#include <linux/init.h>
#include <linux/kthread.h>
#include <linux/poll.h>
-#include <linux/miscdevice.h>
+#include <linux/device.h>
+#include <linux/cdev.h>
#include <lustre_log.h>
+#include <uapi/linux/lustre/lustre_ioctl.h>
#include "mdc_internal.h"
struct chlg_registered_dev {
/* Device name of the form "changelog-{MDTNAME}" */
- char ced_name[32];
- /* Misc device descriptor */
- struct miscdevice ced_misc;
+ char ced_name[32];
+ /* changelog char device */
+ struct cdev ced_cdev;
+ struct device ced_device;
/* OBDs referencing this device (multiple mount point) */
- struct list_head ced_obds;
+ struct list_head ced_obds;
/* Reference counter for proper deregistration */
- struct kref ced_refs;
+ struct kref ced_refs;
/* Link within the global chlg_registered_devices */
- struct list_head ced_link;
+ struct list_head ced_link;
};
struct chlg_reader_state {
/* Shortcut to the corresponding OBD device */
- struct obd_device *crs_obd;
+ struct obd_device *crs_obd;
+ /* the corresponding chlg_registered_dev */
+ struct chlg_registered_dev *crs_ced;
/* Producer thread (if any) */
- struct task_struct *crs_prod_task;
+ struct task_struct *crs_prod_task;
/* An error occurred that prevents from reading further */
- int crs_err;
+ int crs_err;
/* EOF, no more records available */
- bool crs_eof;
+ bool crs_eof;
/* Desired start position */
- __u64 crs_start_offset;
+ __u64 crs_start_offset;
/* Wait queue for the catalog processing thread */
- wait_queue_head_t crs_waitq_prod;
+ wait_queue_head_t crs_waitq_prod;
/* Wait queue for the record copy threads */
- wait_queue_head_t crs_waitq_cons;
+ wait_queue_head_t crs_waitq_cons;
/* Mutex protecting crs_rec_count and crs_rec_queue */
- struct mutex crs_lock;
+ struct mutex crs_lock;
/* Number of item in the list */
- __u64 crs_rec_count;
+ __u64 crs_rec_count;
/* List of prefetched enqueued_record::enq_linkage_items */
- struct list_head crs_rec_queue;
+ struct list_head crs_rec_queue;
+ unsigned int crs_last_catidx;
+ unsigned int crs_last_idx;
+ bool crs_poll;
};
struct chlg_rec_entry {
CDEV_CHLG_MAX_PREFETCH = 1024,
};
+DEFINE_IDR(mdc_changelog_minor_idr);
+static DEFINE_SPINLOCK(chlg_minor_lock);
+
+static int chlg_minor_alloc(int *pminor)
+{
+ void *minor_allocated = (void *)-1;
+ int minor;
+
+ idr_preload(GFP_KERNEL);
+ spin_lock(&chlg_minor_lock);
+ minor = idr_alloc(&mdc_changelog_minor_idr, minor_allocated, 0,
+ MDC_CHANGELOG_DEV_COUNT, GFP_NOWAIT);
+ spin_unlock(&chlg_minor_lock);
+ idr_preload_end();
+
+ if (minor < 0)
+ return minor;
+
+ *pminor = minor;
+ return 0;
+}
+
+static void chlg_minor_free(int minor)
+{
+ spin_lock(&chlg_minor_lock);
+ idr_remove(&mdc_changelog_minor_idr, minor);
+ spin_unlock(&chlg_minor_lock);
+}
+
+static void chlg_device_release(struct device *dev)
+{
+ struct chlg_registered_dev *entry = dev_get_drvdata(dev);
+
+ chlg_minor_free(MINOR(entry->ced_cdev.dev));
+ OBD_FREE_PTR(entry);
+}
+
+/**
+ * Deregister a changelog character device whose refcount has reached zero.
