From: Chad Talbott <ctalbott@google.com>
To: vgoyal@redhat.com, jens.axboe@oracle.com
Cc: mrubin@google.com, guijianfeng@cn.fujitsu.com,
Li Zefan <lizf@cn.fujitsu.com>,
linux-kernel@vger.kernel.org, dpshah@google.com,
Nauman Rafique <nauman@google.com>
Subject: [PATCH 3/4] io-controller: Add a new interface "weight_device" for IO-Controller
Date: Thu, 25 Mar 2010 11:04:33 -0700 [thread overview]
Message-ID: <20100325180433.25299.44167.stgit@meat.mtv.corp.google.com> (raw)
In-Reply-To: <20100325180310.25299.64877.stgit@meat.mtv.corp.google.com>
This is Gui Jianfeng's original patch.
---
block/blk-cgroup.c | 240 +++++++++++++++++++++++++++++++++++++++++++++++++++
block/blk-cgroup.h | 10 ++
block/cfq-iosched.c | 2
3 files changed, 251 insertions(+), 1 deletions(-)
diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c
index 809451f..bf77b99 100644
--- a/block/blk-cgroup.c
+++ b/block/blk-cgroup.c
@@ -24,6 +24,13 @@ static LIST_HEAD(blkio_list);
struct blkio_cgroup blkio_root_cgroup = { .weight = 2*BLKIO_WEIGHT_DEFAULT };
EXPORT_SYMBOL_GPL(blkio_root_cgroup);
+static inline void policy_insert_node(struct blkio_cgroup *blkcg,
+ struct blkio_policy_node *pn);
+static inline void policy_delete_node(struct blkio_policy_node *pn);
+static struct blkio_policy_node *policy_search_node(const struct blkio_cgroup *blkcg,
+ dev_t dev);
+
+
static struct cgroup_subsys_state *blkiocg_create(struct cgroup_subsys *,
struct cgroup *);
static int blkiocg_can_attach(struct cgroup_subsys *, struct cgroup *,
@@ -154,6 +161,7 @@ blkiocg_weight_write(struct cgroup *cgroup, struct cftype *cftype, u64 val)
struct blkio_group *blkg;
struct hlist_node *n;
struct blkio_policy_type *blkiop;
+ struct blkio_policy_node *pn;
if (val < BLKIO_WEIGHT_MIN || val > BLKIO_WEIGHT_MAX)
return -EINVAL;
@@ -162,7 +170,13 @@ blkiocg_weight_write(struct cgroup *cgroup, struct cftype *cftype, u64 val)
spin_lock(&blkio_list_lock);
spin_lock_irq(&blkcg->lock);
blkcg->weight = (unsigned int)val;
+
hlist_for_each_entry(blkg, n, &blkcg->blkg_list, blkcg_node) {
+ pn = policy_search_node(blkcg, blkg->dev);
+
+ if (pn)
+ continue;
+
list_for_each_entry(blkiop, &blkio_list, list)
blkiop->ops.blkio_update_group_weight_fn(blkg,
blkcg->weight);
@@ -213,8 +227,227 @@ void blkiocg_update_blkio_group_dequeue_stats(struct blkio_group *blkg,
EXPORT_SYMBOL_GPL(blkiocg_update_blkio_group_dequeue_stats);
#endif
+static int check_dev_num(dev_t dev)
+{
+ int part = 0;
+ struct gendisk *disk;
+
+ disk = get_gendisk(dev, &part);
+ if (!disk || part)
+ return -ENODEV;
+
+ return 0;
+}
+
+static int policy_parse_and_set(char *buf, struct blkio_policy_node *newpn)
+{
+ char *s[4], *p, *major_s = NULL, *minor_s = NULL;
+ int ret;
+ unsigned long major, minor, temp;
+ int i = 0;
+ dev_t dev;
+
+ memset(s, 0, sizeof(s));
+
+ while ((p = strsep(&buf, " ")) != NULL) {
+ if (!*p)
+ continue;
+
+ s[i++] = p;
+
+ /* Prevent from inputing too many things */
+ if (i == 3)
