Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 3bc303c2 authored by Jeff Layton's avatar Jeff Layton Committed by Steve French
Browse files

cifs: convert oplock breaks to use slow_work facility (try #4) 



This is the fourth respin of the patch to convert oplock breaks to
use the slow_work facility.

A customer of ours was testing a backport of one of the earlier
patchsets, and hit a "Busy inodes after umount..." problem. An oplock
break job had raced with a umount, and the superblock got torn down and
its memory reused. When the oplock break job tried to dereference the
inode->i_sb, the kernel oopsed.

This patchset has the oplock break job hold an inode and vfsmount
reference until the oplock break completes.  With this, there should be
no need to take a tcon reference (the vfsmount implicitly holds one
already).

Currently, when an oplock break comes in there's a chance that the
oplock break job won't occur if the allocation of the oplock_q_entry
fails. There are also some rather nasty races in the allocation and
handling these structs.

Rather than allocating oplock queue entries when an oplock break comes
in, add a few extra fields to the cifsFileInfo struct. Get rid of the
dedicated cifs_oplock_thread as well and queue the oplock break job to
the slow_work thread pool.

This approach also has the advantage that the oplock break jobs can
potentially run in parallel rather than be serialized like they are
today.

Signed-off-by: default avatarJeff Layton <jlayton@redhat.com>
Signed-off-by: default avatarSteve French <sfrench@us.ibm.com>
parent 48541bd3

fs/cifs/Kconfig

+1 −0
Original line number Diff line number Diff line
@@ -2,6 +2,7 @@ config CIFS 
	tristate "CIFS support (advanced network filesystem, SMBFS successor)"

	depends on INET

	select NLS

	select SLOW_WORK

	help

	  This is the client VFS module for the Common Internet File System

	  (CIFS) protocol which is the successor to the Server Message Block

fs/cifs/cifsfs.c

+5 −90
Original line number Diff line number Diff line
@@ -64,9 +64,6 @@ unsigned int multiuser_mount = 0; 
unsigned int extended_security = CIFSSEC_DEF;

/* unsigned int ntlmv2_support = 0; */

unsigned int sign_CIFS_PDUs = 1;

extern struct task_struct *oplockThread; /* remove sparse warning */

struct task_struct *oplockThread = NULL;

/* extern struct task_struct * dnotifyThread; remove sparse warning */

static const struct super_operations cifs_super_ops;

unsigned int CIFSMaxBufSize = CIFS_MAX_MSGSIZE;

module_param(CIFSMaxBufSize, int, 0);
@@ -973,89 +970,12 @@ cifs_destroy_mids(void) 
	kmem_cache_destroy(cifs_oplock_cachep);

}



static int cifs_oplock_thread(void *dummyarg)

{

	struct oplock_q_entry *oplock_item;

	struct cifsTconInfo *pTcon;

	struct inode *inode;

	__u16  netfid;

	int rc, waitrc = 0;



	set_freezable();

	do {

		if (try_to_freeze())

			continue;



		spin_lock(&cifs_oplock_lock);

		if (list_empty(&cifs_oplock_list)) {

			spin_unlock(&cifs_oplock_lock);

			set_current_state(TASK_INTERRUPTIBLE);

			schedule_timeout(39*HZ);

		} else {

			oplock_item = list_entry(cifs_oplock_list.next,

						struct oplock_q_entry, qhead);

			cFYI(1, ("found oplock item to write out"));

			pTcon = oplock_item->tcon;

			inode = oplock_item->pinode;

			netfid = oplock_item->netfid;

			spin_unlock(&cifs_oplock_lock);

			DeleteOplockQEntry(oplock_item);

			/* can not grab inode sem here since it would

				deadlock when oplock received on delete

				since vfs_unlink holds the i_mutex across

				the call */

			/* mutex_lock(&inode->i_mutex);*/

			if (S_ISREG(inode->i_mode)) {

#ifdef CONFIG_CIFS_EXPERIMENTAL

				if (CIFS_I(inode)->clientCanCacheAll == 0)

					break_lease(inode, FMODE_READ);

				else if (CIFS_I(inode)->clientCanCacheRead == 0)

					break_lease(inode, FMODE_WRITE);

#endif

				rc = filemap_fdatawrite(inode->i_mapping);

				if (CIFS_I(inode)->clientCanCacheRead == 0) {

					waitrc = filemap_fdatawait(

							      inode->i_mapping);

					invalidate_remote_inode(inode);

				}

				if (rc == 0)

					rc = waitrc;

			} else

				rc = 0;

			/* mutex_unlock(&inode->i_mutex);*/

			if (rc)

				CIFS_I(inode)->write_behind_rc = rc;

			cFYI(1, ("Oplock flush inode %p rc %d",

				inode, rc));



				/* releasing stale oplock after recent reconnect

				of smb session using a now incorrect file

				handle is not a data integrity issue but do

				not bother sending an oplock release if session

				to server still is disconnected since oplock

				already released by the server in that case */

			if (!pTcon->need_reconnect) {

				rc = CIFSSMBLock(0, pTcon, netfid,

						0 /* len */ , 0 /* offset */, 0,

						0, LOCKING_ANDX_OPLOCK_RELEASE,

						false /* wait flag */);

				cFYI(1, ("Oplock release rc = %d", rc));

			}

			set_current_state(TASK_INTERRUPTIBLE);

			schedule_timeout(1);  /* yield in case q were corrupt */

		}

	} while (!kthread_should_stop());



	return 0;

}



static int __init

init_cifs(void)

{

	int rc = 0;

	cifs_proc_init();

	INIT_LIST_HEAD(&cifs_tcp_ses_list);

	INIT_LIST_HEAD(&cifs_oplock_list);