+ */
+static void chlg_dev_clear(struct kref *kref)
+{
+ struct chlg_registered_dev *entry;
+
+ ENTRY;
+ entry = container_of(kref, struct chlg_registered_dev,
+ ced_refs);
+
+ list_del(&entry->ced_link);
+ cdev_device_del(&entry->ced_cdev, &entry->ced_device);
+ put_device(&entry->ced_device);
+ EXIT;
+}
+
+static inline struct obd_device* chlg_obd_get(struct chlg_registered_dev *dev)
+{
+ struct obd_device *obd;
+
+ mutex_lock(&chlg_registered_dev_lock);
+ if (list_empty(&dev->ced_obds))
+ return NULL;
+
+ obd = list_first_entry(&dev->ced_obds, struct obd_device,
+ u.cli.cl_chg_dev_linkage);
+ class_incref(obd, "changelog", dev);
+ mutex_unlock(&chlg_registered_dev_lock);
+ return obd;
+}
+
+static inline void chlg_obd_put(struct chlg_registered_dev *dev,
+ struct obd_device *obd)
+{
+ class_decref(obd, "changelog", dev);
+}
+
/**
* ChangeLog catalog processing callback invoked on each record.
* If the current record is eligible to userland delivery, push
rec = container_of(hdr, struct llog_changelog_rec, cr_hdr);
+ crs->crs_last_catidx = llh->lgh_hdr->llh_cat_idx;
+ crs->crs_last_idx = hdr->lrh_index;
+
if (rec->cr_hdr.lrh_type != CHANGELOG_REC) {
rc = -EINVAL;
CERROR("%s: not a changelog rec %x/%d in llog "DFID" rc = %d\n",
static int chlg_load(void *args)
{
struct chlg_reader_state *crs = args;
- struct obd_device *obd = crs->crs_obd;
+ struct chlg_registered_dev *ced = crs->crs_ced;
+ struct obd_device *obd = NULL;
struct llog_ctxt *ctx = NULL;
struct llog_handle *llh = NULL;
int rc;
ENTRY;
+ crs->crs_last_catidx = -1;
+ crs->crs_last_idx = 0;
+
+again:
+ obd = chlg_obd_get(ced);
+ if (obd == NULL)
+ RETURN(-ENODEV);
+
+ crs->crs_obd = obd;
+
ctx = llog_get_context(obd, LLOG_CHANGELOG_REPL_CTXT);
if (ctx == NULL)
GOTO(err_out, rc = -ENOENT);
GOTO(err_out, rc);
}
+
rc = llog_init_handle(NULL, llh,
LLOG_F_IS_CAT |
LLOG_F_EXT_JOBID |
GOTO(err_out, rc);
}
- rc = llog_cat_process(NULL, llh, chlg_read_cat_process_cb, crs, 0, 0);
+ rc = llog_cat_process(NULL, llh, chlg_read_cat_process_cb, crs,
+ crs->crs_last_catidx, crs->crs_last_idx);
if (rc < 0) {
CERROR("%s: fail to process llog: rc = %d\n", obd->obd_name, rc);
GOTO(err_out, rc);
}
+ if (!kthread_should_stop() && crs->crs_poll) {
+ llog_cat_close(NULL, llh);
+ llog_ctxt_put(ctx);
+ class_decref(obd, "changelog", crs);
+ schedule_timeout_interruptible(cfs_time_seconds(1));
+ goto again;
+ }
crs->crs_eof = true;
if (ctx != NULL)
llog_ctxt_put(ctx);
+ crs->crs_obd = NULL;
+ chlg_obd_put(ced, obd);
wait_event_interruptible(crs->crs_waitq_prod, kthread_should_stop());
RETURN(rc);
*/
static int chlg_clear(struct chlg_reader_state *crs, __u32 reader, __u64 record)
{
- struct obd_device *obd = crs->crs_obd;
+ struct obd_device *obd = NULL;
struct changelog_setinfo cs = {
.cs_recno = record,
.cs_id = reader
};
+ int rc;
+
+ obd = chlg_obd_get(crs->crs_ced);
+ if (obd == NULL)
+ return -ENODEV;
- return obd_set_info_async(NULL, obd->obd_self_export,
- strlen(KEY_CHANGELOG_CLEAR),
- KEY_CHANGELOG_CLEAR, sizeof(cs), &cs, NULL);
+ rc = obd_set_info_async(NULL, obd->obd_self_export,
+ strlen(KEY_CHANGELOG_CLEAR),
+ KEY_CHANGELOG_CLEAR, sizeof(cs), &cs, NULL);
+
+ chlg_obd_put(crs->crs_ced, obd);
+ return rc;
}
/** Maximum changelog control command size */
}
/**
- * Find the OBD device associated to a changelog character device.
