[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20061128192236.GA2051@2ka.mipt.ru>
Date: Tue, 28 Nov 2006 22:22:36 +0300
From: Evgeniy Polyakov <johnpol@....mipt.ru>
To: David Miller <davem@...emloft.net>
Cc: drepper@...hat.com, akpm@...l.org, netdev@...r.kernel.org,
zach.brown@...cle.com, hch@...radead.org,
chase.venters@...entec.com, johann.borck@...sedata.com,
linux-kernel@...r.kernel.org, jeff@...zik.org, aviro@...hat.com
Subject: Re: Kevent POSIX timers support.
On Tue, Nov 28, 2006 at 11:13:00AM -0800, David Miller (davem@...emloft.net) wrote:
> From: Evgeniy Polyakov <johnpol@....mipt.ru>
> Date: Tue, 28 Nov 2006 12:16:02 +0300
>
> > Although ukevent has pointer embedded, it is unioned with u64, so there
> > should be no problems until 128 bit arch appeared, which likely will not
> > happen soon. There is also unused in kevent posix timers patch
> > 'u32 ret_val[2]' field, which can store segval's value too.
> >
> > But the fact that ukevent does not and will not in any way have variable
> > size is absolutely.
>
> I believe that in order to be %100 safe you will need to use the
> special aligned_u64 type, as that takes care of a crucial difference
> between x86 and x86_64 API, namely that u64 needs 8-byte alignment on
> x86_64 but not on x86.
>
> You probably know this already :-)
Yep :)
So I put it at the end, where structure is already correctly aligned, so
there is no need for special alignment.
And, btw, last time I checked, aligned_u64 was not exported to
userspace.
--
Evgeniy Polyakov
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists