Re: [RFC] ext4: Don't send extra barrier during fsync if there are no dirty pages. | KernelTrap

Re: [RFC] ext4: Don't send extra barrier during fsync if there are no dirty pages.

view
thread

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

[view in full thread]

From: Christoph Hellwig

Subject: Re: [RFC] ext4: Don't send extra barrier during fsync if there are no dirty pages.

Date: Tuesday, August 3, 2010 - 2:01 am

On Mon, Aug 02, 2010 at 05:09:39PM -0700, Darrick J. Wong wrote:
quoted text> Well... on my fsync-happy workloads, this seems to cut the barrier count down
> by about 20%, and speeds it up by about 20%.

Care to share the test case for this?  I'd be especially interesting on
how it behaves with non-draining barriers / cache flushes in fsync.

--
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:

[RFC] ext4: Don't send extra barrier during fsync if there ..., Darrick J. Wong, (Thu Apr 29, 4:51 pm)

Re: [RFC] ext4: Don't send extra barrier during fsync if t ..., Mingming Cao, (Mon May 3, 5:57 pm)

Re: [RFC] ext4: Don't send extra barrier during fsync if t ..., Ric Wheeler, (Tue May 4, 7:16 am)

Re: [RFC] ext4: Don't send extra barrier during fsync if t ..., Mingming Cao, (Tue May 4, 12:49 pm)

[RFC v2] ext4: Don't send extra barrier during fsync if th ..., Darrick J. Wong, (Tue Jun 29, 1:51 pm)

Re: [RFC] ext4: Don't send extra barrier during fsync if t ..., tytso, (Wed Jun 30, 5:48 am)

Re: [RFC] ext4: Don't send extra barrier during fsync if t ..., Ric Wheeler, (Wed Jun 30, 6:21 am)

Re: [RFC] ext4: Don't send extra barrier during fsync if t ..., tytso, (Wed Jun 30, 6:44 am)

Re: [RFC] ext4: Don't send extra barrier during fsync if t ..., Ric Wheeler, (Wed Jun 30, 6:54 am)

Re: [RFC] ext4: Don't send extra barrier during fsync if t ..., Jan Kara, (Wed Jul 21, 10:16 am)

Re: [RFC] ext4: Don't send extra barrier during fsync if t ..., Darrick J. Wong, (Mon Aug 2, 5:09 pm)

Re: [RFC] ext4: Don't send extra barrier during fsync if t ..., Christoph Hellwig, (Tue Aug 3, 2:01 am)

Re: [RFC] ext4: Don't send extra barrier during fsync if t ..., Jan Kara, (Tue Aug 3, 6:21 am)

Re: [RFC] ext4: Don't send extra barrier during fsync if t ..., Avi Kivity, (Tue Aug 3, 6:24 am)

Re: [RFC] ext4: Don't send extra barrier during fsync if t ..., Darrick J. Wong, (Wed Aug 4, 11:16 am)

Re: [RFC] ext4: Don't send extra barrier during fsync if t ..., Ted Ts'o, (Wed Aug 4, 4:32 pm)

Re: [RFC] ext4: Don't send extra barrier during fsync if t ..., Avi Kivity, (Wed Aug 4, 7:20 pm)

Re: [RFC] ext4: Don't send extra barrier during fsync if t ..., Ted Ts'o, (Thu Aug 5, 9:17 am)

Re: [RFC v2] ext4: Don't send extra barrier during fsync i ..., Ted Ts'o, (Thu Aug 5, 9:40 am)

Re: [RFC v2] ext4: Don't send extra barrier during fsync i ..., Ted Ts'o, (Thu Aug 5, 9:45 am)

Re: [RFC] ext4: Don't send extra barrier during fsync if t ..., Jeff Moyer, (Thu Aug 5, 12:13 pm)

Re: [RFC] ext4: Don't send extra barrier during fsync if t ..., Ted Ts'o, (Thu Aug 5, 1:39 pm)

Re: [RFC] ext4: Don't send extra barrier during fsync if t ..., Jeff Moyer, (Thu Aug 5, 1:44 pm)

Re: [RFC v2] ext4: Don't send extra barrier during fsync i ..., Darrick J. Wong, (Fri Aug 6, 12:04 am)

Re: [RFC v2] ext4: Don't send extra barrier during fsync i ..., Darrick J. Wong, (Fri Aug 6, 12:13 am)

Re: [RFC v2] ext4: Don't send extra barrier during fsync i ..., Ric Wheeler, (Fri Aug 6, 3:17 am)

Re: [RFC v2] ext4: Don't send extra barrier during fsync i ..., Ted Ts'o, (Fri Aug 6, 11:04 am)

Re: [RFC v2] ext4: Don't send extra barrier during fsync i ..., Darrick J. Wong, (Mon Aug 9, 12:36 pm)

[RFC v3] ext4: Combine barrier requests coming from fsync, Darrick J. Wong, (Mon Aug 9, 12:53 pm)

Re: [RFC v3] ext4: Combine barrier requests coming from fsync, Christoph Hellwig, (Mon Aug 9, 2:07 pm)

