Merged fs0 to mm0 for simpler progress on posix api.

mm0 and all other posix dependents are building ok.

conts/posix/mm0/include/bdev.h

@@ -0,0 +1,8 @@
+#ifndef __BDEV_H__
+#define __BDEV_H__
+/*
+ * This is a mock-up compiled in blockdev buffer, to be used temporarily.
+ */
+
+void *vfs_rootdev_open(void);
+#endif/* __BDEV_H__ */

conts/posix/mm0/include/fs.h

@@ -0,0 +1,171 @@
+/*
+ * VFS definitions.
+ *
+ * Copyright (C) 2007, 2008 Bahadir Balban.
+ */
+#ifndef __FS_H__
+#define __FS_H__
+
+#include <l4/lib/list.h>
+#include <l4/macros.h>
+#include <l4lib/types.h>
+#include <stat.h>
+#include <path.h>
+
+typedef void (*vnode_op_t)(void);
+typedef void (*file_op_t)(void);
+
+struct dentry;
+struct file;
+struct file_system_type;
+struct superblock;
+struct vnode;
+
+struct dentry_ops {
+	int (*compare)(struct dentry *d, const char *n);
+};
+
+/* Operations that work on file content */
+struct file_ops {
+
+	/* Read a vnode's contents by page range */
+	int (*read)(struct vnode *v, unsigned long pfn,
+		    unsigned long npages, void *buf);
+
+	/* Write a vnode's contents by page range */
+	int (*write)(struct vnode *v, unsigned long pfn,
+		    unsigned long npages, void *buf);
+	file_op_t open;
+	file_op_t close;
+	file_op_t mmap;
+	file_op_t lseek;
+	file_op_t flush;
+	file_op_t fsync;
+};
+
+/* Operations that work on vnode fields and associations between vnodes */
+struct vnode_ops {
+	vnode_op_t create;
+	struct vnode *(*lookup)(struct vnode *root, struct pathdata *pdata,
+				const char *component);
+	int (*readdir)(struct vnode *v);
+	int (*filldir)(void *buf, struct vnode *v, int count);
+	vnode_op_t link;
+	vnode_op_t unlink;
+	int (*mkdir)(struct vnode *parent, const char *name);
+	struct vnode *(*mknod)(struct vnode *parent, const char *name,
+			       unsigned int mode);
+	vnode_op_t rmdir;
+	vnode_op_t rename;
+	vnode_op_t getattr;
+	vnode_op_t setattr;
+};
+
+struct superblock_ops {
+	int (*write_sb)(struct superblock *sb);
+
+	/*
+	 * Given a vnum, reads the disk-inode and copies its data
+	 * into the vnode's generic fields
+	 */
+	int (*read_vnode)(struct superblock *sb, struct vnode *v);
+
+	/* Writes vnode's generic fields into the disk-inode structure */
+	int (*write_vnode)(struct superblock *sb, struct vnode *v);
+
+	/* Allocates a disk-inode along with a vnode, and associates the two */
+	struct vnode *(*alloc_vnode)(struct superblock *sb);
+
+	/* Frees the vnode and the corresponding on-disk inode */
+	int (*free_vnode)(struct superblock *sb, struct vnode *v);
+};
+
+#define	VFS_DNAME_MAX			256
+struct dentry {
+	int refcnt;
+	char name[VFS_DNAME_MAX];
+	struct dentry *parent;		/* Parent dentry */
+	struct link child;		/* List of dentries with same parent */
+	struct link children;	/* List of children dentries */
+	struct link vref;		/* For vnode's dirent reference list */
+	struct link cache_list;	/* Dentry cache reference */
+	struct vnode *vnode;		/* The vnode associated with dentry */
+	struct dentry_ops ops;
+};
+
+/*
+ * Buffer that maintains directory content for a directory vnode. This is the
+ * only vnode content that fs0 maintains. All other file data is in mm0 page
+ * cache.
+ */
+struct dirbuf {
+	unsigned long npages;
+	int dirty;
+	u8 *buffer;
+};
+
+/* Posix-style dirent format used by userspace. Returned by sys_readdir() */
+#define DIRENT_NAME_MAX			32
+struct dirent {
+	u32 inum;			/* Inode number */
+	u32 offset;			/* Dentry offset in its buffer */
+	u16 rlength;			/* Record length */
+	u8  name[DIRENT_NAME_MAX];	/* Name string */
+};
+
+struct vnode {
+	unsigned long vnum;		/* Filesystem-wide unique vnode id */
+	int refcnt;			/* Reference counter */
+	int links;			/* Number of hard links */
+	struct superblock *sb;		/* Reference to superblock */
+	struct vnode_ops ops;		/* Operations on this vnode */
+	struct file_ops fops;		/* File-related operations on this vnode */
+	struct link dentries;	/* Dirents that refer to this vnode */
+	struct link cache_list;	/* For adding the vnode to vnode cache */
+	struct dirbuf dirbuf;		/* Only directory buffers are kept */
+	u32 mode;			/* Permissions and vnode type */
+	u32 owner;			/* Owner */
+	u64 atime;			/* Last access time */
+	u64 mtime;			/* Last content modification */
+	u64 ctime;			/* Last vnode modification */
+	u64 size;			/* Size of contents */
+	void *inode;			/* Ptr to fs-specific inode */
+};
+
+/* FS0 vfs specific macros */
+
+/* Check if directory */
+#define vfs_isdir(v)		S_ISDIR((v)->mode)
+
+/* Set vnode type */
+#define vfs_set_type(v, type)	{v->mode &= ~S_IFMT; v->mode |= S_IFMT & type; }
+
+struct fstype_ops {
+	struct superblock *(*get_superblock)(void *buf);
+};
+
+#define VFS_FSNAME_MAX			256
+struct file_system_type {
+	char name[VFS_FSNAME_MAX];
+	unsigned long magic;
+	struct fstype_ops ops;
+	struct link list;	/* Member of list of all fs types */
+	struct link sblist; /* List of superblocks with this type */
+};
+struct superblock *get_superblock(void *buf);
+
+struct superblock {
+	u64 fssize;
+	unsigned int blocksize;
+	struct link list;
+	struct file_system_type *fs;
+	struct superblock_ops *ops;
+	struct vnode *root;
+	void *fs_super;
+};
+
+void vfs_mount_fs(struct superblock *sb);
+
+extern struct dentry_ops generic_dentry_operations;
+
+#endif /* __FS_H__ */

