From: Jaegeuk Kim <jaegeuk@kernel.org>
To: Yangtao Li <frank.li@vivo.com>
Cc: chao@kernel.org, linux-f2fs-devel@lists.sourceforge.net,
linux-kernel@vger.kernel.org, Yuwei.Guan@zeekrlife.com
Subject: Re: [PATCH v2] f2fs: introduce discard_io_aware_gran sysfs node
Date: Tue, 3 Jan 2023 18:25:28 -0800 [thread overview]
Message-ID: <Y7TjmBqDo9IgUaZw@google.com> (raw)
In-Reply-To: <20221212140540.5187-1-frank.li@vivo.com>
On 12/12, Yangtao Li wrote:
> The current discard_io_aware_gran is a fixed value, change it to be
> configurable through the sys node.
>
> Signed-off-by: Yangtao Li <frank.li@vivo.com>
> ---
> v2:
> - allow 0
> Documentation/ABI/testing/sysfs-fs-f2fs | 9 +++++++++
> fs/f2fs/f2fs.h | 3 +++
> fs/f2fs/segment.c | 3 ++-
> fs/f2fs/sysfs.c | 13 +++++++++++++
> 4 files changed, 27 insertions(+), 1 deletion(-)
>
> diff --git a/Documentation/ABI/testing/sysfs-fs-f2fs b/Documentation/ABI/testing/sysfs-fs-f2fs
> index 9e3756625a81..7b6cd4cf40ce 100644
> --- a/Documentation/ABI/testing/sysfs-fs-f2fs
> +++ b/Documentation/ABI/testing/sysfs-fs-f2fs
> @@ -669,3 +669,12 @@ Contact: "Ping Xiong" <xiongping1@xiaomi.com>
> Description: When DATA SEPARATION is on, it controls the age threshold to indicate
> the data blocks as warm. By default it was initialized as 2621440 blocks
> (equals to 10GB).
> +
> +What: /sys/fs/f2fs/<disk>/discard_io_aware_gran
> +Date: December 2022
> +Contact: "Yangtao Li" <frank.li@vivo.com>
> +Description: Controls background discard granularity of inner discard thread
> + when is not in idle. Inner thread will not issue discards with size that
> + is smaller than granularity. The unit size is one block(4KB), now only
> + support configuring in range of [0, 512].
> + Default: 512
> diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
> index e8953c3dc81a..bd1430d09c6d 100644
> --- a/fs/f2fs/f2fs.h
> +++ b/fs/f2fs/f2fs.h
> @@ -332,6 +332,8 @@ struct discard_entry {
> #define DEFAULT_DISCARD_GRANULARITY 16
> /* default maximum discard granularity of ordered discard, unit: block count */
> #define DEFAULT_MAX_ORDERED_DISCARD_GRANULARITY 16
> +/* default minimum granularity discard not be aware of I/O, unit: block count */
> +#define DEFAULT_IO_AWARE_DISCARD_GRANULARITY 512
It seems we don't need to add another dependency between this and MAX_PLIST_NUM.
Can we use one?
>
> /* max discard pend list number */
> #define MAX_PLIST_NUM 512
> @@ -410,6 +412,7 @@ struct discard_cmd_control {
> unsigned int min_discard_issue_time; /* min. interval between discard issue */
> unsigned int mid_discard_issue_time; /* mid. interval between discard issue */
> unsigned int max_discard_issue_time; /* max. interval between discard issue */
> + unsigned int discard_io_aware_gran; /* minimum discard granularity not be aware of I/O */
> unsigned int discard_urgent_util; /* utilization which issue discard proactively */
> unsigned int discard_granularity; /* discard granularity */
> unsigned int max_ordered_discard; /* maximum discard granularity issued by lba order */
> diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
> index a9099a754dd2..f4bf39ee31c6 100644
> --- a/fs/f2fs/segment.c
> +++ b/fs/f2fs/segment.c
> @@ -1060,7 +1060,7 @@ static void __init_discard_policy(struct f2fs_sb_info *sbi,
> dpolicy->granularity = granularity;
>
> dpolicy->max_requests = dcc->max_discard_request;
> - dpolicy->io_aware_gran = MAX_PLIST_NUM;
> + dpolicy->io_aware_gran = dcc->discard_io_aware_gran;
> dpolicy->timeout = false;
>
> if (discard_type == DPOLICY_BG) {
> @@ -2066,6 +2066,7 @@ static int create_discard_cmd_control(struct f2fs_sb_info *sbi)
> if (!dcc)
> return -ENOMEM;
>
> + dcc->discard_io_aware_gran = DEFAULT_IO_AWARE_DISCARD_GRANULARITY;
> dcc->discard_granularity = DEFAULT_DISCARD_GRANULARITY;
> dcc->max_ordered_discard = DEFAULT_MAX_ORDERED_DISCARD_GRANULARITY;
> if (F2FS_OPTION(sbi).discard_unit == DISCARD_UNIT_SEGMENT)
> diff --git a/fs/f2fs/sysfs.c b/fs/f2fs/sysfs.c
> index 83a366f3ee80..5ab42da5f2a3 100644
> --- a/fs/f2fs/sysfs.c
> +++ b/fs/f2fs/sysfs.c
> @@ -473,6 +473,17 @@ static ssize_t __sbi_store(struct f2fs_attr *a,
> return count;
> }
>
> + if (!strcmp(a->attr.name, "discard_io_aware_gran")) {
> + if (t > MAX_PLIST_NUM)
> + return -EINVAL;
> + if (!f2fs_block_unit_discard(sbi))
> + return -EINVAL;
> + if (t == *ui)
> + return count;
> + *ui = t;
> + return count;
> + }
> +
> if (!strcmp(a->attr.name, "discard_granularity")) {
> if (t == 0 || t > MAX_PLIST_NUM)
> return -EINVAL;
> @@ -825,6 +836,7 @@ F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, max_discard_request, max_discard_req
> F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, min_discard_issue_time, min_discard_issue_time);
> F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, mid_discard_issue_time, mid_discard_issue_time);
> F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, max_discard_issue_time, max_discard_issue_time);
> +F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, discard_io_aware_gran, discard_io_aware_gran);
> F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, discard_urgent_util, discard_urgent_util);
> F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, discard_granularity, discard_granularity);
> F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, max_ordered_discard, max_ordered_discard);
> @@ -960,6 +972,7 @@ static struct attribute *f2fs_attrs[] = {
> ATTR_LIST(min_discard_issue_time),
> ATTR_LIST(mid_discard_issue_time),
> ATTR_LIST(max_discard_issue_time),
> + ATTR_LIST(discard_io_aware_gran),
> ATTR_LIST(discard_urgent_util),
> ATTR_LIST(discard_granularity),
> ATTR_LIST(max_ordered_discard),
> --
> 2.25.1
prev parent reply other threads:[~2023-01-04 2:25 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-12-12 14:05 [PATCH v2] f2fs: introduce discard_io_aware_gran sysfs node Yangtao Li
2023-01-04 2:25 ` Jaegeuk Kim [this message]
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=Y7TjmBqDo9IgUaZw@google.com \
--to=jaegeuk@kernel.org \
--cc=Yuwei.Guan@zeekrlife.com \
--cc=chao@kernel.org \
--cc=frank.li@vivo.com \
--cc=linux-f2fs-devel@lists.sourceforge.net \
--cc=linux-kernel@vger.kernel.org \
/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