[PATCH v4 0/2] dm: restrict conflicting table loads and improve queue initialization

[PATCH v4 0/2] dm: restrict conflicting table loads and improve queue initialization

[Mike Snitzer]

A mapped_device now has a specific immutable type (md->type) that is
set during the initial table_load (see: patch 1/2).

The DM device's request_queue will only have an elevator allocated and
exposed via sysfs if it is request-based (see: patch 2/2).

Please see the individual patch headers for more details.

This patchset is based on agk's DM "editing" tree:
http://www.kernel.org/pub/linux/kernel/people/agk/patches/2.6/editing/
(apply all patches listed in the series file up to NEXT_PATCHES_END)

This patchset also depends on the following linux-2.6-block.git
'for-linus' branch commits:
01effb0 block: allow initialization of previously allocated request_queue
e36f724 block: Adjust elv_iosched_show to return "none" for bio-based DM

I've made a full quilt series available here (builds on 2.6.34):
http://people.redhat.com/msnitzer/patches/dm-queue-init/latest/

I welcome all comments/review.

v4 change:
- dm_set_md_type takes a specific unsigned type rather than dm_table*

v3 changes:
- reduce table_load's md->type_lock protected critical section
- only call dm_setup_md_queue on first table load
  - simplified dm_setup_md_queue as a result
- use existing DM_TYPE_* definitions instead of duplicating them
- eliminate overly verbose comments as the code speaks for itself much
  more clearly

v2 changes:
- remove do_resume and table_clear interlock
- rebase to agk's latest "editing" tree
- drop the dm_get_verified_mdptr and find_device_noinit patches that
  were at the end of v1; can revisit that code duplication cleanup later

Mike Snitzer (2):
  dm: prevent table type changes after initial table load
  dm: only initialize full request_queue for request-based device

 drivers/md/dm-ioctl.c |   24 +++++++++
 drivers/md/dm.c       |  131 +++++++++++++++++++++++++++++++++++++------------
 drivers/md/dm.h       |    7 +++
 3 files changed, 130 insertions(+), 32 deletions(-)

[PATCH v4 1/2] dm: prevent table type changes after initial table load

[Mike Snitzer]

Before table_load() stages an inactive table atomically check if its
type conflicts with a previously established device type (md->type).

Introduce 'type_lock' in mapped_device structure and use it to protect
md->type access.  table_load() sets md->type without concern for:
- another table_load() racing to set conflicting md->type.
- do_resume() making a conflicting table live.

Allowed transitions:
DM_TYPE_NONE => DM_TYPE_BIO_BASED
DM_TYPE_NONE => DM_TYPE_REQUEST_BASED

Once a table load occurs the DM device's type is completely immutable.

This prevents a table reload from switching the inactive table directly
to a conflicting type (even if the table is explicitly cleared before
load).

Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Acked-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
---
 drivers/md/dm-ioctl.c |   15 +++++++++++++++
 drivers/md/dm.c       |   37 ++++++++++++++++++++++++++++++-------
 drivers/md/dm.h       |    5 +++++
 3 files changed, 50 insertions(+), 7 deletions(-)

Index: linux-2.6/drivers/md/dm-ioctl.c
===================================================================
--- linux-2.6.orig/drivers/md/dm-ioctl.c
+++ linux-2.6/drivers/md/dm-ioctl.c
@@ -1176,6 +1176,21 @@ static int table_load(struct dm_ioctl *p
 		goto out;
 	}
 
+	/* Protect md->type against concurrent table loads. */
+	dm_lock_md_type(md);
+	if (dm_get_md_type(md) == DM_TYPE_NONE) {
+		/* initial table load, set md's type based on table's type */
+		dm_set_md_type(md, dm_table_get_type(t));
+	} else if (dm_get_md_type(md) != dm_table_get_type(t)) {
+		DMWARN("can't change device type after initial table load.");
+		dm_table_destroy(t);
+		dm_unlock_md_type(md);
+		r = -EINVAL;
+		goto out;
+	}
+	dm_unlock_md_type(md);
+
+	/* stage inactive table */
 	down_write(&_hash_lock);
 	hc = dm_get_mdptr(md);
 	if (!hc || hc->md != md) {
Index: linux-2.6/drivers/md/dm.c
===================================================================
--- linux-2.6.orig/drivers/md/dm.c
+++ linux-2.6/drivers/md/dm.c
@@ -123,6 +123,10 @@ struct mapped_device {
 	unsigned long flags;
 
 	struct request_queue *queue;
+	unsigned type;
+	/* Protect type against concurrent access. */
+	struct mutex type_lock;
+
 	struct gendisk *disk;
 	char name[16];
 
