Re: [RFC] [WIP] Unbork fs.h, 1 of 3

From: Daniel Phillips (phillips@bonn-fries.net)
Date: Mon Dec 31 2001 - 03:41:04 EST


Fixes an error at line 381 of hostfs_kernel.c, spotted by Andreas Dilger, thanks.
Also turned linewrap off this time ;)

Daniel

--- ../2.4.16.uml.clean/arch/um/fs/hostfs/hostfs_kern.c Mon Dec 31 03:15:49 2001
+++ ./arch/um/fs/hostfs/hostfs_kern.c Mon Dec 31 08:19:46 2001
@@ -378,7 +378,7 @@
         char *name;
         int type, err = 0, rdev;
 
- inode = get_empty_inode();
+ inode = get_empty_inode(sb);
         if(inode == NULL) return(NULL);
         inode->u.hostfs_i.host_filename = NULL;
         inode->u.hostfs_i.fd = -1;
@@ -395,7 +395,6 @@
                 kfree(name);
         }
         else type = HOSTFS_DIR;
- inode->i_sb = sb;
 
         if(type == HOSTFS_SYMLINK)
                 inode->i_op = &page_symlink_inode_operations;
--- ../2.4.16.uml.clean/drivers/block/rd.c Mon Dec 31 02:55:36 2001
+++ ./drivers/block/rd.c Mon Dec 31 02:31:21 2001
@@ -673,7 +673,7 @@
 #endif
         ram_device = MKDEV(MAJOR_NR, unit);
 
- if ((inode = get_empty_inode()) == NULL)
+ if ((inode = get_empty_inode(NULL)) == NULL)
                 return;
         memset(&infile, 0, sizeof(infile));
         memset(&in_dentry, 0, sizeof(in_dentry));
@@ -683,7 +683,7 @@
         infile.f_op = &def_blk_fops;
         init_special_inode(inode, S_IFBLK | S_IRUSR, kdev_t_to_nr(device));
 
- if ((out_inode = get_empty_inode()) == NULL)
+ if ((out_inode = get_empty_inode(NULL)) == NULL)
                 goto free_inode;
         memset(&outfile, 0, sizeof(outfile));
         memset(&out_dentry, 0, sizeof(out_dentry));
--- ../2.4.16.uml.clean/fs/inode.c Mon Dec 31 02:55:36 2001
+++ ./fs/inode.c Mon Dec 31 02:31:21 2001
@@ -75,16 +75,23 @@
 
 static kmem_cache_t * inode_cachep;
 
-#define alloc_inode() \
- ((struct inode *) kmem_cache_alloc(inode_cachep, SLAB_KERNEL))
-static void destroy_inode(struct inode *inode)
+int foo_count;
+
+static inline struct inode *alloc_inode (struct super_block *sb)
 {
+ kmem_cache_t *cache = sb? sb->s_type->inode_cache: NULL;
+ return (struct inode *) kmem_cache_alloc (cache? cache: inode_cachep, SLAB_KERNEL);
+}
+
+static void destroy_inode (struct inode *inode)
+{
+ struct super_block *sb = inode->i_sb;
+ kmem_cache_t *cache = sb? sb->s_type->inode_cache: NULL;
         if (inode_has_buffers(inode))
                 BUG();
- kmem_cache_free(inode_cachep, (inode));
+ kmem_cache_free (cache? cache: inode_cachep, inode);
 }
 
-
 /*
  * These are initializations that only need to be done
  * once, because the fields are idempotent across use
@@ -661,7 +668,7 @@
          !inode_has_buffers(inode))
 #define INODE(entry) (list_entry(entry, struct inode, i_list))
 
-void prune_icache(int goal)
+void prune_icache (int goal)
 {
         LIST_HEAD(list);
         struct list_head *entry, *freeable = &list;
@@ -710,6 +717,7 @@
 
 int shrink_icache_memory(int priority, int gfp_mask)
 {
+ struct file_system_type *fs;
         int count = 0;
 
         /*
@@ -725,7 +733,15 @@
         count = inodes_stat.nr_unused / priority;
 
         prune_icache(count);
+#if 0
+ /* Manfred thinks this isn't necessary */
         kmem_cache_shrink(inode_cachep);
+ write_lock(&file_systems_lock);
+ for (fs = file_systems; fs; fs = fs->next)
+ kmem_cache_shrink (fs->inode_cache);
+ write_unlock(&file_systems_lock);
+ kmem_cache_shrink(inode_cachep);
+#endif
         return 0;
 }
 
