diff --git a/fs/smb/client/cifs_debug.c b/fs/smb/client/cifs_debug.c index 6d8804fb6535..5596c9f30ccb 100644 --- a/fs/smb/client/cifs_debug.c +++ b/fs/smb/client/cifs_debug.c @@ -136,6 +136,11 @@ cifs_dump_channel(struct seq_file *m, int i, struct cifs_chan *chan) { struct TCP_Server_Info *server = chan->server; + if (!server) { + seq_printf(m, "\n\n\t\tChannel: %d DISABLED", i+1); + return; + } + seq_printf(m, "\n\n\t\tChannel: %d ConnectionId: 0x%llx" "\n\t\tNumber of credits: %d,%d,%d Dialect 0x%x" "\n\t\tTCP status: %d Instance: %d" @@ -279,6 +284,8 @@ static int cifs_debug_data_proc_show(struct seq_file *m, void *v) struct cifs_ses *ses; struct cifs_tcon *tcon; struct cifs_server_iface *iface; + size_t iface_weight = 0, iface_min_speed = 0; + struct cifs_server_iface *last_iface = NULL; int c, i, j; seq_puts(m, @@ -544,11 +551,25 @@ skip_rdma: "\tLast updated: %lu seconds ago", ses->iface_count, (jiffies - ses->iface_last_update) / HZ); + + last_iface = list_last_entry(&ses->iface_list, + struct cifs_server_iface, + iface_head); + iface_min_speed = last_iface->speed; + j = 0; list_for_each_entry(iface, &ses->iface_list, iface_head) { seq_printf(m, "\n\t%d)", ++j); cifs_dump_iface(m, iface); + + iface_weight = iface->speed / iface_min_speed; + seq_printf(m, "\t\tWeight (cur,total): (%zu,%zu)" + "\n\t\tAllocated channels: %u\n", + iface->weight_fulfilled, + iface_weight, + iface->num_channels); + if (is_ses_using_iface(ses, iface)) seq_puts(m, "\t\t[CONNECTED]\n"); } @@ -746,14 +767,14 @@ static ssize_t name##_write(struct file *file, const char __user *buffer, \ size_t count, loff_t *ppos) \ { \ int rc; \ - rc = kstrtoint_from_user(buffer, count, 10, & name); \ + rc = kstrtoint_from_user(buffer, count, 10, &name); \ if (rc) \ return rc; \ return count; \ } \ static int name##_proc_show(struct seq_file *m, void *v) \ { \ - seq_printf(m, "%d\n", name ); \ + seq_printf(m, "%d\n", name); \ return 0; \ } \ static int name##_open(struct inode *inode, struct file *file) \ diff --git a/fs/smb/client/cifs_ioctl.h b/fs/smb/client/cifs_ioctl.h index 332588e77c31..26327442e383 100644 --- a/fs/smb/client/cifs_ioctl.h +++ b/fs/smb/client/cifs_ioctl.h @@ -26,6 +26,11 @@ struct smb_mnt_fs_info { __u64 cifs_posix_caps; } __packed; +struct smb_mnt_tcon_info { + __u32 tid; + __u64 session_id; +} __packed; + struct smb_snapshot_array { __u32 number_of_snapshots; __u32 number_of_snapshots_returned; @@ -108,6 +113,7 @@ struct smb3_notify_info { #define CIFS_IOC_NOTIFY _IOW(CIFS_IOCTL_MAGIC, 9, struct smb3_notify) #define CIFS_DUMP_FULL_KEY _IOWR(CIFS_IOCTL_MAGIC, 10, struct smb3_full_key_debug_info) #define CIFS_IOC_NOTIFY_INFO _IOWR(CIFS_IOCTL_MAGIC, 11, struct smb3_notify_info) +#define CIFS_IOC_GET_TCON_INFO _IOR(CIFS_IOCTL_MAGIC, 12, struct smb_mnt_tcon_info) #define CIFS_IOC_SHUTDOWN _IOR('X', 125, __u32) /* diff --git a/fs/smb/client/cifsfs.h b/fs/smb/client/cifsfs.h index 8ca3d7606bb4..3adea10aa9da 100644 --- a/fs/smb/client/cifsfs.h +++ b/fs/smb/client/cifsfs.h @@ -152,6 +152,6 @@ extern const struct export_operations cifs_export_ops; #endif /* CONFIG_CIFS_NFSD_EXPORT */ /* when changing internal version - update following two lines at same time */ -#define SMB3_PRODUCT_BUILD 45 -#define CIFS_VERSION "2.45" +#define SMB3_PRODUCT_BUILD 46 +#define CIFS_VERSION "2.46" #endif /* _CIFSFS_H */ diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index 02082621d8e0..6ffbd81bd109 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -650,6 +650,7 @@ struct TCP_Server_Info { bool noautotune; /* do not autotune send buf sizes */ bool nosharesock; bool tcp_nodelay; + bool terminate; unsigned int credits; /* send no more requests at once */ unsigned int max_credits; /* can override large 32000 default at mnt */ unsigned int in_flight; /* number of requests on the wire to server */ @@ -969,6 +970,8 @@ struct cifs_server_iface { struct list_head iface_head; struct kref refcount; size_t speed; + size_t weight_fulfilled; + unsigned int num_channels; unsigned int rdma_capable : 1; unsigned int rss_capable : 1; unsigned int is_active : 1; /* unset if non existent */ @@ -1050,6 +1053,7 @@ struct cifs_ses { spinlock_t chan_lock; /* ========= begin: protected by chan_lock ======== */ #define CIFS_MAX_CHANNELS 16 +#define CIFS_INVAL_CHAN_INDEX (-1) #define CIFS_ALL_CHANNELS_SET(ses) \ ((1UL << (ses)->chan_count) - 1) #define CIFS_ALL_CHANS_GOOD(ses) \ @@ -2143,6 +2147,7 @@ static inline int cifs_get_num_sgs(const struct smb_rqst *rqst, unsigned int len, skip; unsigned int nents = 0; unsigned long addr; + size_t data_size; int i, j; /* @@ -2158,17 +2163,21 @@ static inline int cifs_get_num_sgs(const struct smb_rqst *rqst, * rqst[1+].rq_iov[0+] data to be encrypted/decrypted */ for (i = 0; i < num_rqst; i++) { + data_size = iov_iter_count(&rqst[i].rq_iter); + /* We really don't want a mixture of pinned and unpinned pages * in the sglist. It's hard to keep track of which is what. * Instead, we convert to a BVEC-type iterator higher up. */ - if (WARN_ON_ONCE(user_backed_iter(&rqst[i].rq_iter))) + if (data_size && + WARN_ON_ONCE(user_backed_iter(&rqst[i].rq_iter))) return -EIO; /* We also don't want to have any extra refs or pins to clean * up in the sglist. */ - if (WARN_ON_ONCE(iov_iter_extract_will_pin(&rqst[i].rq_iter))) + if (data_size && + WARN_ON_ONCE(iov_iter_extract_will_pin(&rqst[i].rq_iter))) return -EIO; for (j = 0; j < rqst[i].rq_nvec; j++) { @@ -2184,7 +2193,8 @@ static inline int cifs_get_num_sgs(const struct smb_rqst *rqst, } skip = 0; } - nents += iov_iter_npages(&rqst[i].rq_iter, INT_MAX); + if (data_size) + nents += iov_iter_npages(&rqst[i].rq_iter, INT_MAX); } nents += DIV_ROUND_UP(offset_in_page(sig) + SMB2_SIGNATURE_SIZE, PAGE_SIZE); return nents; diff --git a/fs/smb/client/cifsproto.h b/fs/smb/client/cifsproto.h index 890ceddae07e..d87e2c26cce2 100644 --- a/fs/smb/client/cifsproto.h +++ b/fs/smb/client/cifsproto.h @@ -132,6 +132,7 @@ extern int SendReceiveBlockingLock(const unsigned int xid, struct smb_hdr *in_buf, struct smb_hdr *out_buf, int *bytes_returned); + void cifs_signal_cifsd_for_reconnect(struct TCP_Server_Info *server, bool all_channels); @@ -610,13 +611,13 @@ void cifs_free_hash(struct shash_desc **sdesc); struct cifs_chan * cifs_ses_find_chan(struct cifs_ses *ses, struct TCP_Server_Info *server); -int cifs_try_adding_channels(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses); +int cifs_try_adding_channels(struct cifs_ses *ses); bool is_server_using_iface(struct TCP_Server_Info *server, struct cifs_server_iface *iface); bool is_ses_using_iface(struct cifs_ses *ses, struct cifs_server_iface *iface); void cifs_ses_mark_for_reconnect(struct cifs_ses *ses); -unsigned int +int cifs_ses_get_chan_index(struct cifs_ses *ses, struct TCP_Server_Info *server); void @@ -640,6 +641,8 @@ cifs_chan_needs_reconnect(struct cifs_ses *ses, bool cifs_chan_is_iface_active(struct cifs_ses *ses, struct TCP_Server_Info *server); +void +cifs_disable_secondary_channels(struct cifs_ses *ses); int cifs_chan_update_iface(struct