Date: Wed, 6 Apr 2016 20:46:28 -0500 (CDT) From: Bob Friesenhahn <bfriesen@simple.dallas.tx.us> To: Leo Prasath Arulraj <leo.prasath@gmail.com> Cc: freebsd-fs@freebsd.org Subject: Re: Does ZFS or UFS classify FS Metatada I/O requests with specific BIO_META like flag? Message-ID: <alpine.GSO.2.20.1604062041310.23612@freddy.simplesystems.org> In-Reply-To: <CAFcURvMJ4i0qOJ2cvXuiwoK0tQume1j9rQTr09W0TqQkCow3Yg@mail.gmail.com> References: <CAFcURvMJ4i0qOJ2cvXuiwoK0tQume1j9rQTr09W0TqQkCow3Yg@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 6 Apr 2016, Leo Prasath Arulraj wrote: > > I want to know if the FreeBSD filesystem UFS and ZFS set any such flag. > From a skim through the source code, it does not seem so because the bio > structure itself does not have any such flag in the first place : > http://fxr.watson.org/fxr/source/sys/bio.h?v=FREEBSD10#L143,144 > > I approciate any help regarding this. Zfs would have no use for any such thing since it does everything in terms of transaction groups (delimited by a disk cache sync of data for involved disks, followed by a TXG write, followed by a disk cache sync). Only synchronous writes get any priority, but they are prioritized by writing to a separate synchronous write log (as well as async/sync data targeted for the current transaction group) so that the request is not lost if the transaction group is otherwise lost. Bob -- Bob Friesenhahn bfriesen@simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer, http://www.GraphicsMagick.org/
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?alpine.GSO.2.20.1604062041310.23612>