minix/servers/vfs/mount.c
David van Moolenbroek 28f2a169da VFS: bugfixes for handling block-special files:
- on driver restarts, reopen devices on a per-file basis, not per-mount
- do not assume that there is just one vnode per block-special device
- update block-special files in the uncommon mounting success paths, too
- upon mount, sync but also invalidate affected buffers on the root FS
- upon unmount, check whether a vnode is in use before updating it
2011-03-25 10:56:43 +00:00

528 lines
15 KiB
C

/* This file performs the MOUNT and UMOUNT system calls.
*
* The entry points into this file are
* do_fslogin: perform the FSLOGIN system call
* do_mount: perform the MOUNT system call
* do_umount: perform the UMOUNT system call
* unmount: unmount a file system
*/
#include "fs.h"
#include <fcntl.h>
#include <string.h>
#include <minix/callnr.h>
#include <minix/com.h>
#include <minix/keymap.h>
#include <minix/const.h>
#include <minix/endpoint.h>
#include <minix/syslib.h>
#include <minix/bitmap.h>
#include <minix/ds.h>
#include <unistd.h>
#include <sys/stat.h>
#include <sys/mount.h>
#include <dirent.h>
#include "file.h"
#include "fproc.h"
#include "param.h"
#include <minix/vfsif.h>
#include "vnode.h"
#include "vmnt.h"
/* Allow the root to be replaced before the first 'real' mount. */
PRIVATE int allow_newroot = 1;
/* Bitmap of in-use "none" pseudo devices. */
PRIVATE bitchunk_t nonedev[BITMAP_CHUNKS(NR_NONEDEVS)] = { 0 };
#define alloc_nonedev(dev) SET_BIT(nonedev, minor(dev) - 1)
#define free_nonedev(dev) UNSET_BIT(nonedev, minor(dev) - 1)
FORWARD _PROTOTYPE( dev_t name_to_dev, (int allow_mountpt) );
FORWARD _PROTOTYPE( int mount_fs, (endpoint_t fs_e) );
FORWARD _PROTOTYPE( int is_nonedev, (dev_t dev) );
FORWARD _PROTOTYPE( dev_t find_free_nonedev, (void) );
/*===========================================================================*
* update_bspec *
*===========================================================================*/
PRIVATE void update_bspec(dev_t dev, endpoint_t fs_e)
{
/* Update all block special files for a certain device, to use a new FS endpt
* to route raw block I/O requests through.
*/
struct vnode *vp;
for (vp = &vnode[0]; vp < &vnode[NR_VNODES]; ++vp)
if (vp->v_ref_count > 0 && S_ISBLK(vp->v_mode) && vp->v_sdev == dev)
vp->v_bfs_e = fs_e;
}
/*===========================================================================*
* do_fslogin *
*===========================================================================*/
PUBLIC int do_fslogin()
{
/* deprecated */
return(SUSPEND);
}
/*===========================================================================*
* do_mount *
*===========================================================================*/
PUBLIC int do_mount()
{
endpoint_t fs_e;
int r, proc_nr;
/* Only the super-user may do MOUNT. */
if (!super_user) return(EPERM);
/* FS process' endpoint number */
if (m_in.mount_flags & MS_LABEL16) {
/* Get the label from the caller, and ask DS for the endpoint. */
r = sys_datacopy(who_e, (vir_bytes) m_in.fs_label, SELF,
(vir_bytes) mount_label, (phys_bytes) sizeof(mount_label));
if (r != OK) return(r);
mount_label[sizeof(mount_label)-1] = 0;
r = ds_retrieve_label_endpt(mount_label, &fs_e);
if (r != OK) return(r);
if (isokendpt(fs_e, &proc_nr) != OK) return(EINVAL);
} else {
/* Legacy support: get the endpoint from the request itself. */
fs_e = (endpoint_t) m_in.fs_label;
mount_label[0] = 0;
}
/* Sanity check on process number. */
if(fs_e <= 0) {
printf("VFS: warning: got process number %d for mount call.\n", fs_e);
