diff options
Diffstat (limited to 'fs')
68 files changed, 1997 insertions, 957 deletions
diff --git a/fs/ceph/Kconfig b/fs/ceph/Kconfig index 7f7d92d6b024..cf235f6eacf9 100644 --- a/fs/ceph/Kconfig +++ b/fs/ceph/Kconfig @@ -36,3 +36,15 @@ config CEPH_FS_POSIX_ACL groups beyond the owner/group/world scheme. If you don't know what Access Control Lists are, say N + +config CEPH_FS_SECURITY_LABEL + bool "CephFS Security Labels" + depends on CEPH_FS && SECURITY + help + Security labels support alternative access control models + implemented by security modules like SELinux. This option + enables an extended attribute handler for file security + labels in the Ceph filesystem. + + If you are not using a security module that requires using + extended attributes for file security labels, say N. diff --git a/fs/ceph/acl.c b/fs/ceph/acl.c index 8a19c249036c..aa55f412a6e3 100644 --- a/fs/ceph/acl.c +++ b/fs/ceph/acl.c @@ -159,7 +159,7 @@ out: } int ceph_pre_init_acls(struct inode *dir, umode_t *mode, - struct ceph_acls_info *info) + struct ceph_acl_sec_ctx *as_ctx) { struct posix_acl *acl, *default_acl; size_t val_size1 = 0, val_size2 = 0; @@ -234,9 +234,9 @@ int ceph_pre_init_acls(struct inode *dir, umode_t *mode, kfree(tmp_buf); - info->acl = acl; - info->default_acl = default_acl; - info->pagelist = pagelist; + as_ctx->acl = acl; + as_ctx->default_acl = default_acl; + as_ctx->pagelist = pagelist; return 0; out_err: @@ -248,18 +248,10 @@ out_err: return err; } -void ceph_init_inode_acls(struct inode* inode, struct ceph_acls_info *info) +void ceph_init_inode_acls(struct inode *inode, struct ceph_acl_sec_ctx *as_ctx) { if (!inode) return; - ceph_set_cached_acl(inode, ACL_TYPE_ACCESS, info->acl); - ceph_set_cached_acl(inode, ACL_TYPE_DEFAULT, info->default_acl); -} - -void ceph_release_acls_info(struct ceph_acls_info *info) -{ - posix_acl_release(info->acl); - posix_acl_release(info->default_acl); - if (info->pagelist) - ceph_pagelist_release(info->pagelist); + ceph_set_cached_acl(inode, ACL_TYPE_ACCESS, as_ctx->acl); + ceph_set_cached_acl(inode, ACL_TYPE_DEFAULT, as_ctx->default_acl); } diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index a47c541f8006..e078cc55b989 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -10,6 +10,7 @@ #include <linux/pagevec.h> #include <linux/task_io_accounting_ops.h> #include <linux/signal.h> +#include <linux/iversion.h> #include "super.h" #include "mds_client.h" @@ -1576,6 +1577,7 @@ static vm_fault_t ceph_page_mkwrite(struct vm_fault *vmf) /* Update time before taking page lock */ file_update_time(vma->vm_file); + inode_inc_iversion_raw(inode); do { lock_page(page); diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index 0176241eaea7..d98dcd976c80 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -8,6 +8,7 @@ #include <linux/vmalloc.h> #include <linux/wait.h> #include <linux/writeback.h> +#include <linux/iversion.h> #include "super.h" #include "mds_client.h" @@ -1138,8 +1139,9 @@ struct cap_msg_args { u64 ino, cid, follows; u64 flush_tid, oldest_flush_tid, size, max_size; u64 xattr_version; + u64 change_attr; struct ceph_buffer *xattr_buf; - struct timespec64 atime, mtime, ctime; + struct timespec64 atime, mtime, ctime, btime; int op, caps, wanted, dirty; u32 seq, issue_seq, mseq, time_warp_seq; u32 flags; @@ -1160,7 +1162,6 @@ static int send_cap_msg(struct cap_msg_args *arg) struct ceph_msg *msg; void *p; size_t extra_len; - struct timespec64 zerotime = {0}; struct ceph_osd_client *osdc = &arg->session->s_mdsc->fsc->client->osdc; dout("send_cap_msg %s %llx %llx caps %s wanted %s dirty %s" @@ -1245,15 +1246,10 @@ static int send_cap_msg(struct cap_msg_args *arg) /* pool namespace (version 8) (mds always ignores this) */ ceph_encode_32(&p, 0); - /* - * btime and change_attr (version 9) - * - * We just zero these out for now, as the MDS ignores them unless - * the requisite feature flags are set (which we don't do yet). - */ - ceph_encode_timespec64(p, &zerotime); + /* btime and change_attr (version 9) */ + ceph_encode_timespec64(p, &arg->btime); p += sizeof(struct ceph_timespec); - ceph_encode_64(&p, 0); + ceph_encode_64(&p, arg->change_attr); /* Advisory flags (version 10) */ ceph_encode_32(&p, arg->flags); @@ -1263,20 +1259,22 @@ static int send_cap_msg(struct cap_msg_args *arg) } /* - * Queue cap releases when an inode is dropped from our cache. Since - * inode is about to be destroyed, there is no need for i_ceph_lock. + * Queue cap releases when an inode is dropped from our cache. */ -void __ceph_remove_caps(struct inode *inode) +void __ceph_remove_caps(struct ceph_inode_info *ci) { - struct ceph_inode_info *ci = ceph_inode(inode); struct rb_node *p; + /* lock i_ceph_lock, because ceph_d_revalidate(..., LOOKUP_RCU) + * may call __ceph_caps_issued_mask() on a freeing inode. */ + spin_lock(&ci->i_ceph_lock); p = rb_first(&ci->i_caps); while (p) { struct ceph_cap *cap = rb_entry(p, struct ceph_cap, ci_node); p = rb_next(p); __ceph_remove_cap(cap, true); } + spin_unlock(&ci->i_ceph_lock); } /* @@ -1297,7 +1295,7 @@ void __ceph_remove_caps(struct inode *inode) * caller should hold snap_rwsem (read), s_mutex. */ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap, - int op, bool sync, int used, int want, int retain, + int op, int flags, int used, int want, int retain, int flushing, u64 flush_tid, u64 oldest_flush_tid) __releases(cap->ci->i_ceph_lock) { @@ -1377,6 +1375,8 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap, arg.mtime = inode->i_mtime; arg.atime = inode->i_atime; arg.ctime = inode->i_ctime; + arg.btime = ci->i_btime; + arg.change_attr = inode_peek_iversion_raw(inode); arg.op = op; arg.caps = cap->implemented; @@ -1393,12 +1393,19 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap, arg.mode = inode->i_mode; arg.inline_data = ci->i_inline_version != CEPH_INLINE_NONE; - if (list_empty(&ci->i_cap_snaps)) - arg.flags = CEPH_CLIENT_CAPS_NO_CAPSNAP; - else - arg.flags = CEPH_CLIENT_CAPS_PENDING_CAPSNAP; - if (sync) - arg.flags |= CEPH_CLIENT_CAPS_SYNC; + if (!(flags & CEPH_CLIENT_CAPS_PENDING_CAPSNAP) && + !list_empty(&ci->i_cap_snaps)) { + struct ceph_cap_snap *capsnap; + list_for_each_entry_reverse(capsnap, &ci->i_cap_snaps, ci_item) { + if (capsnap->cap_flush.tid) + break; + if (capsnap->need_flush) { + flags |= CEPH_CLIENT_CAPS_PENDING_CAPSNAP; + break; + } + } + } + arg.flags = flags; spin_unlock(&ci->i_ceph_lock); @@ -1436,6 +1443,8 @@ static inline int __send_flush_snap(struct inode *inode, arg.atime = capsnap->atime; arg.mtime = capsnap->mtime; arg.ctime = capsnap->ctime; + arg.btime = capsnap->btime; + arg.change_attr = capsnap->change_attr; arg.op = CEPH_CAP_OP_FLUSHSNAP; arg.caps = capsnap->issued; @@ -1603,10 +1612,8 @@ retry: } // make sure flushsnap messages are sent in proper order. - if (ci->i_ceph_flags & CEPH_I_KICK_FLUSH) { + if (ci->i_ceph_flags & CEPH_I_KICK_FLUSH) __kick_flushing_caps(mdsc, session, ci, 0); - ci->i_ceph_flags &= ~CEPH_I_KICK_FLUSH; - } __ceph_flush_snaps(ci, session); out: @@ -2048,10 +2055,8 @@ ack: if (cap == ci->i_auth_cap && (ci->i_ceph_flags & (CEPH_I_KICK_FLUSH | CEPH_I_FLUSH_SNAPS))) { - if (ci->i_ceph_flags & CEPH_I_KICK_FLUSH) { + if (ci->i_ceph_flags & CEPH_I_KICK_FLUSH) __kick_flushing_caps(mdsc, session, ci, 0); - ci->i_ceph_flags &= ~CEPH_I_KICK_FLUSH; - } if (ci->i_ceph_flags & CEPH_I_FLUSH_SNAPS) __ceph_flush_snaps(ci, session); @@ -2087,7 +2092,7 @@ ack: sent++; /* __send_cap drops i_ceph_lock */ - delayed += __send_cap(mdsc, cap, CEPH_CAP_OP_UPDATE, false, + delayed += __send_cap(mdsc, cap, CEPH_CAP_OP_UPDATE, 0, cap_used, want, retain, flushing, flush_tid, oldest_flush_tid); goto retry; /* retake i_ceph_lock and restart our cap scan. */ @@ -2121,6 +2126,7 @@ static int try_flush_caps(struct inode *inode, u64 *ptid) retry: spin_lock(&ci->i_ceph_lock); +retry_locked: if (ci->i_ceph_flags & CEPH_I_NOFLUSH) { spin_unlock(&ci->i_ceph_lock); dout("try_flush_caps skipping %p I_NOFLUSH set\n", inode); @@ -2128,8 +2134,6 @@ retry: } if (ci->i_dirty_caps && ci->i_auth_cap) { struct ceph_cap *cap = ci->i_auth_cap; - int used = __ceph_caps_used(ci); - int want = __ceph_caps_wanted(ci); int delayed; if (!session || session != cap->session) { @@ -2145,13 +2149,25 @@ retry: goto out; } + if (ci->i_ceph_flags & + (CEPH_I_KICK_FLUSH | CEPH_I_FLUSH_SNAPS)) { + if (ci->i_ceph_flags & CEPH_I_KICK_FLUSH) + __kick_flushing_caps(mdsc, session, ci, 0); + if (ci->i_ceph_flags & CEPH_I_FLUSH_SNAPS) + __ceph_flush_snaps(ci, session); + goto retry_locked; + } + flushing = __mark_caps_flushing(inode, session, true, &flush_tid, &oldest_flush_tid); /* __send_cap drops i_ceph_lock */ - delayed = __send_cap(mdsc, cap, CEPH_CAP_OP_FLUSH, true, - used, want, (cap->issued | cap->implemented), - flushing, flush_tid, oldest_flush_tid); + delayed = __send_cap(mdsc, cap, CEPH_CAP_OP_FLUSH, + CEPH_CLIENT_CAPS_SYNC, + __ceph_caps_used(ci), + __ceph_caps_wanted(ci), + (cap->issued | cap->implemented), + flushing, flush_tid, oldest_flush_tid); if (delayed) { spin_lock(&ci->i_ceph_lock); @@ -2320,6 +2336,16 @@ static void __kick_flushing_caps(struct ceph_mds_client *mdsc, struct ceph_cap_flush *cf; int ret; u64 first_tid = 0; + u64 last_snap_flush = 0; + + ci->i_ceph_flags &= ~CEPH_I_KICK_FLUSH; + + list_for_each_entry_reverse(cf, &ci->i_cap_flush_list, i_list) { + if (!cf->caps) { + last_snap_flush = cf->tid; + break; + } + } list_for_each_entry(cf, &ci->i_cap_flush_list, i_list) { if (cf->tid < first_tid) @@ -2338,10 +2364,13 @@ static void __kick_flushing_caps(struct ceph_mds_client *mdsc, dout("kick_flushing_caps %p cap %p tid %llu %s\n", inode, cap, cf->tid, ceph_cap_string(cf->caps)); ci->i_ceph_flags |= CEPH_I_NODELAY; + ret = __send_cap(mdsc, cap, CEPH_CAP_OP_FLUSH, - false, __ceph_caps_used(ci), + (cf->tid < last_snap_flush ? + CEPH_CLIENT_CAPS_PENDING_CAPSNAP : 0), + __ceph_caps_used(ci), __ceph_caps_wanted(ci), - cap->issued | cap->implemented, + (cap->issued | cap->implemented), cf->caps, cf->tid, oldest_flush_tid); if (ret) { pr_err("kick_flushing_caps: error sending " @@ -2410,7 +2439,6 @@ void ceph_early_kick_flushing_caps(struct ceph_mds_client *mdsc, */ if ((cap->issued & ci->i_flushing_caps) != ci->i_flushing_caps) { - ci->i_ceph_flags &= ~CEPH_I_KICK_FLUSH; /* encode_caps_cb() also will reset these sequence * numbers. make sure sequence numbers in cap flush * message match later reconnect message */ @@ -2450,7 +2478,6 @@ void ceph_kick_flushing_caps(struct ceph_mds_client *mdsc, continue; } if (ci->i_ceph_flags & CEPH_I_KICK_FLUSH) { - ci->i_ceph_flags &= ~CEPH_I_KICK_FLUSH; __kick_flushing_caps(mdsc, session, ci, oldest_flush_tid); } @@ -2478,7 +2505,6 @@ static void kick_flushing_inode_caps(struct ceph_mds_client *mdsc, oldest_flush_tid = __get_oldest_flush_tid(mdsc); spin_unlock(&mdsc->cap_dirty_lock); - ci->i_ceph_flags &= ~CEPH_I_KICK_FLUSH; __kick_flushing_caps(mdsc, session, ci, oldest_flush_tid); spin_unlock(&ci->i_ceph_lock); } else { @@ -3040,8 +3066,10 @@ struct cap_extra_info { bool dirstat_valid; u64 nfiles; u64 nsubdirs; + u64 change_attr; /* currently issued */ int issued; + struct timespec64 btime; }; /* @@ -3123,11 +3151,14 @@ static void handle_cap_grant(struct inode *inode, __check_cap_issue(ci, cap, newcaps); + inode_set_max_iversion_raw(inode, extra_info->change_attr); + if ((newcaps & CEPH_CAP_AUTH_SHARED) && (extra_info->issued & CEPH_CAP_AUTH_EXCL) == 0) { inode->i_mode = le32_to_cpu(grant->mode); inode->i_uid = make_kuid(&init_user_ns, le32_to_cpu(grant->uid)); inode->i_gid = make_kgid(&init_user_ns, le32_to_cpu(grant->gid)); + ci->i_btime = extra_info->btime; dout("%p mode 0%o uid.gid %d.%d\n", inode, inode->i_mode, from_kuid(&init_user_ns, inode->i_uid), from_kgid(&init_user_ns, inode->i_gid)); @@ -3154,6 +3185,7 @@ static void handle_cap_grant(struct inode *inode, ci->i_xattrs.blob = ceph_buffer_get(xattr_buf); ci->i_xattrs.version = version; ceph_forget_all_cached_acls(inode); + ceph_security_invalidate_secctx(inode); } } @@ -3848,17 +3880,19 @@ void ceph_handle_caps(struct ceph_mds_session *session, } } - if (msg_version >= 11) { + if (msg_version >= 9) { struct ceph_timespec *btime; - u64 change_attr; - u32 flags; - /* version >= 9 */ if (p + sizeof(*btime) > end) goto bad; btime = p; + ceph_decode_timespec64(&extra_info.btime, btime); p += sizeof(*btime); - ceph_decode_64_safe(&p, end, change_attr, bad); + ceph_decode_64_safe(&p, end, extra_info.change_attr, bad); + } + + if (msg_version >= 11) { + u32 flags; /* version >= 10 */ ceph_decode_32_safe(&p, end, flags, bad); /* version >= 11 */ diff --git a/fs/ceph/debugfs.c b/fs/ceph/debugfs.c index 83cd41fa2b01..2eb88ed22993 100644 --- a/fs/ceph/debugfs.c +++ b/fs/ceph/debugfs.c @@ -52,7 +52,7 @@ static int mdsc_show(struct seq_file *s, void *p) struct ceph_mds_client *mdsc = fsc->mdsc; struct ceph_mds_request *req; struct rb_node *rp; - int pathlen; + int pathlen = 0; u64 pathbase; char *path; diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index 0637149fb9f9..aab29f48c62d 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -825,7 +825,7 @@ static int ceph_mknod(struct inode *dir, struct dentry *dentry, struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb); struct ceph_mds_client *mdsc = fsc->mdsc; struct ceph_mds_request *req; - struct ceph_acls_info acls = {}; + struct ceph_acl_sec_ctx as_ctx = {}; int err; if (ceph_snap(dir) != CEPH_NOSNAP) @@ -836,7 +836,10 @@ static int ceph_mknod(struct inode *dir, struct dentry *dentry, goto out; } - err = ceph_pre_init_acls(dir, &mode, &acls); + err = ceph_pre_init_acls(dir, &mode, &as_ctx); + if (err < 0) + goto out; + err = ceph_security_init_secctx(dentry, mode, &as_ctx); if (err < 0) goto out; @@ -855,9 +858,9 @@ static int ceph_mknod(struct inode *dir, struct dentry *dentry, req->r_args.mknod.rdev = cpu_to_le32(rdev); req->r_dentry_drop = CEPH_CAP_FILE_SHARED | CEPH_CAP_AUTH_EXCL; req->r_dentry_unless = CEPH_CAP_FILE_EXCL; - if (acls.pagelist) { - req->r_pagelist = acls.pagelist; - acls.pagelist = NULL; + if (as_ctx.pagelist) { + req->r_pagelist = as_ctx.pagelist; + as_ctx.pagelist = NULL; } err = ceph_mdsc_do_request(mdsc, dir, req); if (!err && !req->r_reply_info.head->is_dentry) @@ -865,10 +868,10 @@ static int ceph_mknod(struct inode *dir, struct dentry *dentry, ceph_mdsc_put_request(req); out: if (!err) - ceph_init_inode_acls(d_inode(dentry), &acls); + ceph_init_inode_acls(d_inode(dentry), &as_ctx); else d_drop(dentry); - ceph_release_acls_info(&acls); + ceph_release_acl_sec_ctx(&as_ctx); return err; } @@ -884,6 +887,7 @@ static int ceph_symlink(struct inode *dir, struct dentry *dentry, struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb); struct ceph_mds_client *mdsc = fsc->mdsc; struct ceph_mds_request *req; + struct ceph_acl_sec_ctx as_ctx = {}; int err; if (ceph_snap(dir) != CEPH_NOSNAP) @@ -894,6 +898,10 @@ static int ceph_symlink(struct inode *dir, struct dentry *dentry, goto out; } + err = ceph_security_init_secctx(dentry, S_IFLNK | 0777, &as_ctx); + if (err < 0) + goto out; + dout("symlink in dir %p dentry %p to '%s'\n", dir, dentry, dest); req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_SYMLINK, USE_AUTH_MDS); if (IS_ERR(req)) { @@ -919,6 +927,7 @@ static int ceph_symlink(struct inode *dir, struct dentry *dentry, out: if (err) d_drop(dentry); + ceph_release_acl_sec_ctx(&as_ctx); return err; } @@ -927,7 +936,7 @@ static int ceph_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb); struct ceph_mds_client *mdsc = fsc->mdsc; struct ceph_mds_request *req; - struct ceph_acls_info acls = {}; + struct ceph_acl_sec_ctx as_ctx = {}; int err = -EROFS; int op; @@ -950,7 +959,10 @@ static int ceph_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) } mode |= S_IFDIR; - err = ceph_pre_init_acls(dir, &mode, &acls); + err = ceph_pre_init_acls(dir, &mode, &as_ctx); + if (err < 0) + goto out; + err = ceph_security_init_secctx(dentry, mode, &as_ctx); if (err < 0) goto out; @@ -967,9 +979,9 @@ static int ceph_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) req->r_args.mkdir.mode = cpu_to_le32(mode); req->r_dentry_drop = CEPH_CAP_FILE_SHARED | CEPH_CAP_AUTH_EXCL; req->r_dentry_unless = CEPH_CAP_FILE_EXCL; - if (acls.pagelist) { - req->r_pagelist = acls.pagelist; - acls.pagelist = NULL; + if (as_ctx.pagelist) { + req->r_pagelist = as_ctx.pagelist; + as_ctx.pagelist = NULL; } err = ceph_mdsc_do_request(mdsc, dir, req); if (!err && @@ -979,10 +991,10 @@ static int ceph_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) ceph_mdsc_put_request(req); out: if (!err) - ceph_init_inode_acls(d_inode(dentry), &acls); + ceph_init_inode_acls(d_inode(dentry), &as_ctx); else d_drop(dentry); - ceph_release_acls_info(&acls); + ceph_release_acl_sec_ctx(&as_ctx); return err; } @@ -1433,8 +1445,7 @@ static bool __dentry_lease_is_valid(struct ceph_dentry_info *di) return false; } -static int dentry_lease_is_valid(struct dentry *dentry, unsigned int flags, - struct inode *dir) +static int dentry_lease_is_valid(struct dentry *dentry, unsigned int flags) { struct ceph_dentry_info *di; struct ceph_mds_session *session = NULL; @@ -1466,7 +1477,7 @@ static int dentry_lease_is_valid(struct dentry *dentry, unsigned int flags, spin_unlock(&dentry->d_lock); if (session) { - ceph_mdsc_lease_send_msg(session, dir, dentry, + ceph_mdsc_lease_send_msg(session, dentry, CEPH_MDS_LEASE_RENEW, seq); ceph_put_mds_session(session); } @@ -1512,18 +1523,26 @@ static int __dir_lease_try_check(const struct dentry *dentry) static int dir_lease_is_valid(struct inode *dir, struct dentry *dentry) { struct ceph_inode_info *ci = ceph_inode(dir); - struct ceph_dentry_info *di = ceph_dentry(dentry); - int valid = 0; + int valid; + int shared_gen; spin_lock(&ci->i_ceph_lock); - if (atomic_read(&ci->i_shared_gen) == di->lease_shared_gen) - valid = __ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1); + valid = __ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1); + shared_gen = atomic_read(&ci->i_shared_gen); spin_unlock(&ci->i_ceph_lock); - if (valid) - __ceph_dentry_dir_lease_touch(di); - dout("dir_lease_is_valid dir %p v%u dentry %p v%u = %d\n", - dir, (unsigned)atomic_read(&ci->i_shared_gen), - dentry, (unsigned)di->lease_shared_gen, valid); + if (valid) { + struct ceph_dentry_info *di; + spin_lock(&dentry->d_lock); + di = ceph_dentry(dentry); + if (dir == d_inode(dentry->d_parent) && + di && di->lease_shared_gen == shared_gen) + __ceph_dentry_dir_lease_touch(di); + else + valid = 0; + spin_unlock(&dentry->d_lock); + } + dout("dir_lease_is_valid dir %p v%u dentry %p = %d\n", + dir, (unsigned)atomic_read(&ci->i_shared_gen), dentry, valid); return valid; } @@ -1558,7 +1577,7 @@ static int ceph_d_revalidate(struct dentry *dentry, unsigned int flags) ceph_snap(d_inode(dentry)) == CEPH_SNAPDIR) { valid = 1; } else { - valid = dentry_lease_is_valid(dentry, flags, dir); + valid = dentry_lease_is_valid(dentry, flags); if (valid == -ECHILD) return valid; if (valid || dir_lease_is_valid(dir, dentry)) { diff --git a/fs/ceph/export.c b/fs/ceph/export.c index d3ef7ee429ec..15ff1b09cfa2 100644 --- a/fs/ceph/export.c +++ b/fs/ceph/export.c @@ -368,7 +368,7 @@ static struct dentry *ceph_get_parent(struct dentry *child) } out: dout("get_parent %p ino %llx.%llx err=%ld\n", - child, ceph_vinop(inode), (IS_ERR(dn) ? PTR_ERR(dn) : 0)); + child, ceph_vinop(inode), (long)PTR_ERR_OR_ZERO(dn)); return dn; } diff --git a/fs/ceph/file.c b/fs/ceph/file.c index c5517ffeb11c..685a03cc4b77 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -10,6 +10,7 @@ #include <linux/namei.h> #include <linux/writeback.h> #include <linux/falloc.h> +#include <linux/iversion.h> #include "super.h" #include "mds_client.h" @@ -437,7 +438,7 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry, struct ceph_mds_client *mdsc = fsc->mdsc; struct ceph_mds_request *req; struct dentry *dn; - struct ceph_acls_info acls = {}; + struct ceph_acl_sec_ctx as_ctx = {}; int mask; int err; @@ -451,25 +452,28 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry, if (flags & O_CREAT) { if (ceph_quota_is_max_files_exceeded(dir)) return -EDQUOT; - err = ceph_pre_init_acls(dir, &mode, &acls); + err = ceph_pre_init_acls(dir, &mode, &as_ctx); if (err < 0) return err; + err = ceph_security_init_secctx(dentry, mode, &as_ctx); + if (err < 0) + goto out_ctx; } /* do the open */ req = prepare_open_request(dir->i_sb, flags, mode); if (IS_ERR(req)) { err = PTR_ERR(req); - goto out_acl; + goto out_ctx; } req->r_dentry = dget(dentry); req->r_num_caps = 2; if (flags & O_CREAT) { req->r_dentry_drop = CEPH_CAP_FILE_SHARED | CEPH_CAP_AUTH_EXCL; req->r_dentry_unless = CEPH_CAP_FILE_EXCL; - if (acls.pagelist) { - req->r_pagelist = acls.pagelist; - acls.pagelist = NULL; + if (as_ctx.pagelist) { + req->r_pagelist = as_ctx.pagelist; + as_ctx.pagelist = NULL; } } @@ -507,7 +511,7 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry, } else { dout("atomic_open finish_open on dn %p\n", dn); if (req->r_op == CEPH_MDS_OP_CREATE && req->r_reply_info.has_create_ino) { - ceph_init_inode_acls(d_inode(dentry), &acls); + ceph_init_inode_acls(d_inode(dentry), &as_ctx); file->f_mode |= FMODE_CREATED; } err = finish_open(file, dentry, ceph_open); @@ -516,8 +520,8 @@ out_req: if (!req->r_err && req->r_target_inode) ceph_put_fmode(ceph_inode(req->r_target_inode), req->r_fmode); ceph_mdsc_put_request(req); -out_acl: - ceph_release_acls_info(&acls); +out_ctx: + ceph_release_acl_sec_ctx(&as_ctx); dout("atomic_open result=%d\n", err); return err; } @@ -1007,7 +1011,7 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter, * may block. */ truncate_inode_pages_range(inode->i_mapping, pos, - (pos+len) | (PAGE_SIZE - 1)); + PAGE_ALIGN(pos + len) - 1); req->r_mtime = mtime; } @@ -1022,7 +1026,7 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter, req->r_callback = ceph_aio_complete_req; req->r_inode = inode; req->r_priv = aio_req; - list_add_tail(&req->r_unsafe_item, &aio_req->osd_reqs); + list_add_tail(&req->r_private_item, &aio_req->osd_reqs); pos += len; continue; @@ -1082,8 +1086,8 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter, while (!list_empty(&osd_reqs)) { req = list_first_entry(&osd_reqs, struct ceph_osd_request, - r_unsafe_item); - list_del_init(&req->r_unsafe_item); + r_private_item); + list_del_init(&req->r_private_item); if (ret >= 0) ret = ceph_osdc_start_request(req->r_osdc, req, false); @@ -1432,6 +1436,8 @@ retry_snap: if (err) goto out; + inode_inc_iversion_raw(inode); + if (ci->i_inline_version != CEPH_INLINE_NONE) { err = ceph_uninline_data(file, NULL); if (err < 0) @@ -2063,6 +2069,8 @@ static ssize_t __ceph_copy_file_range(struct file *src_file, loff_t src_off, do_final_copy = true; file_update_time(dst_file); + inode_inc_iversion_raw(dst_inode); + if (endoff > size) { int caps_flags = 0; diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index 761451f36e2d..791f84a13bb8 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -13,6 +13,7 @@ #include <linux/posix_acl.h> #include <linux/random.h> #include <linux/sort.h> +#include <linux/iversion.h> #include "super.h" #include "mds_client.h" @@ -42,6 +43,7 @@ static int ceph_set_ino_cb(struct inode *inode, void *data) { ceph_inode(inode)->i_vino = *(struct ceph_vino *)data; inode->i_ino = ceph_vino_to_ino(*(struct ceph_vino *)data); + inode_set_iversion_raw(inode, 0); return 0; } @@ -509,6 +511,7 @@ struct inode *ceph_alloc_inode(struct super_block *sb) INIT_WORK(&ci->i_work, ceph_inode_work); ci->i_work_mask = 0; + memset(&ci->i_btime, '\0', sizeof(ci->i_btime)); ceph_fscache_inode_init(ci); @@ -523,17 +526,20 @@ void ceph_free_inode(struct inode *inode) kmem_cache_free(ceph_inode_cachep, ci); } -void ceph_destroy_inode(struct inode *inode) +void ceph_evict_inode(struct inode *inode) { struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_inode_frag *frag; struct rb_node *n; - dout("destroy_inode %p ino %llx.%llx\n", inode, ceph_vinop(inode)); + dout("evict_inode %p ino %llx.%llx\n", inode, ceph_vinop(inode)); + + truncate_inode_pages_final(&inode->i_data); + clear_inode(inode); ceph_fscache_unregister_inode_cookie(ci); - __ceph_remove_caps(inode); + __ceph_remove_caps(ci); if (__ceph_has_any_quota(ci)) ceph_adjust_quota_realms_count(inode, false); @@ -578,16 +584,6 @@ void ceph_destroy_inode(struct inode *inode) ceph_put_string(rcu_dereference_raw(ci->i_layout.pool_ns)); } -int ceph_drop_inode(struct inode *inode) -{ - /* - * Positve dentry and corresponding inode are always accompanied - * in MDS reply. So no need to keep inode in the cache after - * dropping all its aliases. - */ - return 1; -} - static inline blkcnt_t calc_inode_blocks(u64 size) { return (size + (1<<9) - 1) >> 9; @@ -795,6 +791,9 @@ static int fill_inode(struct inode *inode, struct page *locked_page, le64_to_cpu(info->version) > (ci->i_version & ~1))) new_version = true; + /* Update change_attribute */ + inode_set_max_iversion_raw(inode, iinfo->change_attr); + __ceph_caps_issued(ci, &issued); issued |= __ceph_caps_dirty(ci); new_issued = ~issued & info_caps; @@ -813,6 +812,8 @@ static int fill_inode(struct inode *inode, struct page *locked_page, dout("%p mode 0%o uid.gid %d.