+ break;
+ }
+
+ if (i != 2)
+ return -EINVAL;
+
+ p = strsep(&s[0], ":");
+ if (p != NULL)
+ major_s = p;
+ else
+ return -EINVAL;
+
+ minor_s = s[0];
+ if (!minor_s)
+ return -EINVAL;
+
+ ret = strict_strtoul(major_s, 10, &major);
+ if (ret)
+ return -EINVAL;
+
+ ret = strict_strtoul(minor_s, 10, &minor);
+ if (ret)
+ return -EINVAL;
+
+ dev = MKDEV(major, minor);
+
+ ret = check_dev_num(dev);
+ if (ret)
+ return ret;
+
+ newpn->dev = dev;
+
+ if (s[1] == NULL)
+ return -EINVAL;
+
+ ret = strict_strtoul(s[1], 10, &temp);
+ if (ret || (temp < BLKIO_WEIGHT_MIN && temp > 0) ||
+ temp > BLKIO_WEIGHT_MAX)
+ return -EINVAL;
+
+ newpn->weight = temp;
+
+ return 0;
+}
+
+unsigned int blkcg_get_weight(struct blkio_cgroup *blkcg,
+ dev_t dev)
+{
+ struct blkio_policy_node *pn;
+
+ pn = policy_search_node(blkcg, dev);
+ if (pn)
+ return pn->weight;
+ else
+ return blkcg->weight;
+}
+EXPORT_SYMBOL_GPL(blkcg_get_weight);
+
+static inline void policy_insert_node(struct blkio_cgroup *blkcg,
+ struct blkio_policy_node *pn)
+{
+ list_add(&pn->node, &blkcg->policy_list);
+}
+
+/* Must be called with blkcg->lock held */
+static inline void policy_delete_node(struct blkio_policy_node *pn)
+{
+ list_del(&pn->node);
+}
+
+/* Must be called with blkcg->lock held */
+static struct blkio_policy_node *policy_search_node(const struct blkio_cgroup *blkcg,
+ dev_t dev)
+{
+ struct blkio_policy_node *pn;
+
+ list_for_each_entry(pn, &blkcg->policy_list, node) {
+ if (pn->dev == dev)
+ return pn;
+ }
+
+ return NULL;
+}
+
+
+static int blkiocg_weight_device_write(struct cgroup *cgrp, struct cftype *cft,
+ const char *buffer)
+{
+ int ret = 0;
+ char *buf;
+ struct blkio_policy_node *newpn, *pn;
+ struct blkio_cgroup *blkcg;
+ struct blkio_group *blkg;
+ int keep_newpn = 0;
+ struct hlist_node *n;
+ struct blkio_policy_type *blkiop;
+
+ buf = kstrdup(buffer, GFP_KERNEL);
+ if (!buf)
+ return -ENOMEM;
+
+ newpn = kzalloc(sizeof(*newpn), GFP_KERNEL);
+ if (!newpn) {
+ ret = -ENOMEM;
+ goto free_buf;
+ }
+
+ ret = policy_parse_and_set(buf, newpn);
+ if (ret)
+ goto free_newpn;
+
+ blkcg = cgroup_to_blkio_cgroup(cgrp);
+
+ spin_lock_irq(&blkcg->lock);
+
+ pn = policy_search_node(blkcg, newpn->dev);
+ if (!pn) {
+ if (newpn->weight != 0) {
+ policy_insert_node(blkcg, newpn);
+ keep_newpn = 1;
+ }
+ spin_unlock_irq(&blkcg->lock);
+ goto update_io_group;
+ }
+
+ if (newpn->weight == 0) {
+ /* weight == 0 means deleteing a specific weight */
+ policy_delete_node(pn);
+ spin_unlock_irq(&blkcg->lock);
+ goto update_io_group;
+ }
+ spin_unlock_irq(&blkcg->lock);
+
+ pn->weight = newpn->weight;
+
+update_io_group:
+ /* update weight for each cfqg */
+ spin_lock(&blkio_list_lock);
+ spin_lock_irq(&blkcg->lock);
+
+ hlist_for_each_entry(blkg, n, &blkcg->blkg_list, blkcg_node) {
+ if (newpn->dev == blkg->dev) {
+ list_for_each_entry(blkiop, &blkio_list, list)
+ blkiop->ops.blkio_update_group_weight_fn(blkg,
+ newpn->weight ?