conts/posix/mm0/include/lib/bit.h

@@ -9,8 +9,8 @@ int find_and_set_first_free_contig_bits(u32 *word,  unsigned int limit,
 					int nbits);
 int check_and_clear_bit(u32 *word, int bit);
 int check_and_clear_contig_bits(u32 *word, int first, int nbits);
-int check_and_set_bit(u32 *word, int bit);
 
+int check_and_set_bit(u32 *word, int bit);
 
 /* Set */
 static inline void setbit(unsigned int *w, unsigned int flags)

conts/posix/mm0/include/lib/idpool.h

@@ -3,6 +3,8 @@
 
 #include <lib/bit.h>
 #include <lib/spinlock.h>
+#include <string.h>
+#include INC_GLUE(memory.h)
 
 struct id_pool {
 	int nwords;
@@ -10,6 +12,14 @@ struct id_pool {
 	u32 bitmap[];
 };
 
+/* Copy one id pool to another by calculating its size */
+static inline void id_pool_copy(struct id_pool *to, struct id_pool *from, int totalbits)
+{
+	int nwords = BITWISE_GETWORD(totalbits);
+
+	memcpy(to, from, nwords * SZ_WORD + sizeof(struct id_pool));
+}
+
 struct id_pool *id_pool_new_init(int mapsize);
 int id_new(struct id_pool *pool);
 int id_del(struct id_pool *pool, int id);

conts/posix/mm0/include/lib/malloc.h

@@ -3,7 +3,6 @@
 
 #include <stddef.h>
 #include <string.h>
-#include <stdio.h>
 
 void *kmalloc(size_t size);
 void kfree(void *blk);

conts/posix/mm0/include/lib/pathstr.h

@@ -0,0 +1,9 @@
+#ifndef __LIB_PATHSTR_H__
+#define __LIB_PATHSTR_H__
+
+char *strreverse(char *str);
+char *splitpath_end(char **path, char sep);
+char *splitpath(char **str, char sep);
+
+
+#endif /* __LIB_PATHSTR_H__ */

conts/posix/mm0/include/lib/spinlock.h

@@ -4,13 +4,11 @@
 #ifndef __MM0_SPINLOCK_H__
 #define __MM0_SPINLOCK_H__
 
-
-
 struct spinlock {
 	int lock;
 };
 
-static inline void spin_lock_init(struct spinlock *s) { } 
+static inline void spin_lock_init(struct spinlock *s) { }
 static inline void spin_lock(struct spinlock *s) { }
 static inline void spin_unlock(struct spinlock *s) { }
 

conts/posix/mm0/include/lib/vaddr.h

@@ -0,0 +1,17 @@
+/*
+ * Virtual address allocation pool (for shm)
+ *
+ * Copyright (C) 2007 Bahadir Balban
+ */
+#ifndef __VADDR_H__
+#define __VADDR_H__
+
+#include <lib/idpool.h>
+
+void vaddr_pool_init(struct id_pool *pool, unsigned long start,
+		     unsigned long end);
+void *vaddr_new(struct id_pool *pool, int npages);
+int vaddr_del(struct id_pool *, void *vaddr, int npages);
+
+#endif /* __VADDR_H__ */
+

conts/posix/mm0/include/linker.lds.in

@@ -34,6 +34,9 @@ SECTIONS
 	.data : AT (ADDR(.data) - pager_offset)
 	{
 		*(.data)
+		_start_bdev = .;
+		*(.data.memfs)
+		_end_bdev = .;
 	}
 	.bss : AT (ADDR(.bss) - pager_offset) { *(.bss) }
 	. = ALIGN(4K);

conts/posix/mm0/include/memfs/file.h

@@ -0,0 +1,7 @@
+#ifndef __MEMFS_FILE_H__
+#define __MEMFS_FILE_H__
+
+extern struct file_ops memfs_file_operations;
+extern struct dentry_ops memfs_dentry_operations;
+
+#endif /* __MEMFS_FILE_H__ */

conts/posix/mm0/include/memfs/memfs.h

@@ -0,0 +1,97 @@
+/*
+ * The disk layout of our simple unix-like filesystem.
+ *
+ * Copyright (C) 2007, 2008 Bahadir Balban
+ */
+
+#ifndef __MEMFS_LAYOUT_H__
+#define __MEMFS_LAYOUT_H__
+
+#include <l4lib/types.h>
+#include <l4/lib/list.h>
+#include <l4/macros.h>
+#include <l4/config.h>
+#include INC_GLUE(memory.h)
+#include <memcache/memcache.h>
+#include <lib/idpool.h>
+
+/*
+ *
+ * Filesystem layout:
+ *
+ * |---------------|
+ * |  Superblock   |
+ * |---------------|
+ *
+ * Superblock layout:
+ *
+ * |---------------|
+ * | inode cache   |
+ * |---------------|
+ * | dentry cache  |
+ * |---------------|
+ * | block cache   |
+ * |---------------|
+ *
+ */
+
+/*
+ * These fixed filesystem limits make it much easier to implement
+ * filesystem space allocation.
+ */
+#define MEMFS_TOTAL_SIZE		SZ_4MB
+#define MEMFS_TOTAL_INODES		128
+#define MEMFS_TOTAL_BLOCKS		2000
+#define MEMFS_FMAX_BLOCKS		40
+#define MEMFS_BLOCK_SIZE		PAGE_SIZE
+#define MEMFS_MAGIC			0xB
+#define MEMFS_NAME			"memfs"
+#define MEMFS_NAME_SIZE			8
+
+struct memfs_inode {
+	u32 inum;	/* Inode number */
+	u32 mode;	/* File permissions */
+	u32 owner;	/* File owner */
+	u64 atime;	/* Last access time */
+	u64 mtime;	/* Last content modification */
+	u64 ctime;	/* Last inode modification */
+	u64 size;	/* Size of contents */
+	void *block[MEMFS_FMAX_BLOCKS]; /* Number of blocks */
+};
+
+struct memfs_superblock {
+	u32 magic;		/* Filesystem magic number */
+	char name[8];
+	u32 blocksize;		/* Filesystem block size */
+	u64 fmaxblocks;		/* Maximum number of blocks per file */
+	u64 fssize;		/* Total size of filesystem */
+	unsigned long root_vnum;	/* The root vnum of this superblock */
+	struct link inode_cache_list;	/* Chain of alloc caches */
+	struct link block_cache_list;	/* Chain of alloc caches */
+	struct id_pool *ipool;			/* Index pool for inodes */
+	struct id_pool *bpool;			/* Index pool for blocks */
+	struct memfs_inode *inode[MEMFS_TOTAL_INODES];	/* Table of inodes */
+	void *block[MEMFS_TOTAL_BLOCKS]; 	/* Table of fs blocks */
+} __attribute__ ((__packed__));
+
+#define MEMFS_DNAME_MAX			32
+struct memfs_dentry {
+	u32 inum;			/* Inode number */
+	u32 offset;			/* Dentry offset in its buffer */
+	u16 rlength;			/* Record length */
+	u8  name[MEMFS_DNAME_MAX];	/* Name string */
+};
+
+extern struct vnode_ops memfs_vnode_operations;
+extern struct superblock_ops memfs_superblock_operations;
+extern struct file_ops memfs_file_operations;
+
+int memfs_format_filesystem(void *buffer);
+struct memfs_inode *memfs_create_inode(struct memfs_superblock *sb);
+void memfs_register_fstype(struct link *);
+struct superblock *memfs_get_superblock(void *block);
+int memfs_generate_superblock(void *block);
+
+void *memfs_alloc_block(struct memfs_superblock *sb);
+int memfs_free_block(struct memfs_superblock *sb, void *block);
+#endif /* __MEMFS_LAYOUT_H__ */

