2006-10-25 15:40:36 +02:00
|
|
|
/* This file performs the MOUNT and UMOUNT system calls.
|
|
|
|
*
|
|
|
|
* The entry points into this file are
|
|
|
|
* do_mount: perform the MOUNT system call
|
|
|
|
* do_umount: perform the UMOUNT system call
|
|
|
|
*
|
|
|
|
* Changes for VFS:
|
|
|
|
* Jul 2006 (Balazs Gerofi)
|
|
|
|
*/
|
|
|
|
|
|
|
|
#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 <unistd.h>
|
|
|
|
#include <sys/stat.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;
|
|
|
|
|
|
|
|
FORWARD _PROTOTYPE( dev_t name_to_dev, (void) );
|
|
|
|
FORWARD _PROTOTYPE( int mount_fs, (endpoint_t fs_e) );
|
|
|
|
|
|
|
|
/*===========================================================================*
|
|
|
|
* do_fslogin *
|
|
|
|
*===========================================================================*/
|
|
|
|
PUBLIC int do_fslogin()
|
|
|
|
{
|
|
|
|
/* Login before mount request */
|
|
|
|
if ((unsigned long)mount_m_in.m1_p3 != who_e) {
|
|
|
|
last_login_fs_e = who_e;
|
|
|
|
return SUSPEND;
|
|
|
|
}
|
|
|
|
/* Login after a suspended mount */
|
|
|
|
else {
|
|
|
|
/* Copy back original mount request message */
|
|
|
|
m_in = mount_m_in;
|
|
|
|
|
|
|
|
/* Set up last login FS */
|
|
|
|
last_login_fs_e = who_e;
|
|
|
|
|
|
|
|
/* Set up endpoint and call nr */
|
|
|
|
who_e = m_in.m_source;
|
|
|
|
who_p = _ENDPOINT_P(who_e);
|
|
|
|
call_nr = m_in.m_type;
|
|
|
|
fp = &fproc[who_p]; /* pointer to proc table struct */
|
|
|
|
super_user = (fp->fp_effuid == SU_UID ? TRUE : FALSE); /* su? */
|
|
|
|
|
|
|
|
|
|
|
|
return do_mount();
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*===========================================================================*
|
|
|
|
* do_mount *
|
|
|
|
*===========================================================================*/
|
|
|
|
PUBLIC int do_mount()
|
|
|
|
{
|
|
|
|
endpoint_t fs_e;
|
2007-01-16 14:57:35 +01:00
|
|
|
|
|
|
|
/* Only the super-user may do MOUNT. */
|
|
|
|
if (!super_user) return(EPERM);
|
2006-10-25 15:40:36 +02:00
|
|
|
|
|
|
|
/* FS process' endpoint number */
|
|
|
|
fs_e = (unsigned long)m_in.m1_p3;
|
2006-11-27 15:21:43 +01:00
|
|
|
|
2007-01-16 14:57:35 +01:00
|
|
|
/* Sanity check on process number. */
|
|
|
|
if(fs_e <= 0) {
|
|
|
|
printf("vfs: warning: got process number %d for mount call.\n", fs_e);
|
|
|
|
return EINVAL;
|
|
|
|
}
|
|
|
|
|
2006-10-25 15:40:36 +02:00
|
|
|
/* Do the actual job */
|
2007-01-22 16:25:41 +01:00
|
|
|
return mount_fs(fs_e);
|
2006-10-25 15:40:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
/*===========================================================================*
|
|
|
|
* mount *
|
|
|
|
*===========================================================================*/
|
|
|
|
PRIVATE int mount_fs(endpoint_t fs_e)
|
|
|
|
{
|
|
|
|
/* Perform the mount(name, mfile, rd_only) system call. */
|
|
|
|
int rdir, mdir; /* TRUE iff {root|mount} file is dir */
|
|
|
|
int i, r, found;
|
|
|
|
struct fproc *tfp;
|
|
|
|
struct dmap *dp;
|
|
|
|
dev_t dev;
|
|
|
|
message m;
|
|
|
|
struct vnode *vp, *root_node, *mounted_on, *bspec;
|
|
|
|
struct vmnt *vmp, *vmp2;
|
|
|
|
struct mountpoint_req mreq;
|
|
|
|
