linux/drivers/xen/xenbus/xenbus_dev_frontend.c

687 lines
16 KiB
C
Raw Normal View History

/*
* Driver giving user-space access to the kernel's xenbus connection
* to xenstore.
*
* Copyright (c) 2005, Christian Limpach
* Copyright (c) 2005, Rusty Russell, IBM Corporation
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License version 2
* as published by the Free Software Foundation; or, when distributed
* separately from the Linux kernel or incorporated into other
* software packages, subject to the following license:
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this source file (the "Software"), to deal in the Software without
* restriction, including without limitation the rights to use, copy, modify,
* merge, publish, distribute, sublicense, and/or sell copies of the Software,
* and to permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*
* Changes:
* 2008-10-07 Alex Zeffertt Replaced /proc/xen/xenbus with xenfs filesystem
* and /proc/xen compatibility mount point.
* Turned xenfs into a loadable module.
*/
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/uio.h>
#include <linux/notifier.h>
#include <linux/wait.h>
#include <linux/fs.h>
#include <linux/poll.h>
#include <linux/mutex.h>
#include <linux/sched.h>
#include <linux/spinlock.h>
#include <linux/mount.h>
#include <linux/pagemap.h>
#include <linux/uaccess.h>
#include <linux/init.h>
#include <linux/namei.h>
#include <linux/string.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
#include <linux/slab.h>
#include <linux/miscdevice.h>
#include <xen/xenbus.h>
#include <xen/xen.h>
#include <asm/xen/hypervisor.h>
#include "xenbus.h"
/*
* An element of a list of outstanding transactions, for which we're
* still waiting a reply.
*/
struct xenbus_transaction_holder {
struct list_head list;
struct xenbus_transaction handle;
};
/*
* A buffer of data on the queue.
*/
struct read_buffer {
struct list_head list;
unsigned int cons;
unsigned int len;
char msg[];
};
struct xenbus_file_priv {
/*
* msgbuffer_mutex is held while partial requests are built up
* and complete requests are acted on. It therefore protects
* the "transactions" and "watches" lists, and the partial
* request length and buffer.
*
* reply_mutex protects the reply being built up to return to
* usermode. It nests inside msgbuffer_mutex but may be held
* alone during a watch callback.
*/
struct mutex msgbuffer_mutex;
/* In-progress transactions */
struct list_head transactions;
/* Active watches. */
struct list_head watches;
/* Partial request. */
unsigned int len;
union {
struct xsd_sockmsg msg;
char buffer[XENSTORE_PAYLOAD_MAX];
} u;
/* Response queue. */
struct mutex reply_mutex;
struct list_head read_buffers;
wait_queue_head_t read_waitq;
xen: optimize xenbus driver for multiple concurrent xenstore accesses Handling of multiple concurrent Xenstore accesses through xenbus driver either from the kernel or user land is rather lame today: xenbus is capable to have one access active only at one point of time. Rewrite xenbus to handle multiple requests concurrently by making use of the request id of the Xenstore protocol. This requires to: - Instead of blocking inside xb_read() when trying to read data from the xenstore ring buffer do so only in the main loop of xenbus_thread(). - Instead of doing writes to the xenstore ring buffer in the context of the caller just queue the request and do the write in the dedicated xenbus thread. - Instead of just forwarding the request id specified by the caller of xenbus to xenstore use a xenbus internal unique request id. This will allow multiple outstanding requests. - Modify the locking scheme in order to allow multiple requests being active in parallel. - Instead of waiting for the reply of a user's xenstore request after writing the request to the xenstore ring buffer return directly to the caller and do the waiting in the read path. Additionally signal handling was optimized by avoiding waking up the xenbus thread or sending an event to Xenstore in case the addressed entity is known to be running already. As a result communication with Xenstore is sped up by a factor of up to 5: depending on the request type (read or write) and the amount of data transferred the gain was at least 20% (small reads) and went up to a factor of 5 for large writes. In the end some more rough edges of xenbus have been smoothed: - Handling of memory shortage when reading from xenstore ring buffer in the xenbus driver was not optimal: it was busy looping and issuing a warning in each loop. - In case of xenstore not running in dom0 but in a stubdom we end up with two xenbus threads running as the initialization of xenbus in dom0 expecting a local xenstored will be redone later when connecting to the xenstore domain. Up to now this was no problem as locking would prevent the two xenbus threads interfering with each other, but this was just a waste of kernel resources. - An out of memory situation while writing to or reading from the xenstore ring buffer no longer will lead to a possible loss of synchronization with xenstore. - The user read and write part are now interruptible by signals. Signed-off-by: Juergen Gross <jgross@suse.com> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
2017-02-09 21:39:58 +08:00
struct kref kref;
};
/* Read out any raw xenbus messages queued up. */
static ssize_t xenbus_file_read(struct file *filp,
char __user *ubuf,
size_t len, loff_t *ppos)
{
struct xenbus_file_priv *u = filp->private_data;
struct read_buffer *rb;
unsigned i;
int ret;
mutex_lock(&u->reply_mutex);
again:
while (list_empty(&u->read_buffers)) {
mutex_unlock(&u->reply_mutex);
if (filp->f_flags & O_NONBLOCK)
return -EAGAIN;
ret = wait_event_interruptible(u->read_waitq,
!list_empty(&u->read_buffers));
if (ret)
return ret;
mutex_lock(&u->reply_mutex);
}
rb = list_entry(u->read_buffers.next, struct read_buffer, list);
i = 0;
while (i < len) {
unsigned sz = min((unsigned)len - i, rb->len - rb->cons);
ret = copy_to_user(ubuf + i, &rb->msg[rb->cons], sz);
i += sz - ret;
rb->cons += sz - ret;
if (ret != 0) {
if (i == 0)
i = -EFAULT;
goto out;
}
/* Clear out buffer if it has been consumed */
if (rb->cons == rb->len) {
list_del(&rb->list);
kfree(rb);
if (list_empty(&u->read_buffers))
break;
rb = list_entry(u->read_buffers.next,
struct read_buffer, list);
}
}
if (i == 0)
goto again;
out:
mutex_unlock(&u->reply_mutex);
return i;
}
/*
* Add a buffer to the queue. Caller must hold the appropriate lock
* if the queue is not local. (Commonly the caller will build up
* multiple queued buffers on a temporary local list, and then add it
* to the appropriate list under lock once all the buffers have een
* successfully allocated.)