#ifdef CONFIG_CIFS_EXPERIMENTAL

	INIT_LIST_HEAD(&GlobalDnotifyReqList);

	INIT_LIST_HEAD(&GlobalDnotifyRsp_Q);
@@ -1084,7 +1004,6 @@ init_cifs(void) 
	rwlock_init(&GlobalSMBSeslock);

	rwlock_init(&cifs_tcp_ses_lock);

	spin_lock_init(&GlobalMid_Lock);

	spin_lock_init(&cifs_oplock_lock);



	if (cifs_max_pending < 2) {

		cifs_max_pending = 2;
@@ -1119,18 +1038,15 @@ init_cifs(void) 
	if (rc)

		goto out_unregister_key_type;

#endif

	oplockThread = kthread_run(cifs_oplock_thread, NULL, "cifsoplockd");

	if (IS_ERR(oplockThread)) {

		rc = PTR_ERR(oplockThread);

		cERROR(1, ("error %d create oplock thread", rc));

		goto out_unregister_dfs_key_type;

	}

	rc = slow_work_register_user();

	if (rc)

		goto out_unregister_resolver_key;



	return 0;



 out_unregister_dfs_key_type:

#ifdef CONFIG_CIFS_DFS_UPCALL

 out_unregister_resolver_key:

	unregister_key_type(&key_type_dns_resolver);

#ifdef CONFIG_CIFS_DFS_UPCALL

 out_unregister_key_type:

#endif

#ifdef CONFIG_CIFS_UPCALL
@@ -1165,7 +1081,6 @@ exit_cifs(void) 
	cifs_destroy_inodecache();

	cifs_destroy_mids();

	cifs_destroy_request_bufs();

	kthread_stop(oplockThread);

}



MODULE_AUTHOR("Steve French <sfrench@us.ibm.com>");

fs/cifs/cifsglob.h

+5 −7
Original line number Diff line number Diff line
@@ -18,6 +18,7 @@ 
 */

#include <linux/in.h>

#include <linux/in6.h>

#include <linux/slow-work.h>

#include "cifs_fs_sb.h"

#include "cifsacl.h"

/*
@@ -346,14 +347,16 @@ struct cifsFileInfo { 
	/* lock scope id (0 if none) */

	struct file *pfile; /* needed for writepage */

	struct inode *pInode; /* needed for oplock break */

	struct vfsmount *mnt;

	struct mutex lock_mutex;

	struct list_head llist; /* list of byte range locks we have. */

	bool closePend:1;	/* file is marked to close */

	bool invalidHandle:1;	/* file closed via session abend */

	bool messageMode:1;	/* for pipes: message vs byte mode */

	bool oplock_break_cancelled:1;

	atomic_t count;		/* reference count */

	struct mutex fh_mutex; /* prevents reopen race after dead ses*/

	struct cifs_search_info srch_inf;

	struct slow_work oplock_break; /* slow_work job for oplock breaks */

};



/* Take a reference on the file private data */
@@ -670,12 +673,6 @@ GLOBAL_EXTERN rwlock_t cifs_tcp_ses_lock; 
 */

GLOBAL_EXTERN rwlock_t GlobalSMBSeslock;



/* Global list of oplocks */

GLOBAL_EXTERN struct list_head cifs_oplock_list;



/* Protects the cifs_oplock_list */

GLOBAL_EXTERN spinlock_t cifs_oplock_lock;



/* Outstanding dir notify requests */

GLOBAL_EXTERN struct list_head GlobalDnotifyReqList;

/* DirNotify response queue */
@@ -726,3 +723,4 @@ GLOBAL_EXTERN unsigned int cifs_min_rcv; /* min size of big ntwrk buf pool */ 
GLOBAL_EXTERN unsigned int cifs_min_small;  /* min size of small buf pool */

GLOBAL_EXTERN unsigned int cifs_max_pending; /* MAX requests at once to server*/



extern const struct slow_work_ops cifs_oplock_break_ops;

fs/cifs/cifsproto.h

+1 −5
Original line number Diff line number Diff line
@@ -86,17 +86,13 @@ extern int CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, 
			     const int stage,

			     const struct nls_table *nls_cp);

extern __u16 GetNextMid(struct TCP_Server_Info *server);

extern struct oplock_q_entry *AllocOplockQEntry(struct inode *, u16,

						 struct cifsTconInfo *);

extern void DeleteOplockQEntry(struct oplock_q_entry *);

extern void DeleteTconOplockQEntries(struct cifsTconInfo *);

extern struct timespec cifs_NTtimeToUnix(__le64 utc_nanoseconds_since_1601);

extern u64 cifs_UnixTimeToNT(struct timespec);

extern struct timespec cnvrtDosUnixTm(__le16 le_date, __le16 le_time,

				      int offset);



extern int cifs_posix_open(char *full_path, struct inode **pinode,

			   struct super_block *sb, int mode, int oflags,

			   struct vfsmount *mnt, int mode, int oflags,

			   __u32 *poplock, __u16 *pnetfid, int xid);

extern void cifs_unix_basic_to_fattr(struct cifs_fattr *fattr,

				     FILE_UNIX_BASIC_INFO *info,

fs/cifs/cifssmb.c

+1 −0
Original line number Diff line number Diff line
@@ -94,6 +94,7 @@ static void mark_open_files_invalid(struct cifsTconInfo *pTcon) 
	list_for_each_safe(tmp, tmp1, &pTcon->openFileList) {

		open_file = list_entry(tmp, struct cifsFileInfo, tlist);

		open_file->invalidHandle = true;

		open_file->oplock_break_cancelled = true;

	}

	write_unlock(&GlobalSMBSeslock);

	/* BB Add call to invalidate_inodes(sb) for all superblocks mounted