On Tue, Nov 20, 2007 at 02:16:17PM +0100, Damien Wyart wrote:
> Hello,
>
> > > > Ok, so it's not synchronous writes that we are doing - we're just
> > > > submitting bio's tagged as WRITE_SYNC to get the I/O issued quickly.
> > > > The "synchronous" nature appears to be coming from higher level
>
Hello,
> > > Ok, so it's not synchronous writes that we are doing - we're just
> > > submitting bio's tagged as WRITE_SYNC to get the I/O issued
> > > quickly. The "synchronous" nature appears to be coming from higher
> > > level locking when reclaiming inodes (on the flush lock). It
> > >
Hello,
Ok, so it's not synchronous writes that we are doing - we're just
submitting bio's tagged as WRITE_SYNC to get the I/O issued
quickly. The synchronous nature appears to be coming from higher
level locking when reclaiming inodes (on the flush lock). It
appears that inode
On Tue, Nov 20, 2007 at 02:16:17PM +0100, Damien Wyart wrote:
Hello,
Ok, so it's not synchronous writes that we are doing - we're just
submitting bio's tagged as WRITE_SYNC to get the I/O issued quickly.
The synchronous nature appears to be coming from higher level
locking when
On Fri, Nov 09, 2007 at 09:10:47AM -0700, Jonathan Corbet wrote:
> Adrian Bunk <[EMAIL PROTECTED]> wrote:
>
> > What's missing is a definition which of them are formal tags that must
> > be explicitely given (look at point 13 in SubmittingPatches).
> >
> > Signed-off-by: and Reviewed-by: are
Adrian Bunk <[EMAIL PROTECTED]> wrote:
> What's missing is a definition which of them are formal tags that must
> be explicitely given (look at point 13 in SubmittingPatches).
>
> Signed-off-by: and Reviewed-by: are the formal tags someone must have
> explicitely given and that correspond to
Adrian Bunk [EMAIL PROTECTED] wrote:
What's missing is a definition which of them are formal tags that must
be explicitely given (look at point 13 in SubmittingPatches).
Signed-off-by: and Reviewed-by: are the formal tags someone must have
explicitely given and that correspond to some
On Fri, Nov 09, 2007 at 09:10:47AM -0700, Jonathan Corbet wrote:
Adrian Bunk [EMAIL PROTECTED] wrote:
What's missing is a definition which of them are formal tags that must
be explicitely given (look at point 13 in SubmittingPatches).
Signed-off-by: and Reviewed-by: are the formal
On Wed, Nov 07, 2007 at 08:15:06AM +0100, Torsten Kaiser wrote:
> On 11/7/07, David Chinner <[EMAIL PROTECTED]> wrote:
> > Ok, so it's not synchronous writes that we are doing - we're just
> > submitting bio's tagged as WRITE_SYNC to get the I/O issued quickly.
> > The "synchronous" nature appears
On Wed, Nov 07, 2007 at 08:15:06AM +0100, Torsten Kaiser wrote:
On 11/7/07, David Chinner [EMAIL PROTECTED] wrote:
Ok, so it's not synchronous writes that we are doing - we're just
submitting bio's tagged as WRITE_SYNC to get the I/O issued quickly.
The synchronous nature appears to be
On 11/7/07, David Chinner <[EMAIL PROTECTED]> wrote:
> Ok, so it's not synchronous writes that we are doing - we're just
> submitting bio's tagged as WRITE_SYNC to get the I/O issued quickly.
> The "synchronous" nature appears to be coming from higher level
> locking when reclaiming inodes (on the
On Wed, Nov 07, 2007 at 10:31:14AM +1100, David Chinner wrote:
> On Tue, Nov 06, 2007 at 10:53:25PM +0100, Torsten Kaiser wrote:
> > On 11/6/07, David Chinner <[EMAIL PROTECTED]> wrote:
> > > Rather than vmstat, can you use something like iostat to show how busy
> > > your
> > > disks are? i.e.