conts/posix/mm0/include/memfs/vnode.h

@@ -0,0 +1,7 @@
+#ifndef __MEMFS_VNODE_H__
+#define __MEMFS_VNODE_H__
+
+#include <fs.h>
+
+
+#endif /* __MEMFS_VNODE_H__ */

conts/posix/mm0/include/path.h

@@ -0,0 +1,41 @@
+/*
+ * Copyright (C) 2008 Bahadir Balban
+ *
+ * Path lookup related information
+ */
+#ifndef __PATH_H__
+#define __PATH_H__
+
+#include <l4/lib/list.h>
+#include <task.h>
+
+/*
+ * FIXME:
+ * These ought to be strings and split/comparison functions should
+ * always use strings because formats like UTF-8 wouldn't work.
+ */
+#define VFS_CHAR_SEP		'/'
+#define VFS_STR_ROOTDIR		"/"
+#define VFS_STR_PARDIR		".."
+#define VFS_STR_CURDIR		"."
+#define VFS_STR_XATDIR		"...."
+
+struct pathdata {
+	struct link list;
+	struct vnode *vstart;
+};
+
+struct pathcomp {
+	struct link list;
+	const char *str;
+};
+
+struct pathdata *pathdata_parse(const char *pathname, char *pathbuf,
+				struct tcb *task);
+void pathdata_destroy(struct pathdata *p);
+
+/* Destructive, i.e. unlinks those components from list */
+const char *pathdata_next_component(struct pathdata *pdata);
+const char *pathdata_last_component(struct pathdata *pdata);
+
+#endif /* __PATH_H__ */

