minix/servers/vfs/protect.c

262 lines
7.2 KiB
C
Raw Normal View History

/* This file deals with protection in the file system. It contains the code
* for four system calls that relate to protection.
*
* The entry points into this file are
* do_chmod: perform the CHMOD and FCHMOD system calls
* do_chown: perform the CHOWN and FCHOWN system calls
* do_umask: perform the UMASK system call
* do_access: perform the ACCESS system call
*
* Changes for VFS:
* Jul 2006 (Balazs Gerofi)
*/
#include "fs.h"
#include <unistd.h>
#include <minix/callnr.h>
#include "file.h"
#include "fproc.h"
#include "param.h"
#include <minix/vfsif.h>
#include "vnode.h"
#include "vmnt.h"
/*===========================================================================*
* do_chmod *
*===========================================================================*/
PUBLIC int do_chmod()
{
struct filp *flp;
struct chmod_req req;
struct lookup_req lookup_req;
struct node_details res;
Mostly bugfixes of bugs triggered by the test set. bugfixes: SYSTEM: . removed rc->p_priv->s_flags = 0; for the priv struct shared by all user processes in get_priv(). this should only be done once. doing a SYS_PRIV_USER in sys_privctl() caused the flags of all user processes to be reset, so they were no longer PREEMPTIBLE. this happened when RS executed a policy script. (this broke test1 in the test set) VFS/MFS: . chown can change the mode of a file, and chmod arguments are only part of the full file mode so the full filemode is slightly magic. changed these calls so that the final modes are returned to VFS, so that the vnode can be kept up-to-date. (this broke test11 in the test set) MFS: . lookup() checked for sizeof(string) instead of sizeof(user_path), truncating long path names (caught by test 23) . truncate functions neglected to update ctime (this broke test16) VFS: . corner case of an empty filename lookup caused fields of a request not to be filled in in the lookup functions, not making it clear that the lookup had failed, causing messages to garbage processes, causing strange failures. (caught by test 30) . trust v_size in vnode when doing reads or writes on non-special files, truncating i/o where necessary; this is necessary for pipes, as MFS can't tell when a pipe has been truncated without it being told explicitly each time. when the last reader/writer on a pipe closes, tell FS about the new size using truncate_vn(). (this broke test 25, among others) . permission check for chdir() had disappeared; added a forbidden() call (caught by test 23) new code, shouldn't change anything: . introduced RTS_SET, RTS_UNSET, and RTS_ISSET macro's, and their LOCK variants. These macros set and clear the p_rts_flags field, causing a lot of duplicated logic like old_flags = rp->p_rts_flags; /* save value of the flags */ rp->p_rts_flags &= ~NO_PRIV; if (old_flags != 0 && rp->p_rts_flags == 0) lock_enqueue(rp); to change into the simpler RTS_LOCK_UNSET(rp, NO_PRIV); so the macros take care of calling dequeue() and enqueue() (or lock_*()), as the case may be). This makes the code a bit more readable and a bit less fragile. . removed return code from do_clocktick in CLOCK as it currently never replies . removed some debug code from VFS . fixed grant debug message in device.c preemptive checks, tests, changes: . added return code checks of receive() to SYSTEM and CLOCK . O_TRUNC should never arrive at MFS (added sanity check and removed O_TRUNC code) . user_path declared with PATH_MAX+1 to let it be null-terminated . checks in MFS to see if strings passed by VFS are null-terminated IS: . static irq name table thrown out
2007-02-01 18:50:02 +01:00
struct vnode *vp;
int r, ch_mode;
if (call_nr == CHMOD) {
/* Perform the chmod(name, mode) system call. */
if (fetch_name(m_in.name, m_in.name_length, M3) != 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;
req.inode_nr = res.inode_nr;
req.fs_e = res.fs_e;
}
else if (call_nr == FCHMOD) {
if (!(flp = get_filp(m_in.m3_i1))) return err_code;
req.inode_nr = flp->filp_vno->v_inode_nr;
req.fs_e = flp->filp_vno->v_fs_e;
}
else panic(__FILE__, "do_chmod called with strange call_nr", call_nr);
