2007-11-02 17:05:39

by Latchesar Ionkov

[permalink] [raw]
Subject: [PATCH 7/10] 9p: sysfs support for in-kenel servers

Sysfs support for 9P servers.
Every server type is represented as a directory in /sys/fs/9p/srv. Initially
there is a single file in the directory -- 'clone'. Reading from the clone
file creates a new instance of the file server and returns its id. Each
instance is represented as a directory in /sys/fs/9p/srv/<file server>/. The
file server instance can be controlled by the 'ctl' file in its directory.
Currently the ctl file accepts the following commands:

listen-add <trans> <options>
listen-del <trans> <options>
destroy

Example:
echo 'listen-add tcp port=564' > /sys/fs/9p/srv/ramfs/0/ctl

Signed-off-by: Latchesar Ionkov <[email protected]>

---
include/net/9p/srv.h | 24 +++-
net/9p/srv.c | 386 +++++++++++++++++++++++++++++++++++++++++++++++++-
2 files changed, 405 insertions(+), 5 deletions(-)

diff --git a/include/net/9p/srv.h b/include/net/9p/srv.h
index 72d011e..a81ae8f 100644
--- a/include/net/9p/srv.h
+++ b/include/net/9p/srv.h
@@ -69,11 +69,17 @@ struct p9srv {
void (*wstat)(struct p9srv_req *);

/* implementation specific */
+ int id;
atomic_t refcount;
spinlock_t lock; /* covers all srv fields */
+ unsigned long status;
struct list_head conn_list; /* all connections for srv */
struct list_head req_list; /* requests not yet worked on */
struct list_head workreq_list; /* requests being worked on */
+ struct kobject kobj;
+ struct list_head srv_list; /* all servers of a type */
+ struct list_head lis_list; /* all listeners for the srv */
+ struct work_struct shutwork;
};

struct p9srv_conn {
@@ -90,6 +96,16 @@ struct p9srv_conn {
wait_queue_head_t reset_wqueue;
};

+struct p9srv_type {
+ spinlock_t lock;
+ char *name;
+ int nextid;
+ struct kobject kobj;
+ struct list_head srv_list; /* all servers of a type */
+ struct list_head stype_list; /* all server types */
+ struct p9srv* (*create)(void);
+};
+
enum {
/* connection status values */
Reset = 1, /* resetting */
@@ -149,11 +165,11 @@ extern char *Enotfound;
extern char *Eopen;
extern char *Eexist;
extern char *Enotempty;
-
+extern struct kset p9srv_subsys;

struct p9srv *p9srv_srv_create(void);
void p9srv_srv_stop(struct p9srv *srv);
-void p9srv_srv_destroy(struct p9srv *srv);
+void p9srv_srv_shutdown(struct p9srv *srv);
void p9srv_srv_start(struct p9srv *srv);
void p9srv_respond(struct p9srv_req *req, struct p9_fcall *rc);
void p9srv_respond_error(struct p9srv_req *req, char *ename, int ecode);
@@ -165,5 +181,9 @@ void p9srv_conn_destroy(struct p9srv_conn *conn);
struct p9srv_fid *p9srv_fid_find(struct p9srv_fidpool *fp, u32 fid);
void p9srv_fid_incref(struct p9srv_fid *fid);
void p9srv_fid_decref(struct p9srv_fid *fid);
+struct p9srv_type *p9srv_type_register(char *name, struct p9srv *(*create)(void));
+void p9srv_type_unregister(char *name);
+int p9srv_listener_add(struct p9srv *, char *, char *);
+int p9srv_listener_del(struct p9srv *, char *, char *);

#endif
diff --git a/net/9p/srv.c b/net/9p/srv.c
index 855ac70..e0c359b 100644
--- a/net/9p/srv.c
+++ b/net/9p/srv.c
@@ -37,6 +37,18 @@ struct p9srv_fidpool {
struct hlist_head hash[FID_HTABLE_SIZE];
};