struct node_details res;
|
|
|
|
struct readsuper_req sreq;
|
|
|
|
struct readsuper_res sres;
|
|
|
|
struct lookup_req lookup_req;
|
2007-01-05 17:36:55 +01:00
|
|
|
node_req_t node_req;
|
2006-10-25 15:40:36 +02:00
|
|
|
|
|
|
|
/* Only the super-user may do MOUNT. */
|
|
|
|
if (!super_user) return(EPERM);
|
|
|
|
|
|
|
|
/* If FS not yet logged in, save message and suspend mount */
|
|
|
|
if (last_login_fs_e != fs_e) {
|
|
|
|
mount_m_in = m_in;
|
|
|
|
return SUSPEND;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Mount request got after FS login or
|
|
|
|
* FS login arrived after a suspended mount */
|
2007-01-16 14:57:35 +01:00
|
|
|
last_login_fs_e = NONE;
|
2006-10-25 15:40:36 +02:00
|
|
|
|
|
|
|
/* Clear endpoint field */
|
2007-01-16 14:57:35 +01:00
|
|
|
mount_m_in.m1_p3 = (char *) NONE;
|
2006-10-25 15:40:36 +02:00
|
|
|
|
|
|
|
/* 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);
|
|
|
|
|
|
|
|
/* Get free vnode for the mountpoint */
|
2007-01-05 17:36:55 +01:00
|
|
|
if ((mounted_on = get_free_vnode(__FILE__, __LINE__)) == NIL_VNODE) {
|
2006-10-25 15:40:36 +02:00
|
|
|
printf("VFSmount: not free vnode available\n");
|
|
|
|
return ENFILE;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Mark it as used so that we won't find the same for the root_node */
|
2007-01-05 17:36:55 +01:00
|
|
|
mounted_on->v_ref_count = 1;
|
2006-10-25 15:40:36 +02:00
|
|
|
|
|
|
|
/* Convert name to device number */
|
|
|
|
if ((dev = name_to_dev()) == 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) {
|
2007-01-05 17:36:55 +01:00
|
|
|
if (bspec->v_ref_count > 0 && bspec->v_sdev == dev) {
|
2006-10-25 15:40:36 +02:00
|
|
|
/* Found, sync the buffer cache */
|
|
|
|
req_sync(bspec->v_fs_e);
|
|
|
|
break;
|
|
|
|
/* Note: there are probably some blocks in the FS process' buffer
|
|
|
|
* cache which contain data on this minor, although they will be
|
|
|
|
* purged since the handling moves to the new FS process (if
|
|
|
|
* everything goes well with the mount...)
|
|
|
|
*/
|
|
|
|
}
|
|
|
|
}
|
|
|
|
/* Didn't find? */
|
|
|
|
if (bspec == &vnode[NR_VNODES] && bspec->v_sdev != dev)
|
|
|
|
bspec = NULL;
|
|
|
|
|
|
|
|
/* Scan vmnt table to see if dev already mounted, if not,
|
|
|
|
* find a free slot.*/
|
|
|
|
found = FALSE;
|
|
|
|
vmp = NIL_VMNT;
|
|
|
|
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. */
|
|
|
|
if (vmp->m_mounted_on != vmp->m_root_node ||
|
|
|
|
vmp->m_mounted_on == fproc[FS_PROC_NR].fp_rd) {
|
|
|
|
/* Normally, m_mounted_on refers to the mount point. For a root
|
|
|
|
* filesystem, m_mounted_on is equal to the root vnode. We assume
|
|
|
|
* that the root of FS is always the real root. If the two
|
|
|
|
* vnodes are different or if the root of FS is equal two the
|
|
|
|
* root of the filesystem we found, we found a filesystem that
|
|
|
|
* is in use. */
|
2007-01-05 17:36:55 +01:00
|
|
|
mounted_on->v_ref_count = 0;
|
2006-10-25 15:40:36 +02:00
|
|
|
return EBUSY; /* already mounted */
|
|
|
|
}
|
|
|
|
|
|
|
|
if (root_dev == vmp->m_dev)
|
|
|
|
panic("fs", "inconsistency remounting old root", NO_NUM);
|
|
|
|
|
|
|
|