conts/posix/mm0/include/stat.h

@@ -0,0 +1,82 @@
+#ifndef __FS0_STAT_H__
+#define __FS0_STAT_H__
+
+/* Posix definitions for file mode flags (covers type and access permissions) */
+#define S_IFMT  00170000
+#define S_IFSOCK 0140000
+#define S_IFLNK	 0120000
+#define S_IFREG  0100000
+#define S_IFBLK  0060000
+#define S_IFDIR  0040000
+#define S_IFCHR  0020000
+#define S_IFIFO  0010000
+#define S_ISUID  0004000
+#define S_ISGID  0002000
+#define S_ISVTX  0001000
+
+#define S_ISLNK(m)	(((m) & S_IFMT) == S_IFLNK)
+#define S_ISREG(m)	(((m) & S_IFMT) == S_IFREG)
+#define S_ISDIR(m)	(((m) & S_IFMT) == S_IFDIR)
+#define S_ISCHR(m)	(((m) & S_IFMT) == S_IFCHR)
+#define S_ISBLK(m)	(((m) & S_IFMT) == S_IFBLK)
+#define S_ISFIFO(m)	(((m) & S_IFMT) == S_IFIFO)
+#define S_ISSOCK(m)	(((m) & S_IFMT) == S_IFSOCK)
+
+#define S_IRWXU 00700
+#define S_IRUSR 00400
+#define S_IWUSR 00200
+#define S_IXUSR 00100
+
+#define S_IRWXG 00070
+#define S_IRGRP 00040
+#define S_IWGRP 00020
+#define S_IXGRP 00010
+
+#define S_IRWXO 00007
+#define S_IROTH 00004
+#define S_IWOTH 00002
+#define S_IXOTH 00001
+
+#define S_IRWXUGO	(S_IRWXU|S_IRWXG|S_IRWXO)
+#define S_IALLUGO	(S_ISUID|S_ISGID|S_ISVTX|S_IRWXUGO)
+#define S_IRUGO		(S_IRUSR|S_IRGRP|S_IROTH)
+#define S_IWUGO		(S_IWUSR|S_IWGRP|S_IWOTH)
+#define S_IXUGO		(S_IXUSR|S_IXGRP|S_IXOTH)
+
+#define O_ACCMODE	00000003
+#define O_RDONLY	00000000
+#define O_WRONLY	00000001
+#define O_RDWR		00000002
+#define O_CREAT		00000100
+#define O_EXCL		00000200
+#define O_NOCTTY	00000400
+#define O_TRUNC		00001000
+#define O_APPEND	00002000
+#define O_NONBLOCK	00004000
+#define O_SYNC		00010000
+#define FASYNC		00020000
+#define O_DIRECT	00040000
+#define O_LARGEFILE	00100000
+#define O_DIRECTORY	00200000
+#define O_NOFOLLOW	00400000
+#define O_NOATIME	01000000
+#define O_NDELAY	O_NONBLOCK
+
+/*
+ * Internal codezero-specific stat structure.
+ * This is converted to posix stat in userspace
+ */
+struct kstat {
+	u64 vnum;
+	u32 mode;
+	int links;
+	u16 uid;
+	u16 gid;
+	u64 size;
+	int blksize;
+	u64 atime;
+	u64 mtime;
+	u64 ctime;
+};
+
+#endif /* __FS0_STAT_H__ */

conts/posix/mm0/include/syscalls.h

@@ -36,5 +36,30 @@ int sys_fork(struct tcb *parent);
 int sys_clone(struct tcb *parent, void *child_stack, unsigned int clone_flags);
 void sys_exit(struct tcb *task, int status);
 
+/* Posix calls */
+int sys_open(struct tcb *sender, const char *pathname, int flags, u32 mode);
+int sys_readdir(struct tcb *sender, int fd, void *buf, int count);
+int sys_mkdir(struct tcb *sender, const char *pathname, unsigned int mode);
+int sys_chdir(struct tcb *sender, const char *pathname);
+
+/* Calls from pager that completes a posix call */
+int pager_open_bypath(struct tcb *pager, char *pathname);
+int pager_sys_open(struct tcb *sender, l4id_t opener, int fd);
+int pager_sys_read(struct tcb *sender, unsigned long vnum, unsigned long f_offset,
+		   unsigned long npages, void *pagebuf);
+
+int pager_sys_write(struct tcb *sender, unsigned long vnum, unsigned long f_offset,
+		   unsigned long npages, void *pagebuf);
+
+int pager_sys_close(struct tcb *sender, l4id_t closer, int fd);
+int pager_update_stats(struct tcb *sender, unsigned long vnum,
+		       unsigned long newsize);
+
+int pager_notify_fork(struct tcb *sender, l4id_t parentid,
+		      l4id_t childid, unsigned long utcb_address,
+		      unsigned int flags);
+
+int pager_notify_exit(struct tcb *sender, l4id_t tid);
+
 #endif /* __MM0_SYSARGS_H__ */
 

conts/posix/mm0/include/task.h

@@ -47,6 +47,7 @@ struct file_descriptor {
 
 struct task_fd_head {
 	struct file_descriptor fd[TASK_FILES_MAX];
+	struct id_pool *fdpool;
 	int tcb_refs;
 };
 