+struct p9srv_listener {
+ struct p9_trans_listener *listener;
+ char *name;
+ char *options;
+ struct list_head lis_list;
+};
+
+enum {
+ /* p9srv status */
+ Shutdown = 1,
+};
+
enum {
/* p9srv_req status */
Respond = 1,
@@ -45,6 +57,8 @@ enum {
};

struct workqueue_struct *p9srv_wq;
+static DEFINE_SPINLOCK(p9srv_type_lock);
+static LIST_HEAD(p9srv_type_list);

char *Eunknownfid = "unknown fid";
EXPORT_SYMBOL(Eunknownfid);
@@ -78,9 +92,12 @@ char *Eexist = "file or directory already exists";
EXPORT_SYMBOL(Eexist);
char *Enotempty = "directory not empty";
EXPORT_SYMBOL(Enotempty);
+decl_subsys_name(p9srv, srv, NULL, NULL);
+EXPORT_SYMBOL(p9srv_subsys);

static void p9srv_srv_ref(struct p9srv *srv);
static void p9srv_srv_unref(struct p9srv *srv);
+static void p9srv_shut_work(struct work_struct *work);
static void p9srv_conn_reset(struct p9srv_conn *conn, struct p9srv_req *vreq);
static void p9srv_conn_respond(struct p9srv_req *req);
static struct p9srv_fidpool *p9srv_fidpool_create(void);
@@ -109,6 +126,14 @@ static void p9srv_stat(struct p9srv_req *);
static void p9srv_wstat(struct p9srv_req *);
static void p9srv_preq_work(struct work_struct *work);
static void p9srv_conn_request(struct p9_trans *, struct p9_trans_req *req);
+static void p9srv_type_release(struct kobject *);
+static ssize_t p9srv_type_show(struct kobject *, struct attribute *, char *);
+static ssize_t p9srv_type_store(struct kobject *, struct attribute *,
+ const char *, size_t);
+static void p9srv_srv_release(struct kobject *);
+static ssize_t p9srv_srv_show(struct kobject *, struct attribute *, char *);
+static ssize_t p9srv_srv_store(struct kobject *, struct attribute *,
+ const char *, size_t);

static void (*p9srv_fcall[])(struct p9srv_req *) = {
p9srv_version, /* Tversion */
@@ -144,6 +169,36 @@ static void (*p9srv_fcall_post[])(struct p9srv_req *) = {
NULL, /* Twstat */
};

+static struct sysfs_ops p9srv_type_ops = {
+ .show = p9srv_type_show,
+ .store = p9srv_type_store,
+};
+
+static struct kobj_type p9srv_type_ktype = {
+ .release = p9srv_type_release,
+ .sysfs_ops = &p9srv_type_ops,
+};
+
+static struct attribute p9srv_clone_attr = {
+ .name = "clone",
+ .mode = 0444,
+};
+
+static struct sysfs_ops p9srv_srv_ops = {
+ .show = p9srv_srv_show,
+ .store = p9srv_srv_store,
+};
+
+static struct kobj_type p9srv_srv_ktype = {
+ .release = p9srv_srv_release,
+ .sysfs_ops = &p9srv_srv_ops,
+};
+
+static struct attribute p9srv_ctl_attr = {
+ .name = "ctl",
+ .mode = 0644,
+};
+
struct p9srv *p9srv_srv_create(void)
{
struct p9srv *srv;
@@ -156,10 +211,14 @@ struct p9srv *p9srv_srv_create(void)

memset(srv, 0, sizeof(struct p9srv));
spin_lock_init(&srv->lock);
+ srv->id = -1;
atomic_set(&srv->refcount, 1);
+ srv->status = 0;
INIT_LIST_HEAD(&srv->conn_list);
INIT_LIST_HEAD(&srv->req_list);
INIT_LIST_HEAD(&srv->workreq_list);
+ INIT_LIST_HEAD(&srv->lis_list);
+ INIT_WORK(&srv->shutwork, p9srv_shut_work);

/* the values below may be overwritten by the creator */
srv->msize = 65536 + P9_IOHDRSZ;
@@ -171,12 +230,32 @@ struct p9srv *p9srv_srv_create(void)
}
EXPORT_SYMBOL(p9srv_srv_create);

