> Only if the queue depth is not bound. Queue depths are bound and so
> the distance we can go over the threshold is limited. This is the
> fundamental principle on which the throttling is based.
>
> Hence, if the queue is not full, then we will have either written
> dirty pages to it (i.e wb
On Tue, Mar 13, 2007 at 09:21:59AM +0100, Miklos Szeredi wrote:
> > > read request
> > > sys_write
> > > mutex_lock(i_mutex)
> > > ...
> > > balance_dirty_pages
> > > submit write requests
> > > loop ... write requests completed ... dirty still over limit ...
> > > ... l
> > > IIUC, your problem is that there's another bdi that holds all the
> > > dirty pages, and this throttle loop never flushes pages from that
> > > other bdi and we sleep instead. It seems to me that the fundamental
> > > problem is that to clean the pages we need to flush both bdi's, not
> > > j
On Mon, Mar 12, 2007 at 11:36:16PM +0100, Miklos Szeredi wrote:
> I'll try to explain the reason for the deadlock first.
Ah, thanks for that.
> > IIUC, your problem is that there's another bdi that holds all the
> > dirty pages, and this throttle loop never flushes pages from that
> > other bdi a
I'll try to explain the reason for the deadlock first.
> IIUC, your problem is that there's another bdi that holds all the
> dirty pages, and this throttle loop never flushes pages from that
> other bdi and we sleep instead. It seems to me that the fundamental
> problem is that to clean the pages
On Mon, Mar 12, 2007 at 12:40:47PM +0100, Miklos Szeredi wrote:
> > > I have no idea how serious the scalability problems with this are. If
> > > they are serious, different solutions can probably be found for the
> > > above, but this is certainly the simplest.
> >
> > Atomic operations to a sin
> > I have no idea how serious the scalability problems with this are. If
> > they are serious, different solutions can probably be found for the
> > above, but this is certainly the simplest.
>
> Atomic operations to a single per-backing device from all CPUs at once?
> That's a pretty serious sc
On Tue, Mar 06, 2007 at 07:04:46PM +0100, Miklos Szeredi wrote:
> From: Andrew Morton <[EMAIL PROTECTED]>
>
> [EMAIL PROTECTED]: bugfix]
>
> Miklos Szeredi <[EMAIL PROTECTED]>:
>
> Changes:
> - updated to apply after clear_page_dirty_for_io() race fix
>
> This is needed for
>
> - balance_dir
From: Andrew Morton <[EMAIL PROTECTED]>
[EMAIL PROTECTED]: bugfix]
Miklos Szeredi <[EMAIL PROTECTED]>:
Changes:
- updated to apply after clear_page_dirty_for_io() race fix
This is needed for
- balance_dirty_pages() deadlock fix
- fuse dirty page accounting
I have no idea how serious the sc
9 matches
Mail list logo