On Tue, Nov 06, 2007 at 10:53:25PM +0100, Torsten Kaiser wrote:
> On 11/6/07, David Chinner <[EMAIL PROTECTED]> wrote:
> > Rather than vmstat, can you use something like iostat to show how busy your
> > disks are? i.e. are we seeing RMW cycles in the raid5 or some such issue.
>
> Both "vmstat
On Tue, Nov 06, 2007 at 09:25:12AM -0700, Jonathan Corbet wrote:
> Andrew wrote:
>
> > > Reviewed-by: Fengguang Wu <[EMAIL PROTECTED]>
> >
> > I would prefer Tested-by: :(
>
> This seems like as good an opportunity as any to toss my patch tags
> document out there one more time. I still think
On 11/6/07, Fengguang Wu <[EMAIL PROTECTED]> wrote:
> --
> Subject: writeback: remove pages_skipped accounting in
> __block_write_full_page()
> From: Fengguang Wu <[EMAIL PROTECTED]>
>
> Miklos Szeredi <[EMAIL PROTECTED]> and me identified a
On 11/6/07, Peter Zijlstra <[EMAIL PROTECTED]> wrote:
> On Tue, 2007-11-06 at 15:25 +1100, David Chinner wrote:
>
> > I'm struggling to understand what possible changed in XFS or writeback that
> > would lead to stalls like this, esp. as you appear to be removing files when
> > the stalls occur.
>
On Tue, 2007-11-06 at 15:25 +1100, David Chinner wrote:
> I'm struggling to understand what possible changed in XFS or writeback that
> would lead to stalls like this, esp. as you appear to be removing files when
> the stalls occur.
Just a crazy idea,..
Could there be a set_page_dirty() that
> This seems like as good an opportunity as any to toss my patch tags
> document out there one more time. I still think it's a good idea to
> codify some sort of consensus on what these tags mean...
>
> jon
>
[snip]
> +By offering my Reviewed-by: tag, I state that:
> +
> + (a) I have carried
Andrew wrote:
> > Reviewed-by: Fengguang Wu <[EMAIL PROTECTED]>
>
> I would prefer Tested-by: :(
This seems like as good an opportunity as any to toss my patch tags
document out there one more time. I still think it's a good idea to
codify some sort of consensus on what these tags mean...
On Mon, 2007-11-05 at 15:57 -0800, Andrew Morton wrote:
> > > Subject: mm: speed up writeback ramp-up on clean systems
> > >
> > > We allow violation of bdi limits if there is a lot of room on the
> > > system. Once we hit half the total limit we start enforcing bdi limits
> > > and bdi ramp-up
On Fri, Nov 02, 2007 at 08:22:10PM +0100, Torsten Kaiser wrote:
> [ 547.20] mm/page-writeback.c 676 wb_kupdate: pdflush(285) 58858 >
> global 12829 72 0 wc __ tw 0 sk 0
> [ 550.48] mm/page-writeback.c 676 wb_kupdate: pdflush(285) 57834 >
> global 12017 62 0 wc __ tw 0 sk 0
> [
On Fri, Nov 02, 2007 at 08:22:10PM +0100, Torsten Kaiser wrote:
[ 547.20] mm/page-writeback.c 676 wb_kupdate: pdflush(285) 58858
global 12829 72 0 wc __ tw 0 sk 0
[ 550.48] mm/page-writeback.c 676 wb_kupdate: pdflush(285) 57834
global 12017 62 0 wc __ tw 0 sk 0
[ 552.71]
On Mon, 2007-11-05 at 15:57 -0800, Andrew Morton wrote:
Subject: mm: speed up writeback ramp-up on clean systems
We allow violation of bdi limits if there is a lot of room on the
system. Once we hit half the total limit we start enforcing bdi limits
and bdi ramp-up should happen.
