Re: [patch] Converting writeback linked lists to a tree based data structure

view
thread

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

From: Andrew Morton

Subject: Re: [patch] Converting writeback linked lists to a tree based data structure

Date: Wednesday, January 16, 2008 - 1:13 am

On Wed, 16 Jan 2008 18:55:38 +1100 David Chinner <dgc@sgi.com> wrote:

quoted text> On Tue, Jan 15, 2008 at 07:44:15PM -0800, Andrew Morton wrote:
> > On Wed, 16 Jan 2008 11:01:08 +0800 Fengguang Wu <wfg@mail.ustc.edu.cn> wrote:
> > 
> > > On Tue, Jan 15, 2008 at 09:53:42AM -0800, Michael Rubin wrote:
> > > > On Jan 15, 2008 12:46 AM, Peter Zijlstra <a.p.zijlstra@chello.nl> wrote:
> > > > > Just a quick question, how does this interact/depend-uppon etc.. with
> > > > > Fengguangs patches I still have in my mailbox? (Those from Dec 28th)
> > > > 
> > > > They don't. They apply to a 2.6.24rc7 tree. This is a candidte for 2.6.25.
> > > > 
> > > > This work was done before Fengguang's patches. I am trying to test
> > > > Fengguang's for comparison but am having problems with getting mm1 to
> > > > boot on my systems.
> > > 
> > > Yeah, they are independent ones. The initial motivation is to fix the
> > > bug "sluggish writeback on small+large files". Michael introduced
> > > a new rbtree, and me introduced a new list(s_more_io_wait).
> > > 
> > > Basically I think rbtree is an overkill to do time based ordering.
> > > Sorry, Michael. But s_dirty would be enough for that. Plus, s_more_io
> > > provides fair queuing between small/large files, and s_more_io_wait
> > > provides waiting mechanism for blocked inodes.
> > > 
> > > The time ordered rbtree may delay io for a blocked inode simply by
> > > modifying its dirtied_when and reinsert it. But it would no longer be
> > > that easy if it is to be ordered by location.
> > 
> > What does the term "ordered by location" mean?  Attemting to sort inodes by
> > physical disk address?  By using their i_ino as a key?
> > 
> > That sounds optimistic.
> 
> In XFS, inode number is an encoding of it's location on disk, so
> ordering inode writeback by inode number *does* make sense.

This code is mainly concerned with writing pagecache data, not inodes.
--
unsubscribe noticeTo unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:

[patch] Converting writeback linked lists to a tree based ..., Michael Rubin, (Tue Jan 15, 1:09 am)

Re: [patch] Converting writeback linked lists to a tree ba ..., Peter Zijlstra, (Tue Jan 15, 1:46 am)

Re: [patch] Converting writeback linked lists to a tree ba ..., Michael Rubin, (Tue Jan 15, 10:53 am)

Re: [patch] Converting writeback linked lists to a tree ba ..., Fengguang Wu, (Tue Jan 15, 8:01 pm)

Re: [patch] Converting writeback linked lists to a tree ba ..., Andrew Morton, (Tue Jan 15, 8:44 pm)

Re: [patch] Converting writeback linked lists to a tree ba ..., Fengguang Wu, (Tue Jan 15, 9:25 pm)

Re: [patch] Converting writeback linked lists to a tree ba ..., Andrew Morton, (Tue Jan 15, 9:42 pm)

Re: [patch] Converting writeback linked lists to a tree ba ..., Fengguang Wu, (Tue Jan 15, 9:55 pm)

Re: [patch] Converting writeback linked lists to a tree ba ..., Andrew Morton, (Tue Jan 15, 10:51 pm)

Re: [patch] Converting writeback linked lists to a tree ba ..., David Chinner, (Wed Jan 16, 12:55 am)

Re: [patch] Converting writeback linked lists to a tree ba ..., Andrew Morton, (Wed Jan 16, 1:13 am)

Re: [patch] Converting writeback linked lists to a tree ba ..., Fengguang Wu, (Wed Jan 16, 2:07 am)

Re: [patch] Converting writeback linked lists to a tree ba ..., Fengguang Wu, (Wed Jan 16, 6:06 am)

