Re: [md PATCH 2/3] md: remove md_super_wait() call after bitmap_flush()

[NeilBrown <neilb@suse.com>]

[-- Attachment #1: Type: text/plain, Size: 1293 bytes --]

On Thu, Nov 10 2016, Shaohua Li wrote:

> On Wed, Nov 09, 2016 at 10:21:32AM +1100, Neil Brown wrote:
>> bitmap_flush() finishes with bitmap_update_sb(), and that finishes
>> with write_page(..., 1), so write_page() will wait for all writes
>> to complete.  So there is no point calling md_super_wait()
>> immediately afterwards.
>> 
>> Signed-off-by: NeilBrown <neilb@suse.com>
>> ---
>>  drivers/md/md.c |    1 -
>>  1 file changed, 1 deletion(-)
>> 
>> diff --git a/drivers/md/md.c b/drivers/md/md.c
>> index f389d8abe137..1f1c7f007b68 100644
>> --- a/drivers/md/md.c
>> +++ b/drivers/md/md.c
>> @@ -5472,7 +5472,6 @@ static void __md_stop_writes(struct mddev *mddev)
>>  	del_timer_sync(&mddev->safemode_timer);
>>  
>>  	bitmap_flush(mddev);
>> -	md_super_wait(mddev);
>
> bitmap_flush() could be null if there is no bitmap, is this safe?

Good question.
If there is no bitmap, then all metadata updates (both superblock
and bad-block-list) are synchronous in md_update_sb(), which is always
called under ->reconfig_mutex and so which cannot race with this code.

So yes, it is safe.  That md_super_wait() was only ever intended to wait
for things that bitmap_flush() might have flushed, so it should have
been inside that function.

Thanks,
NeilBrown

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 800 bytes --]