From: "Huang\, Ying" <ying.huang@intel.com>
To: Jaegeuk Kim <jaegeuk@kernel.org>
Cc: "Huang\, Ying" <ying.huang@intel.com>,
kernel test robot <xiaolong.ye@intel.com>, <lkp@01.org>,
LKML <linux-kernel@vger.kernel.org>,
<linux-f2fs-devel@lists.sourceforge.net>
Subject: Re: [LKP] [lkp] [f2fs] ec795418c4: fsmark.files_per_sec -36.3% regression
Date: Thu, 04 Aug 2016 13:36:13 -0700 [thread overview]
Message-ID: <87r3a4b7b6.fsf@yhuang-mobile.sh.intel.com> (raw)
In-Reply-To: <20160804185251.GA13813@jaegeuk> (Jaegeuk Kim's message of "Thu, 4 Aug 2016 11:52:51 -0700")
Jaegeuk Kim <jaegeuk@kernel.org> writes:
> On Thu, Aug 04, 2016 at 10:44:20AM -0700, Huang, Ying wrote:
>> Jaegeuk Kim <jaegeuk@kernel.org> writes:
>>
>> > Hi Huang,
>> >
>> > On Thu, Aug 04, 2016 at 10:00:41AM -0700, Huang, Ying wrote:
>> >> Hi, Jaegeuk,
>> >>
>> >> "Huang, Ying" <ying.huang@intel.com> writes:
>> >> > Hi,
>> >> >
>> >> > I checked the comparison result below and found this is a regression for
>> >> > fsmark.files_per_sec, not fsmark.app_overhead.
>> >> >
>> >> > Best Regards,
>> >> > Huang, Ying
>> >> >
>> >> > kernel test robot <xiaolong.ye@intel.com> writes:
>> >> >
>> >> >> FYI, we noticed a -36.3% regression of fsmark.files_per_sec due to commit:
>> >> >>
>> >> >> commit ec795418c41850056feb956534edf059dc1155d4 ("f2fs: use percpu_rw_semaphore")
>> >> >> https://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs.git dev-test
>> >>
>> >> I found this has been merged by upstream. Do you have some plan to fix
>> >> it? Or you think the test itself has some problem?
>> >
>> > Sorry, too busy to take a look at this.
>> > The patch implements percpu_rw_semaphore which is intended to enhance FS
>> > scalability. Since I couldn't see any big regression in my test cases, could you
>> > check any debugging options which may give some overheads?
>>
>> The kernel config related with F2FS is as follow in our test,
>>
>> CONFIG_F2FS_FS=m
>> CONFIG_F2FS_STAT_FS=y
>> CONFIG_F2FS_FS_XATTR=y
>> CONFIG_F2FS_FS_POSIX_ACL=y
>> # CONFIG_F2FS_FS_SECURITY is not set
>> # CONFIG_F2FS_CHECK_FS is not set
>> # CONFIG_F2FS_FS_ENCRYPTION is not set
>> # CONFIG_F2FS_IO_TRACE is not set
>> # CONFIG_F2FS_FAULT_INJECTION is not set
>>
>> What do you think we need to change? Or do you mean some other
>> debugging options? Anyway, you can check our kernel config attached.
>>
>> > Let me recheck this with whole my tests.
>>
>> Maybe you can try our kernel config? Or if our kernel config is not
>> reasonable, can you help us to revise it? The full kernel config we
>> used is attached with the email.
>
> I could reproduce the fsmark regression in my machine and confirm there is
> another small regression as well.
> I'll revert this patch. Thank you.
>
> [lkp] [f2fs] 3bdad3c7ee: aim7.jobs-per-min -25.3% regression
> [lkp] [f2fs] b93f771286: aim7.jobs-per-min -81.2% regression
>
> In terms of the above regression, I could check that _reproduce_ procedure
> includes mounting filesystem only. Is that correct?
Sorry, our test system failed to generate reproduce steps for aim7. We
will implement it.
- [lkp] [f2fs] 3bdad3c7ee: aim7.jobs-per-min -25.3% regression
The disk is one 48G ram disk. The steps for aim7 is,
cat > workfile <<EOF
FILESIZE: 1M
POOLSIZE: 10M
10 disk_cp
EOF
(
echo $HOSTNAME
echo disk_cp
echo 1
echo 3000
echo 2
echo 3000
echo 1
) | ./multitask -t &
- [lkp] [f2fs] b93f771286: aim7.jobs-per-min -81.2% regression
The disk is 4 12G ram disk, and setup RAID0 on them via mdadm. The
steps for aim7 is,
cat > workfile <<EOF
FILESIZE: 1M
POOLSIZE: 10M
10 sync_disk_rw
EOF
(
echo $HOSTNAME
echo sync_disk_rw
echo 1
echo 600
echo 2
echo 600
echo 1
) | ./multitask -t &
Best Regards,
Huang, Ying
next prev parent reply other threads:[~2016-08-04 20:36 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-07-18 2:09 [lkp] [f2fs] ec795418c4: fsmark.app_overhead -36.3% regression kernel test robot
2016-07-18 20:27 ` [LKP] " Huang, Ying
2016-08-04 17:00 ` [LKP] [lkp] [f2fs] ec795418c4: fsmark.files_per_sec " Huang, Ying
2016-08-04 17:24 ` Jaegeuk Kim
2016-08-04 17:44 ` Huang, Ying
2016-08-04 18:52 ` Jaegeuk Kim
2016-08-04 20:36 ` Huang, Ying [this message]
2016-08-11 22:49 ` Huang, Ying
2016-08-12 1:22 ` Jaegeuk Kim
2016-08-24 16:51 ` huang ying
2016-08-27 0:52 ` Jaegeuk Kim
2016-08-27 2:13 ` Fengguang Wu
2016-08-30 2:30 ` Jaegeuk Kim
2016-08-30 16:44 ` Huang, Ying
2016-09-26 6:26 ` Huang, Ying
2016-09-26 18:23 ` Jaegeuk Kim
2016-09-27 0:50 ` Huang, Ying
2016-09-27 1:41 ` Jaegeuk Kim
2016-10-31 3:14 ` Huang, Ying
2016-10-31 17:42 ` Jaegeuk Kim
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=87r3a4b7b6.fsf@yhuang-mobile.sh.intel.com \
--to=ying.huang@intel.com \
--cc=jaegeuk@kernel.org \
--cc=linux-f2fs-devel@lists.sourceforge.net \
--cc=linux-kernel@vger.kernel.org \
--cc=lkp@01.org \
--cc=xiaolong.ye@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).