/* 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);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Fill in lookup request fields */
|
|
|
|
lookup_req.path = user_fullpath;
|
|
|
|
lookup_req.lastc = NULL;
|
|
|
|
lookup_req.flags = EAT_PATH;
|
|
|
|
|
|
|
|
/* Request lookup */
|
|
|
|
if ((r = lookup(&lookup_req, &res)) != OK) return r;
|
|
|
|
|
|
|
|
/* Fill in request message fields.*/
|
|
|
|
mreq.fs_e = res.fs_e;
|
|
|
|
mreq.inode_nr = res.inode_nr;
|
|
|
|
mreq.uid = fp->fp_effuid;
|
|
|
|
mreq.gid = fp->fp_effgid;
|
|
|
|
|
|
|
|
/* Issue request */
|
|
|
|
if ((r = req_mountpoint(&mreq, &res)) != OK) return r;
|
|
|
|
|
|
|
|
mounted_on->v_fs_e = res.fs_e;
|
|
|
|
mounted_on->v_inode_nr = res.inode_nr;
|
|
|
|
mounted_on->v_mode = res.fmode;
|
|
|
|
mounted_on->v_size = res.fsize;
|
|
|
|
mounted_on->v_sdev = NO_DEV;
|
2007-01-05 17:36:55 +01:00
|
|
|
mounted_on->v_fs_count = 1;
|
|
|
|
mounted_on->v_ref_count = 1;
|
2006-10-25 15:40:36 +02:00
|
|
|
|
|
|
|
/* Find the vmnt for the vnode */
|
|
|
|
if ( (vmp2 = find_vmnt(mounted_on->v_fs_e)) == NIL_VMNT)
|
|
|
|
printf("VFS: vmnt not found by mount()\n");
|
|
|
|
mounted_on->v_vmnt = vmp2;
|
|
|
|
mounted_on->v_dev = vmp2->m_dev;
|
|
|
|
|
|
|
|
/* Get the root inode of the mounted file system. */
|
|
|
|
root_node = vmp->m_root_node;
|
|
|
|
|
|
|
|
/* File types may not conflict. */
|
|
|
|
if (r == OK) {
|
|
|
|
mdir = ((mounted_on->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 mount point. */
|
|
|
|
if (r != OK) {
|
|
|
|
put_vnode(mounted_on);
|
|
|
|
|
|
|
|
return(r);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Nothing else can go wrong. Perform the mount. */
|
|
|
|
put_vnode(vmp->m_mounted_on);
|
|
|
|
vmp->m_mounted_on = mounted_on;
|
|
|
|
vmp->m_flags = m_in.rd_only;
|
|
|
|
allow_newroot = 0; /* The root is now fixed */
|
|
|
|
|
|
|
|
return(OK);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* We'll need a vnode for the root inode, check whether there is one */
|
2007-01-05 17:36:55 +01:00
|
|
|
if ((root_node = get_free_vnode(__FILE__, __LINE__)) == NIL_VNODE) {
|
2006-10-25 15:40:36 +02:00
|
|
|
printf("VFSmount: no free vnode available\n");
|
|
|
|
return ENFILE;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Set it back to zero so that if st goes wrong it won't be kept in use */
|
2007-01-05 17:36:55 +01:00
|
|
|
mounted_on->v_ref_count = 0;
|
2006-10-25 15:40:36 +02:00
|
|
|
|
|
|
|
/* Fetch the name of the mountpoint */
|
|
|
|
if (fetch_name(m_in.name2, m_in.name2_length, M1) != OK) {
|
|
|
|
return(err_code);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Get driver process' endpoint */
|
|
|
|
dp = &dmap[(dev >> MAJOR) & BYTE];
|
|
|
|
if (dp->dmap_driver == NONE) {
|
|
|
|
printf("VFSmount: no driver for dev %x\n", dev);
|
|
|
|
return(EINVAL);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Open the device the file system lives on. */
|
|
|
|
if (dev_open(dev, fs_e, m_in.rd_only ? R_BIT : (R_BIT|W_BIT)) != OK) {
|
|
|
|
return(EINVAL);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Request for reading superblock and root inode */
|
|
|
|
sreq.fs_e = fs_e;
|
|
|
|
sreq.readonly = m_in.rd_only;
|
|
|
|
sreq.boottime = boottime;
|
|
|
|
sreq.driver_e = dp->dmap_driver;