Mostly bugfixes of bugs triggered by the test set. bugfixes: SYSTEM: . removed rc->p_priv->s_flags = 0; for the priv struct shared by all user processes in get_priv(). this should only be done once. doing a SYS_PRIV_USER in sys_privctl() caused the flags of all user processes to be reset, so they were no longer PREEMPTIBLE. this happened when RS executed a policy script. (this broke test1 in the test set) VFS/MFS: . chown can change the mode of a file, and chmod arguments are only part of the full file mode so the full filemode is slightly magic. changed these calls so that the final modes are returned to VFS, so that the vnode can be kept up-to-date. (this broke test11 in the test set) MFS: . lookup() checked for sizeof(string) instead of sizeof(user_path), truncating long path names (caught by test 23) . truncate functions neglected to update ctime (this broke test16) VFS: . corner case of an empty filename lookup caused fields of a request not to be filled in in the lookup functions, not making it clear that the lookup had failed, causing messages to garbage processes, causing strange failures. (caught by test 30) . trust v_size in vnode when doing reads or writes on non-special files, truncating i/o where necessary; this is necessary for pipes, as MFS can't tell when a pipe has been truncated without it being told explicitly each time. when the last reader/writer on a pipe closes, tell FS about the new size using truncate_vn(). (this broke test 25, among others) . permission check for chdir() had disappeared; added a forbidden() call (caught by test 23) new code, shouldn't change anything: . introduced RTS_SET, RTS_UNSET, and RTS_ISSET macro's, and their LOCK variants. These macros set and clear the p_rts_flags field, causing a lot of duplicated logic like old_flags = rp->p_rts_flags; /* save value of the flags */ rp->p_rts_flags &= ~NO_PRIV; if (old_flags != 0 && rp->p_rts_flags == 0) lock_enqueue(rp); to change into the simpler RTS_LOCK_UNSET(rp, NO_PRIV); so the macros take care of calling dequeue() and enqueue() (or lock_*()), as the case may be). This makes the code a bit more readable and a bit less fragile. . removed return code from do_clocktick in CLOCK as it currently never replies . removed some debug code from VFS . fixed grant debug message in device.c preemptive checks, tests, changes: . added return code checks of receive() to SYSTEM and CLOCK . O_TRUNC should never arrive at MFS (added sanity check and removed O_TRUNC code) . user_path declared with PATH_MAX+1 to let it be null-terminated . checks in MFS to see if strings passed by VFS are null-terminated IS: . static irq name table thrown out
2007-02-01 18:50:02 +01:00
/* Find vnode, if it's in use. */
vp = find_vnode(req.fs_e, req.inode_nr);
/* Fill in request message fields.*/
req.uid = fp->fp_effuid;
req.gid = fp->fp_effgid;
req.rmode = m_in.mode;
/* Issue request */
Mostly bugfixes of bugs triggered by the test set. bugfixes: SYSTEM: . removed rc->p_priv->s_flags = 0; for the priv struct shared by all user processes in get_priv(). this should only be done once. doing a SYS_PRIV_USER in sys_privctl() caused the flags of all user processes to be reset, so they were no longer PREEMPTIBLE. this happened when RS executed a policy script. (this broke test1 in the test set) VFS/MFS: . chown can change the mode of a file, and chmod arguments are only part of the full file mode so the full filemode is slightly magic. changed these calls so that the final modes are returned to VFS, so that the vnode can be kept up-to-date. (this broke test11 in the test set) MFS: . lookup() checked for sizeof(string) instead of sizeof(user_path), truncating long path names (caught by test 23) . truncate functions neglected to update ctime (this broke test16) VFS: . corner case of an empty filename lookup caused fields of a request not to be filled in in the lookup functions, not making it clear that the lookup had failed, causing messages to garbage processes, causing strange failures. (caught by test 30) . trust v_size in vnode when doing reads or writes on non-special files, truncating i/o where necessary; this is necessary for pipes, as MFS can't tell when a pipe has been truncated without it being told explicitly each time. when the last reader/writer on a pipe closes, tell FS about the new size using truncate_vn(). (this broke test 25, among others) . permission