X-Git-Url: http://ftp.safe.ca/?p=safe%2Fjmp%2Flinux-2.6;a=blobdiff_plain;f=fs%2Fnfsd%2Fvfs.c;h=6ad76a4cfc012261cb1fab19581671ca4c56c2b6;hp=d5a238e6e6ed433a92b819cedee752d09394a0da;hb=d911df7b8d44de41661363a4e29ee710180ba025;hpb=f2b0dee2ec8d562ad9ced2b7481be72d356c6cfc diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index d5a238e..6ad76a4 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -83,7 +83,6 @@ struct raparm_hbucket { spinlock_t pb_lock; } ____cacheline_aligned_in_smp; -static struct raparms * raparml; #define RAPARM_HASH_BITS 4 #define RAPARM_HASH_SIZE (1<ex_flags & NFSEXP_CROSSMOUNT) || EX_NOHIDE(exp2)) { /* successfully crossed mount point */ - exp_put(exp); - *expp = exp2; + /* + * This is subtle: dentry is *not* under mnt at this point. + * The only reason we are safe is that original mnt is pinned + * down by exp, so we should dput before putting exp. + */ dput(dentry); *dpp = mounts; + exp_put(exp); + *expp = exp2; } else { exp_put(exp2); dput(mounts); @@ -144,7 +148,7 @@ nfsd_lookup_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp, dprintk("nfsd: nfsd_lookup(fh %s, %.*s)\n", SVCFH_fmt(fhp), len,name); /* Obtain dentry and export. */ - err = fh_verify(rqstp, fhp, S_IFDIR, MAY_EXEC); + err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_EXEC); if (err) return err; @@ -262,15 +266,14 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, { struct dentry *dentry; struct inode *inode; - int accmode = MAY_SATTR; + int accmode = NFSD_MAY_SATTR; int ftype = 0; - int imode; __be32 err; int host_err; int size_change = 0; if (iap->ia_valid & (ATTR_ATIME | ATTR_MTIME | ATTR_SIZE)) - accmode |= MAY_WRITE|MAY_OWNER_OVERRIDE; + accmode |= NFSD_MAY_WRITE|NFSD_MAY_OWNER_OVERRIDE; if (iap->ia_valid & ATTR_SIZE) ftype = S_IFREG; @@ -332,7 +335,8 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, */ if (iap->ia_valid & ATTR_SIZE) { if (iap->ia_size < inode->i_size) { - err = nfsd_permission(rqstp, fhp->fh_export, dentry, MAY_TRUNC|MAY_OWNER_OVERRIDE); + err = nfsd_permission(rqstp, fhp->fh_export, dentry, + NFSD_MAY_TRUNC|NFSD_MAY_OWNER_OVERRIDE); if (err) goto out; } @@ -357,28 +361,29 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, put_write_access(inode); goto out_nfserr; } - DQUOT_INIT(inode); + vfs_dq_init(inode); } - imode = inode->i_mode; + /* sanitize the mode change */ if (iap->ia_valid & ATTR_MODE) { iap->ia_mode &= S_IALLUGO; - imode = iap->ia_mode |= (imode & ~S_IALLUGO); - /* if changing uid/gid revoke setuid/setgid in mode */ - if ((iap->ia_valid & ATTR_UID) && iap->ia_uid != inode->i_uid) { - iap->ia_valid |= ATTR_KILL_PRIV; + iap->ia_mode |= (inode->i_mode & ~S_IALLUGO); + } + + /* Revoke setuid/setgid on chown */ + if (!S_ISDIR(inode->i_mode) && + (((iap->ia_valid & ATTR_UID) && iap->ia_uid != inode->i_uid) || + ((iap->ia_valid & ATTR_GID) && iap->ia_gid != inode->i_gid))) { + iap->ia_valid |= ATTR_KILL_PRIV; + if (iap->ia_valid & ATTR_MODE) { + /* we're setting mode too, just clear the s*id bits */ iap->ia_mode &= ~S_ISUID; + if (iap->ia_mode & S_IXGRP) + iap->ia_mode &= ~S_ISGID; + } else { + /* set ATTR_KILL_* bits and let VFS handle it */ + iap->ia_valid |= (ATTR_KILL_SUID | ATTR_KILL_SGID); } - if ((iap->ia_valid & ATTR_GID) && iap->ia_gid != inode->i_gid) - iap->ia_mode &= ~S_ISGID; - } else { - /* - * Revoke setuid/setgid bit on chown/chgrp - */ - if ((iap->ia_valid & ATTR_UID) && iap->ia_uid != inode->i_uid) - iap->ia_valid |= ATTR_KILL_SUID | ATTR_KILL_PRIV; - if ((iap->ia_valid & ATTR_GID) && iap->ia_gid != inode->i_gid) - iap->ia_valid |= ATTR_KILL_SGID; } /* Change the attributes. */ @@ -411,6 +416,7 @@ out_nfserr: static ssize_t nfsd_getxattr(struct dentry *dentry, char *key, void **buf) { ssize_t buflen; + ssize_t ret; buflen = vfs_getxattr(dentry, key, NULL, 0); if (buflen <= 0) @@ -420,7 +426,10 @@ static ssize_t nfsd_getxattr(struct dentry *dentry, char *key, void **buf) if (!