nfs4xdr.c

linux 内核源代码

			if (argp->pagelen < 4) {
				/* There isn't an opcode still on the wire */
				op->opnum = OP_WRITE + 1;
				op->status = nfserr_bad_xdr;
				argp->opcnt = i+1;
				break;
			}

			/*
			 * False alarm. We just hit a page boundary, but there
			 * is still data available.  Move pointer across page
			 * boundary.  *snip from READ_BUF*
			 */
			argp->p = page_address(argp->pagelist[0]);
			argp->pagelist++;
			if (argp->pagelen < PAGE_SIZE) {
				argp->end = p + (argp->pagelen>>2);
				argp->pagelen = 0;
			} else {
				argp->end = p + (PAGE_SIZE>>2);
				argp->pagelen -= PAGE_SIZE;
			}
		}
		op->opnum = ntohl(*argp->p++);

		switch (op->opnum) {
		case 2: /* Reserved operation */
			op->opnum = OP_ILLEGAL;
			if (argp->minorversion == 0)
				op->status = nfserr_op_illegal;
			else
				op->status = nfserr_minor_vers_mismatch;
			break;
		case OP_ACCESS:
			op->status = nfsd4_decode_access(argp, &op->u.access);
			break;
		case OP_CLOSE:
			op->status = nfsd4_decode_close(argp, &op->u.close);
			break;
		case OP_COMMIT:
			op->status = nfsd4_decode_commit(argp, &op->u.commit);
			break;
		case OP_CREATE:
			op->status = nfsd4_decode_create(argp, &op->u.create);
			break;
		case OP_DELEGRETURN:
			op->status = nfsd4_decode_delegreturn(argp, &op->u.delegreturn);
			break;
		case OP_GETATTR:
			op->status = nfsd4_decode_getattr(argp, &op->u.getattr);
			break;
		case OP_GETFH:
			op->status = nfs_ok;
			break;
		case OP_LINK:
			op->status = nfsd4_decode_link(argp, &op->u.link);
			break;
		case OP_LOCK:
			op->status = nfsd4_decode_lock(argp, &op->u.lock);
			break;
		case OP_LOCKT:
			op->status = nfsd4_decode_lockt(argp, &op->u.lockt);
			break;
		case OP_LOCKU:
			op->status = nfsd4_decode_locku(argp, &op->u.locku);
			break;
		case OP_LOOKUP:
			op->status = nfsd4_decode_lookup(argp, &op->u.lookup);
			break;
		case OP_LOOKUPP:
			op->status = nfs_ok;
			break;
		case OP_NVERIFY:
			op->status = nfsd4_decode_verify(argp, &op->u.nverify);
			break;
		case OP_OPEN:
			op->status = nfsd4_decode_open(argp, &op->u.open);
			break;
		case OP_OPEN_CONFIRM:
			op->status = nfsd4_decode_open_confirm(argp, &op->u.open_confirm);
			break;
		case OP_OPEN_DOWNGRADE:
			op->status = nfsd4_decode_open_downgrade(argp, &op->u.open_downgrade);
			break;
		case OP_PUTFH:
			op->status = nfsd4_decode_putfh(argp, &op->u.putfh);
			break;
		case OP_PUTROOTFH:
			op->status = nfs_ok;
			break;
		case OP_READ:
			op->status = nfsd4_decode_read(argp, &op->u.read);
			break;
		case OP_READDIR:
			op->status = nfsd4_decode_readdir(argp, &op->u.readdir);
			break;
		case OP_READLINK:
			op->status = nfs_ok;
			break;
		case OP_REMOVE:
			op->status = nfsd4_decode_remove(argp, &op->u.remove);
			break;
		case OP_RENAME:
			op->status = nfsd4_decode_rename(argp, &op->u.rename);
			break;
		case OP_RESTOREFH:
			op->status = nfs_ok;
			break;
		case OP_RENEW:
			op->status = nfsd4_decode_renew(argp, &op->u.renew);
			break;
		case OP_SAVEFH:
			op->status = nfs_ok;
			break;
		case OP_SECINFO:
			op->status = nfsd4_decode_secinfo(argp, &op->u.secinfo);
			break;
		case OP_SETATTR:
			op->status = nfsd4_decode_setattr(argp, &op->u.setattr);
			break;
		case OP_SETCLIENTID:
			op->status = nfsd4_decode_setclientid(argp, &op->u.setclientid);
			break;
		case OP_SETCLIENTID_CONFIRM:
			op->status = nfsd4_decode_setclientid_confirm(argp, &op->u.setclientid_confirm);
			break;
		case OP_VERIFY:
			op->status = nfsd4_decode_verify(argp, &op->u.verify);
			break;
		case OP_WRITE:
			op->status = nfsd4_decode_write(argp, &op->u.write);
			break;
		case OP_RELEASE_LOCKOWNER:
			op->status = nfsd4_decode_release_lockowner(argp, &op->u.release_lockowner);
			break;
		default:
			op->opnum = OP_ILLEGAL;
			op->status = nfserr_op_illegal;
			break;
		}

