From: Anand Jain <anand.jain@oracle.com>
To: linux-btrfs@vger.kernel.org
Cc: josef@toxicpanda.com, dsterba@suse.com
Subject: [PATCH v5 2/5] btrfs: create read policy framework
Date: Mon, 17 Feb 2020 19:12:42 +0800 [thread overview]
Message-ID: <1581937965-16569-3-git-send-email-anand.jain@oracle.com> (raw)
In-Reply-To: <1581937965-16569-1-git-send-email-anand.jain@oracle.com>
As of now we use %pid method to read stripped mirrored data, which means
process id determines the stripe id to be read. This type of routing
typically helps in a system with many small independent processes tying
to read random data. On the other hand the %pid based read IO policy is
inefficient because if there is a single process trying to read large
data the overall disk bandwidth remains under-utilized.
So this patch introduces read policy framework so that we could add more
read policies, such as IO routing based on device's wait-queue or manual
when we have a read-preferred device or a policy based on the target
storage caching.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
---
v5: Title renamed from:- btrfs: add read_policy framework
Change log updated.
Unnecessary comment dropped, added more where necessary.
Optimize code in the switch remove duplicate code.
Define BTRFS_READ_POLICY_DEFAULT dropped.
Rename enum btrfs_read_policy_type to enum btrfs_read_policy.
Rename BTRFS_READ_BY_PID to BTRFS_READ_POLICY_PID.
v4: -
v3: Declare fs_devices::readmirror as enum btrfs_readmirror_policy_type
v2: Declare fs_devices::readmirror as u8 instead of atomic_t
A small change in comment and change log wordings.
fs/btrfs/volumes.c | 13 ++++++++++++-
fs/btrfs/volumes.h | 14 ++++++++++++++
2 files changed, 26 insertions(+), 1 deletion(-)
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 387f80656476..b6efb87bb0ae 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -1209,6 +1209,7 @@ static int open_fs_devices(struct btrfs_fs_devices *fs_devices,
fs_devices->opened = 1;
fs_devices->latest_bdev = latest_dev->bdev;
fs_devices->total_rw_bytes = 0;
+ fs_devices->read_policy = BTRFS_READ_POLICY_PID;
out:
return ret;
}
@@ -5358,7 +5359,17 @@ static int find_live_mirror(struct btrfs_fs_info *fs_info,
else
num_stripes = map->num_stripes;
- preferred_mirror = first + current->pid % num_stripes;
+ switch (fs_info->fs_devices->read_policy) {
+ default:
+ /*
+ * Shouldn't happen, just warn and use pid instead of failing.
+ */
+ btrfs_warn_rl(fs_info,
+ "unknown read_policy type %u, fallback to pid",
+ fs_info->fs_devices->read_policy);
+ case BTRFS_READ_POLICY_PID:
+ preferred_mirror = first + current->pid % num_stripes;
+ }
if (dev_replace_is_ongoing &&
fs_info->dev_replace.cont_reading_from_srcdev_mode ==
diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
index f01552a0785e..ed2bba741b6e 100644
--- a/fs/btrfs/volumes.h
+++ b/fs/btrfs/volumes.h
@@ -209,6 +209,15 @@ struct btrfs_device {
BTRFS_DEVICE_GETSET_FUNCS(disk_total_bytes);
BTRFS_DEVICE_GETSET_FUNCS(bytes_used);
+/*
+ * Read policies for the mirrored block groups, read picks the stripe based
+ * on these policies.
+ */
+enum btrfs_read_policy {
+ BTRFS_READ_POLICY_PID,
+ BTRFS_NR_READ_POLICY,
+};
+
struct btrfs_fs_devices {
u8 fsid[BTRFS_FSID_SIZE]; /* FS specific uuid */
u8 metadata_uuid[BTRFS_FSID_SIZE];
@@ -260,6 +269,11 @@ struct btrfs_fs_devices {
struct kobject *devices_kobj;
struct kobject *devinfo_kobj;
struct completion kobj_unregister;
+
+ /*
+ * policy used to read the mirrored stripes
+ */
+ enum btrfs_read_policy read_policy;
};
#define BTRFS_BIO_INLINE_CSUM_SIZE 64
--
1.8.3.1
next prev parent reply other threads:[~2020-02-17 11:13 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-17 11:12 [PATCH v5 0/5] readmirror feature (sysfs and in-memory only approach; with new read_policy device) Anand Jain
2020-02-17 11:12 ` [PATCH v5 1/5] btrfs: add btrfs_strmatch helper Anand Jain
2020-02-17 11:12 ` Anand Jain [this message]
2020-02-19 7:41 ` [PATCH v5 2/5] btrfs: create read policy framework kbuild test robot
2020-02-19 11:40 ` Anand Jain
2020-02-19 12:35 ` David Sterba
2020-02-17 11:12 ` [PATCH v5 3/5] btrfs: create read policy sysfs attribute, pid Anand Jain
2020-02-17 11:12 ` [PATCH v5 4/5] btrfs: introduce new device-state read_preferred Anand Jain
2020-02-17 11:12 ` [PATCH v5 5/5] btrfs: introduce new read_policy device Anand Jain
2020-02-18 16:35 ` [PATCH v5 0/5] readmirror feature (sysfs and in-memory only approach; with new read_policy device) David Sterba
2020-02-19 11:42 ` Anand Jain
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=1581937965-16569-3-git-send-email-anand.jain@oracle.com \
--to=anand.jain@oracle.com \
--cc=dsterba@suse.com \
--cc=josef@toxicpanda.com \
--cc=linux-btrfs@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;
as well as URLs for NNTP newsgroup(s).