Andrew wrote:
Reviewed-by: Fengguang Wu [EMAIL PROTECTED]
I would prefer Tested-by: :(
This seems like as good an opportunity as any to toss my patch tags
document out there one more time. I still think it's a good idea to
codify some sort of consensus on what these tags mean...
jon
This seems like as good an opportunity as any to toss my patch tags
document out there one more time. I still think it's a good idea to
codify some sort of consensus on what these tags mean...
jon
[snip]
+By offering my Reviewed-by: tag, I state that:
+
+ (a) I have carried out a
On Tue, 2007-11-06 at 15:25 +1100, David Chinner wrote:
I'm struggling to understand what possible changed in XFS or writeback that
would lead to stalls like this, esp. as you appear to be removing files when
the stalls occur.
Just a crazy idea,..
Could there be a set_page_dirty() that
On 11/6/07, Peter Zijlstra [EMAIL PROTECTED] wrote:
On Tue, 2007-11-06 at 15:25 +1100, David Chinner wrote:
I'm struggling to understand what possible changed in XFS or writeback that
would lead to stalls like this, esp. as you appear to be removing files when
the stalls occur.
Just a
On 11/6/07, Fengguang Wu [EMAIL PROTECTED] wrote:
--
Subject: writeback: remove pages_skipped accounting in
__block_write_full_page()
From: Fengguang Wu [EMAIL PROTECTED]
Miklos Szeredi [EMAIL PROTECTED] and me identified a writeback bug:
On Tue, Nov 06, 2007 at 09:25:12AM -0700, Jonathan Corbet wrote:
Andrew wrote:
Reviewed-by: Fengguang Wu [EMAIL PROTECTED]
I would prefer Tested-by: :(
This seems like as good an opportunity as any to toss my patch tags
document out there one more time. I still think it's a good
On Tue, Nov 06, 2007 at 10:53:25PM +0100, Torsten Kaiser wrote:
On 11/6/07, David Chinner [EMAIL PROTECTED] wrote:
Rather than vmstat, can you use something like iostat to show how busy your
disks are? i.e. are we seeing RMW cycles in the raid5 or some such issue.
Both vmstat 10 and
On Wed, Nov 07, 2007 at 10:31:14AM +1100, David Chinner wrote:
On Tue, Nov 06, 2007 at 10:53:25PM +0100, Torsten Kaiser wrote:
On 11/6/07, David Chinner [EMAIL PROTECTED] wrote:
Rather than vmstat, can you use something like iostat to show how busy
your
disks are? i.e. are we seeing
On 11/7/07, David Chinner [EMAIL PROTECTED] wrote:
Ok, so it's not synchronous writes that we are doing - we're just
submitting bio's tagged as WRITE_SYNC to get the I/O issued quickly.
The synchronous nature appears to be coming from higher level
locking when reclaiming inodes (on the flush
On 11/6/07, David Chinner <[EMAIL PROTECTED]> wrote:
> On Mon, Nov 05, 2007 at 07:27:16PM +0100, Torsten Kaiser wrote:
> > On 11/5/07, David Chinner <[EMAIL PROTECTED]> wrote:
> > > Ok, so it's probably a side effect of the writeback changes.
> > >
> > > Attached are two patches (two because one
On Mon, Nov 05, 2007 at 07:27:16PM +0100, Torsten Kaiser wrote:
> On 11/5/07, David Chinner <[EMAIL PROTECTED]> wrote:
> > Ok, so it's probably a side effect of the writeback changes.
> >
> > Attached are two patches (two because one was in a separate patchset as
> > a standalone change) that
On Fri, 2 Nov 2007 18:33:29 +0800
Fengguang Wu <[EMAIL PROTECTED]> wrote:
> On Fri, Nov 02, 2007 at 11:15:32AM +0100, Peter Zijlstra wrote:
> > On Fri, 2007-11-02 at 10:21 +0800, Fengguang Wu wrote:
> >
> > > Interestingly, no background_writeout() appears, but only
> > > balance_dirty_pages()
On 11/5/07, David Chinner <[EMAIL PROTECTED]> wrote:
> Ok, so it's probably a side effect of the writeback changes.