- * @param[in] cdev character device instance descriptor
- * @return corresponding OBD device or NULL if none was found.
- */
-static struct obd_device *chlg_obd_get(dev_t cdev)
-{
- int minor = MINOR(cdev);
- struct obd_device *obd = NULL;
- struct chlg_registered_dev *curr;
-
- mutex_lock(&chlg_registered_dev_lock);
- list_for_each_entry(curr, &chlg_registered_devices, ced_link) {
- if (curr->ced_misc.minor == minor) {
- /* take the first available OBD device attached */
- obd = list_first_entry(&curr->ced_obds,
- struct obd_device,
- u.cli.cl_chg_dev_linkage);
- break;
- }
- }
- mutex_unlock(&chlg_registered_dev_lock);
- return obd;
-}
-
-/**
* Open handler, initialize internal CRS state and spawn prefetch thread if
* needed.
* @param[in] inode Inode struct for the open character device.
static int chlg_open(struct inode *inode, struct file *file)
{
struct chlg_reader_state *crs;
- struct obd_device *obd = chlg_obd_get(inode->i_rdev);
+ struct chlg_registered_dev *dev;
struct task_struct *task;
int rc;
ENTRY;
- if (!obd)
- RETURN(-ENODEV);
+ dev = container_of(inode->i_cdev, struct chlg_registered_dev, ced_cdev);
OBD_ALLOC_PTR(crs);
if (!crs)
RETURN(-ENOMEM);
- crs->crs_obd = obd;
+ kref_get(&dev->ced_refs);
+ crs->crs_ced = dev;
crs->crs_err = false;
crs->crs_eof = false;
if (IS_ERR(task)) {
rc = PTR_ERR(task);
CERROR("%s: cannot start changelog thread: rc = %d\n",
- obd->obd_name, rc);
+ dev->ced_name, rc);
GOTO(err_crs, rc);
}
crs->crs_prod_task = task;
RETURN(0);
err_crs:
+ kref_put(&dev->ced_refs, chlg_dev_clear);
OBD_FREE_PTR(crs);
return rc;
}
list_for_each_entry_safe(rec, tmp, &crs->crs_rec_queue, enq_linkage)
enq_record_delete(rec);
+ kref_put(&crs->crs_ced->ced_refs, chlg_dev_clear);
OBD_FREE_PTR(crs);
return rc;
return mask;
}
+static long chlg_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+ int rc;
+
+ struct chlg_reader_state *crs = file->private_data;
+ switch (cmd) {
+ case OBD_IOC_CHLG_POLL:
+ crs->crs_poll = !!arg;
+ rc = 0;
+ break;
+ default:
+ rc = -EINVAL;
+ break;
+ }
+ return rc;
+}
+
static const struct file_operations chlg_fops = {
.owner = THIS_MODULE,
.llseek = chlg_llseek,
.open = chlg_open,
.release = chlg_release,
.poll = chlg_poll,
+ .unlocked_ioctl = chlg_ioctl,
};
/**
* This uses obd_name of the form: "testfs-MDT0000-mdc-ffff88006501600"
* and returns a name of the form: "changelog-testfs-MDT0000".