-void p9srv_srv_destroy(struct p9srv *srv)
+void p9srv_srv_shutdown(struct p9srv *srv)
{
- P9_DPRINTK(P9SRV_DEBUG_SRV, "%p\n", srv);
+ struct p9srv_listener *slis, *stmp;
+ LIST_HEAD(lis_list);
+
+ if (test_and_set_bit(Shutdown, &srv->status))
+ return;
+
+ P9_DPRINTK(P9SRV_DEBUG_SRV, "srv %p\n", srv);
+ spin_lock(&srv->lock);
+ list_for_each_entry_safe(slis, stmp, &srv->lis_list, lis_list) {
+ list_move(&slis->lis_list, &lis_list);
+ }
+ spin_unlock(&srv->lock);
+
+ list_for_each_entry_safe(slis, stmp, &lis_list, lis_list) {
+ list_del(&slis->lis_list);
+ (*slis->listener->destroy)(slis->listener);
+ kfree(slis);
+ }
+
+ sysfs_remove_file(&srv->kobj, &p9srv_ctl_attr);
+ kobject_unregister(&srv->kobj);
p9srv_srv_unref(srv);
}
-EXPORT_SYMBOL(p9srv_srv_destroy);
+EXPORT_SYMBOL(p9srv_srv_shutdown);

void p9srv_srv_ref(struct p9srv *srv)
{
@@ -200,6 +279,14 @@ void p9srv_srv_unref(struct p9srv *srv)
}
}

+static void p9srv_shut_work(struct work_struct *work)
+{
+ struct p9srv *srv;
+
+ srv = container_of(work, struct p9srv, shutwork);
+ p9srv_srv_shutdown(srv);
+}
+
void p9srv_srv_start(struct p9srv *srv)
{
P9_DPRINTK(P9SRV_DEBUG_SRV, "%p\n", srv);
@@ -366,6 +453,289 @@ void p9srv_req_unref(struct p9srv_req *req)
}
EXPORT_SYMBOL(p9srv_req_unref);