check for chdir() had disappeared; added a forbidden() call (caught by test 23) new code, shouldn't change anything: . introduced RTS_SET, RTS_UNSET, and RTS_ISSET macro's, and their LOCK variants. These macros set and clear the p_rts_flags field, causing a lot of duplicated logic like old_flags = rp->p_rts_flags; /* save value of the flags */ rp->p_rts_flags &= ~NO_PRIV; if (old_flags != 0 && rp->p_rts_flags == 0) lock_enqueue(rp); to change into the simpler RTS_LOCK_UNSET(rp, NO_PRIV); so the macros take care of calling dequeue() and enqueue() (or lock_*()), as the case may be). This makes the code a bit more readable and a bit less fragile. . removed return code from do_clocktick in CLOCK as it currently never replies . removed some debug code from VFS . fixed grant debug message in device.c preemptive checks, tests, changes: . added return code checks of receive() to SYSTEM and CLOCK . O_TRUNC should never arrive at MFS (added sanity check and removed O_TRUNC code) . user_path declared with PATH_MAX+1 to let it be null-terminated . checks in MFS to see if strings passed by VFS are null-terminated IS: . static irq name table thrown out
2007-02-01 18:50:02 +01:00
if((r = req_chmod(&req, &ch_mode)) != OK) return r;
if(vp != NIL_VNODE)
vp->v_mode = ch_mode;
return OK;
}
/*===========================================================================*
* do_chown *
*===========================================================================*/
PUBLIC int do_chown()
{
int inode_nr;
int fs_e;
struct filp *flp;
struct chown_req req;
struct lookup_req lookup_req;
struct node_details res;
Mostly bugfixes of bugs triggered by the test set. bugfixes: SYSTEM: . removed rc->p_priv->s_flags = 0; for the priv struct shared by all user processes in get_priv(). this should only be done once. doing a SYS_PRIV_USER in sys_privctl() caused the flags of all user processes to be reset, so they were no longer PREEMPTIBLE. this happened when RS executed a policy script. (this broke test1 in the test set) VFS/MFS: . chown can change the mode of a file, and chmod arguments are only part of the full file mode so the full filemode is slightly magic. changed these calls so that the final modes are returned to VFS, so that the vnode can be kept up-to-date. (this broke test11 in the test set) MFS: . lookup() checked for sizeof(string) instead of sizeof(user_path), truncating long path names (caught by test 23) . truncate functions neglected to update ctime (this broke test16) VFS: . corner case of an empty filename lookup caused fields of a request not to be filled in in the lookup functions, not making it clear that the lookup had failed, causing messages to garbage processes, causing strange failures. (caught by test 30) . trust v_size in vnode when doing reads or writes on non-special files, truncating i/o where necessary; this is necessary for pipes, as MFS can't tell when a pipe has been truncated without it being told explicitly each time. when the last reader/writer on a pipe closes, tell FS about the new size using truncate_vn(). (this broke test 25, among others) . permission check for chdir() had disappeared; added a forbidden() call (caught by test 23) new code, shouldn't change anything: . introduced RTS_SET, RTS_UNSET, and RTS_ISSET macro's, and their LOCK variants. These macros set and clear the p_rts_flags field, causing a lot of duplicated logic like old_flags = rp->p_rts_flags; /* save value of the flags */ rp->p_rts_flags &= ~NO_PRIV; if (old_flags != 0 && rp->p_rts_flags == 0) lock_enqueue(rp); to change into the simpler RTS_LOCK_UNSET(rp, NO_PRIV); so the macros take care of calling dequeue() and enqueue() (or lock_*()), as the case may be). This makes the code a bit more readable and a bit less fragile. . removed return code from do_clocktick in CLOCK as it currently never replies . removed some debug code from VFS . fixed grant debug message in device.c preemptive checks, tests, changes: . added return code checks of receive() to SYSTEM and CLOCK . O_TRUNC should never arrive at MFS (added sanity check and removed O_TRUNC code) . user_path declared with PATH_MAX+1 to let it be null-terminated . checks in MFS to see if strings passed by VFS are null-terminated IS: . static irq name table thrown out
