From: Tejun Heo <tj@kernel.org>
To: laijs@cn.fujitsu.com
Cc: axboe@kernel.dk, jack@suse.cz, fengguang.wu@intel.com,
jmoyer@redhat.com, zab@redhat.com, linux-kernel@vger.kernel.org,
herbert@gondor.hengli.com.au, davem@davemloft.net,
linux-crypto@vger.kernel.org, Tejun Heo <tj@kernel.org>
Subject: [PATCH 04/10] workqueue: add workqueue->unbound_attrs
Date: Tue, 19 Mar 2013 17:00:23 -0700 [thread overview]
Message-ID: <1363737629-16745-5-git-send-email-tj@kernel.org> (raw)
In-Reply-To: <1363737629-16745-1-git-send-email-tj@kernel.org>
Currently, when exposing attrs of an unbound workqueue via sysfs, the
workqueue_attrs of first_pwq() is used as that should equal the
current state of the workqueue.
The planned NUMA affinity support will make unbound workqueues make
use of multiple pool_workqueues for different NUMA nodes and the above
assumption will no longer hold. Introduce workqueue->unbound_attrs
which records the current attrs in effect and use it for sysfs instead
of first_pwq()->attrs.
Signed-off-by: Tejun Heo <tj@kernel.org>
---
kernel/workqueue.c | 36 ++++++++++++++++++++++++------------
1 file changed, 24 insertions(+), 12 deletions(-)
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 2768ed2..57e6139 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -247,6 +247,8 @@ struct workqueue_struct {
int nr_drainers; /* WQ: drain in progress */
int saved_max_active; /* PW: saved pwq max_active */
+ struct workqueue_attrs *unbound_attrs; /* WQ: only for unbound wqs */
+
#ifdef CONFIG_SYSFS
struct wq_device *wq_dev; /* I: for sysfs interface */
#endif
@@ -3099,10 +3101,9 @@ static ssize_t wq_nice_show(struct device *dev, struct device_attribute *attr,
struct workqueue_struct *wq = dev_to_wq(dev);
int written;
- rcu_read_lock_sched();
- written = scnprintf(buf, PAGE_SIZE, "%d\n",
- first_pwq(wq)->pool->attrs->nice);
- rcu_read_unlock_sched();
+ mutex_lock(&wq_mutex);
+ written = scnprintf(buf, PAGE_SIZE, "%d\n", wq->unbound_attrs->nice);
+ mutex_unlock(&wq_mutex);
return written;
}
@@ -3116,9 +3117,9 @@ static struct workqueue_attrs *wq_sysfs_prep_attrs(struct workqueue_struct *wq)
if (!attrs)
return NULL;
- rcu_read_lock_sched();
- copy_workqueue_attrs(attrs, first_pwq(wq)->pool->attrs);
- rcu_read_unlock_sched();
+ mutex_lock(&wq_mutex);
+ copy_workqueue_attrs(attrs, wq->unbound_attrs);
+ mutex_unlock(&wq_mutex);
return attrs;
}
@@ -3149,10 +3150,9 @@ static ssize_t wq_cpumask_show(struct device *dev,
struct workqueue_struct *wq = dev_to_wq(dev);
int written;
- rcu_read_lock_sched();
- written = cpumask_scnprintf(buf, PAGE_SIZE,
- first_pwq(wq)->pool->attrs->cpumask);
- rcu_read_unlock_sched();
+ mutex_lock(&wq_mutex);
+ written = cpumask_scnprintf(buf, PAGE_SIZE, wq->unbound_attrs->cpumask);
+ mutex_unlock(&wq_mutex);
written += scnprintf(buf + written, PAGE_SIZE - written, "\n");
return written;
@@ -3585,8 +3585,10 @@ static void pwq_unbound_release_workfn(struct work_struct *work)
* If we're the last pwq going away, @wq is already dead and no one
* is gonna access it anymore. Free it.