+struct p9srv_type *p9srv_type_register(char *name,
+ struct p9srv *(*create)(void))
+{
+ int err;
+ struct p9srv_type *ret, *stype;
+
+ ret = kzalloc(sizeof(*ret), GFP_KERNEL);
+ if (!ret)
+ return ERR_PTR(-ENOMEM);
+
+ spin_lock_init(&ret->lock);
+ ret->name = name;
+ ret->nextid = 0;
+ INIT_LIST_HEAD(&ret->srv_list);
+ INIT_LIST_HEAD(&ret->stype_list);
+ ret->create = create;
+
+ spin_lock(&p9srv_type_lock);
+ list_for_each_entry(stype, &p9srv_type_list, stype_list) {
+ if (strcmp(name, stype->name) == 0) {
+ spin_unlock(&p9srv_type_lock);
+ kfree(ret);
+ return ERR_PTR(-EEXIST);
+ }
+ }
+
+ list_add_tail(&ret->stype_list, &p9srv_type_list);
+ spin_unlock(&p9srv_type_lock);
+
+ kobject_init(&ret->kobj);
+ kobject_set_name(&ret->kobj, "%s", name);
+ ret->kobj.parent = &p9srv_subsys.kobj;
+ ret->kobj.ktype = &p9srv_type_ktype;
+ err = kobject_add(&ret->kobj);
+ if (err)
+ goto error;
+
+ err = sysfs_create_file(&ret->kobj, &p9srv_clone_attr);
+ if (err)
+ goto error;
+
+ return ret;
+
+error:
+ kobject_unregister(&ret->kobj);
+ /* kfree ??? */
+ return ERR_PTR(err);
+}
+EXPORT_SYMBOL(p9srv_type_register);
+
+void p9srv_type_unregister(char *name)
+{
+ struct p9srv_type *stype;
+
+ spin_lock(&p9srv_type_lock);
+ list_for_each_entry(stype, &p9srv_type_list, stype_list) {
+ if (strcmp(name, stype->name) == 0) {
+ list_del(&stype->stype_list);
+ spin_unlock(&p9srv_type_lock);
+ sysfs_remove_file(&stype->kobj, &p9srv_clone_attr);
+ kobject_unregister(&stype->kobj);
+ /* kfree ??? */
+ return;
+ }
+ }
+ spin_unlock(&p9srv_type_lock);
+}
+EXPORT_SYMBOL(p9srv_type_unregister);
+
+static void p9srv_type_release(struct kobject *kobj)
+{
+}
+
+static ssize_t p9srv_type_show(struct kobject *kobj, struct attribute *attr,
+ char *buf)
+{
+ int err;
+ struct p9srv_type *stype;
+ struct p9srv *srv;
+
+ stype = container_of(kobj, struct p9srv_type, kobj);
+ srv = (*stype->create)();
+ if (IS_ERR(srv))
+ return PTR_ERR(srv);
+
+ spin_lock(&stype->lock);
+ srv->id = stype->nextid++;
+ list_add_tail(&srv->srv_list, &stype->srv_list);
+ spin_unlock(&stype->lock);
+
+ kobject_init(&srv->kobj);
+ kobject_set_name(&srv->kobj, "%d", srv->id);
+ srv->kobj.parent = &stype->kobj;
+ srv->kobj.ktype = &p9srv_srv_ktype;
+ err = kobject_add(&srv->kobj);
+ if (err)
+ goto error;
+
+ err = sysfs_create_file(&srv->kobj, &p9srv_ctl_attr);
+ if (err)
+ goto error;
+
+ return snprintf(buf, PAGE_SIZE, "%d", srv->id);
+
+error:
+ spin_lock(&stype->lock);
+ list_del(&srv->srv_list);
+ spin_unlock(&stype->lock);
+ kobject_unregister(&srv->kobj);
+ /* kfree ??? */
+ return err;
+}
+
+static ssize_t p9srv_type_store(struct kobject *kobj, struct attribute *attr,
+ const char *buf, size_t buflen)
+{
+ return -EPERM;
+}
+
+static void p9srv_srv_release(struct kobject *kobj)
+{
+}
+
+static ssize_t p9srv_srv_show(struct kobject *kobj, struct attribute *attr,
+ char *buf)
+{
+ struct p9srv *srv;
+
+ srv = container_of(kobj, struct p9srv, kobj);
+ return snprintf(buf, PAGE_SIZE, "%d\n", srv->id);
+}
+
+static ssize_t p9srv_srv_store(struct kobject *kobj, struct attribute *attr,
+ const char *buf, size_t buflen)
+{
+ int n, err;
+ char *args[7];
+ char *p, *ep;
+ struct p9srv *srv;
+
+ srv = container_of(kobj, struct p9srv, kobj);