*/
static int queue_reply(struct list_head *queue, const void *data, size_t len)
{
struct read_buffer *rb;
if (len == 0)
return 0;
if (len > XENSTORE_PAYLOAD_MAX)
return -EINVAL;
rb = kmalloc(sizeof(*rb) + len, GFP_KERNEL);
if (rb == NULL)
return -ENOMEM;
rb->cons = 0;
rb->len = len;
memcpy(rb->msg, data, len);
list_add_tail(&rb->list, queue);
return 0;
}
/*
* Free all the read_buffer s on a list.
* Caller must have sole reference to list.
*/
static void queue_cleanup(struct list_head *list)
{
struct read_buffer *rb;
while (!list_empty(list)) {
rb = list_entry(list->next, struct read_buffer, list);
list_del(list->next);
kfree(rb);
}
}
struct watch_adapter {
struct list_head list;
struct xenbus_watch watch;
struct xenbus_file_priv *dev_data;
char *token;
};
static void free_watch_adapter(struct watch_adapter *watch)
{
kfree(watch->watch.node);
kfree(watch->token);
kfree(watch);
}
static struct watch_adapter *alloc_watch_adapter(const char *path,
const char *token)
{
struct watch_adapter *watch;
watch = kzalloc(sizeof(*watch), GFP_KERNEL);
if (watch == NULL)
goto out_fail;
watch->watch.node = kstrdup(path, GFP_KERNEL);
if (watch->watch.node == NULL)
goto out_free;
watch->token = kstrdup(token, GFP_KERNEL);
if (watch->token == NULL)
goto out_free;
return watch;
out_free:
free_watch_adapter(watch);
out_fail:
return NULL;
}
static void watch_fired(struct xenbus_watch *watch,
const char *path,
const char *token)
{
struct watch_adapter *adap;
struct xsd_sockmsg hdr;
const char *token_caller;
int path_len, tok_len, body_len;
int ret;
LIST_HEAD(staging_q);
adap = container_of(watch, struct watch_adapter, watch);
token_caller = adap->token;
path_len = strlen(path) + 1;
tok_len = strlen(token_caller) + 1;
body_len = path_len + tok_len;
hdr.type = XS_WATCH_EVENT;
hdr.len = body_len;
mutex_lock(&adap->dev_data->reply_mutex);
ret = queue_reply(&staging_q, &hdr, sizeof(hdr));
if (!ret)
ret = queue_reply(&staging_q, path, path_len);
if (!ret)
ret = queue_reply(&staging_q, token_caller, tok_len);
if (!ret) {
/* success: pass reply list onto watcher */
list_splice_tail(&staging_q, &adap->dev_data->read_buffers);
wake_up(&adap->dev_data->read_waitq);
} else
queue_cleanup(&staging_q);
mutex_unlock(&adap->dev_data->reply_mutex);
}
xen: optimize xenbus driver for multiple concurrent xenstore accesses Handling of multiple concurrent Xenstore accesses through xenbus driver either from the kernel or user land is rather lame today: xenbus is capable to have one access active only at one point of time. Rewrite xenbus to handle multiple requests concurrently by making use of the request id of the Xenstore protocol. This requires to: - Instead of blocking inside xb_read() when trying to read data from the xenstore ring buffer do so only in the main loop of xenbus_thread(). - Instead of doing writes to the xenstore ring buffer in the context of the caller just queue the request and do the write in the dedicated xenbus thread. - Instead of just forwarding the request id specified by the caller of xenbus to xenstore use a xenbus internal unique request id. This will allow multiple outstanding requests. - Modify the locking scheme in order to allow multiple requests being active in parallel. - Instead of waiting for the reply of a user's xenstore request after writing the request to the xenstore ring buffer return directly to the caller and do the waiting in the read path. Additionally signal handling was optimized by avoiding waking up the xenbus thread or sending an event to Xenstore in case the addressed entity is known to be running already. As a result communication with Xenstore is sped up by a factor of up to 5: depending on the request type (read or write) and the amount of data transferred the gain was at least 20% (small reads) and went up to a factor of 5 for large writes. In the end some more rough edges of xenbus have been smoothed: - Handling of memory shortage when reading from xenstore ring buffer in the xenbus driver was not optimal: it was busy looping and issuing a warning in each loop. - In case of xenstore not running in dom0 but in a stubdom we end up with two xenbus threads running as the initialization of xenbus in dom0 expecting a local xenstored will be redone later when connecting to the xenstore domain. Up to now this was no problem as locking would prevent the two xenbus threads interfering with each other, but this was just a waste of kernel resources. - An out of memory situation while writing to or reading from the xenstore ring buffer no longer will lead to a possible loss of synchronization with xenstore. - The user read and write part are now interruptible by signals. Signed-off-by: Juergen Gross <jgross@suse.com> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
2017-02-09 21:39:58 +08:00
static void xenbus_file_free(struct kref *kref)
{
struct xenbus_file_priv *u;
struct xenbus_transaction_holder *trans, *tmp;
struct watch_adapter *watch, *tmp_watch;
struct read_buffer *rb, *tmp_rb;
u = container_of(kref, struct xenbus_file_priv, kref);
/*
* No need for locking here because there are no other users,
* by definition.
