2014-04-22 05:31:33

by Jianyu Zhan

Subject: [PATCH] netclassid_cgroup: explicitly init the early_init field

For a cgroup subsystem who should init early, then it should carefully
take care of the implementation of css_alloc, because it will be called
before mm_init() setup the world.

Luckily we don't, and we better explicitly assign the early_init field
to 0, for document reason.

Signed-off-by: Jianyu Zhan <[email protected]>
---
net/core/netclassid_cgroup.c | 1 +
1 file changed, 1 insertion(+)

diff --git a/net/core/netclassid_cgroup.c b/net/core/netclassid_cgroup.c
index 22931e1..1b07dca 100644
--- a/net/core/netclassid_cgroup.c
+++ b/net/core/netclassid_cgroup.c
@@ -108,4 +108,5 @@ struct cgroup_subsys net_cls_cgrp_subsys = {
.css_free = cgrp_css_free,
.attach = cgrp_attach,
.base_cftypes = ss_files,
+ .early_init = 0,
};
--
2.0.0-rc0

2014-04-22 06:29:54

by Zefan Li

Subject: Re: [PATCH] netclassid_cgroup: explicitly init the early_init field

On 2014/4/22 13:31, Jianyu Zhan wrote:
> For a cgroup subsystem who should init early, then it should carefully
> take care of the implementation of css_alloc, because it will be called
> before mm_init() setup the world.
>
> Luckily we don't, and we better explicitly assign the early_init field
> to 0, for document reason.
>

The initialization is redundant, and your reason is not reasonable.

> Signed-off-by: Jianyu Zhan <[email protected]>

nack

> ---
> net/core/netclassid_cgroup.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/net/core/netclassid_cgroup.c b/net/core/netclassid_cgroup.c
> index 22931e1..1b07dca 100644
> --- a/net/core/netclassid_cgroup.c
> +++ b/net/core/netclassid_cgroup.c
> @@ -108,4 +108,5 @@ struct cgroup_subsys net_cls_cgrp_subsys = {
> .css_free = cgrp_css_free,
> .attach = cgrp_attach,
> .base_cftypes = ss_files,
> + .early_init = 0,
> };
>

2014-04-22 11:03:36

by Neil Horman

Subject: Re: [PATCH] netclassid_cgroup: explicitly init the early_init field

On Tue, Apr 22, 2014 at 01:31:16PM +0800, Jianyu Zhan wrote:
> For a cgroup subsystem who should init early, then it should carefully
> take care of the implementation of css_alloc, because it will be called
> before mm_init() setup the world.
>
> Luckily we don't, and we better explicitly assign the early_init field
> to 0, for document reason.
>
> Signed-off-by: Jianyu Zhan <[email protected]>
> ---
> net/core/netclassid_cgroup.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/net/core/netclassid_cgroup.c b/net/core/netclassid_cgroup.c
> index 22931e1..1b07dca 100644
> --- a/net/core/netclassid_cgroup.c
> +++ b/net/core/netclassid_cgroup.c
> @@ -108,4 +108,5 @@ struct cgroup_subsys net_cls_cgrp_subsys = {
> .css_free = cgrp_css_free,
> .attach = cgrp_attach,
> .base_cftypes = ss_files,
> + .early_init = 0,
> };
> --
> 2.0.0-rc0
>
>
Agreed, this doesn't make sense, early_init is default assigned to zero. What
problem are you trying to solve?
Neil