Add strict cache mode for 2.6.32

sources/2.6.32/README

@@ -433,6 +433,11 @@ A partial list of the supported mount options follows:
 		if oplock (caching token) is granted and held. Note that
 		direct allows write operations larger than page size
 		to be sent to the server.
+  strictcache   Use for switching on strict cache mode. In this mode the
+		client reads from the cache all the time it has Oplock Level II,
+		otherwise - reads from the server. All written data are stored
+		in the cache, but if the client doesn't have Exclusive Oplock,
+		it writes the data to the server.
   acl   	Allow setfacl and getfacl to manage posix ACLs if server
 		supports them.  (default)
   noacl 	Do not allow setfacl and getfacl calls on this mount

sources/2.6.32/cifs_fs_sb.h

@@ -34,6 +34,7 @@
 #define CIFS_MOUNT_NOPOSIXBRL   0x2000 /* mandatory not posix byte range lock */
 #define CIFS_MOUNT_NOSSYNC      0x4000 /* don't do slow SMBflush on every sync*/
 #define CIFS_MOUNT_WINE_MODE    0x8000 /* use pid forwarding for wine apps    */
+#define CIFS_MOUNT_STRICT_IO	0x10000 /* strict cache mode */
 
 struct cifs_sb_info {
 	struct cifsTconInfo *tcon;	/* primary mount */

sources/2.6.32/cifsfs.c

@@ -309,8 +309,7 @@ cifs_alloc_inode(struct super_block *sb)
 	/* Until the file is open and we have gotten oplock
 	info back from the server, can not assume caching of
 	file data or metadata */
-	cifs_inode->clientCanCacheRead = false;
-	cifs_inode->clientCanCacheAll = false;
+	cifs_set_oplock_level(cifs_inode, 0);
 	cifs_inode->delete_pending = false;
 	cifs_inode->vfs_inode.i_blkbits = 14;  /* 2**14 = CIFS_MAX_MSGSIZE */
 	cifs_inode->server_eof = 0;
@@ -628,10 +627,17 @@ static ssize_t cifs_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
 {
 	struct inode *inode = iocb->ki_filp->f_path.dentry->d_inode;
 	ssize_t written;
+	int rc;
 
 	written = generic_file_aio_write(iocb, iov, nr_segs, pos);
-	if (!CIFS_I(inode)->clientCanCacheAll)
-		filemap_fdatawrite(inode->i_mapping);
+
+	if (CIFS_I(inode)->clientCanCacheAll)
+		return written;
+
+	rc = filemap_fdatawrite(inode->i_mapping);
+	if (rc)
+		cFYI(1, ("cifs_file_aio_write: %d rc on %p inode", rc, inode));
+
 	return written;
 }
 
@@ -765,6 +771,25 @@ const struct file_operations cifs_file_ops = {
 #endif /* CONFIG_CIFS_EXPERIMENTAL */
 };
 
+const struct file_operations cifs_file_strict_ops = {
+	.read = do_sync_read,
+	.write = do_sync_write,
+	.aio_read = cifs_strict_readv,
+	.aio_write = cifs_strict_writev,
+	.open = cifs_open,
+	.release = cifs_close,
+	.lock = cifs_lock,
+	.fsync = cifs_strict_fsync,
+	.flush = cifs_flush,
+	.mmap = cifs_file_strict_mmap,
+	.splice_read = generic_file_splice_read,
+	.llseek = cifs_llseek,
+#ifdef CONFIG_CIFS_POSIX
+	.unlocked_ioctl	= cifs_ioctl,
+#endif /* CONFIG_CIFS_POSIX */
+	.setlease = cifs_setlease,
+};
+
 const struct file_operations cifs_file_direct_ops = {
 	/* no aio, no readv -
 	   BB reevaluate whether they can be done with directio, no cache */
@@ -785,6 +810,7 @@ const struct file_operations cifs_file_direct_ops = {
 	.setlease = cifs_setlease,
 #endif /* CONFIG_CIFS_EXPERIMENTAL */
 };
+
 const struct file_operations cifs_file_nobrl_ops = {
 	.read = do_sync_read,
 	.write = do_sync_write,
@@ -806,6 +832,24 @@ const struct file_operations cifs_file_nobrl_ops = {
 #endif /* CONFIG_CIFS_EXPERIMENTAL */
 };
 
