Re: [PATCH 7/7] fs-writeback: only allow one inflight and pending full flush

2017-10-02 Thread Jan Kara
On Thu 28-09-17 20:09:50, Jens Axboe wrote: > On 09/25/2017 11:35 AM, Jan Kara wrote: > > On Thu 21-09-17 10:00:25, Jens Axboe wrote: > >> On 09/21/2017 09:36 AM, Jens Axboe wrote: > But more importantly once we are not guaranteed that we only have > a single global wb_writeback_work per

Re: [PATCH 7/7] fs-writeback: only allow one inflight and pending full flush

2017-09-29 Thread Jens Axboe
On 09/28/2017 08:09 PM, Jens Axboe wrote: > On 09/25/2017 11:35 AM, Jan Kara wrote: >> On Thu 21-09-17 10:00:25, Jens Axboe wrote: >>> On 09/21/2017 09:36 AM, Jens Axboe wrote: > But more importantly once we are not guaranteed that we only have > a single global wb_writeback_work per bdi_wr

Re: [PATCH 7/7] fs-writeback: only allow one inflight and pending full flush

2017-09-28 Thread Jens Axboe
On 09/25/2017 11:35 AM, Jan Kara wrote: > On Thu 21-09-17 10:00:25, Jens Axboe wrote: >> On 09/21/2017 09:36 AM, Jens Axboe wrote: But more importantly once we are not guaranteed that we only have a single global wb_writeback_work per bdi_writeback we should just embedd that into str

Re: [PATCH 7/7] fs-writeback: only allow one inflight and pending full flush

2017-09-25 Thread Jens Axboe
On 09/25/2017 03:35 AM, Jan Kara wrote: > On Thu 21-09-17 10:00:25, Jens Axboe wrote: >> On 09/21/2017 09:36 AM, Jens Axboe wrote: But more importantly once we are not guaranteed that we only have a single global wb_writeback_work per bdi_writeback we should just embedd that into str

Re: [PATCH 7/7] fs-writeback: only allow one inflight and pending full flush

2017-09-25 Thread Jan Kara
On Thu 21-09-17 10:00:25, Jens Axboe wrote: > On 09/21/2017 09:36 AM, Jens Axboe wrote: > >> But more importantly once we are not guaranteed that we only have > >> a single global wb_writeback_work per bdi_writeback we should just > >> embedd that into struct bdi_writeback instead of dynamically >

Re: [PATCH 7/7] fs-writeback: only allow one inflight and pending full flush

2017-09-21 Thread Christoph Hellwig
On Thu, Sep 21, 2017 at 10:00:25AM -0600, Jens Axboe wrote: > Something like the below would fit on top to do that. Gets rid of the > allocation and embeds the work item for global start-all in the > bdi_writeback structure. Something like that. Although if we still kalloc the global wb we wouldn

Re: [PATCH 7/7] fs-writeback: only allow one inflight and pending full flush

2017-09-21 Thread Jens Axboe
On 09/21/2017 09:36 AM, Jens Axboe wrote: >> But more importantly once we are not guaranteed that we only have >> a single global wb_writeback_work per bdi_writeback we should just >> embedd that into struct bdi_writeback instead of dynamically >> allocating it. > > We could do this as a followup.

Re: [PATCH 7/7] fs-writeback: only allow one inflight and pending full flush

2017-09-21 Thread Jens Axboe
On 09/21/2017 09:05 AM, Christoph Hellwig wrote: > On Wed, Sep 20, 2017 at 09:33:02AM -0600, Jens Axboe wrote: >> When someone calls wakeup_flusher_threads() or >> wakeup_flusher_threads_bdi(), they schedule writeback of all dirty >> pages in the system (or on that bdi). If we are tight on memory,

Re: [PATCH 7/7] fs-writeback: only allow one inflight and pending full flush

2017-09-21 Thread Christoph Hellwig
On Wed, Sep 20, 2017 at 09:33:02AM -0600, Jens Axboe wrote: > When someone calls wakeup_flusher_threads() or > wakeup_flusher_threads_bdi(), they schedule writeback of all dirty > pages in the system (or on that bdi). If we are tight on memory, we > can get tons of these queued from kswapd/vmscan.