+ if (test_bit(Shutdown, &srv->status))
+ return -ENOENT;
+
+ p = (char *) buf;
+ ep = (char *) buf + buflen - 1;
+ while (*ep == '\n')
+ ep--;
+
+ ep++;
+ *ep = '\0';
+ P9_DPRINTK(P9SRV_DEBUG_SRV, "buf '%s'\n", p);
+ for(n = 0; n < ARRAY_SIZE(args) && p < ep; n++) {
+ args[n] = p;
+ p = strchr(p, ' ');
+ if (!p)
+ break;
+
+ *p = '\0';
+ p++;
+ }
+
+ while (*args[n] == '\0')
+ n--;
+
+ n++;
+ if (n == 0)
+ return -EINVAL;
+
+ if (strcmp(args[0], "listen-add") == 0) {
+ if (n == 2)
+ args[2] = NULL;
+ else if (n < 2 || n > 3)
+ return -EINVAL;
+
+ err = p9srv_listener_add(srv, args[1], args[2]);
+ if (err)
+ return err;
+ } else if (strcmp(args[0], "listen-del") == 0) {
+ if (n == 2)
+ args[2] = NULL;
+ else if (n < 2 || n > 3)
+ return -EINVAL;
+
+ err = p9srv_listener_del(srv, args[1], args[2]);
+ if (err)
+ return err;
+ } else if (strcmp(args[0], "shutdown") == 0) {
+ /* we can't call p9srv_shutdown directly because
+ kobject_unregister will wait for the kobject
+ that we hold to be freed (and that won't happen) */
+ queue_work(p9srv_wq, &srv->shutwork);
+ } else {
+ return -EINVAL;
+ }
+
+ return buflen;
+}
+
+static void p9srv_newtrans(struct p9_trans_listener *l, struct p9_trans *trans)
+{
+ struct p9srv *srv;
+
+ srv = l->aux;
+ if (!trans) {
+ P9_DPRINTK(P9SRV_DEBUG_SRV, "listener err %d\n", l->err);
+ return;
+ }
+
+ if (test_bit(Shutdown, &srv->status)) {
+ (*trans->destroy)(trans);
+ return;
+ }
+
+ p9srv_conn_create(srv, trans);
+}
+
+
+int p9srv_listener_add(struct p9srv *srv, char *lname, char *options)
+{
+ substring_t tname;
+ struct p9_trans_module *tmod;
+ struct p9_trans_listener *lis;
+ struct p9srv_listener *slis;
+
+ P9_DPRINTK(P9SRV_DEBUG_SRV, "listener '%s' options '%s'\n",
+ lname, options);
+ tname.from = lname;
+ tname.to = lname + strlen(lname);
+ tmod = v9fs_match_trans(&tname);
+ if (!tmod)
+ return -ENOENT;
+
+ lis = tmod->create_listener(options);
+ if (IS_ERR(lis))
+ return PTR_ERR(lis);
+
+ lis->newtrans = p9srv_newtrans;
+ lis->aux = srv;
+
+ slis = kmalloc(sizeof(*slis) + strlen(lname) + strlen(options) + 2,
+ GFP_KERNEL);
+ if (!slis) {
+ (*lis->destroy)(lis);
+ return -ENOMEM;
+ }
+
+ slis->listener = lis;
+ slis->name = (char *) slis + sizeof(*slis);
+ slis->options = slis->name + strlen(lname) + 1;
+ strcpy(slis->name, lname);
+ strcpy(slis->options, options);
+ INIT_LIST_HEAD(&slis->lis_list);
+ spin_lock(&srv->lock);
+ list_add_tail(&slis->lis_list, &srv->lis_list);
+ spin_unlock(&srv->lock);
+
+ return 0;
+}
+EXPORT_SYMBOL(p9srv_listener_add);
+
+int p9srv_listener_del(struct p9srv *srv, char *lname, char *options)
+{
+ struct p9srv_listener *slis;
+
+ P9_DPRINTK(P9SRV_DEBUG_SRV, "listener '%s' options '%s'\n",
+ lname, options);
+ spin_lock(&srv->lock);
+ list_for_each_entry(slis, &srv->lis_list, lis_list) {
+ if (!strcmp(lname, slis->name) &&
+ !strcmp(options, slis->options)) {
+ list_del(&slis->lis_list);
+ spin_unlock(&srv->lock);
+ (*slis->listener->destroy)(slis->listener);
+ kfree(slis);
+ return 0;
+ }
+ }
+
+ return -ENOENT;
+}
+EXPORT_SYMBOL(p9srv_listener_del);
+
static void p9srv_version(struct p9srv_req *req)
{
int msize;
@@ -1258,17 +1628,27 @@ void p9srv_conn_respond(struct p9srv_req *req)