+const struct file_operations cifs_file_strict_nobrl_ops = {
+	.read = do_sync_read,
+	.write = do_sync_write,
+	.aio_read = cifs_strict_readv,
+	.aio_write = cifs_strict_writev,
+	.open = cifs_open,
+	.release = cifs_close,
+	.fsync = cifs_strict_fsync,
+	.flush = cifs_flush,
+	.mmap = cifs_file_strict_mmap,
+	.splice_read = generic_file_splice_read,
+	.llseek = cifs_llseek,
+#ifdef CONFIG_CIFS_POSIX
+	.unlocked_ioctl	= cifs_ioctl,
+#endif /* CONFIG_CIFS_POSIX */
+	.setlease = cifs_setlease,
+};
+
 const struct file_operations cifs_file_direct_nobrl_ops = {
 	/* no aio, no readv -
 	   BB reevaluate whether they can be done with directio, no cache */

sources/2.6.32/cifsfs.h

@@ -62,6 +62,7 @@ extern int cifs_rmdir(struct inode *, struct dentry *);
 extern int cifs_rename(struct inode *, struct dentry *, struct inode *,
 		       struct dentry *);
 extern int cifs_revalidate(struct dentry *);
+extern void cifs_invalidate_mapping(struct inode *inode);
 extern int cifs_getattr(struct vfsmount *, struct dentry *, struct kstat *);
 extern int cifs_setattr(struct dentry *, struct iattr *);
 
@@ -73,19 +74,27 @@ extern const struct inode_operations cifs_dfs_referral_inode_operations;
 /* Functions related to files and directories */
 extern const struct file_operations cifs_file_ops;
 extern const struct file_operations cifs_file_direct_ops; /* if directio mnt */
-extern const struct file_operations cifs_file_nobrl_ops;
-extern const struct file_operations cifs_file_direct_nobrl_ops; /* no brlocks */
+extern const struct file_operations cifs_file_strict_ops; /* if strictio mnt */
+extern const struct file_operations cifs_file_nobrl_ops; /* no brlocks */
+extern const struct file_operations cifs_file_direct_nobrl_ops;
+extern const struct file_operations cifs_file_strict_nobrl_ops;
 extern int cifs_open(struct inode *inode, struct file *file);
 extern int cifs_close(struct inode *inode, struct file *file);
 extern int cifs_closedir(struct inode *inode, struct file *file);
 extern ssize_t cifs_user_read(struct file *file, char __user *read_data,
-			 size_t read_size, loff_t *poffset);
+			      size_t read_size, loff_t *poffset);
+extern ssize_t cifs_strict_readv(struct kiocb *iocb, const struct iovec *iov,
+				 unsigned long nr_segs, loff_t pos);
 extern ssize_t cifs_user_write(struct file *file, const char __user *write_data,
-			 size_t write_size, loff_t *poffset);
+			       size_t write_size, loff_t *poffset);
+extern ssize_t cifs_strict_writev(struct kiocb *iocb, const struct iovec *iov,
+				  unsigned long nr_segs, loff_t pos);
 extern int cifs_lock(struct file *, int, struct file_lock *);
 extern int cifs_fsync(struct file *, struct dentry *, int);
+extern int cifs_strict_fsync(struct file *, struct dentry *, int);
 extern int cifs_flush(struct file *, fl_owner_t id);
 extern int cifs_file_mmap(struct file * , struct vm_area_struct *);
+extern int cifs_file_strict_mmap(struct file * , struct vm_area_struct *);
 extern const struct file_operations cifs_dir_ops;
 extern int cifs_dir_open(struct inode *inode, struct file *file);
 extern int cifs_readdir(struct file *file, void *direntry, filldir_t filldir);