return EINVAL;
}
/* Do the actual job */
return mount_fs(fs_e);
}
/*===========================================================================*
* mount *
*===========================================================================*/
PRIVATE int mount_fs(endpoint_t fs_e)
{
/* Perform the mount(name, mfile, mount_flags) system call. */
int rdir, mdir; /* TRUE iff {root|mount} file is dir */
int i, r, found, rdonly, nodev, isroot, replace_root;
struct fproc *tfp;
struct dmap *dp;
dev_t dev;
struct vnode *root_node, *vp = NULL, *bspec;
struct vmnt *vmp;
char *label;
struct node_details res;
/* Only the super-user may do MOUNT. */
if (!super_user) return(EPERM);
/* Clear endpoint field */
/* Should the file system be mounted read-only? */
rdonly = (m_in.mount_flags & MS_RDONLY);
/* A null string for block special device means don't use a device at all. */
nodev = (m_in.name1_length == 0);
if (!nodev) {
/* If 'name' is not for a block special file, return error. */
if (fetch_name(m_in.name1, m_in.name1_length, M1) != OK)
return(err_code);
if ((dev = name_to_dev(FALSE /*allow_mountpt*/)) == NO_DEV)
return(err_code);
} else {
/* Find a free pseudo-device as substitute for an actual device. */
if ((dev = find_free_nonedev()) == NO_DEV)
return(err_code);
}
/* Check whether there is a block special file open which uses the
* same device (partition) */
for (bspec = &vnode[0]; bspec < &vnode[NR_VNODES]; ++bspec) {
if (bspec->v_ref_count > 0 && bspec->v_sdev == dev) {
/* Found, flush and invalidate any blocks for this device. */
req_flush(bspec->v_fs_e, dev);
break;
}
}
/* Scan vmnt table to see if dev already mounted. If not, find a free slot.*/
found = FALSE;
vmp = NULL;
for (i = 0; i < NR_MNTS; ++i) {
if (vmnt[i].m_dev == dev) {
vmp = &vmnt[i];
found = TRUE;
break;
} else if (!vmp && vmnt[i].m_dev == NO_DEV) {
vmp = &vmnt[i];
}
}
/* Partition was/is already mounted */
if (found) {
/* It is possible that we have an old root lying around that
* needs to be remounted. This could for example be a boot
* ramdisk that has already been replaced by the real root.
*/
if(vmp->m_mounted_on || root_dev == vmp->m_dev) {
return(EBUSY); /* not a root or still mounted */
}
/* Now get the inode of the file to be mounted on. */
if (fetch_name(m_in.name2, m_in.name2_length, M1)!=OK) return(err_code);
if ((vp = eat_path(PATH_NOFLAGS, fp)) == NULL) return(err_code);
if (vp->v_ref_count != 1) {
put_vnode(vp);
return(EBUSY);
}
/* Tell FS on which vnode it is mounted (glue into mount tree) */
if ((r = req_mountpoint(vp->v_fs_e, vp->v_inode_nr)) == OK) {
root_node = vmp->m_root_node;
/* File types of 'vp' and 'root_node' may not conflict. */
mdir = ((vp->v_mode & I_TYPE) == I_DIRECTORY);/* TRUE iff dir*/
rdir = ((root_node->v_mode & I_TYPE) == I_DIRECTORY);
if(!mdir && rdir) r = EISDIR;
}
if (r != OK) {
put_vnode(vp);
return(r);
}
/* Nothing else can go wrong. Perform the mount. */
vmp->m_mounted_on = vp;
vmp->m_flags = rdonly;
strcpy(vmp->m_label, mount_label);
allow_newroot = 0; /* The root is now fixed */
if (nodev) alloc_nonedev(dev); /* Make the allocation final */
update_bspec(dev, fs_e); /* Update open block-special files */
return(OK);
} else if (vmp == NULL) {
/* No free slot available, bail out */
return(ENOMEM);
}
/* Fetch the name of the mountpoint */
if (fetch_name(m_in.name2, m_in.name2_length, M1) != OK) return(err_code);
isroot = (strcmp(user_fullpath, "/") == 0);
replace_root = (isroot && allow_newroot);
if(!replace_root) {
/* Get vnode of mountpoint */