*/
list_for_each_entry_safe(trans, tmp, &u->transactions, list) {
xenbus_transaction_end(trans->handle, 1);
list_del(&trans->list);
kfree(trans);
}
list_for_each_entry_safe(watch, tmp_watch, &u->watches, list) {
unregister_xenbus_watch(&watch->watch);
list_del(&watch->list);
free_watch_adapter(watch);
}
list_for_each_entry_safe(rb, tmp_rb, &u->read_buffers, list) {
list_del(&rb->list);
kfree(rb);
}
kfree(u);
}
static struct xenbus_transaction_holder *xenbus_get_transaction(
struct xenbus_file_priv *u, uint32_t tx_id)
{
struct xenbus_transaction_holder *trans;
list_for_each_entry(trans, &u->transactions, list)
if (trans->handle.id == tx_id)
return trans;
return NULL;
}
void xenbus_dev_queue_reply(struct xb_req_data *req)
{
struct xenbus_file_priv *u = req->par;
struct xenbus_transaction_holder *trans = NULL;
int rc;
LIST_HEAD(staging_q);
xs_request_exit(req);
mutex_lock(&u->msgbuffer_mutex);
if (req->type == XS_TRANSACTION_START) {
trans = xenbus_get_transaction(u, 0);
if (WARN_ON(!trans))
goto out;
if (req->msg.type == XS_ERROR) {
list_del(&trans->list);
kfree(trans);
} else {
rc = kstrtou32(req->body, 10, &trans->handle.id);
if (WARN_ON(rc))
goto out;
}
} else if (req->msg.type == XS_TRANSACTION_END) {
trans = xenbus_get_transaction(u, req->msg.tx_id);
if (WARN_ON(!trans))
goto out;
list_del(&trans->list);
kfree(trans);
}
mutex_unlock(&u->msgbuffer_mutex);
mutex_lock(&u->reply_mutex);
rc = queue_reply(&staging_q, &req->msg, sizeof(req->msg));
if (!rc)
rc = queue_reply(&staging_q, req->body, req->msg.len);
if (!rc) {
list_splice_tail(&staging_q, &u->read_buffers);
wake_up(&u->read_waitq);
} else {
queue_cleanup(&staging_q);
}
mutex_unlock(&u->reply_mutex);
kfree(req->body);
kfree(req);
kref_put(&u->kref, xenbus_file_free);
return;
out:
mutex_unlock(&u->msgbuffer_mutex);
}
static int xenbus_command_reply(struct xenbus_file_priv *u,
unsigned int msg_type, const char *reply)
{
struct {
struct xsd_sockmsg hdr;
const char body[16];
} msg;
int rc;
msg.hdr = u->u.msg;
msg.hdr.type = msg_type;
msg.hdr.len = strlen(reply) + 1;
if (msg.hdr.len > sizeof(msg.body))
return -E2BIG;
mutex_lock(&u->reply_mutex);
rc = queue_reply(&u->read_buffers, &msg, sizeof(msg.hdr) + msg.hdr.len);
wake_up(&u->read_waitq);
mutex_unlock(&u->reply_mutex);
xen: optimize xenbus driver for multiple concurrent xenstore accesses Handling of multiple concurrent Xenstore accesses through xenbus driver either from the kernel or user land is rather lame today: xenbus is capable to have one access active only at one point of time. Rewrite xenbus to handle multiple requests concurrently by making use of the request id of the Xenstore protocol. This requires to: - Instead of blocking inside xb_read() when trying to read data from the xenstore ring buffer do so only in the main loop of xenbus_thread(). - Instead of doing writes to the xenstore ring buffer in the context of the caller just queue the request and do the write in the dedicated xenbus thread. - Instead of just forwarding the request id specified by the caller of xenbus to xenstore use a xenbus internal unique request id. This will allow multiple outstanding requests. - Modify the locking scheme in order to allow multiple requests being active in parallel. - Instead of waiting for the reply of a user's xenstore request after writing the request to the xenstore ring buffer return directly to the caller and do the waiting in the read path. Additionally signal handling was optimized by avoiding waking up the xenbus thread or sending an event to Xenstore in case the addressed entity is known to be running already. As a result communication with Xenstore is sped up by a factor of up to 5: depending on the request type (read or write) and the amount of data transferred the gain was at least 20% (small reads) and went up to a factor of 5 for large writes. In the end some more rough edges of xenbus have been smoothed: - Handling of memory shortage when reading from xenstore ring buffer in the xenbus driver was not optimal: it was busy looping and issuing a warning in each loop. - In case of xenstore not running in dom0 but in a stubdom we end up with two xenbus threads running as the initialization of xenbus in dom0 expecting a local xenstored will be redone later when connecting to the xenstore domain. Up to now this was no problem as locking would prevent the two xenbus threads interfering with each other, but this was just a waste of kernel resources. - An out of memory situation while writing to or reading from the xenstore ring buffer no longer will lead to a possible loss of synchronization with xenstore. - The user read and write part are now interruptible by signals. Signed-off-by: Juergen Gross <jgross@suse.com> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
2017-02-09 21:39:58 +08:00
if (!rc)
kref_put(&u->kref, xenbus_file_free);
return rc;
}
static int xenbus_write_transaction(unsigned msg_type,
struct xenbus_file_priv *u)
{
int rc;
struct xenbus_transaction_holder *trans = NULL;
if (msg_type == XS_TRANSACTION_START) {
