From: David Sterba <dsterba@suse.cz>
To: Naohiro Aota <naohiro.aota@wdc.com>
Cc: linux-btrfs@vger.kernel.org, wangyugui@e16-tech.com,
clm@meta.com, hch@lst.de
Subject: Re: [PATCH v2] btrfs: introduce sync_csum_mode to tweak sync checksum behavior
Date: Wed, 31 Jan 2024 20:04:59 +0100 [thread overview]
Message-ID: <20240131190459.GS31555@twin.jikos.cz> (raw)
In-Reply-To: <75b81282919c566735f80f71c57343e282c40bed.1706685025.git.naohiro.aota@wdc.com>
On Wed, Jan 31, 2024 at 04:13:45PM +0900, Naohiro Aota wrote:
> We disable offloading checksum to workqueues and do it synchronously when
> the checksum algorithm is fast. However, as reported in the link below,
> RAID0 with multiple devices may suffer from the sync checksum, because
> "fast checksum" is still not fast enough to catch up RAID0 writing.
>
> To measure the effectiveness of sync checksum for developers, it would be
> better to have a switch for the sync checksum under CONFIG_BTRFS_DEBUG
> hood.
>
> This commit introduces fs_devices->sync_csum_mode for CONFIG_BTRFS_DEBUG,
Please rename it to offload_checksums, this also inverts the logic but
is IMHO clear what it does.
> so that a btrfs developer can change the behavior by writing to
> /sys/fs/btrfs/<uuid>/sync_csum. The default is "auto" which is the same as
> the previous behavior. Or, you can set "on" or "off" to always/never use
> sync checksum.
>
> More benchmark should be collected with this knob to implement a proper
> criteria to enable/disable sync checksum.
>
> Link: https://lore.kernel.org/linux-btrfs/20230731152223.4EFB.409509F4@e16-tech.com/
> Link: https://lore.kernel.org/linux-btrfs/p3vo3g7pqn664mhmdhlotu5dzcna6vjtcoc2hb2lsgo2fwct7k@xzaxclba5tae/
> Signed-off-by: Naohiro Aota <naohiro.aota@wdc.com>
> ---
> v2:
> - Call it "sync checksum" properly
> - Removed a patch to automatically change checksum behavior
> - Hide the sysfs interface under CONFIG_BTRFS_DEBUG
> ---
> fs/btrfs/bio.c | 13 ++++++++++++-
> fs/btrfs/sysfs.c | 43 +++++++++++++++++++++++++++++++++++++++++++
> fs/btrfs/volumes.h | 23 +++++++++++++++++++++++
> 3 files changed, 78 insertions(+), 1 deletion(-)
>
> diff --git a/fs/btrfs/bio.c b/fs/btrfs/bio.c
> index 960b81718e29..c896d3cd792b 100644
> --- a/fs/btrfs/bio.c
> +++ b/fs/btrfs/bio.c
> @@ -608,8 +608,19 @@ static void run_one_async_done(struct btrfs_work *work, bool do_free)
>
> static bool should_async_write(struct btrfs_bio *bbio)
> {
> + bool auto_csum_mode = true;
> +
> +#ifdef CONFIG_BTRFS_DEBUG
> + struct btrfs_fs_devices *fs_devices = bbio->fs_info->fs_devices;
> +
> + if (fs_devices->sync_csum_mode == BTRFS_SYNC_CSUM_FORCE_ON)
> + return false;
> +
> + auto_csum_mode = fs_devices->sync_csum_mode == BTRFS_SYNC_CSUM_AUTO;
> +#endif
> +
> /* Submit synchronously if the checksum implementation is fast. */
> - if (test_bit(BTRFS_FS_CSUM_IMPL_FAST, &bbio->fs_info->flags))
> + if (auto_csum_mode && test_bit(BTRFS_FS_CSUM_IMPL_FAST, &bbio->fs_info->flags))
> return false;
>
> /*
> diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c
> index 84c05246ffd8..ea1e54149ef4 100644
> --- a/fs/btrfs/sysfs.c
> +++ b/fs/btrfs/sysfs.c
> @@ -1306,6 +1306,46 @@ static ssize_t btrfs_bg_reclaim_threshold_store(struct kobject *kobj,
> BTRFS_ATTR_RW(, bg_reclaim_threshold, btrfs_bg_reclaim_threshold_show,
> btrfs_bg_reclaim_threshold_store);
>
> +#ifdef CONFIG_BTRFS_DEBUG
> +static ssize_t btrfs_sync_csum_show(struct kobject *kobj,
> + struct kobj_attribute *a, char *buf)
> +{
> + struct btrfs_fs_devices *fs_devices = to_fs_devs(kobj);
> +
> + switch (fs_devices->sync_csum_mode) {
> + case BTRFS_SYNC_CSUM_AUTO:
> + return sysfs_emit(buf, "auto\n");
> + case BTRFS_SYNC_CSUM_FORCE_ON:
> + return sysfs_emit(buf, "on\n");
> + case BTRFS_SYNC_CSUM_FORCE_OFF:
> + return sysfs_emit(buf, "off\n");
We're using numeric indicators for on/off in other sysfs files, though
here it's a bit more readable.
> + default:
> + WARN_ON(1);
> + return -EINVAL;
> + }
> +}
> +
> +static ssize_t btrfs_sync_csum_store(struct kobject *kobj,
> + struct kobj_attribute *a, const char *buf,
> + size_t len)
> +{
> + struct btrfs_fs_devices *fs_devices = to_fs_devs(kobj);
> +
> + if (sysfs_streq(buf, "auto"))
Please use kstrobool, it accepts awide range of "yes/no" values and
check for "auto" only after it returns -EINVAL.
> + fs_devices->sync_csum_mode = BTRFS_SYNC_CSUM_AUTO;
> + else if (sysfs_streq(buf, "on"))
> + fs_devices->sync_csum_mode = BTRFS_SYNC_CSUM_FORCE_ON;
> + else if (sysfs_streq(buf, "off"))
> + fs_devices->sync_csum_mode = BTRFS_SYNC_CSUM_FORCE_OFF;
> + else
> + return -EINVAL;
> +
> + return len;
> + return -EINVAL;
> +}
next prev parent reply other threads:[~2024-01-31 19:05 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-01-31 7:13 [PATCH v2] btrfs: introduce sync_csum_mode to tweak sync checksum behavior Naohiro Aota
2024-01-31 14:15 ` Johannes Thumshirn
2024-01-31 18:58 ` David Sterba
2024-02-01 1:16 ` Naohiro Aota
2024-02-01 2:11 ` David Sterba
2024-01-31 19:04 ` David Sterba [this message]
2024-02-01 1:28 ` Naohiro Aota
2024-02-01 2:14 ` David Sterba
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=20240131190459.GS31555@twin.jikos.cz \
--to=dsterba@suse.cz \
--cc=clm@meta.com \
--cc=hch@lst.de \
--cc=linux-btrfs@vger.kernel.org \
--cc=naohiro.aota@wdc.com \
--cc=wangyugui@e16-tech.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