
                    association
           Product: gcc
           Version: 4.9.2
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
          Assignee: unassigned at gcc dot gnu.org
          Reporter: maurits.de.jong at ericsson dot com

Created attachment 34387
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=34387&action=edit
The offending source separately

I would have expected this to work (basically insert a conversion function
call if the type doesn't match):


-----
typedef struct {
        int x;
} mystruct;

extern int __f(int);
extern int __stoi(mystruct);

void call(void)
{
        mystruct S = {1};

        double z = _Generic((S), mystruct: __f(__stoi(S)), default: __f(S));
}
-----

but the typechecker is not happy about __f(S), which is in an association 
not relevant for the controlling expression:

The standard says:

  None of the expressions from any other generic association of the generic
  selection is evaluated.

Of course, evaluation is different from typechecking, but since the construct
is intended to conditionalise based on types, wouldn't it make sense to only
typecheck the selected association?




bash$ ./bin/gcc -std=c11 -v generic.c
Using built-in specs.
COLLECT_GCC=./bin/gcc
COLLECT_LTO_WRAPPER=/space/radmmj/gcc492/gcc-4.9.2/libexec/gcc/x86_64-unknown-linux-gnu/4.9.2/lto-wrapper
Target: x86_64-unknown-linux-gnu
Configured with: ../configure --prefix=/space/radmmj/gcc492/gcc-4.9.2 :
(reconfigured) ../configure --prefix=/space/radmmj/gcc492/gcc-4.9.2
--enable-languages=c
Thread model: posix
gcc version 4.9.2 (GCC) 
COLLECT_GCC_OPTIONS='-std=c11' '-v' '-mtune=generic' '-march=x86-64'
 /space/radmmj/gcc492/gcc-4.9.2/libexec/gcc/x86_64-unknown-linux-gnu/4.9.2/cc1
-quiet -v -imultiarch x86_64-linux-gnu generic.c -quiet -dumpbase generic.c
-mtune=generic -march=x86-64 -auxbase generic -std=c11 -version -o
/tmp/ccEshK6s.s
GNU C (GCC) version 4.9.2 (x86_64-unknown-linux-gnu)
        compiled by GNU C version 4.9.2, GMP version 4.3.2, MPFR version 2.4.2,
MPC version 0.8.1
GGC heuristics: --param ggc-min-expand=100 --param ggc-min-heapsize=131072
ignoring nonexistent directory "/usr/local/include/x86_64-linux-gnu"
ignoring nonexistent directory
"/space/radmmj/gcc492/gcc-4.9.2/lib/gcc/x86_64-unknown-linux-gnu/4.9.2/../../../../x86_64-unknown-linux-gnu/include"
#include "..." search starts here:
#include <...> search starts here:
 /space/radmmj/gcc492/gcc-4.9.2/lib/gcc/x86_64-unknown-linux-gnu/4.9.2/include
 /usr/local/include
 /space/radmmj/gcc492/gcc-4.9.2/include