static int __init p9srv_init(void)
{
+ int err;
+
p9srv_wq = create_workqueue("9psrv");
if (!p9srv_wq) {
printk(KERN_WARNING "9psrv: creating workqueue failed\n");
return -ENOMEM;
}

+ kobj_set_kset_s(&p9srv_subsys, p9_subsys);
+ err = subsystem_register(&p9srv_subsys);
+ if (err) {
+ destroy_workqueue(p9srv_wq);
+ return err;
+ }
+
return 0;
}

static void __exit p9srv_exit(void)
{
+ subsystem_unregister(&p9srv_subsys);
destroy_workqueue(p9srv_wq);
p9srv_wq = NULL;
}


2007-11-03 03:45:46

by Greg KH

[permalink] [raw]
Subject: Re: [PATCH 7/10] 9p: sysfs support for in-kenel servers

On Fri, Nov 02, 2007 at 11:05:27AM -0600, Latchesar Ionkov wrote:
> Sysfs support for 9P servers.
> Every server type is represented as a directory in /sys/fs/9p/srv. Initially
> there is a single file in the directory -- 'clone'. Reading from the clone
> file creates a new instance of the file server and returns its id. Each
> instance is represented as a directory in /sys/fs/9p/srv/<file server>/. The
> file server instance can be controlled by the 'ctl' file in its directory.
> Currently the ctl file accepts the following commands:
>
> listen-add <trans> <options>
> listen-del <trans> <options>
> destroy
>
> Example:
> echo 'listen-add tcp port=564' > /sys/fs/9p/srv/ramfs/0/ctl

Traditionally sysfs has not been used for "configuration" stuff like
this, as that is what configfs is for.

Also, a sysfs file that causes an action to happen just by reading the
file is not a safe thing to have. Lots of scripts have been known to
just walk the whole sysfs tree and open and read everything they can for
no real good reason. So you should seriously reconsider this kind of
interaction.


> +static struct attribute p9srv_clone_attr = {
> + .name = "clone",
> + .mode = 0444,
> +};

What's wrong with the __ATTR macro? Using a "raw" struct attribute
seems a bit odd here.

> +static struct sysfs_ops p9srv_srv_ops = {
> + .show = p9srv_srv_show,
> + .store = p9srv_srv_store,
> +};
> +
> +static struct kobj_type p9srv_srv_ktype = {
> + .release = p9srv_srv_release,
> + .sysfs_ops = &p9srv_srv_ops,
> +};
> +
> +static struct attribute p9srv_ctl_attr = {
> + .name = "ctl",
> + .mode = 0644,
> +};

Same here.

> +struct p9srv_type *p9srv_type_register(char *name,
> + struct p9srv *(*create)(void))
> +{
> + int err;
> + struct p9srv_type *ret, *stype;
> +
> + ret = kzalloc(sizeof(*ret), GFP_KERNEL);
> + if (!ret)
> + return ERR_PTR(-ENOMEM);
> +
> + spin_lock_init(&ret->lock);
> + ret->name = name;
> + ret->nextid = 0;
> + INIT_LIST_HEAD(&ret->srv_list);
> + INIT_LIST_HEAD(&ret->stype_list);
> + ret->create = create;
> +
> + spin_lock(&p9srv_type_lock);
> + list_for_each_entry(stype, &p9srv_type_list, stype_list) {
> + if (strcmp(name, stype->name) == 0) {
> + spin_unlock(&p9srv_type_lock);
> + kfree(ret);
> + return ERR_PTR(-EEXIST);
> + }
> + }
> +
> + list_add_tail(&ret->stype_list, &p9srv_type_list);
> + spin_unlock(&p9srv_type_lock);
> +
> + kobject_init(&ret->kobj);
> + kobject_set_name(&ret->kobj, "%s", name);
> + ret->kobj.parent = &p9srv_subsys.kobj;
> + ret->kobj.ktype = &p9srv_type_ktype;
> + err = kobject_add(&ret->kobj);
> + if (err)
> + goto error;
> +
> + err = sysfs_create_file(&ret->kobj, &p9srv_clone_attr);
> + if (err)
> + goto error;
> +
> + return ret;
> +
> +error:
> + kobject_unregister(&ret->kobj);
> + /* kfree ??? */

Yes, otherwise you just leaked memory :)

> + return ERR_PTR(err);
> +}
> +EXPORT_SYMBOL(p9srv_type_register);
> +
> +void p9srv_type_unregister(char *name)
> +{
> + struct p9srv_type *stype;
> +
> + spin_lock(&p9srv_type_lock);
> + list_for_each_entry(stype, &p9srv_type_list, stype_list) {
> + if (strcmp(name, stype->name) == 0) {
> + list_del(&stype->stype_list);
> + spin_unlock(&p9srv_type_lock);
> + sysfs_remove_file(&stype->kobj, &p9srv_clone_attr);
> + kobject_unregister(&stype->kobj);
> + /* kfree ??? */
> + return;
> + }
> + }
> + spin_unlock(&p9srv_type_lock);
> +}
> +EXPORT_SYMBOL(p9srv_type_unregister);
> +
> +static void p9srv_type_release(struct kobject *kobj)
> +{
> +}

{sigh}

Yes, the kernel complains if you do not have a release function for your
kobject, but did you think that just providing an empty function is
somehow the proper thing to do here?

