public inbox for fstests@vger.kernel.org
 help / color / mirror / Atom feed
From: Ross Zwisler <ross.zwisler@linux.intel.com>
To: Xiong Zhou <xzhou@redhat.com>
Cc: ross.zwisler@linux.intel.com, fstests@vger.kernel.org,
	linux-nvdimm@ml01.01.org, jmoyer@redhat.com, eguan@redhat.com
Subject: Re: [PATCH v5 2/2] generic: test mmap io through DAX and non-DAX
Date: Mon, 6 Feb 2017 12:17:17 -0700	[thread overview]
Message-ID: <20170206191717.GB12290@linux.intel.com> (raw)
In-Reply-To: <1486275704-18917-3-git-send-email-xzhou@redhat.com>

On Sun, Feb 05, 2017 at 02:21:44PM +0800, Xiong Zhou wrote:
> Mount TEST_DEV as non-DAX, SCRATCH_DEV as DAX, then
> do some IO between them. In this case we use mmap
> and dio/buffered IO read/write test programme.
> 
> Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com>

Yep, this is fine.

> Signed-off-by: Xiong Zhou <xzhou@redhat.com>
<>
> diff --git a/tests/generic/407 b/tests/generic/407
<>
> +do_tests()
> +{
> +	# less than page size
> +	t_mmap_dio_dax 1024
> +	# page size
> +	t_mmap_dio_dax `src/feature -s`
> +	# bigger sizes, for PMD faults
> +	t_mmap_dio_dax $((16 * 1024 * 1024))
> +	t_mmap_dio_dax $((64 * 1024 * 1024))
> +}
> +
> +# make fs 2Mb aligned for PMD fault testing
> +if [[ $FSTYP =~ ext ]] ; then
> +	_scratch_mkfs "-E stride=512,stripe_width=1" \
> +		> /dev/null 2>&1
> +elif [[ $FSTYP =~ xfs ]] ; then
> +	_scratch_mkfs "-d su=2m,sw=1" > /dev/null 2>&1
> +fi
> +
> +# mount SCRATCH_DEV with dax option, TEST_DEV not
> +export MOUNT_OPTIONS=""
> +export TEST_FS_MOUNT_OPTS=""
> +_test_cycle_mount
> +_fs_options $TEST_DEV | grep -qw "dax" && \
> +	_notrun "we need $TEST_DEV mount without dax"
> +_scratch_mount "-o dax"
> +
> +tsize=$((64 * 1024 * 1024))

Ugh - for some reason 64 MiB files for me don't get PMDs on ext4 because the
ext4 block allocator gives us misaligned blocks.  I'll follow up with them in
a new thread on the ext4 list and I'll cc you, Xiong.

If we up this to 128 MiB, so

tsize=$((128 * 1024 * 1024))

I get PMDs for both ext4 and XFS.  Can we make this one change before
applying?

The rest looks good to me.  Thanks for the tests!

Reviewed-by: Ross Zwisler <ross.zwisler@linux.intel.com>

  reply	other threads:[~2017-02-06 19:17 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-01-18  3:07 [PATCH 0/4] mmap dio and DAX Xiong Zhou
2017-01-18  3:07 ` [PATCH 1/4] common/rc: add _require_scratch_dax Xiong Zhou
2017-01-19  4:17   ` Eryu Guan
2017-01-18  3:07 ` [PATCH 2/4] src/t_mmap_dio: add mmap dio test Xiong Zhou
2017-01-18  3:07 ` [PATCH 3/4] xfs: test per-inode DAX flag by IO Xiong Zhou
2017-01-19  4:24   ` Eryu Guan
2017-01-18  3:07 ` [PATCH 4/4] generic: test mmap dio through DAX and non-DAX Xiong Zhou
2017-01-19  5:54   ` Eryu Guan
2017-01-19 10:13     ` [PATCH v2 0/2] mmap dio and DAX Xiong Zhou
2017-01-19 10:13       ` [PATCH v2 1/2] xfs: test per-inode DAX flag by IO Xiong Zhou
2017-01-19 20:49         ` Ross Zwisler
2017-01-20  1:15           ` Xiong Zhou
2017-01-20  2:21           ` [PATCH v3] " Xiong Zhou
2017-01-20  6:15             ` [PATCH v4 0/2] mmap dio and DAX Xiong Zhou
2017-01-20  6:15               ` [PATCH v4 1/2] xfs: test per-inode DAX flag by IO Xiong Zhou
2017-02-04 15:17                 ` [PATCH] fixup! " Ross Zwisler
2017-02-05  6:21                   ` [PATCH v5 0/2] mmap dio and DAX Xiong Zhou
2017-02-05  6:21                     ` [PATCH v5 1/2] xfs: test per-inode DAX flag by IO Xiong Zhou
2017-02-06 19:13                       ` Ross Zwisler
2017-02-05  6:21                     ` [PATCH v5 2/2] generic: test mmap io through DAX and non-DAX Xiong Zhou
2017-02-06 19:17                       ` Ross Zwisler [this message]
2017-02-08  4:11                         ` [PATCH v6 0/2] mmap dio and DAX Xiong Zhou
2017-02-08  4:11                           ` [PATCH v6 1/2] xfs: test per-inode DAX flag by IO Xiong Zhou
2017-02-08  4:11                           ` [PATCH v6 2/2] generic: test mmap io through DAX and non-DAX Xiong Zhou
2017-02-17  6:46                             ` Eryu Guan
2017-02-17  6:52                               ` Xiong Zhou
2017-01-20  6:15               ` [PATCH v4 " Xiong Zhou
2017-01-24 22:28               ` [PATCH v4 0/2] mmap dio and DAX Ross Zwisler
2017-02-03  5:57                 ` Xiong Zhou
2017-02-03  6:29                   ` Eryu Guan
2017-02-03 16:57                   ` Ross Zwisler
2017-02-04 10:14                     ` Xiong Zhou
2017-01-19 10:13       ` [PATCH v2 2/2] generic: test mmap dio through DAX and non-DAX Xiong Zhou

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=20170206191717.GB12290@linux.intel.com \
    --to=ross.zwisler@linux.intel.com \
    --cc=eguan@redhat.com \
    --cc=fstests@vger.kernel.org \
    --cc=jmoyer@redhat.com \
    --cc=linux-nvdimm@ml01.01.org \
    --cc=xzhou@redhat.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