		if (op->status) {
			argp->opcnt = i+1;
			break;
		}
	}

	DECODE_TAIL;
}
/*
 * END OF "GENERIC" DECODE ROUTINES.
 */

/*
 * START OF "GENERIC" ENCODE ROUTINES.
 *   These may look a little ugly since they are imported from a "generic"
 * set of XDR encode/decode routines which are intended to be shared by
 * all of our NFSv4 implementations (OpenBSD, MacOS X...).
 *
 * If the pain of reading these is too great, it should be a straightforward
 * task to translate them into Linux-specific versions which are more
 * consistent with the style used in NFSv2/v3...
 */
#define ENCODE_HEAD              __be32 *p

#define WRITE32(n)               *p++ = htonl(n)
#define WRITE64(n)               do {				\
	*p++ = htonl((u32)((n) >> 32));				\
	*p++ = htonl((u32)(n));					\
} while (0)
#define WRITEMEM(ptr,nbytes)     do {				\
	*(p + XDR_QUADLEN(nbytes) -1) = 0;                      \
	memcpy(p, ptr, nbytes);					\
	p += XDR_QUADLEN(nbytes);				\
} while (0)
#define WRITECINFO(c)		do {				\
	*p++ = htonl(c.atomic);					\
	*p++ = htonl(c.before_ctime_sec);				\
	*p++ = htonl(c.before_ctime_nsec);				\
	*p++ = htonl(c.after_ctime_sec);				\
	*p++ = htonl(c.after_ctime_nsec);				\
} while (0)

#define RESERVE_SPACE(nbytes)	do {				\
	p = resp->p;						\
	BUG_ON(p + XDR_QUADLEN(nbytes) > resp->end);		\
} while (0)
#define ADJUST_ARGS()		resp->p = p

/*
 * Header routine to setup seqid operation replay cache
 */
#define ENCODE_SEQID_OP_HEAD					\
	__be32 *p;						\
	__be32 *save;						\
								\
	save = resp->p;

/*
 * Routine for encoding the result of a "seqid-mutating" NFSv4 operation.  This
 * is where sequence id's are incremented, and the replay cache is filled.
 * Note that we increment sequence id's here, at the last moment, so we're sure
 * we know whether the error to be returned is a sequence id mutating error.
 */

#define ENCODE_SEQID_OP_TAIL(stateowner) do {			\
	if (seqid_mutating_err(nfserr) && stateowner) { 	\
		stateowner->so_seqid++;				\
		stateowner->so_replay.rp_status = nfserr;   	\
		stateowner->so_replay.rp_buflen = 		\
			  (((char *)(resp)->p - (char *)save)); \
		memcpy(stateowner->so_replay.rp_buf, save,      \
 			stateowner->so_replay.rp_buflen); 	\
	} } while (0);