*/
- if (list_empty(&wq->pwqs))
+ if (list_empty(&wq->pwqs)) {
+ free_workqueue_attrs(wq->unbound_attrs);
kfree(wq);
+ }
}
/**
@@ -3656,6 +3658,9 @@ static void init_and_link_pwq(struct pool_workqueue *pwq,
/* link in @pwq */
list_add_rcu(&pwq->pwqs_node, &wq->pwqs);
+ if (wq->flags & WQ_UNBOUND)
+ copy_workqueue_attrs(wq->unbound_attrs, pool->attrs);
+
spin_unlock_irq(&pwq_lock);
mutex_unlock(&wq->flush_mutex);
}
@@ -3764,6 +3769,12 @@ struct workqueue_struct *__alloc_workqueue_key(const char *fmt,
if (!wq)
return NULL;
+ if (flags & WQ_UNBOUND) {
+ wq->unbound_attrs = alloc_workqueue_attrs(GFP_KERNEL);
+ if (!wq->unbound_attrs)
+ goto err_free_wq;
+ }
+
vsnprintf(wq->name, namelen, fmt, args1);
va_end(args);
va_end(args1);
@@ -3832,6 +3843,7 @@ struct workqueue_struct *__alloc_workqueue_key(const char *fmt,
return wq;
err_free_wq:
+ free_workqueue_attrs(wq->unbound_attrs);
kfree(wq);
return NULL;
err_destroy:
--
1.8.1.4
next prev parent reply other threads:[~2013-03-20 0:00 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-03-20 0:00 [PATCHSET wq/for-3.10] workqueue: NUMA affinity for unbound workqueues Tejun Heo
2013-03-20 0:00 ` [PATCH 01/10] workqueue: add wq_numa_tbl_len and wq_numa_possible_cpumask[] Tejun Heo
2013-03-20 14:08 ` JoonSoo Kim
2013-03-20 14:48 ` Tejun Heo
2013-03-20 15:43 ` Lai Jiangshan
2013-03-20 15:48 ` Tejun Heo
2013-03-20 16:43 ` Lai Jiangshan
2013-03-20 0:00 ` [PATCH 02/10] workqueue: drop 'H' from kworker names of unbound worker pools Tejun Heo
2013-03-20 0:00 ` [PATCH 03/10] workqueue: determine NUMA node of workers accourding to the allowed cpumask Tejun Heo
2013-03-20 0:00 ` Tejun Heo [this message]
2013-03-20 0:00 ` [PATCH 05/10] workqueue: make workqueue->name[] fixed len Tejun Heo
2013-03-20 0:00 ` [PATCH 06/10] workqueue: move hot fields of workqueue_struct to the end Tejun Heo
2013-03-20 0:00 ` [PATCH 07/10] workqueue: map an unbound workqueues to multiple per-node pool_workqueues Tejun Heo
2013-03-20 0:00 ` [PATCH 08/10] workqueue: break init_and_link_pwq() into two functions and introduce alloc_unbound_pwq() Tejun Heo
2013-03-20 15:52 ` Lai Jiangshan
2013-03-20 16:04 ` Tejun Heo
2013-03-20 0:00 ` [PATCH 09/10] workqueue: implement NUMA affinity for unbound workqueues Tejun Heo
2013-03-20 15:03 ` Lai Jiangshan
2013-03-20 15:05 ` Tejun Heo
2013-03-20 15:26 ` Lai Jiangshan
2013-03-20 15:32 ` Tejun Heo
2013-03-20 17:08 ` [PATCH v2 " Tejun Heo
2013-03-20 18:54 ` [PATCH v2 UPDATED " Tejun Heo
2013-03-20 0:00 ` [PATCH 10/10] workqueue: update sysfs interface to reflect NUMA awareness and a kernel param to disable NUMA affinity Tejun Heo
2013-03-20 12:14 ` [PATCHSET wq/for-3.10] workqueue: NUMA affinity for unbound workqueues Lai Jiangshan
2013-03-20 17:08 ` [PATCH 11/10] workqueue: use NUMA-aware allocation for pool_workqueues workqueues Tejun Heo
2013-03-20 18:57 ` [PATCHSET wq/for-3.10] workqueue: NUMA affinity for unbound workqueues Tejun Heo
2013-03-24 16:04 ` Lai Jiangshan
2013-03-24 18:55 ` Tejun Heo
2013-03-25 19:15 ` Tejun Heo
2013-03-25 20:48 ` Tejun Heo
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=1363737629-16745-5-git-send-email-tj@kernel.org \
--to=tj@kernel.org \
--cc=axboe@kernel.dk \
--cc=davem@davemloft.net \
--cc=fengguang.wu@intel.com \
--cc=herbert@gondor.hengli.com.au \
--cc=jack@suse.cz \
--cc=jmoyer@redhat.com \
--cc=laijs@cn.fujitsu.com \
--cc=linux-crypto@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=zab@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
Powered by JetHome