2007-02-01 18:50:02 +01:00
struct vnode *vp;
int r, ch_mode;
if (call_nr == CHOWN) {
/* Perform the chmod(name, mode) system call. */
if (fetch_name(m_in.name1, m_in.name1_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;
req.inode_nr = res.inode_nr;
req.fs_e = res.fs_e;
}
else if (call_nr == FCHOWN) {
if (!(flp = get_filp(m_in.m1_i1))) return err_code;
req.inode_nr = flp->filp_vno->v_inode_nr;
req.fs_e = flp->filp_vno->v_fs_e;
}
else panic(__FILE__, "do_chmod called with strange call_nr", call_nr);
Mostly bugfixes of bugs triggered by the test set. bugfixes: SYSTEM: . removed rc->p_priv->s_flags = 0; for the priv struct shared by all user processes in get_priv(). this should only be done once. doing a SYS_PRIV_USER in sys_privctl() caused the flags of all user processes to be reset, so they were no longer PREEMPTIBLE. this happened when RS executed a policy script. (this broke test1 in the test set) VFS/MFS: . chown can change the mode of a file, and chmod arguments are only part of the full file mode so the full filemode is slightly magic. changed these calls so that the final modes are returned to VFS, so that the vnode can be kept up-to-date. (this broke test11 in the test set) MFS: . lookup() checked for sizeof(string) instead of sizeof(user_path), truncating long path names (caught by test 23) . truncate functions neglected to update ctime (this broke test16) VFS: . corner case of an empty filename lookup caused fields of a request not to be filled in in the lookup functions, not making it clear that the lookup had failed, causing messages to garbage processes, causing strange failures. (caught by test 30) . trust v_size in vnode when doing reads or writes on non-special files, truncating i/o where necessary; this is necessary for pipes, as MFS can't tell when a pipe has been truncated without it being told explicitly each time. when the last reader/writer on a pipe closes, tell FS about the new size using truncate_vn(). (this broke test 25, among others) . permission check for chdir() had disappeared; added a forbidden() call (caught by test 23) new code, shouldn't change anything: . introduced RTS_SET, RTS_UNSET, and RTS_ISSET macro's, and their LOCK variants. These macros set and clear the p_rts_flags field, causing a lot of duplicated logic like old_flags = rp->p_rts_flags; /* save value of the flags */ rp->p_rts_flags &= ~NO_PRIV; if (old_flags != 0 && rp->p_rts_flags == 0) lock_enqueue(rp); to change into the simpler RTS_LOCK_UNSET(rp, NO_PRIV); so the macros take care of calling dequeue() and enqueue() (or lock_*()), as the case may be). This makes the code a bit more readable and a bit less fragile. . removed return code from do_clocktick in CLOCK as it currently never replies . removed some debug code from VFS . fixed grant debug message in device.c preemptive checks, tests, changes: . added return code checks of receive() to SYSTEM and CLOCK . O_TRUNC should never arrive at MFS (added sanity check and removed O_TRUNC code) . user_path declared with PATH_MAX+1 to let it be null-terminated . checks in MFS to see if strings passed by VFS are null-terminated IS: . static irq name table thrown out
2007-02-01 18:50:02 +01:00
/* Find vnode, if it's in use. */
vp = find_vnode(req.fs_e, req.inode_nr);
/* Fill in request message fields.*/
req.uid = fp->fp_effuid;
req.gid = fp->fp_effgid;
req.newuid = m_in.owner;
req.newgid = m_in.group;
/* Issue request */
Mostly bugfixes of bugs triggered by the test set. bugfixes: SYSTEM: . removed rc->p_priv->s_flags = 0; for the priv struct shared by all user processes in get_priv(). this should only be done once. doing a SYS_PRIV_USER in sys_privctl() caused the flags of all user processes to be reset, so they were no longer PREEMPTIBLE. this happened when RS executed a policy script. (this broke test1 in the test set) VFS/MFS: . chown can change the mode of a file, and chmod arguments are only part of the full file mode so the full filemode is slightly magic. changed these calls so that the final modes are returned to VFS, so that the vnode can be kept up-to-date. (this broke test11 in the test set) MFS: . lookup() checked for sizeof(string) instead of sizeof(user_path), truncating long path names (caught by test 23) . truncate