*buf) return -ENOMEM; - return vfs_getxattr(dentry, key, *buf, buflen); + ret = vfs_getxattr(dentry, key, *buf, buflen); + if (ret < 0) + kfree(*buf); + return ret; } #endif @@ -463,7 +472,7 @@ nfsd4_set_nfs4_acl(struct svc_rqst *rqstp, struct svc_fh *fhp, unsigned int flags = 0; /* Get inode */ - error = fh_verify(rqstp, fhp, 0 /* S_IFREG */, MAY_SATTR); + error = fh_verify(rqstp, fhp, 0 /* S_IFREG */, NFSD_MAY_SATTR); if (error) return error; @@ -564,20 +573,20 @@ struct accessmap { int how; }; static struct accessmap nfs3_regaccess[] = { - { NFS3_ACCESS_READ, MAY_READ }, - { NFS3_ACCESS_EXECUTE, MAY_EXEC }, - { NFS3_ACCESS_MODIFY, MAY_WRITE|MAY_TRUNC }, - { NFS3_ACCESS_EXTEND, MAY_WRITE }, + { NFS3_ACCESS_READ, NFSD_MAY_READ }, + { NFS3_ACCESS_EXECUTE, NFSD_MAY_EXEC }, + { NFS3_ACCESS_MODIFY, NFSD_MAY_WRITE|NFSD_MAY_TRUNC }, + { NFS3_ACCESS_EXTEND, NFSD_MAY_WRITE }, { 0, 0 } }; static struct accessmap nfs3_diraccess[] = { - { NFS3_ACCESS_READ, MAY_READ }, - { NFS3_ACCESS_LOOKUP, MAY_EXEC }, - { NFS3_ACCESS_MODIFY, MAY_EXEC|MAY_WRITE|MAY_TRUNC }, - { NFS3_ACCESS_EXTEND, MAY_EXEC|MAY_WRITE }, - { NFS3_ACCESS_DELETE, MAY_REMOVE }, + { NFS3_ACCESS_READ, NFSD_MAY_READ }, + { NFS3_ACCESS_LOOKUP, NFSD_MAY_EXEC }, + { NFS3_ACCESS_MODIFY, NFSD_MAY_EXEC|NFSD_MAY_WRITE|NFSD_MAY_TRUNC}, + { NFS3_ACCESS_EXTEND, NFSD_MAY_EXEC|NFSD_MAY_WRITE }, + { NFS3_ACCESS_DELETE, NFSD_MAY_REMOVE }, { 0, 0 } }; @@ -590,10 +599,10 @@ static struct accessmap nfs3_anyaccess[] = { * mainly at mode bits, and we make sure to ignore read-only * filesystem checks */ - { NFS3_ACCESS_READ, MAY_READ }, - { NFS3_ACCESS_EXECUTE, MAY_EXEC }, - { NFS3_ACCESS_MODIFY, MAY_WRITE|MAY_LOCAL_ACCESS }, - { NFS3_ACCESS_EXTEND, MAY_WRITE|MAY_LOCAL_ACCESS }, + { NFS3_ACCESS_READ, NFSD_MAY_READ }, + { NFS3_ACCESS_EXECUTE, NFSD_MAY_EXEC }, + { NFS3_ACCESS_MODIFY, NFSD_MAY_WRITE|NFSD_MAY_LOCAL_ACCESS }, + { NFS3_ACCESS_EXTEND, NFSD_MAY_WRITE|NFSD_MAY_LOCAL_ACCESS }, { 0, 0 } }; @@ -607,7 +616,7 @@ nfsd_access(struct svc_rqst *rqstp, struct svc_fh *fhp, u32 *access, u32 *suppor u32 query, result = 0, sresult = 0; __be32 error; - error = fh_verify(rqstp, fhp, 0, MAY_NOP); + error = fh_verify(rqstp, fhp, 0, NFSD_MAY_NOP); if (error) goto out; @@ -668,6 +677,7 @@ __be32 nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access, struct file **filp) { + const struct cred *cred = current_cred(); struct dentry *dentry; struct inode *inode; int flags = O_RDONLY|O_LARGEFILE; @@ -679,7 +689,7 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, * and (hopefully) checked permission - so allow OWNER_OVERRIDE * in case a chmod has now revoked permission. */ - err = fh_verify(rqstp, fhp, type, access | MAY_OWNER_OVERRIDE); + err = fh_verify(rqstp, fhp, type, access | NFSD_MAY_OWNER_OVERRIDE); if (err) goto out; @@ -690,7 +700,7 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, * or any access when mandatory locking enabled */ err = nfserr_perm; - if (IS_APPEND(inode) && (access & MAY_WRITE)) + if (IS_APPEND(inode) && (access & NFSD_MAY_WRITE)) goto out; /* * We must ignore files (but only files) which might have mandatory @@ -707,22 +717,22 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, * Check to see if there are any leases on this file. * This may block while leases are broken. */ - host_err = break_lease(inode, O_NONBLOCK | ((access & MAY_WRITE) ? FMODE_WRITE : 0)); + host_err = break_lease(inode, O_NONBLOCK | ((access & NFSD_MAY_WRITE) ? FMODE_WRITE : 0)); if (host_err == -EWOULDBLOCK) host_err = -ETIMEDOUT; if (host_err) /* NOMEM or WOULDBLOCK */ goto out_nfserr; - if (access & MAY_WRITE) { - if (access & MAY_READ) + if (access & NFSD_MAY_WRITE) { + if (access & NFSD_MAY_READ) flags = O_RDWR|O_LARGEFILE; else flags = O_WRONLY|O_LARGEFILE; - DQUOT_INIT(inode); + vfs_dq_init(inode); } *filp = dentry_open(dget(dentry), mntget(fhp->fh_export->ex_path.mnt), - flags); + flags, cred); if (IS_ERR(*filp)) host_err = PTR_ERR(*filp); out_nfserr: @@ -760,7 +770,6 @@ static inline int nfsd_dosync(struct file *filp, struct dentry *dp, return err; } - static int nfsd_sync(struct file *filp) @@ -954,10 +963,47 @@ static void kill_suid(struct dentry *dentry) mutex_unlock(&dentry->d_inode->i_mutex); } +/* + * Gathered writes: If another process is currently writing to the file, + * there's a high chance this is another nfsd (triggered by a bulk write + * from a client's biod). Rather than syncing the file with each write + * request, we sleep for 10 msec. + * + * I don't know if this roughly approximates C. Juszak's idea of + * gathered writes, but it's a nice and simple solution (IMHO), and it + * seems to work:-) + * + * Note: we do this only in the NFSv2 case, since v3 and higher have a + * better tool (separate unstable writes and commits) for solving this + * problem. + */ +static int wait_for_concurrent_writes(struct file *file) +{ + struct inode *inode = file->f_path.dentry->d_inode; + static ino_t last_ino; + static dev_t last_dev; + int err = 0; + + if (atomic_read(&inode->i_writecount) > 1 + || (last_ino == inode->i_ino && last_dev == inode->i_sb->s_dev)) { + dprintk("nfsd: write defer %d\n", task_pid_nr(current)); + msleep(10); + dprintk("nfsd: write resume %d\n", task_pid_nr(current)); + } + + if (inode->i_state & I_DIRTY) { + dprintk("nfsd: write sync %d\n", task_pid_nr(current)); + err = nfsd_sync(file); + } + last_ino = inode->i_ino; + last_dev = inode->i_sb->s_dev; + return err; +} + static __be32 nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, loff_t offset, struct kvec *vec, int vlen, - unsigned long cnt, int *stablep) + unsigned long *cnt, int *stablep) { struct svc_export *exp; struct dentry *dentry; @@ -966,12 +1012,13 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, __be32 err = 0; int host_err; int stable = *stablep; + int use_wgather; #ifdef MSNFS err = nfserr_perm; if ((fhp->fh_export->ex_flags & NFSEXP_MSNFS) && - (!lock_may_write(file->f_path.dentry->d_inode, offset, cnt))) + (!lock_may_write(file->f_path.dentry->d_inode, offset, *cnt))) goto out; #endif @@ -984,26 +1031,31 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, * - the sync export option has been set, or * - the client requested O_SYNC behavior (NFSv3 feature). * - The file system doesn't support fsync(). - * When gathered writes have been configured for this volume, + * When NFSv2 gathered writes have been configured for this volume, * flushing the data to disk is handled separately below. */ + use_wgather = (rqstp->rq_vers == 2) && EX_WGATHER(exp); - if (file->f_op->fsync == 0) {/* COMMIT3 cannot work */ + if (!file->f_op->fsync) {/* COMMIT3 cannot work */ stable = 2; *stablep = 2; /* FILE_SYNC */ } if (!EX_ISSYNC(exp)) stable = 0; - if (stable && !EX_WGATHER(exp)) + if (stable && !use_wgather) { + spin_lock(&file->f_lock); file->f_flags |= O_SYNC; + spin_unlock(&file->f_lock); + } /* Write the data. */ oldfs = get_fs(); set_fs(KERNEL_DS); host_err = vfs_writev(file, (struct iovec __user *)vec, vlen, &offset); set_fs(oldfs); if (host_err >= 0) { - nfsdstats.io_write += cnt; + *cnt = host_err; + nfsdstats.io_write += host_err; fsnotify_modify(file->f_path.dentry); } @@ -1011,46 +1063,13 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, if (host_err >= 0 && (inode->i_mode & (S_ISUID | S_ISGID))) kill_suid(dentry); - if (host_err >= 0 && stable) { - static ino_t last_ino; - static dev_t last_dev; - - /* - * Gathered writes: If another process is currently - * writing to the file, there's a high