[PATCH] xen/blkback: use rb_entry()

[PATCH] xen/blkback: use rb_entry()

[Geliang Tang]

To make the code clearer, use rb_entry() instead of container_of() to
deal with rbtree.

Signed-off-by: Geliang Tang <geliangtang@gmail.com>
---
 drivers/block/xen-blkback/blkback.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c
index 726c32e..7e59fae 100644
--- a/drivers/block/xen-blkback/blkback.c
+++ b/drivers/block/xen-blkback/blkback.c
@@ -191,10 +191,10 @@ static void make_response(struct xen_blkif_ring *ring, u64 id,
 			  unsigned short op, int st);
 
 #define foreach_grant_safe(pos, n, rbtree, node) \
-	for ((pos) = container_of(rb_first((rbtree)), typeof(*(pos)), node), \
+	for ((pos) = rb_entry(rb_first((rbtree)), typeof(*(pos)), node), \
 	     (n) = (&(pos)->node != NULL) ? rb_next(&(pos)->node) : NULL; \
 	     &(pos)->node != NULL; \
-	     (pos) = container_of(n, typeof(*(pos)), node), \
+	     (pos) = rb_entry(n, typeof(*(pos)), node), \
 	     (n) = (&(pos)->node != NULL) ? rb_next(&(pos)->node) : NULL)
 
 
@@ -223,7 +223,7 @@ static int add_persistent_gnt(struct xen_blkif_ring *ring,
 	/* Figure out where to put new node */
 	new = &ring->persistent_gnts.rb_node;
 	while (*new) {
-		this = container_of(*new, struct persistent_gnt, node);
+		this = rb_entry(*new, struct persistent_gnt, node);
 
 		parent = *new;
 		if (persistent_gnt->gnt < this->gnt)
@@ -254,7 +254,7 @@ static struct persistent_gnt *get_persistent_gnt(struct xen_blkif_ring *ring,
 
 	node = ring->persistent_gnts.rb_node;
 	while (node) {
-		data = container_of(node, struct persistent_gnt, node);
+		data = rb_entry(node, struct persistent_gnt, node);
 
 		if (gref < data->gnt)
 			node = node->rb_left;
-- 
2.9.3


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

[PATCH] xen/evtchn: use rb_entry()

[Geliang Tang]

To make the code clearer, use rb_entry() instead of container_of() to
deal with rbtree.

Signed-off-by: Geliang Tang <geliangtang@gmail.com>
---
 drivers/xen/evtchn.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/xen/evtchn.c b/drivers/xen/evtchn.c
index e8c7f09..6890897 100644
--- a/drivers/xen/evtchn.c
+++ b/drivers/xen/evtchn.c
@@ -125,7 +125,7 @@ static int add_evtchn(struct per_user_data *u, struct user_evtchn *evtchn)
 	while (*new) {
 		struct user_evtchn *this;
 
-		this = container_of(*new, struct user_evtchn, node);
+		this = rb_entry(*new, struct user_evtchn, node);
 
 		parent = *new;
 		if (this->port < evtchn->port)
@@ -157,7 +157,7 @@ static struct user_evtchn *find_evtchn(struct per_user_data *u, unsigned port)
 	while (node) {
 		struct user_evtchn *evtchn;
 
-		evtchn = container_of(node, struct user_evtchn, node);
+		evtchn = rb_entry(node, struct user_evtchn, node);
 
 		if (evtchn->port < port)
 			node = node->rb_left;
-- 
2.9.3


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [PATCH] xen/evtchn: use rb_entry()

[Juergen Gross]

On 20/12/16 15:02, Geliang Tang wrote:
> To make the code clearer, use rb_entry() instead of container_of() to
> deal with rbtree.
> 
> Signed-off-by: Geliang Tang <geliangtang@gmail.com>

Reviewed-by: Juergen Gross <jgross@suse.com>

> ---
>  drivers/xen/evtchn.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/xen/evtchn.c b/drivers/xen/evtchn.c
> index e8c7f09..6890897 100644
> --- a/drivers/xen/evtchn.c
> +++ b/drivers/xen/evtchn.c
> @@ -125,7 +125,7 @@ static int add_evtchn(struct per_user_data *u, struct user_evtchn *evtchn)
>  	while (*new) {
>  		struct user_evtchn *this;
>  
> -		this = container_of(*new, struct user_evtchn, node);
> +		this = rb_entry(*new, struct user_evtchn, node);
>  
>  		parent = *new;
>  		if (this->port < evtchn->port)
> @@ -157,7 +157,7 @@ static struct user_evtchn *find_evtchn(struct per_user_data *u, unsigned port)
>  	while (node) {
>  		struct user_evtchn *evtchn;
>  
> -		evtchn = container_of(node, struct user_evtchn, node);
> +		evtchn = rb_entry(node, struct user_evtchn, node);
>  
>  		if (evtchn->port < port)
>  			node = node->rb_left;
> 


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [PATCH] xen/blkback: use rb_entry()

[Konrad Rzeszutek Wilk]

On Tue, Dec 20, 2016 at 10:02:19PM +0800, Geliang Tang wrote:
> To make the code clearer, use rb_entry() instead of container_of() to
> deal with rbtree.

That is OK but I think 'container_of' is more clear.

Roger, thoughts?

> 
> Signed-off-by: Geliang Tang <geliangtang@gmail.com>
> ---
>  drivers/block/xen-blkback/blkback.c | 8 ++++----
>  1 file changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c
> index 726c32e..7e59fae 100644
> --- a/drivers/block/xen-blkback/blkback.c
> +++ b/drivers/block/xen-blkback/blkback.c
> @@ -191,10 +191,10 @@ static void make_response(struct xen_blkif_ring *ring, u64 id,
>  			  unsigned short op, int st);
>  
>  #define foreach_grant_safe(pos, n, rbtree, node) \
> -	for ((pos) = container_of(rb_first((rbtree)), typeof(*(pos)), node), \
> +	for ((pos) = rb_entry(rb_first((rbtree)), typeof(*(pos)), node), \
>  	     (n) = (&(pos)->node != NULL) ? rb_next(&(pos)->node) : NULL; \
>  	     &(pos)->node != NULL; \
> -	     (pos) = container_of(n, typeof(*(pos)), node), \
> +	     (pos) = rb_entry(n, typeof(*(pos)), node), \
>  	     (n) = (&(pos)->node != NULL) ? rb_next(&(pos)->node) : NULL)
>  
>  
> @@ -223,7 +223,7 @@ static int add_persistent_gnt(struct xen_blkif_ring *ring,
>  	/* Figure out where to put new node */
>  	new = &ring->persistent_gnts.rb_node;
>  	while (*new) {
> -		this = container_of(*new, struct persistent_gnt, node);
> +		this = rb_entry(*new, struct persistent_gnt, node);
>  
>  		parent = *new;
>  		if (persistent_gnt->gnt < this->gnt)
> @@ -254,7 +254,7 @@ static struct persistent_gnt *get_persistent_gnt(struct xen_blkif_ring *ring,
>  
>  	node = ring->persistent_gnts.rb_node;
>  	while (node) {
> -		data = container_of(node, struct persistent_gnt, node);
> +		data = rb_entry(node, struct persistent_gnt, node);
>  
>  		if (gref < data->gnt)
>  			node = node->rb_left;
> -- 
> 2.9.3
> 

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [PATCH] xen/blkback: use rb_entry()

[Roger Pau Monné]

On Tue, Dec 20, 2016 at 11:47:03AM -0500, Konrad Rzeszutek Wilk wrote:
> On Tue, Dec 20, 2016 at 10:02:19PM +0800, Geliang Tang wrote:
> > To make the code clearer, use rb_entry() instead of container_of() to
> > deal with rbtree.
> 
> That is OK but I think 'container_of' is more clear.
> 
> Roger, thoughts?

I think so, container_of is a global macro that's widely used and everyone
knows, rb_entry OTOH it's not and it's use doesn't really simply the code at
all. I'm not really opposed, but it seems kind of a pointless change (not that
it's wrong).

Roger.

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [PATCH] xen/blkback: use rb_entry()

[Konrad Rzeszutek Wilk]

On Tue, Dec 20, 2016 at 05:44:06PM +0000, Roger Pau Monné wrote:
> On Tue, Dec 20, 2016 at 11:47:03AM -0500, Konrad Rzeszutek Wilk wrote:
> > On Tue, Dec 20, 2016 at 10:02:19PM +0800, Geliang Tang wrote:
> > > To make the code clearer, use rb_entry() instead of container_of() to
> > > deal with rbtree.
> > 
> > That is OK but I think 'container_of' is more clear.
> > 
> > Roger, thoughts?
> 
> I think so, container_of is a global macro that's widely used and everyone
> knows, rb_entry OTOH it's not and it's use doesn't really simply the code at
> all. I'm not really opposed, but it seems kind of a pointless change (not that
> it's wrong).

<nods> I concur.

Geliang Tang,

Thank you for the patch but there is no need for it.

Thanks again!
> 
> Roger.