>
> Attached are two patches (two because one was in a separate patchset as
> a standalone change) that should prevent async writeback from blocking
> on locked inode cluster
On 11/5/07, David Chinner [EMAIL PROTECTED] wrote:
Ok, so it's probably a side effect of the writeback changes.
Attached are two patches (two because one was in a separate patchset as
a standalone change) that should prevent async writeback from blocking
on locked inode cluster buffers. Apply
On Fri, 2 Nov 2007 18:33:29 +0800
Fengguang Wu [EMAIL PROTECTED] wrote:
On Fri, Nov 02, 2007 at 11:15:32AM +0100, Peter Zijlstra wrote:
On Fri, 2007-11-02 at 10:21 +0800, Fengguang Wu wrote:
Interestingly, no background_writeout() appears, but only
balance_dirty_pages() and
On Mon, Nov 05, 2007 at 07:27:16PM +0100, Torsten Kaiser wrote:
On 11/5/07, David Chinner [EMAIL PROTECTED] wrote:
Ok, so it's probably a side effect of the writeback changes.
Attached are two patches (two because one was in a separate patchset as
a standalone change) that should prevent
On 11/6/07, David Chinner [EMAIL PROTECTED] wrote:
On Mon, Nov 05, 2007 at 07:27:16PM +0100, Torsten Kaiser wrote:
On 11/5/07, David Chinner [EMAIL PROTECTED] wrote:
Ok, so it's probably a side effect of the writeback changes.
Attached are two patches (two because one was in a separate
On 11/5/07, David Chinner <[EMAIL PROTECTED]> wrote:
> On Sun, Nov 04, 2007 at 12:19:19PM +0100, Torsten Kaiser wrote:
> > I can now confirm, that I see this also with the current
> > mainline-git-version
> > I used 2.6.24-rc1-git-b4f555081fdd27d13e6ff39d455d5aefae9d2c0c
> > plus the fix for the
On Sun, Nov 04, 2007 at 12:19:19PM +0100, Torsten Kaiser wrote:
> On 11/2/07, David Chinner <[EMAIL PROTECTED]> wrote:
> > That's stalled waiting on the inode cluster buffer lock. That implies
> > that the inode lcuser is already being written out and the inode has
> > been redirtied during
On 11/2/07, David Chinner <[EMAIL PROTECTED]> wrote:
> That's stalled waiting on the inode cluster buffer lock. That implies
> that the inode lcuser is already being written out and the inode has
> been redirtied during writeout.
>
> Does the kernel you are testing have the "flush inodes in
On 11/2/07, David Chinner [EMAIL PROTECTED] wrote:
That's stalled waiting on the inode cluster buffer lock. That implies
that the inode lcuser is already being written out and the inode has
been redirtied during writeout.
Does the kernel you are testing have the flush inodes in ascending
On Sun, Nov 04, 2007 at 12:19:19PM +0100, Torsten Kaiser wrote:
On 11/2/07, David Chinner [EMAIL PROTECTED] wrote:
That's stalled waiting on the inode cluster buffer lock. That implies
that the inode lcuser is already being written out and the inode has
been redirtied during writeout.
On 11/5/07, David Chinner [EMAIL PROTECTED] wrote:
On Sun, Nov 04, 2007 at 12:19:19PM +0100, Torsten Kaiser wrote:
I can now confirm, that I see this also with the current
mainline-git-version
I used 2.6.24-rc1-git-b4f555081fdd27d13e6ff39d455d5aefae9d2c0c
plus the fix for the sg changes
On 11/2/07, David Chinner <[EMAIL PROTECTED]> wrote:
> On Fri, Nov 02, 2007 at 08:22:10PM +0100, Torsten Kaiser wrote:
> > [ 630.00] SysRq : Emergency Sync
> > [ 630.12] Emergency Sync complete
> > [ 632.85] SysRq : Show Blocked State
> > [ 632.85] task
On Fri, Nov 02, 2007 at 08:22:10PM +0100, Torsten Kaiser wrote:
> [ 630.00] SysRq : Emergency Sync
> [ 630.12] Emergency Sync complete
> [ 632.85] SysRq : Show Blocked State
> [ 632.85] taskPC stack pid father
> [ 632.85] pdflush D
On 11/2/07, Peter Zijlstra <[EMAIL PROTECTED]> wrote:
> On Fri, 2007-11-02 at 10:21 +0800, Fengguang Wu wrote:
>
> > Interestingly, no background_writeout() appears, but only
> > balance_dirty_pages() and wb_kupdate. Obviously wb_kupdate won't
> > block the process.