chance - * this is another nfsd (triggered by a bulk write - * from a client's biod). Rather than syncing the - * file with each write request, we sleep for 10 msec. - * - * I don't know if this roughly approximates - * C. Juszak's idea of gathered writes, but it's a - * nice and simple solution (IMHO), and it seems to - * work:-) - */ - if (EX_WGATHER(exp)) { - if (atomic_read(&inode->i_writecount) > 1 - || (last_ino == inode->i_ino && last_dev == inode->i_sb->s_dev)) { - dprintk("nfsd: write defer %d\n", task_pid_nr(current)); - msleep(10); - dprintk("nfsd: write resume %d\n", task_pid_nr(current)); - } - - if (inode->i_state & I_DIRTY) { - dprintk("nfsd: write sync %d\n", task_pid_nr(current)); - host_err=nfsd_sync(file); - } -#if 0 - wake_up(&inode->i_wait); -#endif - } - last_ino = inode->i_ino; - last_dev = inode->i_sb->s_dev; - } + if (host_err >= 0 && stable && use_wgather) + host_err = wait_for_concurrent_writes(file); dprintk("nfsd: write complete host_err=%d\n", host_err); if (host_err >= 0) err = 0; - else + else err = nfserrno(host_err); out: return err; @@ -1070,12 +1089,12 @@ nfsd_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, if (file) { err = nfsd_permission(rqstp, fhp->fh_export, fhp->fh_dentry, - MAY_READ|MAY_OWNER_OVERRIDE); + NFSD_MAY_READ|NFSD_MAY_OWNER_OVERRIDE); if (err) goto out; err = nfsd_vfs_read(rqstp, fhp, file, offset, vec, vlen, count); } else { - err = nfsd_open(rqstp, fhp, S_IFREG, MAY_READ, &file); + err = nfsd_open(rqstp, fhp, S_IFREG, NFSD_MAY_READ, &file); if (err) goto out; err = nfsd_vfs_read(rqstp, fhp, file, offset, vec, vlen, count); @@ -1092,20 +1111,20 @@ out: */ __be32 nfsd_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, - loff_t offset, struct kvec *vec, int vlen, unsigned long cnt, + loff_t offset, struct kvec *vec, int vlen, unsigned long *cnt, int *stablep) { __be32 err = 0; if (file) { err = nfsd_permission(rqstp, fhp->fh_export, fhp->fh_dentry, - MAY_WRITE|MAY_OWNER_OVERRIDE); + NFSD_MAY_WRITE|NFSD_MAY_OWNER_OVERRIDE); if (err) goto out; err = nfsd_vfs_write(rqstp, fhp, file, offset, vec, vlen, cnt, stablep); } else { - err = nfsd_open(rqstp, fhp, S_IFREG, MAY_WRITE, &file); + err = nfsd_open(rqstp, fhp, S_IFREG, NFSD_MAY_WRITE, &file); if (err) goto out; @@ -1137,7 +1156,8 @@ nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp, if ((u64)count > ~(u64)offset) return nfserr_inval; - if ((err = nfsd_open(rqstp, fhp, S_IFREG, MAY_WRITE, &file)) != 0) + err = nfsd_open(rqstp, fhp, S_IFREG, NFSD_MAY_WRITE, &file); + if (err) return err; if (EX_ISSYNC(fhp->fh_export)) { if (file->f_op && file->f_op->fsync) { @@ -1165,13 +1185,28 @@ nfsd_create_setattr(struct svc_rqst *rqstp, struct svc_fh *resfhp, * send along the gid on create when it tries to implement * setgid directories via NFS: */ - if (current->fsuid != 0) + if (current_fsuid() != 0) iap->ia_valid &= ~(ATTR_UID|ATTR_GID); if (iap->ia_valid) return nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); return 0; } +/* HPUX client sometimes creates a file in mode 000, and sets size to 0. + * setting size to 0 may fail for some specific file systems by the permission + * checking which requires WRITE permission but the mode is 000. + * we ignore the resizing(to 0) on the just new created file, since the size is + * 0 after file created. + * + * call this only after vfs_create() is called. + * */ +static void +nfsd_check_ignore_resizing(struct iattr *iap) +{ + if ((iap->ia_valid & ATTR_SIZE) && (iap->ia_size == 0)) + iap->ia_valid &= ~ATTR_SIZE; +} + /* * Create a file (regular, directory, device, fifo); UNIX sockets * not yet implemented. @@ -1198,7 +1233,7 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, if (isdotent(fname, flen)) goto out; - err = fh_verify(rqstp, fhp, S_IFDIR, MAY_CREATE); + err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_CREATE); if (err) goto out; @@ -1206,7 +1241,7 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, dirp = dentry->d_inode; err = nfserr_notdir; - if(!dirp->i_op || !dirp->i_op->lookup) + if (!dirp->i_op->lookup) goto out; /* * Check whether the response file handle has been verified yet. @@ -1249,36 +1284,36 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, iap->ia_mode = 0; iap->ia_mode = (iap->ia_mode & S_IALLUGO) | type; + err = nfserr_inval; + if (!S_ISREG(type) && !S_ISDIR(type) && !special_file(type)) { + printk(KERN_WARNING "nfsd: bad file type %o in nfsd_create\n", + type); + goto out; + } + + host_err = mnt_want_write(fhp->fh_export->ex_path.mnt); + if (host_err) + goto out_nfserr; + /* * Get the dir op function pointer. */ err = 0; switch (type) { case S_IFREG: - host_err = mnt_want_write(fhp->fh_export->ex_path.mnt); - if (host_err) - goto out_nfserr; host_err = vfs_create(dirp, dchild, iap->ia_mode, NULL); + if (!host_err) + nfsd_check_ignore_resizing(iap); break; case S_IFDIR: - host_err = mnt_want_write(fhp->fh_export->ex_path.mnt); - if (host_err) - goto out_nfserr; host_err = vfs_mkdir(dirp, dchild, iap->ia_mode); break; case S_IFCHR: case S_IFBLK: case S_IFIFO: case S_IFSOCK: - host_err = mnt_want_write(fhp->fh_export->ex_path.mnt); - if (host_err) - goto out_nfserr; host_err = vfs_mknod(dirp, dchild, iap->ia_mode, rdev); break; - default: - printk("nfsd: bad file type %o in nfsd_create\n", type); - host_err = -EINVAL; - goto out_nfserr; } if (host_err < 0) { mnt_drop_write(fhp->fh_export->ex_path.mnt); @@ -1290,7 +1325,6 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, write_inode_now(dchild->d_inode, 1); } - err2 = nfsd_create_setattr(rqstp, resfhp, iap); if (err2) err = err2; @@ -1335,7 +1369,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, goto out; if (!(iap->ia_valid & ATTR_MODE)) iap->ia_mode = 0; - err = fh_verify(rqstp, fhp, S_IFDIR, MAY_CREATE); + err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_CREATE); if (err) goto out; @@ -1345,7 +1379,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, /* Get all the sanity checks out of the way before * we lock the parent. */ err = nfserr_notdir; - if(!dirp->i_op || !dirp->i_op->lookup) + if (!dirp->i_op->lookup) goto out; fh_lock_nested(fhp, I_MUTEX_PARENT); @@ -1423,6 +1457,8 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, /* setattr will sync the child (or not) */ } + nfsd_check_ignore_resizing(iap); + if (createmode == NFS3_CREATE_EXCLUSIVE) { /* Cram the verifier into atime/mtime */ iap->ia_valid = ATTR_MTIME|ATTR_ATIME @@ -1472,7 +1508,7 @@ nfsd_readlink(struct svc_rqst *rqstp, struct svc_fh *fhp, char *buf, int *lenp) __be32 err; int host_err; - err = fh_verify(rqstp, fhp, S_IFLNK, MAY_NOP); + err = fh_verify(rqstp, fhp, S_IFLNK, NFSD_MAY_NOP); if (err) goto out; @@ -1480,7 +1516,7 @@ nfsd_readlink(struct svc_rqst *rqstp, struct svc_fh *fhp, char *buf, int *lenp) inode = dentry->d_inode; err = nfserr_inval; - if (!inode->i_op || !inode->i_op->readlink) + if (!inode->i_op->readlink) goto out; touch_atime(fhp->fh_export->ex_path.mnt, dentry); @@ -1518,7 +1554,6 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, struct dentry *dentry, *dnew; __be32 err, cerr; int host_err; - umode_t mode; err = nfserr_noent; if (!flen || !plen) @@ -1527,7 +1562,7 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, if (isdotent(fname, flen)) goto out; - err = fh_verify(rqstp, fhp, S_IFDIR, MAY_CREATE); + err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_CREATE); if (err) goto out; fh_lock(fhp); @@ -1537,11 +1572,6 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, if (IS_ERR(dnew)) goto out_nfserr; - mode = S_IALLUGO; - /* Only the MODE ATTRibute is even vaguely meaningful */ - if (iap && (iap->ia_valid & ATTR_MODE)) - mode = iap->ia_mode & S_IALLUGO; - host_err = mnt_want_write(fhp->fh_export->ex_path.mnt); if (host_err) goto out_nfserr; @@ -1553,11 +1583,11 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, else { strncpy(path_alloced, path, plen); path_alloced[plen] = 0; - host_err = vfs_symlink(dentry->d_inode, dnew, path_alloced, mode); + host_err = vfs_symlink(dentry->d_inode, dnew, path_alloced); kfree(path_alloced); } } else - host_err = vfs_symlink(dentry->d_inode, dnew, path, mode); + host_err = vfs_symlink(dentry->d_inode, dnew, path); if (!host_err) { if (EX_ISSYNC(fhp->fh_export)) @@ -1592,10 +1622,10 @@ nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp, __be32 err; int host_err; - err = fh_verify(rqstp, ffhp, S_IFDIR, MAY_CREATE); + err = fh_verify(rqstp, ffhp, S_IFDIR, NFSD_MAY_CREATE); if (err) goto out; - err = fh_verify(rqstp, tfhp, -S_IFDIR, MAY_NOP); + err = fh_verify(rqstp, tfhp, -S_IFDIR, NFSD_MAY_NOP); if (err) goto out; @@ -1662,10 +1692,10 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, __be32 err; int host_err; - err = fh_verify(rqstp, ffhp, S_IFDIR, MAY_REMOVE); + err = fh_verify(rqstp, ffhp, S_IFDIR, NFSD_MAY_REMOVE); if (err) goto out; - err = fh_verify(rqstp, tfhp, S_IFDIR, MAY_CREATE); + err = fh_verify(rqstp, tfhp, S_IFDIR, NFSD_MAY_CREATE); if (err) goto out; @@ -1769,7 +1799,7 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, err = nfserr_acces; if (!flen || isdotent(fname, flen)) goto out; - err = fh_verify(rqstp, fhp, S_IFDIR, MAY_REMOVE); + err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_REMOVE); if (err) goto out; @@ -1823,6 +1853,126 @@ out: } /* + * We do this buffering because we must not call back into the file + * system's ->lookup() method from the filldir callback. That may well + * deadlock a number of file systems. + * + * This is based heavily on the implementation of same in XFS. + */ +struct buffered_dirent { + u64 ino; + loff_t offset; + int namlen; + unsigned int d_type; + char name[]; +}; + +struct readdir_data { + char *dirent; + size_t used; + int full; +}; + +static int nfsd_buffered_filldir(void *__buf, const char *name, int namlen, + loff_t offset, u64 ino, unsigned int d_type) +{ + struct readdir_data *buf = __buf; + struct buffered_dirent *de = (void *)(buf->dirent + buf->used); + unsigned int reclen; + + reclen = ALIGN(sizeof(struct buffered_dirent) + namlen, sizeof(u64)); + if (buf->used + reclen > PAGE_SIZE) { + buf->full = 1; + return -EINVAL; + } + + de->namlen = namlen; + de->offset = offset; + de->ino = ino; + de->d_type = d_type; + memcpy(de->name, name, namlen); + buf->used += reclen; + + return 0; +} + +static __be32 nfsd_buffered_readdir(struct file *file, filldir_t func, + struct readdir_cd *cdp, loff_t *offsetp) +{ + struct readdir_data buf; + struct buffered_dirent *de; + int host_err; + int size; + loff_t offset; + + buf.dirent = (void *)__get_free_page(GFP_KERNEL); + if (!buf.dirent) + return nfserrno(-ENOMEM); + + offset = *offsetp; + + while (1) { + struct inode *dir_inode = file->f_path.dentry->d_inode; + unsigned int reclen; + + cdp->err = nfserr_eof; /* will be cleared on successful read */ + buf.used = 0; + buf.full = 0; + + host_err = vfs_readdir(file, nfsd_buffered_filldir, &buf); + if (buf.full) + host_err = 0; + + if (host_err < 0) + break; + + size = buf.used; + + if (!size) + break; + + /* + * Various filldir functions may end up calling back into + * lookup_one_len() and the file system's ->lookup() method. + * These expect i_mutex to be held, as it would within readdir. + */ + host_err = mutex_lock_killable(&dir_inode->i_mutex); + if (host_err) + break; + + de = (struct buffered_dirent *)buf.dirent; + while (size > 0) { + offset = de->offset; + + if (func(cdp, de->name, de->namlen, de->offset, + de->ino, de->d_type)) + break; + + if (cdp->err != nfs_ok) + break; + + reclen = ALIGN(sizeof(*de) + de->namlen, + sizeof(u64)); + size -= reclen; + de = (struct buffered_dirent *)((char *)de + reclen); + } + mutex_unlock(&dir_inode->i_mutex); + if (size > 0) /* We bailed out early */ + break; + + offset = vfs_llseek(file, 0, SEEK_CUR); + } + + free_page((unsigned long)(buf.dirent)); + + if (host_err) + return nfserrno(host_err); + + *offsetp = offset; + return cdp->err; +} + +/* * Read entries from a directory. * The NFSv3/4 verifier we ignore for now. */ @@ -1831,11 +1981,10 @@ nfsd_readdir(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t *offsetp, struct readdir_cd *cdp, filldir_t func) { __be32 err; - int host_err; struct file *file; loff_t offset = *offsetp; - err = nfsd_open(rqstp, fhp, S_IFDIR, MAY_READ, &file); + err = nfsd_open(rqstp, fhp, S_IFDIR, NFSD_MAY_READ, &file); if (err) goto out; @@ -1845,21 +1994,7 @@ nfsd_readdir(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t *offsetp, goto out_close; } - /* - * Read the directory entries. This silly loop is necessary because - * readdir() is not guaranteed to fill up the entire buffer, but - * may choose to do less. - */ - - do { - cdp->err = nfserr_eof; /* will be cleared on successful read */ - host_err = vfs_readdir(file, func, cdp); - } while (host_err >=0 && cdp->err == nfs_ok); - if (host_err) - err = nfserrno(host_err); - else - err = cdp->err; - *offsetp = vfs_llseek(file, 0, 1); + err = nfsd_buffered_readdir(file, func, cdp, offsetp); if (err == nfserr_eof || err == nfserr_toosmall) err = nfs_ok; /* can still be found in ->err */ @@ -1874,9 +2009,9 @@ out: * N.B. After this call fhp needs an fh_put */ __be32 -nfsd_statfs(struct svc_rqst *rqstp, struct svc_fh *fhp, struct kstatfs *stat) +nfsd_statfs(struct svc_rqst *rqstp, struct svc_fh *fhp, struct kstatfs *stat, int access) { - __be32 err = fh_verify(rqstp, fhp, 0, MAY_NOP); + __be32 err = fh_verify(rqstp, fhp, 0, NFSD_MAY_NOP | access); if (!err && vfs_statfs(fhp->fh_dentry,stat)) err = nfserr_io; return err; @@ -1897,42 +2032,42 @@ nfsd_permission(struct svc_rqst *rqstp, struct svc_export *exp, struct inode *inode = dentry->d_inode; int err; - if (acc == MAY_NOP) + if (acc == NFSD_MAY_NOP) return 0; #if 0 dprintk("nfsd: permission 0x%x%s%s%s%s%s%s%s mode 0%o%s%s%s\n", acc, - (acc & MAY_READ)? " read" : "", - (acc & MAY_WRITE)? " write" : "", - (acc & MAY_EXEC)? " exec" : "", - (acc & MAY_SATTR)? " sattr" : "", - (acc & MAY_TRUNC)? " trunc" : "", - (acc & MAY_LOCK)? " lock" : "", - (acc & MAY_OWNER_OVERRIDE)? " owneroverride" : "", + (acc & NFSD_MAY_READ)? " read" : "", + (acc & NFSD_MAY_WRITE)? " write" : "", + (acc & NFSD_MAY_EXEC)? " exec" : "", + (acc & NFSD_MAY_SATTR)? " sattr" : "", + (acc & NFSD_MAY_TRUNC)? " trunc" : "", + (acc & NFSD_MAY_LOCK)? " lock" : "", + (acc & NFSD_MAY_OWNER_OVERRIDE)? " owneroverride" : "", inode->i_mode, IS_IMMUTABLE(inode)? " immut" : "", IS_APPEND(inode)? " append" : "", __mnt_is_readonly(exp->ex_path.mnt)? " ro" : ""); dprintk(" owner %d/%d user %d/%d\n", - inode->i_uid, inode->i_gid, current->fsuid, current->fsgid); + inode->i_uid, inode->i_gid, current_fsuid(), current_fsgid()); #endif /* Normally we reject any write/sattr etc access on a read-only file * system. But if it is IRIX doing check on write-access for a * device special file, we ignore rofs. */ - if (!(acc & MAY_LOCAL_ACCESS)) - if (acc & (MAY_WRITE | MAY_SATTR | MAY_TRUNC)) { + if (!