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [PATCH] xen/blkback: use rb_entry()

[Eric Dumazet]

On Tue, 2016-12-20 at 12:51 -0500, Konrad Rzeszutek Wilk wrote:
> On Tue, Dec 20, 2016 at 05:44:06PM +0000, Roger Pau Monné wrote:
> > On Tue, Dec 20, 2016 at 11:47:03AM -0500, Konrad Rzeszutek Wilk wrote:
> > > On Tue, Dec 20, 2016 at 10:02:19PM +0800, Geliang Tang wrote:
> > > > To make the code clearer, use rb_entry() instead of container_of() to
> > > > deal with rbtree.
> > > 
> > > That is OK but I think 'container_of' is more clear.
> > > 
> > > Roger, thoughts?
> > 
> > I think so, container_of is a global macro that's widely used and everyone
> > knows, rb_entry OTOH it's not and it's use doesn't really simply the code at
> > all. I'm not really opposed, but it seems kind of a pointless change (not that
> > it's wrong).
> 
> <nods> I concur.
> 
> Geliang Tang,
> 
> Thank you for the patch but there is no need for it.

The same could be said of list_entry()

#define hlist_entry(ptr, type, member) container_of(ptr,type,member)

#define list_entry(ptr, type, member) container_of(ptr, type, member)

# git grep -n list_entry | wc -l
3636

rb_entry() will probably make its way everywhere.




_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [PATCH] xen/blkback: use rb_entry()

[Konrad Rzeszutek Wilk]

On Tue, Dec 20, 2016 at 01:53:21PM -0800, Eric Dumazet wrote:
> On Tue, 2016-12-20 at 12:51 -0500, Konrad Rzeszutek Wilk wrote:
> > On Tue, Dec 20, 2016 at 05:44:06PM +0000, Roger Pau Monné wrote:
> > > On Tue, Dec 20, 2016 at 11:47:03AM -0500, Konrad Rzeszutek Wilk wrote:
> > > > On Tue, Dec 20, 2016 at 10:02:19PM +0800, Geliang Tang wrote:
> > > > > To make the code clearer, use rb_entry() instead of container_of() to
> > > > > deal with rbtree.
> > > > 
> > > > That is OK but I think 'container_of' is more clear.
> > > > 
> > > > Roger, thoughts?
> > > 
> > > I think so, container_of is a global macro that's widely used and everyone
> > > knows, rb_entry OTOH it's not and it's use doesn't really simply the code at
> > > all. I'm not really opposed, but it seems kind of a pointless change (not that
> > > it's wrong).
> > 
> > <nods> I concur.
> > 
> > Geliang Tang,
> > 
> > Thank you for the patch but there is no need for it.
> 
> The same could be said of list_entry()
> 

Sure. And I am used to that as well :-)

> #define hlist_entry(ptr, type, member) container_of(ptr,type,member)
> 
> #define list_entry(ptr, type, member) container_of(ptr, type, member)
> 
> # git grep -n list_entry | wc -l
> 3636
> 
> rb_entry() will probably make its way everywhere.

That is good to know.

But for right now this patch is not necessary. Thank you!


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [PATCH] xen/evtchn: use rb_entry()

[Juergen Gross]

On 20/12/16 15:02, Geliang Tang wrote:
> To make the code clearer, use rb_entry() instead of container_of() to
> deal with rbtree.
> 
> Signed-off-by: Geliang Tang <geliangtang@gmail.com>

Committed to xen/tip.git for-linus-4.10


Juergen

> ---
>  drivers/xen/evtchn.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/xen/evtchn.c b/drivers/xen/evtchn.c
> index e8c7f09..6890897 100644
> --- a/drivers/xen/evtchn.c
> +++ b/drivers/xen/evtchn.c
> @@ -125,7 +125,7 @@ static int add_evtchn(struct per_user_data *u, struct user_evtchn *evtchn)
>  	while (*new) {
>  		struct user_evtchn *this;
>  
> -		this = container_of(*new, struct user_evtchn, node);
> +		this = rb_entry(*new, struct user_evtchn, node);
>  
>  		parent = *new;
>  		if (this->port < evtchn->port)
> @@ -157,7 +157,7 @@ static struct user_evtchn *find_evtchn(struct per_user_data *u, unsigned port)
>  	while (node) {
>  		struct user_evtchn *evtchn;
>  
> -		evtchn = container_of(node, struct user_evtchn, node);
> +		evtchn = rb_entry(node, struct user_evtchn, node);
>  
>  		if (evtchn->port < port)
>  			node = node->rb_left;
> 


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel