Re: [RFC PATCH 0/5] locks: implement "filp-private" (aka UNPOSIX) locks

From: Scott Lovenberg
Date: Sat Oct 12 2013 - 16:56:26 EST


On Fri, Oct 11, 2013 at 8:42 PM, Jeff Layton <jlayton@xxxxxxxxxx> wrote:
> On Fri, 11 Oct 2013 20:18:58 -0400
> Scott Lovenberg <scott.lovenberg@xxxxxxxxx> wrote:
>
>>
>> On Oct 11, 2013, at 19:49, Jeremy Allison <jra@xxxxxxxxx> wrote:
>>
>> > On Fri, 11 Oct 2013 15:36:43 -0600 Andreas Dilger <adilger@xxxxxxxxx> wrote:
>> >>>
>> >>> At this point, my main questions are:
>> >>>
>> >>> 1) does this look useful, particularly for fileserver implementors?
>> >
>> > Yes from the Samba perspective. We'll have to keep the old
>> > code around for compatibility with non-Linux OS'es, but this
>> > will allow Linux Samba to short-circuit a bunch of logic
>> > we have to get around the insane POSIX locking semantics
>> > on close.
>> >
>> > Jeremy.
>>
>> From the peanut gallery, IIRC from college a few years back, wasn't the POSIX file locking stuff passed by all parties because they intended to do their own thing regardless of the standard? The reason that all locks are blown on a release is mostly because there were already implementations and no one wanted to push the issue, or am I misunderstanding/forgetting the history of file locks in POSIX?
>
> This blog post of Jeremy's explains some of the history:
>
> http://www.samba.org/samba/news/articles/low_point/tale_two_stds_os2.html
>
> See the section entitled "First Implementation Past the Post".
>
> --
> Jeff Layton <jlayton@xxxxxxxxxx>

Thanks, Jeff. That was actually the exact article I was referencing
but forgetting the details of. Jeremy, thanks for writing that up so
many years ago (I used to eat that stuff up in college).


--
Peace and Blessings,
-Scott.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/