%d\n", inode, inode->i_mode, from_kuid(&init_user_ns, inode->i_uid), from_kgid(&init_user_ns, inode->i_gid)); + ceph_decode_timespec64(&ci->i_btime, &iinfo->btime); + ceph_decode_timespec64(&ci->i_snap_btime, &iinfo->snap_btime); } if ((new_version || (new_issued & CEPH_CAP_LINK_SHARED)) && @@ -887,6 +888,7 @@ static int fill_inode(struct inode *inode, struct page *locked_page, iinfo->xattr_data, iinfo->xattr_len); ci->i_xattrs.version = le64_to_cpu(info->xattr_version); ceph_forget_all_cached_acls(inode); + ceph_security_invalidate_secctx(inode); xattr_blob = NULL; } @@ -1027,59 +1029,38 @@ out: } /* - * caller should hold session s_mutex. + * caller should hold session s_mutex and dentry->d_lock. */ -static void update_dentry_lease(struct dentry *dentry, - struct ceph_mds_reply_lease *lease, - struct ceph_mds_session *session, - unsigned long from_time, - struct ceph_vino *tgt_vino, - struct ceph_vino *dir_vino) +static void __update_dentry_lease(struct inode *dir, struct dentry *dentry, + struct ceph_mds_reply_lease *lease, + struct ceph_mds_session *session, + unsigned long from_time, + struct ceph_mds_session **old_lease_session) { struct ceph_dentry_info *di = ceph_dentry(dentry); long unsigned duration = le32_to_cpu(lease->duration_ms); long unsigned ttl = from_time + (duration * HZ) / 1000; long unsigned half_ttl = from_time + (duration * HZ / 2) / 1000; - struct inode *dir; - struct ceph_mds_session *old_lease_session = NULL; - /* - * Make sure dentry's inode matches tgt_vino. NULL tgt_vino means that - * we expect a negative dentry. - */ - if (!tgt_vino && d_really_is_positive(dentry)) - return; - - if (tgt_vino && (d_really_is_negative(dentry) || - !ceph_ino_compare(d_inode(dentry), tgt_vino))) - return; - - spin_lock(&dentry->d_lock); dout("update_dentry_lease %p duration %lu ms ttl %lu\n", dentry, duration, ttl); - dir = d_inode(dentry->d_parent); - - /* make sure parent matches dir_vino */ - if (!ceph_ino_compare(dir, dir_vino)) - goto out_unlock; - /* only track leases on regular dentries */ if (ceph_snap(dir) != CEPH_NOSNAP) - goto out_unlock; + return; di->lease_shared_gen = atomic_read(&ceph_inode(dir)->i_shared_gen); if (duration == 0) { __ceph_dentry_dir_lease_touch(di); - goto out_unlock; + return; } if (di->lease_gen == session->s_cap_gen && time_before(ttl, di->time)) - goto out_unlock; /* we already have a newer lease. */ + return; /* we already have a newer lease. */ if (di->lease_session && di->lease_session != session) { - old_lease_session = di->lease_session; + *old_lease_session = di->lease_session; di->lease_session = NULL; } @@ -1092,6 +1073,62 @@ static void update_dentry_lease(struct dentry *dentry, di->time = ttl; __ceph_dentry_lease_touch(di); +} + +static inline void update_dentry_lease(struct inode *dir, struct dentry *dentry, + struct ceph_mds_reply_lease *lease, + struct ceph_mds_session *session, + unsigned long from_time) +{ + struct ceph_mds_session *old_lease_session = NULL; + spin_lock(&dentry->d_lock); + __update_dentry_lease(dir, dentry, lease, session, from_time, + &old_lease_session); + spin_unlock(&dentry->d_lock); + if (old_lease_session) + ceph_put_mds_session(old_lease_session); +} + +/* + * update dentry lease without having parent inode locked + */ +static void update_dentry_lease_careful(struct dentry *dentry, + struct ceph_mds_reply_lease *lease, + struct ceph_mds_session *session, + unsigned long from_time, + char *dname, u32 dname_len, + struct ceph_vino *pdvino, + struct ceph_vino *ptvino) + +{ + struct inode *dir; + struct ceph_mds_session *old_lease_session = NULL; + + spin_lock(&dentry->d_lock); + /* make sure dentry's name matches target */ + if (dentry->d_name.len != dname_len || + memcmp(dentry->d_name.name, dname, dname_len)) + goto out_unlock; + + dir = d_inode(dentry->d_parent); + /* make sure parent matches dvino */ + if (!ceph_ino_compare(dir, pdvino)) + goto out_unlock; + + /* make sure dentry's inode matches target. NULL ptvino means that + * we expect a negative dentry */ + if (ptvino) { + if (d_really_is_negative(dentry)) + goto out_unlock; + if (!ceph_ino_compare(d_inode(dentry), ptvino)) + goto out_unlock; + } else { + if (d_really_is_positive(dentry)) + goto out_unlock; + } + + __update_dentry_lease(dir, dentry, lease, session, + from_time, &old_lease_session); out_unlock: spin_unlock(&dentry->d_lock); if (old_lease_session) @@ -1156,19 +1193,6 @@ static int splice_dentry(struct dentry **pdn, struct inode *in) return 0; } -static int d_name_cmp(struct dentry *dentry, const char *name, size_t len) -{ - int ret; - - /* take d_lock to ensure dentry->d_name stability */ - spin_lock(&dentry->d_lock); - ret = dentry->d_name.len - len; - if (!ret) - ret = memcmp(dentry->d_name.name, name, len); - spin_unlock(&dentry->d_lock); - return ret; -} - /* * Incorporate results into the local cache. This is either just * one inode, or a directory, dentry, and possibly linked-to inode (e.g., @@ -1371,10 +1395,9 @@ retry_lookup: } else if (have_lease) { if (d_unhashed(dn)) d_add(dn, NULL); - update_dentry_lease(dn, rinfo->dlease, - session, - req->r_request_started, - NULL, &dvino); + update_dentry_lease(dir, dn, + rinfo->dlease, session, + req->r_request_started); } goto done; } @@ -1396,11 +1419,9 @@ retry_lookup: } if (have_lease) { - tvino.ino = le64_to_cpu(rinfo->targeti.in->ino); - tvino.snap = le64_to_cpu(rinfo->targeti.in->snapid); - update_dentry_lease(dn, rinfo->dlease, session, - req->r_request_started, - &tvino, &dvino); + update_dentry_lease(dir, dn, + rinfo->dlease, session, + req->r_request_started); } dout(" final dn %p\n", dn); } else if ((req->r_op == CEPH_MDS_OP_LOOKUPSNAP || @@ -1418,27 +1439,20 @@ retry_lookup: err = splice_dentry(&req->r_dentry, in); if (err < 0) goto done; - } else if (rinfo->head->is_dentry && - !d_name_cmp(req->r_dentry, rinfo->dname, rinfo->dname_len)) { + } else if (rinfo->head->is_dentry && req->r_dentry) { + /* parent inode is not locked, be carefull */ struct ceph_vino *ptvino = NULL; - - if ((le32_to_cpu(rinfo->diri.in->cap.caps) & CEPH_CAP_FILE_SHARED) || - le32_to_cpu(rinfo->dlease->duration_ms)) { - dvino.ino = le64_to_cpu(rinfo->diri.in->ino); - dvino.snap = le64_to_cpu(rinfo->diri.in->snapid); - - if (rinfo->head->is_target) { - tvino.ino = le64_to_cpu(rinfo->targeti.in->ino); - tvino.snap = le64_to_cpu(rinfo->targeti.in->snapid); - ptvino = &tvino; - } - - update_dentry_lease(req->r_dentry, rinfo->dlease, - session, req->r_request_started, ptvino, - &dvino); - } else { - dout("%s: no dentry lease or dir cap\n", __func__); + dvino.ino = le64_to_cpu(rinfo->diri.in->ino); + dvino.snap = le64_to_cpu(rinfo->diri.in->snapid); + if (rinfo->head->is_target) { + tvino.ino = le64_to_cpu(rinfo->targeti.in->ino); + tvino.snap = le64_to_cpu(rinfo->targeti.in->snapid); + ptvino = &tvino; } + update_dentry_lease_careful(req->r_dentry, rinfo->dlease, + session, req->r_request_started, + rinfo->dname, rinfo->dname_len, + &dvino, ptvino); } done: dout("fill_trace done err=%d\n", err); @@ -1600,7 +1614,7 @@ int ceph_readdir_prepopulate(struct ceph_mds_request *req, /* FIXME: release caps/leases if error occurs */ for (i = 0; i < rinfo->dir_nr; i++) { struct ceph_mds_reply_dir_entry *rde = rinfo->dir_entries + i; - struct ceph_vino tvino, dvino; + struct ceph_vino tvino; dname.name = rde->name; dname.len = rde->name_len; @@ -1701,9 +1715,9 @@ retry_lookup: ceph_dentry(dn)->offset = rde->offset; - dvino = ceph_vino(d_inode(parent)); - update_dentry_lease(dn, rde->lease, req->r_session, - req->r_request_started, &tvino, &dvino); + update_dentry_lease(d_inode(parent), dn, + rde->lease, req->r_session, + req->r_request_started); if (err == 0 && skipped == 0 && cache_ctl.index >= 0) { ret = fill_readdir_cache(d_inode(parent), dn, @@ -2282,7 +2296,7 @@ static int statx_to_caps(u32 want) { int mask = 0; - if (want & (STATX_MODE|STATX_UID|STATX_GID|STATX_CTIME)) + if (want & (STATX_MODE|STATX_UID|STATX_GID|STATX_CTIME|STATX_BTIME)) mask |= CEPH_CAP_AUTH_SHARED; if (want & (STATX_NLINK|STATX_CTIME)) @@ -2307,6 +2321,7 @@ int ceph_getattr(const struct path *path, struct kstat *stat, { struct inode *inode = d_inode(path->dentry); struct ceph_inode_info *ci = ceph_inode(inode); + u32 valid_mask = STATX_BASIC_STATS; int err = 0; /* Skip the getattr altogether if we're asked not to sync */ @@ -2319,6 +2334,16 @@ int ceph_getattr(const struct path *path, struct kstat *stat, generic_fillattr(inode, stat); stat->ino = ceph_translate_ino(inode->i_sb, inode->i_ino); + + /* + * btime on newly-allocated inodes is 0, so if this is still set to + * that, then assume that it's not valid. + */ + if (ci->i_btime.tv_sec || ci->i_btime.tv_nsec) { + stat->btime = ci->i_btime; + valid_mask |= STATX_BTIME; + } + if (ceph_snap(inode) == CEPH_NOSNAP) stat->dev = inode->i_sb->s_dev; else @@ -2342,7 +2367,6 @@ int ceph_getattr(const struct path *path, struct kstat *stat, stat->nlink = 1 + 1 + ci->i_subdirs; } - /* Mask off any higher bits (e.g. btime) until we have support */ - stat->result_mask = request_mask & STATX_BASIC_STATS; + stat->result_mask = request_mask & valid_mask; return err; } diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index c8a9b89b922d..920e9f048bd8 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -150,14 +150,13 @@ static int parse_reply_info_in(void **p, void *end, info->pool_ns_data = *p; *p += info->pool_ns_len; } - /* btime, change_attr */ - { - struct ceph_timespec btime; - u64 change_attr; - ceph_decode_need(p, end, sizeof(btime), bad); - ceph_decode_copy(p, &btime, sizeof(btime)); - ceph_decode_64_safe(p, end, change_attr, bad); - } + + /* btime */ + ceph_decode_need(p, end, sizeof(info->btime), bad); + ceph_decode_copy(p, &info->btime, sizeof(info->btime)); + + /* change attribute */ + ceph_decode_64_safe(p, end, info->change_attr, bad); /* dir pin */ if (struct_v >= 2) { @@ -166,6 +165,15 @@ static int parse_reply_info_in(void **p, void *end, info->dir_pin = -ENODATA; } + /* snapshot birth time, remains zero for v<=2 */ + if (struct_v >= 3) { + ceph_decode_need(p, end, sizeof(info->snap_btime), bad); + ceph_decode_copy(p, &info->snap_btime, + sizeof(info->snap_btime)); + } else { + memset(&info->snap_btime, 0, sizeof(info->snap_btime)); + } + *p = end; } else { if (features & CEPH_FEATURE_MDS_INLINE_DATA) { @@ -197,7 +205,14 @@ static int parse_reply_info_in(void **p, void *end, } } + if (features & CEPH_FEATURE_FS_BTIME) { + ceph_decode_need(p, end, sizeof(info->btime), bad); + ceph_decode_copy(p, &info->btime, sizeof(info->btime)); + ceph_decode_64_safe(p, end, info->change_attr, bad); + } + info->dir_pin = -ENODATA; + /* info->snap_btime remains zero */ } return 0; bad: @@ -717,6 +732,7 @@ void ceph_mdsc_release_request(struct kref *kref) ceph_pagelist_release(req->r_pagelist); put_request_session(req); ceph_unreserve_caps(req->r_mdsc, &req->r_caps_reservation); + WARN_ON_ONCE(!list_empty(&req->r_wait)); kfree(req); } @@ -903,7 +919,7 @@ static int __choose_mds(struct ceph_mds_client *mdsc, struct inode *dir; rcu_read_lock(); - parent = req->r_dentry->d_parent; + parent = READ_ONCE(req->r_dentry->d_parent); dir = req->r_parent ? : d_inode_rcu(parent); if (!dir || dir->i_sb != mdsc->fsc->sb) { @@ -2135,7 +2151,7 @@ retry: memcpy(path + pos, temp->d_name.name, temp->d_name.len); } spin_unlock(&temp->d_lock); - temp = temp->d_parent; + temp = READ_ONCE(temp->d_parent); /* Are we at the root? */ if (IS_ROOT(temp)) @@ -3727,42 +3743,35 @@ static void check_new_map(struct ceph_mds_client *mdsc, ceph_mdsmap_is_laggy(newmap, i) ? " (laggy)" : "", ceph_session_state_name(s->s_state)); - if (i >= newmap->m_num_mds || - memcmp(ceph_mdsmap_get_addr(oldmap, i), - ceph_mdsmap_get_addr(newmap, i), - sizeof(struct ceph_entity_addr))) { - if (s->s_state == CEPH_MDS_SESSION_OPENING) { - /* the session never opened, just close it - * out now */ - get_session(s); - __unregister_session(mdsc, s); - __wake_requests(mdsc, &s->s_waiting); - ceph_put_mds_session(s); - } else if (i >= newmap->m_num_mds) { - /* force close session for stopped mds */ - get_session(s); - __unregister_session(mdsc, s); - __wake_requests(mdsc, &s->s_waiting); - kick_requests(mdsc, i); - mutex_unlock(&mdsc->mutex); + if (i >= newmap->m_num_mds) { + /* force close session for stopped mds */ + get_session(s); + __unregister_session(mdsc, s); + __wake_requests(mdsc, &s->s_waiting); + mutex_unlock(&mdsc->mutex); - mutex_lock(&s->s_mutex); - cleanup_session_requests(mdsc, s); - remove_session_caps(s); - mutex_unlock(&s->s_mutex); + mutex_lock(&s->s_mutex); + cleanup_session_requests(mdsc, s); + remove_session_caps(s); + mutex_unlock(&s->s_mutex); - ceph_put_mds_session(s); + ceph_put_mds_session(s); - mutex_lock(&mdsc->mutex); - } else { - /* just close it */ - mutex_unlock(&mdsc->mutex); - mutex_lock(&s->s_mutex); - mutex_lock(&mdsc->mutex); - ceph_con_close(&s->s_con); - mutex_unlock(&s->s_mutex); - s->s_state = CEPH_MDS_SESSION_RESTARTING; - } + mutex_lock(&mdsc->mutex); + kick_requests(mdsc, i); + continue; + } + + if (memcmp(ceph_mdsmap_get_addr(oldmap, i), + ceph_mdsmap_get_addr(newmap, i), + sizeof(struct ceph_entity_addr))) { + /* just close it */ + mutex_unlock(&mdsc->mutex); + mutex_lock(&s->s_mutex); + mutex_lock(&mdsc->mutex); + ceph_con_close(&s->s_con); + mutex_unlock(&s->s_mutex); + s->s_state = CEPH_MDS_SESSION_RESTARTING; } else if (oldstate == newstate) { continue; /* nothing new with this mds */ } @@ -3931,31 +3940,33 @@ bad: } void ceph_mdsc_lease_send_msg(struct ceph_mds_session *session, - struct inode *inode, struct dentry *dentry, char action, u32 seq) { struct ceph_msg *msg; struct ceph_mds_lease *lease; - int len = sizeof(*lease) + sizeof(u32); - int dnamelen = 0; + struct inode *dir; + int len = sizeof(*lease) + sizeof(u32) + NAME_MAX; - dout("lease_send_msg inode %p dentry %p %s to mds%d\n", - inode, dentry, ceph_lease_op_name(action), session->s_mds); - dnamelen = dentry->d_name.len; - len += dnamelen; + dout("lease_send_msg identry %p %s to mds%d\n", + dentry, ceph_lease_op_name(action), session->s_mds); msg = ceph_msg_new(CEPH_MSG_CLIENT_LEASE, len, GFP_NOFS, false); if (!msg) return; lease = msg->front.iov_base; lease->action = action; - lease->ino = cpu_to_le64(ceph_vino(inode).ino); - lease->first = lease->last = cpu_to_le64(ceph_vino(inode).snap); lease->seq = cpu_to_le32(seq); - put_unaligned_le32(dnamelen, lease + 1); - memcpy((void *)(lease + 1) + 4, dentry->d_name.name, dnamelen); + spin_lock(&dentry->d_lock); + dir = d_inode(dentry->d_parent); + lease->ino = cpu_to_le64(ceph_ino(dir)); + lease->first = lease->last = cpu_to_le64(ceph_snap(dir)); + + put_unaligned_le32(dentry->d_name.len, lease + 1); + memcpy((void *)(lease + 1) + 4, + dentry->d_name.name, dentry->d_name.len); + spin_unlock(&dentry->d_lock); /* * if this is a preemptive lease RELEASE, no need to * flush request stream, since the actual request will @@ -4157,6 +4168,7 @@ static void wait_requests(struct ceph_mds_client *mdsc) while ((req = __get_oldest_req(mdsc))) { dout("wait_requests timed out on tid %llu\n", req->r_tid); + list_del_init(&req->r_wait); __unregister_request(mdsc, req); } } diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h index a83f28bc2387..f7c8603484fe 100644 --- a/fs/ceph/mds_client.h +++ b/fs/ceph/mds_client.h @@ -69,6 +69,9 @@ struct ceph_mds_reply_info_in { u64 max_bytes; u64 max_files; s32 dir_pin; + struct ceph_timespec btime; + struct ceph_timespec snap_btime; + u64 change_attr; }; struct ceph_mds_reply_dir_entry { @@ -504,7 +507,6 @@ extern char *ceph_mdsc_build_path(struct dentry *dentry, int *plen, u64 *base, extern void __ceph_mdsc_drop_dentry_lease(struct dentry *dentry); extern void ceph_mdsc_lease_send_msg(struct ceph_mds_session *session, - struct inode *inode, struct dentry *dentry, char action, u32 seq); diff --git a/fs/ceph/mdsmap.c b/fs/ceph/mdsmap.c index 701b4fb0fb5a..ce2d00da5096 100644 --- a/fs/ceph/mdsmap.c +++ b/fs/ceph/mdsmap.c @@ -107,7 +107,7 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end) struct ceph_mdsmap *m; const void *start = *p; int i, j, n; - int err = -EINVAL; + int err; u8 mdsmap_v, mdsmap_cv; u16 mdsmap_ev; @@ -183,8 +183,9 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end) inc = ceph_decode_32(p); state = ceph_decode_32(p); state_seq = ceph_decode_64(p); - ceph_decode_copy(p, &addr, sizeof(addr)); - ceph_decode_addr(&addr); + err = ceph_decode_entity_addr(p, end, &addr); + if (err) + goto corrupt; ceph_decode_copy(p, &laggy_since, sizeof(laggy_since)); *p += sizeof(u32); ceph_decode_32_safe(p, end, namelen, bad); @@ -357,7 +358,7 @@ bad_ext: nomem: err = -ENOMEM; goto out_err; -bad: +corrupt: pr_err("corrupt mdsmap\n"); print_hex_dump(KERN_DEBUG, "mdsmap: ", DUMP_PREFIX_OFFSET, 16, 1, @@ -365,6 +366,9 @@ bad: out_err: ceph_mdsmap_destroy(m); return ERR_PTR(err); +bad: + err = -EINVAL; + goto corrupt; } void ceph_mdsmap_destroy(struct ceph_mdsmap *m) diff --git a/fs/ceph/quota.c b/fs/ceph/quota.c index d629fc857450..de56dee60540 100644 --- a/fs/ceph/quota.c +++ b/fs/ceph/quota.c @@ -135,7 +135,7 @@ static struct inode *lookup_quotarealm_inode(struct ceph_mds_client *mdsc, return NULL; mutex_lock(&qri->mutex); - if (qri->inode) { + if (qri->inode && ceph_is_any_caps(qri->inode)) { /* A request has already returned the inode */ mutex_unlock(&qri->mutex); return qri->inode; @@ -146,7 +146,18 @@ static struct inode *lookup_quotarealm_inode(struct ceph_mds_client *mdsc, mutex_unlock(&qri->mutex); return NULL; } - in = ceph_lookup_inode(sb, realm->ino); + if (qri->inode) { + /* get caps */ + int ret = __ceph_do_getattr(qri->inode, NULL, + CEPH_STAT_CAP_INODE, true); + if (ret >= 0) + in = qri->inode; + else + in = ERR_PTR(ret); + } else { + in = ceph_lookup_inode(sb, realm->ino); + } + if (IS_ERR(in)) { pr_warn("Can't lookup inode %llx (err: %ld)\n", realm->ino, PTR_ERR(in)); diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c index 72c6c022f02b..4c6494eb02b5 100644 --- a/fs/ceph/snap.c +++ b/fs/ceph/snap.c @@ -3,6 +3,7 @@ #include <linux/sort.h> #include <linux/slab.h> +#include <linux/iversion.h> #include "super.h" #include "mds_client.h" #include <linux/ceph/decode.h> @@ -606,6 +607,8 @@ int __ceph_finish_cap_snap(struct ceph_inode_info *ci, capsnap->mtime = inode->i_mtime; capsnap->atime = inode->i_atime; capsnap->ctime = inode->i_ctime; + capsnap->btime = ci->i_btime; + capsnap->change_attr = inode_peek_iversion_raw(inode); capsnap->time_warp_seq = ci->i_time_warp_seq; capsnap->truncate_size = ci->i_truncate_size; capsnap->truncate_seq = ci->i_truncate_seq; diff --git a/fs/ceph/super.c b/fs/ceph/super.c index ed1b65a6c2c3..ab4868c7308e 100644 --- a/fs/ceph/super.c +++ b/fs/ceph/super.c @@ -840,10 +840,10 @@ static int ceph_remount(struct super_block *sb, int *flags, char *data) static const struct super_operations ceph_super_ops = { .alloc_inode = ceph_alloc_inode, - .destroy_inode = ceph_destroy_inode, .free_inode = ceph_free_inode, .write_inode = ceph_write_inode, - .drop_inode = ceph_drop_inode, + .drop_inode = generic_delete_inode, + .evict_inode = ceph_evict_inode, .sync_fs = ceph_sync_fs, .put_super = ceph_put_super, .remount_fs = ceph_remount, @@ -978,7 +978,7 @@ static int ceph_set_super(struct super_block *s, void *data) s->s_d_op = &ceph_dentry_ops; s->s_export_op = &ceph_export_ops; - s->s_time_gran = 1000; /* 1000 ns == 1 us */ + s->s_time_gran = 1; ret = set_anon_super(s, NULL); /* what is that second arg for? */ if (ret != 0) @@ -1159,17 +1159,15 @@ static int __init init_ceph(void) goto out; ceph_flock_init(); - ceph_xattr_init(); ret = register_filesystem(&ceph_fs_type); if (ret) - goto out_xattr; + goto out_caches; pr_info("loaded (mds proto %d)\n", CEPH_MDSC_PROTOCOL); return 0; -out_xattr: - ceph_xattr_exit(); +out_caches: destroy_caches(); out: return ret; @@ -1179,7 +1177,6 @@ static void __exit exit_ceph(void) { dout("exit_ceph\n"); unregister_filesystem(&ceph_fs_type); - ceph_xattr_exit(); destroy_caches(); } diff --git a/fs/ceph/super.h b/fs/ceph/super.h index fbe6869a3f95..d2352fd95dbc 100644 --- a/fs/ceph/super.h +++ b/fs/ceph/super.h @@ -197,7 +197,8 @@ struct ceph_cap_snap { u64 xattr_version; u64 size; - struct timespec64 mtime, atime, ctime; + u64 change_attr; + struct timespec64 mtime, atime, ctime, btime; u64 time_warp_seq; u64 truncate_size; u32 truncate_seq; @@ -384,6 +385,8 @@ struct ceph_inode_info { int i_snap_realm_counter; /* snap realm (if caps) */ struct list_head i_snap_realm_item; struct list_head i_snap_flush_item; + struct timespec64 i_btime; + struct timespec64 i_snap_btime; struct work_struct i_work; unsigned long i_work_mask; @@ -544,7 +547,12 @@ static inline void __ceph_dir_set_complete(struct ceph_inode_info *ci, long long release_count, long long ordered_count) { - smp_mb__before_atomic(); + /* + * Makes sure operations that setup readdir cache (update page + * cache and i_size) are strongly ordered w.r.t. the following + * atomic64_set() operations. + */ + smp_mb(); atomic64_set(&ci->i_complete_seq[0], release_count); atomic64_set(&ci->i_complete_seq[1], ordered_count); } @@ -876,9 +884,8 @@ static inline bool __ceph_have_pending_cap_snap(struct ceph_inode_info *ci) extern const struct inode_operations ceph_file_iops; extern struct inode *ceph_alloc_inode(struct super_block *sb); -extern void ceph_destroy_inode(struct inode *inode); +extern void ceph_evict_inode(struct inode *inode); extern void ceph_free_inode(struct inode *inode); -extern int ceph_drop_inode(struct inode *inode); extern struct inode *ceph_get_inode(struct super_block *sb, struct ceph_vino vino); @@ -921,10 +928,20 @@ ssize_t __ceph_getxattr(struct inode *, const char *, void *, size_t); extern ssize_t ceph_listxattr(struct dentry *, char *, size_t); extern void __ceph_build_xattrs_blob(struct ceph_inode_info *ci); extern void __ceph_destroy_xattrs(struct ceph_inode_info *ci); -extern void __init ceph_xattr_init(void); -extern void ceph_xattr_exit(void); extern const struct xattr_handler *ceph_xattr_handlers[]; +struct ceph_acl_sec_ctx { +#ifdef CONFIG_CEPH_FS_POSIX_ACL + void *default_acl; + void *acl; +#endif +#ifdef CONFIG_CEPH_FS_SECURITY_LABEL + void *sec_ctx; + u32 sec_ctxlen; +#endif + struct ceph_pagelist *pagelist; +}; + #ifdef CONFIG_SECURITY extern bool ceph_security_xattr_deadlock(struct inode *in); extern bool ceph_security_xattr_wanted(struct inode *in); @@ -939,21 +956,32 @@ static inline bool ceph_security_xattr_wanted(struct inode *in) } #endif -/* acl.c */ -struct ceph_acls_info { - void *default_acl; - void *acl; - struct ceph_pagelist *pagelist; -}; +#ifdef CONFIG_CEPH_FS_SECURITY_LABEL +extern int ceph_security_init_secctx(struct dentry *dentry, umode_t mode, + struct ceph_acl_sec_ctx *ctx); +extern void ceph_security_invalidate_secctx(struct inode *inode); +#else +static inline int ceph_security_init_secctx(struct dentry *dentry, umode_t mode, + struct ceph_acl_sec_ctx *ctx) +{ + return 0; +} +static inline void ceph_security_invalidate_secctx(struct inode *inode) +{ +} +#endif + +void ceph_release_acl_sec_ctx(struct ceph_acl_sec_ctx *as_ctx); +/* acl.c */ #ifdef CONFIG_CEPH_FS_POSIX_ACL struct posix_acl *ceph_get_acl(struct inode *, int); int ceph_set_acl(struct inode *inode, struct posix_acl *acl, int type); int ceph_pre_init_acls(struct inode *dir, umode_t *mode, - struct ceph_acls_info *info); -void ceph_init_inode_acls(struct inode *inode, struct ceph_acls_info *info); -void ceph_release_acls_info(struct ceph_acls_info *info); + struct ceph_acl_sec_ctx *as_ctx); +void ceph_init_inode_acls(struct inode *inode, + struct ceph_acl_sec_ctx *as_ctx); static inline void ceph_forget_all_cached_acls(struct inode *inode) { @@ -966,15 +994,12 @@ static inline void ceph_forget_all_cached_acls(struct inode *inode) #define ceph_set_acl NULL static inline int ceph_pre_init_acls(struct inode *dir, umode_t *mode, - struct ceph_acls_info *info) + struct ceph_acl_sec_ctx *as_ctx) { return 0; } static inline void ceph_init_inode_acls(struct inode *inode, - struct ceph_acls_info *info) -{ -} -static inline void ceph_release_acls_info(struct ceph_acls_info *info) + struct ceph_acl_sec_ctx *as_ctx) { } static inline int ceph_acl_chmod(struct dentry *dentry, struct inode *inode) @@ -1000,7 +1025,7 @@ extern void ceph_add_cap(struct inode *inode, unsigned cap, unsigned seq, u64 realmino, int flags, struct ceph_cap **new_cap); extern void __ceph_remove_cap(struct ceph_cap *cap, bool queue_release); -extern void __ceph_remove_caps(struct inode* inode); +extern void __ceph_remove_caps(struct ceph_inode_info *ci); extern void ceph_put_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap); extern int ceph_is_any_caps(struct inode *inode); diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c index 0cc42c8879e9..37b458a9af3a 100644 --- a/fs/ceph/xattr.c +++ b/fs/ceph/xattr.c @@ -8,6 +8,7 @@ #include <linux/ceph/decode.h> #include <linux/xattr.h> +#include <linux/security.h> #include <linux/posix_acl_xattr.h> #include <linux/slab.h> @@ -17,26 +18,9 @@ static int __remove_xattr(struct ceph_inode_info *ci, struct ceph_inode_xattr *xattr); -static const struct xattr_handler ceph_other_xattr_handler; - -/* - * List of handlers for synthetic system.