xen: optimize xenbus driver for multiple concurrent xenstore accesses Handling of multiple concurrent Xenstore accesses through xenbus driver either from the kernel or user land is rather lame today: xenbus is capable to have one access active only at one point of time. Rewrite xenbus to handle multiple requests concurrently by making use of the request id of the Xenstore protocol. This requires to: - Instead of blocking inside xb_read() when trying to read data from the xenstore ring buffer do so only in the main loop of xenbus_thread(). - Instead of doing writes to the xenstore ring buffer in the context of the caller just queue the request and do the write in the dedicated xenbus thread. - Instead of just forwarding the request id specified by the caller of xenbus to xenstore use a xenbus internal unique request id. This will allow multiple outstanding requests. - Modify the locking scheme in order to allow multiple requests being active in parallel. - Instead of waiting for the reply of a user's xenstore request after writing the request to the xenstore ring buffer return directly to the caller and do the waiting in the read path. Additionally signal handling was optimized by avoiding waking up the xenbus thread or sending an event to Xenstore in case the addressed entity is known to be running already. As a result communication with Xenstore is sped up by a factor of up to 5: depending on the request type (read or write) and the amount of data transferred the gain was at least 20% (small reads) and went up to a factor of 5 for large writes. In the end some more rough edges of xenbus have been smoothed: - Handling of memory shortage when reading from xenstore ring buffer in the xenbus driver was not optimal: it was busy looping and issuing a warning in each loop. - In case of xenstore not running in dom0 but in a stubdom we end up with two xenbus threads running as the initialization of xenbus in dom0 expecting a local xenstored will be redone later when connecting to the xenstore domain. Up to now this was no problem as locking would prevent the two xenbus threads interfering with each other, but this was just a waste of kernel resources. - An out of memory situation while writing to or reading from the xenstore ring buffer no longer will lead to a possible loss of synchronization with xenstore. - The user read and write part are now interruptible by signals. Signed-off-by: Juergen Gross <jgross@suse.com> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
2017-02-09 21:39:58 +08:00
trans = kzalloc(sizeof(*trans), GFP_KERNEL);
if (!trans) {
rc = -ENOMEM;
goto out;
}
xen: optimize xenbus driver for multiple concurrent xenstore accesses Handling of multiple concurrent Xenstore accesses through xenbus driver either from the kernel or user land is rather lame today: xenbus is capable to have one access active only at one point of time. Rewrite xenbus to handle multiple requests concurrently by making use of the request id of the Xenstore protocol. This requires to: - Instead of blocking inside xb_read() when trying to read data from the xenstore ring buffer do so only in the main loop of xenbus_thread(). - Instead of doing writes to the xenstore ring buffer in the context of the caller just queue the request and do the write in the dedicated xenbus thread. - Instead of just forwarding the request id specified by the caller of xenbus to xenstore use a xenbus internal unique request id. This will allow multiple outstanding requests. - Modify the locking scheme in order to allow multiple requests being active in parallel. - Instead of waiting for the reply of a user's xenstore request after writing the request to the xenstore ring buffer return directly to the caller and do the waiting in the read path. Additionally signal handling was optimized by avoiding waking up the xenbus thread or sending an event to Xenstore in case the addressed entity is known to be running already. As a result communication with Xenstore is sped up by a factor of up to 5: depending on the request type (read or write) and the amount of data transferred the gain was at least 20% (small reads) and went up to a factor of 5 for large writes. In the end some more rough edges of xenbus have been smoothed: - Handling of memory shortage when reading from xenstore ring buffer in the xenbus driver was not optimal: it was busy looping and issuing a warning in each loop. - In case of xenstore not running in dom0 but in a stubdom we end up with two xenbus threads running as the initialization of xenbus in dom0 expecting a local xenstored will be redone later when connecting to the xenstore domain. Up to now this was no problem as locking would prevent the two xenbus threads interfering with each other, but this was just a waste of kernel resources. - An out of memory situation while writing to or reading from the xenstore ring buffer no longer will lead to a possible loss of synchronization with xenstore. - The user read and write part are now interruptible by signals. Signed-off-by: Juergen Gross <jgross@suse.com> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
2017-02-09 21:39:58 +08:00
list_add(&trans->list, &u->transactions);
} else if (u->u.msg.tx_id != 0 &&
!xenbus_get_transaction(u, u->u.msg.tx_id))
return xenbus_command_reply(u, XS_ERROR, "ENOENT");