sources/2.6.32/cifsglob.h

@@ -388,6 +388,7 @@ struct cifsInodeInfo {
 	bool clientCanCacheRead:1;	/* read oplock */
 	bool clientCanCacheAll:1;	/* read and writebehind oplock */
 	bool delete_pending:1;		/* DELETE_ON_CLOSE is set */
+	bool invalid_mapping:1;		/* uptodate cache flag */
 	u64  server_eof;		/* current file size on server */
 	u64  uniqueid;			/* server inode number */
 	struct inode vfs_inode;

sources/2.6.32/cifsproto.h

@@ -70,6 +70,8 @@ extern struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *);
 #ifdef CONFIG_CIFS_EXPERIMENTAL
 extern struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *);
 #endif
+extern void cifs_update_eof(struct cifsInodeInfo *cifsi, loff_t offset,
+			    unsigned int bytes_written);
 extern unsigned int smbCalcSize(struct smb_hdr *ptr);
 extern unsigned int smbCalcSize_LE(struct smb_hdr *ptr);
 extern int decode_negTokenInit(unsigned char *security_blob, int length,
@@ -90,10 +92,12 @@ 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 void cifs_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock);
 
 extern struct cifsFileInfo *cifs_new_fileinfo(struct inode *newinode,
 				__u16 fileHandle, struct file *file,
-				struct vfsmount *mnt, unsigned int oflags);
+				struct vfsmount *mnt, unsigned int oflags,
+				__u32 oplock);
 extern int cifs_posix_open(char *full_path, struct inode **pinode,
 				struct vfsmount *mnt,
 				struct super_block *sb,

sources/2.6.32/connect.c

@@ -82,6 +82,7 @@ struct smb_vol {
 	bool no_xattr:1;   /* set if xattr (EA) support should be disabled*/
 	bool server_ino:1; /* use inode numbers from server ie UniqueId */
 	bool direct_io:1;
+	bool strict_io:1; /* strict cache behavior */
 	bool remap:1;      /* set to remap seven reserved chars in filenames */
 	bool posix_paths:1; /* unset to not ask for posix pathnames. */
 	bool no_linux_ext:1;
@@ -1303,7 +1304,7 @@ cifs_parse_mount_options(char *options, const char *devname,
 		} else if (strnicmp(data, "wine", 4) == 0) {
 			vol->wine_mode = 1;
 			vol->mand_lock = 1;
-			vol->direct_io = 1;
+			vol->strict_io = 1;
 		} else if (strnicmp(data, "cifsacl", 7) == 0) {
 			vol->cifs_acl = 1;
 		} else if (strnicmp(data, "nocifsacl", 9) == 0) {
@@ -1328,6 +1329,8 @@ cifs_parse_mount_options(char *options, const char *devname,
 			vol->direct_io = 1;
 		} else if (strnicmp(data, "forcedirectio", 13) == 0) {
 			vol->direct_io = 1;
+		} else if (strnicmp(data, "strictcache", 11) == 0) {
+			vol->strict_io = 1;
 		} else if (strnicmp(data, "noac", 4) == 0) {
 			printk(KERN_WARNING "CIFS: Mount option noac not "
 				"supported. Instead set "
@@ -2191,6 +2194,8 @@ static void setup_cifs_sb(struct smb_vol *pvolume_info,
 		cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_OVERR_GID;
 	if (pvolume_info->dynperm)
 		cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_DYNPERM;
+	if (pvolume_info->strict_io)
+		cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_STRICT_IO;
 	if (pvolume_info->direct_io) {
 		cFYI(1, ("mounting share using direct i/o"));
 		cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_DIRECT_IO;

sources/2.6.32/dir.c

@@ -132,9 +132,9 @@ cifs_bp_rename_retry:
 
