orangefs: delay freeing slot until cancel completes
Make cancels reuse the aborted read/write op, to make sure they do not
fail on lack of memory.
Don't issue a cancel unless the daemon has seen our read/write, has not
replied and isn't being shut down.
If cancel *is* issued, don't wait for it to complete; stash the slot
in there and just have it freed when cancel is finally replied to or
purged (and delay dropping the reference until then, obviously).
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
diff --git a/fs/orangefs/orangefs-kernel.h b/fs/orangefs/orangefs-kernel.h
index a8cde90..3ceeeae 100644
--- a/fs/orangefs/orangefs-kernel.h
+++ b/fs/orangefs/orangefs-kernel.h
@@ -190,9 +190,14 @@
/*
* Set uses_shared_memory to 1 if this operation uses shared memory.
* If true, then a retry on the op must also get a new shared memory
- * buffer and re-populate it.
+ * buffer and re-populate it. Cancels don't care - it only matters
+ * for service_operation() retry logics and cancels don't go through
+ * it anymore.
*/
- int uses_shared_memory;
+ union {
+ int uses_shared_memory;
+ int slot_to_free;
+ };
struct orangefs_upcall_s upcall;
struct orangefs_downcall_s downcall;
@@ -219,17 +224,13 @@
op->op_state = OP_VFS_STATE_SERVICED;
wake_up_interruptible(&op->waitq);
}
-static inline void set_op_state_purged(struct orangefs_kernel_op_s *op)
-{
- op->op_state |= OP_VFS_STATE_PURGED;
- wake_up_interruptible(&op->waitq);
-}
#define op_state_waiting(op) ((op)->op_state & OP_VFS_STATE_WAITING)
#define op_state_in_progress(op) ((op)->op_state & OP_VFS_STATE_INPROGR)
#define op_state_serviced(op) ((op)->op_state & OP_VFS_STATE_SERVICED)
#define op_state_purged(op) ((op)->op_state & OP_VFS_STATE_PURGED)
#define op_state_given_up(op) ((op)->op_state & OP_VFS_STATE_GIVEN_UP)
+#define op_is_cancel(op) ((op)->upcall.type == ORANGEFS_VFS_OP_CANCEL)
static inline void get_op(struct orangefs_kernel_op_s *op)
{
@@ -249,6 +250,27 @@
}
}
+extern void orangefs_bufmap_put(int);
+static inline void put_cancel(struct orangefs_kernel_op_s *op)
+{
+ orangefs_bufmap_put(op->slot_to_free);
+ op_release(op);
+}
+
+static inline void set_op_state_purged(struct orangefs_kernel_op_s *op)
+{
+ spin_lock(&op->lock);
+ if (unlikely(op_is_cancel(op))) {
+ list_del(&op->list);
+ spin_unlock(&op->lock);
+ put_cancel(op);
+ } else {
+ op->op_state |= OP_VFS_STATE_PURGED;
+ wake_up_interruptible(&op->waitq);
+ spin_unlock(&op->lock);
+ }
+}
+
/* per inode private orangefs info */
struct orangefs_inode_s {
struct orangefs_object_kref refn;
@@ -448,6 +470,7 @@
int op_cache_initialize(void);
int op_cache_finalize(void);
struct orangefs_kernel_op_s *op_alloc(__s32 type);
+void orangefs_new_tag(struct orangefs_kernel_op_s *op);
char *get_opname_string(struct orangefs_kernel_op_s *new_op);
int orangefs_inode_cache_initialize(void);
@@ -528,6 +551,7 @@
int orangefs_dev_init(void);
void orangefs_dev_cleanup(void);
int is_daemon_in_service(void);
+bool __is_daemon_in_service(void);
int fs_mount_pending(__s32 fsid);
/*
@@ -562,7 +586,7 @@
int orangefs_unmount_sb(struct super_block *sb);
-int orangefs_cancel_op_in_progress(__u64 tag);
+bool orangefs_cancel_op_in_progress(struct orangefs_kernel_op_s *op);
static inline __u64 orangefs_convert_time_field(const struct timespec *ts)
{