xen: optimize xenbus driver for multiple concurrent xenstore accesses Handling of multiple concurrent Xenstore accesses through xenbus driver either from the kernel or user land is rather lame today: xenbus is capable to have one access active only at one point of time. Rewrite xenbus to handle multiple requests concurrently by making use of the request id of the Xenstore protocol. This requires to: - Instead of blocking inside xb_read() when trying to read data from the xenstore ring buffer do so only in the main loop of xenbus_thread(). - Instead of doing writes to the xenstore ring buffer in the context of the caller just queue the request and do the write in the dedicated xenbus thread. - Instead of just forwarding the request id specified by the caller of xenbus to xenstore use a xenbus internal unique request id. This will allow multiple outstanding requests. - Modify the locking scheme in order to allow multiple requests being active in parallel. - Instead of waiting for the reply of a user's xenstore request after writing the request to the xenstore ring buffer return directly to the caller and do the waiting in the read path. Additionally signal handling was optimized by avoiding waking up the xenbus thread or sending an event to Xenstore in case the addressed entity is known to be running already. As a result communication with Xenstore is sped up by a factor of up to 5: depending on the request type (read or write) and the amount of data transferred the gain was at least 20% (small reads) and went up to a factor of 5 for large writes. In the end some more rough edges of xenbus have been smoothed: - Handling of memory shortage when reading from xenstore ring buffer in the xenbus driver was not optimal: it was busy looping and issuing a warning in each loop. - In case of xenstore not running in dom0 but in a stubdom we end up with two xenbus threads running as the initialization of xenbus in dom0 expecting a local xenstored will be redone later when connecting to the xenstore domain. Up to now this was no problem as locking would prevent the two xenbus threads interfering with each other, but this was just a waste of kernel resources. - An out of memory situation while writing to or reading from the xenstore ring buffer no longer will lead to a possible loss of synchronization with xenstore. - The user read and write part are now interruptible by signals. Signed-off-by: Juergen Gross <jgross@suse.com> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
2017-02-09 21:39:58 +08:00
rc = xenbus_dev_request_and_reply(&u->u.msg, u);
if (rc && trans) {
list_del(&trans->list);
kfree(trans);
}
out:
return rc;
}
static int xenbus_write_watch(unsigned msg_type, struct xenbus_file_priv *u)
{
struct watch_adapter *watch;
char *path, *token;
int err, rc;
LIST_HEAD(staging_q);
path = u->u.buffer + sizeof(u->u.msg);
token = memchr(path, 0, u->u.msg.len);
if (token == NULL) {
rc = xenbus_command_reply(u, XS_ERROR, "EINVAL");
goto out;
}
token++;
if (memchr(token, 0, u->u.msg.len - (token - path)) == NULL) {
rc = xenbus_command_reply(u, XS_ERROR, "EINVAL");
goto out;
}
if (msg_type == XS_WATCH) {
watch = alloc_watch_adapter(path, token);
if (watch == NULL) {
rc = -ENOMEM;
goto out;
}
watch->watch.callback = watch_fired;
watch->dev_data = u;
err = register_xenbus_watch(&watch->watch);
if (err) {
free_watch_adapter(watch);
rc = err;
goto out;
}
list_add(&watch->list, &u->watches);
} else {
list_for_each_entry(watch, &u->watches, list) {
if (!strcmp(watch->token, token) &&
!strcmp(watch->watch.node, path)) {
unregister_xenbus_watch(&watch->watch);
list_del(&watch->list);
free_watch_adapter(watch);
break;
}
}
}
/* Success. Synthesize a reply to say all is OK. */
rc = xenbus_command_reply(u, msg_type, "OK");
out:
return rc;
}
static ssize_t xenbus_file_write(struct file *filp,
const char __user *ubuf,
size_t len, loff_t *ppos)
{
struct xenbus_file_priv *u = filp->private_data;
uint32_t msg_type;
int rc = len;
int ret;
LIST_HEAD(staging_q);
/*
* We're expecting usermode to be writing properly formed
* xenbus messages. If they write an incomplete message we
* buffer it up. Once it is complete, we act on it.
*/
/*
* Make sure concurrent writers can't stomp all over each
* other's messages and make a mess of our partial message
* buffer. We don't make any attemppt to stop multiple
* writers from making a mess of each other's incomplete
* messages; we're just trying to guarantee our own internal
* consistency and make sure that single writes are handled
* atomically.
*/
mutex_lock(&u->msgbuffer_mutex);
/* Get this out of the way early to avoid confusion */
if (len == 0)
goto out;
/* Can't write a xenbus message larger we can buffer */
if (len > sizeof(u->u.buffer) - u->len) {
/* On error, dump existing buffer */
u->len = 0;
rc = -EINVAL;
goto out;
}
ret = copy_from_user(u->u.buffer + u->len, ubuf, len);
if (ret != 0) {
rc = -EFAULT;
goto out;
}
/* Deal with a partial copy. */
len -= ret;
rc = len;
u->len += len;
/* Return if we haven't got a full message yet */
if (u->len < sizeof(u->u.msg))
goto out; /* not even the header yet */
/* If we're expecting a message that's larger than we can
possibly send, dump what we have and return an error. */
if ((sizeof(u->u.msg) + u->u.msg.len) > sizeof(u->u.buffer)) {
rc = -E2BIG;
u->len = 0;
goto out;
}
if (u->len < (sizeof(u->u.msg) + u->u.msg.len))
goto out; /* incomplete data portion */
/*
* OK, now we have a complete message. Do something with it.
