Re: [PATCH 2/6] writeback: wb_start_writeback() should use WB_SYNC_ALL for WB_REASON_SYNC

From	Jens Axboe <axboe@fb.com>
Newsgroups	linux.kernel
Subject	Re: [PATCH 2/6] writeback: wb_start_writeback() should use WB_SYNC_ALL for WB_REASON_SYNC
Date	2016-03-22 23:00 +0100
Message-ID	<rfCvg-276-13@gated-at.bofh.it> (permalink)
References	<rfyKZ-88A-3@gated-at.bofh.it> <rfyL0-88A-31@gated-at.bofh.it> <rfCbU-20m-7@gated-at.bofh.it> <rfClA-23H-5@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On 03/22/2016 03:40 PM, Jens Axboe wrote:
> On 03/22/2016 03:34 PM, Dave Chinner wrote:
>> On Tue, Mar 22, 2016 at 11:55:16AM -0600, Jens Axboe wrote:
>>> If you call sync, the initial call to wakeup_flusher_threads() ends up
>>> calling wb_start_writeback() with reason=WB_REASON_SYNC, but
>>> wb_start_writeback() always uses WB_SYNC_NONE as the writeback mode.
>>> Ensure that we use WB_SYNC_ALL for a sync operation.
>>
>> This seems wrong to me. We want background write to happen as
>> quickly as possible and /not block/ when we first kick sync.
>
> It's not going to block. wakeup_flusher_threads() async queues writeback
> work through wb_start_writeback().

For block here, you mean the async work ending up doing 
wait_on_page_writeback() because we're doing WB_SYNC_ALL instead of 
WB_SYNC_NONE?

And if so:

>> The latter blocking passes of sync use WB_SYNC_ALL to ensure that we
>> block waiting for all remaining IO to be issued and waited on, but
>> the background writeback doesn't need to do this.

why not have it do that?

-- 
Jens Axboe

Thread

[PATCH 2/6] writeback: wb_start_writeback() should use WB_SYNC_ALL for WB_REASON_SYNC Jens Axboe <axboe@fb.com> - 2016-03-22 19:00 +0100
  Re: [PATCH 2/6] writeback: wb_start_writeback() should use  WB_SYNC_ALL for WB_REASON_SYNC Dave Chinner <david@fromorbit.com> - 2016-03-22 22:40 +0100
    Re: [PATCH 2/6] writeback: wb_start_writeback() should use  WB_SYNC_ALL for WB_REASON_SYNC Jens Axboe <axboe@fb.com> - 2016-03-22 22:50 +0100
      Re: [PATCH 2/6] writeback: wb_start_writeback() should use  WB_SYNC_ALL for WB_REASON_SYNC Jens Axboe <axboe@fb.com> - 2016-03-22 23:00 +0100
      Re: [PATCH 2/6] writeback: wb_start_writeback() should use  WB_SYNC_ALL for WB_REASON_SYNC Dave Chinner <david@fromorbit.com> - 2016-03-22 23:10 +0100
        Re: [PATCH 2/6] writeback: wb_start_writeback() should use  WB_SYNC_ALL for WB_REASON_SYNC Jens Axboe <axboe@fb.com> - 2016-03-22 23:10 +0100

csiph-web