if ((vp = eat_path(PATH_NOFLAGS, fp)) == NULL) return(err_code);
if (vp->v_ref_count != 1) {
put_vnode(vp);
return(EBUSY);
}
/* Tell FS on which vnode it is mounted (glue into mount tree) */
if ((r = req_mountpoint(vp->v_fs_e, vp->v_inode_nr)) != OK) {
put_vnode(vp);
return r;
}
}
/* We'll need a vnode for the root inode, check whether there is one */
if ((root_node = get_free_vnode()) == NULL) {
if (vp != NULL) put_vnode(vp);
return(ENFILE);
}
label = "";
if (!nodev) {
/* Get driver process' endpoint */
dp = &dmap[(dev >> MAJOR) & BYTE];
if (dp->dmap_driver == NONE) {
printf("VFS: no driver for dev %x\n", dev);
if (vp != NULL) put_vnode(vp);
return(EINVAL);
}
label = dp->dmap_label;
if (strlen(label) == 0)
panic("VFS mount_fs: no label for major: 0x%x", dev >> MAJOR);
}
/* Tell FS which device to mount */
if ((r = req_readsuper(fs_e, label, dev, rdonly, isroot, &res)) != OK) {
if (vp != NULL) put_vnode(vp);
return(r);
}
/* Fill in root node's fields */
root_node->v_fs_e = res.fs_e;
root_node->v_inode_nr = res.inode_nr;
root_node->v_mode = res.fmode;
root_node->v_uid = res.uid;
root_node->v_gid = res.gid;
root_node->v_size = res.fsize;
root_node->v_sdev = NO_DEV;
root_node->v_fs_count = 1;
root_node->v_ref_count = 1;
/* Fill in max file size and blocksize for the vmnt */
vmp->m_fs_e = res.fs_e;
vmp->m_dev = dev;
vmp->m_flags = rdonly;
/* Root node is indeed on the partition */
root_node->v_vmnt = vmp;
root_node->v_dev = vmp->m_dev;
if(replace_root) {
/* Superblock and root node already read.
* Nothing else can go wrong. Perform the mount. */
vmp->m_root_node = root_node;
vmp->m_mounted_on = NULL;
strcpy(vmp->m_label, mount_label);
if (nodev) alloc_nonedev(dev);
update_bspec(dev, fs_e);
root_dev = dev;
ROOT_FS_E = fs_e;
/* Replace all root and working directories */
for (i= 0, tfp= fproc; i<NR_PROCS; i++, tfp++) {
if (tfp->fp_pid == PID_FREE)
continue;
#define MAKEROOT(what) { \
put_vnode(what); \
dup_vnode(root_node); \
what = root_node; \
}
if(tfp->fp_rd) MAKEROOT(tfp->fp_rd);
if(tfp->fp_wd) MAKEROOT(tfp->fp_wd);
}
return(OK);
}
/* File types may not conflict. */
mdir = ((vp->v_mode & I_TYPE) == I_DIRECTORY); /*TRUE iff dir*/
rdir = ((root_node->v_mode & I_TYPE) == I_DIRECTORY);
if (!mdir && rdir) r = EISDIR;
/* If error, return the super block and both inodes; release the vmnt. */
if (r != OK) {
put_vnode(vp);
put_vnode(root_node);
vmp->m_dev = NO_DEV;
return(r);
}
/* Nothing else can go wrong. Perform the mount. */
vmp->m_mounted_on = vp;
vmp->m_root_node = root_node;
strcpy(vmp->m_label, mount_label);
/* The root is now fixed */
allow_newroot = 0;
/* Allocate the pseudo device that was found, if not using a real device. */
if (nodev) alloc_nonedev(dev);
/* The new FS will handle block I/O requests for its device now. */
update_bspec(dev, fs_e);
return(OK);
}
/*===========================================================================*
* do_umount *
*===========================================================================*/
PUBLIC int do_umount(void)
{
/* Perform the umount(name) system call. */
char label[LABEL_MAX];
dev_t dev;
int r;
/* Only the super-user may do umount. */
if (!super_user) return(EPERM);
/* If 'name' is not for a block special file or mountpoint, return error. */
if(fetch_name(m_in.name, m_in.name_length, M3) != OK) return(err_code);
if((dev = name_to_dev(TRUE /*allow_mountpt*/)) == NO_DEV) return(err_code);
if((r = unmount(dev, label)) != OK) return(r);
/* Return the label of the mounted file system, so that the caller
* can shut down the corresponding server process.