@@ -1874,8 +1878,10 @@ static struct mapped_device *alloc_dev(i
 	if (r < 0)
 		goto bad_minor;
 
+	md->type = DM_TYPE_NONE;
 	init_rwsem(&md->io_lock);
 	mutex_init(&md->suspend_lock);
+	mutex_init(&md->type_lock);
 	spin_lock_init(&md->deferred_lock);
 	spin_lock_init(&md->barrier_error_lock);
 	rwlock_init(&md->map_lock);
@@ -2128,6 +2134,30 @@ int dm_create(int minor, struct mapped_d
 	return 0;
 }
 
+/*
+ * Functions to manage md->type.
+ * All are required to hold md->type_lock.
+ */
+void dm_lock_md_type(struct mapped_device *md)
+{
+	mutex_lock(&md->type_lock);
+}
+
+void dm_unlock_md_type(struct mapped_device *md)
+{
+	mutex_unlock(&md->type_lock);
+}
+
+void dm_set_md_type(struct mapped_device *md, unsigned type)
+{
+	md->type = type;
+}
+
+unsigned dm_get_md_type(struct mapped_device *md)
+{
+	return md->type;
+}
+
 static struct mapped_device *dm_find_md(dev_t dev)
 {
 	struct mapped_device *md;
@@ -2403,13 +2433,6 @@ struct dm_table *dm_swap_table(struct ma
 		goto out;
 	}
 
-	/* cannot change the device type, once a table is bound */
-	if (md->map &&
-	    (dm_table_get_type(md->map) != dm_table_get_type(table))) {
-		DMWARN("can't change the device type after a table is bound");
-		goto out;
-	}
-
 	map = __bind(md, table, &limits);
 
 out:
Index: linux-2.6/drivers/md/dm.h
===================================================================
--- linux-2.6.orig/drivers/md/dm.h
+++ linux-2.6/drivers/md/dm.h
@@ -66,6 +66,11 @@ int dm_table_alloc_md_mempools(struct dm
 void dm_table_free_md_mempools(struct dm_table *t);
 struct dm_md_mempools *dm_table_get_md_mempools(struct dm_table *t);
 
+void dm_lock_md_type(struct mapped_device *md);
+void dm_unlock_md_type(struct mapped_device *md);
+void dm_set_md_type(struct mapped_device *md, unsigned type);
+unsigned dm_get_md_type(struct mapped_device *md);
+
 /*
  * To check the return value from dm_table_find_target().
  */

Re: [PATCH v4 1/2] dm: prevent table type changes after initial table load

[Alasdair G Kergon]

On Thu, May 27, 2010 at 08:47:48AM -0400, Mike Snitzer wrote:
> Introduce 'type_lock' in mapped_device structure and use it to protect
> md->type access.  table_load() sets md->type without concern for:

> - another table_load() racing to set conflicting md->type.
which leads to non-deterministic behaviour that is of no practical use and
so there is no need to support it.

> - do_resume() making a conflicting table live.
Any table being made live must already have passed through table_load so I
don't see how there can be conflicting types.

The mutex here seems to be overkill: instead of protecting one field to
cope with a useless race, we should be making the output of the race
deterministic (e.g. perhaps giving an error on the parallel table_load
attempt).

Anyway, I can take this patch for now and we can deal with the mutex/race
differently later.

Alasdair
 

Re: [PATCH v4 1/2] dm: prevent table type changes after initial table load

[Mike Snitzer]

On Tue, Jun 29 2010 at  7:40am -0400,
Alasdair G Kergon <agk@redhat.com> wrote:

> On Thu, May 27, 2010 at 08:47:48AM -0400, Mike Snitzer wrote:
> > Introduce 'type_lock' in mapped_device structure and use it to protect
> > md->type access.  table_load() sets md->type without concern for:
> 
> > - another table_load() racing to set conflicting md->type.
> which leads to non-deterministic behaviour that is of no practical use and
> so there is no need to support it.
> 
> > - do_resume() making a conflicting table live.
> Any table being made live must already have passed through table_load so I
> don't see how there can be conflicting types.

As we talked about on irc.  There is potential for do_resume() to
destage the "inactive" table slot, in the process of making the table
"live", and in parallel another table load can arrive to stage another
"inactive" table (which can have a conflicting type).  Without a
critical section there is room for this to occur.