|
|
|
|
sreq.dev = dev;
|
|
|
|
sreq.slink_storage = user_fullpath;
|
|
|
|
|
|
|
|
if (!strcmp(user_fullpath, "/")) sreq.isroot = 1;
|
|
|
|
else sreq.isroot = 0;
|
|
|
|
|
|
|
|
/* Issue request */
|
|
|
|
if ((r = req_readsuper(&sreq, &sres)) != OK) {
|
2006-12-22 12:54:42 +01:00
|
|
|
printf("VFSmount: reading superblock error dev: %d\n", dev);
|
2006-10-25 15:40:36 +02:00
|
|
|
dev_close(dev);
|
|
|
|
return r;
|
|
|
|
}
|
|
|
|
|
2007-01-05 17:36:55 +01:00
|
|
|
/* Fill in request message fields.*/
|
|
|
|
node_req.fs_e = sres.fs_e;
|
|
|
|
node_req.inode_nr = sres.inode_nr;
|
|
|
|
|
|
|
|
/* Issue request */
|
|
|
|
if ((r = req_getnode(&node_req, &res)) != OK)
|
|
|
|
{
|
|
|
|
printf("mount: req_getnode failed: %d\n", r);
|
|
|
|
dev_close(dev);
|
|
|
|
return r;
|
|
|
|
}
|
|
|
|
|
2006-10-25 15:40:36 +02:00
|
|
|
/* Fill in root node's fields */
|
2007-01-05 17:36:55 +01:00
|
|
|
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;
|
2006-10-25 15:40:36 +02:00
|
|
|
root_node->v_sdev = NO_DEV;
|
2007-01-05 17:36:55 +01:00
|
|
|
root_node->v_fs_count = 2;
|
|
|
|
root_node->v_ref_count = 1;
|
2006-10-25 15:40:36 +02:00
|
|
|
|
|
|
|
/* Fill in max file size and blocksize for the vmnt */
|
|
|
|
vmp->m_fs_e = sres.fs_e;
|
|
|
|
vmp->m_dev = dev;
|
|
|
|
vmp->m_block_size = sres.blocksize;
|
|
|
|
vmp->m_max_file_size = sres.maxsize;
|
|
|
|
vmp->m_flags = m_in.rd_only;
|
|
|
|
vmp->m_driver_e = dp->dmap_driver;
|
|
|
|
|
|
|
|
/* Root node is indeed on the partition */
|
|
|
|
root_node->v_vmnt = vmp;
|
|
|
|
root_node->v_dev = vmp->m_dev;
|
|
|
|
|
|
|
|
if (strcmp(user_fullpath, "/") == 0 && allow_newroot) {
|
|
|
|
printf("Replacing root\n");
|
|
|
|
|
|
|
|
/* Superblock and root node already read.
|
|
|
|
* Nothing else can go wrong. Perform the mount. */
|
|
|
|
vmp->m_root_node = root_node;
|
2007-01-05 17:36:55 +01:00
|
|
|
root_node->v_ref_count++;
|
2006-10-25 15:40:36 +02:00
|
|
|
vmp->m_mounted_on = root_node;
|
|
|
|
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;
|
|
|
|
|
|
|
|
if (tfp->fp_rd == NULL)
|
|
|
|
panic("fs", "do_mount: null rootdir", i);
|
|
|
|
put_vnode(tfp->fp_rd);
|
|
|
|
dup_vnode(root_node);
|
|
|
|
tfp->fp_rd = root_node;
|
|
|
|
|
|
|
|
if (tfp->fp_wd == NULL)
|
|
|
|
panic("fs", "do_mount: null workdir", i);
|
|
|
|
put_vnode(tfp->fp_wd);
|
|
|
|
dup_vnode(root_node);
|
|
|
|
tfp->fp_wd = root_node;
|
|
|
|
}
|
|
|
|
|
|
|
|
return(OK);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Fill in lookup request fields */
|
|
|
|
lookup_req.path = user_fullpath;
|
|
|
|
lookup_req.lastc = NULL;
|
|
|
|
lookup_req.flags = EAT_PATH;
|
|
|
|
|
|
|
|
/* Request lookup */
|
|
|
|
if ((r = lookup(&lookup_req, &res)) != OK) {
|
|
|
|
put_vnode(root_node);
|
|
|
|
return r;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Fill in request fields.*/
|
|
|
|
mreq.fs_e = res.fs_e;
|
|
|
|
mreq.inode_nr = res.inode_nr;
|
|
|
|
mreq.uid = fp->fp_effuid;
|
|
|
|
mreq.gid = fp->fp_effgid;
|
|
|
|
|
|
|
|
/* Issue request */
|
|
|
|
if ((r = req_mountpoint(&mreq, &res)) != OK) {
|
|
|
|
put_vnode(root_node);
|
|
|
|
return r;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Fill in vnode's fields */