/space/radmmj/gcc492/gcc-4.9.2/lib/gcc/x86_64-unknown-linux-gnu/4.9.2/include-fixed
 /usr/include/x86_64-linux-gnu
 /usr/include
End of search list.
GNU C (GCC) version 4.9.2 (x86_64-unknown-linux-gnu)
        compiled by GNU C version 4.9.2, GMP version 4.3.2, MPFR version 2.4.2,
MPC version 0.8.1
GGC heuristics: --param ggc-min-expand=100 --param ggc-min-heapsize=131072
Compiler executable checksum: 1bfa4fb39c457521c17e15d89fa44581
generic.c: In function ‘call’:
generic.c:12:69: error: incompatible type for argument 1 of ‘__f’
         double z = _Generic((S), mystruct: __f(__stoi(S)), default: __f(S));
                                                                     ^
generic.c:5:12: note: expected ‘int’ but argument is of type ‘mystruct’
 extern int __f(int);
            ^
>From gcc-bugs-return-472295-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Tue Jan 06 10:43:30 2015
Return-Path: <gcc-bugs-return-472295-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 7614 invoked by alias); 6 Jan 2015 10:43:30 -0000
Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc-bugs.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc-bugs/>
List-Post: <mailto:gcc-bugs@gcc.gnu.org>
List-Help: <mailto:gcc-bugs-help@gcc.gnu.org>
Sender: gcc-bugs-owner@gcc.gnu.org
Delivered-To: mailing list gcc-bugs@gcc.gnu.org
Received: (qmail 7570 invoked by uid 48); 6 Jan 2015 10:43:26 -0000
From: "olegendo at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/64479] [4.8 Regression][SH] wrong optimization delayed-branch
Date: Tue, 06 Jan 2015 10:43:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: target
X-Bugzilla-Version: 4.8.4
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: olegendo at gcc dot gnu.org
X-Bugzilla-Status: NEW
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: ---
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields: cf_gcctarget bug_status cf_reconfirmed_on cc short_desc everconfirmed
Message-ID: <bug-64479-4-kL5txm4e15@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-64479-4@http.gcc.gnu.org/bugzilla/>
References: <bug-64479-4@http.gcc.gnu.org/bugzilla/>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 7bit
X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/
Auto-Submitted: auto-generated
MIME-Version: 1.0
X-SW-Source: 2015-01/txt/msg00289.txt.bz2
Content-length: 1197

https://gcc.gnu.org/bugzilla/show_bug.cgi?idd479

Oleg Endo <olegendo at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Target|sh3                         |sh*-*-*
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2015-01-06
                 CC|                            |kkojima at gcc dot gnu.org,
                   |                            |olegendo at gcc dot gnu.org
            Summary|wrong optimization          |[4.8 Regression][SH] wrong
                   |delayed-branch  for SH      |optimization delayed-branch
     Ever confirmed|0                           |1

--- Comment #1 from Oleg Endo <olegendo at gcc dot gnu.org> ---
Thanks for pinpointing and reporting this.

trunk (GCC 5): OK
4.9 branch: OK
4.8 branch: NG

This is a problem of the flaky cbranch optimization in 4.8 in the *cbranch_t
splitter code.

For some reason, modified_between_p seems to ignore the fact that the T_REG is
clobbered in call insns.  Thus it thinks that it's OK to extend the T_REG
lifetime across a call insn.

[Bug c/64509] _Generic throws error in unselected generic association

[mpolacek at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64509

Marek Polacek <mpolacek at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |mpolacek at gcc dot gnu.org

--- Comment #1 from Marek Polacek <mpolacek at gcc dot gnu.org> ---
I think it's fine to reject that testcase.  The type checking is performed when
parsing while building the association expression...

[Bug c/64509] _Generic throws error in unselected generic association

[joseph at codesourcery dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64509

--- Comment #3 from joseph at codesourcery dot com <joseph at codesourcery dot com> ---
_Generic is intended for cases like <tgmath.h>, where calls to all 
functions are still valid for all argument types.  This is a case where 
the expression __f(S) violates a constraint, and that constraint applies 
to all function calls, not just evaluated ones - this is just the same as 
a constraint violation inside if (0), or if 
(expression_that_evaluates_to_0_at_runtime).  (It's also required that 
functions called in unevaluated generic associations are defined somewhere 
in the program; only sizeof and _Alignof have exemptions from that.)

[Bug c/64509] _Generic throws error in unselected generic association

[mpolacek at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64509

Marek Polacek <mpolacek at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|---                         |INVALID

--- Comment #5 from Marek Polacek <mpolacek at gcc dot gnu.org> ---
(In reply to Martien de Jong from comment #4)
> BTW, I notice that
> 
> ./gcc/ginclude/tgmath.h
> 
> uses
> 
> __builtin_classify_type
> __builtin_types_compatible_p
> __builtin_choose_expr
> 
> Will it move to using _Generic() ?

I think it will over time (with a proper __GNUC_PREREQ guard).

Anyway, there is nothing to do for this bug, so closing.