* attributes. Other - * attributes are handled directly. - */ -const struct xattr_handler *ceph_xattr_handlers[] = { -#ifdef CONFIG_CEPH_FS_POSIX_ACL - &posix_acl_access_xattr_handler, - &posix_acl_default_xattr_handler, -#endif - &ceph_other_xattr_handler, - NULL, -}; - static bool ceph_is_valid_xattr(const char *name) { return !strncmp(name, XATTR_CEPH_PREFIX, XATTR_CEPH_PREFIX_LEN) || - !strncmp(name, XATTR_SECURITY_PREFIX, - XATTR_SECURITY_PREFIX_LEN) || !strncmp(name, XATTR_TRUSTED_PREFIX, XATTR_TRUSTED_PREFIX_LEN) || !strncmp(name, XATTR_USER_PREFIX, XATTR_USER_PREFIX_LEN); } @@ -48,8 +32,8 @@ static bool ceph_is_valid_xattr(const char *name) struct ceph_vxattr { char *name; size_t name_size; /* strlen(name) + 1 (for '\0') */ - size_t (*getxattr_cb)(struct ceph_inode_info *ci, char *val, - size_t size); + ssize_t (*getxattr_cb)(struct ceph_inode_info *ci, char *val, + size_t size); bool (*exists_cb)(struct ceph_inode_info *ci); unsigned int flags; }; @@ -68,8 +52,8 @@ static bool ceph_vxattrcb_layout_exists(struct ceph_inode_info *ci) rcu_dereference_raw(fl->pool_ns) != NULL); } -static size_t ceph_vxattrcb_layout(struct ceph_inode_info *ci, char *val, - size_t size) +static ssize_t ceph_vxattrcb_layout(struct ceph_inode_info *ci, char *val, + size_t size) { struct ceph_fs_client *fsc = ceph_sb_to_client(ci->vfs_inode.i_sb); struct ceph_osd_client *osdc = &fsc->client->osdc; @@ -79,7 +63,7 @@ static size_t ceph_vxattrcb_layout(struct ceph_inode_info *ci, char *val, const char *ns_field = " pool_namespace="; char buf[128]; size_t len, total_len = 0; - int ret; + ssize_t ret; pool_ns = ceph_try_get_string(ci->i_layout.pool_ns); @@ -96,18 +80,15 @@ static size_t ceph_vxattrcb_layout(struct ceph_inode_info *ci, char *val, len = snprintf(buf, sizeof(buf), "stripe_unit=%u stripe_count=%u object_size=%u pool=%lld", ci->i_layout.stripe_unit, ci->i_layout.stripe_count, - ci->i_layout.object_size, (unsigned long long)pool); + ci->i_layout.object_size, pool); total_len = len; } if (pool_ns) total_len += strlen(ns_field) + pool_ns->len; - if (!size) { - ret = total_len; - } else if (total_len > size) { - ret = -ERANGE; - } else { + ret = total_len; + if (size >= total_len) { memcpy(val, buf, len); ret = len; if (pool_name) { @@ -128,28 +109,55 @@ static size_t ceph_vxattrcb_layout(struct ceph_inode_info *ci, char *val, return ret; } -static size_t ceph_vxattrcb_layout_stripe_unit(struct ceph_inode_info *ci, - char *val, size_t size) +/* + * The convention with strings in xattrs is that they should not be NULL + * terminated, since we're returning the length with them. snprintf always + * NULL terminates however, so call it on a temporary buffer and then memcpy + * the result into place. + */ +static int ceph_fmt_xattr(char *val, size_t size, const char *fmt, ...) { - return snprintf(val, size, "%u", ci->i_layout.stripe_unit); + int ret; + va_list args; + char buf[96]; /* NB: reevaluate size if new vxattrs are added */ + + va_start(args, fmt); + ret = vsnprintf(buf, size ? sizeof(buf) : 0, fmt, args); + va_end(args); + + /* Sanity check */ + if (size && ret + 1 > sizeof(buf)) { + WARN_ONCE(true, "Returned length too big (%d)", ret); + return -E2BIG; + } + + if (ret <= size) + memcpy(val, buf, ret); + return ret; } -static size_t ceph_vxattrcb_layout_stripe_count(struct ceph_inode_info *ci, +static ssize_t ceph_vxattrcb_layout_stripe_unit(struct ceph_inode_info *ci, char *val, size_t size) { - return snprintf(val, size, "%u", ci->i_layout.stripe_count); + return ceph_fmt_xattr(val, size, "%u", ci->i_layout.stripe_unit); +} + +static ssize_t ceph_vxattrcb_layout_stripe_count(struct ceph_inode_info *ci, + char *val, size_t size) +{ + return ceph_fmt_xattr(val, size, "%u", ci->i_layout.stripe_count); } -static size_t ceph_vxattrcb_layout_object_size(struct ceph_inode_info *ci, - char *val, size_t size) +static ssize_t ceph_vxattrcb_layout_object_size(struct ceph_inode_info *ci, + char *val, size_t size) { - return snprintf(val, size, "%u", ci->i_layout.object_size); + return ceph_fmt_xattr(val, size, "%u", ci->i_layout.object_size); } -static size_t ceph_vxattrcb_layout_pool(struct ceph_inode_info *ci, - char *val, size_t size) +static ssize_t ceph_vxattrcb_layout_pool(struct ceph_inode_info *ci, + char *val, size_t size) { - int ret; + ssize_t ret; struct ceph_fs_client *fsc = ceph_sb_to_client(ci->vfs_inode.i_sb); struct ceph_osd_client *osdc = &fsc->client->osdc; s64 pool = ci->i_layout.pool_id; @@ -157,21 +165,27 @@ static size_t ceph_vxattrcb_layout_pool(struct ceph_inode_info *ci, down_read(&osdc->lock); pool_name = ceph_pg_pool_name_by_id(osdc->osdmap, pool); - if (pool_name) - ret = snprintf(val, size, "%s", pool_name); - else - ret = snprintf(val, size, "%lld", (unsigned long long)pool); + if (pool_name) { + ret = strlen(pool_name); + if (ret <= size) + memcpy(val, pool_name, ret); + } else { + ret = ceph_fmt_xattr(val, size, "%lld", pool); + } up_read(&osdc->lock); return ret; } -static size_t ceph_vxattrcb_layout_pool_namespace(struct ceph_inode_info *ci, - char *val, size_t size) +static ssize_t ceph_vxattrcb_layout_pool_namespace(struct ceph_inode_info *ci, + char *val, size_t size) { - int ret = 0; + ssize_t ret = 0; struct ceph_string *ns = ceph_try_get_string(ci->i_layout.pool_ns); + if (ns) { - ret = snprintf(val, size, "%.*s", (int)ns->len, ns->str); + ret = ns->len; + if (ret <= size) + memcpy(val, ns->str, ret); ceph_put_string(ns); } return ret; @@ -179,53 +193,54 @@ static size_t ceph_vxattrcb_layout_pool_namespace(struct ceph_inode_info *ci, /* directories */ -static size_t ceph_vxattrcb_dir_entries(struct ceph_inode_info *ci, char *val, - size_t size) +static ssize_t ceph_vxattrcb_dir_entries(struct ceph_inode_info *ci, char *val, + size_t size) { - return snprintf(val, size, "%lld", ci->i_files + ci->i_subdirs); + return ceph_fmt_xattr(val, size, "%lld", ci->i_files + ci->i_subdirs); } -static size_t ceph_vxattrcb_dir_files(struct ceph_inode_info *ci, char *val, - size_t size) +static ssize_t ceph_vxattrcb_dir_files(struct ceph_inode_info *ci, char *val, + size_t size) { - return snprintf(val, size, "%lld", ci->i_files); + return ceph_fmt_xattr(val, size, "%lld", ci->i_files); } -static size_t ceph_vxattrcb_dir_subdirs(struct ceph_inode_info *ci, char *val, - size_t size) +static ssize_t ceph_vxattrcb_dir_subdirs(struct ceph_inode_info *ci, char *val, + size_t size) { - return snprintf(val, size, "%lld", ci->i_subdirs); + return ceph_fmt_xattr(val, size, "%lld", ci->i_subdirs); } -static size_t ceph_vxattrcb_dir_rentries(struct ceph_inode_info *ci, char *val, - size_t size) +static ssize_t ceph_vxattrcb_dir_rentries(struct ceph_inode_info *ci, char *val, + size_t size) { - return snprintf(val, size, "%lld", ci->i_rfiles + ci->i_rsubdirs); + return ceph_fmt_xattr(val, size, "%lld", + ci->i_rfiles + ci->i_rsubdirs); } -static size_t ceph_vxattrcb_dir_rfiles(struct ceph_inode_info *ci, char *val, - size_t size) +static ssize_t ceph_vxattrcb_dir_rfiles(struct ceph_inode_info *ci, char *val, + size_t size) { - return snprintf(val, size, "%lld", ci->i_rfiles); + return ceph_fmt_xattr(val, size, "%lld", ci->i_rfiles); } -static size_t ceph_vxattrcb_dir_rsubdirs(struct ceph_inode_info *ci, char *val, - size_t size) +static ssize_t ceph_vxattrcb_dir_rsubdirs(struct ceph_inode_info *ci, char *val, + size_t size) { - return snprintf(val, size, "%lld", ci->i_rsubdirs); + return ceph_fmt_xattr(val, size, "%lld", ci->i_rsubdirs); } -static size_t ceph_vxattrcb_dir_rbytes(struct ceph_inode_info *ci, char *val, - size_t size) +static ssize_t ceph_vxattrcb_dir_rbytes(struct ceph_inode_info *ci, char *val, + size_t size) { - return snprintf(val, size, "%lld", ci->i_rbytes); + return ceph_fmt_xattr(val, size, "%lld", ci->i_rbytes); } -static size_t ceph_vxattrcb_dir_rctime(struct ceph_inode_info *ci, char *val, - size_t size) +static ssize_t ceph_vxattrcb_dir_rctime(struct ceph_inode_info *ci, char *val, + size_t size) { - return snprintf(val, size, "%lld.09%ld", ci->i_rctime.tv_sec, - ci->i_rctime.tv_nsec); + return ceph_fmt_xattr(val, size, "%lld.%09ld", ci->i_rctime.tv_sec, + ci->i_rctime.tv_nsec); } /* dir pin */ @@ -234,10 +249,10 @@ static bool ceph_vxattrcb_dir_pin_exists(struct ceph_inode_info *ci) return ci->i_dir_pin != -ENODATA; } -static size_t ceph_vxattrcb_dir_pin(struct ceph_inode_info *ci, char *val, - size_t size) +static ssize_t ceph_vxattrcb_dir_pin(struct ceph_inode_info *ci, char *val, + size_t size) { - return snprintf(val, size, "%d", (int)ci->i_dir_pin); + return ceph_fmt_xattr(val, size, "%d", (int)ci->i_dir_pin); } /* quotas */ @@ -254,23 +269,36 @@ static bool ceph_vxattrcb_quota_exists(struct ceph_inode_info *ci) return ret; } -static size_t ceph_vxattrcb_quota(struct ceph_inode_info *ci, char *val, - size_t size) +static ssize_t ceph_vxattrcb_quota(struct ceph_inode_info *ci, char *val, + size_t size) +{ + return ceph_fmt_xattr(val, size, "max_bytes=%llu max_files=%llu", + ci->i_max_bytes, ci->i_max_files); +} + +static ssize_t ceph_vxattrcb_quota_max_bytes(struct ceph_inode_info *ci, + char *val, size_t size) { - return snprintf(val, size, "max_bytes=%llu max_files=%llu", - ci->i_max_bytes, ci->i_max_files); + return ceph_fmt_xattr(val, size, "%llu", ci->i_max_bytes); } -static size_t ceph_vxattrcb_quota_max_bytes(struct ceph_inode_info *ci, - char *val, size_t size) +static ssize_t ceph_vxattrcb_quota_max_files(struct ceph_inode_info *ci, + char *val, size_t size) { - return snprintf(val, size, "%llu", ci->i_max_bytes); + return ceph_fmt_xattr(val, size, "%llu", ci->i_max_files); } -static size_t ceph_vxattrcb_quota_max_files(struct ceph_inode_info *ci, - char *val, size_t size) +/* snapshots */ +static bool ceph_vxattrcb_snap_btime_exists(struct ceph_inode_info *ci) { - return snprintf(val, size, "%llu", ci->i_max_files); + return (ci->i_snap_btime.tv_sec != 0 || ci->i_snap_btime.tv_nsec != 0); +} + +static ssize_t ceph_vxattrcb_snap_btime(struct ceph_inode_info *ci, char *val, + size_t size) +{ + return ceph_fmt_xattr(val, size, "%lld.%09ld", ci->i_snap_btime.tv_sec, + ci->i_snap_btime.tv_nsec); } #define CEPH_XATTR_NAME(_type, _name) XATTR_CEPH_PREFIX #_type "." #_name @@ -327,7 +355,7 @@ static struct ceph_vxattr ceph_dir_vxattrs[] = { XATTR_RSTAT_FIELD(dir, rctime), { .name = "ceph.dir.pin", - .name_size = sizeof("ceph.dir_pin"), + .name_size = sizeof("ceph.dir.pin"), .getxattr_cb = ceph_vxattrcb_dir_pin, .exists_cb = ceph_vxattrcb_dir_pin_exists, .flags = VXATTR_FLAG_HIDDEN, @@ -341,9 +369,15 @@ static struct ceph_vxattr ceph_dir_vxattrs[] = { }, XATTR_QUOTA_FIELD(quota, max_bytes), XATTR_QUOTA_FIELD(quota, max_files), + { + .name = "ceph.snap.btime", + .name_size = sizeof("ceph.snap.btime"), + .getxattr_cb = ceph_vxattrcb_snap_btime, + .exists_cb = ceph_vxattrcb_snap_btime_exists, + .flags = VXATTR_FLAG_READONLY, + }, { .name = NULL, 0 } /* Required table terminator */ }; -static size_t ceph_dir_vxattrs_name_size; /* total size of all names */ /* files */ @@ -360,9 +394,15 @@ static struct ceph_vxattr ceph_file_vxattrs[] = { XATTR_LAYOUT_FIELD(file, layout, object_size), XATTR_LAYOUT_FIELD(file, layout, pool), XATTR_LAYOUT_FIELD(file, layout, pool_namespace), + { + .name = "ceph.snap.btime", + .name_size = sizeof("ceph.snap.btime"), + .getxattr_cb = ceph_vxattrcb_snap_btime, + .exists_cb = ceph_vxattrcb_snap_btime_exists, + .flags = VXATTR_FLAG_READONLY, + }, { .name = NULL, 0 } /* Required table terminator */ }; -static size_t ceph_file_vxattrs_name_size; /* total size of all names */ static struct ceph_vxattr *ceph_inode_vxattrs(struct inode *inode) { @@ -373,47 +413,6 @@ static struct ceph_vxattr *ceph_inode_vxattrs(struct inode *inode) return NULL; } -static size_t ceph_vxattrs_name_size(struct ceph_vxattr *vxattrs) -{ - if (vxattrs == ceph_dir_vxattrs) - return ceph_dir_vxattrs_name_size; - if (vxattrs == ceph_file_vxattrs) - return ceph_file_vxattrs_name_size; - BUG_ON(vxattrs); - return 0; -} - -/* - * Compute the aggregate size (including terminating '\0') of all - * virtual extended attribute names in the given vxattr table. - */ -static size_t __init vxattrs_name_size(struct ceph_vxattr *vxattrs) -{ - struct ceph_vxattr *vxattr; - size_t size = 0; - - for (vxattr = vxattrs; vxattr->name; vxattr++) { - if (!(vxattr->flags & VXATTR_FLAG_HIDDEN)) - size += vxattr->name_size; - } - - return size; -} - -/* Routines called at initialization and exit time */ - -void __init ceph_xattr_init(void) -{ - ceph_dir_vxattrs_name_size = vxattrs_name_size(ceph_dir_vxattrs); - ceph_file_vxattrs_name_size = vxattrs_name_size(ceph_file_vxattrs); -} - -void ceph_xattr_exit(void) -{ - ceph_dir_vxattrs_name_size = 0; - ceph_file_vxattrs_name_size = 0; -} - static struct ceph_vxattr *ceph_match_vxattr(struct inode *inode, const char *name) { @@ -523,8 +522,8 @@ static int __set_xattr(struct ceph_inode_info *ci, dout("__set_xattr_val p=%p\n", p); } - dout("__set_xattr_val added %llx.%llx xattr %p %s=%.*s\n", - ceph_vinop(&ci->vfs_inode), xattr, name, val_len, val); + dout("__set_xattr_val added %llx.%llx xattr %p %.*s=%.*s\n", + ceph_vinop(&ci->vfs_inode), xattr, name_len, name, val_len, val); return 0; } @@ -823,7 +822,7 @@ ssize_t __ceph_getxattr(struct inode *inode, const char *name, void *value, struct ceph_inode_xattr *xattr; struct ceph_vxattr *vxattr = NULL; int req_mask; - int err; + ssize_t err; /* let's see if a virtual xattr was requested */ vxattr = ceph_match_vxattr(inode, name); @@ -835,8 +834,11 @@ ssize_t __ceph_getxattr(struct inode *inode, const char *name, void *value, if (err) return err; err = -ENODATA; - if (!(vxattr->exists_cb && !vxattr->exists_cb(ci))) + if (!(vxattr->exists_cb && !vxattr->exists_cb(ci))) { err = vxattr->getxattr_cb(ci, value, size); + if (size && size < err) + err = -ERANGE; + } return err; } @@ -897,10 +899,9 @@ ssize_t ceph_listxattr(struct dentry *dentry, char *names, size_t size) struct inode *inode = d_inode(dentry); struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_vxattr *vxattrs = ceph_inode_vxattrs(inode); - u32 vir_namelen = 0; + bool len_only = (size == 0); u32 namelen; int err; - u32 len; int i; spin_lock(&ci->i_ceph_lock); @@ -919,38 +920,45 @@ ssize_t ceph_listxattr(struct dentry *dentry, char *names, size_t size) err = __build_xattrs(inode); if (err < 0) goto out; - /* - * Start with virtual dir xattr names (if any) (including - * terminating '\0' characters for each). - */ - vir_namelen = ceph_vxattrs_name_size(vxattrs); - /* adding 1 byte per each variable due to the null termination */ + /* add 1 byte for each xattr due to the null termination */ namelen = ci->i_xattrs.names_size + ci->i_xattrs.count; - err = -ERANGE; - if (size && vir_namelen + namelen > size) - goto out; - - err = namelen + vir_namelen; - if (size == 0) - goto out; + if (!len_only) { + if (namelen > size) { + err = -ERANGE; + goto out; + } + names = __copy_xattr_names(ci, names); + size -= namelen; + } - names = __copy_xattr_names(ci, names); /* virtual xattr names, too */ - err = namelen; if (vxattrs) { for (i = 0; vxattrs[i].name; i++) { - if (!(vxattrs[i].flags & VXATTR_FLAG_HIDDEN) && - !(vxattrs[i].exists_cb && - !vxattrs[i].exists_cb(ci))) { - len = sprintf(names, "%s", vxattrs[i].name); - names += len + 1; - err += len + 1; + size_t this_len; + + if (vxattrs[i].flags & VXATTR_FLAG_HIDDEN) + continue; + if (vxattrs[i].exists_cb && !vxattrs[i].exists_cb(ci)) + continue; + + this_len = strlen(vxattrs[i].name) + 1; + namelen += this_len; + if (len_only) + continue; + + if (this_len > size) { + err = -ERANGE; + goto out; } + + memcpy(names, vxattrs[i].name, this_len); + names += this_len; + size -= this_len; } } - + err = namelen; out: spin_unlock(&ci->i_ceph_lock); return err; @@ -1206,4 +1214,138 @@ bool ceph_security_xattr_deadlock(struct inode *in) spin_unlock(&ci->i_ceph_lock); return ret; } + +#ifdef CONFIG_CEPH_FS_SECURITY_LABEL +int ceph_security_init_secctx(struct dentry *dentry, umode_t mode, + struct ceph_acl_sec_ctx *as_ctx) +{ + struct ceph_pagelist *pagelist = as_ctx->pagelist; + const char *name; + size_t name_len; + int err; + + err = security_dentry_init_security(dentry, mode, &dentry->d_name, + &as_ctx->sec_ctx, + &as_ctx->sec_ctxlen); + if (err < 0) { + WARN_ON_ONCE(err != -EOPNOTSUPP); + err = 0; /* do nothing */ + goto out; + } + + err = -ENOMEM; + if (!pagelist) { + pagelist = ceph_pagelist_alloc(GFP_KERNEL); + if (!pagelist) + goto out; + err = ceph_pagelist_reserve(pagelist, PAGE_SIZE); + if (err) + goto out; + ceph_pagelist_encode_32(pagelist, 1); + } + + /* + * FIXME: Make security_dentry_init_security() generic. Currently + * It only supports single security module and only selinux has + * dentry_init_security hook. + */ + name = XATTR_NAME_SELINUX; + name_len = strlen(name); + err = ceph_pagelist_reserve(pagelist, + 4 * 2 + name_len + as_ctx->sec_ctxlen); + if (err) + goto out; + + if (as_ctx->pagelist) { + /* update count of KV pairs */ + BUG_ON(pagelist->length <= sizeof(__le32)); + if (list_is_singular(&pagelist->head)) { + le32_add_cpu((__le32*)pagelist->mapped_tail, 1); + } else { + struct page *page = list_first_entry(&pagelist->head, + struct page, lru); + void *addr = kmap_atomic(page); + le32_add_cpu((__le32*)addr, 1); + kunmap_atomic(addr); + } + } else { + as_ctx->pagelist = pagelist; + } + + ceph_pagelist_encode_32(pagelist, name_len); + ceph_pagelist_append(pagelist, name, name_len); + + ceph_pagelist_encode_32(pagelist, as_ctx->sec_ctxlen); + ceph_pagelist_append(pagelist, as_ctx->sec_ctx, as_ctx->sec_ctxlen); + + err = 0; +out: + if (pagelist && !as_ctx->pagelist) + ceph_pagelist_release(pagelist); + return err; +} + +void ceph_security_invalidate_secctx(struct inode *inode) +{ + security_inode_invalidate_secctx(inode); +} + +static int ceph_xattr_set_security_label(const struct xattr_handler *handler, + struct dentry *unused, struct inode *inode, + const char *key, const void *buf, + size_t buflen, int flags) +{ + if (security_ismaclabel(key)) { + const char *name = xattr_full_name(handler, key); + return __ceph_setxattr(inode, name, buf, buflen, flags); + } + return -EOPNOTSUPP; +} + +static int ceph_xattr_get_security_label(const struct xattr_handler *handler, + struct dentry *unused, struct inode *inode, + const char *key, void *buf, size_t buflen) +{ + if (security_ismaclabel(key)) { + const char *name = xattr_full_name(handler, key); + return __ceph_getxattr(inode, name, buf, buflen); + } + return -EOPNOTSUPP; +} + +static const struct xattr_handler ceph_security_label_handler = { + .prefix = XATTR_SECURITY_PREFIX, + .get = ceph_xattr_get_security_label, + .set = ceph_xattr_set_security_label, +}; +#endif #endif + +void ceph_release_acl_sec_ctx(struct ceph_acl_sec_ctx *as_ctx) +{ +#ifdef CONFIG_CEPH_FS_POSIX_ACL + posix_acl_release(as_ctx->acl); + posix_acl_release(as_ctx->default_acl); +#endif +#ifdef CONFIG_CEPH_FS_SECURITY_LABEL + security_release_secctx(as_ctx->sec_ctx, as_ctx->sec_ctxlen); +#endif + if (as_ctx->pagelist) + ceph_pagelist_release(as_ctx->pagelist); +} + +/* + * List of handlers for synthetic system.* attributes. Other + * attributes are handled directly. + */ +const struct xattr_handler *ceph_xattr_handlers[] = { +#ifdef CONFIG_CEPH_FS_POSIX_ACL + &posix_acl_access_xattr_handler, + &posix_acl_default_xattr_handler, +#endif +#ifdef CONFIG_CEPH_FS_SECURITY_LABEL + &ceph_security_label_handler, +#endif + &ceph_other_xattr_handler, + NULL, +}; diff --git a/fs/cifs/Kconfig b/fs/cifs/Kconfig index 523e9ea78a28..b16219e5dac9 100644 --- a/fs/cifs/Kconfig +++ b/fs/cifs/Kconfig @@ -13,9 +13,11 @@ config CIFS select CRYPTO_LIB_ARC4 select CRYPTO_AEAD2 select CRYPTO_CCM + select CRYPTO_GCM select CRYPTO_ECB select CRYPTO_AES select CRYPTO_DES + select KEYS help This is the client VFS module for the SMB3 family of NAS protocols, (including support for the most recent, most secure dialect SMB3.1.1) @@ -109,7 +111,7 @@ config CIFS_WEAK_PW_HASH config CIFS_UPCALL bool "Kerberos/SPNEGO advanced session setup" - depends on CIFS && KEYS + depends on CIFS select DNS_RESOLVER help Enables an upcall mechanism for CIFS which accesses userspace helper @@ -144,14 +146,6 @@ config CIFS_POSIX (such as Samba 3.10 and later) which can negotiate CIFS POSIX ACL support. If unsure, say N. -config CIFS_ACL - bool "Provide CIFS ACL support" - depends on CIFS_XATTR && KEYS - help - Allows fetching CIFS/NTFS ACL from the server. The DACL blob - is handed over to the application/caller. See the man - page for getcifsacl for more information. If unsure, say Y. - config CIFS_DEBUG bool "Enable CIFS debugging routines" default y @@ -184,7 +178,7 @@ config CIFS_DEBUG_DUMP_KEYS config CIFS_DFS_UPCALL bool "DFS feature support" - depends on CIFS && KEYS + depends on CIFS select DNS_RESOLVER help Distributed File System (DFS) support is used to access shares @@ -203,10 +197,10 @@ config CIFS_NFSD_EXPORT Allows NFS server to export a CIFS mounted share (nfsd over cifs) config CIFS_SMB_DIRECT - bool "SMB Direct support (Experimental)" + bool "SMB Direct support" depends on CIFS=m && INFINIBAND && INFINIBAND_ADDR_TRANS || CIFS=y && INFINIBAND=y && INFINIBAND_ADDR_TRANS=y help - Enables SMB Direct experimental support for SMB 3.0, 3.02 and 3.1.1. + Enables SMB Direct support for SMB 3.0, 3.02 and 3.1.1. SMB Direct allows transferring SMB packets over RDMA. If unsure, say N. diff --git a/fs/cifs/Makefile b/fs/cifs/Makefile index 51af69a1a328..41332f20055b 100644 --- a/fs/cifs/Makefile +++ b/fs/cifs/Makefile @@ -10,10 +10,9 @@ cifs-y := trace.o cifsfs.o cifssmb.o cifs_debug.o connect.o dir.o file.o \ cifs_unicode.o nterr.o cifsencrypt.o \ readdir.o ioctl.o sess.o export.o smb1ops.o winucase.o \ smb2ops.o smb2maperror.o smb2transport.o \ - smb2misc.o smb2pdu.o smb2inode.o smb2file.o + smb2misc.o smb2pdu.o smb2inode.o smb2file.o cifsacl.o cifs-$(CONFIG_CIFS_XATTR) += xattr.o -cifs-$(CONFIG_CIFS_ACL) += cifsacl.o cifs-$(CONFIG_CIFS_UPCALL) += cifs_spnego.o diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index ec933fb0b36e..a38d796f5ffe 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -240,9 +240,7 @@ static int cifs_debug_data_proc_show(struct seq_file *m, void *v) #ifdef CONFIG_CIFS_XATTR seq_printf(m, ",XATTR"); #endif -#ifdef CONFIG_CIFS_ACL seq_printf(m, ",ACL"); -#endif seq_putc(m, '\n'); seq_printf(m, "CIFSMaxBufSize: %d\n", CIFSMaxBufSize); seq_printf(m, "Active VFS Requests: %d\n", GlobalTotalActiveXid); diff --git a/fs/cifs/cifs_fs_sb.h b/fs/cifs/cifs_fs_sb.h index ed49222abecb..b326d2ca3765 100644 --- a/fs/cifs/cifs_fs_sb.h +++ b/fs/cifs/cifs_fs_sb.h @@ -52,6 +52,7 @@ #define CIFS_MOUNT_UID_FROM_ACL 0x2000000 /* try to get UID via special SID */ #define CIFS_MOUNT_NO_HANDLE_CACHE 0x4000000 /* disable caching dir handles */ #define CIFS_MOUNT_NO_DFS 0x8000000 /* disable DFS resolving */ +#define CIFS_MOUNT_MODE_FROM_SID 0x10000000 /* retrieve mode from special ACE */ struct cifs_sb_info { struct rb_root tlink_tree; @@ -83,5 +84,10 @@ struct cifs_sb_info { * failover properly. */ char *origin_fullpath; /* \\HOST\SHARE\[OPTIONAL PATH] */ + /* + * Indicate whether serverino option was turned off later + * (cifs_autodisable_serverino) in order to match new mounts. + */ + bool mnt_cifs_serverino_autodisabled; }; #endif /* _CIFS_FS_SB_H */ diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 24635b65effa..270d3c58fb3b 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -526,6 +526,8 @@ cifs_show_options(struct seq_file *s, struct dentry *root) seq_puts(s, ",nobrl"); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_HANDLE_CACHE) seq_puts(s, ",nohandlecache"); + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MODE_FROM_SID) + seq_puts(s, ",modefromsid"); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) seq_puts(s, ",cifsacl"); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DYNPERM) @@ -554,6 +556,11 @@ cifs_show_options(struct seq_file *s, struct dentry *root) seq_printf(s, ",bsize=%u", cifs_sb->bsize); seq_printf(s, ",echo_interval=%lu", tcon->ses->server->echo_interval / HZ); + + /* Only display max_credits if it was overridden on mount */ + if (tcon->ses->server->max_credits != SMB2_MAX_CREDITS_AVAILABLE) + seq_printf(s, ",max_credits=%u", tcon->ses->server->max_credits); + if (tcon->snapshot_time) seq_printf(s, ",snapshot=%llu", tcon->snapshot_time); if (tcon->handle_timeout) @@ -1517,11 +1524,9 @@ init_cifs(void) goto out_destroy_dfs_cache; #endif /* CONFIG_CIFS_UPCALL */ -#ifdef CONFIG_CIFS_ACL rc = init_cifs_idmap(); if (rc) goto out_register_key_type; -#endif /* CONFIG_CIFS_ACL */ rc = register_filesystem(&cifs_fs_type); if (rc) @@ -1536,10 +1541,8 @@ init_cifs(void) return 0; out_init_cifs_idmap: -#ifdef CONFIG_CIFS_ACL exit_cifs_idmap(); out_register_key_type: -#endif #ifdef CONFIG_CIFS_UPCALL exit_cifs_spnego(); out_destroy_dfs_cache: @@ -1571,9 +1574,7 @@ exit_cifs(void) unregister_filesystem(&cifs_fs_type); unregister_filesystem(&smb3_fs_type); cifs_dfs_release_automount_timer(); -#ifdef CONFIG_CIFS_ACL exit_cifs_idmap(); -#endif #ifdef CONFIG_CIFS_UPCALL exit_cifs_spnego(); #endif @@ -1607,5 +1608,6 @@ MODULE_SOFTDEP("pre: sha256"); MODULE_SOFTDEP("pre: sha512"); MODULE_SOFTDEP("pre: aead2"); MODULE_SOFTDEP("pre: ccm"); +MODULE_SOFTDEP("pre: gcm"); module_init(init_cifs) module_exit(exit_cifs) diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 4777b3c4a92c..fe610e7e3670 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -550,6 +550,7 @@ struct smb_vol { bool override_gid:1; bool dynperm:1; bool noperm:1; + bool mode_ace:1; bool no_psx_acl:1; /* set if posix acl support should be disabled */ bool cifs_acl:1; bool backupuid_specified; /* mount option backupuid is specified */ @@ -600,6 +601,7 @@ struct smb_vol { __u64 snapshot_time; /* needed for timewarp tokens */ __u32 handle_timeout; /* persistent and durable handle timeout in ms */ unsigned int max_credits; /* smb3 max_credits 10 < credits < 60000 */ + __u16 compression; /* compression algorithm 0xFFFF default 0=disabled */ }; /** @@ -617,7 +619,8 @@ struct smb_vol { CIFS_MOUNT_FSCACHE | CIFS_MOUNT_MF_SYMLINKS | \ CIFS_MOUNT_MULTIUSER | CIFS_MOUNT_STRICT_IO | \ CIFS_MOUNT_CIFS_BACKUPUID | CIFS_MOUNT_CIFS_BACKUPGID | \ - CIFS_MOUNT_NO_DFS) + CIFS_MOUNT_UID_FROM_ACL | CIFS_MOUNT_NO_HANDLE_CACHE | \ + CIFS_MOUNT_NO_DFS | CIFS_MOUNT_MODE_FROM_SID) /** * Generic VFS superblock mount flags (s_flags) to consider when @@ -1870,7 +1873,6 @@ extern unsigned int cifs_min_small; /* min size of small buf pool */ extern unsigned int cifs_max_pending; /* MAX requests at once to server*/ extern bool disable_legacy_dialects; /* forbid vers=1.0 and vers=2.0 mounts */ -#ifdef CONFIG_CIFS_ACL GLOBAL_EXTERN struct rb_root uidtree; GLOBAL_EXTERN struct rb_root gidtree; GLOBAL_EXTERN spinlock_t siduidlock; @@ -1879,7 +1881,6 @@ GLOBAL_EXTERN struct rb_root siduidtree; GLOBAL_EXTERN struct rb_root sidgidtree; GLOBAL_EXTERN spinlock_t uidsidlock; GLOBAL_EXTERN spinlock_t gidsidlock; -#endif /* CONFIG_CIFS_ACL */ void cifs_oplock_break(struct work_struct *work); void cifs_queue_oplock_break(struct cifsFileInfo *cfile); diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 1fbd92843a73..e2f95965065d 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -3600,11 +3600,9 @@ static int cifs_copy_posix_acl(char *trgt, char *src, const int buflen, return size; } -static __u16 convert_ace_to_cifs_ace(struct cifs_posix_ace *cifs_ace, +static void convert_ace_to_cifs_ace(struct cifs_posix_ace *cifs_ace, const struct posix_acl_xattr_entry *local_ace) { - __u16 rc = 0; /* 0 = ACL converted ok */ - cifs_ace->cifs_e_perm = le16_to_cpu(local_ace->e_perm); cifs_ace->cifs_e_tag = le16_to_cpu(local_ace->e_tag); /* BB is there a better way to handle the large uid? */ @@ -3617,7 +3615,6 @@ static __u16 convert_ace_to_cifs_ace(struct cifs_posix_ace *cifs_ace, cifs_dbg(FYI, "perm %d tag %d id %d\n", ace->e_perm, ace->e_tag, ace->e_id); */ - return rc; } /* Convert ACL from local Linux POSIX xattr to CIFS POSIX ACL wire format */ @@ -3653,13 +3650,8 @@ static __u16 ACL_to_cifs_posix(char *parm_data, const char *pACL, cifs_dbg(FYI, "unknown ACL type %d\n", acl_type); return 0; } - for (i = 0; i < count; i++) { - rc = convert_ace_to_cifs_ace(&cifs_acl->ace_array[i], &ace[i]); - if (rc != 0) { - /* ACE not converted */ - break; - } - } + for (i = 0; i < count; i++) + convert_ace_to_cifs_ace(&cifs_acl->ace_array[i], &ace[i]); if (rc == 0) { rc = (__u16)(count * sizeof(struct cifs_posix_ace)); rc += sizeof(struct cifs_posix_acl); @@ -3920,7 +3912,6 @@ GetExtAttrOut: #endif /* CONFIG_POSIX */ -#ifdef CONFIG_CIFS_ACL /* * Initialize NT TRANSACT SMB into small smb request buffer. This assumes that * all NT TRANSACTS that we init here have total parm and data under about 400 @@ -4164,7 +4155,6 @@ setCifsAclRetry: return (rc); } -#endif /* CONFIG_CIFS_ACL */ /* Legacy Query Path Information call for lookup to old servers such as Win9x/WinME */ diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 714a359c7c8d..a4830ced0f98 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -96,7 +96,8 @@ enum { Opt_multiuser, Opt_sloppy, Opt_nosharesock, Opt_persistent, Opt_nopersistent, Opt_resilient, Opt_noresilient, - Opt_domainauto, Opt_rdma, + Opt_domainauto, Opt_rdma, Opt_modesid, + Opt_compress, /* Mount options which take numeric value */ Opt_backupuid, Opt_backupgid, Opt_uid, @@ -175,6 +176,7 @@ static const match_table_t cifs_mount_option_tokens = { { Opt_serverino, "serverino" }, { Opt_noserverino, "noserverino" }, { Opt_rwpidforward, "rwpidforward" }, + { Opt_modesid, "modefromsid" }, { Opt_cifsacl, "cifsacl" }, { Opt_nocifsacl, "nocifsacl" }, { Opt_acl, "acl" }, @@ -212,6 +214,7 @@ static const match_table_t cifs_mount_option_tokens = { { Opt_echo_interval, "echo_interval=%s" }, { Opt_max_credits, "max_credits=%s" }, { Opt_snapshot, "snapshot=%s" }, + { Opt_compress, "compress=%s" }, { Opt_blank_user, "user=" }, { Opt_blank_user, "username=" }, @@ -706,10 +709,10 @@ static bool server_unresponsive(struct TCP_Server_Info *server) { /* - * We need to wait 2 echo intervals to make sure we handle such + * We need to wait 3 echo intervals to make sure we handle such * situations right: * 1s client sends a normal SMB request - * 2s client gets a response + * 3s client gets a response * 30s echo workqueue job pops, and decides we got a response recently * and don't need to send another * ... @@ -718,9 +721,9 @@ server_unresponsive(struct TCP_Server_Info *server) */ if ((server->tcpStatus == CifsGood || server->tcpStatus == CifsNeedNegotiate) && - time_after(jiffies, server->lstrp + 2 * server->echo_interval)) { + time_after(jiffies, server->lstrp + 3 * server->echo_interval)) { cifs_dbg(VFS, "Server %s has not responded in %lu seconds. Reconnecting...