>
> Yeah, the background
On 11/2/07, Fengguang Wu <[EMAIL PROTECTED]> wrote:
> I guess the new debug printks will provide more hints on it.
The "throttle_vm_writeout" did not trigger for my new workload.
Except one (the first) "balance_dirty_pages" came from line 445, the
newly added.
But I found an other workload that
On Fri, Nov 02, 2007 at 11:15:32AM +0100, Peter Zijlstra wrote:
> On Fri, 2007-11-02 at 10:21 +0800, Fengguang Wu wrote:
>
> > Interestingly, no background_writeout() appears, but only
> > balance_dirty_pages() and wb_kupdate. Obviously wb_kupdate won't
> > block the process.
>
> Yeah, the
On Fri, 2007-11-02 at 10:21 +0800, Fengguang Wu wrote:
> Interestingly, no background_writeout() appears, but only
> balance_dirty_pages() and wb_kupdate. Obviously wb_kupdate won't
> block the process.
Yeah, the background threshold is not (yet) scaled. So it can happen
that the bdi_dirty
On Fri, Nov 02, 2007 at 08:42:05AM +0100, Torsten Kaiser wrote:
> The Subject is still missleading, I'm using 2.6.23-mm1.
>
> On 11/2/07, Fengguang Wu <[EMAIL PROTECTED]> wrote:
> > On Thu, Nov 01, 2007 at 07:20:51PM +0100, Torsten Kaiser wrote:
> > > On 11/1/07, Fengguang Wu <[EMAIL PROTECTED]>
On 11/2/07, Fengguang Wu <[EMAIL PROTECTED]> wrote:
> On Thu, Nov 01, 2007 at 08:00:10PM +0100, Torsten Kaiser wrote:
> > On 11/1/07, Torsten Kaiser <[EMAIL PROTECTED]> wrote:
> > > On 11/1/07, Fengguang Wu <[EMAIL PROTECTED]> wrote:
> > > > Thank you. Maybe we can start by the applied debug patch
The Subject is still missleading, I'm using 2.6.23-mm1.
On 11/2/07, Fengguang Wu <[EMAIL PROTECTED]> wrote:
> On Thu, Nov 01, 2007 at 07:20:51PM +0100, Torsten Kaiser wrote:
> > On 11/1/07, Fengguang Wu <[EMAIL PROTECTED]> wrote:
> > > On Wed, Oct 31, 2007 at 04:22:10PM +0100, Torsten Kaiser
The Subject is still missleading, I'm using 2.6.23-mm1.
On 11/2/07, Fengguang Wu [EMAIL PROTECTED] wrote:
On Thu, Nov 01, 2007 at 07:20:51PM +0100, Torsten Kaiser wrote:
On 11/1/07, Fengguang Wu [EMAIL PROTECTED] wrote:
On Wed, Oct 31, 2007 at 04:22:10PM +0100, Torsten Kaiser wrote:
On Fri, Nov 02, 2007 at 08:42:05AM +0100, Torsten Kaiser wrote:
The Subject is still missleading, I'm using 2.6.23-mm1.
On 11/2/07, Fengguang Wu [EMAIL PROTECTED] wrote:
On Thu, Nov 01, 2007 at 07:20:51PM +0100, Torsten Kaiser wrote:
On 11/1/07, Fengguang Wu [EMAIL PROTECTED] wrote:
On 11/2/07, Fengguang Wu [EMAIL PROTECTED] wrote:
I guess the new debug printks will provide more hints on it.