@@ -765,12 +781,14 @@
  * i_sb, i_ino, i_count, i_state and the lists have
  * been initialized elsewhere..
  */
-static void clean_inode(struct inode *inode)
+static void clean_inode (struct super_block *sb, struct inode *inode)
 {
         static struct address_space_operations empty_aops;
         static struct inode_operations empty_iops;
         static struct file_operations empty_fops;
- memset(&inode->u, 0, sizeof(inode->u));
+ unsigned given = sb? sb->s_type->inode_private: 0; // only rd.c has null sb
+ memset(&inode->u, 0, given? given: sizeof(inode->u));
+ inode->i_sb = sb;
         inode->i_sock = 0;
         inode->i_op = &empty_iops;
         inode->i_fop = &empty_fops;
@@ -802,20 +820,19 @@
  * lists.
  */
  
-struct inode * get_empty_inode(void)
+struct inode *get_empty_inode (struct super_block *sb)
 {
         static unsigned long last_ino;
- struct inode * inode;
+ struct inode *inode;
 
         spin_lock_prefetch(&inode_lock);
-
- inode = alloc_inode();
+ inode = alloc_inode (sb);
         if (inode)
         {
                 spin_lock(&inode_lock);
                 inodes_stat.nr_inodes++;
                 list_add(&inode->i_list, &inode_in_use);
- inode->i_sb = NULL;
+ inode->i_sb = NULL; // need this?
                 inode->i_dev = 0;
                 inode->i_blkbits = 0;
                 inode->i_ino = ++last_ino;
@@ -823,7 +840,7 @@
                 atomic_set(&inode->i_count, 1);
                 inode->i_state = 0;
                 spin_unlock(&inode_lock);
- clean_inode(inode);
+ clean_inode(sb, inode);
         }
         return inode;
 }
@@ -838,7 +855,7 @@
 {
         struct inode * inode;
 
- inode = alloc_inode();
+ inode = alloc_inode (sb);
         if (inode) {
                 struct inode * old;
 
@@ -849,7 +866,6 @@
                         inodes_stat.nr_inodes++;
                         list_add(&inode->i_list, &inode_in_use);
                         list_add(&inode->i_hash, head);
- inode->i_sb = sb;
                         inode->i_dev = sb->s_dev;
                         inode->i_blkbits = sb->s_blocksize_bits;
                         inode->i_ino = ino;
@@ -857,8 +873,7 @@
                         atomic_set(&inode->i_count, 1);
                         inode->i_state = I_LOCK;
                         spin_unlock(&inode_lock);
-
- clean_inode(inode);
+ clean_inode(sb, inode);
 
                         /* reiserfs specific hack right here. We don't
                         ** want this to last, and are looking for VFS changes
@@ -897,6 +912,22 @@
                 wait_on_inode(inode);
         }
         return inode;
+}
+
+int create_inode_cache (struct file_system_type *fs)
+{
+ if (fs->inode_private)
+ if (!(fs->inode_cache = kmem_cache_create (fs->name,
+ fs->inode_private + sizeof(struct inode) - sizeof(get_empty_inode(0)->u),
+ 0, SLAB_HWCACHE_ALIGN, init_once, NULL)))
+ return -ENOSPC;
+ return 0;
+}
+
+int destroy_inode_cache (struct file_system_type *fs)
+{
+ printk("Destroying inode cache for %s\n", fs->name);
+ return kmem_cache_destroy (fs->inode_cache)? -EBUSY: 0;
 }
 
 static inline unsigned long hash(struct super_block *sb, unsigned long i_ino)
--- ../2.4.16.uml.clean/fs/super.c Mon Dec 31 02:55:36 2001
+++ ./fs/super.c Mon Dec 31 02:31:21 2001
@@ -67,8 +67,8 @@
  * Once the reference is obtained we can drop the spinlock.
  */
 
-static struct file_system_type *file_systems;
-static rwlock_t file_systems_lock = RW_LOCK_UNLOCKED;
+struct file_system_type *file_systems;
+rwlock_t file_systems_lock = RW_LOCK_UNLOCKED;
 
 /* WARNING: This can be used only if we _already_ own a reference */
 static void get_filesystem(struct file_system_type *fs)
@@ -105,10 +105,10 @@
  * unregistered.
  */
  