Re: [patch] Converting writeback linked lists to a tree ba ..., Michael Rubin, (Wed Jan 16, 11:55 am)

Re: [patch] Converting writeback linked lists to a tree ba ..., Fengguang Wu, (Wed Jan 16, 8:31 pm)

Re: [patch] Converting writeback linked lists to a tree ba ..., Fengguang Wu, (Thu Jan 17, 2:41 am)

Re: [patch] Converting writeback linked lists to a tree ba ..., Michael Rubin, (Thu Jan 17, 2:07 pm)

Re: [patch] Converting writeback linked lists to a tree ba ..., Fengguang Wu, (Thu Jan 17, 9:56 pm)

Re: [patch] Converting writeback linked lists to a tree ba ..., David Chinner, (Thu Jan 17, 10:01 pm)

Re: [patch] Converting writeback linked lists to a tree ba ..., Andi Kleen, (Thu Jan 17, 10:41 pm)

Re: [patch] Converting writeback linked lists to a tree ba ..., Fengguang Wu, (Thu Jan 17, 10:41 pm)

Re: [patch] Converting writeback linked lists to a tree ba ..., Fengguang Wu, (Thu Jan 17, 11:01 pm)

Re: [patch] Converting writeback linked lists to a tree ba ..., Michael Rubin, (Thu Jan 17, 11:43 pm)

Re: [patch] Converting writeback linked lists to a tree ba ..., Mike Waychison, (Fri Jan 18, 12:36 am)

Re: [patch] Converting writeback linked lists to a tree ba ..., Mike Waychison, (Fri Jan 18, 12:48 am)

Re: [patch] Converting writeback linked lists to a tree ba ..., David Chinner, (Fri Jan 18, 1:54 am)

Re: [patch] Converting writeback linked lists to a tree ba ..., Michael Rubin, (Fri Jan 18, 2:26 am)

Re: [patch] Converting writeback linked lists to a tree ba ..., Fengguang Wu, (Fri Jan 18, 2:32 am)

Navigation

Popular discussions


linux-kernel:
Greg Kroah-Hartman	[PATCH 041/196] kobject: add kobject_init_and_add function
Lukas Hejtmanek	Re: Another libata error related to OCZ SSD
Greg Kroah-Hartman	[PATCH 023/196] MCP_UCB1200: Convert from class_device to device
Florian Fainelli	Re: System clock runs too fast after 2.6.27 -> 2.6.28.1 upgrade
Christoph Lameter	[patch 1/4] mmu_notifier: Core code
git:
Johannes Schindelin	Re: [PATCH 1/2] Add strbuf_initf()
John Bito	[EGIT] Push to GitHub caused corruption
Jakub Narebski	Re: [PATCH 0/2] gitweb: patch view
Junio C Hamano	Re: [PATCH] When a remote is added but not fetched, tell the user.
Andy Parkins	Re: [RFC] Submodules in GIT
git-commits-head:
Linux Kernel Mailing List	ahci: Workaround HW bug for SB600/700 SATA controller PMP support
Linux Kernel Mailing List	V4L/DVB (11086): au0828: rename macro for currently non-function VBI support
Linux Kernel Mailing List	ceph: client types
Linux Kernel Mailing List	ceph: on-wire types
Linux Kernel Mailing List	crypto: chainiv - Use kcrypto_wq instead of keventd_wq
linux-netdev:
Andrew Morton	Re: [Bugme-new] [Bug 14969] New: b44: WOL does not work in suspended state
Giuseppe CAVALLARO	Re: [PATCH 03/13] stmmac: add the new Header file for stmmac platform data
Taku Izumi	[PATCH 3/3] ixgbe: add registers etc. printout code just before resetting adapters
Eric Dumazet	rps: some comments
Thomas Gleixner	Re: [RFC PATCH 02/12] On Tue, 23 Sep 2008, David Miller wrote:
openbsd-misc:
Stephan Andreas	problems with login after xlock in OpenBSD release 4.7
pmc	Make A Change. Alcoholism and Drug Addiction Treatment
ropers	Re: what exactly is enc0?
Fuad NAHDI	Re: What does your environment look like?
Matthew Szudzik	Typo on OpenBSD 4.4 CD Set

Colocation donated by:

Syndicate