\n", - server->hostname, (2 * server->echo_interval) / HZ); + server->hostname, (3 * server->echo_interval) / HZ); cifs_reconnect(server); wake_up(&server->response_q); return true; @@ -1223,11 +1226,11 @@ next_pdu: atomic_read(&midCount)); cifs_dump_mem("Received Data is: ", bufs[i], HEADER_SIZE(server)); + smb2_add_credits_from_hdr(bufs[i], server); #ifdef CONFIG_CIFS_DEBUG2 if (server->ops->dump_detail) server->ops->dump_detail(bufs[i], server); - smb2_add_credits_from_hdr(bufs[i], server); cifs_dump_mids(server); #endif /* CIFS_DEBUG2 */ } @@ -1830,6 +1833,9 @@ cifs_parse_mount_options(const char *mountdata, const char *devname, case Opt_rwpidforward: vol->rwpidforward = 1; break; + case Opt_modesid: + vol->mode_ace = 1; + break; case Opt_cifsacl: vol->cifs_acl = 1; break; @@ -1911,6 +1917,11 @@ cifs_parse_mount_options(const char *mountdata, const char *devname, case Opt_rdma: vol->rdma = true; break; + case Opt_compress: + vol->compression = UNKNOWN_TYPE; + cifs_dbg(VFS, + "SMB3 compression support is experimental\n"); + break; /* Numeric Values */ case Opt_backupuid: @@ -2544,8 +2555,15 @@ static int match_server(struct TCP_Server_Info *server, struct smb_vol *vol) if (vol->nosharesock) return 0; - /* BB update this for smb3any and default case */ - if ((server->vals != vol->vals) || (server->ops != vol->ops)) + /* If multidialect negotiation see if existing sessions match one */ + if (strcmp(vol->vals->version_string, SMB3ANY_VERSION_STRING) == 0) { + if (server->vals->protocol_id < SMB30_PROT_ID) + return 0; + } else if (strcmp(vol->vals->version_string, + SMBDEFAULT_VERSION_STRING) == 0) { + if (server->vals->protocol_id < SMB21_PROT_ID) + return 0; + } else if ((server->vals != vol->vals) || (server->ops != vol->ops)) return 0; if (!net_eq(cifs_net_ns(server), current->nsproxy->net_ns)) @@ -2680,6 +2698,7 @@ cifs_get_tcp_session(struct smb_vol *volume_info) tcp_ses->sequence_number = 0; tcp_ses->reconnect_instance = 1; tcp_ses->lstrp = jiffies; + tcp_ses->compress_algorithm = cpu_to_le16(volume_info->compression); spin_lock_init(&tcp_ses->req_lock); INIT_LIST_HEAD(&tcp_ses->tcp_ses_list); INIT_LIST_HEAD(&tcp_ses->smb_ses_list); @@ -3460,12 +3479,16 @@ compare_mount_options(struct super_block *sb, struct cifs_mnt_data *mnt_data) { struct cifs_sb_info *old = CIFS_SB(sb); struct cifs_sb_info *new = mnt_data->cifs_sb; + unsigned int oldflags = old->mnt_cifs_flags & CIFS_MOUNT_MASK; + unsigned int newflags = new->mnt_cifs_flags & CIFS_MOUNT_MASK; if ((sb->s_flags & CIFS_MS_MASK) != (mnt_data->flags & CIFS_MS_MASK)) return 0; - if ((old->mnt_cifs_flags & CIFS_MOUNT_MASK) != - (new->mnt_cifs_flags & CIFS_MOUNT_MASK)) + if (old->mnt_cifs_serverino_autodisabled) + newflags &= ~CIFS_MOUNT_SERVER_INUM; + + if (oldflags != newflags) return 0; /* @@ -3965,6 +3988,8 @@ int cifs_setup_cifs_sb(struct smb_vol *pvolume_info, cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_NOPOSIXBRL; if (pvolume_info->rwpidforward) cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_RWPIDFORWARD; + if (pvolume_info->mode_ace) + cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_MODE_FROM_SID; if (pvolume_info->cifs_acl) cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_CIFS_ACL; if (pvolume_info->backupuid_specified) { @@ -4459,11 +4484,13 @@ cifs_are_all_path_components_accessible(struct TCP_Server_Info *server, unsigned int xid, struct cifs_tcon *tcon, struct cifs_sb_info *cifs_sb, - char *full_path) + char *full_path, + int added_treename) { int rc; char *s; char sep, tmp; + int skip = added_treename ? 1 : 0; sep = CIFS_DIR_SEP(cifs_sb); s = full_path; @@ -4478,7 +4505,14 @@ cifs_are_all_path_components_accessible(struct TCP_Server_Info *server, /* next separator */ while (*s && *s != sep) s++; - + /* + * if the treename is added, we then have to skip the first + * part within the separators + */ + if (skip) { + skip = 0; + continue; + } /* * temporarily null-terminate the path at the end of * the current component @@ -4526,8 +4560,7 @@ static int is_path_remote(struct cifs_sb_info *cifs_sb, struct smb_vol *vol, if (rc != -EREMOTE) { rc = cifs_are_all_path_components_accessible(server, xid, tcon, - cifs_sb, - full_path); + cifs_sb, full_path, tcon->Flags & SMB_SHARE_IS_IN_DFS); if (rc != 0) { cifs_dbg(VFS, "cannot query dirs between root and final path, " "enabling CIFS_MOUNT_USE_PREFIX_PATH\n"); diff --git a/fs/cifs/dfs_cache.c b/fs/cifs/dfs_cache.c index e3e1c13df439..1692c0c6c23a 100644 --- a/fs/cifs/dfs_cache.c +++ b/fs/cifs/dfs_cache.c @@ -492,7 +492,7 @@ static struct dfs_cache_entry *__find_cache_entry(unsigned int hash, #ifdef CONFIG_CIFS_DEBUG2 char *name = get_tgt_name(ce); - if (unlikely(IS_ERR(name))) { + if (IS_ERR(name)) { rcu_read_unlock(); return ERR_CAST(name); } diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index d7cc62252634..1bffe029fb66 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -892,7 +892,6 @@ cifs_get_inode_info(struct inode **inode, const char *full_path, cifs_dbg(FYI, "cifs_sfu_type failed: %d\n", tmprc); } -#ifdef CONFIG_CIFS_ACL /* fill in 0777 bits from ACL */ if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) { rc = cifs_acl_to_fattr(cifs_sb, &fattr, *inode, full_path, fid); @@ -902,7 +901,6 @@ cifs_get_inode_info(struct inode **inode, const char *full_path, goto cgii_exit; } } -#endif /* CONFIG_CIFS_ACL */ /* fill in remaining high mode bits e.g. SUID, VTX */ if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL) @@ -2415,7 +2413,7 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) xid = get_xid(); - cifs_dbg(FYI, "setattr on file %pd attrs->iavalid 0x%x\n", + cifs_dbg(FYI, "setattr on file %pd attrs->ia_valid 0x%x\n", direntry, attrs->ia_valid); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_PERM) @@ -2466,7 +2464,6 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) if (attrs->ia_valid & ATTR_GID) gid = attrs->ia_gid; -#ifdef CONFIG_CIFS_ACL if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) { if (uid_valid(uid) || gid_valid(gid)) { rc = id_mode_to_cifs_acl(inode, full_path, NO_CHANGE_64, @@ -2478,7 +2475,6 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) } } } else -#endif /* CONFIG_CIFS_ACL */ if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID)) attrs->ia_valid &= ~(ATTR_UID | ATTR_GID); @@ -2489,7 +2485,6 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) if (attrs->ia_valid & ATTR_MODE) { mode = attrs->ia_mode; rc = 0; -#ifdef CONFIG_CIFS_ACL if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) { rc = id_mode_to_cifs_acl(inode, full_path, mode, INVALID_UID, INVALID_GID); @@ -2499,7 +2494,6 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) goto cifs_setattr_exit; } } else -#endif /* CONFIG_CIFS_ACL */ if (((mode & S_IWUGO) == 0) && (cifsInode->cifsAttrs & ATTR_READONLY) == 0) { diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index b1a696a73f7c..f383877a6511 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -539,6 +539,7 @@ cifs_autodisable_serverino(struct cifs_sb_info *cifs_sb) tcon = cifs_sb_master_tcon(cifs_sb); cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_SERVER_INUM; + cifs_sb->mnt_cifs_serverino_autodisabled = true; cifs_dbg(VFS, "Autodisabling the use of server inode numbers on %s.\n", tcon ? tcon->treeName : "new server"); cifs_dbg(VFS, "The server doesn't seem to support them properly or the files might be on different servers (DFS).\n"); diff --git a/fs/cifs/smb1ops.c b/fs/cifs/smb1ops.c index 9e430ae9314f..b7421a096319 100644 --- a/fs/cifs/smb1ops.c +++ b/fs/cifs/smb1ops.c @@ -1223,16 +1223,15 @@ struct smb_version_operations smb1_operations = { .query_all_EAs = CIFSSMBQAllEAs, .set_EA = CIFSSMBSetEA, #endif /* CIFS_XATTR */ -#ifdef CONFIG_CIFS_ACL .get_acl = get_cifs_acl, .get_acl_by_fid = get_cifs_acl_by_fid, .set_acl = set_cifs_acl, -#endif /* CIFS_ACL */ .make_node = cifs_make_node, }; struct smb_version_values smb1_values = { .version_string = SMB1_VERSION_STRING, + .protocol_id = SMB10_PROT_ID, .large_lock_type = LOCKING_ANDX_LARGE_FILES, .exclusive_lock_type = 0, .shared_lock_type = LOCKING_ANDX_SHARED_LOCK, diff --git a/fs/cifs/smb2inode.c b/fs/cifs/smb2inode.c index 278405d26c47..d8d9cdfa30b6 100644 --- a/fs/cifs/smb2inode.c +++ b/fs/cifs/smb2inode.c @@ -120,6 +120,8 @@ smb2_compound_op(const unsigned int xid, struct cifs_tcon *tcon, SMB2_O_INFO_FILE, 0, sizeof(struct smb2_file_all_info) + PATH_MAX * 2, 0, NULL); + if (rc) + goto finished; smb2_set_next_command(tcon, &rqst[num_rqst]); smb2_set_related(&rqst[num_rqst++]); trace_smb3_query_info_compound_enter(xid, ses->Suid, tcon->tid, @@ -147,6 +149,8 @@ smb2_compound_op(const unsigned int xid, struct cifs_tcon *tcon, COMPOUND_FID, current->tgid, FILE_DISPOSITION_INFORMATION, SMB2_O_INFO_FILE, 0, data, size); + if (rc) + goto finished; smb2_set_next_command(tcon, &rqst[num_rqst]); smb2_set_related(&rqst[num_rqst++]); trace_smb3_rmdir_enter(xid, ses->Suid, tcon->tid, full_path); @@ -163,6 +167,8 @@ smb2_compound_op(const unsigned int xid, struct cifs_tcon *tcon, COMPOUND_FID, current->tgid, FILE_END_OF_FILE_INFORMATION, SMB2_O_INFO_FILE, 0, data, size); + if (rc) + goto finished; smb2_set_next_command(tcon, &rqst[num_rqst]); smb2_set_related(&rqst[num_rqst++]); trace_smb3_set_eof_enter(xid, ses->Suid, tcon->tid, full_path); @@ -180,6 +186,8 @@ smb2_compound_op(const unsigned int xid, struct cifs_tcon *tcon, COMPOUND_FID, current->tgid, FILE_BASIC_INFORMATION, SMB2_O_INFO_FILE, 0, data, size); + if (rc) + goto finished; smb2_set_next_command(tcon, &rqst[num_rqst]); smb2_set_related(&rqst[num_rqst++]); trace_smb3_set_info_compound_enter(xid, ses->Suid, tcon->tid, @@ -206,6 +214,8 @@ smb2_compound_op(const unsigned int xid, struct cifs_tcon *tcon, COMPOUND_FID, current->tgid, FILE_RENAME_INFORMATION, SMB2_O_INFO_FILE, 0, data, size); + if (rc) + goto finished; smb2_set_next_command(tcon, &rqst[num_rqst]); smb2_set_related(&rqst[num_rqst++]); trace_smb3_rename_enter(xid, ses->Suid, tcon->tid, full_path); @@ -231,6 +241,8 @@ smb2_compound_op(const unsigned int xid, struct cifs_tcon *tcon, COMPOUND_FID, current->tgid, FILE_LINK_INFORMATION, SMB2_O_INFO_FILE, 0, data, size); + if (rc) + goto finished; smb2_set_next_command(tcon, &rqst[num_rqst]); smb2_set_related(&rqst[num_rqst++]); trace_smb3_hardlink_enter(xid, ses->Suid, tcon->tid, full_path); diff --git a/fs/cifs/smb2ops.c b/fs/cifs/smb2ops.c index 9fd56b0acd7e..0cdc4e47ca87 100644 --- a/fs/cifs/smb2ops.c +++ b/fs/cifs/smb2ops.c @@ -2027,6 +2027,10 @@ smb2_set_related(struct smb_rqst *rqst) struct smb2_sync_hdr *shdr; shdr = (struct smb2_sync_hdr *)(rqst->rq_iov[0].iov_base); + if (shdr == NULL) { + cifs_dbg(FYI, "shdr NULL in smb2_set_related\n"); + return; + } shdr->Flags |= SMB2_FLAGS_RELATED_OPERATIONS; } @@ -2041,6 +2045,12 @@ smb2_set_next_command(struct cifs_tcon *tcon, struct smb_rqst *rqst) unsigned long len = smb_rqst_len(server, rqst); int i, num_padding; + shdr = (struct smb2_sync_hdr *)(rqst->rq_iov[0].iov_base); + if (shdr == NULL) { + cifs_dbg(FYI, "shdr NULL in smb2_set_next_command\n"); + return; + } + /* SMB headers in a compound are 8 byte aligned. */ /* No padding needed */ @@ -2080,7 +2090,6 @@ smb2_set_next_command(struct cifs_tcon *tcon, struct smb_rqst *rqst) } finished: - shdr = (struct smb2_sync_hdr *)(rqst->rq_iov[0].iov_base); shdr->NextCommand = cpu_to_le32(len); } @@ -2374,6 +2383,34 @@ smb2_get_dfs_refer(const unsigned int xid, struct cifs_ses *ses, } static int +parse_reparse_posix(struct reparse_posix_data *symlink_buf, + u32 plen, char **target_path, + struct cifs_sb_info *cifs_sb) +{ + unsigned int len; + + /* See MS-FSCC 2.1.2.6 for the 'NFS' style reparse tags */ + len = le16_to_cpu(symlink_buf->ReparseDataLength); + + if (le64_to_cpu(symlink_buf->InodeType) != NFS_SPECFILE_LNK) { + cifs_dbg(VFS, "%lld not a supported symlink type\n", + le64_to_cpu(symlink_buf->InodeType)); + return -EOPNOTSUPP; + } + + *target_path = cifs_strndup_from_utf16( + symlink_buf->PathBuffer, + len, true, cifs_sb->local_nls); + if (!(*target_path)) + return -ENOMEM; + + convert_delimiter(*target_path, '/'); + cifs_dbg(FYI, "%s: target path: %s\n", __func__, *target_path); + + return 0; +} + +static int parse_reparse_symlink(struct reparse_symlink_data_buffer *symlink_buf, u32 plen, char **target_path, struct cifs_sb_info *cifs_sb) @@ -2381,11 +2418,7 @@ parse_reparse_symlink(struct reparse_symlink_data_buffer *symlink_buf, unsigned int sub_len; unsigned int sub_offset; - /* We only handle Symbolic Link : MS-FSCC 2.1.2.4 */ - if (le32_to_cpu(symlink_buf->ReparseTag) != IO_REPARSE_TAG_SYMLINK) { - cifs_dbg(VFS, "srv returned invalid symlink buffer\n"); - return -EIO; - } + /* We handle Symbolic Link reparse tag here. See: MS-FSCC 2.1.2.4 */ sub_offset = le16_to_cpu(symlink_buf->SubstituteNameOffset); sub_len = le16_to_cpu(symlink_buf->SubstituteNameLength); @@ -2407,6 +2440,41 @@ parse_reparse_symlink(struct reparse_symlink_data_buffer *symlink_buf, return 0; } +static int +parse_reparse_point(struct reparse_data_buffer *buf, + u32 plen, char **target_path, + struct cifs_sb_info *cifs_sb) +{ + if (plen < sizeof(struct reparse_data_buffer)) { + cifs_dbg(VFS, "reparse buffer is too small. Must be " + "at least 8 bytes but was %d\n", plen); + return -EIO; + } + + if (plen < le16_to_cpu(buf->ReparseDataLength) + + sizeof(struct reparse_data_buffer)) { + cifs_dbg(VFS, "srv returned invalid reparse buf " + "length: %d\n", plen); + return -EIO; + } + + /* See MS-FSCC 2.1.2 */ + switch (le32_to_cpu(buf->ReparseTag)) { + case IO_REPARSE_TAG_NFS: + return parse_reparse_posix( + (struct reparse_posix_data *)buf, + plen, target_path, cifs_sb); + case IO_REPARSE_TAG_SYMLINK: + return parse_reparse_symlink( + (struct reparse_symlink_data_buffer *)buf, + plen, target_path, cifs_sb); + default: + cifs_dbg(VFS, "srv returned unknown symlink buffer " + "tag:0x%08x\n", le32_to_cpu(buf->ReparseTag)); + return -EOPNOTSUPP; + } +} + #define SMB2_SYMLINK_STRUCT_SIZE \ (sizeof(struct smb2_err_rsp) - 1 + sizeof(struct smb2_symlink_err_rsp)) @@ -2533,23 +2601,8 @@ smb2_query_symlink(const unsigned int xid, struct cifs_tcon *tcon, goto querty_exit; } - if (plen < 8) { - cifs_dbg(VFS, "reparse buffer is too small. Must be " - "at least 8 bytes but was %d\n", plen); - rc = -EIO; - goto querty_exit; - } - - if (plen < le16_to_cpu(reparse_buf->ReparseDataLength) + 8) { - cifs_dbg(VFS, "srv returned invalid reparse buf " - "length: %d\n", plen); - rc = -EIO; - goto querty_exit; - } - - rc = parse_reparse_symlink( - (struct reparse_symlink_data_buffer *)reparse_buf, - plen, target_path, cifs_sb); + rc = parse_reparse_point(reparse_buf, plen, target_path, + cifs_sb); goto querty_exit; } @@ -2561,26 +2614,32 @@ smb2_query_symlink(const unsigned int xid, struct cifs_tcon *tcon, err_buf = err_iov.iov_base; if (le32_to_cpu(err_buf->ByteCount) < sizeof(struct smb2_symlink_err_rsp) || err_iov.iov_len < SMB2_SYMLINK_STRUCT_SIZE) { - rc = -ENOENT; + rc = -EINVAL; + goto querty_exit; + } + + symlink = (struct smb2_symlink_err_rsp *)err_buf->ErrorData; + if (le32_to_cpu(symlink->SymLinkErrorTag) != SYMLINK_ERROR_TAG || + le32_to_cpu(symlink->ReparseTag) != IO_REPARSE_TAG_SYMLINK) { + rc = -EINVAL; goto querty_exit; } /* open must fail on symlink - reset rc */ rc = 0; - symlink = (struct smb2_symlink_err_rsp *)err_buf->ErrorData; sub_len = le16_to_cpu(symlink->SubstituteNameLength); sub_offset = le16_to_cpu(symlink->SubstituteNameOffset); print_len = le16_to_cpu(symlink->PrintNameLength); print_offset = le16_to_cpu(symlink->PrintNameOffset); if (err_iov.iov_len < SMB2_SYMLINK_STRUCT_SIZE + sub_offset + sub_len) { - rc = -ENOENT; + rc = -EINVAL; goto querty_exit; } if (err_iov.iov_len < SMB2_SYMLINK_STRUCT_SIZE + print_offset + print_len) { - rc = -ENOENT; + rc = -EINVAL; goto querty_exit; } @@ -2606,7 +2665,6 @@ smb2_query_symlink(const unsigned int xid, struct cifs_tcon *tcon, return rc; } -#ifdef CONFIG_CIFS_ACL static struct cifs_ntsd * get_smb2_acl_by_fid(struct cifs_sb_info *cifs_sb, const struct cifs_fid *cifsfid, u32 *pacllen) @@ -2691,7 +2749,6 @@ get_smb2_acl_by_path(struct cifs_sb_info *cifs_sb, return pntsd; } -#ifdef CONFIG_CIFS_ACL static int set_smb2_acl(struct cifs_ntsd *pnntsd, __u32 acllen, struct inode *inode, const char *path, int aclflag) @@ -2749,7 +2806,6 @@ set_smb2_acl(struct cifs_ntsd *pnntsd, __u32 acllen, free_xid(xid); return rc; } -#endif /* CIFS_ACL */ /* Retrieve an ACL from the server */ static struct cifs_ntsd * @@ -2769,7 +2825,6 @@ get_smb2_acl(struct cifs_sb_info *cifs_sb, cifsFileInfo_put(open_file); return pntsd; } -#endif static long smb3_zero_range(struct file *file, struct cifs_tcon *tcon, loff_t offset, loff_t len, bool keep_size) @@ -3367,7 +3422,7 @@ smb2_dir_needs_close(struct cifsFileInfo *cfile) static void fill_transform_hdr(struct smb2_transform_hdr *tr_hdr, unsigned int orig_len, - struct smb_rqst *old_rq) + struct smb_rqst *old_rq, __le16 cipher_type) { struct smb2_sync_hdr *shdr = (struct smb2_sync_hdr *)old_rq->rq_iov[0].iov_base; @@ -3376,7 +3431,10 @@ fill_transform_hdr(struct smb2_transform_hdr *tr_hdr, unsigned int orig_len, tr_hdr->ProtocolId = SMB2_TRANSFORM_PROTO_NUM; tr_hdr->OriginalMessageSize = cpu_to_le32(orig_len); tr_hdr->Flags = cpu_to_le16(0x01); - get_random_bytes(&tr_hdr->Nonce, SMB3_AES128CMM_NONCE); + if (cipher_type == SMB2_ENCRYPTION_AES128_GCM) + get_random_bytes(&tr_hdr->Nonce, SMB3_AES128GCM_NONCE); + else + get_random_bytes(&tr_hdr->Nonce, SMB3_AES128CCM_NONCE); memcpy(&tr_hdr->SessionId, &shdr->SessionId, 8); } @@ -3534,8 +3592,13 @@ crypt_message(struct TCP_Server_Info *server, int num_rqst, rc = -ENOMEM; goto free_sg; } - iv[0] = 3; - memcpy(iv + 1, (char *)tr_hdr->Nonce, SMB3_AES128CMM_NONCE); + + if (server->cipher_type == SMB2_ENCRYPTION_AES128_GCM) + memcpy(iv, (char *)tr_hdr->Nonce, SMB3_AES128GCM_NONCE); + else { + iv[0] = 3; + memcpy(iv + 1, (char *)tr_hdr->Nonce, SMB3_AES128CCM_NONCE); + } aead_request_set_crypt(req, sg, sg, crypt_len, iv); aead_request_set_ad(req, assoc_data_len); @@ -3635,7 +3698,7 @@ smb3_init_transform_rq(struct TCP_Server_Info *server, int num_rqst, } /* fill the 1st iov with a transform header */ - fill_transform_hdr(tr_hdr, orig_len, old_rq); + fill_transform_hdr(tr_hdr, orig_len, old_rq, server->cipher_type); rc = crypt_message(server, num_rqst, new_rq, 1); cifs_dbg(FYI, "Encrypt message returned %d\n", rc); @@ -4284,11 +4347,9 @@ struct smb_version_operations smb20_operations = { .query_all_EAs = smb2_query_eas, .set_EA = smb2_set_ea, #endif /* CIFS_XATTR */ -#ifdef CONFIG_CIFS_ACL .get_acl = get_smb2_acl, .get_acl_by_fid = get_smb2_acl_by_fid, .set_acl = set_smb2_acl, -#endif /* CIFS_ACL */ .next_header = smb2_next_header, .ioctl_query_info = smb2_ioctl_query_info, .make_node = smb2_make_node, @@ -4385,11 +4446,9 @@ struct smb_version_operations smb21_operations = { .query_all_EAs = smb2_query_eas, .set_EA = smb2_set_ea, #endif /* CIFS_XATTR */ -#ifdef CONFIG_CIFS_ACL .get_acl = get_smb2_acl, .get_acl_by_fid = get_smb2_acl_by_fid, .set_acl = set_smb2_acl, -#endif /* CIFS_ACL */ .next_header = smb2_next_header, .ioctl_query_info = smb2_ioctl_query_info, .make_node = smb2_make_node, @@ -4495,11 +4554,9 @@ struct smb_version_operations smb30_operations = { .query_all_EAs = smb2_query_eas, .set_EA = smb2_set_ea, #endif /* CIFS_XATTR */ -#ifdef CONFIG_CIFS_ACL .get_acl = get_smb2_acl, .get_acl_by_fid = get_smb2_acl_by_fid, .set_acl = set_smb2_acl, -#endif /* CIFS_ACL */ .next_header = smb2_next_header, .ioctl_query_info = smb2_ioctl_query_info, .make_node = smb2_make_node, @@ -4606,11 +4663,9 @@ struct smb_version_operations smb311_operations = { .query_all_EAs = smb2_query_eas, .set_EA = smb2_set_ea, #endif /* CIFS_XATTR */ -#ifdef CONFIG_CIFS_ACL .get_acl = get_smb2_acl, .get_acl_by_fid = get_smb2_acl_by_fid, .set_acl = set_smb2_acl, -#endif /* CIFS_ACL */ .next_header = smb2_next_header, .ioctl_query_info = smb2_ioctl_query_info, .make_node = smb2_make_node, diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c index 75311a8a68bf..f58e4dc3987b 100644 --- a/fs/cifs/smb2pdu.c +++ b/fs/cifs/smb2pdu.c @@ -489,10 +489,25 @@ static void build_encrypt_ctxt(struct smb2_encryption_neg_context *pneg_ctxt) { pneg_ctxt->ContextType = SMB2_ENCRYPTION_CAPABILITIES; - pneg_ctxt->DataLength = cpu_to_le16(4); /* Cipher Count + le16 cipher */ - pneg_ctxt->CipherCount = cpu_to_le16(1); -/* pneg_ctxt->Ciphers[0] = SMB2_ENCRYPTION_AES128_GCM;*/ /* not supported yet */ - pneg_ctxt->Ciphers[0] = SMB2_ENCRYPTION_AES128_CCM; + pneg_ctxt->DataLength = cpu_to_le16(6); /* Cipher Count + two ciphers */ + pneg_ctxt->CipherCount = cpu_to_le16(2); + pneg_ctxt->Ciphers[0] = SMB2_ENCRYPTION_AES128_GCM; + pneg_ctxt->Ciphers[1] = SMB2_ENCRYPTION_AES128_CCM; +} + +static unsigned int +build_netname_ctxt(struct smb2_netname_neg_context *pneg_ctxt, char *hostname) +{ + struct nls_table *cp = load_nls_default(); + + pneg_ctxt->ContextType = SMB2_NETNAME_NEGOTIATE_CONTEXT_ID; + + /* copy up to max of first 100 bytes of server name to NetName field */ + pneg_ctxt->DataLength = cpu_to_le16(2 + + (2 * cifs_strtoUTF16(pneg_ctxt->NetName, hostname, 100, cp))); + /* context size is DataLength + minimal smb2_neg_context */ + return DIV_ROUND_UP(le16_to_cpu(pneg_ctxt->DataLength) + + sizeof(struct smb2_neg_context), 8) * 8; } static void @@ -521,7 +536,7 @@ build_posix_ctxt(struct smb2_posix_neg_context *pneg_ctxt) static void assemble_neg_contexts(struct smb2_negotiate_req *req, - unsigned int *total_len) + struct TCP_Server_Info *server, unsigned int *total_len) { char *pneg_ctxt = (char *)req; unsigned int ctxt_len; @@ -551,17 +566,25 @@ assemble_neg_contexts(struct smb2_negotiate_req *req, *total_len += ctxt_len; pneg_ctxt += ctxt_len; - build_compression_ctxt((struct smb2_compression_capabilities_context *) + if (server->compress_algorithm) { + build_compression_ctxt((struct smb2_compression_capabilities_context *) pneg_ctxt); - ctxt_len = DIV_ROUND_UP( - sizeof(struct smb2_compression_capabilities_context), 8) * 8; + ctxt_len = DIV_ROUND_UP( + sizeof(struct smb2_compression_capabilities_context), + 8) * 8; + *total_len += ctxt_len; + pneg_ctxt += ctxt_len; + req->NegotiateContextCount = cpu_to_le16(5); + } else + req->NegotiateContextCount = cpu_to_le16(4); + + ctxt_len = build_netname_ctxt((struct smb2_netname_neg_context *)pneg_ctxt, + server->hostname); *total_len += ctxt_len; pneg_ctxt += ctxt_len; build_posix_ctxt((struct smb2_posix_neg_context *)pneg_ctxt); *total_len += sizeof(struct smb2_posix_neg_context); - - req->NegotiateContextCount = cpu_to_le16(4); } static void decode_preauth_context(struct smb2_preauth_neg_context *ctxt) @@ -829,7 +852,7 @@ SMB2_negotiate(const unsigned int xid, struct cifs_ses *ses) if ((ses->server->vals->protocol_id == SMB311_PROT_ID) || (strcmp(ses->server->vals->version_string, SMBDEFAULT_VERSION_STRING) == 0)) - assemble_neg_contexts(req, &total_len); + assemble_neg_contexts(req, server, &total_len); } iov[0].iov_base = (char *)req; iov[0].iov_len = total_len; @@ -2095,6 +2118,48 @@ add_twarp_context(struct kvec *iov, unsigned int *num_iovec, __u64 timewarp) return 0; } +static struct crt_query_id_ctxt * +create_query_id_buf(void) +{ + struct crt_query_id_ctxt *buf; + + buf = kzalloc(sizeof(struct crt_query_id_ctxt), GFP_KERNEL); + if (!buf) + return NULL; + + buf->ccontext.DataOffset = cpu_to_le16(0); + buf->ccontext.DataLength = cpu_to_le32(0); + buf->ccontext.NameOffset = cpu_to_le16(offsetof + (struct crt_query_id_ctxt, Name)); + buf->ccontext.NameLength = cpu_to_le16(4); + /* SMB2_CREATE_QUERY_ON_DISK_ID is "QFid" */ + buf->Name[0] = 'Q'; + buf->Name[1] = 'F'; + buf->Name[2] = 'i'; + buf->Name[3] = 'd'; + return buf; +} + +/* See MS-SMB2 2.2.13.2.9 */ +static int +add_query_id_context(struct kvec *iov, unsigned int *num_iovec) +{ + struct smb2_create_req *req = iov[0].iov_base; + unsigned int num = *num_iovec; + + iov[num].iov_base = create_query_id_buf(); + if (iov[num].iov_base == NULL) + return -ENOMEM; + iov[num].iov_len = sizeof(struct crt_query_id_ctxt); + if (!req->CreateContextsOffset) + req->CreateContextsOffset = cpu_to_le32( + sizeof(struct smb2_create_req) + + iov[num - 1].iov_len); + le32_add_cpu(&req->CreateContextsLength, sizeof(struct crt_query_id_ctxt)); + *num_iovec = num + 1; + return 