It wants you to do something in the release function, like free the
memory of the object. Not just provide empty functions to try to shut
it up...

Please fix this.

> +static ssize_t p9srv_type_show(struct kobject *kobj, struct attribute *attr,
> + char *buf)
> +{
> + int err;
> + struct p9srv_type *stype;
> + struct p9srv *srv;
> +
> + stype = container_of(kobj, struct p9srv_type, kobj);
> + srv = (*stype->create)();

Again, creating stuff in a show function is not a good idea.

> + if (IS_ERR(srv))
> + return PTR_ERR(srv);
> +
> + spin_lock(&stype->lock);
> + srv->id = stype->nextid++;
> + list_add_tail(&srv->srv_list, &stype->srv_list);
> + spin_unlock(&stype->lock);
> +
> + kobject_init(&srv->kobj);
> + kobject_set_name(&srv->kobj, "%d", srv->id);
> + srv->kobj.parent = &stype->kobj;
> + srv->kobj.ktype = &p9srv_srv_ktype;
> + err = kobject_add(&srv->kobj);
> + if (err)
> + goto error;
> +
> + err = sysfs_create_file(&srv->kobj, &p9srv_ctl_attr);
> + if (err)
> + goto error;
> +
> + return snprintf(buf, PAGE_SIZE, "%d", srv->id);
> +
> +error:
> + spin_lock(&stype->lock);
> + list_del(&srv->srv_list);
> + spin_unlock(&stype->lock);
> + kobject_unregister(&srv->kobj);
> + /* kfree ??? */

Well, no, not if your release function frees the memory, otherwise it is
still present the way the code is written today...

But watch out, you are going here if kobject_add() fails, in which case,
kobject_unregister makes no sense as the kobject was never properly
added.

> +static ssize_t p9srv_type_store(struct kobject *kobj, struct attribute *attr,
> + const char *buf, size_t buflen)
> +{
> + return -EPERM;
> +}
> +
> +static void p9srv_srv_release(struct kobject *kobj)
> +{
> +}

Again with the empty release :(


> static int __init p9srv_init(void)
> {
> + int err;
> +
> p9srv_wq = create_workqueue("9psrv");
> if (!p9srv_wq) {
> printk(KERN_WARNING "9psrv: creating workqueue failed\n");
> return -ENOMEM;
> }
>
> + kobj_set_kset_s(&p9srv_subsys, p9_subsys);

This macro is now gone away in the next -mm release :)

thanks,

greg k-h

2007-11-03 10:33:20

by Peter Zijlstra

[permalink] [raw]
Subject: Re: [PATCH 7/10] 9p: sysfs support for in-kenel servers

On Fri, 2007-11-02 at 20:48 -0700, Greg KH wrote:

> Also, a sysfs file that causes an action to happen just by reading the
> file is not a safe thing to have. Lots of scripts have been known to
> just walk the whole sysfs tree and open and read everything they can for
> no real good reason. So you should seriously reconsider this kind of
> interaction.

Christoph, doesn't SLUB do exactly that?

2007-11-03 18:22:27

by Christoph Lameter

[permalink] [raw]
Subject: Re: [PATCH 7/10] 9p: sysfs support for in-kenel servers

On Sat, 3 Nov 2007, Peter Zijlstra wrote:

> On Fri, 2007-11-02 at 20:48 -0700, Greg KH wrote:
>
> > Also, a sysfs file that causes an action to happen just by reading the
> > file is not a safe thing to have. Lots of scripts have been known to
> > just walk the whole sysfs tree and open and read everything they can for
> > no real good reason. So you should seriously reconsider this kind of
> > interaction.
>
> Christoph, doesn't SLUB do exactly that?

SLUB gathers statistics and summaries and displays them on read.