+ newpn->weight :
+ blkcg->weight);
+ }
+ }
+
+ spin_unlock_irq(&blkcg->lock);
+ spin_unlock(&blkio_list_lock);
+
+free_newpn:
+ if (!keep_newpn)
+ kfree(newpn);
+free_buf:
+ kfree(buf);
+ return ret;
+}
+
+static int blkiocg_weight_device_read(struct cgroup *cgrp, struct cftype *cft,
+ struct seq_file *m)
+{
+ struct blkio_cgroup *blkcg;
+ struct blkio_policy_node *pn;
+
+ seq_printf(m, "dev\tweight\n");
+
+ blkcg = cgroup_to_blkio_cgroup(cgrp);
+ if (list_empty(&blkcg->policy_list))
+ goto out;
+
+ spin_lock_irq(&blkcg->lock);
+ list_for_each_entry(pn, &blkcg->policy_list, node) {
+ seq_printf(m, "%u:%u\t%u\n", MAJOR(pn->dev),
+ MINOR(pn->dev), pn->weight);
+ }
+ spin_unlock_irq(&blkcg->lock);
+out:
+ return 0;
+}
+
struct cftype blkio_files[] = {
{
+ .name = "weight_device",
+ .read_seq_string = blkiocg_weight_device_read,
+ .write_string = blkiocg_weight_device_write,
+ .max_write_len = 256,
+ },
+ {
.name = "weight",
.read_u64 = blkiocg_weight_read,
.write_u64 = blkiocg_weight_write,
@@ -247,6 +480,7 @@ static void blkiocg_destroy(struct cgroup_subsys *subsys, struct cgroup *cgroup)
unsigned long flags;
struct blkio_group *blkg;
struct blkio_policy_type *blkiop;
+ struct blkio_policy_node *pn, *pntmp;
rcu_read_lock();
remove_entry:
@@ -276,7 +510,12 @@ remove_entry:
blkiop->ops.blkio_unlink_group_fn(blkg);
spin_unlock(&blkio_list_lock);
goto remove_entry;
+
done:
+ list_for_each_entry_safe(pn, pntmp, &blkcg->policy_list, node) {
+ policy_delete_node(pn);
+ kfree(pn);
+ }
free_css_id(&blkio_subsys, &blkcg->css);
rcu_read_unlock();
if (blkcg != &blkio_root_cgroup)
@@ -307,6 +546,7 @@ done:
spin_lock_init(&blkcg->lock);
INIT_HLIST_HEAD(&blkcg->blkg_list);
+ INIT_LIST_HEAD(&blkcg->policy_list);
return &blkcg->css;
}
diff --git a/block/blk-cgroup.h b/block/blk-cgroup.h
index 9e70c03..92296b8 100644
--- a/block/blk-cgroup.h
+++ b/block/blk-cgroup.h
@@ -29,6 +29,7 @@ struct blkio_cgroup {
unsigned int weight;
spinlock_t lock;
struct hlist_head blkg_list;
+ struct list_head policy_list; /* list of blkio_policy_node */
};
struct blkio_group {
@@ -50,6 +51,15 @@ struct blkio_group {
unsigned long sectors;
};
+struct blkio_policy_node {
+ struct list_head node;
+ dev_t dev;
+ unsigned int weight;
+};
+
+extern unsigned int blkcg_get_weight(struct blkio_cgroup *blkcg,
+ dev_t dev);
+
typedef void (blkio_unlink_group_fn) (struct blkio_group *blkg);
typedef void (blkio_update_group_weight_fn) (struct blkio_group *blkg,
unsigned int weight);
diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index 864b39a..d4d57a3 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -954,7 +954,6 @@ cfq_find_alloc_cfqg(struct cfq_data *cfqd, struct cgroup *cgroup, int create)
if (!cfqg)
goto done;
- cfqg->weight = blkcg->weight;
for_each_cfqg_st(cfqg, i, j, st)
*st = CFQ_RB_ROOT;
RB_CLEAR_NODE(&cfqg->rb_node);
@@ -971,6 +970,7 @@ cfq_find_alloc_cfqg(struct cfq_data *cfqd, struct cgroup *cgroup, int create)
sscanf(dev_name(bdi->dev), "%u:%u", &major, &minor);
blkiocg_add_blkio_group(blkcg, &cfqg->blkg, cfqd->queue,
MKDEV(major, minor));
+ cfqg->weight = blkcg_get_weight(blkcg, cfqg->blkg.dev);
/* Add group on cfqd list */
hlist_add_head(&cfqg->cfqd_node, &cfqd->cfqg_list);
next prev parent reply other threads:[~2010-03-25 18:05 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-03-25 18:04 [PATCH 0/4] io-controller: Use names rather than major:minor Chad Talbott
2010-03-25 18:04 ` [PATCH 1/4] blkio_group key change: void * -> request_queue * Chad Talbott
2010-03-25 23:25 ` Vivek Goyal
2010-03-26 0:17 ` Chad Talbott
2010-03-25 18:04 ` [PATCH 2/4] Adds an RCU-protected pointer to request_queue that makes it easy to Chad Talbott
2010-03-25 23:02 ` Vivek Goyal
2010-03-25 18:04 ` Chad Talbott [this message]
2010-03-25 18:04 ` [PATCH 4/4] Use disk-names to set blkio.weight_device policy Chad Talbott
2010-03-26 1:31 ` [PATCH 0/4] io-controller: Use names rather than major:minor Gui Jianfeng
2010-03-26 15:20 ` Vivek Goyal
2010-03-26 22:54 ` Chad Talbott
2010-03-26 23:21 ` Divyesh Shah
2010-03-27 0:28 ` Vivek Goyal
2010-03-27 0:20 ` Vivek Goyal
2010-03-27 0:24 ` Vivek Goyal
2010-03-27 0:30 ` Vivek Goyal
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=20100325180433.25299.44167.stgit@meat.mtv.corp.google.com \
--to=ctalbott@google.com \
--cc=dpshah@google.com \
--cc=guijianfeng@cn.fujitsu.com \
--cc=jens.axboe@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=lizf@cn.fujitsu.com \
--cc=mrubin@google.com \
--cc=nauman@google.com \
--cc=vgoyal@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.