Re: Recent kernel "mount" slow

From: Jeff Chua
Date: Fri Nov 23 2012 - 18:47:54 EST


On Sat, Nov 24, 2012 at 7:31 AM, Jeff Chua <jeff.chua.linux@xxxxxxxxx> wrote:
> On Sat, Nov 24, 2012 at 6:21 AM, Jeff Chua <jeff.chua.linux@xxxxxxxxx> wrote:
>> On Fri, Nov 23, 2012 at 9:24 PM, Jens Axboe <axboe@xxxxxxxxx> wrote:
>>> On 2012-11-22 20:21, Linus Torvalds wrote:
>>>> Doesn't sound like a fsdevel issue since it seems to be independent of
>>>> filesystems. More like some generic block layer thing. Adding Jens
>>>> (and quoting the whole thing)
>>>>
>>>> Jens, any ideas? Most of your stuff came in after -rc2, which would
>>>> fit with the fact that most of the slowdown seems to be after -rc2
>>>> according to Jeff.
>>>
>>> No ideas. Looking at what went in from my side, only the rq plug sorting
>>> is a core change, and that should not cause any change in behaviour for
>>> a single device. That's commit 975927b9.
>>>
>>>> Jeff, more bisecting would be good, though.
>>>
>>> Probably required, yes...
>>
>>
>> This one slows mount from 0.012s to 0.168s.
>>
>> commit 62ac665ff9fc07497ca524bd20d6a96893d11071
>> Author: Mikulas Patocka <mpatocka@xxxxxxxxxx>
>> Date: Wed Sep 26 07:46:43 2012 +0200
>>
>> blockdev: turn a rw semaphore into a percpu rw semaphore
>>
>>
>> There were couple of more changes to percpu-rw-semaphores after
>> 3.7.0-rc2 and those slows mount further from 0.168s to 0.500s. I don't
>> really know, but I'm suspecting these. Still bisecting.
>>
>> commit 5c1eabe68501d1e1b1586c7f4c46cc531828c4ab
>> Author: Mikulas Patocka <mpatocka@xxxxxxxxxx>
>> Date: Mon Oct 22 19:37:47 2012 -0400
>>
>> percpu-rw-semaphores: use light/heavy barriers
>>
>>
>> commit 1bf11c53535ab87e3bf14ecdf6747bf46f601c5d
>> Author: Mikulas Patocka <mpatocka@xxxxxxxxxx>
>> Date: Mon Oct 22 19:39:16 2012 -0400
>>
>> percpu-rw-semaphores: use rcu_read_lock_sched
>
>
> I reverted these 3 patches and mount time is now 0.012s.
>
> # time mount /dev/sda1 /mnt; sync; sync; umount /mnt
>
>
>> commit 1a25b1c4ce189e3926f2981f3302352a930086db
>> Author: Mikulas Patocka <mpatocka@xxxxxxxxxx>
>> Date: Mon Oct 15 17:20:17 2012 -0400
>>
>> Lock splice_read and splice_write functions
>
> Looks like this one is not the problem. But I reverted it anyway
> because it's part of the same chunk.
>
>
> Happy Thanksgiving!

I'm on latest git 3.7.0-rc6 5e351cdc998db82935d1248a053a1be37d1160fd
and with the above patches reverted and mount time is now 0.012s.

Jeff
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/