-int register_filesystem(struct file_system_type * fs)
+int register_filesystem (struct file_system_type *fs)
 {
- int res = 0;
- struct file_system_type ** p;
+ struct file_system_type **p;
+ int err = 0;
 
         if (!fs)
                 return -EINVAL;
@@ -118,11 +118,12 @@
         write_lock(&file_systems_lock);
         p = find_filesystem(fs->name);
         if (*p)
- res = -EBUSY;
+ err = -EBUSY;
         else
- *p = fs;
+ if (!(err = create_inode_cache (fs)))
+ *p = fs;
         write_unlock(&file_systems_lock);
- return res;
+ return err;
 }
 
 /**
@@ -137,23 +138,25 @@
  * may be freed or reused.
  */
  
-int unregister_filesystem(struct file_system_type * fs)
+int unregister_filesystem (struct file_system_type *fs)
 {
- struct file_system_type ** tmp;
-
+ struct file_system_type **p;
+ int err = -EINVAL;
         write_lock(&file_systems_lock);
- tmp = &file_systems;
- while (*tmp) {
- if (fs == *tmp) {
- *tmp = fs->next;
+ p = &file_systems;
+ while (*p) {
+ if (*p == fs) {
+ if (fs->inode_cache && (err = destroy_inode_cache (fs)))
+ break;
+ *p = fs->next;
                         fs->next = NULL;
- write_unlock(&file_systems_lock);
- return 0;
+ err = 0;
+ break;
                 }
- tmp = &(*tmp)->next;
+ p = &(*p)->next;
         }
         write_unlock(&file_systems_lock);
- return -EINVAL;
+ return err;
 }
 
 static int fs_index(const char * __name)
--- ../2.4.16.uml.clean/include/linux/fs.h Mon Dec 31 02:55:36 2001
+++ ./include/linux/fs.h Mon Dec 31 08:13:45 2001
@@ -693,6 +693,9 @@
 #include <linux/cramfs_fs_sb.h>
 #include <linux/jffs2_fs_sb.h>
 
+extern struct file_system_type *file_systems;
+extern rwlock_t file_systems_lock;
+
 extern struct list_head super_blocks;
 extern spinlock_t sb_lock;
 
@@ -952,16 +955,28 @@
         struct module *owner;
         struct file_system_type * next;
         struct list_head fs_supers;
+ unsigned super_private, inode_private;
+ struct kmem_cache_s *inode_cache;
 };
 
-#define DECLARE_FSTYPE(var,type,read,flags) \
+#define FILESYSTEM_TYPE(var,type,read,ssize,isize,flags) \
 struct file_system_type var = { \
         name: type, \
         read_super: read, \
+ super_private: ssize, \
+ inode_private: isize, \
         fs_flags: flags, \
         owner: THIS_MODULE, \
 }
 
+#define FILESYSTEM(var,type,read,ssize,isize) \
+ FILESYSTEM_TYPE(var,type,read,ssize,isize,FS_REQUIRES_DEV)
+
+/* Backward compatible declarations, remove when all updated */
+
+#define DECLARE_FSTYPE(var,type,read,flags) \
+ FILESYSTEM_TYPE(var,type,read,0,0,flags)
+
 #define DECLARE_FSTYPE_DEV(var,type,read) \
         DECLARE_FSTYPE(var,type,read,FS_REQUIRES_DEV)
 
@@ -1327,18 +1342,21 @@
 }
 
 extern void clear_inode(struct inode *);
-extern struct inode * get_empty_inode(void);
+extern struct inode *get_empty_inode (struct super_block *sb);
 
-static inline struct inode * new_inode(struct super_block *sb)
+static inline struct inode *new_inode (struct super_block *sb)
 {
- struct inode *inode = get_empty_inode();
+ struct inode *inode = get_empty_inode(sb);
         if (inode) {
- inode->i_sb = sb;
                 inode->i_dev = sb->s_dev;
                 inode->i_blkbits = sb->s_blocksize_bits;
         }
         return inode;
 }
+
+int create_inode_cache (struct file_system_type *fs);
+int destroy_inode_cache (struct file_system_type *fs);
+
 extern void remove_suid(struct inode *inode);
 
 extern void insert_inode_hash(struct inode *);
--- ../2.4.16.uml.clean/net/socket.c Mon Dec 31 02:55:36 2001
+++ ./net/socket.c Mon Dec 31 02:31:21 2001
@@ -440,11 +440,10 @@
         struct inode * inode;
         struct socket * sock;
 
- inode = get_empty_inode();
+ inode = get_empty_inode(sock_mnt->mnt_sb);
         if (!inode)
                 return NULL;
 
- inode->i_sb = sock_mnt->mnt_sb;
         sock = socki_lookup(inode);
 
         inode->i_mode = S_IFSOCK|S_IRWXUGO;
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Mon Dec 31 2001 - 21:00:24 EST