The throttle_vm_writeout did not trigger for my new workload.
Except one (the first) balance_dirty_pages came from line 445, the
newly added.
But I found an other workload that looks
On Fri, Nov 02, 2007 at 11:15:32AM +0100, Peter Zijlstra wrote:
On Fri, 2007-11-02 at 10:21 +0800, Fengguang Wu wrote:
Interestingly, no background_writeout() appears, but only
balance_dirty_pages() and wb_kupdate. Obviously wb_kupdate won't
block the process.
Yeah, the background
On 11/2/07, Fengguang Wu [EMAIL PROTECTED] wrote:
On Thu, Nov 01, 2007 at 08:00:10PM +0100, Torsten Kaiser wrote:
On 11/1/07, Torsten Kaiser [EMAIL PROTECTED] wrote:
On 11/1/07, Fengguang Wu [EMAIL PROTECTED] wrote:
Thank you. Maybe we can start by the applied debug patch :-)
Will
On Fri, 2007-11-02 at 10:21 +0800, Fengguang Wu wrote:
Interestingly, no background_writeout() appears, but only
balance_dirty_pages() and wb_kupdate. Obviously wb_kupdate won't
block the process.
Yeah, the background threshold is not (yet) scaled. So it can happen
that the bdi_dirty limit
On 11/2/07, Peter Zijlstra [EMAIL PROTECTED] wrote:
On Fri, 2007-11-02 at 10:21 +0800, Fengguang Wu wrote:
Interestingly, no background_writeout() appears, but only
balance_dirty_pages() and wb_kupdate. Obviously wb_kupdate won't
block the process.
Yeah, the background threshold is not
On Fri, Nov 02, 2007 at 08:22:10PM +0100, Torsten Kaiser wrote:
[ 630.00] SysRq : Emergency Sync
[ 630.12] Emergency Sync complete
[ 632.85] SysRq : Show Blocked State
[ 632.85] taskPC stack pid father
[ 632.85] pdflush D
On 11/2/07, David Chinner [EMAIL PROTECTED] wrote:
On Fri, Nov 02, 2007 at 08:22:10PM +0100, Torsten Kaiser wrote:
[ 630.00] SysRq : Emergency Sync
[ 630.12] Emergency Sync complete
[ 632.85] SysRq : Show Blocked State
[ 632.85] taskPC stack
On Thu, Nov 01, 2007 at 08:00:10PM +0100, Torsten Kaiser wrote:
> On 11/1/07, Torsten Kaiser <[EMAIL PROTECTED]> wrote:
> > On 11/1/07, Fengguang Wu <[EMAIL PROTECTED]> wrote:
> > > Thank you. Maybe we can start by the applied debug patch :-)
> >
> > Will applied it and try to recreate this.
>
>
On Thu, Nov 01, 2007 at 07:20:51PM +0100, Torsten Kaiser wrote:
> On 11/1/07, Fengguang Wu <[EMAIL PROTECTED]> wrote:
> > On Wed, Oct 31, 2007 at 04:22:10PM +0100, Torsten Kaiser wrote:
> > > Since 2.6.23-mm1 I also experience strange hangs during heavy writeouts.
> > > Each time I noticed this I
On Thu, Nov 01, 2007 at 07:20:51PM +0100, Torsten Kaiser wrote:
On 11/1/07, Fengguang Wu [EMAIL PROTECTED] wrote:
On Wed, Oct 31, 2007 at 04:22:10PM +0100, Torsten Kaiser wrote:
Since 2.6.23-mm1 I also experience strange hangs during heavy writeouts.
Each time I noticed this I was using
On Thu, Nov 01, 2007 at 08:00:10PM +0100, Torsten Kaiser wrote:
On 11/1/07, Torsten Kaiser [EMAIL PROTECTED] wrote:
On 11/1/07, Fengguang Wu [EMAIL PROTECTED] wrote:
Thank you. Maybe we can start by the applied debug patch :-)
Will applied it and try to recreate this.
Patch applied,
68 matches
Mail list logo