And the critical section must span more than setting md->type; as
md->queue initialization may fail (elevator memory allocation fails
ENOMEM).  So competing table_load needs protection from concurrent queue
initialization.

(do understand that I know you're commenting on this patch header in
isolation; but I figured I'd give context for "why the mutex").  I have
an updated patch header too:

Before table_load() stages an inactive table check if its type
conflicts with a previously established device type (md->type).

Allowed md->type transitions:
DM_TYPE_NONE => DM_TYPE_BIO_BASED
DM_TYPE_NONE => DM_TYPE_REQUEST_BASED

Once a table load occurs the DM device's type is completely immutable.
This prevents a table reload from switching the inactive table directly
to a conflicting type (even if the table is explicitly cleared before
load).

Introduce 'type_lock' in mapped_device structure and use it to protect
md->type access.  Use of mutex prepares for follow-on DM device queue
initialization changes that will allocate memory while md->type_lock
is held.

Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Acked-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>


> The mutex here seems to be overkill: instead of protecting one field to
> cope with a useless race, we should be making the output of the race
> deterministic (e.g. perhaps giving an error on the parallel table_load
> attempt).

We're protecting 2 fields in the end (md->type and md->queue): once 2/2
is applied.

I agree 100% on improving the DM core to be trained to error out on
competing table loads (at a higher level).. given the parallelism that
we have right now in DM operations (across devices) we get this awkward
inability to _know_ the state transitions of:
no table -> inactive table -> live table

As you pointed out that is an unwanted side-effect of supporting
operations we do want.

> Anyway, I can take this patch for now and we can deal with the mutex/race
> differently later.

OK, sounds good.

Mike

Re: [PATCH v4 1/2] dm: prevent table type changes after initial table load

[Alasdair G Kergon]

On Tue, Jun 29, 2010 at 09:52:17AM -0400, Mike Snitzer wrote:
> As we talked about on irc.  There is potential for do_resume() to
> destage the "inactive" table slot, in the process of making the table
> "live", and in parallel another table load can arrive to stage another
> "inactive" table (which can have a conflicting type).  

But the type already got set earlier and, once set, is an immutable table
property.  There cannot be conflicts with do_resume - it doesn't touch
this.

Alasdair

Re: [PATCH v4 1/2] dm: prevent table type changes after initial table load

[Mike Snitzer]

On Tue, Jun 29 2010 at 10:19am -0400,
Alasdair G Kergon <agk@redhat.com> wrote:

> On Tue, Jun 29, 2010 at 09:52:17AM -0400, Mike Snitzer wrote:
> > As we talked about on irc.  There is potential for do_resume() to
> > destage the "inactive" table slot, in the process of making the table
> > "live", and in parallel another table load can arrive to stage another
> > "inactive" table (which can have a conflicting type).  
> 
> But the type already got set earlier and, once set, is an immutable table
> property.  There cannot be conflicts with do_resume - it doesn't touch
> this.

Yes, do_resume doesn't operate with any concern for md->type or
md->queue changing because table_load safely sets these immutable
properties.

That safety comes at the expense of protecting a critical section in
table_load.  And that protection must span the access to both md->type
and md->queue.

Would you rather I ignore do_resume entirely?  I'm not understanding why
you keep going back to this notion that do_resume doesn't matter.  The
only reason it doesn't matter is because table_load takes care to make
it that way.

Without appropriate table_load serialization do_resume would be prone to
the races I explained above.  That's all I've been trying to convey.

Mike

[PATCH v4 2/2 "v11"] dm: only initialize full request_queue for request-based device

[Mike Snitzer]

Allocate a minimalist request_queue structure initially (needed for both
bio and request-based DM).  A bio-based DM device no longer defaults to
having a fully initialized request_queue (request_fn, elevator, etc).
So bio-based DM devices no longer register elevator sysfs attributes
('iosched/' tree or 'scheduler' other than "none").

Initialization of a full request_queue (request_fn, elevator, etc) is
deferred until it is known that the DM device is request-based -- at the
end of the table load sequence.

Factor DM device's request_queue initialization:
- common to both request-based and bio-based into dm_init_md_queue().
- specific to request-based into dm_init_request_based_queue().

md->type_lock is also used to protect md->queue during table_load().
md->queue is setup without concern for:
- another table_load() racing to setup conflicting queue state.
- do_resume() making a conflicting table live.

Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Acked-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
---
 drivers/md/dm-ioctl.c |   11 +++++
 drivers/md/dm.c       |   93 ++++++++++++++++++++++++++++++++++++--------------
 drivers/md/dm.h       |    2 +
 3 files changed, 79 insertions(+), 27 deletions(-)

Index: linux-2.6/drivers/md/dm-ioctl.c
===================================================================
--- linux-2.6.orig/drivers/md/dm-ioctl.c
+++ linux-2.6/drivers/md/dm-ioctl.c
@@ -1176,11 +1176,20 @@ static int table_load(struct dm_ioctl *p
 		goto out;
 	}
 
-	/* Protect md->type against concurrent table loads. */
+	/* Protect md->type and md->queue against concurrent table loads. */
 	dm_lock_md_type(md);
 	if (dm_get_md_type(md) == DM_TYPE_NONE) {
 		/* initial table load, set md's type based on table's type */
 		dm_set_md_type(md, dm_table_get_type(t));
+
+		/* setup md->queue to reflect md's type (may block) */
+		r = dm_setup_md_queue(md);
+		if (r) {
+			DMWARN("unable to setup device queue for this table.");
+			dm_table_destroy(t);
+			dm_unlock_md_type(md);
+			goto out;
+		}
 	} else if (dm_get_md_type(md) != dm_table_get_type(t)) {
 		DMWARN("can't change device type after initial table load.");
 		dm_table_destroy(t);
Index: linux-2.6/drivers/md/dm.c
===================================================================
--- linux-2.6.orig/drivers/md/dm.c
+++ linux-2.6/drivers/md/dm.c
@@ -124,7 +124,7 @@ struct mapped_device {
 
 	struct request_queue *queue;
 	unsigned type;
-	/* Protect type against concurrent access. */
+	/* Protect queue and type against concurrent access. */
 	struct mutex type_lock;
 
 	struct gendisk *disk;
@@ -1853,6 +1853,28 @@ static const struct block_device_operati
 static void dm_wq_work(struct work_struct *work);
 static void dm_rq_barrier_work(struct work_struct *work);
 
+static void dm_init_md_queue(struct mapped_device *md)
+{
+	/*
+	 * Request-based dm devices cannot be stacked on top of bio-based dm
+	 * devices.  The type of this dm device has not been decided yet.
+	 * The type is decided at the first table loading time.
+	 * To prevent problematic device stacking, clear the queue flag
+	 * for request stacking support until then.
+	 *
+	 * This queue is new, so no concurrency on the queue_flags.
+	 */
+	queue_flag_clear_unlocked(QUEUE_FLAG_STACKABLE, md->queue);
+
+	md->queue->queuedata = md;
+	md->queue->backing_dev_info.congested_fn = dm_any_congested;
+	md->queue->backing_dev_info.congested_data = md;
+	blk_queue_make_request(md->queue, dm_request);
+	blk_queue_bounce_limit(md->queue, BLK_BOUNCE_ANY);
+	md->queue->unplug_fn = dm_unplug_all;
+	blk_queue_merge_bvec(md->queue, dm_merge_bvec);
+}
+
 /*
  * Allocate and initialise a blank device with a given minor.
  */
@@ -1892,34 +1914,11 @@ static struct mapped_device *alloc_dev(i
 	INIT_LIST_HEAD(&md->uevent_list);
 	spin_lock_init(&md->uevent_lock);
 
-	md->queue = blk_init_queue(dm_request_fn, NULL);
+	md->queue = blk_alloc_queue(GFP_KERNEL);
 	if (!md->queue)
 		goto bad_queue;
 
-	/*
-	 * Request-based dm devices cannot be stacked on top of bio-based dm
-	 * devices.  The type of this dm device has not been decided yet,
-	 * although we initialized the queue using blk_init_queue().
-	 * The type is decided at the first table loading time.
-	 * To prevent problematic device stacking, clear the queue flag
-	 * for request stacking support until then.
-	 *
-	 * This queue is new, so no concurrency on the queue_flags.
-	 */
-	queue_flag_clear_unlocked(QUEUE_FLAG_STACKABLE, md->queue);
-	md->saved_make_request_fn = md->queue->make_request_fn;
-	md->queue->queuedata = md;
-	md->queue->backing_dev_info.congested_fn = dm_any_congested;
-	md->queue->backing_dev_info.congested_data = md;
-	blk_queue_make_request(md->queue, dm_request);
-	blk_queue_bounce_limit(md->queue, BLK_BOUNCE_ANY);
-	md->queue->unplug_fn = dm_unplug_all;
-	blk_queue_merge_bvec(md->queue, dm_merge_bvec);
-	blk_queue_softirq_done(md->queue, dm_softirq_done);
-	blk_queue_prep_rq(md->queue, dm_prep_fn);
-	blk_queue_lld_busy(md->queue, dm_lld_busy);
-	blk_queue_ordered(md->queue, QUEUE_ORDERED_DRAIN_FLUSH,
-			  dm_rq_prepare_flush);
+	dm_init_md_queue(md);
 
