From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mout.gmx.net ([212.227.15.19]:53967 "EHLO mout.gmx.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752097AbdFHFGi (ORCPT ); Thu, 8 Jun 2017 01:06:38 -0400 Message-ID: <1496898372.8617.27.camel@gmx.de> Subject: Re: [PATCH] Fix loop device flush before configure v2 From: Mike Galbraith To: James Wang , axboe@fb.com, ming.lei@redhat.com Cc: hare@suse.com, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, mgorman@suse.com Date: Thu, 08 Jun 2017 07:06:12 +0200 In-Reply-To: <20170608021734.5025-1-jnwang@suse.com> References: <20170608021734.5025-1-jnwang@suse.com> Content-Type: text/plain; charset="UTF-8" Mime-Version: 1.0 Sender: linux-block-owner@vger.kernel.org List-Id: linux-block@vger.kernel.org On Thu, 2017-06-08 at 10:17 +0800, James Wang wrote: > This condition check was exist at before commit b5dd2f6047ca ("block: loop: > improve performance via blk-mq") When add MQ support to loop device, it be > removed because the member of '->lo_thread' be removed. And then upstream > add '->worker_task', I think they forget add it to here. > > When I install SLES-12 product is base on 4.4 kernel I found installer will > hang +60 second at scan disks. and I found LVM tools would take this action. > finally I found this problem is more obvious on AMD platform. This problem > will impact all scenarios that scan loop devcie. > > When the loop device didn't configure backing file or Request Queue, we > shouldn't to cost a lot of time to flush it. The changelog sounds odd to me, perhaps reword/condense a bit?... While installing SLES-12 (based on v4.4), I found that the installer will stall for 60+ seconds during LVM disk scan. The root cause was determined to be the removal of a bound device check in loop_flush() by commit b5dd2f6047ca ("block: loop: improve performance via blk-mq"). Restoring this check, examining ->lo_state as set by loop_set_fd() eliminates the bad behavior. Test method: modprobe loop max_loop=64 dd if=/dev/zero of=disk bs=512 count=200K for((i=0;i<4;i++))do losetup -f disk; done mkfs.ext4 -F /dev/loop0 for((i=0;i<4;i++))do mkdir t$i; mount /dev/loop$i t$i;done for f in `ls /dev/loop[0-9]*|sort`; do \ echo $f; dd if=$f of=/dev/null bs=512 count=1; \ done Test output: stock patched /dev/loop0 18.1217e-05 8.3842e-05 /dev/loop1 6.1114e-05 0.000147979 /dev/loop10 0.414701 0.000116564 /dev/loop11 0.7474 6.7942e-05 /dev/loop12 0.747986 8.9082e-05 /dev/loop13 0.746532 7.4799e-05 /dev/loop14 0.480041 9.3926e-05 /dev/loop15 1.26453 7.2522e-05 Note that from loop10 onward, the device is not mounted, yet the stock kernel consumes several orders of magnitude more wall time than it does for a mounted device. Reviewed-by: Hannes Reinecke Signed-off-by: James Wang Fixes: b5dd2f6047ca ("block: loop: improve performance via blk-mq") --- > drivers/block/loop.c | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/drivers/block/loop.c b/drivers/block/loop.c > index 48f6fa6f810e..2e5b8538760c 100644 > --- a/drivers/block/loop.c > +++ b/drivers/block/loop.c > @@ -625,6 +625,9 @@ static int loop_switch(struct loop_device *lo, struct file *file) > */ > static int loop_flush(struct loop_device *lo) > { > + /* loop not yet configured, no running thread, nothing to flush */ > + if (lo->lo_state != Lo_bound) > + return 0; > return loop_switch(lo, NULL); > } >