Date: Wed, 09 May 2007 09:14:53 -0700 From: Garrett Cooper <yanegomi@gmail.com> To: Fredrik Lindberg <fli@shapeshifter.se> Cc: Perforce Change Reviews <perforce@freebsd.org>, Ivan Voras <ivoras@gmail.com> Subject: Re: PERFORCE change 119527 for review Message-ID: <4641F37D.50808@gmail.com> In-Reply-To: <4641E19F.80809@shapeshifter.se> References: <200705082229.l48MTrbh069129@repoman.freebsd.org> <20070509070626.GA41419@freebsd.org> <4641818B.4030507@FreeBSD.org> <20070509103404.D71759@fledge.watson.org> <464199D9.7020908@shapeshifter.se> <9bbcef730705090710h653dc15bjb7a1159484c1c48b@mail.gmail.com> <4641E19F.80809@shapeshifter.se>
next in thread | previous in thread | raw e-mail | index | archive | help
Fredrik Lindberg wrote:
> Ivan Voras wrote:
>> On 09/05/07, Fredrik Lindberg <fli@shapeshifter.se> wrote:
>>
>>> Ok, I'll take a look at the various queue.h shipped with the major BSDs,
>>> and see if I can find a common subset that suits my needs (without too
>>> many compromises) and based on the findings I'll reconsider my choice.
>>
>> I use sys/queue.h in my userland applications all the time, there's no
>> problem there.
>>
>
> And it works on {Net,Open}BSD?, I could swear that I've been bitten by
> sys/queue.h in the past when moving stuff between FreeBSD and NetBSD.
> Could have been those TAILQ-macros Robert Watson talked about though.
>
> Fredrik Lindberg
It may work with programs under normal conditions but are the queues
guaranteed to work with thread sychronization and other non-simple
conditions? I'd think that the FreeBSD queue is closer to that.
Plus the FreeBSD Queue's API is simple and the maintainer doesn't have
to peruse another set of source looking for the required function /
method they need.
Furthermore, it's a requirement made by style(9) to use methods that
aren't rolled by the implementor :).
Just a few thoughts.
-Garrett
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4641F37D.50808>