|
|
|
|
mounted_on->v_fs_e = res.fs_e;
|
|
|
|
mounted_on->v_inode_nr = res.inode_nr;
|
|
|
|
mounted_on->v_mode = res.fmode;
|
|
|
|
mounted_on->v_size = res.fsize;
|
|
|
|
mounted_on->v_sdev = NO_DEV;
|
2007-01-05 17:36:55 +01:00
|
|
|
mounted_on->v_fs_count = 1;
|
|
|
|
mounted_on->v_ref_count = 1;
|
2006-10-25 15:40:36 +02:00
|
|
|
|
|
|
|
/* Find the vmnt for the vnode */
|
|
|
|
if ( (vmp2 = find_vmnt(mounted_on->v_fs_e)) == NIL_VMNT)
|
|
|
|
printf("VFS: vmnt not found by mount()");
|
|
|
|
mounted_on->v_vmnt = vmp2;
|
|
|
|
mounted_on->v_dev = vmp2->m_dev;
|
|
|
|
|
|
|
|
/* File types may not conflict. */
|
|
|
|
if (r == OK) {
|
|
|
|
mdir = ((mounted_on->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(mounted_on);
|
|
|
|
put_vnode(root_node);
|
|
|
|
|
|
|
|
vmp->m_dev = NO_DEV;
|
|
|
|
dev_close(dev);
|
|
|
|
return(r);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Nothing else can go wrong. Perform the mount. */
|
|
|
|
vmp->m_mounted_on = mounted_on;
|
|
|
|
vmp->m_root_node = root_node;
|
|
|
|
|
|
|
|
/* The root is now fixed */
|
|
|
|
allow_newroot = 0;
|
|
|
|
|
|
|
|
/* There was a block spec file open, and it should be handled by the
|
|
|
|
* new FS proc now */
|
|
|
|
if (bspec) {
|
|
|
|
printf("VFSmount: moving opened block spec to new FS_e: %d...\n", fs_e);
|
|
|
|
bspec->v_bfs_e = fs_e;
|
|
|
|
}
|
|
|
|
return(OK);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*===========================================================================*
|
|
|
|
* do_umount *
|
|
|
|
*===========================================================================*/
|
|
|
|
PUBLIC int do_umount()
|
|
|
|
{
|
|
|
|
/* Perform the umount(name) system call. */
|
|
|
|
dev_t dev;
|
|
|
|
|
|
|
|
/* Only the super-user may do UMOUNT. */
|
|
|
|
if (!super_user) return(EPERM);
|
|
|
|
|
|
|
|
/* If 'name' is not for a block special file, return error. */
|
|
|
|
if (fetch_name(m_in.name, m_in.name_length, M3) != OK) return(err_code);
|
|
|
|
if ( (dev = name_to_dev()) == NO_DEV) return(err_code);
|
|
|
|
|
|
|
|
return(unmount(dev));
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
/*===========================================================================*
|
|
|
|
* unmount *
|
|
|
|
*===========================================================================*/
|
|
|
|
PUBLIC int unmount(dev)
|
|
|
|
Dev_t dev;
|
|
|
|
{
|
|
|
|
struct vnode *vp;
|
2007-01-16 14:57:35 +01:00
|
|
|
struct vmnt *vmp_i = NULL, *vmp = NULL;
|
2006-10-25 15:40:36 +02:00
|
|
|
struct dmap *dp;
|
|
|
|
int count, r;
|
|
|
|
int fs_e;
|
|
|
|
|
|
|
|
/* Find vmnt */
|
2007-01-16 14:57:35 +01:00
|
|
|
for (vmp_i = &vmnt[0]; vmp_i < &vmnt[NR_MNTS]; ++vmp_i) {
|
2007-01-22 16:25:41 +01:00
|
|
|
if (vmp_i->m_dev == dev) {
|
2007-01-16 14:57:35 +01:00
|
|
|
if(vmp) panic(__FILE__, "device mounted more than once", dev);
|
|
|
|
vmp = vmp_i;
|
|
|
|
}
|
2006-10-25 15:40:36 +02:00
|
|
|
}
|
|
|
|
|
2007-01-16 14:57:35 +01:00
|
|
|
/* Device mounted? */
|
|
|
|
if(!vmp)
|
|
|
|
return EINVAL;
|
|
|
|
|
2006-10-25 15:40:36 +02:00
|
|
|
/* 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++) {
|
2007-01-05 17:36:55 +01:00
|
|
|