 struct cifsFileInfo *
 cifs_new_fileinfo(struct inode *newinode, __u16 fileHandle,
-		  struct file *file, struct vfsmount *mnt, unsigned int oflags)
+		  struct file *file, struct vfsmount *mnt, unsigned int oflags,
+		  __u32 oplock)
 {
-	int oplock = 0;
 	struct cifsFileInfo *pCifsFile;
 	struct cifsInodeInfo *pCifsInode;
 	struct cifs_sb_info *cifs_sb = CIFS_SB(mnt->mnt_sb);
@@ -143,9 +143,6 @@ cifs_new_fileinfo(struct inode *newinode, __u16 fileHandle,
 	if (pCifsFile == NULL)
 		return pCifsFile;
 
-	if (oplockEnabled)
-		oplock = REQ_OPLOCK;
-
 	pCifsFile->netfid = fileHandle;
 	pCifsFile->pid = current->tgid;
 	pCifsFile->pInode = igrab(newinode);
@@ -159,9 +156,12 @@ cifs_new_fileinfo(struct inode *newinode, __u16 fileHandle,
 	atomic_set(&pCifsFile->count, 1);
 	slow_work_init(&pCifsFile->oplock_break, &cifs_oplock_break_ops);
 
+	pCifsInode = CIFS_I(newinode);
+	if (pCifsInode)
+		cifs_set_oplock_level(pCifsInode, oplock);
 	write_lock(&GlobalSMBSeslock);
 	list_add(&pCifsFile->tlist, &cifs_sb->tcon->openFileList);
-	pCifsInode = CIFS_I(newinode);
+
 	if (pCifsInode) {
 		/* if readable file instance put first in list*/
 		if (oflags & FMODE_READ)
@@ -169,13 +169,6 @@ cifs_new_fileinfo(struct inode *newinode, __u16 fileHandle,
 		else
 			list_add_tail(&pCifsFile->flist,
 				      &pCifsInode->openFileList);
-
-		if ((oplock & 0xF) == OPLOCK_EXCLUSIVE) {
-			pCifsInode->clientCanCacheAll = true;
-			pCifsInode->clientCanCacheRead = true;
-			cFYI(1, ("Exclusive Oplock inode %p", newinode));
-		} else if ((oplock & 0xF) == OPLOCK_READ)
-				pCifsInode->clientCanCacheRead = true;
 	}
 	write_unlock(&GlobalSMBSeslock);
 
@@ -252,7 +245,8 @@ int cifs_posix_open(char *full_path, struct inode **pinode,
 	}
 
 	if (mnt)
-		cifs_new_fileinfo(*pinode, *pnetfid, NULL, mnt, oflags);
+		cifs_new_fileinfo(*pinode, *pnetfid, NULL, mnt, oflags,
+				  *poplock);
 
 posix_open_ret:
 	kfree(presp_data);
@@ -468,7 +462,7 @@ cifs_create_set_dentry:
 		CIFSSMBClose(xid, tcon, fileHandle);
 	} else if (!(posix_create) && (newinode)) {
 			cifs_new_fileinfo(newinode, fileHandle, NULL,
-						nd->path.mnt, oflags);
+					  nd->path.mnt, oflags, oplock);
 	}
 cifs_create_out:
 	kfree(buf);

sources/2.6.32/inode.c

@@ -42,13 +42,17 @@ static void cifs_set_ops(struct inode *inode, const bool is_dfs_referral)
 				inode->i_fop = &cifs_file_direct_nobrl_ops;
 			else
 				inode->i_fop = &cifs_file_direct_ops;
+		} else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_STRICT_IO) {
+			if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL)
+				inode->i_fop = &cifs_file_strict_nobrl_ops;
+			else
+				inode->i_fop = &cifs_file_strict_ops;
 		} else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL)
 			inode->i_fop = &cifs_file_nobrl_ops;
 		else { /* not direct, send byte range locks */
 			inode->i_fop = &cifs_file_ops;
 		}
 