*/
if (strlen(label) >= M3_LONG_STRING) /* should never evaluate to true */
label[M3_LONG_STRING-1] = 0;
strcpy(m_out.umount_label, label);
return(OK);
}
/*===========================================================================*
* unmount *
*===========================================================================*/
PUBLIC int unmount(
dev_t dev, /* block-special device */
char *label /* buffer to retrieve label, or NULL */
)
{
struct vnode *vp;
struct vmnt *vmp_i = NULL, *vmp = NULL;
struct dmap *dp;
int count, r;
/* Find vmnt that is to be unmounted */
for(vmp_i = &vmnt[0]; vmp_i < &vmnt[NR_MNTS]; ++vmp_i) {
if (vmp_i->m_dev == dev) {
if(vmp) panic("device mounted more than once: %d", dev);
vmp = vmp_i;
}
}
/* Did we find the vmnt (i.e., was dev a mounted device)? */
if(!vmp) return(EINVAL);
/* See if the mounted device is busy. Only 1 vnode using it should be
* open -- the root vnode -- and that inode only 1 time. */
count = 0;
for(vp = &vnode[0]; vp < &vnode[NR_VNODES]; vp++)
if(vp->v_ref_count > 0 && vp->v_dev == dev) count += vp->v_ref_count;
if(count > 1) return(EBUSY); /* can't umount a busy file system */
/* Tell FS to drop all inode references for root inode except 1. */
vnode_clean_refs(vmp->m_root_node);
if (vmp->m_mounted_on) {
put_vnode(vmp->m_mounted_on);
vmp->m_mounted_on = NULL;
}
/* Tell FS to unmount */
if(vmp->m_fs_e <= 0 || vmp->m_fs_e == NONE)
panic("unmount: strange fs endpoint: %d", vmp->m_fs_e);
if ((r = req_unmount(vmp->m_fs_e)) != OK) /* Not recoverable. */
printf("VFS: ignoring failed umount attempt (%d) fs pid: %d\n", r,
_ENDPOINT_P(vmp->m_fs_e));
if (is_nonedev(vmp->m_dev))
free_nonedev(vmp->m_dev);
if (label != NULL)
strcpy(label, vmp->m_label);
vmp->m_root_node->v_ref_count = 0;
vmp->m_root_node->v_fs_count = 0;
vmp->m_root_node->v_sdev = NO_DEV;
vmp->m_root_node = NULL;
vmp->m_dev = NO_DEV;
vmp->m_fs_e = NONE;
/* Is there a block special file that was handled by that partition? */
for (vp = &vnode[0]; vp < &vnode[NR_VNODES]; vp++) {
if(vp->v_ref_count > 0 && S_ISBLK(vp->v_mode) &&
vp->v_bfs_e == vmp->m_fs_e) {
/* Get the driver endpoint of the block spec device */
dp = &dmap[(dev >> MAJOR) & BYTE];
if (dp->dmap_driver == NONE) {
printf("VFS: driver not found for device %d\n", dev);
continue;
}
printf("VFS: umount moving block spec %d to root FS\n", dev);
vp->v_bfs_e = ROOT_FS_E;
/* Send the (potentially new) driver endpoint */
r = req_newdriver(vp->v_bfs_e, vp->v_sdev, dp->dmap_driver);
if (r != OK)
printf("VFS: error sending driver endpoint for"
" moved block spec\n");
}
}
return(OK);
}
/*===========================================================================*
* name_to_dev *
*===========================================================================*/
PRIVATE dev_t name_to_dev(int allow_mountpt)
{
/* Convert the block special file in 'user_fullpath' to a device number.
* If the given path is not a block special file, but 'allow_mountpt' is set
* and the path is the root node of a mounted file system, return that device
* number. In all other cases, return NO_DEV and an error code in 'err_code'.
*/
dev_t dev;
struct vnode *vp;
/* Request lookup */
if ((vp = eat_path(PATH_NOFLAGS, fp)) == NULL) {
return(NO_DEV);
}
if ((vp->v_mode & I_TYPE) == I_BLOCK_SPECIAL) {
dev = vp->v_sdev;
} else if (allow_mountpt && vp->v_vmnt->m_root_node == vp) {
dev = vp->v_dev;
} else {
err_code = ENOTBLK;
dev = NO_DEV;
}
put_vnode(vp);
return(dev);
}
/*===========================================================================*
* is_nonedev *
*===========================================================================*/
PRIVATE int is_nonedev(dev_t dev)
{
/* Return whether the given device is a "none" pseudo device.
*/
return (major(dev) == NONE_MAJOR &&
minor(dev) > 0 && minor(dev) <= NR_NONEDEVS);
}
/*===========================================================================*
* find_free_nonedev *
*===========================================================================*/
PRIVATE dev_t find_free_nonedev(void)
{
/* Find a free "none" pseudo device. Do not allocate it yet.
*/
int i;
for (i = 0; i < NR_NONEDEVS; i++)
if (!GET_BIT(nonedev, i))
return makedev(NONE_MAJOR, i + 1);
err_code = EMFILE;
return NO_DEV;
}