 	md->disk = alloc_disk(1);
 	if (!md->disk)
@@ -2158,6 +2157,48 @@ unsigned dm_get_md_type(struct mapped_de
 	return md->type;
 }
 
+/*
+ * Fully initialize a request-based queue (->elevator, ->request_fn, etc).
+ */
+static int dm_init_request_based_queue(struct mapped_device *md)
+{
+	struct request_queue *q = NULL;
+
+	BUG_ON(md->queue->elevator);
+
+	/* Fully initialize the queue */
+	q = blk_init_allocated_queue(md->queue, dm_request_fn, NULL);
+	if (!q)
+		return 0;
+
+	md->queue = q;
+	md->saved_make_request_fn = md->queue->make_request_fn;
+	dm_init_md_queue(md);
+	blk_queue_softirq_done(md->queue, dm_softirq_done);
+	blk_queue_prep_rq(md->queue, dm_prep_fn);
+	blk_queue_lld_busy(md->queue, dm_lld_busy);
+	blk_queue_ordered(md->queue, QUEUE_ORDERED_DRAIN_FLUSH,
+			  dm_rq_prepare_flush);
+
+	elv_register_queue(md->queue);
+
+	return 1;
+}
+
+/*
+ * Setup the DM device's queue based on md's type
+ */
+int dm_setup_md_queue(struct mapped_device *md)
+{
+	if ((dm_get_md_type(md) == DM_TYPE_REQUEST_BASED) &&
+	    !dm_init_request_based_queue(md)) {
+		DMWARN("Cannot initialize queue for Request-based dm");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 static struct mapped_device *dm_find_md(dev_t dev)
 {
 	struct mapped_device *md;
Index: linux-2.6/drivers/md/dm.h
===================================================================
--- linux-2.6.orig/drivers/md/dm.h
+++ linux-2.6/drivers/md/dm.h
@@ -71,6 +71,8 @@ void dm_unlock_md_type(struct mapped_dev
 void dm_set_md_type(struct mapped_device *md, unsigned type);
 unsigned dm_get_md_type(struct mapped_device *md);
 
+int dm_setup_md_queue(struct mapped_device *md);
+
 /*
  * To check the return value from dm_table_find_target().
  */

[PATCH 3/2] dm: table load must always try dm_setup_md_queue

[Mike Snitzer]

A DM device's first table_load will establish the immutable md->type.
But md->queue initialization, based on md->type, may fail at that time
(if blk_init_allocated_queue cannot allocate memory).

Therefore any subsequent table_load must (re)try dm_setup_md_queue
independent of establishing md->type.

Signed-off-by: Mike Snitzer <snitzer@redhat.com>
---
 drivers/md/dm-ioctl.c |   18 +++++++++---------
 drivers/md/dm.c       |    3 ++-
 2 files changed, 11 insertions(+), 10 deletions(-)

NOTE: coping with blk_init_allocated_queue failure was mistakenly
"optimized" away during review of the patch that introduced
dm_setup_md_queue:
https://www.redhat.com/archives/dm-devel/2010-May/msg00173.html

Index: linux-2.6/drivers/md/dm-ioctl.c
===================================================================
--- linux-2.6.orig/drivers/md/dm-ioctl.c
+++ linux-2.6/drivers/md/dm-ioctl.c
@@ -1181,15 +1181,6 @@ static int table_load(struct dm_ioctl *p
 	if (dm_get_md_type(md) == DM_TYPE_NONE) {
 		/* initial table load, set md's type based on table's type */
 		dm_set_md_type(md, dm_table_get_type(t));
-
-		/* setup md->queue to reflect md's type (may block) */
-		r = dm_setup_md_queue(md);
-		if (r) {
-			DMWARN("unable to setup device queue for this table.");
-			dm_table_destroy(t);
-			dm_unlock_md_type(md);
-			goto out;
-		}
 	} else if (dm_get_md_type(md) != dm_table_get_type(t)) {
 		DMWARN("can't change device type after initial table load.");
 		dm_table_destroy(t);
@@ -1197,6 +1188,15 @@ static int table_load(struct dm_ioctl *p
 		r = -EINVAL;
 		goto out;
 	}
+
+	/* setup md->queue to reflect md's type (may block) */
+	r = dm_setup_md_queue(md);
+	if (r) {
+		DMWARN("unable to setup device queue for this table.");
+		dm_table_destroy(t);
+		dm_unlock_md_type(md);
+		goto out;
+	}
 	dm_unlock_md_type(md);
 
