From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([208.118.235.92]:33561)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <paolo.bonzini@gmail.com>) id 1TUg0n-00019T-KS
	for qemu-devel@nongnu.org; Sat, 03 Nov 2012 11:51:55 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <paolo.bonzini@gmail.com>) id 1TUg0m-0003uH-CE
	for qemu-devel@nongnu.org; Sat, 03 Nov 2012 11:51:53 -0400
Received: from mail-da0-f45.google.com ([209.85.210.45]:34137)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <paolo.bonzini@gmail.com>) id 1TUg0m-0003uA-6H
	for qemu-devel@nongnu.org; Sat, 03 Nov 2012 11:51:52 -0400
Received: by mail-da0-f45.google.com with SMTP id n15so1998782dad.4
	for <qemu-devel@nongnu.org>; Sat, 03 Nov 2012 08:51:51 -0700 (PDT)
Sender: Paolo Bonzini <paolo.bonzini@gmail.com>
Message-ID: <50953D90.1010002@redhat.com>
Date: Sat, 03 Nov 2012 16:51:44 +0100
From: Paolo Bonzini <pbonzini@redhat.com>
MIME-Version: 1.0
References: <1351867404-25510-1-git-send-email-pbonzini@redhat.com>
	<CAFEAcA_8RMHPsrvCbnL9rvtzp7LJxKSAWhoG6F8+KYKngonRbQ@mail.gmail.com>
	<CAAu8pHv_PTsfcdwkuODPvUHEQ8E4xwj5qwxxtpzqUE_QWnvFbg@mail.gmail.com>
	<509527DF.6040801@redhat.com>
	<CAAu8pHunFDKWzkCosZrNfPM6d-Wmg3gTGe4g3qmNHT82DRsn4w@mail.gmail.com>
In-Reply-To: <CAAu8pHunFDKWzkCosZrNfPM6d-Wmg3gTGe4g3qmNHT82DRsn4w@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Subject: Re: [Qemu-devel] [PATCH v2 0/5] Fixes for thread pool patches.
List-Id: <qemu-devel.nongnu.org>
List-Unsubscribe: <https://lists.nongnu.org/mailman/options/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.nongnu.org/archive/html/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <https://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: Blue Swirl <blauwirbel@gmail.com>
Cc: Peter Maydell <peter.maydell@linaro.org>, qemu-devel@nongnu.org, jan.kiszka@siemens.com

Il 03/11/2012 16:26, Blue Swirl ha scritto:
>> > It's a clang bug.  The error should be suppressed, since the function is
>> > used with the weak alias.
>> >
>> > Or try if adding "|| defined __clang__" to compiler.h fixes it.
> It does.

Yeah, I tested now Clang/Linux myself and reported it as
http://llvm.org/bugs/show_bug.cgi?id=14250.  With a simple testcase such
as this:

static int g()
{
    return 42;
}
typeof(g) f __attribute__((__weak__, __alias__("g")));


it's easy to see that the function is actually emitted---and with "g" as
the name, even:

    .file    "g2.c"
    .text
    .align    16, 0x90
    .type    g,@function
g:                                      # @g
    .cfi_startproc
# BB#0:
    movl    $42, %eax
    ret
.Ltmp0:
    .size    g, .Ltmp0-g
    .cfi_endproc

    .weak    f
f = g
    .section    ".note.GNU-stack","",@progbits

> Maybe __APPLE__ should be replaced by __clang__, or can the
> problem also occur with GCC on Apple systems?

It can occur with GCC, though it will be just a "note" if I understand
correctly and not fail with -Werror.  No idea whether the generated code
would run, probably not.

My favorite fix is to just disable the warning, otherwise make the
function inline.

Paolo