0; +} + static int alloc_path_with_tree_prefix(__le16 **out_path, int *out_size, int *out_len, const char *treename, const __le16 *path) @@ -2423,6 +2488,12 @@ SMB2_open_init(struct cifs_tcon *tcon, struct smb_rqst *rqst, __u8 *oplock, return rc; } + if (n_iov > 2) { + struct create_context *ccontext = + (struct create_context *)iov[n_iov-1].iov_base; + ccontext->Next = cpu_to_le32(iov[n_iov-1].iov_len); + } + add_query_id_context(iov, &n_iov); rqst->rq_nvec = n_iov; return 0; @@ -2550,12 +2621,11 @@ SMB2_ioctl_init(struct cifs_tcon *tcon, struct smb_rqst *rqst, * indatalen is usually small at a couple of bytes max, so * just allocate through generic pool */ - in_data_buf = kmalloc(indatalen, GFP_NOFS); + in_data_buf = kmemdup(in_data, indatalen, GFP_NOFS); if (!in_data_buf) { cifs_small_buf_release(req); return -ENOMEM; } - memcpy(in_data_buf, in_data, indatalen); } req->CtlCode = cpu_to_le32(opcode); diff --git a/fs/cifs/smb2pdu.h b/fs/cifs/smb2pdu.h index 858353d20c39..7e2e782f8edd 100644 --- a/fs/cifs/smb2pdu.h +++ b/fs/cifs/smb2pdu.h @@ -123,7 +123,7 @@ struct smb2_sync_pdu { __le16 StructureSize2; /* size of wct area (varies, request specific) */ } __packed; -#define SMB3_AES128CMM_NONCE 11 +#define SMB3_AES128CCM_NONCE 11 #define SMB3_AES128GCM_NONCE 12 struct smb2_transform_hdr { @@ -166,6 +166,8 @@ struct smb2_err_rsp { __u8 ErrorData[1]; /* variable length */ } __packed; +#define SYMLINK_ERROR_TAG 0x4c4d5953 + struct smb2_symlink_err_rsp { __le32 SymLinkLength; __le32 SymLinkErrorTag; @@ -227,6 +229,7 @@ struct smb2_negotiate_req { } __packed; /* Dialects */ +#define SMB10_PROT_ID 0x0000 /* local only, not sent on wire w/CIFS negprot */ #define SMB20_PROT_ID 0x0202 #define SMB21_PROT_ID 0x0210 #define SMB30_PROT_ID 0x0300 @@ -293,7 +296,7 @@ struct smb2_encryption_neg_context { __le16 DataLength; __le32 Reserved; __le16 CipherCount; /* AES-128-GCM and AES-128-CCM */ - __le16 Ciphers[1]; /* Ciphers[0] since only one used now */ + __le16 Ciphers[2]; } __packed; /* See MS-SMB2 2.2.3.1.3 */ @@ -316,6 +319,12 @@ struct smb2_compression_capabilities_context { * For smb2_netname_negotiate_context_id See MS-SMB2 2.2.3.1.4. * Its struct simply contains NetName, an array of Unicode characters */ +struct smb2_netname_neg_context { + __le16 ContextType; /* 0x100 */ + __le16 DataLength; + __le32 Reserved; + __le16 NetName[0]; /* hostname of target converted to UCS-2 */ +} __packed; #define POSIX_CTXT_DATA_LEN 16 struct smb2_posix_neg_context { @@ -640,6 +649,7 @@ struct smb2_tree_disconnect_rsp { #define SMB2_CREATE_DURABLE_HANDLE_REQUEST_V2 "DH2Q" #define SMB2_CREATE_DURABLE_HANDLE_RECONNECT_V2 "DH2C" #define SMB2_CREATE_APP_INSTANCE_ID 0x45BCA66AEFA7F74A9008FA462E144D74 +#define SMB2_CREATE_APP_INSTANCE_VERSION 0xB982D0B73B56074FA07B524A8116A010 #define SVHDX_OPEN_DEVICE_CONTEX 0x9CCBCF9E04C1E643980E158DA1F6EC83 #define SMB2_CREATE_TAG_POSIX 0x93AD25509CB411E7B42383DE968BCD7C @@ -654,9 +664,10 @@ struct smb2_tree_disconnect_rsp { * [3] : durable context * [4] : posix context * [5] : time warp context - * [6] : compound padding + * [6] : query id context + * [7] : compound padding */ -#define SMB2_CREATE_IOV_SIZE 7 +#define SMB2_CREATE_IOV_SIZE 8 struct smb2_create_req { struct smb2_sync_hdr sync_hdr; @@ -680,10 +691,10 @@ struct smb2_create_req { /* * Maximum size of a SMB2_CREATE response is 64 (smb2 header) + - * 88 (fixed part of create response) + 520 (path) + 150 (contexts) + + * 88 (fixed part of create response) + 520 (path) + 208 (contexts) + * 2 bytes of padding. */ -#define MAX_SMB2_CREATE_RESPONSE_SIZE 824 +#define MAX_SMB2_CREATE_RESPONSE_SIZE 880 struct smb2_create_rsp { struct smb2_sync_hdr sync_hdr; @@ -806,6 +817,13 @@ struct durable_reconnect_context_v2 { __le32 Flags; /* see above DHANDLE_FLAG_PERSISTENT */ } __packed; +/* See MS-SMB2 2.2.14.2.9 */ +struct on_disk_id { + __le64 DiskFileId; + __le64 VolumeId; + __u32 Reserved[4]; +} __packed; + /* See MS-SMB2 2.2.14.2.12 */ struct durable_reconnect_context_v2_rsp { __le32 Timeout; @@ -826,6 +844,12 @@ struct crt_twarp_ctxt { } __packed; +/* See MS-SMB2 2.2.13.2.9 */ +struct crt_query_id_ctxt { + struct create_context ccontext; + __u8 Name[8]; +} __packed; + #define COPY_CHUNK_RES_KEY_SIZE 24 struct resume_key_req { char ResumeKey[COPY_CHUNK_RES_KEY_SIZE]; diff --git a/fs/cifs/smb2transport.c b/fs/cifs/smb2transport.c index d1181572758b..1ccbcf9c2c3b 100644 --- a/fs/cifs/smb2transport.c +++ b/fs/cifs/smb2transport.c @@ -734,7 +734,10 @@ smb3_crypto_aead_allocate(struct TCP_Server_Info *server) struct crypto_aead *tfm; if (!server->secmech.ccmaesencrypt) { - tfm = crypto_alloc_aead("ccm(aes)", 0, 0); + if (server->cipher_type == SMB2_ENCRYPTION_AES128_GCM) + tfm = crypto_alloc_aead("gcm(aes)", 0, 0); + else + tfm = crypto_alloc_aead("ccm(aes)", 0, 0); if (IS_ERR(tfm)) { cifs_dbg(VFS, "%s: Failed to alloc encrypt aead\n", __func__); @@ -744,7 +747,10 @@ smb3_crypto_aead_allocate(struct TCP_Server_Info *server) } if (!server->secmech.ccmaesdecrypt) { - tfm = crypto_alloc_aead("ccm(aes)", 0, 0); + if (server->cipher_type == SMB2_ENCRYPTION_AES128_GCM) + tfm = crypto_alloc_aead("gcm(aes)", 0, 0); + else + tfm = crypto_alloc_aead("ccm(aes)", 0, 0); if (IS_ERR(tfm)) { crypto_free_aead(server->secmech.ccmaesencrypt); server->secmech.ccmaesencrypt = NULL; diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 60661b3f983a..5d6d44bfe10a 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -979,6 +979,7 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses, }; unsigned int instance; char *buf; + struct TCP_Server_Info *server; optype = flags & CIFS_OP_MASK; @@ -990,7 +991,8 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses, return -EIO; } - if (ses->server->tcpStatus == CifsExiting) + server = ses->server; + if (server->tcpStatus == CifsExiting) return -ENOENT; /* @@ -1001,7 +1003,7 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses, * other requests. * This can be handled by the eventual session reconnect. */ - rc = wait_for_compound_request(ses->server, num_rqst, flags, + rc = wait_for_compound_request(server, num_rqst, flags, &instance); if (rc) return rc; @@ -1017,7 +1019,7 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses, * of smb data. */ - mutex_lock(&ses->server->srv_mutex); + mutex_lock(&server->srv_mutex); /* * All the parts of the compound chain belong obtained credits from the @@ -1026,24 +1028,24 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses, * we obtained credits and return -EAGAIN in such cases to let callers * handle it. */ - if (instance != ses->server->reconnect_instance) { - mutex_unlock(&ses->server->srv_mutex); + if (instance != server->reconnect_instance) { + mutex_unlock(&server->srv_mutex); for (j = 0; j < num_rqst; j++) - add_credits(ses->server, &credits[j], optype); + add_credits(server, &credits[j], optype); return -EAGAIN; } for (i = 0; i < num_rqst; i++) { - midQ[i] = ses->server->ops->setup_request(ses, &rqst[i]); + midQ[i] = server->ops->setup_request(ses, &rqst[i]); if (IS_ERR(midQ[i])) { - revert_current_mid(ses->server, i); + revert_current_mid(server, i); for (j = 0; j < i; j++) cifs_delete_mid(midQ[j]); - mutex_unlock(&ses->server->srv_mutex); + mutex_unlock(&server->srv_mutex); /* Update # of requests on wire to server */ for (j = 0; j < num_rqst; j++) - add_credits(ses->server, &credits[j], optype); + add_credits(server, &credits[j], optype); return PTR_ERR(midQ[i]); } @@ -1059,19 +1061,19 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses, else midQ[i]->callback = cifs_compound_last_callback; } - cifs_in_send_inc(ses->server); - rc = smb_send_rqst(ses->server, num_rqst, rqst, flags); - cifs_in_send_dec(ses->server); + cifs_in_send_inc(server); + rc = smb_send_rqst(server, num_rqst, rqst, flags); + cifs_in_send_dec(server); for (i = 0; i < num_rqst; i++) cifs_save_when_sent(midQ[i]); if (rc < 0) { - revert_current_mid(ses->server, num_rqst); - ses->server->sequence_number -= 2; + revert_current_mid(server, num_rqst); + server->sequence_number -= 2; } - mutex_unlock(&ses->server->srv_mutex); + mutex_unlock(&server->srv_mutex); /* * If sending failed for some reason or it is an oplock break that we @@ -1079,7 +1081,7 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses, */ if (rc < 0 || (flags & CIFS_NO_SRV_RSP)) { for (i = 0; i < num_rqst; i++) - add_credits(ses->server, &credits[i], optype); + add_credits(server, &credits[i], optype); goto out; } @@ -1099,7 +1101,7 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses, rqst[0].rq_nvec); for (i = 0; i < num_rqst; i++) { - rc = wait_for_response(ses->server, midQ[i]); + rc = wait_for_response(server, midQ[i]); if (rc != 0) break; } @@ -1107,7 +1109,7 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses, for (; i < num_rqst; i++) { cifs_dbg(VFS, "Cancelling wait for mid %llu cmd: %d\n", midQ[i]->mid, le16_to_cpu(midQ[i]->command)); - send_cancel(ses->server, &rqst[i], midQ[i]); + send_cancel(server, &rqst[i], midQ[i]); spin_lock(&GlobalMid_Lock); if (midQ[i]->mid_state == MID_REQUEST_SUBMITTED) { midQ[i]->mid_flags |= MID_WAIT_CANCELLED; @@ -1123,7 +1125,7 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses, if (rc < 0) goto out; - rc = cifs_sync_mid_result(midQ[i], ses->server); + rc = cifs_sync_mid_result(midQ[i], server); if (rc != 0) { /* mark this mid as cancelled to not free it below */ cancelled_mid[i] = true; @@ -1140,14 +1142,14 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses, buf = (char *)midQ[i]->resp_buf; resp_iov[i].iov_base = buf; resp_iov[i].iov_len = midQ[i]->resp_buf_size + - ses->server->vals->header_preamble_size; + server->vals->header_preamble_size; if (midQ[i]->large_buf) resp_buf_type[i] = CIFS_LARGE_BUFFER; else resp_buf_type[i] = CIFS_SMALL_BUFFER; - rc = ses->server->ops->check_receive(midQ[i], ses->server, + rc = server->ops->check_receive(midQ[i], server, flags & CIFS_LOG_ERROR); /* mark it so buf will not be freed by cifs_delete_mid */ diff --git a/fs/cifs/xattr.c b/fs/cifs/xattr.c index 50ddb795aaeb..9076150758d8 100644 --- a/fs/cifs/xattr.c +++ b/fs/cifs/xattr.c @@ -96,7 +96,6 @@ static int cifs_xattr_set(const struct xattr_handler *handler, break; case XATTR_CIFS_ACL: { -#ifdef CONFIG_CIFS_ACL struct cifs_ntsd *pacl; if (!value) @@ -117,7 +116,6 @@ static int cifs_xattr_set(const struct xattr_handler *handler, CIFS_I(inode)->time = 0; kfree(pacl); } -#endif /* CONFIG_CIFS_ACL */ break; } @@ -247,7 +245,6 @@ static int cifs_xattr_get(const struct xattr_handler *handler, break; case XATTR_CIFS_ACL: { -#ifdef CONFIG_CIFS_ACL u32 acllen; struct cifs_ntsd *pacl; @@ -270,7 +267,6 @@ static int cifs_xattr_get(const struct xattr_handler *handler, rc = acllen; kfree(pacl); } -#endif /* CONFIG_CIFS_ACL */ break; } @@ -124,6 +124,15 @@ static int dax_is_empty_entry(void *entry) } /* + * true if the entry that was found is of a smaller order than the entry + * we were looking for + */ +static bool dax_is_conflict(void *entry) +{ + return entry == XA_RETRY_ENTRY; +} + +/* * DAX page cache entry locking */ struct exceptional_entry_key { @@ -195,11 +204,13 @@ static void dax_wake_entry(struct xa_state *xas, void *entry, bool wake_all) * Look up entry in page cache, wait for it to become unlocked if it * is a DAX entry and return it. The caller must subsequently call * put_unlocked_entry() if it did not lock the entry or dax_unlock_entry() - * if it did. + * if it did. The entry returned may have a larger order than @order. + * If @order is larger than the order of the entry found in i_pages, this + * function returns a dax_is_conflict entry. * * Must be called with the i_pages lock held. */ -static void *get_unlocked_entry(struct xa_state *xas) +static void *get_unlocked_entry(struct xa_state *xas, unsigned int order) { void *entry; struct wait_exceptional_entry_queue ewait; @@ -210,6 +221,8 @@ static void *get_unlocked_entry(struct xa_state *xas) for (;;) { entry = xas_find_conflict(xas); + if (dax_entry_order(entry) < order) + return XA_RETRY_ENTRY; if (!entry || WARN_ON_ONCE(!xa_is_value(entry)) || !dax_is_locked(entry)) return entry; @@ -254,7 +267,7 @@ static void wait_entry_unlocked(struct xa_state *xas, void *entry) static void put_unlocked_entry(struct xa_state *xas, void *entry) { /* If we were the only waiter woken, wake the next one */ - if (entry) + if (entry && dax_is_conflict(entry)) dax_wake_entry(xas, entry, false); } @@ -461,7 +474,7 @@ void dax_unlock_page(struct page *page, dax_entry_t cookie) * overlap with xarray value entries. */ static void *grab_mapping_entry(struct xa_state *xas, - struct address_space *mapping, unsigned long size_flag) + struct address_space *mapping, unsigned int order) { unsigned long index = xas->xa_index; bool pmd_downgrade = false; /* splitting PMD entry into PTE entries? */ @@ -469,20 +482,17 @@ static void *grab_mapping_entry(struct xa_state *xas, retry: xas_lock_irq(xas); - entry = get_unlocked_entry(xas); + entry = get_unlocked_entry(xas, order); if (entry) { + if (dax_is_conflict(entry)) + goto fallback; if (!xa_is_value(entry)) { xas_set_err(xas, EIO); goto out_unlock; } - if (size_flag & DAX_PMD) { - if (dax_is_pte_entry(entry)) { - put_unlocked_entry(xas, entry); - goto fallback; - } - } else { /* trying to grab a PTE entry */ + if (order == 0) { if (dax_is_pmd_entry(entry) && (dax_is_zero_entry(entry) || dax_is_empty_entry(entry))) { @@ -523,7 +533,11 @@ retry: if (entry) { dax_lock_entry(xas, entry); } else { - entry = dax_make_entry(pfn_to_pfn_t(0), size_flag | DAX_EMPTY); + unsigned long flags = DAX_EMPTY; + + if (order > 0) + flags |= DAX_PMD; + entry = dax_make_entry(pfn_to_pfn_t(0), flags); dax_lock_entry(xas, entry); if (xas_error(xas)) goto out_unlock; @@ -594,7 +608,7 @@ struct page *dax_layout_busy_page(struct address_space *mapping) if (WARN_ON_ONCE(!xa_is_value(entry))) continue; if (unlikely(dax_is_locked(entry))) - entry = get_unlocked_entry(&xas); + entry = get_unlocked_entry(&xas, 0); if (entry) page = dax_busy_page(entry); put_unlocked_entry(&xas, entry); @@ -621,7 +635,7 @@ static int __dax_invalidate_entry(struct address_space *mapping, void *entry; xas_lock_irq(&xas); - entry = get_unlocked_entry(&xas); + entry = get_unlocked_entry(&xas, 0); if (!entry || WARN_ON_ONCE(!xa_is_value(entry))) goto out; if (!trunc && @@ -848,7 +862,7 @@ static int dax_writeback_one(struct xa_state *xas, struct dax_device *dax_dev, if (unlikely(dax_is_locked(entry))) { void *old_entry = entry; - entry = get_unlocked_entry(xas); + entry = get_unlocked_entry(xas, 0); /* Entry got punched out / reallocated? */ if (!entry || WARN_ON_ONCE(!xa_is_value(entry))) @@ -1509,7 +1523,7 @@ static vm_fault_t dax_iomap_pmd_fault(struct vm_fault *vmf, pfn_t *pfnp, * entry is already in the array, for instance), it will return * VM_FAULT_FALLBACK. */ - entry = grab_mapping_entry(&xas, mapping, DAX_PMD); + entry = grab_mapping_entry(&xas, mapping, PMD_ORDER); if (xa_is_internal(entry)) { result = xa_to_internal(entry); goto fallback; @@ -1658,11 +1672,10 @@ dax_insert_pfn_mkwrite(struct vm_fault *vmf, pfn_t pfn, unsigned int order) vm_fault_t ret; xas_lock_irq(&xas); - entry = get_unlocked_entry(&xas); + entry = get_unlocked_entry(&xas, order); /* Did we race with someone splitting entry or so? */ - if (!entry || - (order == 0 && !dax_is_pte_entry(entry)) || - (order == PMD_ORDER && !dax_is_pmd_entry(entry))) { + if (!entry || dax_is_conflict(entry) || + (order == 0 && !dax_is_pte_entry(entry))) { put_unlocked_entry(&xas, entry); xas_unlock_irq(&xas); trace_dax_insert_pfn_mkwrite_no_entry(mapping->host, vmf, diff --git a/fs/ext4/file.c b/fs/ext4/file.c index f4a24a46245e..70b0438dbc94 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -371,15 +371,17 @@ static const struct vm_operations_struct ext4_file_vm_ops = { static int ext4_file_mmap(struct file *file, struct vm_area_struct *vma) { struct inode *inode = file->f_mapping->host; + struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); + struct dax_device *dax_dev = sbi->s_daxdev; - if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb)))) + if (unlikely(ext4_forced_shutdown(sbi))) return -EIO; /* - * We don't support synchronous mappings for non-DAX files. At least - * until someone comes with a sensible use case. + * We don't support synchronous mappings for non-DAX files and + * for DAX files if underneath dax_device is not synchronous. */ - if (!IS_DAX(file_inode(file)) && (vma->vm_flags & VM_SYNC)) + if (!daxdev_mapping_supported(vma, dax_dev)) return -EOPNOTSUPP; file_accessed(file); diff --git a/fs/nfs/Makefile b/fs/nfs/Makefile index c587e3c4c6a6..34cdeaecccf6 100644 --- a/fs/nfs/Makefile +++ b/fs/nfs/Makefile @@ -8,7 +8,8 @@ obj-$(CONFIG_NFS_FS) += nfs.o CFLAGS_nfstrace.o += -I$(src) nfs-y := client.o dir.o file.o getroot.o inode.o super.o \ io.o direct.o pagelist.o read.o symlink.o unlink.o \ - write.o namespace.o mount_clnt.o nfstrace.o export.o + write.o namespace.o mount_clnt.o nfstrace.o \ + export.o sysfs.o nfs-$(CONFIG_ROOT_NFS) += nfsroot.o nfs-$(CONFIG_SYSCTL) += sysctl.o nfs-$(CONFIG_NFS_FSCACHE) += fscache.o fscache-index.o diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c index 315967354954..f39924ba050b 100644 --- a/fs/nfs/callback_proc.c +++ b/fs/nfs/callback_proc.c @@ -414,27 +414,39 @@ static __be32 validate_seqid(const struct nfs4_slot_table *tbl, const struct nfs4_slot *slot, const struct cb_sequenceargs * args) { + __be32 ret; + + ret = cpu_to_be32(NFS4ERR_BADSLOT); if (args->csa_slotid > tbl->server_highest_slotid) - return htonl(NFS4ERR_BADSLOT); + goto out_err; /* Replay */ if (args->csa_sequenceid == slot->seq_nr) { + ret = cpu_to_be32(NFS4ERR_DELAY); if (nfs4_test_locked_slot(tbl, slot->slot_nr)) - return htonl(NFS4ERR_DELAY); + goto out_err; + /* Signal process_op to set this error on next op */ + ret = cpu_to_be32(NFS4ERR_RETRY_UNCACHED_REP); if (args->csa_cachethis == 0) - return htonl(NFS4ERR_RETRY_UNCACHED_REP); + goto out_err; /* Liar! We never allowed you to set csa_cachethis != 0 */ - return htonl(NFS4ERR_SEQ_FALSE_RETRY); + ret = cpu_to_be32(NFS4ERR_SEQ_FALSE_RETRY); + goto out_err; } /* Note: wraparound relies on seq_nr being of type u32 */ - if (likely(args->csa_sequenceid == slot->seq_nr + 1)) - return htonl(NFS4_OK); - /* Misordered request */ - return htonl(NFS4ERR_SEQ_MISORDERED); + ret = cpu_to_be32(NFS4ERR_SEQ_MISORDERED); + if (args->csa_sequenceid != slot->seq_nr + 1) + goto out_err; + + return cpu_to_be32(NFS4_OK); + +out_err: + trace_nfs4_cb_seqid_err(args, ret); + return ret; } /* diff --git a/fs/nfs/client.c b/fs/nfs/client.c index d7e4f0848e28..30838304a0bf 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -49,6 +49,7 @@ #include "pnfs.h" #include "nfs.h" #include "netns.h" +#include "sysfs.h" #define NFSDBG_FACILITY NFSDBG_CLIENT @@ -175,6 +176,7 @@ struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_init) clp->cl_rpcclient = ERR_PTR(-EINVAL); clp->cl_proto = cl_init->proto; + clp->cl_nconnect = cl_init->nconnect; clp->cl_net = get_net(cl_init->net); clp->cl_principal = "*"; @@ -192,7 +194,7 @@ error_0: EXPORT_SYMBOL_GPL(nfs_alloc_client); #if IS_ENABLED(CONFIG_NFS_V4) -void nfs_cleanup_cb_ident_idr(struct net *net) +static void nfs_cleanup_cb_ident_idr(struct net *net) { struct nfs_net *nn = net_generic(net, nfs_net_id); @@ -214,7 +216,7 @@ static void pnfs_init_server(struct nfs_server *server) } #else -void nfs_cleanup_cb_ident_idr(struct net *net) +static void nfs_cleanup_cb_ident_idr(struct net *net) { } @@ -406,10 +408,10 @@ struct nfs_client *nfs_get_client(const struct nfs_client_initdata *cl_init) clp = nfs_match_client(cl_init); if (clp) { spin_unlock(&nn->nfs_client_lock); - if (IS_ERR(clp)) - return clp; if (new) new->rpc_ops->free_client(new); + if (IS_ERR(clp)) + return clp; return nfs_found_client(cl_init, clp); } if (new) { @@ -493,6 +495,7 @@ int nfs_create_rpc_client(struct nfs_client *clp, struct rpc_create_args args = { .net = clp->cl_net, .protocol = clp->cl_proto, + .nconnect = clp->cl_nconnect, .address = (struct sockaddr *)&clp->cl_addr, .addrsize = clp->cl_addrlen, .timeout = cl_init->timeparms, @@ -658,6 +661,7 @@ static int nfs_init_server(struct nfs_server *server, .net = data->net, .timeparms = &timeparms, .cred = server->cred, + .nconnect = data->nfs_server.nconnect, }; struct nfs_client *clp; int error; @@ -1072,6 +1076,18 @@ void nfs_clients_init(struct net *net) #endif spin_lock_init(&nn->nfs_client_lock); nn->boot_time = ktime_get_real(); + + nfs_netns_sysfs_setup(nn, net); +} + +void nfs_clients_exit(struct net *net) +{ + struct nfs_net *nn = net_generic(net, nfs_net_id); + + nfs_netns_sysfs_destroy(nn); + nfs_cleanup_cb_ident_idr(net); + WARN_ON_ONCE(!list_empty(&nn->nfs_client_list)); + WARN_ON_ONCE(!list_empty(&nn->nfs_volume_list)); } #ifdef CONFIG_PROC_FS diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 57b6a45576ad..8d501093660f 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -80,6 +80,10 @@ static struct nfs_open_dir_context *alloc_nfs_open_dir_context(struct inode *dir ctx->dup_cookie = 0; ctx->cred = get_cred(cred); spin_lock(&dir->i_lock); + if (list_empty(&nfsi->open_files) && + (nfsi->cache_validity & NFS_INO_DATA_INVAL_DEFER)) + nfsi->cache_validity |= NFS_INO_INVALID_DATA | + NFS_INO_REVAL_FORCED; list_add(&ctx->list, &nfsi->open_files); spin_unlock(&dir->i_lock); return ctx; @@ -140,19 +144,12 @@ struct nfs_cache_array { struct nfs_cache_array_entry array[0]; }; -struct readdirvec { - unsigned long nr; - unsigned long index; - struct page *pages[NFS_MAX_READDIR_RAPAGES]; -}; - typedef int (*decode_dirent_t)(struct xdr_stream *, struct nfs_entry *, bool); typedef struct { struct file *file; struct page *page; struct dir_context *ctx; unsigned long page_index; - struct readdirvec pvec; u64 *dir_cookie; u64 last_cookie; loff_t current_index; @@ -532,10 +529,6 @@ int nfs_readdir_page_filler(nfs_readdir_descriptor_t *desc, struct nfs_entry *en struct nfs_cache_array *array; unsigned int count = 0; int status; - int max_rapages = NFS_MAX_READDIR_RAPAGES; - - desc->pvec.index = desc->page_index; - desc->pvec.nr = 0; scratch = alloc_page(GFP_KERNEL); if (scratch == NULL) @@ -560,40 +553,20 @@ int nfs_readdir_page_filler(nfs_readdir_descriptor_t *desc, struct nfs_entry *en if (desc->plus) nfs_prime_dcache(file_dentry(desc->file), entry); - status = nfs_readdir_add_to_array(entry, desc->pvec.pages[desc->pvec.nr]); - if (status == -ENOSPC) { - desc->pvec.nr++; - if (desc->pvec.nr == max_rapages) - break; - status = nfs_readdir_add_to_array(entry, desc->pvec.pages[desc->pvec.nr]); - } + status = nfs_readdir_add_to_array(entry, page); if (status != 0) break; } while (!entry->eof); - /* - * page and desc->pvec.pages[0] are valid, don't need to check - * whether or not to be NULL. - */ - copy_highpage(page, desc->pvec.pages[0]); - out_nopages: if (count == 0 || (status == -EBADCOOKIE && entry->eof != 0)) { - array = kmap_atomic(desc->pvec.pages[desc->pvec.nr]); + array = kmap(page); array->eof_index = array->size; status = 0; - kunmap_atomic(array); + kunmap(page); } put_page(scratch); - - /* - * desc->pvec.nr > 0 means at least one page was completely filled, - * we should return -ENOSPC. Otherwise function - * nfs_readdir_xdr_to_array will enter infinite loop. - */ - if (desc->pvec.nr > 0) - return -ENOSPC; return status; } @@ -627,24 +600,6 @@ out_freepages: return -ENOMEM; } -/* - * nfs_readdir_rapages_init initialize rapages by nfs_cache_array structure. - */ -static -void nfs_readdir_rapages_init(nfs_readdir_descriptor_t *desc) -{ - struct nfs_cache_array *array; - int max_rapages = NFS_MAX_READDIR_RAPAGES; - int index; - - for (index = 0; index < max_rapages; index++) { - array = kmap_atomic(desc->pvec.pages[index]); - memset(array, 0, sizeof(struct nfs_cache_array)); - array->eof_index = -1; - kunmap_atomic(array); - } -} - static int nfs_readdir_xdr_to_array(nfs_readdir_descriptor_t *desc, struct page *page, struct inode *inode) { @@ -655,12 +610,6 @@ int nfs_readdir_xdr_to_array(nfs_readdir_descriptor_t *desc, struct page *page, int status = -ENOMEM; unsigned int array_size = ARRAY_SIZE(pages); - /* - * This means we hit readdir rdpages miss, the preallocated rdpages - * are useless, the preallocate rdpages should be reinitialized. - */ - nfs_readdir_rapages_init(desc); - entry.prev_cookie = 0; entry.cookie = desc->last_cookie; entry.eof = 0; @@ -721,24 +670,9 @@ int nfs_readdir_filler(void *data, struct page* page) struct inode *inode = file_inode(desc->file); int ret; - /* - * If desc->page_index in range desc->pvec.index and - * desc->pvec.index + desc->pvec.nr, we get readdir cache hit. - */ - if (desc->page_index >= desc->pvec.index && - desc->page_index < (desc->pvec.index + desc->pvec.nr)) { - /* - * page and desc->pvec.pages[x] are valid, don't need to check - * whether or not to be NULL. - */ - copy_highpage(page, desc->pvec.pages[desc->page_index - desc->pvec.index]); - ret = 0; - } else { - ret = nfs_readdir_xdr_to_array(desc, page, inode); - if (ret < 0) - goto error; - } - + ret = nfs_readdir_xdr_to_array(desc, page, inode); + if (ret < 0) + goto error; SetPageUptodate(page); if (invalidate_inode_pages2_range(inode->i_mapping, page->index + 1, -1) < 0) { @@ -903,7 +837,6 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx) *desc = &my_desc; struct nfs_open_dir_context *dir_ctx = file->private_data; int res = 0; - int max_rapages = NFS_MAX_READDIR_RAPAGES; dfprintk(FILE, "NFS: readdir(%pD2) starting at cookie %llu\n", file, (long long)ctx->pos); @@ -923,12 +856,6 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx) desc->decode = NFS_PROTO(inode)->decode_dirent; desc->plus = nfs_use_readdirplus(inode, ctx); - res = nfs_readdir_alloc_pages(desc->pvec.pages, max_rapages); - if (res < 0) - return -ENOMEM; - - nfs_readdir_rapages_init(desc); - if (ctx->pos == 0 || nfs_attribute_cache_expired(inode)) res = nfs_revalidate_mapping(inode, file->f_mapping); if (res < 0) @@ -964,7 +891,6 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx) break; } while (!desc->eof); out: - nfs_readdir_free_pages(desc->pvec.pages, max_rapages); if (res > 0) res = 0; dfprintk(FILE, "NFS: readdir(%pD2) returns %d\n", file, res); diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c index bcff3bf5ae09..b04e20d28162 100644 --- a/fs/nfs/flexfilelayout/flexfilelayout.c +++ b/fs/nfs/flexfilelayout/flexfilelayout.c @@ -934,6 +934,10 @@ out_nolseg: if (pgio->pg_error < 0) return; out_mds: + trace_pnfs_mds_fallback_pg_init_read(pgio->pg_inode, + 0, NFS4_MAX_UINT64, IOMODE_READ, + NFS_I(pgio->pg_inode)->layout, + pgio->pg_lseg); pnfs_put_lseg(pgio->pg_lseg); pgio->pg_lseg = NULL; nfs_pageio_reset_read_mds(pgio); @@ -1000,6 +1004,10 @@ retry: return; out_mds: + trace_pnfs_mds_fallback_pg_init_write(pgio->pg_inode, + 0, NFS4_MAX_UINT64, IOMODE_RW, + NFS_I(pgio->pg_inode)->layout, + pgio->pg_lseg); pnfs_put_lseg(pgio->pg_lseg); pgio->pg_lseg = NULL; nfs_pageio_reset_write_mds(pgio); @@ -1026,6 +1034,10 @@ ff_layout_pg_get_mirror_count_write(struct nfs_pageio_descriptor *pgio, if (pgio->pg_lseg) return FF_LAYOUT_MIRROR_COUNT(pgio->pg_lseg); + trace_pnfs_mds_fallback_pg_get_mirror_count(pgio->pg_inode, + 0, NFS4_MAX_UINT64, IOMODE_RW, + NFS_I(pgio->pg_inode)->layout, + pgio->pg_lseg); /* no lseg means that pnfs is not in use, so no mirroring here */ nfs_pageio_reset_write_mds(pgio); out: @@ -1075,6 +1087,10 @@ static void ff_layout_reset_write(struct nfs_pgio_header *hdr, bool retry_pnfs) hdr->args.count, (unsigned long long)hdr->args.offset); + trace_pnfs_mds_fallback_write_done(hdr->inode, + hdr->args.offset, hdr->args.count, + IOMODE_RW, NFS_I(hdr->inode)->layout, + hdr->lseg); task->tk_status = pnfs_write_done_resend_to_mds(hdr); } } @@ -1094,6 +1110,10 @@ static void ff_layout_reset_read(struct nfs_pgio_header *hdr) hdr->args.count, (unsigned long long)hdr->args.offset); + trace_pnfs_mds_fallback_read_done(hdr->inode, + hdr->args.offset, hdr->args.count, + IOMODE_READ, NFS_I(hdr->inode)->layout, + hdr->lseg); task->tk_status = pnfs_read_done_resend_to_mds(hdr); } } @@ -1827,6 +1847,9 @@ ff_layout_read_pagelist(struct nfs_pgio_header *hdr) out_failed: if (ff_layout_avoid_mds_available_ds(lseg)) return PNFS_TRY_AGAIN; + trace_pnfs_mds_fallback_read_pagelist(hdr->inode, + hdr->args.offset, hdr->args.count, + IOMODE_READ, NFS_I(hdr->inode)->layout, lseg); return PNFS_NOT_ATTEMPTED; } @@ -1892,6 +1915,9 @@ ff_layout_write_pagelist(struct nfs_pgio_header *hdr, int sync) out_failed: if (ff_layout_avoid_mds_available_ds(lseg)) return PNFS_TRY_AGAIN; + trace_pnfs_mds_fallback_write_pagelist(hdr->inode, + hdr->args.offset, hdr->args.count, + IOMODE_RW, NFS_I(hdr->inode)->layout, lseg); return PNFS_NOT_ATTEMPTED; } diff --git a/fs/nfs/flexfilelayout/flexfilelayoutdev.c b/fs/nfs/flexfilelayout/flexfilelayoutdev.c index 19f856f45689..3eda40a320a5 100644 --- a/fs/nfs/flexfilelayout/flexfilelayoutdev.c +++ b/fs/nfs/flexfilelayout/flexfilelayoutdev.c @@ -257,7 +257,7 @@ int ff_layout_track_ds_error(struct nfs4_flexfile_layout *flo, if (status == 0) return 0; - if (mirror->mirror_ds == NULL) + if (IS_ERR_OR_NULL(mirror->mirror_ds)) return -EINVAL; dserr = kmalloc(sizeof(*dserr), gfp_flags); diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 0b4a1a974411..8a1758200b57 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -51,6 +51,7 @@ #include "pnfs.h" #include "nfs.h" #include "netns.h" +#include "sysfs.h" #include "nfstrace.h" @@ -208,7 +209,7 @@ static void nfs_set_cache_invalid(struct inode *inode, unsigned long flags) } if (inode->i_mapping->nrpages == 0) - flags &= ~NFS_INO_INVALID_DATA; + flags &= ~(NFS_INO_INVALID_DATA|NFS_INO_DATA_INVAL_DEFER); nfsi->cache_validity |= flags; if (flags & NFS_INO_INVALID_DATA) nfs_fscache_invalidate(inode); @@ -652,7 +653,8 @@ static int nfs_vmtruncate(struct inode * inode, loff_t offset) i_size_write(inode, offset); /* Optimisation */ if (offset == 0) - NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_DATA; + NFS_I(inode)->cache_validity &= ~(NFS_INO_INVALID_DATA | + NFS_INO_DATA_INVAL_DEFER); NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_SIZE; spin_unlock(&inode->i_lock); @@ -1032,6 +1034,10 @@ void nfs_inode_attach_open_context(struct nfs_open_context *ctx) struct nfs_inode *nfsi = NFS_I(inode); spin_lock(&inode->i_lock); + if (list_empty(&nfsi->open_files) && + (nfsi->cache_validity & NFS_INO_DATA_INVAL_DEFER)) + nfsi->cache_validity |= NFS_INO_INVALID_DATA | + NFS_INO_REVAL_FORCED; list_add_tail_rcu(&ctx->list, &nfsi->open_files); spin_unlock(&inode->i_lock); } @@ -1100,6 +1106,7 @@ int nfs_open(struct inode *inode, struct file *filp) nfs_fscache_open_file(inode, filp); return 0; } +EXPORT_SYMBOL_GPL(nfs_open); /* * This function is called whenever some part of NFS notices that @@ -1312,7 +1319,8 @@ int nfs_revalidate_mapping(struct inode *inode, set_bit(NFS_INO_INVALIDATING, bitlock); smp_wmb(); - nfsi->cache_validity &= ~NFS_INO_INVALID_DATA; + nfsi->cache_validity &= ~(NFS_INO_INVALID_DATA| + NFS_INO_DATA_INVAL_DEFER); spin_unlock(&inode->i_lock); trace_nfs_invalidate_mapping_enter(inode); ret = nfs_invalidate_mapping(inode, mapping); @@ -1870,7 +1878,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) dprintk("NFS: change_attr change on server for file %s/%ld\n", inode->i_sb->s_id, inode->i_ino); - } + } else if (!have_delegation) + nfsi->cache_validity |= NFS_INO_DATA_INVAL_DEFER; inode_set_iversion_raw(inode, fattr->change_attr); attr_changed = true; } @@ -2159,12 +2168,8 @@ static int nfs_net_init(struct net *net) static void nfs_net_exit(struct net *net) { - struct nfs_net *nn = net_generic(net, nfs_net_id); - nfs_fs_proc_net_exit(net); - nfs_cleanup_cb_ident_idr(net); - WARN_ON_ONCE(!list_empty(&nn->nfs_client_list)); - WARN_ON_ONCE(!list_empty(&nn->nfs_volume_list)); + nfs_clients_exit(net); } static struct pernet_operations nfs_net_ops = { @@ -2181,6 +2186,10 @@ static int __init init_nfs_fs(void) { int err; + err = nfs_sysfs_init(); + if (err < 0) + goto out10; + err = register_pernet_subsys(&nfs_net_ops); if (err < 0) goto out9; @@ -2244,6 +2253,8 @@ out7: out8: unregister_pernet_subsys(&nfs_net_ops); out9: + nfs_sysfs_exit(); +out10: return err; } @@ -2260,6 +2271,7 @@ static void __exit exit_nfs_fs(void) unregister_nfs_fs(); nfs_fs_proc_exit(); nfsiod_stop(); + nfs_sysfs_exit(); } /* Not quite true; I just maintain it */ diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index 498fab72f70b..a2346a2f8361 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -69,8 +69,7 @@ struct nfs_clone_mount { * Maximum number of pages that readdir can use for creating * a vmapped array of pages. */ -#define NFS_MAX_READDIR_PAGES 64 -#define NFS_MAX_READDIR_RAPAGES 8 +#define NFS_MAX_READDIR_PAGES 8 struct nfs_client_initdata { unsigned long init_flags; @@ -82,6 +81,7 @@ struct nfs_client_initdata { struct nfs_subversion *nfs_mod; int proto; u32 minorversion; + unsigned int nconnect; struct net *net; const struct rpc_timeout *timeparms; const struct cred *cred; @@ -123,6 +123,7 @@ struct nfs_parsed_mount_data { char *export_path; int port; unsigned short protocol; + unsigned short nconnect; } nfs_server; void *lsm_opts; @@ -158,6 +159,7 @@ extern void nfs_umount(const struct nfs_mount_request *info); /* client.c */ extern const struct rpc_program nfs_program; extern void nfs_clients_init(struct net *net); +extern void nfs_clients_exit(struct net *net); extern struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *); int nfs_create_rpc_client(struct nfs_client *, const struct nfs_client_initdata *, rpc_authflavor_t); struct nfs_client *nfs_get_client(const struct nfs_client_initdata *); @@ -170,7 +172,6 @@ int nfs_init_server_rpcclient(struct nfs_server *, const struct rpc_timeout *t, struct nfs_server *nfs_alloc_server(void); void nfs_server_copy_userdata(struct nfs_server *, struct nfs_server *); -extern void nfs_cleanup_cb_ident_idr(struct net *); extern void nfs_put_client(struct nfs_client *); extern void nfs_free_client(struct nfs_client *); extern struct nfs_client *nfs4_find_client_ident(struct net *, int); diff --git a/fs/nfs/netns.h b/fs/nfs/netns.h index fc9978c58265..c8374f74dce1 100644 --- a/fs/nfs/netns.h +++ b/fs/nfs/netns.h @@ -15,6 +15,8 @@ struct bl_dev_msg { uint32_t major, minor; }; +struct nfs_netns_client; + struct nfs_net { struct cache_detail *nfs_dns_resolve; struct rpc_pipe *bl_device_pipe; @@ -29,6 +31,7 @@ struct nfs_net { unsigned short nfs_callback_tcpport6; int cb_users[NFS4_MAX_MINOR_VERSION + 1]; #endif + struct nfs_netns_client *nfs_client; spinlock_t nfs_client_lock; ktime_t boot_time; #ifdef CONFIG_PROC_FS diff --git a/fs/nfs/nfs2xdr.c b/fs/nfs/nfs2xdr.c index 572794dab4b1..cbc17a203248 100644 --- a/fs/nfs/nfs2xdr.c +++ b/fs/nfs/nfs2xdr.c @@ -151,7 +151,7 @@ static int decode_stat(struct xdr_stream *xdr, enum nfs_stat *status) return 0; out_status: *status = be32_to_cpup(p); - trace_nfs_xdr_status((int)*status); + trace_nfs_xdr_status(xdr, (int)*status); return 0; } diff --git a/fs/nfs/nfs3client.c b/fs/nfs/nfs3client.c index fb0c425b5d45..148ceb74d27c 100644 --- a/fs/nfs/nfs3client.c +++ b/fs/nfs/nfs3client.c @@ -102,6 +102,9 @@ struct nfs_client *nfs3_set_ds_client(struct nfs_server *mds_srv, return ERR_PTR(-EINVAL); cl_init.hostname = buf; + if (mds_clp->cl_nconnect > 1 && ds_proto == XPRT_TRANSPORT_TCP) + cl_init.nconnect = mds_clp->cl_nconnect; + if (mds_srv->flags & NFS_MOUNT_NORESVPORT) set_bit(NFS_CS_NORESVPORT, &cl_init.init_flags); diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index abbbdde97e31..602767850b36 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -343,7 +343,7 @@ static int decode_nfsstat3(struct xdr_stream *xdr, enum nfs_stat *status) return 0; out_status: *status = be32_to_cpup(p); - trace_nfs_xdr_status((int)*status); + trace_nfs_xdr_status(xdr, (int)*status); return 0; } diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index 8a38a254f516..d778dad9a75e 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -312,12 +312,12 @@ extern int nfs4_set_rw_stateid(nfs4_stateid *stateid, const struct nfs_lock_context *l_ctx, fmode_t fmode); +extern int nfs4_proc_get_lease_time(struct nfs_client *clp, + struct nfs_fsinfo *fsinfo); #if defined(CONFIG_NFS_V4_1) extern int nfs41_sequence_done(struct rpc_task *, struct nfs4_sequence_res *); extern int nfs4_proc_create_session(struct nfs_client *, const struct cred *); extern int nfs4_proc_destroy_session(struct nfs4_session *, const struct cred *); -extern int nfs4_proc_get_lease_time(struct nfs_client *clp, - struct nfs_fsinfo *fsinfo); extern int nfs4_proc_layoutcommit(struct nfs4_layoutcommit_data *data, bool sync); extern int nfs4_detect_session_trunking(struct nfs_client *clp, diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c index 81b9b6d7927a..616393a01c06 100644 --- a/fs/nfs/nfs4client.c +++ b/fs/nfs/nfs4client.c @@ -859,7 +859,8 @@ static int nfs4_set_client(struct nfs_server *server, const size_t addrlen, const char *ip_addr, int proto, const struct rpc_timeout *timeparms, - u32 minorversion, struct net *net) + u32 minorversion, unsigned int nconnect, + struct net *net) { struct nfs_client_initdata cl_init = { .hostname = hostname, @@ -875,6 +876,8 @@ static int nfs4_set_client(struct nfs_server *server, }; struct nfs_client *clp; + if (minorversion > 0 && proto == XPRT_TRANSPORT_TCP) + cl_init.nconnect = nconnect; if (server->flags & NFS_MOUNT_NORESVPORT) set_bit(NFS_CS_NORESVPORT, &cl_init.init_flags); if (server->options & NFS_OPTION_MIGRATION) @@ -941,6 +944,9 @@ struct nfs_client *nfs4_set_ds_client(struct nfs_server *mds_srv, return ERR_PTR(-EINVAL); cl_init.hostname = buf; + if (mds_clp->cl_nconnect > 1 && ds_proto == XPRT_TRANSPORT_TCP) + cl_init.nconnect = mds_clp->cl_nconnect; + if (mds_srv->flags & NFS_MOUNT_NORESVPORT) __set_bit(NFS_CS_NORESVPORT, &cl_init.init_flags); @@ -1074,6 +1080,7 @@ static int nfs4_init_server(struct nfs_server *server, data->nfs_server.protocol, &timeparms, data->minorversion, + data->nfs_server.nconnect, data->net); if (error < 0) return error; @@ -1163,6 +1170,7 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data, XPRT_TRANSPORT_RDMA, parent_server->client->cl_timeout, parent_client->cl_mvops->minor_version, + parent_client->cl_nconnect, parent_client->cl_net); if (!error) goto init_server; @@ -1176,6 +1184,7 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data, XPRT_TRANSPORT_TCP, parent_server->client->cl_timeout, parent_client->cl_mvops->minor_version, + parent_client->cl_nconnect, parent_client->cl_net); if (error < 0) goto error; @@ -1271,7 +1280,8 @@ int nfs4_update_server(struct nfs_server *server, const char *hostname, set_bit(NFS_MIG_TSM_POSSIBLE, &server->mig_status); error = nfs4_set_client(server, hostname, sap, salen, buf, clp->cl_proto, clnt->cl_timeout, - clp->cl_minorversion, net); + clp->cl_minorversion, + clp->cl_nconnect, net); clear_bit(NFS_MIG_TSM_POSSIBLE, &server->mig_status); if (error != 0) { nfs_server_insert_lists(server); diff --git a/fs/nfs/nfs4file.c b/fs/nfs/nfs4file.c index f4157eb1f69d..96db471ca2e5 100644 --- a/fs/nfs/nfs4file.c +++ b/fs/nfs/nfs4file.c @@ -49,7 +49,7 @@ nfs4_file_open(struct inode *inode, struct file *filp) return err; if ((openflags & O_ACCMODE) == 3) - openflags--; + return nfs_open(inode, filp); /* We can't create new files here */ openflags &= ~(O_CREAT|O_EXCL); @@ -204,7 +204,11 @@ static loff_t nfs42_remap_file_range(struct file *src_file, loff_t src_off, bool same_inode = false; int ret; - if (remap_flags & ~(REMAP_FILE_DEDUP | REMAP_FILE_ADVISORY)) + /* NFS does not support deduplication. */ + if (remap_flags & REMAP_FILE_DEDUP) + return -EOPNOTSUPP; + + if (remap_flags & ~REMAP_FILE_ADVISORY) return -EINVAL; /* check alignment w.r.t. clone_blksize */ diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 6418cb6c079b..39896afc6edf 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -428,6 +428,22 @@ static int nfs4_delay(long *timeout, bool interruptible) return nfs4_delay_killable(timeout); } +static const nfs4_stateid * +nfs4_recoverable_stateid(const nfs4_stateid *stateid) +{ + if (!stateid) + return NULL; + switch (stateid->type) { + case NFS4_OPEN_STATEID_TYPE: + case NFS4_LOCK_STATEID_TYPE: + case NFS4_DELEGATION_STATEID_TYPE: + return stateid; + default: + break; + } + return NULL; +} + /* This is the error handling routine for processes that are allowed * to sleep. */ @@ -436,7 +452,7 @@ static int nfs4_do_handle_exception(struct nfs_server *server, { struct nfs_client *clp = server->nfs_client; struct nfs4_state *state = exception->state; - const nfs4_stateid *stateid = exception->stateid; + const nfs4_stateid *stateid; struct inode *inode = exception->inode; int ret = errorcode; @@ -444,8 +460,9 @@ static int nfs4_do_handle_exception(struct nfs_server *server, exception->recovering = 0; exception->retry = 0; + stateid = nfs4_recoverable_stateid(exception->stateid); if (stateid == NULL && state != NULL) - stateid = &state->stateid; + stateid = nfs4_recoverable_stateid(&state->stateid); switch(errorcode) { case 0: @@ -1165,6 +1182,18 @@ static bool nfs4_clear_cap_atomic_open_v1(struct nfs_server *server, return true; } +static fmode_t _nfs4_ctx_to_accessmode(const struct nfs_open_context *ctx) +{ + return ctx->mode & (FMODE_READ|FMODE_WRITE|FMODE_EXEC); +} + +static fmode_t _nfs4_ctx_to_openmode(const struct nfs_open_context *ctx) +{ + fmode_t ret = ctx->mode & (FMODE_READ|FMODE_WRITE); + + return (ctx->mode & FMODE_EXEC) ? FMODE_READ | ret : ret; +} + static u32 nfs4_map_atomic_open_share(struct nfs_server *server, fmode_t fmode, int openflags) @@ -2900,14 +2929,13 @@ static unsigned nfs4_exclusive_attrset(struct nfs4_opendata *opendata, } static int _nfs4_open_and_get_state(struct nfs4_opendata *opendata, - fmode_t fmode, - int flags, - struct nfs_open_context *ctx) + int flags, struct nfs_open_context *ctx) { struct nfs4_state_owner *sp = opendata->owner; struct nfs_server *server = sp->so_server; struct dentry *dentry; struct nfs4_state *state; + fmode_t acc_mode = _nfs4_ctx_to_accessmode(ctx); unsigned int seq; int ret; @@ -2946,7 +2974,8 @@ static int _nfs4_open_and_get_state(struct nfs4_opendata *opendata, /* Parse layoutget results before we check for access */ pnfs_parse_lgopen(state->inode, opendata->lgp, ctx); - ret = nfs4_opendata_access(sp->so_cred, opendata, state, fmode, flags); + ret = nfs4_opendata_access(sp->so_cred, opendata, state, + acc_mode, flags); if (ret != 0) goto out; @@ -2978,7 +3007,7 @@ static int _nfs4_do_open(struct inode *dir, struct dentry *dentry = ctx->dentry; const struct cred *cred = ctx->cred; struct nfs4_threshold **ctx_th = &ctx->mdsthreshold; - fmode_t fmode = ctx->mode & (FMODE_READ|FMODE_WRITE|FMODE_EXEC); + fmode_t fmode = _nfs4_ctx_to_openmode(ctx); enum open_claim_type4 claim = NFS4_OPEN_CLAIM_NULL; struct iattr *sattr = c->sattr; struct nfs4_label *label = c->label; @@ -3024,7 +3053,7 @@ static int _nfs4_do_open(struct inode *dir, if (d_really_is_positive(dentry)) opendata->state = nfs4_get_open_state(d_inode(dentry), sp); - status = _nfs4_open_and_get_state(opendata, fmode, flags, ctx); + status = _nfs4_open_and_get_state(opendata, flags, ctx); if (status != 0) goto err_free_label; state = ctx->state; @@ -3594,9 +3623,9 @@ static void nfs4_close_context(struct nfs_open_context *ctx, int is_sync) if (ctx->state == NULL) return; if (is_sync) - nfs4_close_sync(ctx->state, ctx->mode); + nfs4_close_sync(ctx->state, _nfs4_ctx_to_openmode(ctx)); else - nfs4_close_state(ctx->state, ctx->mode); + nfs4_close_state(ctx->state, _nfs4_ctx_to_openmode(ctx)); } #define FATTR4_WORD1_NFS40_MASK (2*FATTR4_WORD1_MOUNTED_ON_FILEID - 1UL) @@ -5980,7 +6009,7 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, .rpc_message = &msg, .callback_ops = &nfs4_setclientid_ops, .callback_data = &setclientid, - .flags = RPC_TASK_TIMEOUT, + .flags = RPC_TASK_TIMEOUT | RPC_TASK_NO_ROUND_ROBIN, }; int status; @@ -6046,7 +6075,8 @@ int nfs4_proc_setclientid_confirm(struct nfs_client *clp, dprintk("NFS call setclientid_confirm auth=%s, (client ID %llx)\n", clp->cl_rpcclient->cl_auth->au_ops->au_name, clp->cl_clientid); - status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT); + status = rpc_call_sync(clp->cl_rpcclient, &msg, + RPC_TASK_TIMEOUT | RPC_TASK_NO_ROUND_ROBIN); trace_nfs4_setclientid_confirm(clp, status); dprintk("NFS reply setclientid_confirm: %d\n", status); return status; @@ -7627,7 +7657,7 @@ static int _nfs4_proc_secinfo(struct inode *dir, const struct qstr *name, struct NFS_SP4_MACH_CRED_SECINFO, &clnt, &msg); status = nfs4_call_sync(clnt, NFS_SERVER(dir), &msg, &args.seq_args, - &res.seq_res, 0); + &res.seq_res, RPC_TASK_NO_ROUND_ROBIN); dprintk("NFS reply secinfo: %d\n", status); put_cred(cred); @@ -7965,7 +7995,7 @@ nfs4_run_exchange_id(struct nfs_client *clp, const struct cred *cred, .rpc_client = clp->cl_rpcclient, .callback_ops = &nfs4_exchange_id_call_ops, .rpc_message = &msg, - .flags = RPC_TASK_TIMEOUT, + .flags = RPC_TASK_TIMEOUT | RPC_TASK_NO_ROUND_ROBIN, }; struct nfs41_exchange_id_data *calldata; int status; @@ -8190,7 +8220,8 @@ static int _nfs4_proc_destroy_clientid(struct nfs_client *clp, }; int status; - status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT); + status = rpc_call_sync(clp->cl_rpcclient, &msg, + RPC_TASK_TIMEOUT | RPC_TASK_NO_ROUND_ROBIN); trace_nfs4_destroy_clientid(clp, status); if (status) dprintk("NFS: Got error %d from the server %s on " @@ -8241,6 +8272,8 @@ out: return ret; } +#endif /* CONFIG_NFS_V4_1 */ + struct nfs4_get_lease_time_data { struct nfs4_get_lease_time_args *args; struct nfs4_get_lease_time_res *res; @@ -8273,7 +8306,7 @@ static void nfs4_get_lease_time_done(struct rpc_task *task, void *calldata) (struct nfs4_get_lease_time_data *)calldata; dprintk("--> %s\n", __func__); - if (!nfs41_sequence_done(task, &data->res->lr_seq_res)) + if (!nfs4_sequence_done(task, &data->res->lr_seq_res)) return; switch (task->tk_status) { case -NFS4ERR_DELAY: @@ -8331,6 +8364,8 @@ int nfs4_proc_get_lease_time(struct nfs_client *clp, struct nfs_fsinfo *fsinfo) return status; } +#ifdef CONFIG_NFS_V4_1 + /* * Initialize the values to be used by the client in CREATE_SESSION * If nfs4_init_session set the fore channel request and response sizes, @@ -8345,6 +8380,7 @@ static void nfs4_init_channel_attrs(struct nfs41_create_session_args *args, { unsigned int max_rqst_sz, max_resp_sz; unsigned int max_bc_payload = rpc_max_bc_payload(clnt); + unsigned int max_bc_slots = rpc_num_bc_slots(clnt); max_rqst_sz = NFS_MAX_FILE_IO_SIZE + nfs41_maxwrite_overhead; max_resp_sz = NFS_MAX_FILE_IO_SIZE + nfs41_maxread_overhead; @@ -8367,6 +8403,8 @@ static void nfs4_init_channel_attrs(struct nfs41_create_session_args *args, args->bc_attrs.max_resp_sz_cached = 0; args->bc_attrs.max_ops = NFS4_MAX_BACK_CHANNEL_OPS; args->bc_attrs.max_reqs = max_t(unsigned short, max_session_cb_slots, 1); + if (args->bc_attrs.max_reqs > max_bc_slots) + args->bc_attrs.max_reqs = max_bc_slots; dprintk("%s: Back Channel : max_rqst_sz=%u max_resp_sz=%u " "max_resp_sz_cached=%u max_ops=%u max_reqs=%u\n", @@ -8469,7 +8507,8 @@ static int _nfs4_proc_create_session(struct nfs_client *clp, nfs4_init_channel_attrs(&args, clp->cl_rpcclient); args.flags = (SESSION4_PERSIST | SESSION4_BACK_CHAN); - status = rpc_call_sync(session->clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT); + status = rpc_call_sync(session->clp->cl_rpcclient, &msg, + RPC_TASK_TIMEOUT | RPC_TASK_NO_ROUND_ROBIN); trace_nfs4_create_session(clp, status); switch (status) { @@ -8545,7 +8584,8 @@ int nfs4_proc_destroy_session(struct nfs4_session *session, if (!test_and_clear_bit(NFS4_SESSION_ESTABLISHED, &session->session_state)) return 0; - status = rpc_call_sync(session->clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT); + status = rpc_call_sync(session->clp->cl_rpcclient, &msg, + RPC_TASK_TIMEOUT | RPC_TASK_NO_ROUND_ROBIN); trace_nfs4_destroy_session(session->clp, status); if (status) @@ -8799,7 +8839,7 @@ static int nfs41_proc_reclaim_complete(struct nfs_client *clp, .rpc_client = clp->cl_rpcclient, .rpc_message = &msg, .callback_ops = &nfs4_reclaim_complete_call_ops, - .flags = RPC_TASK_ASYNC, + .flags = RPC_TASK_ASYNC | RPC_TASK_NO_ROUND_ROBIN, }; int status = -ENOMEM; @@ -9318,7 +9358,7 @@ _nfs41_proc_secinfo_no_name(struct nfs_server *server, struct nfs_fh *fhandle, dprintk("--> %s\n", __func__); status = nfs4_call_sync(clnt, server, &msg, &args.seq_args, - &res.seq_res, 0); + &res.seq_res, RPC_TASK_NO_ROUND_ROBIN); dprintk("<-- %s status=%d\n", __func__, status); put_cred(cred); diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index e2e3c4f04d3e..9afd051a4876 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -87,6 +87,27 @@ const nfs4_stateid current_stateid = { static DEFINE_MUTEX(nfs_clid_init_mutex); +static int nfs4_setup_state_renewal(struct nfs_client *clp) +{ + int status; + struct nfs_fsinfo fsinfo; + unsigned long now; + + if (!test_bit(NFS_CS_CHECK_LEASE_TIME, &clp->cl_res_state)) { + nfs4_schedule_state_renewal(clp); + return 0; + } + + now = jiffies; + status = nfs4_proc_get_lease_time(clp, &fsinfo); + if (status == 0) { + nfs4_set_lease_period(clp, fsinfo.lease_time * HZ, now); + nfs4_schedule_state_renewal(clp); + } + + return status; +} + int nfs4_init_clientid(struct nfs_client *clp, const struct cred *cred) { struct nfs4_setclientid_res clid = { @@ -114,7 +135,7 @@ do_confirm: if (status != 0) goto out; clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); - nfs4_schedule_state_renewal(clp); + nfs4_setup_state_renewal(clp); out: return status; } @@ -286,34 +307,13 @@ static int nfs4_begin_drain_session(struct nfs_client *clp) #if defined(CONFIG_NFS_V4_1) -static int nfs41_setup_state_renewal(struct nfs_client *clp) -{ - int status; - struct nfs_fsinfo fsinfo; - unsigned long now; - - if (!test_bit(NFS_CS_CHECK_LEASE_TIME, &clp->cl_res_state)) { - nfs4_schedule_state_renewal(clp); - return 0; - } - - now = jiffies; - status = nfs4_proc_get_lease_time(clp, &fsinfo); - if (status == 0) { - nfs4_set_lease_period(clp, fsinfo.lease_time * HZ, now); - nfs4_schedule_state_renewal(clp); - } - - return status; -} - static void nfs41_finish_session_reset(struct nfs_client *clp) { clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); clear_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state); /* create_session negotiated new slot table */ clear_bit(NFS4CLNT_BIND_CONN_TO_SESSION, &clp->cl_state); - nfs41_setup_state_renewal(clp); + nfs4_setup_state_renewal(clp); } int nfs41_init_clientid(struct nfs_client *clp, const struct cred *cred) @@ -1064,8 +1064,7 @@ int nfs4_select_rw_stateid(struct nfs4_state *state, * choose to use. */ goto out; - nfs4_copy_open_stateid(dst, state); - ret = 0; + ret = nfs4_copy_open_stateid(dst, state) ? 