/* Encode as an array of strings the string given with components
 * seperated @sep.
 */
static __be32 nfsd4_encode_components(char sep, char *components,
				   __be32 **pp, int *buflen)
{
	__be32 *p = *pp;
	__be32 *countp = p;
	int strlen, count=0;
	char *str, *end;

	dprintk("nfsd4_encode_components(%s)\n", components);
	if ((*buflen -= 4) < 0)
		return nfserr_resource;
	WRITE32(0); /* We will fill this in with @count later */
	end = str = components;
	while (*end) {
		for (; *end && (*end != sep); end++)
			; /* Point to end of component */
		strlen = end - str;
		if (strlen) {
			if ((*buflen -= ((XDR_QUADLEN(strlen) << 2) + 4)) < 0)
				return nfserr_resource;
			WRITE32(strlen);
			WRITEMEM(str, strlen);
			count++;
		}
		else
			end++;
		str = end;
	}
	*pp = p;
	p = countp;
	WRITE32(count);
	return 0;
}

/*
 * encode a location element of a fs_locations structure
 */
static __be32 nfsd4_encode_fs_location4(struct nfsd4_fs_location *location,
				    __be32 **pp, int *buflen)
{
	__be32 status;
	__be32 *p = *pp;

	status = nfsd4_encode_components(':', location->hosts, &p, buflen);
	if (status)
		return status;
	status = nfsd4_encode_components('/', location->path, &p, buflen);
	if (status)
		return status;
	*pp = p;
	return 0;
}

/*
 * Return the path to an export point in the pseudo filesystem namespace
 * Returned string is safe to use as long as the caller holds a reference
 * to @exp.
 */
static char *nfsd4_path(struct svc_rqst *rqstp, struct svc_export *exp, __be32 *stat)
{
	struct svc_fh tmp_fh;
	char *path, *rootpath;

	fh_init(&tmp_fh, NFS4_FHSIZE);
	*stat = exp_pseudoroot(rqstp, &tmp_fh);
	if (*stat)
		return NULL;
	rootpath = tmp_fh.fh_export->ex_path;

	path = exp->ex_path;

	if (strncmp(path, rootpath, strlen(rootpath))) {
		dprintk("nfsd: fs_locations failed;"
			"%s is not contained in %s\n", path, rootpath);
		*stat = nfserr_notsupp;
		return NULL;
	}

	return path + strlen(rootpath);
}

/*
 *  encode a fs_locations structure
 */
static __be32 nfsd4_encode_fs_locations(struct svc_rqst *rqstp,
				     struct svc_export *exp,
				     __be32 **pp, int *buflen)
{
	__be32 status;
	int i;
	__be32 *p = *pp;
	struct nfsd4_fs_locations *fslocs = &exp->ex_fslocs;
	char *root = nfsd4_path(rqstp, exp, &status);

	if (status)
		return status;
	status = nfsd4_encode_components('/', root, &p, buflen);
	if (status)
		return status;
	if ((*buflen -= 4) < 0)
		return nfserr_resource;
	WRITE32(fslocs->locations_count);
	for (i=0; i<fslocs->locations_count; i++) {
		status = nfsd4_encode_fs_location4(&fslocs->locations[i],
						   &p, buflen);
		if (status)
			return status;
	}
	*pp = p;
	return 0;
}

static u32 nfs4_ftypes[16] = {
        NF4BAD,  NF4FIFO, NF4CHR, NF4BAD,
        NF4DIR,  NF4BAD,  NF4BLK, NF4BAD,
        NF4REG,  NF4BAD,  NF4LNK, NF4BAD,
        NF4SOCK, NF4BAD,  NF4LNK, NF4BAD,
};

static __be32
nfsd4_encode_name(struct svc_rqst *rqstp, int whotype, uid_t id, int group,
			__be32 **p, int *buflen)
{
	int status;