 	/* stage inactive table */
Index: linux-2.6/drivers/md/dm.c
===================================================================
--- linux-2.6.orig/drivers/md/dm.c
+++ linux-2.6/drivers/md/dm.c
@@ -2164,7 +2164,8 @@ static int dm_init_request_based_queue(s
 {
 	struct request_queue *q = NULL;
 
-	BUG_ON(md->queue->elevator);
+	if (unlikely(md->queue->elevator))
+		return 1;
 
 	/* Fully initialize the queue */
 	q = blk_init_allocated_queue(md->queue, dm_request_fn, NULL);

Re: [PATCH 3/2] dm: table load must always try dm_setup_md_queue

[Kiyoshi Ueda]

Hi Mike,

On 06/05/2010 05:15 AM +0900, Mike Snitzer wrote:
> @@ -2164,7 +2164,8 @@ static int dm_init_request_based_queue(s
>  {
>  	struct request_queue *q = NULL;
>  
> -	BUG_ON(md->queue->elevator);
> +	if (unlikely(md->queue->elevator))
> +		return 1;

I think the "unlikely" should be rather "likely", since
dm_init_request_based_queue() is now called whenever request-based
table is loaded even after the actual initialization has been done.

For others,
Acked-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>

Thanks,
Kiyoshi Ueda

Re: [PATCH 3/2] dm: table load must always try dm_setup_md_queue

[Mike Snitzer]

On Wed, Jun 09 2010 at  1:38am -0400,
Kiyoshi Ueda <k-ueda@ct.jp.nec.com> wrote:

> Hi Mike,
> 
> On 06/05/2010 05:15 AM +0900, Mike Snitzer wrote:
> > @@ -2164,7 +2164,8 @@ static int dm_init_request_based_queue(s
> >  {
> >  	struct request_queue *q = NULL;
> >  
> > -	BUG_ON(md->queue->elevator);
> > +	if (unlikely(md->queue->elevator))
> > +		return 1;
> 
> I think the "unlikely" should be rather "likely", since
> dm_init_request_based_queue() is now called whenever request-based
> table is loaded even after the actual initialization has been done.

Yes, good point.  Though I'd expect reloading a request-based table is
actually fairly rare.  Is it really all that worthwhile to have any
branch prediction here?  Should we just remove the likely/unlikely
entirely?

Thanks,
Mike

Re: [PATCH 3/2] dm: table load must always try dm_setup_md_queue

[Kiyoshi Ueda]

Hi Mike,

On 06/09/2010 05:53 PM +0900, Mike Snitzer wrote:
> On Wed, Jun 09 2010 at  1:38am -0400, Kiyoshi Ueda wrote:
>> On 06/05/2010 05:15 AM +0900, Mike Snitzer wrote:
>>> @@ -2164,7 +2164,8 @@ static int dm_init_request_based_queue(s
>>>  {
>>>  	struct request_queue *q = NULL;
>>>  
>>> -	BUG_ON(md->queue->elevator);
>>> +	if (unlikely(md->queue->elevator))
>>> +		return 1;
>> 
>> I think the "unlikely" should be rather "likely", since
>> dm_init_request_based_queue() is now called whenever request-based
>> table is loaded even after the actual initialization has been done.
> 
> Yes, good point.  Though I'd expect reloading a request-based table is
> actually fairly rare.  Is it really all that worthwhile to have any
> branch prediction here?  Should we just remove the likely/unlikely
> entirely?

For multipath, table reloading happens when path connection state is
changed (e.g. link-up => link-down), since if a path state becomes down,
the device for the path disappears and multipath configuration is changed.

Although I'm not sure "it's actually fairly rare", I believe it's rare.
So I don't object to remove the branch prediction.

Thanks,
Kiyoshi Ueda