(acc & NFSD_MAY_LOCAL_ACCESS)) + if (acc & (NFSD_MAY_WRITE | NFSD_MAY_SATTR | NFSD_MAY_TRUNC)) { if (exp_rdonly(rqstp, exp) || __mnt_is_readonly(exp->ex_path.mnt)) return nfserr_rofs; - if (/* (acc & MAY_WRITE) && */ IS_IMMUTABLE(inode)) + if (/* (acc & NFSD_MAY_WRITE) && */ IS_IMMUTABLE(inode)) return nfserr_perm; } - if ((acc & MAY_TRUNC) && IS_APPEND(inode)) + if ((acc & NFSD_MAY_TRUNC) && IS_APPEND(inode)) return nfserr_perm; - if (acc & MAY_LOCK) { + if (acc & NFSD_MAY_LOCK) { /* If we cannot rely on authentication in NLM requests, * just allow locks, otherwise require read permission, or * ownership @@ -1940,7 +2075,7 @@ nfsd_permission(struct svc_rqst *rqstp, struct svc_export *exp, if (exp->ex_flags & NFSEXP_NOAUTHNLM) return 0; else - acc = MAY_READ | MAY_OWNER_OVERRIDE; + acc = NFSD_MAY_READ | NFSD_MAY_OWNER_OVERRIDE; } /* * The file owner always gets access permission for accesses that @@ -1956,16 +2091,17 @@ nfsd_permission(struct svc_rqst *rqstp, struct svc_export *exp, * We must trust the client to do permission checking - using "ACCESS" * with NFSv3. */ - if ((acc & MAY_OWNER_OVERRIDE) && - inode->i_uid == current->fsuid) + if ((acc & NFSD_MAY_OWNER_OVERRIDE) && + inode->i_uid == current_fsuid()) return 0; - err = permission(inode, acc & (MAY_READ|MAY_WRITE|MAY_EXEC), NULL); + /* This assumes NFSD_MAY_{READ,WRITE,EXEC} == MAY_{READ,WRITE,EXEC} */ + err = inode_permission(inode, acc & (MAY_READ|MAY_WRITE|MAY_EXEC)); /* Allow read access to binaries even when mode 111 */ if (err == -EACCES && S_ISREG(inode->i_mode) && - acc == (MAY_READ | MAY_OWNER_OVERRIDE)) - err = permission(inode, MAY_EXEC, NULL); + acc == (NFSD_MAY_READ | NFSD_MAY_OWNER_OVERRIDE)) + err = inode_permission(inode, MAY_EXEC); return err? nfserrno(err) : 0; } @@ -1973,11 +2109,20 @@ nfsd_permission(struct svc_rqst *rqstp, struct svc_export *exp, void nfsd_racache_shutdown(void) { - if (!raparml) - return; + struct raparms *raparm, *last_raparm; + unsigned int i; + dprintk("nfsd: freeing readahead buffers.\n"); - kfree(raparml); - raparml = NULL; + + for (i = 0; i < RAPARM_HASH_SIZE; i++) { + raparm = raparm_hash[i].pb_head; + while(raparm) { + last_raparm = raparm; + raparm = raparm->p_next; + kfree(last_raparm); + } + raparm_hash[i].pb_head = NULL; + } } /* * Initialize readahead param cache @@ -1988,35 +2133,38 @@ nfsd_racache_init(int cache_size) int i; int j = 0; int nperbucket; + struct raparms **raparm = NULL; - if (raparml) + if (raparm_hash[0].pb_head) return 0; - if (cache_size < 2*RAPARM_HASH_SIZE) - cache_size = 2*RAPARM_HASH_SIZE; - raparml = kcalloc(cache_size, sizeof(struct raparms), GFP_KERNEL); - - if (!raparml) { - printk(KERN_WARNING - "nfsd: Could not allocate memory read-ahead cache.\n"); - return -ENOMEM; - } + nperbucket = DIV_ROUND_UP(cache_size, RAPARM_HASH_SIZE); + if (nperbucket < 2) + nperbucket = 2; + cache_size = nperbucket * RAPARM_HASH_SIZE; dprintk("nfsd: allocating %d readahead buffers.\n", cache_size); - for (i = 0 ; i < RAPARM_HASH_SIZE ; i++) { - raparm_hash[i].pb_head = NULL; + + for (i = 0; i < RAPARM_HASH_SIZE; i++) { spin_lock_init(&raparm_hash[i].pb_lock); - } - nperbucket = DIV_ROUND_UP(cache_size, RAPARM_HASH_SIZE); - for (i = 0; i < cache_size - 1; i++) { - if (i % nperbucket == 0) - raparm_hash[j++].pb_head = raparml + i; - if (i % nperbucket < nperbucket-1) - raparml[i].p_next = raparml + i + 1; + + raparm = &raparm_hash[i].pb_head; + for (j = 0; j < nperbucket; j++) { + *raparm = kzalloc(sizeof(struct raparms), GFP_KERNEL); + if (!*raparm) + goto out_nomem; + raparm = &(*raparm)->p_next; + } + *raparm = NULL; } nfsdstats.ra_size = cache_size; return 0; + +out_nomem: + dprintk("nfsd: kmalloc failed, freeing readahead buffers\n"); + nfsd_racache_shutdown(); + return -ENOMEM; } #if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) @@ -2061,7 +2209,7 @@ nfsd_set_posix_acl(struct svc_fh *fhp, int type, struct posix_acl *acl) size_t size; int error; - if (!IS_POSIXACL(inode) || !inode->i_op || + if (!IS_POSIXACL(inode) || !inode->i_op->setxattr || !inode->i_op->removexattr) return -EOPNOTSUPP; switch(type) {