	if (*buflen < (XDR_QUADLEN(IDMAP_NAMESZ) << 2) + 4)
		return nfserr_resource;
	if (whotype != NFS4_ACL_WHO_NAMED)
		status = nfs4_acl_write_who(whotype, (u8 *)(*p + 1));
	else if (group)
		status = nfsd_map_gid_to_name(rqstp, id, (u8 *)(*p + 1));
	else
		status = nfsd_map_uid_to_name(rqstp, id, (u8 *)(*p + 1));
	if (status < 0)
		return nfserrno(status);
	*p = xdr_encode_opaque(*p, NULL, status);
	*buflen -= (XDR_QUADLEN(status) << 2) + 4;
	BUG_ON(*buflen < 0);
	return 0;
}

static inline __be32
nfsd4_encode_user(struct svc_rqst *rqstp, uid_t uid, __be32 **p, int *buflen)
{
	return nfsd4_encode_name(rqstp, NFS4_ACL_WHO_NAMED, uid, 0, p, buflen);
}

static inline __be32
nfsd4_encode_group(struct svc_rqst *rqstp, uid_t gid, __be32 **p, int *buflen)
{
	return nfsd4_encode_name(rqstp, NFS4_ACL_WHO_NAMED, gid, 1, p, buflen);
}

static inline __be32
nfsd4_encode_aclname(struct svc_rqst *rqstp, int whotype, uid_t id, int group,
		__be32 **p, int *buflen)
{
	return nfsd4_encode_name(rqstp, whotype, id, group, p, buflen);
}

#define WORD0_ABSENT_FS_ATTRS (FATTR4_WORD0_FS_LOCATIONS | FATTR4_WORD0_FSID | \
			      FATTR4_WORD0_RDATTR_ERROR)
#define WORD1_ABSENT_FS_ATTRS FATTR4_WORD1_MOUNTED_ON_FILEID

static __be32 fattr_handle_absent_fs(u32 *bmval0, u32 *bmval1, u32 *rdattr_err)
{
	/* As per referral draft:  */
	if (*bmval0 & ~WORD0_ABSENT_FS_ATTRS ||
	    *bmval1 & ~WORD1_ABSENT_FS_ATTRS) {
		if (*bmval0 & FATTR4_WORD0_RDATTR_ERROR ||
	            *bmval0 & FATTR4_WORD0_FS_LOCATIONS)
			*rdattr_err = NFSERR_MOVED;
		else
			return nfserr_moved;
	}
	*bmval0 &= WORD0_ABSENT_FS_ATTRS;
	*bmval1 &= WORD1_ABSENT_FS_ATTRS;
	return 0;
}

/*
 * Note: @fhp can be NULL; in this case, we might have to compose the filehandle
 * ourselves.
 *
 * @countp is the buffer size in _words_; upon successful return this becomes
 * replaced with the number of words written.
 */
__be32
nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp,
		struct dentry *dentry, __be32 *buffer, int *countp, u32 *bmval,
		struct svc_rqst *rqstp)
{
	u32 bmval0 = bmval[0];
	u32 bmval1 = bmval[1];
	struct kstat stat;
	struct svc_fh tempfh;
	struct kstatfs statfs;
	int buflen = *countp << 2;
	__be32 *attrlenp;
	u32 dummy;
	u64 dummy64;
	u32 rdattr_err = 0;
	__be32 *p = buffer;
	__be32 status;
	int err;
	int aclsupport = 0;
	struct nfs4_acl *acl = NULL;

