veth netdevice defines own rx queues and allocates array containing
up to 4095 ~750-bytes-long 'struct veth_rq' elements. Such allocation
is quite huge and should be accounted to memcg.

Signed-off-by: Vasily Averin <vvs@openvz.org>
---
 drivers/net/veth.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/veth.c b/drivers/net/veth.c
index d29fb9759cc9..bd67f458641a 100644
--- a/drivers/net/veth.c
+++ b/drivers/net/veth.c
@@ -1310,7 +1310,7 @@ static int veth_alloc_queues(struct net_device *dev)
 	struct veth_priv *priv = netdev_priv(dev);
 	int i;
 
-	priv->rq = kcalloc(dev->num_rx_queues, sizeof(*priv->rq), GFP_KERNEL);
+	priv->rq = kcalloc(dev->num_rx_queues, sizeof(*priv->rq), GFP_KERNEL_ACCOUNT);
 	if (!priv->rq)
 		return -ENOMEM;
 
-- 
2.31.1

On Wed, 27 Apr 2022 13:34:29 +0300 Vasily Averin wrote:
> Subject: [PATCH] memcg: enable accounting for veth queues

This is a pure networking patch, right? The prefix should be "net: ",
I think.

> veth netdevice defines own rx queues and allocates array containing
> up to 4095 ~750-bytes-long 'struct veth_rq' elements. Such allocation
> is quite huge and should be accounted to memcg.
> 
> Signed-off-by: Vasily Averin <vvs@openvz.org>

On 4/27/22 19:58, Jakub Kicinski wrote:
> On Wed, 27 Apr 2022 13:34:29 +0300 Vasily Averin wrote:
>> Subject: [PATCH] memcg: enable accounting for veth queues
> 
> This is a pure networking patch, right? The prefix should be "net: ",
> I think.
Thank you for the remark, I'll fix it.

Initially it was a part of the patch accounted resources accounted 
when creating a new netdevice, but then I moved this piece to
a separate patch, because unlike other cases, it is specific to veth.
 
Thank you,
	Vasily Averin

veth netdevice defines own rx queues and allocates array containing
up to 4095 ~750-bytes-long 'struct veth_rq' elements. Such allocation
is quite huge and should be accounted to memcg.

Signed-off-by: Vasily Averin <vvs@openvz.org>
---
 drivers/net/veth.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/veth.c b/drivers/net/veth.c
index d29fb9759cc9..bd67f458641a 100644
--- a/drivers/net/veth.c
+++ b/drivers/net/veth.c
@@ -1310,7 +1310,7 @@ static int veth_alloc_queues(struct net_device *dev)
 	struct veth_priv *priv = netdev_priv(dev);
 	int i;
 
-	priv->rq = kcalloc(dev->num_rx_queues, sizeof(*priv->rq), GFP_KERNEL);
+	priv->rq = kcalloc(dev->num_rx_queues, sizeof(*priv->rq), GFP_KERNEL_ACCOUNT);
 	if (!priv->rq)
 		return -ENOMEM;
 
-- 
2.31.1

Hello:

This patch was applied to netdev/net-next.git (master)
by David S. Miller <davem@davemloft.net>:

On Fri, 29 Apr 2022 08:17:35 +0300 you wrote:
> veth netdevice defines own rx queues and allocates array containing
> up to 4095 ~750-bytes-long 'struct veth_rq' elements. Such allocation
> is quite huge and should be accounted to memcg.
> 
> Signed-off-by: Vasily Averin <vvs@openvz.org>
> ---
>  drivers/net/veth.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)

Here is the summary with links:
  - [net] net: enable memcg accounting for veth queues
    https://git.kernel.org/netdev/net-next/c/961c6136359e

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html