*/
xen: optimize xenbus driver for multiple concurrent xenstore accesses Handling of multiple concurrent Xenstore accesses through xenbus driver either from the kernel or user land is rather lame today: xenbus is capable to have one access active only at one point of time. Rewrite xenbus to handle multiple requests concurrently by making use of the request id of the Xenstore protocol. This requires to: - Instead of blocking inside xb_read() when trying to read data from the xenstore ring buffer do so only in the main loop of xenbus_thread(). - Instead of doing writes to the xenstore ring buffer in the context of the caller just queue the request and do the write in the dedicated xenbus thread. - Instead of just forwarding the request id specified by the caller of xenbus to xenstore use a xenbus internal unique request id. This will allow multiple outstanding requests. - Modify the locking scheme in order to allow multiple requests being active in parallel. - Instead of waiting for the reply of a user's xenstore request after writing the request to the xenstore ring buffer return directly to the caller and do the waiting in the read path. Additionally signal handling was optimized by avoiding waking up the xenbus thread or sending an event to Xenstore in case the addressed entity is known to be running already. As a result communication with Xenstore is sped up by a factor of up to 5: depending on the request type (read or write) and the amount of data transferred the gain was at least 20% (small reads) and went up to a factor of 5 for large writes. In the end some more rough edges of xenbus have been smoothed: - Handling of memory shortage when reading from xenstore ring buffer in the xenbus driver was not optimal: it was busy looping and issuing a warning in each loop. - In case of xenstore not running in dom0 but in a stubdom we end up with two xenbus threads running as the initialization of xenbus in dom0 expecting a local xenstored will be redone later when connecting to the xenstore domain. Up to now this was no problem as locking would prevent the two xenbus threads interfering with each other, but this was just a waste of kernel resources. - An out of memory situation while writing to or reading from the xenstore ring buffer no longer will lead to a possible loss of synchronization with xenstore. - The user read and write part are now interruptible by signals. Signed-off-by: Juergen Gross <jgross@suse.com> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
2017-02-09 21:39:58 +08:00
kref_get(&u->kref);
msg_type = u->u.msg.type;
switch (msg_type) {
case XS_WATCH:
case XS_UNWATCH:
/* (Un)Ask for some path to be watched for changes */
ret = xenbus_write_watch(msg_type, u);
break;
default:
/* Send out a transaction */
ret = xenbus_write_transaction(msg_type, u);
break;
}
xen: optimize xenbus driver for multiple concurrent xenstore accesses Handling of multiple concurrent Xenstore accesses through xenbus driver either from the kernel or user land is rather lame today: xenbus is capable to have one access active only at one point of time. Rewrite xenbus to handle multiple requests concurrently by making use of the request id of the Xenstore protocol. This requires to: - Instead of blocking inside xb_read() when trying to read data from the xenstore ring buffer do so only in the main loop of xenbus_thread(). - Instead of doing writes to the xenstore ring buffer in the context of the caller just queue the request and do the write in the dedicated xenbus thread. - Instead of just forwarding the request id specified by the caller of xenbus to xenstore use a xenbus internal unique request id. This will allow multiple outstanding requests. - Modify the locking scheme in order to allow multiple requests being active in parallel. - Instead of waiting for the reply of a user's xenstore request after writing the request to the xenstore ring buffer return directly to the caller and do the waiting in the read path. Additionally signal handling was optimized by avoiding waking up the xenbus thread or sending an event to Xenstore in case the addressed entity is known to be running already. As a result communication with Xenstore is sped up by a factor of up to 5: depending on the request type (read or write) and the amount of data transferred the gain was at least 20% (small reads) and went up to a factor of 5 for large writes. In the end some more rough edges of xenbus have been smoothed: - Handling of memory shortage when reading from xenstore ring buffer in the xenbus driver was not optimal: it was busy looping and issuing a warning in each loop. - In case of xenstore not running in dom0 but in a stubdom we end up with two xenbus threads running as the initialization of xenbus in dom0 expecting a local xenstored will be redone later when connecting to the xenstore domain. Up to now this was no problem as locking would prevent the two xenbus threads interfering with each other, but this was just a waste of kernel resources. - An out of memory situation while writing to or reading from the xenstore ring buffer no longer will lead to a possible loss of synchronization with xenstore. - The user read and write part are now interruptible by signals. Signed-off-by: Juergen Gross <jgross@suse.com> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
2017-02-09 21:39:58 +08:00
if (ret != 0) {
rc = ret;
xen: optimize xenbus driver for multiple concurrent xenstore accesses Handling of multiple concurrent Xenstore accesses through xenbus driver either from the kernel or user land is rather lame today: xenbus is capable to have one access active only at one point of time. Rewrite xenbus to handle multiple requests concurrently by making use of the request id of the Xenstore protocol. This requires to: - Instead of blocking inside xb_read() when trying to read data from the xenstore ring buffer do so only in the main loop of xenbus_thread(). - Instead of doing writes to the xenstore ring buffer in the context of the caller just queue the request and do the write in the dedicated xenbus thread. - Instead of just forwarding the request id specified by the caller of xenbus to xenstore use a xenbus internal unique request id. This will allow multiple outstanding requests. - Modify the locking scheme in order to allow multiple requests being active in parallel. - Instead of waiting for the reply of a user's xenstore request after writing the request to the xenstore ring buffer return directly to the caller and do the waiting in the read path. Additionally signal handling was optimized by avoiding waking up the xenbus thread or sending an event to Xenstore in case the addressed entity is known to be running already. As a result communication with Xenstore is sped up by a factor of up to 5: depending on the request type (read or write) and the amount of data transferred the gain was at least 20% (small reads) and went up to a factor of 5 for large writes. In the end some more rough edges of xenbus have been smoothed: - Handling of memory shortage when reading from xenstore ring buffer in the xenbus driver was not optimal: it was busy looping and issuing a warning in each loop. - In case of xenstore not running in dom0 but in a stubdom we end up with two xenbus threads running as the initialization of xenbus in dom0 expecting a local xenstored will be redone later when connecting to the xenstore domain. Up to now this was no problem as locking would prevent the two xenbus threads interfering with each other, but this was just a waste of kernel resources. - An out of memory situation while writing to or reading from the xenstore ring buffer no longer will lead to a possible loss of synchronization with xenstore. - The user read and write part are now interruptible by signals. Signed-off-by: Juergen Gross <jgross@suse.com> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
2017-02-09 21:39:58 +08:00
kref_put(&u->kref, xenbus_file_free);
}
/* Buffered message consumed */
u->len = 0;
out:
mutex_unlock(&u->msgbuffer_mutex);
return rc;
}
static int xenbus_file_open(struct inode *inode, struct file *filp)
{
struct xenbus_file_priv *u;
if (xen_store_evtchn == 0)
return -ENOENT;
nonseekable_open(inode, filp);
filp->f_mode &= ~FMODE_ATOMIC_POS; /* cdev-style semantics */
u = kzalloc(sizeof(*u), GFP_KERNEL);
if (u == NULL)
return -ENOMEM;
xen: optimize xenbus driver for multiple concurrent xenstore accesses Handling of multiple concurrent Xenstore accesses through xenbus driver either from the kernel or user land is rather lame today: xenbus is capable to have one access active only at one point of time. Rewrite xenbus to handle multiple requests concurrently by making use of the request id of the Xenstore protocol. This requires to: - Instead of blocking inside xb_read() when trying to read data from the xenstore ring buffer do so only in the main loop of xenbus_thread(). - Instead of doing writes to the xenstore ring buffer in the context of the caller just queue the request and do the write in the dedicated xenbus thread. - Instead of just forwarding the request id specified by the caller of xenbus to xenstore use a xenbus internal unique request id. This will allow multiple outstanding requests. - Modify the locking scheme in order to allow multiple requests being active in parallel. - Instead of waiting for the reply of a user's xenstore request after writing the request to the xenstore ring buffer return directly to the caller and do the waiting in the read path. Additionally signal handling was optimized by avoiding waking up the xenbus thread or sending an event to Xenstore in case the addressed entity is known to be running already. As a result communication with Xenstore is sped up by a factor of up to 5: depending on the request type (read or write) and the amount of data transferred the gain was at least 20% (small reads) and went up to a factor of 5 for large writes. In the end some more rough edges of xenbus have been smoothed: - Handling of memory shortage when reading from xenstore ring buffer in the xenbus driver was not optimal: it was busy looping and issuing a warning in each loop. - In case of xenstore not running in dom0 but in a stubdom we end up with two xenbus threads running as the initialization of xenbus in dom0 expecting a local xenstored will be redone later when connecting to the xenstore domain. Up to now this was no problem as locking would prevent the two xenbus threads interfering with each other, but this was just a waste of kernel resources. - An out of memory situation while writing to or reading from the xenstore ring buffer no longer will lead to a possible loss of synchronization with xenstore. - The user read and write part are now interruptible by signals. Signed-off-by: Juergen Gross <jgross@suse.com> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
2017-02-09 21:39:58 +08:00
kref_init(&u->kref);
INIT_LIST_HEAD(&u->transactions);
INIT_LIST_HEAD(&u->watches);
INIT_LIST_HEAD(&u->read_buffers);
init_waitqueue_head(&u->read_waitq);
mutex_init(&u->reply_mutex);
mutex_init(&u->msgbuffer_mutex);
filp->private_data = u;
return 0;
}
static int xenbus_file_release(struct inode *inode, struct file *filp)
{
struct xenbus_file_priv *u = filp->private_data;
xen: optimize xenbus driver for multiple concurrent xenstore accesses Handling of multiple concurrent Xenstore accesses through xenbus driver either from the kernel or user land is rather lame today: xenbus is capable to have one access active only at one point of time. Rewrite xenbus to handle multiple requests concurrently by making use of the request id of the Xenstore protocol. This requires to: - Instead of blocking inside xb_read() when trying to read data from the xenstore ring buffer do so only in the main loop of xenbus_thread(). - Instead of doing writes to the xenstore ring buffer in the context of the caller just queue the request and do the write in the dedicated xenbus thread. - Instead of just forwarding the request id specified by the caller of xenbus to xenstore use a xenbus internal unique request id. This will allow multiple outstanding requests. - Modify the locking scheme in order to allow multiple requests being active in parallel. - Instead of waiting for the reply of a user's xenstore request after writing the request to the xenstore ring buffer return directly to the caller and do the waiting in the read path. Additionally signal handling was optimized by avoiding waking up the xenbus thread or sending an event to Xenstore in case the addressed entity is known to be running already. As a result communication with Xenstore is sped up by a factor of up to 5: depending on the request type (read or write) and the amount of data transferred the gain was at least 20% (small reads) and went up to a factor of 5 for large writes. In the end some more rough edges of xenbus have been smoothed: - Handling of memory shortage when reading from xenstore ring buffer in the xenbus driver was not optimal: it was busy looping and issuing a warning in each loop. - In case of xenstore not running in dom0 but in a stubdom we end up with two xenbus threads running as the initialization of xenbus in dom0 expecting a local xenstored will be redone later when connecting to the xenstore domain. Up to now this was no problem as locking would prevent the two xenbus threads interfering with each other, but this was just a waste of kernel resources. - An out of memory situation while writing to or reading from the xenstore ring buffer no longer will lead to a possible loss of synchronization with xenstore. - The user read and write part are now interruptible by signals. Signed-off-by: Juergen Gross <jgross@suse.com> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
2017-02-09 21:39:58 +08:00
kref_put(&u->kref, xenbus_file_free);
return 0;
}
static unsigned int xenbus_file_poll(struct file *file, poll_table *wait)
{
struct xenbus_file_priv *u = file->private_data;
poll_wait(file, &u->read_waitq, wait);
if (!list_empty(&u->read_buffers))
return POLLIN | POLLRDNORM;
return 0;
}
const struct file_operations xen_xenbus_fops = {
.read = xenbus_file_read,
.write = xenbus_file_write,
.open = xenbus_file_open,
.release = xenbus_file_release,
.poll = xenbus_file_poll,
llseek: automatically add .llseek fop All file_operations should get a .llseek operation so we can make nonseekable_open the default for future file operations without a .llseek pointer. The three cases that we can automatically detect are no_llseek, seq_lseek and default_llseek. For cases where we can we can automatically prove that the file offset is always ignored, we use noop_llseek, which maintains the current behavior of not returning an error from a seek. New drivers should normally not use noop_llseek but instead use no_llseek and call nonseekable_open at open time. Existing drivers can be converted to do the same when the maintainer knows for certain that no user code relies on calling seek on the device file. The generated code is often incorrectly indented and right now contains comments that clarify for each added line why a specific variant was chosen. In the version that gets submitted upstream, the comments will be gone and I will manually fix the indentation, because there does not seem to be a way to do that using coccinelle. Some amount of new code is currently sitting in linux-next that should get the same modifications, which I will do at the end of the merge window. Many thanks to Julia Lawall for helping me learn to write a semantic patch that does all this. ===== begin semantic patch ===== // This adds an llseek= method to all file operations, // as a preparation for making no_llseek the default. // // The rules are // - use no_llseek explicitly if we do nonseekable_open // - use seq_lseek for sequential files // - use default_llseek if we know we access f_pos // - use noop_llseek if we know we don't access f_pos, // but we still want to allow users to call lseek // @ open1 exists @ identifier nested_open; @@ nested_open(...) { <+... nonseekable_open(...) ...+> } @ open exists@ identifier open_f; identifier i, f; identifier open1.nested_open; @@ int open_f(struct inode *i, struct file *f) { <+... ( nonseekable_open(...) | nested_open(...) ) ...+> } @ read disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ read_no_fpos disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { ... when != off } @ write @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ write_no_fpos @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { ... when != off } @ fops0 @ identifier fops; @@ struct file_operations fops = { ... }; @ has_llseek depends on fops0 @ identifier fops0.fops; identifier llseek_f; @@ struct file_operations fops = { ... .llseek = llseek_f, ... }; @ has_read depends on fops0 @ identifier fops0.fops; identifier read_f; @@ struct file_operations fops = { ... .read = read_f, ... }; @ has_write depends on fops0 @ identifier fops0.fops; identifier write_f; @@ struct file_operations fops = { ... .write = write_f, ... }; @ has_open depends on fops0 @ identifier fops0.fops; identifier open_f; @@ struct file_operations fops = { ... .open = open_f, ... }; // use no_llseek if we call nonseekable_open //////////////////////////////////////////// @ nonseekable1 depends on !has_llseek && has_open @ identifier fops0.fops; identifier nso ~= "nonseekable_open"; @@ struct file_operations fops = { ... .open = nso, ... +.llseek = no_llseek, /* nonseekable */ }; @ nonseekable2 depends on !has_llseek @ identifier fops0.fops; identifier open.open_f; @@ struct file_operations fops = { ... .open = open_f, ... +.llseek = no_llseek, /* open uses nonseekable */ }; // use seq_lseek for sequential files ///////////////////////////////////// @ seq depends on !has_llseek @ identifier fops0.fops; identifier sr ~= "seq_read"; @@ struct file_operations fops = { ... .read = sr, ... +.llseek = seq_lseek, /* we have seq_read */ }; // use default_llseek if there is a readdir /////////////////////////////////////////// @ fops1 depends on !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier readdir_e; @@ // any other fop is used that changes pos struct file_operations fops = { ... .readdir = readdir_e, ... +.llseek = default_llseek, /* readdir is present */ }; // use default_llseek if at least one of read/write touches f_pos ///////////////////////////////////////////////////////////////// @ fops2 depends on !fops1 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read.read_f; @@ // read fops use offset struct file_operations fops = { ... .read = read_f, ... +.llseek = default_llseek, /* read accesses f_pos */ }; @ fops3 depends on !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, ... + .llseek = default_llseek, /* write accesses f_pos */ }; // Use noop_llseek if neither read nor write accesses f_pos /////////////////////////////////////////////////////////// @ fops4 depends on !fops1 && !fops2 && !fops3 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; identifier write_no_fpos.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, .read = read_f, ... +.llseek = noop_llseek, /* read and write both use no f_pos */ }; @ depends on has_write && !has_read && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write_no_fpos.write_f; @@ struct file_operations fops = { ... .write = write_f, ... +.llseek = noop_llseek, /* write uses no f_pos */ }; @ depends on has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; @@ struct file_operations fops = { ... .read = read_f, ... +.llseek = noop_llseek, /* read uses no f_pos */ }; @ depends on !has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; @@ struct file_operations fops = { ... +.llseek = noop_llseek, /* no read or write fn */ }; ===== End semantic patch ===== Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Julia Lawall <julia@diku.dk> Cc: Christoph Hellwig <hch@infradead.org>
2010-08-16 00:52:59 +08:00
.llseek = no_llseek,
};
EXPORT_SYMBOL_GPL(xen_xenbus_fops);
static struct miscdevice xenbus_dev = {
.minor = MISC_DYNAMIC_MINOR,
.name = "xen/xenbus",
.fops = &xen_xenbus_fops,
};
static int __init xenbus_init(void)
{
int err;
if (!xen_domain())
return -ENODEV;
err = misc_register(&xenbus_dev);
if (err)
pr_err("Could not register xenbus frontend device\n");
return err;
}
device_initcall(xenbus_init);