Mailing-List: contact dev-help@stdcxx.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@stdcxx.apache.org
Received-SPF: pass (nike.apache.org: local policy)
Message-ID: <5054F77C.9030406@hates.ms>
Date: Sat, 15 Sep 2012 17:47:40 -0400
From: Liviu Nicoara <nikkoara@hates.ms>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7;
 rv:15.0) Gecko/20120907 Thunderbird/15.0.1
MIME-Version: 1.0
To: dev@stdcxx.apache.org
Subject: Re: STDCXX-1066 [was: Re: STDCXX forks]
References: <5054B4B1.8040502@hates.ms>
 <CALdE9OCKkiTRCjWZsTdGWFyFyeRAym5gXvwwLt4MY5_O128vGA@mail.gmail.com>
 <5054EBD1.2010802@hates.ms>
 <CALdE9OCSfwMJ13FH0LKLYr6d+2QmsX0FzQCaRHVzZOvGp11m+g@mail.gmail.com>
In-Reply-To: 
 <CALdE9OCSfwMJ13FH0LKLYr6d+2QmsX0FzQCaRHVzZOvGp11m+g@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit

On 9/15/12 5:19 PM, Stefan Teleman wrote:
> On Sat, Sep 15, 2012 at 4:57 PM, Liviu Nicoara <nikkoara@hates.ms> wrote:
>
>> Yes, but it restores the default packing, not an arbitrary one, potentially
>> set by the user prior to including our headers. Say, the user sets 2, the
>> default is 4 and we set 8. When we set it to default it goes back to 4,
>> instead of the expected 2. Did I get this right?
>
> This is true, but leaving some arbirary pragma pack(N) (for N != 0) in
> effect for the duration of a program, and expecting it to work sounds
> like a very defective programming approach to me. It will certainly
> not work on SPARC at all.  if the program needs to pack something in a
> certain specific way, it need to do so for that specific something
> only. Otherwise the side-effects of globally setting a non-default
> packing will destroy the program anyway.

I merely wanted to point out that restoring the default packing is not the same 
as restoring the packing to the previous value in effect.

Given this, I thought about an alternative way of forcing this alignment, e.g., 
via a union, aligned on an appropriate type. I see an advantage here in that 
most of the changes would occur where we define the 'primitive' mutex and 
condition wrappers, saving a few pre-processor conditionals and pragmas along 
the way.

What do you think?

Liviu