functions neglected to update ctime (this broke test16) VFS: . corner case of an empty filename lookup caused fields of a request not to be filled in in the lookup functions, not making it clear that the lookup had failed, causing messages to garbage processes, causing strange failures. (caught by test 30) . trust v_size in vnode when doing reads or writes on non-special files, truncating i/o where necessary; this is necessary for pipes, as MFS can't tell when a pipe has been truncated without it being told explicitly each time. when the last reader/writer on a pipe closes, tell FS about the new size using truncate_vn(). (this broke test 25, among others) . permission check for chdir() had disappeared; added a forbidden() call (caught by test 23) new code, shouldn't change anything: . introduced RTS_SET, RTS_UNSET, and RTS_ISSET macro's, and their LOCK variants. These macros set and clear the p_rts_flags field, causing a lot of duplicated logic like old_flags = rp->p_rts_flags; /* save value of the flags */ rp->p_rts_flags &= ~NO_PRIV; if (old_flags != 0 && rp->p_rts_flags == 0) lock_enqueue(rp); to change into the simpler RTS_LOCK_UNSET(rp, NO_PRIV); so the macros take care of calling dequeue() and enqueue() (or lock_*()), as the case may be). This makes the code a bit more readable and a bit less fragile. . removed return code from do_clocktick in CLOCK as it currently never replies . removed some debug code from VFS . fixed grant debug message in device.c preemptive checks, tests, changes: . added return code checks of receive() to SYSTEM and CLOCK . O_TRUNC should never arrive at MFS (added sanity check and removed O_TRUNC code) . user_path declared with PATH_MAX+1 to let it be null-terminated . checks in MFS to see if strings passed by VFS are null-terminated IS: . static irq name table thrown out
2007-02-01 18:50:02 +01:00
r = req_chown(&req, &ch_mode);
if(r == OK && vp) {
vp->v_uid = m_in.owner;
vp->v_gid = m_in.group;
vp->v_mode = ch_mode;
}
return r;
}
/*===========================================================================*
* do_umask *
*===========================================================================*/
PUBLIC int do_umask()
{
/* Perform the umask(co_mode) system call. */
register mode_t r;
r = ~fp->fp_umask; /* set 'r' to complement of old mask */
fp->fp_umask = ~(m_in.co_mode & RWX_MODES);
return(r); /* return complement of old mask */
}
/*===========================================================================*
* do_access *
*===========================================================================*/
PUBLIC int do_access()
{
/* Perform the access(name, mode) system call. */
struct access_req req;
struct lookup_req lookup_req;
struct node_details res;
int r;
/* First check to see if the mode is correct. */
if ( (m_in.mode & ~(R_OK | W_OK | X_OK)) != 0 && m_in.mode != F_OK)
return(EINVAL);
if (fetch_name(m_in.name, m_in.name_length, M3) != 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 fields */
req.fs_e = res.fs_e;
req.amode = m_in.mode;
req.inode_nr = res.inode_nr;
req.uid = fp->fp_realuid; /* real user and group id */
req.gid = fp->fp_realgid;
/* Issue request */
return req_access(&req);
}
2007-01-05 17:36:55 +01:00
/*===========================================================================*
* forbidden *
*===========================================================================*/
PUBLIC int forbidden(struct vnode *vp, mode_t access_desired)
{
/* Given a pointer to an inode, 'rip', and the access desired, determine
* if the access is allowed, and if not why not. The routine looks up the
* caller's uid in the 'fproc' table. If access is allowed, OK is returned
* if it is forbidden, EACCES is returned.
*/
register struct super_block *sp;
register mode_t bits, perm_bits;
int r, shift, type;
if (vp->v_uid == (uid_t)-1 || vp->v_gid == (gid_t)-1)
{
printf("forbidden: bad uid/gid in vnode\n");
printf("forbidden: last allocated at %s, %d\n", vp->v_file, vp->v_line);
return EACCES;
}
/* Isolate the relevant rwx bits from the mode. */
bits = vp->v_mode;
if (fp->fp_effuid == SU_UID) {
/* Grant read and write permission. Grant search permission for
* directories. Grant execute permission (for non-directories) if
* and only if one of the 'X' bits is set.