0 : -EAGAIN; out: if (nfs_server_capable(state->inode, NFS_CAP_STATEID_NFSV41)) dst->seqid = 0; diff --git a/fs/nfs/nfs4trace.c b/fs/nfs/nfs4trace.c index e9fb3e50a999..1a8f376b3f73 100644 --- a/fs/nfs/nfs4trace.c +++ b/fs/nfs/nfs4trace.c @@ -16,4 +16,12 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(nfs4_pnfs_read); EXPORT_TRACEPOINT_SYMBOL_GPL(nfs4_pnfs_write); EXPORT_TRACEPOINT_SYMBOL_GPL(nfs4_pnfs_commit_ds); + +EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_pg_init_read); +EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_pg_init_write); +EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_pg_get_mirror_count); +EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_read_done); +EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_write_done); +EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_read_pagelist); +EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_write_pagelist); #endif diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index cd1a5c08da9a..b2f395fa7350 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -156,7 +156,7 @@ TRACE_DEFINE_ENUM(NFS4ERR_WRONG_TYPE); TRACE_DEFINE_ENUM(NFS4ERR_XDEV); #define show_nfsv4_errors(error) \ - __print_symbolic(-(error), \ + __print_symbolic(error, \ { NFS4_OK, "OK" }, \ /* Mapped by nfs4_stat_to_errno() */ \ { EPERM, "EPERM" }, \ @@ -348,7 +348,7 @@ DECLARE_EVENT_CLASS(nfs4_clientid_event, TP_STRUCT__entry( __string(dstaddr, clp->cl_hostname) - __field(int, error) + __field(unsigned long, error) ), TP_fast_assign( @@ -357,8 +357,8 @@ DECLARE_EVENT_CLASS(nfs4_clientid_event, ), TP_printk( - "error=%d (%s) dstaddr=%s", - __entry->error, + "error=%ld (%s) dstaddr=%s", + -__entry->error, show_nfsv4_errors(__entry->error), __get_str(dstaddr) ) @@ -420,7 +420,7 @@ TRACE_EVENT(nfs4_sequence_done, __field(unsigned int, highest_slotid) __field(unsigned int, target_highest_slotid) __field(unsigned int, status_flags) - __field(int, error) + __field(unsigned long, error) ), TP_fast_assign( @@ -435,10 +435,10 @@ TRACE_EVENT(nfs4_sequence_done, __entry->error = res->sr_status; ), TP_printk( - "error=%d (%s) session=0x%08x slot_nr=%u seq_nr=%u " + "error=%ld (%s) session=0x%08x slot_nr=%u seq_nr=%u " "highest_slotid=%u target_highest_slotid=%u " "status_flags=%u (%s)", - __entry->error, + -__entry->error, show_nfsv4_errors(__entry->error), __entry->session, __entry->slot_nr, @@ -467,7 +467,7 @@ TRACE_EVENT(nfs4_cb_sequence, __field(unsigned int, seq_nr) __field(unsigned int, highest_slotid) __field(unsigned int, cachethis) - __field(int, error) + __field(unsigned long, error) ), TP_fast_assign( @@ -476,13 +476,13 @@ TRACE_EVENT(nfs4_cb_sequence, __entry->seq_nr = args->csa_sequenceid; __entry->highest_slotid = args->csa_highestslotid; __entry->cachethis = args->csa_cachethis; - __entry->error = -be32_to_cpu(status); + __entry->error = be32_to_cpu(status); ), TP_printk( - "error=%d (%s) session=0x%08x slot_nr=%u seq_nr=%u " + "error=%ld (%s) session=0x%08x slot_nr=%u seq_nr=%u " "highest_slotid=%u", - __entry->error, + -__entry->error, show_nfsv4_errors(__entry->error), __entry->session, __entry->slot_nr, @@ -490,6 +490,44 @@ TRACE_EVENT(nfs4_cb_sequence, __entry->highest_slotid ) ); + +TRACE_EVENT(nfs4_cb_seqid_err, + TP_PROTO( + const struct cb_sequenceargs *args, + __be32 status + ), + TP_ARGS(args, status), + + TP_STRUCT__entry( + __field(unsigned int, session) + __field(unsigned int, slot_nr) + __field(unsigned int, seq_nr) + __field(unsigned int, highest_slotid) + __field(unsigned int, cachethis) + __field(unsigned long, error) + ), + + TP_fast_assign( + __entry->session = nfs_session_id_hash(&args->csa_sessionid); + __entry->slot_nr = args->csa_slotid; + __entry->seq_nr = args->csa_sequenceid; + __entry->highest_slotid = args->csa_highestslotid; + __entry->cachethis = args->csa_cachethis; + __entry->error = be32_to_cpu(status); + ), + + TP_printk( + "error=%ld (%s) session=0x%08x slot_nr=%u seq_nr=%u " + "highest_slotid=%u", + -__entry->error, + show_nfsv4_errors(__entry->error), + __entry->session, + __entry->slot_nr, + __entry->seq_nr, + __entry->highest_slotid + ) +); + #endif /* CONFIG_NFS_V4_1 */ TRACE_EVENT(nfs4_setup_sequence, @@ -526,26 +564,37 @@ TRACE_EVENT(nfs4_setup_sequence, TRACE_EVENT(nfs4_xdr_status, TP_PROTO( + const struct xdr_stream *xdr, u32 op, int error ), - TP_ARGS(op, error), + TP_ARGS(xdr, op, error), TP_STRUCT__entry( + __field(unsigned int, task_id) + __field(unsigned int, client_id) + __field(u32, xid) __field(u32, op) - __field(int, error) + __field(unsigned long, error) ), TP_fast_assign( + const struct rpc_rqst *rqstp = xdr->rqst; + const struct rpc_task *task = rqstp->rq_task; + + __entry->task_id = task->tk_pid; + __entry->client_id = task->tk_client->cl_clid; + __entry->xid = be32_to_cpu(rqstp->rq_xid); __entry->op = op; - __entry->error = -error; + __entry->error = error; ), TP_printk( - "operation %d: nfs status %d (%s)", - __entry->op, - __entry->error, show_nfsv4_errors(__entry->error) + "task:%u@%d xid=0x%08x error=%ld (%s) operation=%u", + __entry->task_id, __entry->client_id, __entry->xid, + -__entry->error, show_nfsv4_errors(__entry->error), + __entry->op ) ); @@ -559,7 +608,7 @@ DECLARE_EVENT_CLASS(nfs4_open_event, TP_ARGS(ctx, flags, error), TP_STRUCT__entry( - __field(int, error) + __field(unsigned long, error) __field(unsigned int, flags) __field(unsigned int, fmode) __field(dev_t, dev) @@ -577,7 +626,7 @@ DECLARE_EVENT_CLASS(nfs4_open_event, const struct nfs4_state *state = ctx->state; const struct inode *inode = NULL; - __entry->error = error; + __entry->error = -error; __entry->flags = flags; __entry->fmode = (__force unsigned int)ctx->mode; __entry->dev = ctx->dentry->d_sb->s_dev; @@ -609,11 +658,11 @@ DECLARE_EVENT_CLASS(nfs4_open_event, ), TP_printk( - "error=%d (%s) flags=%d (%s) fmode=%s " + "error=%ld (%s) flags=%d (%s) fmode=%s " "fileid=%02x:%02x:%llu fhandle=0x%08x " "name=%02x:%02x:%llu/%s stateid=%d:0x%08x " "openstateid=%d:0x%08x", - __entry->error, + -__entry->error, show_nfsv4_errors(__entry->error), __entry->flags, show_open_flags(__entry->flags), @@ -695,7 +744,7 @@ TRACE_EVENT(nfs4_close, __field(u32, fhandle) __field(u64, fileid) __field(unsigned int, fmode) - __field(int, error) + __field(unsigned long, error) __field(int, stateid_seq) __field(u32, stateid_hash) ), @@ -715,9 +764,9 @@ TRACE_EVENT(nfs4_close, ), TP_printk( - "error=%d (%s) fmode=%s fileid=%02x:%02x:%llu " + "error=%ld (%s) fmode=%s fileid=%02x:%02x:%llu " "fhandle=0x%08x openstateid=%d:0x%08x", - __entry->error, + -__entry->error, show_nfsv4_errors(__entry->error), __entry->fmode ? show_fmode_flags(__entry->fmode) : "closed", @@ -757,7 +806,7 @@ DECLARE_EVENT_CLASS(nfs4_lock_event, TP_ARGS(request, state, cmd, error), TP_STRUCT__entry( - __field(int, error) + __field(unsigned long, error) __field(int, cmd) __field(char, type) __field(loff_t, start) @@ -787,10 +836,10 @@ DECLARE_EVENT_CLASS(nfs4_lock_event, ), TP_printk( - "error=%d (%s) cmd=%s:%s range=%lld:%lld " + "error=%ld (%s) cmd=%s:%s range=%lld:%lld " "fileid=%02x:%02x:%llu fhandle=0x%08x " "stateid=%d:0x%08x", - __entry->error, + -__entry->error, show_nfsv4_errors(__entry->error), show_lock_cmd(__entry->cmd), show_lock_type(__entry->type), @@ -827,7 +876,7 @@ TRACE_EVENT(nfs4_set_lock, TP_ARGS(request, state, lockstateid, cmd, error), TP_STRUCT__entry( - __field(int, error) + __field(unsigned long, error) __field(int, cmd) __field(char, type) __field(loff_t, start) @@ -863,10 +912,10 @@ TRACE_EVENT(nfs4_set_lock, ), TP_printk( - "error=%d (%s) cmd=%s:%s range=%lld:%lld " + "error=%ld (%s) cmd=%s:%s range=%lld:%lld " "fileid=%02x:%02x:%llu fhandle=0x%08x " "stateid=%d:0x%08x lockstateid=%d:0x%08x", - __entry->error, + -__entry->error, show_nfsv4_errors(__entry->error), show_lock_cmd(__entry->cmd), show_lock_type(__entry->type), @@ -932,7 +981,7 @@ TRACE_EVENT(nfs4_delegreturn_exit, TP_STRUCT__entry( __field(dev_t, dev) __field(u32, fhandle) - __field(int, error) + __field(unsigned long, error) __field(int, stateid_seq) __field(u32, stateid_hash) ), @@ -948,9 +997,9 @@ TRACE_EVENT(nfs4_delegreturn_exit, ), TP_printk( - "error=%d (%s) dev=%02x:%02x fhandle=0x%08x " + "error=%ld (%s) dev=%02x:%02x fhandle=0x%08x " "stateid=%d:0x%08x", - __entry->error, + -__entry->error, show_nfsv4_errors(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), __entry->fhandle, @@ -969,7 +1018,7 @@ DECLARE_EVENT_CLASS(nfs4_test_stateid_event, TP_ARGS(state, lsp, error), TP_STRUCT__entry( - __field(int, error) + __field(unsigned long, error) __field(dev_t, dev) __field(u32, fhandle) __field(u64, fileid) @@ -991,9 +1040,9 @@ DECLARE_EVENT_CLASS(nfs4_test_stateid_event, ), TP_printk( - "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " "stateid=%d:0x%08x", - __entry->error, + -__entry->error, show_nfsv4_errors(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, @@ -1026,7 +1075,7 @@ DECLARE_EVENT_CLASS(nfs4_lookup_event, TP_STRUCT__entry( __field(dev_t, dev) - __field(int, error) + __field(unsigned long, error) __field(u64, dir) __string(name, name->name) ), @@ -1034,13 +1083,13 @@ DECLARE_EVENT_CLASS(nfs4_lookup_event, TP_fast_assign( __entry->dev = dir->i_sb->s_dev; __entry->dir = NFS_FILEID(dir); - __entry->error = error; + __entry->error = -error; __assign_str(name, name->name); ), TP_printk( - "error=%d (%s) name=%02x:%02x:%llu/%s", - __entry->error, + "error=%ld (%s) name=%02x:%02x:%llu/%s", + -__entry->error, show_nfsv4_errors(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->dir, @@ -1076,7 +1125,7 @@ TRACE_EVENT(nfs4_lookupp, TP_STRUCT__entry( __field(dev_t, dev) __field(u64, ino) - __field(int, error) + __field(unsigned long, error) ), TP_fast_assign( @@ -1086,8 +1135,8 @@ TRACE_EVENT(nfs4_lookupp, ), TP_printk( - "error=%d (%s) inode=%02x:%02x:%llu", - __entry->error, + "error=%ld (%s) inode=%02x:%02x:%llu", + -__entry->error, show_nfsv4_errors(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->ino @@ -1107,7 +1156,7 @@ TRACE_EVENT(nfs4_rename, TP_STRUCT__entry( __field(dev_t, dev) - __field(int, error) + __field(unsigned long, error) __field(u64, olddir) __string(oldname, oldname->name) __field(u64, newdir) @@ -1124,9 +1173,9 @@ TRACE_EVENT(nfs4_rename, ), TP_printk( - "error=%d (%s) oldname=%02x:%02x:%llu/%s " + "error=%ld (%s) oldname=%02x:%02x:%llu/%s " "newname=%02x:%02x:%llu/%s", - __entry->error, + -__entry->error, show_nfsv4_errors(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->olddir, @@ -1149,19 +1198,19 @@ DECLARE_EVENT_CLASS(nfs4_inode_event, __field(dev_t, dev) __field(u32, fhandle) __field(u64, fileid) - __field(int, error) + __field(unsigned long, error) ), TP_fast_assign( __entry->dev = inode->i_sb->s_dev; __entry->fileid = NFS_FILEID(inode); __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); - __entry->error = error; + __entry->error = error < 0 ? -error : 0; ), TP_printk( - "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x", - __entry->error, + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x", + -__entry->error, show_nfsv4_errors(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, @@ -1200,7 +1249,7 @@ DECLARE_EVENT_CLASS(nfs4_inode_stateid_event, __field(dev_t, dev) __field(u32, fhandle) __field(u64, fileid) - __field(int, error) + __field(unsigned long, error) __field(int, stateid_seq) __field(u32, stateid_hash) ), @@ -1217,9 +1266,9 @@ DECLARE_EVENT_CLASS(nfs4_inode_stateid_event, ), TP_printk( - "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " "stateid=%d:0x%08x", - __entry->error, + -__entry->error, show_nfsv4_errors(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, @@ -1257,7 +1306,7 @@ DECLARE_EVENT_CLASS(nfs4_getattr_event, __field(u32, fhandle) __field(u64, fileid) __field(unsigned int, valid) - __field(int, error) + __field(unsigned long, error) ), TP_fast_assign( @@ -1269,9 +1318,9 @@ DECLARE_EVENT_CLASS(nfs4_getattr_event, ), TP_printk( - "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " "valid=%s", - __entry->error, + -__entry->error, show_nfsv4_errors(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, @@ -1304,7 +1353,7 @@ DECLARE_EVENT_CLASS(nfs4_inode_callback_event, TP_ARGS(clp, fhandle, inode, error), TP_STRUCT__entry( - __field(int, error) + __field(unsigned long, error) __field(dev_t, dev) __field(u32, fhandle) __field(u64, fileid) @@ -1325,9 +1374,9 @@ DECLARE_EVENT_CLASS(nfs4_inode_callback_event, ), TP_printk( - "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " "dstaddr=%s", - __entry->error, + -__entry->error, show_nfsv4_errors(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, @@ -1359,7 +1408,7 @@ DECLARE_EVENT_CLASS(nfs4_inode_stateid_callback_event, TP_ARGS(clp, fhandle, inode, stateid, error), TP_STRUCT__entry( - __field(int, error) + __field(unsigned long, error) __field(dev_t, dev) __field(u32, fhandle) __field(u64, fileid) @@ -1386,9 +1435,9 @@ DECLARE_EVENT_CLASS(nfs4_inode_stateid_callback_event, ), TP_printk( - "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " "stateid=%d:0x%08x dstaddr=%s", - __entry->error, + -__entry->error, show_nfsv4_errors(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, @@ -1422,7 +1471,7 @@ DECLARE_EVENT_CLASS(nfs4_idmap_event, TP_ARGS(name, len, id, error), TP_STRUCT__entry( - __field(int, error) + __field(unsigned long, error) __field(u32, id) __dynamic_array(char, name, len > 0 ? len + 1 : 1) ), @@ -1437,8 +1486,8 @@ DECLARE_EVENT_CLASS(nfs4_idmap_event, ), TP_printk( - "error=%d id=%u name=%s", - __entry->error, + "error=%ld (%s) id=%u name=%s", + -__entry->error, show_nfsv4_errors(__entry->error), __entry->id, __get_str(name) ) @@ -1471,7 +1520,7 @@ DECLARE_EVENT_CLASS(nfs4_read_event, __field(u64, fileid) __field(loff_t, offset) __field(size_t, count) - __field(int, error) + __field(unsigned long, error) __field(int, stateid_seq) __field(u32, stateid_hash) ), @@ -1485,7 +1534,7 @@ DECLARE_EVENT_CLASS(nfs4_read_event, __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); __entry->offset = hdr->args.offset; __entry->count = hdr->args.count; - __entry->error = error; + __entry->error = error < 0 ? -error : 0; __entry->stateid_seq = be32_to_cpu(state->stateid.seqid); __entry->stateid_hash = @@ -1493,9 +1542,9 @@ DECLARE_EVENT_CLASS(nfs4_read_event, ), TP_printk( - "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " "offset=%lld count=%zu stateid=%d:0x%08x", - __entry->error, + -__entry->error, show_nfsv4_errors(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, @@ -1531,7 +1580,7 @@ DECLARE_EVENT_CLASS(nfs4_write_event, __field(u64, fileid) __field(loff_t, offset) __field(size_t, count) - __field(int, error) + __field(unsigned long, error) __field(int, stateid_seq) __field(u32, stateid_hash) ), @@ -1545,7 +1594,7 @@ DECLARE_EVENT_CLASS(nfs4_write_event, __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); __entry->offset = hdr->args.offset; __entry->count = hdr->args.count; - __entry->error = error; + __entry->error = error < 0 ? -error : 0; __entry->stateid_seq = be32_to_cpu(state->stateid.seqid); __entry->stateid_hash = @@ -1553,9 +1602,9 @@ DECLARE_EVENT_CLASS(nfs4_write_event, ), TP_printk( - "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " "offset=%lld count=%zu stateid=%d:0x%08x", - __entry->error, + -__entry->error, show_nfsv4_errors(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, @@ -1592,7 +1641,7 @@ DECLARE_EVENT_CLASS(nfs4_commit_event, __field(u64, fileid) __field(loff_t, offset) __field(size_t, count) - __field(int, error) + __field(unsigned long, error) ), TP_fast_assign( @@ -1606,9 +1655,9 @@ DECLARE_EVENT_CLASS(nfs4_commit_event, ), TP_printk( - "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " "offset=%lld count=%zu", - __entry->error, + -__entry->error, show_nfsv4_errors(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, @@ -1656,7 +1705,7 @@ TRACE_EVENT(nfs4_layoutget, __field(u32, iomode) __field(u64, offset) __field(u64, count) - __field(int, error) + __field(unsigned long, error) __field(int, stateid_seq) __field(u32, stateid_hash) __field(int, layoutstateid_seq) @@ -1689,10 +1738,10 @@ TRACE_EVENT(nfs4_layoutget, ), TP_printk( - "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " "iomode=%s offset=%llu count=%llu stateid=%d:0x%08x " "layoutstateid=%d:0x%08x", - __entry->error, + -__entry->error, show_nfsv4_errors(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, @@ -1722,6 +1771,7 @@ TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_BLOCKED); TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_INVALID_OPEN); TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_RETRY); TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_SEND_LAYOUTGET); +TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_EXIT); #define show_pnfs_update_layout_reason(reason) \ __print_symbolic(reason, \ @@ -1737,7 +1787,8 @@ TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_SEND_LAYOUTGET); { PNFS_UPDATE_LAYOUT_BLOCKED, "layouts blocked" }, \ { PNFS_UPDATE_LAYOUT_INVALID_OPEN, "invalid open" }, \ { PNFS_UPDATE_LAYOUT_RETRY, "retrying" }, \ - { PNFS_UPDATE_LAYOUT_SEND_LAYOUTGET, "sent layoutget" }) + { PNFS_UPDATE_LAYOUT_SEND_LAYOUTGET, "sent layoutget" }, \ + { PNFS_UPDATE_LAYOUT_EXIT, "exit" }) TRACE_EVENT(pnfs_update_layout, TP_PROTO(struct inode *inode, @@ -1796,6 +1847,78 @@ TRACE_EVENT(pnfs_update_layout, ) ); +DECLARE_EVENT_CLASS(pnfs_layout_event, + TP_PROTO(struct inode *inode, + loff_t pos, + u64 count, + enum pnfs_iomode iomode, + struct pnfs_layout_hdr *lo, + struct pnfs_layout_segment *lseg + ), + TP_ARGS(inode, pos, count, iomode, lo, lseg), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u64, fileid) + __field(u32, fhandle) + __field(loff_t, pos) + __field(u64, count) + __field(enum pnfs_iomode, iomode) + __field(int, layoutstateid_seq) + __field(u32, layoutstateid_hash) + __field(long, lseg) + ), + TP_fast_assign( + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = NFS_FILEID(inode); + __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->pos = pos; + __entry->count = count; + __entry->iomode = iomode; + if (lo != NULL) { + __entry->layoutstateid_seq = + be32_to_cpu(lo->plh_stateid.seqid); + __entry->layoutstateid_hash = + nfs_stateid_hash(&lo->plh_stateid); + } else { + __entry->layoutstateid_seq = 0; + __entry->layoutstateid_hash = 0; + } + __entry->lseg = (long)lseg; + ), + TP_printk( + "fileid=%02x:%02x:%llu fhandle=0x%08x " + "iomode=%s pos=%llu count=%llu " + "layoutstateid=%d:0x%08x lseg=0x%lx", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + show_pnfs_iomode(__entry->iomode), + (unsigned long long)__entry->pos, + (unsigned long long)__entry->count, + __entry->layoutstateid_seq, __entry->layoutstateid_hash, + __entry->lseg + ) +); + +#define DEFINE_PNFS_LAYOUT_EVENT(name) \ + DEFINE_EVENT(pnfs_layout_event, name, \ + TP_PROTO(struct inode *inode, \ + loff_t pos, \ + u64 count, \ + enum pnfs_iomode iomode, \ + struct pnfs_layout_hdr *lo, \ + struct pnfs_layout_segment *lseg \ + ), \ + TP_ARGS(inode, pos, count, iomode, lo, lseg)) + +DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_pg_init_read); +DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_pg_init_write); +DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_pg_get_mirror_count); +DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_read_done); +DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_write_done); +DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_read_pagelist); +DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_write_pagelist); + #endif /* CONFIG_NFS_V4_1 */ #endif /* _TRACE_NFS4_H */ diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 602446158bfb..46a8d636d151 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -837,6 +837,7 @@ static int decode_layoutget(struct xdr_stream *xdr, struct rpc_rqst *req, #define NFS4_dec_sequence_sz \ (compound_decode_hdr_maxsz + \ decode_sequence_maxsz) +#endif #define NFS4_enc_get_lease_time_sz (compound_encode_hdr_maxsz + \ encode_sequence_maxsz + \ encode_putrootfh_maxsz + \ @@ -845,6 +846,7 @@ static int decode_layoutget(struct xdr_stream *xdr, struct rpc_rqst *req, decode_sequence_maxsz + \ decode_putrootfh_maxsz + \ decode_fsinfo_maxsz) +#if defined(CONFIG_NFS_V4_1) #define NFS4_enc_reclaim_complete_sz (compound_encode_hdr_maxsz + \ encode_sequence_maxsz + \ encode_reclaim_complete_maxsz) @@ -2957,6 +2959,8 @@ static void nfs4_xdr_enc_sequence(struct rpc_rqst *req, struct xdr_stream *xdr, encode_nops(&hdr); } +#endif + /* * a GET_LEASE_TIME request */ @@ -2977,6 +2981,8 @@ static void nfs4_xdr_enc_get_lease_time(struct rpc_rqst *req, encode_nops(&hdr); } +#ifdef CONFIG_NFS_V4_1 + /* * a RECLAIM_COMPLETE request */ @@ -3187,7 +3193,7 @@ static bool __decode_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 expected, return true; out_status: nfserr = be32_to_cpup(p); - trace_nfs4_xdr_status(opnum, nfserr); + trace_nfs4_xdr_status(xdr, opnum, nfserr); *nfs_retval = nfs4_stat_to_errno(nfserr); return true; out_bad_operation: @@ -3427,7 +3433,7 @@ static int decode_attr_lease_time(struct xdr_stream *xdr, uint32_t *bitmap, uint *res = be32_to_cpup(p); bitmap[0] &= ~FATTR4_WORD0_LEASE_TIME; } - dprintk("%s: file size=%u\n", __func__, (unsigned int)*res); + dprintk("%s: lease time=%u\n", __func__, (unsigned int)*res); return 0; } @@ -7122,6 +7128,8 @@ static int nfs4_xdr_dec_sequence(struct rpc_rqst *rqstp, return status; } +#endif + /* * Decode GET_LEASE_TIME response */ @@ -7143,6 +7151,8 @@ static int nfs4_xdr_dec_get_lease_time(struct rpc_rqst *rqstp, return status; } +#ifdef CONFIG_NFS_V4_1 + /* * Decode RECLAIM_COMPLETE response */ @@ -7551,7 +7561,7 @@ const struct rpc_procinfo nfs4_procedures[] = { PROC41(CREATE_SESSION, enc_create_session, dec_create_session), PROC41(DESTROY_SESSION, enc_destroy_session, dec_destroy_session), PROC41(SEQUENCE, enc_sequence, dec_sequence), - PROC41(GET_LEASE_TIME, enc_get_lease_time, dec_get_lease_time), + PROC(GET_LEASE_TIME, enc_get_lease_time, dec_get_lease_time), PROC41(RECLAIM_COMPLETE,enc_reclaim_complete, dec_reclaim_complete), PROC41(GETDEVICEINFO, enc_getdeviceinfo, dec_getdeviceinfo), PROC41(LAYOUTGET, enc_layoutget, dec_layoutget), diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h index a0d6910aa03a..976d4089e267 100644 --- a/fs/nfs/nfstrace.h +++ b/fs/nfs/nfstrace.h @@ -11,6 +11,16 @@ #include <linux/tracepoint.h> #include <linux/iversion.h> +TRACE_DEFINE_ENUM(DT_UNKNOWN); +TRACE_DEFINE_ENUM(DT_FIFO); +TRACE_DEFINE_ENUM(DT_CHR); +TRACE_DEFINE_ENUM(DT_DIR); +TRACE_DEFINE_ENUM(DT_BLK); +TRACE_DEFINE_ENUM(DT_REG); +TRACE_DEFINE_ENUM(DT_LNK); +TRACE_DEFINE_ENUM(DT_SOCK); +TRACE_DEFINE_ENUM(DT_WHT); + #define nfs_show_file_type(ftype) \ __print_symbolic(ftype, \ { DT_UNKNOWN, "UNKNOWN" }, \ @@ -23,25 +33,57 @@ { DT_SOCK, "SOCK" }, \ { DT_WHT, "WHT" }) +TRACE_DEFINE_ENUM(NFS_INO_INVALID_DATA); +TRACE_DEFINE_ENUM(NFS_INO_INVALID_ATIME); +TRACE_DEFINE_ENUM(NFS_INO_INVALID_ACCESS); +TRACE_DEFINE_ENUM(NFS_INO_INVALID_ACL); +TRACE_DEFINE_ENUM(NFS_INO_REVAL_PAGECACHE); +TRACE_DEFINE_ENUM(NFS_INO_REVAL_FORCED); +TRACE_DEFINE_ENUM(NFS_INO_INVALID_LABEL); +TRACE_DEFINE_ENUM(NFS_INO_INVALID_CHANGE); +TRACE_DEFINE_ENUM(NFS_INO_INVALID_CTIME); +TRACE_DEFINE_ENUM(NFS_INO_INVALID_MTIME); +TRACE_DEFINE_ENUM(NFS_INO_INVALID_SIZE); +TRACE_DEFINE_ENUM(NFS_INO_INVALID_OTHER); + #define nfs_show_cache_validity(v) \ __print_flags(v, "|", \ - { NFS_INO_INVALID_ATTR, "INVALID_ATTR" }, \ { NFS_INO_INVALID_DATA, "INVALID_DATA" }, \ { NFS_INO_INVALID_ATIME, "INVALID_ATIME" }, \ { NFS_INO_INVALID_ACCESS, "INVALID_ACCESS" }, \ { NFS_INO_INVALID_ACL, "INVALID_ACL" }, \ { NFS_INO_REVAL_PAGECACHE, "REVAL_PAGECACHE" }, \ { NFS_INO_REVAL_FORCED, "REVAL_FORCED" }, \ - { NFS_INO_INVALID_LABEL, "INVALID_LABEL" }) + { NFS_INO_INVALID_LABEL, "INVALID_LABEL" }, \ + { NFS_INO_INVALID_CHANGE, "INVALID_CHANGE" }, \ + { NFS_INO_INVALID_CTIME, "INVALID_CTIME" }, \ + { NFS_INO_INVALID_MTIME, "INVALID_MTIME" }, \ + { NFS_INO_INVALID_SIZE, "INVALID_SIZE" }, \ + { NFS_INO_INVALID_OTHER, "INVALID_OTHER" }) + +TRACE_DEFINE_ENUM(NFS_INO_ADVISE_RDPLUS); +TRACE_DEFINE_ENUM(NFS_INO_STALE); +TRACE_DEFINE_ENUM(NFS_INO_ACL_LRU_SET); +TRACE_DEFINE_ENUM(NFS_INO_INVALIDATING); +TRACE_DEFINE_ENUM(NFS_INO_FSCACHE); +TRACE_DEFINE_ENUM(NFS_INO_FSCACHE_LOCK); +TRACE_DEFINE_ENUM(NFS_INO_LAYOUTCOMMIT); +TRACE_DEFINE_ENUM(NFS_INO_LAYOUTCOMMITTING); +TRACE_DEFINE_ENUM(NFS_INO_LAYOUTSTATS); +TRACE_DEFINE_ENUM(NFS_INO_ODIRECT); #define nfs_show_nfsi_flags(v) \ __print_flags(v, "|", \ - { 1 << NFS_INO_ADVISE_RDPLUS, "ADVISE_RDPLUS" }, \ - { 1 << NFS_INO_STALE, "STALE" }, \ - { 1 << NFS_INO_INVALIDATING, "INVALIDATING" }, \ - { 1 << NFS_INO_FSCACHE, "FSCACHE" }, \ - { 1 << NFS_INO_LAYOUTCOMMIT, "NEED_LAYOUTCOMMIT" }, \ - { 1 << NFS_INO_LAYOUTCOMMITTING, "LAYOUTCOMMIT" }) + { BIT(NFS_INO_ADVISE_RDPLUS), "ADVISE_RDPLUS" }, \ + { BIT(NFS_INO_STALE), "STALE" }, \ + { BIT(NFS_INO_ACL_LRU_SET), "ACL_LRU_SET" }, \ + { BIT(NFS_INO_INVALIDATING), "INVALIDATING" }, \ + { BIT(NFS_INO_FSCACHE), "FSCACHE" }, \ + { BIT(NFS_INO_FSCACHE_LOCK), "FSCACHE_LOCK" }, \ + { BIT(NFS_INO_LAYOUTCOMMIT), "NEED_LAYOUTCOMMIT" }, \ + { BIT(NFS_INO_LAYOUTCOMMITTING), "LAYOUTCOMMIT" }, \ + { BIT(NFS_INO_LAYOUTSTATS), "LAYOUTSTATS" }, \ + { BIT(NFS_INO_ODIRECT), "ODIRECT" }) DECLARE_EVENT_CLASS(nfs_inode_event, TP_PROTO( @@ -83,7 +125,7 @@ DECLARE_EVENT_CLASS(nfs_inode_event_done, TP_ARGS(inode, error), TP_STRUCT__entry( - __field(int, error) + __field(unsigned long, error) __field(dev_t, dev) __field(u32, fhandle) __field(unsigned char, type) @@ -96,7 +138,7 @@ DECLARE_EVENT_CLASS(nfs_inode_event_done, TP_fast_assign( const struct nfs_inode *nfsi = NFS_I(inode); - __entry->error = error; + __entry->error = error < 0 ? -error : 0; __entry->dev = inode->i_sb->s_dev; __entry->fileid = nfsi->fileid; __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); @@ -108,10 +150,10 @@ DECLARE_EVENT_CLASS(nfs_inode_event_done, ), TP_printk( - "error=%d fileid=%02x:%02x:%llu fhandle=0x%08x " + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " "type=%u (%s) version=%llu size=%lld " - "cache_validity=%lu (%s) nfs_flags=%ld (%s)", - __entry->error, + "cache_validity=0x%lx (%s) nfs_flags=0x%lx (%s)", + -__entry->error, nfs_show_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, @@ -158,13 +200,41 @@ DEFINE_NFS_INODE_EVENT_DONE(nfs_fsync_exit); DEFINE_NFS_INODE_EVENT(nfs_access_enter); DEFINE_NFS_INODE_EVENT_DONE(nfs_access_exit); +TRACE_DEFINE_ENUM(LOOKUP_FOLLOW); +TRACE_DEFINE_ENUM(LOOKUP_DIRECTORY); +TRACE_DEFINE_ENUM(LOOKUP_AUTOMOUNT); +TRACE_DEFINE_ENUM(LOOKUP_PARENT); +TRACE_DEFINE_ENUM(LOOKUP_REVAL); +TRACE_DEFINE_ENUM(LOOKUP_RCU); +TRACE_DEFINE_ENUM(LOOKUP_NO_REVAL); +TRACE_DEFINE_ENUM(LOOKUP_NO_EVAL); +TRACE_DEFINE_ENUM(LOOKUP_OPEN); +TRACE_DEFINE_ENUM(LOOKUP_CREATE); +TRACE_DEFINE_ENUM(LOOKUP_EXCL); +TRACE_DEFINE_ENUM(LOOKUP_RENAME_TARGET); +TRACE_DEFINE_ENUM(LOOKUP_JUMPED); +TRACE_DEFINE_ENUM(LOOKUP_ROOT); +TRACE_DEFINE_ENUM(LOOKUP_EMPTY); +TRACE_DEFINE_ENUM(LOOKUP_DOWN); + #define show_lookup_flags(flags) \ - __print_flags((unsigned long)flags, "|", \ - { LOOKUP_AUTOMOUNT, "AUTOMOUNT" }, \ + __print_flags(flags, "|", \ + { LOOKUP_FOLLOW, "FOLLOW" }, \ { LOOKUP_DIRECTORY, "DIRECTORY" }, \ + { LOOKUP_AUTOMOUNT, "AUTOMOUNT" }, \ + { LOOKUP_PARENT, "PARENT" }, \ + { LOOKUP_REVAL, "REVAL" }, \ + { LOOKUP_RCU, "RCU" }, \ + { LOOKUP_NO_REVAL, "NO_REVAL" }, \ + { LOOKUP_NO_EVAL, "NO_EVAL" }, \ { LOOKUP_OPEN, "OPEN" }, \ { LOOKUP_CREATE, "CREATE" }, \ - { LOOKUP_EXCL, "EXCL" }) + { LOOKUP_EXCL, "EXCL" }, \ + { LOOKUP_RENAME_TARGET, "RENAME_TARGET" }, \ + { LOOKUP_JUMPED, "JUMPED" }, \ + { LOOKUP_ROOT, "ROOT" }, \ + { LOOKUP_EMPTY, "EMPTY" }, \ + { LOOKUP_DOWN, "DOWN" }) DECLARE_EVENT_CLASS(nfs_lookup_event, TP_PROTO( @@ -176,7 +246,7 @@ DECLARE_EVENT_CLASS(nfs_lookup_event, TP_ARGS(dir, dentry, flags), TP_STRUCT__entry( - __field(unsigned int, flags) + __field(unsigned long, flags) __field(dev_t, dev) __field(u64, dir) __string(name, dentry->d_name.name) @@ -190,7 +260,7 @@ DECLARE_EVENT_CLASS(nfs_lookup_event, ), TP_printk( - "flags=%u (%s) name=%02x:%02x:%llu/%s", + "flags=0x%lx (%s) name=%02x:%02x:%llu/%s", __entry->flags, show_lookup_flags(__entry->flags), MAJOR(__entry->dev), MINOR(__entry->dev), @@ -219,8 +289,8 @@ DECLARE_EVENT_CLASS(nfs_lookup_event_done, TP_ARGS(dir, dentry, flags, error), TP_STRUCT__entry( - __field(int, error) - __field(unsigned int, flags) + __field(unsigned long, error) + __field(unsigned long, flags) __field(dev_t, dev) __field(u64, dir) __string(name, dentry->d_name.name) @@ -229,14 +299,14 @@ DECLARE_EVENT_CLASS(nfs_lookup_event_done, TP_fast_assign( __entry->dev = dir->i_sb->s_dev; __entry->dir = NFS_FILEID(dir); - __entry->error = error; + __entry->error = error < 0 ? -error : 0; __entry->flags = flags; __assign_str(name, dentry->d_name.name); ), TP_printk( - "error=%d flags=%u (%s) name=%02x:%02x:%llu/%s", - __entry->error, + "error=%ld (%s) flags=0x%lx (%s) name=%02x:%02x:%llu/%s", + -__entry->error, nfs_show_status(__entry->error), __entry->flags, show_lookup_flags(__entry->flags), MAJOR(__entry->dev), MINOR(__entry->dev), @@ -260,15 +330,43 @@ DEFINE_NFS_LOOKUP_EVENT_DONE(nfs_lookup_exit); DEFINE_NFS_LOOKUP_EVENT(nfs_lookup_revalidate_enter); DEFINE_NFS_LOOKUP_EVENT_DONE(nfs_lookup_revalidate_exit); +TRACE_DEFINE_ENUM(O_WRONLY); +TRACE_DEFINE_ENUM(O_RDWR); +TRACE_DEFINE_ENUM(O_CREAT); +TRACE_DEFINE_ENUM(O_EXCL); +TRACE_DEFINE_ENUM(O_NOCTTY); +TRACE_DEFINE_ENUM(O_TRUNC); +TRACE_DEFINE_ENUM(O_APPEND); +TRACE_DEFINE_ENUM(O_NONBLOCK); +TRACE_DEFINE_ENUM(O_DSYNC); +TRACE_DEFINE_ENUM(O_DIRECT); +TRACE_DEFINE_ENUM(O_LARGEFILE); +TRACE_DEFINE_ENUM(O_DIRECTORY); +TRACE_DEFINE_ENUM(O_NOFOLLOW); +TRACE_DEFINE_ENUM(O_NOATIME); +TRACE_DEFINE_ENUM(O_CLOEXEC); + #define show_open_flags(flags) \ - __print_flags((unsigned long)flags, "|", \ + __print_flags(flags, "|", \ + { O_WRONLY, "O_WRONLY" }, \ + { O_RDWR, "O_RDWR" }, \ { O_CREAT, "O_CREAT" }, \ { O_EXCL, "O_EXCL" }, \ + { O_NOCTTY, "O_NOCTTY" }, \ { O_TRUNC, "O_TRUNC" }, \ { O_APPEND, "O_APPEND" }, \ + { O_NONBLOCK, "O_NONBLOCK" }, \ { O_DSYNC, "O_DSYNC" }, \ { O_DIRECT, "O_DIRECT" }, \ - { O_DIRECTORY, "O_DIRECTORY" }) + { O_LARGEFILE, "O_LARGEFILE" }, \ + { O_DIRECTORY, "O_DIRECTORY" }, \ + { O_NOFOLLOW, "O_NOFOLLOW" }, \ + { O_NOATIME, "O_NOATIME" }, \ + { O_CLOEXEC, "O_CLOEXEC" }) + +TRACE_DEFINE_ENUM(FMODE_READ); +TRACE_DEFINE_ENUM(FMODE_WRITE); +TRACE_DEFINE_ENUM(FMODE_EXEC); #define show_fmode_flags(mode) \ __print_flags(mode, "|", \ @@ -286,7 +384,7 @@ TRACE_EVENT(nfs_atomic_open_enter, TP_ARGS(dir, ctx, flags), TP_STRUCT__entry( - __field(unsigned int, flags) + __field(unsigned long, flags) __field(unsigned int, fmode) __field(dev_t, dev) __field(u64, dir) @@ -302,7 +400,7 @@ TRACE_EVENT(nfs_atomic_open_enter, ), TP_printk( - "flags=%u (%s) fmode=%s name=%02x:%02x:%llu/%s", + "flags=0x%lx (%s) fmode=%s name=%02x:%02x:%llu/%s", __entry->flags, show_open_flags(__entry->flags), show_fmode_flags(__entry->fmode), @@ -323,8 +421,8 @@ TRACE_EVENT(nfs_atomic_open_exit, TP_ARGS(dir, ctx, flags, error), TP_STRUCT__entry( - __field(int, error) - __field(unsigned int, flags) + __field(unsigned long, error) + __field(unsigned long, flags) __field(unsigned int, fmode) __field(dev_t, dev) __field(u64, dir) @@ -332,7 +430,7 @@ TRACE_EVENT(nfs_atomic_open_exit, ), TP_fast_assign( - __entry->error = error; + __entry->error = -error; __entry->dev = dir->i_sb->s_dev; __entry->dir = NFS_FILEID(dir); __entry->flags = flags; @@ -341,9 +439,9 @@ TRACE_EVENT(nfs_atomic_open_exit, ), TP_printk( - "error=%d flags=%u (%s) fmode=%s " + "error=%ld (%s) flags=0x%lx (%s) fmode=%s " "name=%02x:%02x:%llu/%s", - __entry->error, + -__entry->error, nfs_show_status(__entry->error), __entry->flags, show_open_flags(__entry->flags), show_fmode_flags(__entry->fmode), @@ -363,7 +461,7 @@ TRACE_EVENT(nfs_create_enter, TP_ARGS(dir, dentry, flags), TP_STRUCT__entry( - __field(unsigned int, flags) + __field(unsigned long, flags) __field(dev_t, dev) __field(u64, dir) __string(name, dentry->d_name.name) @@ -377,7 +475,7 @@ TRACE_EVENT(nfs_create_enter, ), TP_printk( - "flags=%u (%s) name=%02x:%02x:%llu/%s", + "flags=0x%lx (%s) name=%02x:%02x:%llu/%s", __entry->flags, show_open_flags(__entry->flags), MAJOR(__entry->dev), MINOR(__entry->dev), @@ -397,15 +495,15 @@ TRACE_EVENT(nfs_create_exit, TP_ARGS(dir, dentry, flags, error), TP_STRUCT__entry( - __field(int, error) - __field(unsigned int, flags) + __field(unsigned long, error) + __field(unsigned long, flags) __field(dev_t, dev) __field(u64, dir) __string(name, dentry->d_name.name) ), TP_fast_assign( - __entry->error = error; + __entry->error = -error; __entry->dev = dir->i_sb->s_dev; __entry->dir = NFS_FILEID(dir); __entry->flags = flags; @@ -413,8 +511,8 @@ TRACE_EVENT(nfs_create_exit, ), TP_printk( - "error=%d flags=%u (%s) name=%02x:%02x:%llu/%s", - __entry->error, + "error=%ld (%s) flags=0x%lx (%s) name=%02x:%02x:%llu/%s", + -__entry->error, nfs_show_status(__entry->error), __entry->flags, show_open_flags(__entry->flags), MAJOR(__entry->dev), MINOR(__entry->dev), @@ -469,7 +567,7 @@ DECLARE_EVENT_CLASS(nfs_directory_event_done, TP_ARGS(dir, dentry, error), TP_STRUCT__entry( - __field(int, error) + __field(unsigned long, error) __field(dev_t, dev) __field(u64, dir) __string(name, dentry->d_name.name) @@ -478,13 +576,13 @@ DECLARE_EVENT_CLASS(nfs_directory_event_done, TP_fast_assign( __entry->dev = dir->i_sb->s_dev; __entry->dir = NFS_FILEID(dir); - __entry->error = error; + __entry->error = error < 0 ? -error : 0; __assign_str(name, dentry->d_name.name); ), TP_printk( - "error=%d name=%02x:%02x:%llu/%s", - __entry->error, + "error=%ld (%s) name=%02x:%02x:%llu/%s", + -__entry->error, nfs_show_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->dir, __get_str(name) @@ -557,7 +655,7 @@ TRACE_EVENT(nfs_link_exit, TP_ARGS(inode, dir, dentry, error), TP_STRUCT__entry( - __field(int, error) + __field(unsigned long, error) __field(dev_t, dev) __field(u64, fileid) __field(u64, dir) @@ -568,13 +666,13 @@ TRACE_EVENT(nfs_link_exit, __entry->dev = inode->i_sb->s_dev; __entry->fileid = NFS_FILEID(inode); __entry->dir = NFS_FILEID(dir); - __entry->error = error; + __entry->error = error < 0 ? -error : 0; __assign_str(name, dentry->d_name.name); ), TP_printk( - "error=%d fileid=%02x:%02x:%llu name=%02x:%02x:%llu/%s", - __entry->error, + "error=%ld (%s) fileid=%02x:%02x:%llu name=%02x:%02x:%llu/%s", + -__entry->error, nfs_show_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), __entry->fileid, MAJOR(__entry->dev), MINOR(__entry->dev), @@ -642,7 +740,7 @@ DECLARE_EVENT_CLASS(nfs_rename_event_done, TP_STRUCT__entry( __field(dev_t, dev) - __field(int, error) + __field(unsigned long, error) __field(u64, old_dir) __string(old_name, old_dentry->d_name.name) __field(u64, new_dir) @@ -651,17 +749,17 @@ DECLARE_EVENT_CLASS(nfs_rename_event_done, TP_fast_assign( __entry->dev = old_dir->i_sb->s_dev; + __entry->error = -error; __entry->old_dir = NFS_FILEID(old_dir); __entry->new_dir = NFS_FILEID(new_dir); - __entry->error = error; __assign_str(old_name, old_dentry->d_name.name); __assign_str(new_name, new_dentry->d_name.name); ), TP_printk( - "error=%d old_name=%02x:%02x:%llu/%s " + "error=%ld (%s) old_name=%02x:%02x:%llu/%s " "new_name=%02x:%02x:%llu/%s", - __entry->error, + -__entry->error, nfs_show_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->old_dir, __get_str(old_name), @@ -697,7 +795,7 @@ TRACE_EVENT(nfs_sillyrename_unlink, TP_STRUCT__entry( __field(dev_t, dev) - __field(int, error) + __field(unsigned long, error) __field(u64, dir) __dynamic_array(char, name, data->args.name.len + 1) ), @@ -707,15 +805,15 @@ TRACE_EVENT(nfs_sillyrename_unlink, size_t len = data->args.name.len; __entry->dev = dir->i_sb->s_dev; __entry->dir = NFS_FILEID(dir); - __entry->error = error; + __entry->error = -error; memcpy(__get_str(name), data->args.name.name, len); __get_str(name)[len] = 0; ), TP_printk( - "error=%d name=%02x:%02x:%llu/%s", - __entry->error, + "error=%ld (%s) name=%02x:%02x:%llu/%s", + -__entry->error, nfs_show_status(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->dir, __get_str(name) @@ -974,6 +1072,8 @@ TRACE_DEFINE_ENUM(NFSERR_PERM); TRACE_DEFINE_ENUM(NFSERR_NOENT); TRACE_DEFINE_ENUM(NFSERR_IO); TRACE_DEFINE_ENUM(NFSERR_NXIO); +TRACE_DEFINE_ENUM(ECHILD); +TRACE_DEFINE_ENUM(NFSERR_EAGAIN); TRACE_DEFINE_ENUM(NFSERR_ACCES); TRACE_DEFINE_ENUM(NFSERR_EXIST); TRACE_DEFINE_ENUM(NFSERR_XDEV); @@ -985,6 +1085,7 @@ TRACE_DEFINE_ENUM(NFSERR_FBIG); TRACE_DEFINE_ENUM(NFSERR_NOSPC); TRACE_DEFINE_ENUM(NFSERR_ROFS); TRACE_DEFINE_ENUM(NFSERR_MLINK); +TRACE_DEFINE_ENUM(NFSERR_OPNOTSUPP); TRACE_DEFINE_ENUM(NFSERR_NAMETOOLONG); TRACE_DEFINE_ENUM(NFSERR_NOTEMPTY); TRACE_DEFINE_ENUM(NFSERR_DQUOT); @@ -1007,6 +1108,8 @@ TRACE_DEFINE_ENUM(NFSERR_JUKEBOX); { NFSERR_NOENT, "NOENT" }, \ { NFSERR_IO, "IO" }, \ { NFSERR_NXIO, "NXIO" }, \ + { ECHILD, "CHILD" }, \ + { NFSERR_EAGAIN, "AGAIN" }, \ { NFSERR_ACCES, "ACCES" }, \ { NFSERR_EXIST, "EXIST" }, \ { NFSERR_XDEV, "XDEV" }, \ @@ -1018,6 +1121,7 @@ TRACE_DEFINE_ENUM(NFSERR_JUKEBOX); { NFSERR_NOSPC, "NOSPC" }, \ { NFSERR_ROFS, "ROFS" }, \ { NFSERR_MLINK, "MLINK" }, \ + { NFSERR_OPNOTSUPP, "OPNOTSUPP" }, \ { NFSERR_NAMETOOLONG, "NAMETOOLONG" }, \ { NFSERR_NOTEMPTY, "NOTEMPTY" }, \ { NFSERR_DQUOT, "DQUOT" }, \ @@ -1035,22 +1139,33 @@ TRACE_DEFINE_ENUM(NFSERR_JUKEBOX); TRACE_EVENT(nfs_xdr_status, TP_PROTO( + const struct xdr_stream *xdr, int error ), - TP_ARGS(error), + TP_ARGS(xdr, error), TP_STRUCT__entry( - __field(int, error) + __field(unsigned int, task_id) + __field(unsigned int, client_id) + __field(u32, xid) + __field(unsigned long, error) ), TP_fast_assign( + const struct rpc_rqst *rqstp = xdr->rqst; + const struct rpc_task *task = rqstp->rq_task; + + __entry->task_id = task->tk_pid; + __entry->client_id = task->tk_client->cl_clid; + __entry->xid = be32_to_cpu(rqstp->rq_xid); __entry->error = error; ), TP_printk( - "error=%d (%s)", - __entry->error, nfs_show_status(__entry->error) + "task:%u@%d xid=0x%08x error=%ld (%s)", + __entry->task_id, __entry->client_id, __entry->xid, + -__entry->error, nfs_show_status(__entry->error) ) ); diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 6ef5278326b6..ed4e1b07447b 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -77,7 +77,7 @@ void nfs_set_pgio_error(struct nfs_pgio_header *hdr, int error, loff_t pos) static inline struct nfs_page * nfs_page_alloc(void) { - struct nfs_page *p = kmem_cache_zalloc(nfs_page_cachep, GFP_NOIO); + struct nfs_page *p = kmem_cache_zalloc(nfs_page_cachep, GFP_KERNEL); if (p) INIT_LIST_HEAD(&p->wb_list); return p; @@ -775,8 +775,6 @@ int nfs_generic_pgio(struct nfs_pageio_descriptor *desc, if (pagecount <= ARRAY_SIZE(pg_array->page_array)) pg_array->pagevec = pg_array->page_array; else { - if (hdr->rw_mode == FMODE_WRITE) - gfp_flags = GFP_NOIO; pg_array->pagevec = kcalloc(pagecount, sizeof(struct page *), gfp_flags); if (!pg_array->pagevec) { pg_array->npages = 0; @@ -851,7 +849,7 @@ nfs_pageio_alloc_mirrors(struct nfs_pageio_descriptor *desc, desc->pg_mirrors_dynamic = NULL; if (mirror_count == 1) return desc->pg_mirrors_static; - ret = kmalloc_array(mirror_count, sizeof(*ret), GFP_NOFS); + ret = kmalloc_array(mirror_count, sizeof(*ret), GFP_KERNEL); if (ret != NULL) { for (i = 0; i < mirror_count; i++) nfs_pageio_mirror_init(&ret[i], desc->pg_bsize); diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 83722e936b4a..75bd5b552ba4 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -1890,7 +1890,7 @@ lookup_again: spin_unlock(&ino->i_lock); lseg = ERR_PTR(wait_var_event_killable(&lo->plh_outstanding, !atomic_read(&lo->plh_outstanding))); - if (IS_ERR(lseg) || !list_empty(&lo->plh_segs)) + if (IS_ERR(lseg)) goto out_put_layout_hdr; pnfs_put_layout_hdr(lo); goto lookup_again; @@ -1915,6 +1915,7 @@ lookup_again: * stateid. */ if (test_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags)) { + int status; /* * The first layoutget for the file. Need to serialize per @@ -1934,13 +1935,20 @@ lookup_again: } first = true; - if (nfs4_select_rw_stateid(ctx->state, + status = nfs4_select_rw_stateid(ctx->state, iomode == IOMODE_RW ? FMODE_WRITE : FMODE_READ, - NULL, &stateid, NULL) != 0) { + NULL, &stateid, NULL); + if (status != 0) { trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg, PNFS_UPDATE_LAYOUT_INVALID_OPEN); - goto out_unlock; + if (status != -EAGAIN) + goto out_unlock; + spin_unlock(&ino->i_lock); + nfs4_schedule_stateid_recovery(server, ctx->state); + pnfs_clear_first_layoutget(lo); + pnfs_put_layout_hdr(lo); + goto lookup_again; } } else { nfs4_stateid_copy(&stateid, &lo->plh_stateid); @@ -2029,6 +2037,8 @@ lookup_again: out_put_layout_hdr: if (first) pnfs_clear_first_layoutget(lo); + trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg, + PNFS_UPDATE_LAYOUT_EXIT); pnfs_put_layout_hdr(lo); out: dprintk("%s: inode %s/%llu pNFS layout segment %s for " @@ -2468,7 +2478,7 @@ pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *pgio, wb_size, IOMODE_RW, false, - GFP_NOFS); + GFP_KERNEL); if (IS_ERR(pgio->pg_lseg)) { pgio->pg_error = PTR_ERR(pgio->pg_lseg); pgio->pg_lseg = NULL; diff --git a/fs/nfs/super.c b/fs/nfs/super.c index f88ddac2dcdf..3683d2b1cc8e 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -77,6 +77,8 @@ #define NFS_DEFAULT_VERSION 2 #endif +#define NFS_MAX_CONNECTIONS 16 + enum { /* Mount options that take no arguments */ Opt_soft, Opt_softerr, Opt_hard, @@ -108,6 +110,7 @@ enum { Opt_nfsvers, Opt_sec, Opt_proto, Opt_mountproto, Opt_mounthost, Opt_addr, Opt_mountaddr, Opt_clientaddr, + Opt_nconnect, Opt_lookupcache, Opt_fscache_uniq, Opt_local_lock, @@ -181,6 +184,8 @@ static const match_table_t nfs_mount_option_tokens = { { Opt_mounthost, "mounthost=%s" }, { Opt_mountaddr, "mountaddr=%s" }, + { Opt_nconnect, "nconnect=%s" }, + { Opt_lookupcache, "lookupcache=%s" }, { Opt_fscache_uniq, "fsc=%s" }, { Opt_local_lock, "local_lock=%s" }, @@ -582,7 +587,7 @@ static void nfs_show_mountd_options(struct seq_file *m, struct nfs_server *nfss, } default: if (showdefaults) - seq_printf(m, ",mountaddr=unspecified"); + seq_puts(m, ",mountaddr=unspecified"); } if (nfss->mountd_version || showdefaults) @@ -673,6 +678,8 @@ static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss, seq_printf(m, ",proto=%s", rpc_peeraddr2str(nfss->client, RPC_DISPLAY_NETID)); rcu_read_unlock(); + if (clp->cl_nconnect > 0) + seq_printf(m, ",nconnect=%u", clp->cl_nconnect); if (version == 4) { if (nfss->port != NFS_PORT) seq_printf(m, ",port=%u", nfss->port); @@ -690,29 +697,29 @@ static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss, nfs_show_nfsv4_options(m, nfss, showdefaults); if (nfss->options & NFS_OPTION_FSCACHE) - seq_printf(m, ",fsc"); + seq_puts(m, ",fsc"); if (nfss->options & NFS_OPTION_MIGRATION) - seq_printf(m, ",migration"); + seq_puts(m, ",migration"); if (nfss->flags & NFS_MOUNT_LOOKUP_CACHE_NONEG) { if (nfss->flags & NFS_MOUNT_LOOKUP_CACHE_NONE) - seq_printf(m, ",lookupcache=none"); + seq_puts(m, ",lookupcache=none"); else - seq_printf(m, ",lookupcache=pos"); + seq_puts(m, ",lookupcache=pos"); } local_flock = nfss->flags & NFS_MOUNT_LOCAL_FLOCK; local_fcntl = nfss->flags & NFS_MOUNT_LOCAL_FCNTL; if (!local_flock && !local_fcntl) - seq_printf(m, ",local_lock=none"); + seq_puts(m, ",local_lock=none"); else if (local_flock && local_fcntl) - seq_printf(m, ",local_lock=all"); + seq_puts(m, ",local_lock=all"); else if (local_flock) - seq_printf(m, ",local_lock=flock"); + seq_puts(m, ",local_lock=flock"); else - seq_printf(m, ",local_lock=posix"); + seq_puts(m, ",local_lock=posix"); } /* @@ -735,11 +742,21 @@ int nfs_show_options(struct seq_file *m, struct dentry *root) EXPORT_SYMBOL_GPL(nfs_show_options); #if IS_ENABLED(CONFIG_NFS_V4) +static void show_lease(struct seq_file *m, struct nfs_server *server) +{ + struct nfs_client *clp = server->nfs_client; + unsigned long expire; + + seq_printf(m, ",lease_time=%ld", clp->cl_lease_time / HZ); + expire = clp->cl_last_renewal + clp->cl_lease_time; + seq_printf(m, ",lease_expired=%ld", + time_after(expire, jiffies) ? 0 : (jiffies - expire) / HZ); +} #ifdef CONFIG_NFS_V4_1 static void show_sessions(struct seq_file *m, struct nfs_server *server) { if (nfs4_has_session(server->nfs_client)) - seq_printf(m, ",sessions"); + seq_puts(m, ",sessions"); } #else static void show_sessions(struct seq_file *m, struct nfs_server *server) {} @@ -816,7 +833,7 @@ int nfs_show_stats(struct seq_file *m, struct dentry *root) /* * Display all mount option settings */ - seq_printf(m, "\n\topts:\t"); + seq_puts(m, "\n\topts:\t"); seq_puts(m, sb_rdonly(root->d_sb) ? "ro" : "rw"); seq_puts(m, root->d_sb->s_flags & SB_SYNCHRONOUS ? ",sync" : ""); seq_puts(m, root->d_sb->s_flags & SB_NOATIME ? ",noatime" : ""); @@ -827,7 +844,7 @@ int nfs_show_stats(struct seq_file *m, struct dentry *root) show_implementation_id(m, nfss); - seq_printf(m, "\n\tcaps:\t"); + seq_puts(m, "\n\tcaps:\t"); seq_printf(m, "caps=0x%x", nfss->caps); seq_printf(m, ",wtmult=%u", nfss->wtmult); seq_printf(m, ",dtsize=%u", nfss->dtsize); @@ -836,13 +853,14 @@ int nfs_show_stats(struct seq_file *m, struct dentry *root) #if IS_ENABLED(CONFIG_NFS_V4) if (nfss->nfs_client->rpc_ops->version == 4) { - seq_printf(m, "\n\tnfsv4:\t"); + seq_puts(m, "\n\tnfsv4:\t"); seq_printf(m, "bm0=0x%x", nfss->attr_bitmask[0]); seq_printf(m, ",bm1=0x%x", nfss->attr_bitmask[1]); seq_printf(m, ",bm2=0x%x", nfss->attr_bitmask[2]); seq_printf(m, ",acl=0x%x", nfss->acl_bitmask); show_sessions(m, nfss); show_pnfs(m, nfss); + show_lease(m, nfss); } #endif @@ -874,20 +892,20 @@ int nfs_show_stats(struct seq_file *m, struct dentry *root) preempt_enable(); } - seq_printf(m, "\n\tevents:\t"); + seq_puts(m, "\n\tevents:\t"); for (i = 0; i < __NFSIOS_COUNTSMAX; i++) seq_printf(m, "%lu ", totals.events[i]); - seq_printf(m, "\n\tbytes:\t"); + seq_puts(m, "\n\tbytes:\t"); for (i = 0; i < __NFSIOS_BYTESMAX; i++) seq_printf(m, "%Lu ", totals.bytes[i]); #ifdef CONFIG_NFS_FSCACHE if (nfss->options & NFS_OPTION_FSCACHE) { - seq_printf(m, "\n\tfsc:\t"); + seq_puts(m, "\n\tfsc:\t"); for (i = 0; i < __NFSIOS_FSCACHEMAX; i++) seq_printf(m, "%Lu ", totals.fscache[i]); } #endif - seq_printf(m, "\n"); + seq_putc(m, '\n'); rpc_clnt_show_stats(m, nfss->client); @@ -1549,6 +1567,11 @@ static int nfs_parse_mount_options(char *raw, if (mnt->mount_server.addrlen == 0) goto out_invalid_address; break; + case Opt_nconnect: + if (nfs_get_option_ul_bound(args, &option, 1, NFS_MAX_CONNECTIONS)) + goto out_invalid_value; + mnt->nfs_server.nconnect = option; + break; case Opt_lookupcache: string = match_strdup(args); if (string == NULL) diff --git a/fs/nfs/sysfs.c b/fs/nfs/sysfs.c new file mode 100644 index 000000000000..4f3390b20239 --- /dev/null +++ b/fs/nfs/sysfs.c @@ -0,0 +1,187 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2019 Hammerspace Inc + */ + +#include <linux/module.h> +#include <linux/kobject.h> +#include <linux/sysfs.h> +#include <linux/fs.h> +#include <linux/slab.h> +#include <linux/netdevice.h> +#include <linux/string.h> +#include <linux/nfs_fs.h> +#include <linux/rcupdate.h> + +#include "nfs4_fs.h" +#include "netns.h" +#include "sysfs.h" + +struct kobject *nfs_client_kobj; +static struct kset *nfs_client_kset; + +static void nfs_netns_object_release(struct kobject *kobj) +{ + kfree(kobj); +} + +static const struct kobj_ns_type_operations *nfs_netns_object_child_ns_type( + struct kobject *kobj) +{ + return &net_ns_type_operations; +} + +static struct kobj_type nfs_netns_object_type = { + .release = nfs_netns_object_release, + .sysfs_ops = &kobj_sysfs_ops, + .child_ns_type = nfs_netns_object_child_ns_type, +}; + +static struct kobject *nfs_netns_object_alloc(const char *name, + struct kset *kset, struct kobject *parent) +{ + struct kobject *kobj; + + kobj = kzalloc(sizeof(*kobj), GFP_KERNEL); + if (kobj) { + kobj->kset = kset; + if (kobject_init_and_add(kobj, &nfs_netns_object_type, + parent, "%s", name) == 0) + return kobj; + kobject_put(kobj); + } + return NULL; +} + +int nfs_sysfs_init(void) +{ + nfs_client_kset = kset_create_and_add("nfs", NULL, fs_kobj); + if (!nfs_client_kset) + return -ENOMEM; + nfs_client_kobj = nfs_netns_object_alloc("net", nfs_client_kset, NULL); + if (!nfs_client_kobj) { + kset_unregister(nfs_client_kset); + nfs_client_kset = NULL; + return -ENOMEM; + } + return 0; +} + +void nfs_sysfs_exit(void) +{ + kobject_put(nfs_client_kobj); + kset_unregister(nfs_client_kset); +} + +static ssize_t nfs_netns_identifier_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + struct nfs_netns_client *c = container_of(kobj, + struct nfs_netns_client, + kobject); + return scnprintf(buf, PAGE_SIZE, "%s\n", c->identifier); +} + +/* Strip trailing '\n' */ +static size_t nfs_string_strip(const char *c, size_t len) +{ + while (len > 0 && c[len-1] == '\n') + --len; + return len; +} + +static ssize_t nfs_netns_identifier_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct nfs_netns_client *c = container_of(kobj, + struct nfs_netns_client, + kobject); + const char *old; + char *p; + size_t len; + + len = nfs_string_strip(buf, min_t(size_t, count, CONTAINER_ID_MAXLEN)); + if (!len) + return 0; + p = kmemdup_nul(buf, len, GFP_KERNEL); + if (!p) + return -ENOMEM; + old = xchg(&c->identifier, p); + if (old) { + synchronize_rcu(); + kfree(old); + } + return count; +} + +static void nfs_netns_client_release(struct kobject *kobj) +{ + struct nfs_netns_client *c = container_of(kobj, + struct nfs_netns_client, + kobject); + + if (c->identifier) + kfree(c->identifier); + kfree(c); +} + +static const void *nfs_netns_client_namespace(struct kobject *kobj) +{ + return container_of(kobj, struct nfs_netns_client, kobject)->net; +} + +static struct kobj_attribute nfs_netns_client_id = __ATTR(identifier, + 0644, nfs_netns_identifier_show, nfs_netns_identifier_store); + +static struct attribute *nfs_netns_client_attrs[] = { + &nfs_netns_client_id.attr, + NULL, +}; + +static struct kobj_type nfs_netns_client_type = { + .release = nfs_netns_client_release, + .default_attrs = nfs_netns_client_attrs, + .sysfs_ops = &kobj_sysfs_ops, + .namespace = nfs_netns_client_namespace, +}; + +static struct nfs_netns_client *nfs_netns_client_alloc(struct kobject *parent, + struct net *net) +{ + struct nfs_netns_client *p; + + p = kzalloc(sizeof(*p), GFP_KERNEL); + if (p) { + p->net = net; + p->kobject.kset = nfs_client_kset; + if (kobject_init_and_add(&p->kobject, &nfs_netns_client_type, + parent, "nfs_client") == 0) + return p; + kobject_put(&p->kobject); + } + return NULL; +} + +void nfs_netns_sysfs_setup(struct nfs_net *netns, struct net *net) +{ + struct nfs_netns_client *clp; + + clp = nfs_netns_client_alloc(nfs_client_kobj, net); + if (clp) { + netns->nfs_client = clp; + kobject_uevent(&clp->kobject, KOBJ_ADD); + } +} + +void nfs_netns_sysfs_destroy(struct nfs_net *netns) +{ + struct nfs_netns_client *clp = netns->nfs_client; + + if (clp) { + kobject_uevent(&clp->kobject, KOBJ_REMOVE); + kobject_del(&clp->kobject); + kobject_put(&clp->kobject); + netns->nfs_client = NULL; + } +} diff --git a/fs/nfs/sysfs.h b/fs/nfs/sysfs.h new file mode 100644 index 000000000000..f1b27411dcc0 --- /dev/null +++ b/fs/nfs/sysfs.h @@ -0,0 +1,25 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2019 Hammerspace Inc + */ + +#ifndef __NFS_SYSFS_H +#define __NFS_SYSFS_H + +#define CONTAINER_ID_MAXLEN (64) + +struct nfs_netns_client { + struct kobject kobject; + struct net *net; + const char *identifier; +}; + +extern struct kobject *nfs_client_kobj; + +extern int nfs_sysfs_init(void); +extern void nfs_sysfs_exit(void); + +void nfs_netns_sysfs_setup(struct nfs_net *netns, struct net *net); +void nfs_netns_sysfs_destroy(struct nfs_net *netns); + +#endif diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 059a7c38bc4f..92d9cadc6102 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -103,7 +103,7 @@ EXPORT_SYMBOL_GPL(nfs_commit_free); static struct nfs_pgio_header *nfs_writehdr_alloc(void) { - struct nfs_pgio_header *p = mempool_alloc(nfs_wdata_mempool, GFP_NOIO); + struct nfs_pgio_header *p = mempool_alloc(nfs_wdata_mempool, GFP_KERNEL); memset(p, 0, sizeof(*p)); p->rw_mode = FMODE_WRITE; @@ -721,12 +721,11 @@ int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc) struct inode *inode = mapping->host; struct nfs_pageio_descriptor pgio; struct nfs_io_completion *ioc; - unsigned int pflags = memalloc_nofs_save(); int err; nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGES); - ioc = nfs_io_completion_alloc(GFP_NOFS); + ioc = nfs_io_completion_alloc(GFP_KERNEL); if (ioc) nfs_io_completion_init(ioc, nfs_io_completion_commit, inode); @@ -737,8 +736,6 @@ int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc) nfs_pageio_complete(&pgio); nfs_io_completion_put(ioc); - memalloc_nofs_restore(pflags); - if (err < 0) goto out_err; err = pgio.pg_error; diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index b74a47169297..06b68b6115bc 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -49,6 +49,7 @@ xfs-y += $(addprefix libxfs/, \ xfs_refcount_btree.o \ xfs_sb.o \ xfs_symlink_remote.o \ + xfs_trans_inode.o \ xfs_trans_resv.o \ xfs_types.o \ ) @@ -107,8 +108,7 @@ xfs-y += xfs_log.o \ xfs_rmap_item.o \ xfs_log_recover.o \ xfs_trans_ail.o \ - xfs_trans_buf.o \ - xfs_trans_inode.o + xfs_trans_buf.o # optional features xfs-$(CONFIG_XFS_QUOTA) += xfs_dquot.o \ diff --git a/fs/xfs/xfs_trans_inode.c b/fs/xfs/libxfs/xfs_trans_inode.c index 93d14e47269d..a9ad90926b87 100644 --- a/fs/xfs/xfs_trans_inode.c +++ b/fs/xfs/libxfs/xfs_trans_inode.c @@ -66,6 +66,10 @@ xfs_trans_ichgtime( inode->i_mtime = tv; if (flags & XFS_ICHGTIME_CHG) inode->i_ctime = tv; + if (flags & XFS_ICHGTIME_CREATE) { + ip->i_d.di_crtime.t_sec = (int32_t)tv.tv_sec; + ip->i_d.di_crtime.t_nsec = (int32_t)tv.tv_nsec; + } } /* diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index e93bacbd49ae..28101bbc0b78 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -1197,11 +1197,14 @@ xfs_file_mmap( struct file *filp, struct vm_area_struct *vma) { + struct dax_device *dax_dev; + + dax_dev = xfs_find_daxdev_for_inode(file_inode(filp)); /* - * We don't support synchronous mappings for non-DAX files. At least - * until someone comes with a sensible use case. + * We don't support synchronous mappings for non-DAX files and + * for DAX files if underneath dax_device is not synchronous. */ - if (!IS_DAX(file_inode(filp)) && (vma->vm_flags & VM_SYNC)) + if (!daxdev_mapping_supported(vma, dax_dev)) return -EOPNOTSUPP; file_accessed(filp); |