if (vp->v_ref_count > 0 && vp->v_dev == dev) {
|
|
|
|
count += vp->v_ref_count;
|
2006-10-25 15:40:36 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (count > 1) {
|
|
|
|
return(EBUSY); /* can't umount a busy file system */
|
|
|
|
}
|
|
|
|
|
2007-01-05 17:36:55 +01:00
|
|
|
vnode_clean_refs(vmp->m_root_node);
|
|
|
|
|
2006-10-25 15:40:36 +02:00
|
|
|
/* Request FS the unmount */
|
2007-01-16 14:57:35 +01:00
|
|
|
if(vmp->m_fs_e <= 0)
|
|
|
|
panic(__FILE__, "unmount: strange fs endpoint", vmp->m_fs_e);
|
2006-10-25 15:40:36 +02:00
|
|
|
if ((r = req_unmount(vmp->m_fs_e)) != OK) return r;
|
|
|
|
|
|
|
|
/* Close the device the file system lives on. */
|
|
|
|
dev_close(dev);
|
|
|
|
|
|
|
|
/* Is there a block special file that was handled by that partition? */
|
|
|
|
for (vp = &vnode[0]; vp < &vnode[NR_VNODES]; vp++) {
|
|
|
|
if ((vp->v_mode & I_TYPE) == I_BLOCK_SPECIAL &&
|
|
|
|
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("VFSblock_spec_open: driver not found for device %d\n",
|
|
|
|
dev);
|
|
|
|
/* What should be done, panic??? */
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
|
|
|
printf("VFSunmount: moving block spec %d to root FS\n", dev);
|
|
|
|
vp->v_bfs_e = ROOT_FS_E;
|
|
|
|
vp->v_blocksize = _MIN_BLOCK_SIZE;
|
|
|
|
|
|
|
|
/* Send the driver endpoint (even if it is known already...) */
|
|
|
|
if ((r = req_newdriver(vp->v_bfs_e, vp->v_sdev, dp->dmap_driver))
|
|
|
|
!= OK) {
|
|
|
|
printf("VFSunmount: error sending driver endpoint for block spec\n");
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Root device is mounted on itself */
|
|
|
|
if (vmp->m_root_node != vmp->m_mounted_on) {
|
|
|
|
put_vnode(vmp->m_mounted_on);
|
2007-01-05 17:36:55 +01:00
|
|
|
vmp->m_root_node->v_ref_count = 0;
|
2006-10-25 15:40:36 +02:00
|
|
|
}
|
|
|
|
else {
|
2007-01-05 17:36:55 +01:00
|
|
|
vmp->m_mounted_on->v_fs_count--;
|
2006-10-25 15:40:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
fs_e = vmp->m_fs_e;
|
|
|
|
|
|
|
|
vmp->m_root_node = NIL_VNODE;
|
|
|
|
vmp->m_mounted_on = NIL_VNODE;
|
|
|
|
vmp->m_dev = NO_DEV;
|
2007-01-16 14:57:35 +01:00
|
|
|
vmp->m_fs_e = NONE;
|
|
|
|
vmp->m_driver_e = NONE;
|
2006-10-25 15:40:36 +02:00
|
|
|
|
|
|
|
return(OK);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*===========================================================================*
|
|
|
|
* name_to_dev *
|
|
|
|
*===========================================================================*/
|
|
|
|
PRIVATE dev_t name_to_dev()
|
|
|
|
{
|
|
|
|
/* Convert the block special file 'path' to a device number. If 'path'
|
|
|
|
* is not a block special file, return error code in 'err_code'. */
|
|
|
|
struct lookup_req lookup_req;
|
|
|
|
struct node_details res;
|
|
|
|
int r;
|
|
|
|
|
|
|
|
/* Fill in lookup request fields */
|
|
|
|
lookup_req.path = user_fullpath;
|
|
|
|
lookup_req.lastc = NULL;
|
|
|
|
lookup_req.flags = EAT_PATH;
|
|
|
|
|
|
|
|
/* Request lookup */
|
|
|
|
if ((r = lookup(&lookup_req, &res)) != OK) return r;
|
|
|
|
|
|
|
|
if ((res.fmode & I_TYPE) != I_BLOCK_SPECIAL) {
|
|
|
|
err_code = ENOTBLK;
|
|
|
|
return NO_DEV;
|
|
|
|
}
|
|
|
|
|
|
|
|
return res.dev;
|
|
|
|
}
|
|
|
|
|