*/
if ( (bits & I_TYPE) == I_DIRECTORY ||
bits & ((X_BIT << 6) | (X_BIT << 3) | X_BIT))
perm_bits = R_BIT | W_BIT | X_BIT;
else
perm_bits = R_BIT | W_BIT;
} else {
if (fp->fp_effuid == vp->v_uid) shift = 6; /* owner */
else if (fp->fp_effgid == vp->v_gid ) shift = 3; /* group */
else shift = 0; /* other */
perm_bits = (bits >> shift) & (R_BIT | W_BIT | X_BIT);
}
/* If access desired is not a subset of what is allowed, it is refused. */
r = OK;
Mostly bugfixes of bugs triggered by the test set. bugfixes: SYSTEM: . removed rc->p_priv->s_flags = 0; for the priv struct shared by all user processes in get_priv(). this should only be done once. doing a SYS_PRIV_USER in sys_privctl() caused the flags of all user processes to be reset, so they were no longer PREEMPTIBLE. this happened when RS executed a policy script. (this broke test1 in the test set) VFS/MFS: . chown can change the mode of a file, and chmod arguments are only part of the full file mode so the full filemode is slightly magic. changed these calls so that the final modes are returned to VFS, so that the vnode can be kept up-to-date. (this broke test11 in the test set) MFS: . lookup() checked for sizeof(string) instead of sizeof(user_path), truncating long path names (caught by test 23) . truncate functions neglected to update ctime (this broke test16) VFS: . corner case of an empty filename lookup caused fields of a request not to be filled in in the lookup functions, not making it clear that the lookup had failed, causing messages to garbage processes, causing strange failures. (caught by test 30) . trust v_size in vnode when doing reads or writes on non-special files, truncating i/o where necessary; this is necessary for pipes, as MFS can't tell when a pipe has been truncated without it being told explicitly each time. when the last reader/writer on a pipe closes, tell FS about the new size using truncate_vn(). (this broke test 25, among others) . permission check for chdir() had disappeared; added a forbidden() call (caught by test 23) new code, shouldn't change anything: . introduced RTS_SET, RTS_UNSET, and RTS_ISSET macro's, and their LOCK variants. These macros set and clear the p_rts_flags field, causing a lot of duplicated logic like old_flags = rp->p_rts_flags; /* save value of the flags */ rp->p_rts_flags &= ~NO_PRIV; if (old_flags != 0 && rp->p_rts_flags == 0) lock_enqueue(rp); to change into the simpler RTS_LOCK_UNSET(rp, NO_PRIV); so the macros take care of calling dequeue() and enqueue() (or lock_*()), as the case may be). This makes the code a bit more readable and a bit less fragile. . removed return code from do_clocktick in CLOCK as it currently never replies . removed some debug code from VFS . fixed grant debug message in device.c preemptive checks, tests, changes: . added return code checks of receive() to SYSTEM and CLOCK . O_TRUNC should never arrive at MFS (added sanity check and removed O_TRUNC code) . user_path declared with PATH_MAX+1 to let it be null-terminated . checks in MFS to see if strings passed by VFS are null-terminated IS: . static irq name table thrown out
2007-02-01 18:50:02 +01:00
if ((perm_bits | access_desired) != perm_bits) {
r = EACCES;
}
2007-01-05 17:36:55 +01:00
/* Check to see if someone is trying to write on a file system that is
* mounted read-only.
*/
if (r == OK)
if (access_desired & W_BIT)
r = read_only(vp);
return(r);
}
/*===========================================================================*
* read_only *
*===========================================================================*/
PUBLIC int read_only(vp)
struct vnode *vp; /* ptr to inode whose file sys is to be cked */
{
/* Check to see if the file system on which the inode 'ip' resides is mounted
* read only. If so, return EROFS, else return OK.
*/
register struct vmnt *mp;
mp = vp->v_vmnt;
return(mp->m_flags ? EROFS : OK);
}