-
 		/* check if server can support readpages */
 		if (cifs_sb->tcon->ses->server->maxBuf <
 				PAGE_CACHE_SIZE + MAX_CIFS_HDR_SIZE)
@@ -1409,16 +1413,64 @@ cifs_rename_exit:
 	return rc;
 }
 
+static bool cifs_check_inval(struct inode *inode)
+{
+	struct cifsInodeInfo *cifs_i = CIFS_I(inode);
+
+	if (cifs_i->clientCanCacheRead)
+		return false;
+
+	if (!lookupCacheEnabled)
+		return true;
+
+	if (cifs_i->time == 0)
+		return true;
+
+	/* FIXME: the actimeo should be tunable */
+	if (time_after_eq(jiffies, cifs_i->time + HZ))
+		return true;
+
+	/* hardlinked files get "special" treatment */
+	if (S_ISREG(inode->i_mode) && inode->i_nlink != 1)
+		return true;
+
+	return false;
+}
+
+/*
+ * Zap the cache. Called when invalid_mapping flag is set.
+ */
+void
+cifs_invalidate_mapping(struct inode *inode)
+{
+	int rc;
+	struct cifsInodeInfo *cifs_i = CIFS_I(inode);
+
+	cifs_i->invalid_mapping = false;
+
+	/* write back any cached data */
+	if (inode->i_mapping && inode->i_mapping->nrpages != 0) {
+		rc = filemap_write_and_wait(inode->i_mapping);
+		if (rc)
+			cifs_i->write_behind_rc = rc;
+		rc = invalidate_inode_pages2(inode->i_mapping);
+		if (rc) {
+			cERROR(1, ("%s: could not invalidate inode %p",
+			       __func__, inode));
+			cifs_i->invalid_mapping = true;
+		}
+	}
+}
+
 int cifs_revalidate(struct dentry *direntry)
 {
 	int xid;
-	int rc = 0, wbrc = 0;
+	int rc = 0;
 	char *full_path;
 	struct cifs_sb_info *cifs_sb;
 	struct cifsInodeInfo *cifsInode;
 	loff_t local_size;
 	struct timespec local_mtime;
-	bool invalidate_inode = false;
 
 	if (direntry->d_inode == NULL)
 		return -ENOENT;
@@ -1428,10 +1480,6 @@ int cifs_revalidate(struct dentry *direntry)
 	if (cifsInode == NULL)
 		return -ENOENT;
 
-	/* no sense revalidating inode info on file that no one can write */
-	if (CIFS_I(direntry->d_inode)->clientCanCacheRead)
-		return rc;
-
 	xid = GetXid();
 
 	cifs_sb = CIFS_SB(direntry->d_sb);
@@ -1449,19 +1497,8 @@ int cifs_revalidate(struct dentry *direntry)
 		 direntry->d_inode->i_count.counter, direntry,
 		 direntry->d_time, jiffies));
 
-	if (cifsInode->time == 0) {
-		/* was set to zero previously to force revalidate */
-	} else if (time_before(jiffies, cifsInode->time + HZ) &&
-		   lookupCacheEnabled) {
-		if ((S_ISREG(direntry->d_inode->i_mode) == 0) ||
-		    (direntry->d_inode->i_nlink == 1)) {
-			kfree(full_path);
-			FreeXid(xid);
-			return rc;
-		} else {
-			cFYI(1, ("Have to revalidate file due to hardlinks"));
-		}
-	}
+	if (!cifs_check_inval(direntry->d_inode))
+		goto check_inval;
 
 	/* save mtime and size */
 	local_mtime = direntry->d_inode->i_mtime;
@@ -1495,48 +1532,16 @@ int cifs_revalidate(struct dentry *direntry)
 	    (local_size == direntry->d_inode->i_size)) {
 		cFYI(1, ("cifs_revalidate - inode unchanged"));
 	} else {
-		/* file may have changed on server */
-		if (cifsInode->clientCanCacheRead) {
-			/* no need to invalidate inode pages since we were the
-			   only ones who could have modified the file and the
-			   server copy is staler than ours */
-		} else {
-			invalidate_inode = true;
-		}
+		cifsInode->invalid_mapping = true;
 	}
 
