From: Yang Yang <yang.yang@vivo.com>
To: Alasdair Kergon <agk@redhat.com>,
Mike Snitzer <snitzer@kernel.org>,
Mikulas Patocka <mpatocka@redhat.com>,
dm-devel@lists.linux.dev, linux-kernel@vger.kernel.org
Cc: Yang Yang <yang.yang@vivo.com>
Subject: [PATCH 4/5] dm: Avoid sending redundant empty flush bios to the same block device
Date: Tue, 14 May 2024 17:04:43 +0800 [thread overview]
Message-ID: <20240514090445.2847-5-yang.yang@vivo.com> (raw)
In-Reply-To: <20240514090445.2847-1-yang.yang@vivo.com>
If the num_targets is greater than the number of dm_devs in the
dm_table's devices list, __send_empty_flush() might invoke
__send_duplicate_bios() multiple times for the same block device.
This could lead to a substantial decrease in performance when
num_targets significantly exceeds the number of dm_devs.
This patch ensure that __send_duplicate_bios() is only called once
for each dm_dev with different target type.
Signed-off-by: Yang Yang <yang.yang@vivo.com>
---
drivers/md/dm.c | 18 +++++++++++++++---
1 file changed, 15 insertions(+), 3 deletions(-)
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 25215b93c3cf..9dbddc214084 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -1561,6 +1561,7 @@ static void __send_empty_flush(struct clone_info *ci)
{
struct dm_table *t = ci->map;
struct bio flush_bio;
+ struct dm_target *ti;
/*
* Use an on-stack bio for this, it's safe since we don't
@@ -1574,10 +1575,21 @@ static void __send_empty_flush(struct clone_info *ci)
ci->sector_count = 0;
ci->io->tio.clone.bi_iter.bi_size = 0;
- for (unsigned int i = 0; i < t->num_targets; i++) {
- struct dm_target *ti = dm_table_get_target(t, i);
+ if (!t->flush_pass_around) {
+ for (unsigned int i = 0; i < t->num_targets; i++) {
+ ti = dm_table_get_target(t, i);
+ __send_empty_flush_bios(t, ti, ci);
+ }
+ } else {
+ struct list_head *devices = dm_table_get_devices(t);
+ struct dm_dev_internal *dd;
- __send_empty_flush_bios(t, ti, ci);
+ list_for_each_entry(dd, devices, list) {
+ struct list_head *targets = &dd->dm_dev->targets;
+
+ list_for_each_entry(ti, targets, list)
+ __send_empty_flush_bios(t, ti, ci);
+ }
}
/*
--
2.34.1
next prev parent reply other threads:[~2024-05-14 9:06 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-05-14 9:04 [PATCH 0/5] dm: empty flush optimization Yang Yang
2024-05-14 9:04 ` [PATCH 1/5] dm: introduce flush_pass_around flag Yang Yang
2024-05-14 9:04 ` [PATCH 2/5] dm: add __send_empty_flush_bios() helper Yang Yang
2024-05-14 9:04 ` [PATCH 3/5] dm: support retrieving struct dm_target from struct dm_dev Yang Yang
2024-05-15 15:42 ` Benjamin Marzinski
2024-05-15 15:53 ` Mikulas Patocka
2024-05-16 20:49 ` [PATCH] dm: optimize flushes Mikulas Patocka
2024-05-17 8:55 ` YangYang
2024-05-22 16:10 ` Mike Snitzer
2024-05-22 21:05 ` Mikulas Patocka
2024-05-23 17:46 ` [PATCH v2] " Mikulas Patocka
2024-05-23 18:02 ` Mike Snitzer
2024-05-28 11:37 ` Mikulas Patocka
2024-05-23 22:17 ` Eric Wheeler
2024-05-15 16:00 ` [PATCH 3/5] dm: support retrieving struct dm_target from struct dm_dev Benjamin Marzinski
2024-05-16 2:12 ` YangYang
2024-05-16 16:39 ` Benjamin Marzinski
2024-05-16 1:55 ` YangYang
2024-05-16 15:29 ` Benjamin Marzinski
2024-05-17 7:48 ` YangYang
2024-05-17 14:33 ` Benjamin Marzinski
2024-05-20 3:12 ` YangYang
2024-05-14 9:04 ` Yang Yang [this message]
2024-05-14 9:04 ` [PATCH 5/5] dm linear: enable flush optimization function Yang Yang
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=20240514090445.2847-5-yang.yang@vivo.com \
--to=yang.yang@vivo.com \
--cc=agk@redhat.com \
--cc=dm-devel@lists.linux.dev \
--cc=linux-kernel@vger.kernel.org \
--cc=mpatocka@redhat.com \
--cc=snitzer@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
all inboxes | Powered by JetHome®