*/
-static void get_chlg_name(char *name, size_t name_len, struct obd_device *obd)
+static void get_target_name(char *name, size_t name_len, struct obd_device *obd)
{
int i;
- snprintf(name, name_len, "changelog-%s", obd->obd_name);
+ snprintf(name, name_len, "%s", obd->obd_name);
/* Find the 2nd '-' from the end and truncate on it */
for (i = 0; i < 2; i++) {
{
struct chlg_registered_dev *dit;
+ LASSERT(mutex_is_locked(&chlg_registered_dev_lock));
list_for_each_entry(dit, &chlg_registered_devices, ced_link)
if (strcmp(name, dit->ced_name) == 0)
return dit;
struct chlg_registered_dev *dit;
struct obd_device *oit;
+ LASSERT(mutex_is_locked(&chlg_registered_dev_lock));
list_for_each_entry(dit, &chlg_registered_devices, ced_link)
list_for_each_entry(oit, &dit->ced_obds,
u.cli.cl_chg_dev_linkage)
{
struct chlg_registered_dev *exist;
struct chlg_registered_dev *entry;
- int rc;
+ int minor, rc;
ENTRY;
OBD_ALLOC_PTR(entry);
if (entry == NULL)
RETURN(-ENOMEM);
- get_chlg_name(entry->ced_name, sizeof(entry->ced_name), obd);
-
- entry->ced_misc.minor = MISC_DYNAMIC_MINOR;
- entry->ced_misc.name = entry->ced_name;
- entry->ced_misc.fops = &chlg_fops;
+ get_target_name(entry->ced_name, sizeof(entry->ced_name), obd);
kref_init(&entry->ced_refs);
INIT_LIST_HEAD(&entry->ced_obds);
GOTO(out_unlock, rc = 0);
}
- /* Register new character device */
- rc = misc_register(&entry->ced_misc);
- if (rc != 0)
- GOTO(out_unlock, rc);
-
list_add_tail(&obd->u.cli.cl_chg_dev_linkage, &entry->ced_obds);
list_add_tail(&entry->ced_link, &chlg_registered_devices);
+ rc = chlg_minor_alloc(&minor);
+ if (rc)
+ GOTO(out_unlock, rc);
+
+ device_initialize(&entry->ced_device);
+ entry->ced_device.devt = MKDEV(MAJOR(mdc_changelog_dev), minor);
+ entry->ced_device.class = mdc_changelog_class;
+ entry->ced_device.release = chlg_device_release;
+ dev_set_drvdata(&entry->ced_device, entry);
+ rc = dev_set_name(&entry->ced_device, "%s-%s", MDC_CHANGELOG_DEV_NAME,
+ entry->ced_name);
+ if (rc)
+ GOTO(out_minor, rc);
+
+ /* Register new character device */
+ cdev_init(&entry->ced_cdev, &chlg_fops);
+ entry->ced_cdev.owner = THIS_MODULE;
+ rc = cdev_device_add(&entry->ced_cdev, &entry->ced_device);
+ if (rc)
+ GOTO(out_device_name, rc);
+
entry = NULL; /* prevent it from being freed below */
+ GOTO(out_unlock, rc = 0);
+
+out_device_name:
+ kfree_const(entry->ced_device.kobj.name);
+
+out_minor:
+ chlg_minor_free(minor);
+
+ list_del_init(&obd->u.cli.cl_chg_dev_linkage);
+ list_del(&entry->ced_link);
out_unlock:
mutex_unlock(&chlg_registered_dev_lock);
}
/**
- * Deregister a changelog character device whose refcount has reached zero.
- */
-static void chlg_dev_clear(struct kref *kref)
-{
- struct chlg_registered_dev *entry = container_of(kref,
- struct chlg_registered_dev,
- ced_refs);
- ENTRY;
-
- list_del(&entry->ced_link);
- misc_deregister(&entry->ced_misc);
- OBD_FREE_PTR(entry);
- EXIT;
-}
-
-/**
* Release OBD, decrease reference count of the corresponding changelog device.
*/
void mdc_changelog_cdev_finish(struct obd_device *obd)
{
- struct chlg_registered_dev *dev = chlg_registered_dev_find_by_obd(obd);
- ENTRY;
+ struct chlg_registered_dev *dev;
+ ENTRY;
mutex_lock(&chlg_registered_dev_lock);
+ dev = chlg_registered_dev_find_by_obd(obd);
list_del_init(&obd->u.cli.cl_chg_dev_linkage);
kref_put(&dev->ced_refs, chlg_dev_clear);
mutex_unlock(&chlg_registered_dev_lock);