@@ -55,6 +56,18 @@ struct task_vma_head {
 	int tcb_refs;
 };
 
+#define TCB_NO_SHARING				0
+#define	TCB_SHARED_VM				(1 << 0)
+#define	TCB_SHARED_FILES			(1 << 1)
+#define TCB_SHARED_FS				(1 << 2)
+#define TASK_FILES_MAX			32
+
+struct task_fs_data {
+	struct vnode *curdir;
+	struct vnode *rootdir;
+	int tcb_refs;
+};
+
 struct utcb_desc {
 	struct link list;
 	unsigned long utcb_base;
@@ -114,9 +127,6 @@ struct tcb {
 	unsigned long map_start;
 	unsigned long map_end;
 
-	/* Default ipc-shared-page information */
-	void *shared_page;
-
 	/* Chain of utcb descriptors */
 	struct utcb_head *utcb_head;
 
@@ -128,6 +138,8 @@ struct tcb {
 
 	/* File descriptors for this task */
 	struct task_fd_head *files;
+	struct task_fs_data *fs_data;
+
 };
 
 struct tcb_head {

conts/posix/mm0/include/vfs.h

@@ -0,0 +1,90 @@
+#ifndef __VFS_H__
+#define __VFS_H__
+
+#include <fs.h>
+#include <lib/malloc.h>
+#include <l4/lib/list.h>
+#include <memfs/memfs.h>
+#include <l4/macros.h>
+#include <stdio.h>
+#include <task.h>
+#include <path.h>
+
+extern struct link vnode_cache;
+extern struct link dentry_cache;
+
+/*
+ * This is a temporary origacement for page cache support provided by mm0.
+ * Normally mm0 tracks all vnode pages, but this is used to track pages in
+ * directory vnodes, which are normally never mapped by tasks.
+ */
+static inline void *vfs_alloc_dirpage(struct vnode *v)
+{
+	/*
+	 * Urgh, we allocate from the block cache of memfs to store generic vfs directory
+	 * pages. This is currently the quickest we can allocate page-aligned memory.
+	 */
+	return memfs_alloc_block(v->sb->fs_super);
+}
+
+static inline void vfs_free_dirpage(struct vnode *v, void *block)
+{
+	memfs_free_block(v->sb->fs_super, block);
+}
+
+static inline struct dentry *vfs_alloc_dentry(void)
+{
+	struct dentry *d = kzalloc(sizeof(struct dentry));
+
+	link_init(&d->child);
+	link_init(&d->children);
+	link_init(&d->vref);
+	link_init(&d->cache_list);
+
+	return d;
+}
+
+static inline void vfs_free_dentry(struct dentry *d)
+{
+	return kfree(d);
+}
+
+static inline struct vnode *vfs_alloc_vnode(void)
+{
+	struct vnode *v = kzalloc(sizeof(struct vnode));
+
+	link_init(&v->dentries);
+	link_init(&v->cache_list);
+
+	return v;
+}
+
+static inline void vfs_free_vnode(struct vnode *v)
+{
+	BUG(); /* Are the dentries freed ??? */
+	list_remove(&v->cache_list);
+	kfree(v);
+}
+
+static inline struct superblock *vfs_alloc_superblock(void)
+{
+	struct superblock *sb = kmalloc(sizeof(struct superblock));
+	link_init(&sb->list);
+
+	return sb;
+}
+
+struct vfs_mountpoint {
+	struct superblock *sb;	/* The superblock of mounted filesystem */
+	struct vnode *pivot;	/* The dentry upon which we mount */
+};
+
+extern struct vfs_mountpoint vfs_root;
+
+int vfs_mount_root(struct superblock *sb);
+struct vnode *generic_vnode_lookup(struct vnode *thisnode, struct pathdata *p,
+				   const char *component);
+struct vnode *vfs_lookup_bypath(struct pathdata *p);
+struct vnode *vfs_lookup_byvnum(struct superblock *sb, unsigned long vnum);
+
+#endif /* __VFS_H__ */