Re: [RFC v3] ext4: Combine barrier requests coming from fsync, Andreas Dilger, (Mon Aug 9, 2:19 pm)

Re: [RFC v3] ext4: Combine barrier requests coming from fsync, Darrick J. Wong, (Mon Aug 9, 4:38 pm)

Re: [RFC v3] ext4: Combine barrier requests coming from fsync, Darrick J. Wong, (Mon Aug 16, 9:14 am)

Re: [RFC v3] ext4: Combine barrier requests coming from fsync, Darrick J. Wong, (Wed Aug 18, 7:07 pm)

[RFC v4] ext4: Coordinate fsync requests, Darrick J. Wong, (Wed Aug 18, 7:14 pm)

Re: [RFC v3] ext4: Combine barrier requests coming from fsync, Christoph Hellwig, (Thu Aug 19, 1:53 am)

Re: [RFC v3] ext4: Combine barrier requests coming from fsync, Tejun Heo, (Thu Aug 19, 2:17 am)

Re: [RFC v3] ext4: Combine barrier requests coming from fsync, Tejun Heo, (Thu Aug 19, 8:48 am)

Performance testing of various barrier reduction patches [ ..., Darrick J. Wong, (Mon Aug 23, 11:31 am)

Re: Performance testing of various barrier reduction patch ..., Darrick J. Wong, (Thu Sep 23, 4:25 pm)

Re: Performance testing of various barrier reduction patch ..., Andreas Dilger, (Thu Sep 23, 11:24 pm)

Re: Performance testing of various barrier reduction patch ..., Ric Wheeler, (Fri Sep 24, 4:44 am)

Re: Performance testing of various barrier reduction patch ..., Darrick J. Wong, (Mon Sep 27, 4:01 pm)

Re: Performance testing of various barrier reduction patch ..., Darrick J. Wong, (Fri Oct 8, 2:26 pm)

Re: Performance testing of various barrier reduction patch ..., Ric Wheeler, (Fri Oct 8, 2:56 pm)

Re: Performance testing of various barrier reduction patch ..., Ted Ts'o, (Mon Oct 11, 7:33 am)

Re: Performance testing of various barrier reduction patch ..., Darrick J. Wong, (Mon Oct 11, 1:20 pm)

Re: Performance testing of various barrier reduction patch ..., Christoph Hellwig, (Tue Oct 12, 7:14 am)

Re: Performance testing of various barrier reduction patch ..., Darrick J. Wong, (Fri Oct 15, 4:39 pm)

Re: Performance testing of various barrier reduction patch ..., Christoph Hellwig, (Fri Oct 15, 4:40 pm)

Re: Performance testing of various barrier reduction patch ..., Darrick J. Wong, (Fri Oct 15, 5:02 pm)

Re: Performance testing of various barrier reduction patch ..., Darrick J. Wong, (Mon Oct 18, 3:49 pm)

Re: Performance testing of various barrier reduction patch ..., Christoph Hellwig, (Tue Oct 19, 11:28 am)

Navigation

Popular discussions


linux-kernel:
Paul Turner	[tg_shares_up rewrite v4 11/11] sched: update tg->shares after cpu.shares write
Mr. James W. Laferriere	Re: Linux 2.6.25-rc1 , syntax error near unexpected token `;'
Robin Lee Powell	NFS hang + umount -f: better behaviour requested.
Linus Torvalds	Linux 2.6.34-rc4
Colin Cross	[PATCH 12/21] ARM: tegra: Add suspend and hotplug support
git:
Fredrik Kuivinen	Re: fatal: unable to create '.git/index': File exists
Wink Saville	How-to combine several separate git repos?
Emily Ren	How to pull remote branch with specified commit id?
Denis Bueno	Git clone error
pradeep singh	git-update-server-info may be required,cannot clone and pull from a remote reposit...
openbsd-misc:
Sevan / Venture37	Re: This is what Linus Torvalds calls openBSD crowd
Netmaffia.hu	Tini Lányok AKCIÓBAN OTTHON
Siju George	This is what Linus Torvalds calls openBSD crowd
Darrin Chandler	Re: OT: Python (was Re: vi in /bin)
frantisek holop	Re: splassert: vwakeup: and friends
linux-netdev:
Jamie Lokier	Re: POHMELFS high performance network filesystem. Transactions, failover, performa...
Timo Teräs	ip xfrm policy semantics
Ron Mercer	[net-next PATCH 2/2] qlge: Version change to v1.00.00.27
Maciej W. Rozycki	Re: [PATCH] flush_work_sync vs. flush_scheduled_work Re: [PATCH] PHYLIB: IRQ event...
Jarek Poplawski	Re: socket api problem: can't bind an ipv6 socket to ::ffff:0.0.0.0
git-commits-head:
Linux Kernel Mailing List	powerpc/fsl_msi: enable msi allocation in all banks
Linux Kernel Mailing List	ASoC: fix registration of the SoC card in the Freescale MPC8610 drivers
Linux Kernel Mailing List	drivers/acpi: use kasprintf
Linux Kernel Mailing List	bnx2x: Moving includes
Linux Kernel Mailing List	[ARM] mmp: add dma support to mmp2

Colocation donated by:

Syndicate