	BUG_ON(bmval1 & NFSD_WRITEONLY_ATTRS_WORD1);
	BUG_ON(bmval0 & ~NFSD_SUPPORTED_ATTRS_WORD0);
	BUG_ON(bmval1 & ~NFSD_SUPPORTED_ATTRS_WORD1);

	if (exp->ex_fslocs.migrated) {
		status = fattr_handle_absent_fs(&bmval0, &bmval1, &rdattr_err);
		if (status)
			goto out;
	}

	err = vfs_getattr(exp->ex_mnt, dentry, &stat);
	if (err)
		goto out_nfserr;
	if ((bmval0 & (FATTR4_WORD0_FILES_FREE | FATTR4_WORD0_FILES_TOTAL |
			FATTR4_WORD0_MAXNAME)) ||
	    (bmval1 & (FATTR4_WORD1_SPACE_AVAIL | FATTR4_WORD1_SPACE_FREE |
		       FATTR4_WORD1_SPACE_TOTAL))) {
		err = vfs_statfs(dentry, &statfs);
		if (err)
			goto out_nfserr;
	}
	if ((bmval0 & (FATTR4_WORD0_FILEHANDLE | FATTR4_WORD0_FSID)) && !fhp) {
		fh_init(&tempfh, NFS4_FHSIZE);
		status = fh_compose(&tempfh, exp, dentry, NULL);
		if (status)
			goto out;
		fhp = &tempfh;
	}
	if (bmval0 & (FATTR4_WORD0_ACL | FATTR4_WORD0_ACLSUPPORT
			| FATTR4_WORD0_SUPPORTED_ATTRS)) {
		err = nfsd4_get_nfs4_acl(rqstp, dentry, &acl);
		aclsupport = (err == 0);
		if (bmval0 & FATTR4_WORD0_ACL) {
			if (err == -EOPNOTSUPP)
				bmval0 &= ~FATTR4_WORD0_ACL;
			else if (err == -EINVAL) {
				status = nfserr_attrnotsupp;
				goto out;
			} else if (err != 0)
				goto out_nfserr;
		}
	}
	if (bmval0 & FATTR4_WORD0_FS_LOCATIONS) {
		if (exp->ex_fslocs.locations == NULL) {
			bmval0 &= ~FATTR4_WORD0_FS_LOCATIONS;
		}
	}
	if ((buflen -= 16) < 0)
		goto out_resource;

	WRITE32(2);
	WRITE32(bmval0);
	WRITE32(bmval1);
	attrlenp = p++;                /* to be backfilled later */

	if (bmval0 & FATTR4_WORD0_SUPPORTED_ATTRS) {
		u32 word0 = NFSD_SUPPORTED_ATTRS_WORD0;
		if ((buflen -= 12) < 0)
			goto out_resource;
		if (!aclsupport)
			word0 &= ~FATTR4_WORD0_ACL;
		if (!exp->ex_fslocs.locations)
			word0 &= ~FATTR4_WORD0_FS_LOCATIONS;
		WRITE32(2);
		WRITE32(word0);
		WRITE32(NFSD_SUPPORTED_ATTRS_WORD1);
	}
	if (bmval0 & FATTR4_WORD0_TYPE) {
		if ((buflen -= 4) < 0)
			goto out_resource;
		dummy = nfs4_ftypes[(stat.mode & S_IFMT) >> 12];
		if (dummy == NF4BAD)
			goto out_serverfault;
		WRITE32(dummy);
	}
	if (bmval0 & FATTR4_WORD0_FH_EXPIRE_TYPE) {
		if ((buflen -= 4) < 0)
			goto out_resource;
		if (exp->ex_flags & NFSEXP_NOSUBTREECHECK)
			WRITE32(NFS4_FH_PERSISTENT);
		else
			WRITE32(NFS4_FH_PERSISTENT|NFS4_FH_VOL_RENAME);
	}
	if (bmval0 & FATTR4_WORD0_CHANGE) {
		/*
		 * Note: This _must_ be consistent with the scheme for writing
		 * change_info, so any changes made here must be reflected there
		 * as well.  (See xdr4.h:set_change_info() and the WRITECINFO()
		 * macro above.)
		 */
		if ((buflen -= 8) < 0)
			goto out_resource;
		WRITE32(stat.ctime.tv_sec);
		WRITE32(stat.ctime.tv_nsec);
	}