 	/* can not grab this sem since kernel filesys locking documentation
 	   indicates i_mutex may be taken by the kernel on lookup and rename
 	   which could deadlock if we grab the i_mutex here as well */
+check_inval:
 /*	mutex_lock(&direntry->d_inode->i_mutex);*/
-	/* need to write out dirty pages here  */
-	if (direntry->d_inode->i_mapping) {
-		/* do we need to lock inode until after invalidate completes
-		   below? */
-		wbrc = filemap_fdatawrite(direntry->d_inode->i_mapping);
-		if (wbrc)
-			CIFS_I(direntry->d_inode)->write_behind_rc = wbrc;
-	}
-	if (invalidate_inode) {
-	/* shrink_dcache not necessary now that cifs dentry ops
-	are exported for negative dentries */
-/*		if (S_ISDIR(direntry->d_inode->i_mode))
-			shrink_dcache_parent(direntry); */
-		if (S_ISREG(direntry->d_inode->i_mode)) {
-			if (direntry->d_inode->i_mapping) {
-				wbrc = filemap_fdatawait(direntry->d_inode->i_mapping);
-				if (wbrc)
-					CIFS_I(direntry->d_inode)->write_behind_rc = wbrc;
-			}
-			/* may eventually have to do this for open files too */
-			if (list_empty(&(cifsInode->openFileList))) {
-				/* changed on server - flush read ahead pages */
-				cFYI(1, ("Invalidating read ahead data on "
-					 "closed file"));
-				invalidate_remote_inode(direntry->d_inode);
-			}
-		}
-	}
+	if (cifsInode->invalid_mapping)
+		cifs_invalidate_mapping(direntry->d_inode);
 /*	mutex_unlock(&direntry->d_inode->i_mutex); */
 
 	kfree(full_path);

sources/2.6.32/misc.c

@@ -569,6 +569,10 @@ is_valid_oplock_break(struct smb_hdr *buf, struct TCP_Server_Info *srv)
 				if (pSMB->Fid != netfile->netfid)
 					continue;
 
+				cFYI(1, ("file id match, oplock break"));
+				pCifsInode = CIFS_I(netfile->pInode);
+				cifs_set_oplock_level(pCifsInode,
+					pSMB->OplockLevel ? OPLOCK_READ : 0);
 				/*
 				 * don't do anything if file is about to be
 				 * closed anyway.
@@ -579,11 +583,6 @@ is_valid_oplock_break(struct smb_hdr *buf, struct TCP_Server_Info *srv)
 					return true;
 				}
 
-				cFYI(1, ("file id match, oplock break"));
-				pCifsInode = CIFS_I(netfile->pInode);
-				pCifsInode->clientCanCacheAll = false;
-				if (pSMB->OplockLevel == 0)
-					pCifsInode->clientCanCacheRead = false;
 				rc = slow_work_enqueue(&netfile->oplock_break);
 				if (rc) {
 					cERROR(1, ("failed to enqueue oplock "
@@ -729,3 +728,23 @@ cifs_autodisable_serverino(struct cifs_sb_info *cifs_sb)
 			   cifs_sb->tcon->treeName));
 	}
 }
+
+void cifs_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock)
+{
+	oplock &= 0xF;
+
+	if (oplock == OPLOCK_EXCLUSIVE) {
+		cinode->clientCanCacheAll = true;
+		cinode->clientCanCacheRead = true;
+		cFYI(1, ("Exclusive Oplock granted on inode %p",
+		     &cinode->vfs_inode));
+	} else if (oplock == OPLOCK_READ) {
+		cinode->clientCanCacheAll = false;
+		cinode->clientCanCacheRead = true;
+		cFYI(1, ("Level II Oplock granted on inode %p",
+		    &cinode->vfs_inode));
+	} else {
+		cinode->clientCanCacheAll = false;
+		cinode->clientCanCacheRead = false;
+	}
+}