lkml.org 
[lkml]   [2015]   [Mar]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH] nfs: kstrdup() memory handling
On Sat, Mar 21, 2015 at 01:07:28PM -0400, Sanidhya Kashyap wrote:
> index 86d6214..08bf1f1 100644
> --- a/fs/nfs/nfs4client.c
> +++ b/fs/nfs/nfs4client.c
> @@ -1217,8 +1217,15 @@ int nfs4_update_server(struct nfs_server *server, const char *hostname,
> goto out;
> }
>
> - if (server->nfs_client->cl_hostname == NULL)
> + if (server->nfs_client->cl_hostname == NULL) {
> server->nfs_client->cl_hostname = kstrdup(hostname, GFP_KERNEL);
> + if (!server->nfs_client->cl_hostname) {
> + error = -ENOMEM;
> + dprintk("<-- %s(): not enough memory %d\n",
> + __func__, error);
> + goto out;
> + }
> + }
> nfs_server_insert_lists(server);

The function alays ensures to call nfs_server_insert_lists on each exit
path, and you add one that doesn't.

>
> error = nfs_probe_destination(server);
> diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
> index 627f37c..2d8b408 100644
> --- a/fs/nfs/nfs4proc.c
> +++ b/fs/nfs/nfs4proc.c
> @@ -4974,6 +4974,9 @@ nfs4_init_nonuniform_client_string(struct nfs_client *clp,
> RPC_DISPLAY_PROTO));
> rcu_read_unlock();
> clp->cl_owner_id = kstrdup(buf, GFP_KERNEL);
> + if (!clp->cl_owner_id)
> + return -ENOMEM;
> +
> return result;
> }
>
> @@ -4998,6 +5001,9 @@ nfs4_init_uniform_client_string(struct nfs_client *clp,
> clp->rpc_ops->version, clp->cl_minorversion,
> nodename);
> clp->cl_owner_id = kstrdup(buf, GFP_KERNEL);
> + if (!clp->cl_owner_id)
> + return -ENOMEM;
> +

Both these use a scnprintf + kstrdup pattern thast should be replaced by
kasprintf. But I still don't understand why both function get away with
just returning the caller supplied buf in one case, but need a dynamic
allocation in the other case.


\
 
 \ /
  Last update: 2015-03-25 10:01    [W:0.027 / U:0.852 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site