cifs_ses *ses, struct TCP_Server_Info *server); int diff --git a/fs/smb/client/connect.c b/fs/smb/client/connect.c index 1a137b33858a..57c2a7df3457 100644 --- a/fs/smb/client/connect.c +++ b/fs/smb/client/connect.c @@ -132,6 +132,9 @@ static void smb2_query_server_interfaces(struct work_struct *work) free_xid(xid); if (rc) { + if (rc == -EOPNOTSUPP) + return; + cifs_dbg(FYI, "%s: failed to query server interfaces: %d\n", __func__, rc); } @@ -173,8 +176,12 @@ cifs_signal_cifsd_for_reconnect(struct TCP_Server_Info *server, list_for_each_entry(ses, &pserver->smb_ses_list, smb_ses_list) { spin_lock(&ses->chan_lock); for (i = 0; i < ses->chan_count; i++) { + if (!ses->chans[i].server) + continue; + spin_lock(&ses->chans[i].server->srv_lock); - ses->chans[i].server->tcpStatus = CifsNeedReconnect; + if (ses->chans[i].server->tcpStatus != CifsExiting) + ses->chans[i].server->tcpStatus = CifsNeedReconnect; spin_unlock(&ses->chans[i].server->srv_lock); } spin_unlock(&ses->chan_lock); @@ -212,6 +219,14 @@ cifs_mark_tcp_ses_conns_for_reconnect(struct TCP_Server_Info *server, spin_lock(&cifs_tcp_ses_lock); list_for_each_entry_safe(ses, nses, &pserver->smb_ses_list, smb_ses_list) { + /* + * if channel has been marked for termination, nothing to do + * for the channel. in fact, we cannot find the channel for the + * server. So safe to exit here + */ + if (server->terminate) + break; + /* check if iface is still active */ if (!cifs_chan_is_iface_active(ses, server)) cifs_chan_update_iface(ses, server); @@ -246,6 +261,8 @@ cifs_mark_tcp_ses_conns_for_reconnect(struct TCP_Server_Info *server, spin_lock(&tcon->tc_lock); tcon->status = TID_NEED_RECON; spin_unlock(&tcon->tc_lock); + + cancel_delayed_work(&tcon->query_interfaces); } if (ses->tcon_ipc) { ses->tcon_ipc->need_reconnect = true; @@ -385,7 +402,13 @@ static int __cifs_reconnect(struct TCP_Server_Info *server, spin_unlock(&server->srv_lock); cifs_swn_reset_server_dstaddr(server); cifs_server_unlock(server); - mod_delayed_work(cifsiod_wq, &server->reconnect, 0); + + /* increase ref count which reconnect work will drop */ + spin_lock(&cifs_tcp_ses_lock); + server->srv_count++; + spin_unlock(&cifs_tcp_ses_lock); + if (mod_delayed_work(cifsiod_wq, &server->reconnect, 0)) + cifs_put_tcp_session(server, false); } } while (server->tcpStatus == CifsNeedReconnect); @@ -515,7 +538,13 @@ static int reconnect_dfs_server(struct TCP_Server_Info *server) spin_unlock(&server->srv_lock); cifs_swn_reset_server_dstaddr(server); cifs_server_unlock(server); - mod_delayed_work(cifsiod_wq, &server->reconnect, 0); + + /* increase ref count which reconnect work will drop */ + spin_lock(&cifs_tcp_ses_lock); + server->srv_count++; + spin_unlock(&cifs_tcp_ses_lock); + if (mod_delayed_work(cifsiod_wq, &server->reconnect, 0)) + cifs_put_tcp_session(server, false); } while (server->tcpStatus == CifsNeedReconnect); mutex_lock(&server->refpath_lock); @@ -1597,16 +1626,19 @@ cifs_put_tcp_session(struct TCP_Server_Info *server, int from_reconnect) cancel_delayed_work_sync(&server->echo); - if (from_reconnect) + if (from_reconnect) { /* * Avoid deadlock here: reconnect work calls * cifs_put_tcp_session() at its end. Need to be sure * that reconnect work does nothing with server pointer after * that step. */ - cancel_delayed_work(&server->reconnect); - else - cancel_delayed_work_sync(&server->reconnect); + if (cancel_delayed_work(&server->reconnect)) + cifs_put_tcp_session(server, from_reconnect); + } else { + if (cancel_delayed_work_sync(&server->reconnect)) + cifs_put_tcp_session(server, from_reconnect); + } spin_lock(&server->srv_lock); server->tcpStatus = CifsExiting; @@ -3560,7 +3592,7 @@ int cifs_mount(struct cifs_sb_info *cifs_sb, struct smb3_fs_context *ctx) ctx->prepath = NULL; out: - cifs_try_adding_channels(cifs_sb, mnt_ctx.ses); + cifs_try_adding_channels(mnt_ctx.ses); rc = mount_setup_tlink(cifs_sb, mnt_ctx.ses, mnt_ctx.tcon); if (rc) goto error; diff --git a/fs/smb/client/dfs.c b/fs/smb/client/dfs.c index 81b84151450d..a8a1d386da65 100644 --- a/fs/smb/client/dfs.c +++ b/fs/smb/client/dfs.c @@ -263,15 +263,23 @@ out: return rc; } -/* Resolve UNC hostname in @ctx->source and set ip addr in @ctx->dstaddr */ +/* + * If @ctx->dfs_automount, then update @ctx->dstaddr earlier with the DFS root + * server from where we'll start following any referrals. Otherwise rely on the + * value provided by mount(2) as the user might not have dns_resolver key set up + * and therefore failing to upcall to resolve UNC hostname under @ctx->source. + */ static int update_fs_context_dstaddr(struct smb3_fs_context *ctx) { struct sockaddr *addr = (struct sockaddr *)&ctx->dstaddr; - int rc; + int rc = 0; - rc = dns_resolve_server_name_to_ip(ctx->source, addr, NULL); - if (!rc) - cifs_set_port(addr, ctx->port); + if (!ctx->nodfs && ctx->dfs_automount) { + rc = dns_resolve_server_name_to_ip(ctx->source, addr, NULL); + if (!rc) + cifs_set_port(addr, ctx->port); + ctx->dfs_automount = false; + } return rc; } diff --git a/fs/smb/client/fs_context.h b/fs/smb/client/fs_context.h index 9d8d34af0211..cf46916286d0 100644 --- a/fs/smb/client/fs_context.h +++ b/fs/smb/client/fs_context.h @@ -268,6 +268,7 @@ struct smb3_fs_context { bool witness:1; /* use witness protocol */ char *leaf_fullpath; struct cifs_ses *dfs_root_ses; + bool dfs_automount:1; /* set for dfs automount only */ }; extern const struct fs_parameter_spec smb3_fs_parameters[]; diff --git a/fs/smb/client/ioctl.c b/fs/smb/client/ioctl.c index f7160003e0ed..e2f92c21fff5 100644 --- a/fs/smb/client/ioctl.c +++ b/fs/smb/client/ioctl.c @@ -117,6 +117,20 @@ out_drop_write: return rc; } +static long smb_mnt_get_tcon_info(struct cifs_tcon *tcon, void __user *arg) +{ + int rc = 0; + struct smb_mnt_tcon_info tcon_inf; + + tcon_inf.tid = tcon->tid; + tcon_inf.session_id = tcon->ses->Suid; + + if (copy_to_user(arg, &tcon_inf, sizeof(struct smb_mnt_tcon_info))) + rc = -EFAULT; + + return rc; +} + static long smb_mnt_get_fsinfo(unsigned int xid, struct cifs_tcon *tcon, void __user *arg) { @@ -129,6 +143,7 @@ static long smb_mnt_get_fsinfo(unsigned int xid, struct cifs_tcon *tcon, fsinf->version = 1; fsinf->protocol_id = tcon->ses->server->vals->protocol_id; + fsinf->tcon_flags = tcon->Flags; fsinf->device_characteristics = le32_to_cpu(tcon->fsDevInfo.DeviceCharacteristics); fsinf->device_type = le32_to_cpu(tcon->fsDevInfo.DeviceType); @@ -414,6 +429,17 @@ long cifs_ioctl(struct file *filep, unsigned int command, unsigned long arg) tcon = tlink_tcon(pSMBFile->tlink); rc = smb_mnt_get_fsinfo(xid, tcon, (void __user *)arg); break; + case CIFS_IOC_GET_TCON_INFO: + cifs_sb = CIFS_SB(inode->i_sb); + tlink = cifs_sb_tlink(cifs_sb); + if (IS_ERR(tlink)) { + rc = PTR_ERR(tlink); + break; + } + tcon = tlink_tcon(tlink); + rc = smb_mnt_get_tcon_info(tcon, (void __user *)arg); + cifs_put_tlink(tlink); + break; case CIFS_ENUMERATE_SNAPSHOTS: if (pSMBFile == NULL) break; diff --git a/fs/smb/client/namespace.c b/fs/smb/client/namespace.c index c8f5ed8a69f1..a6968573b775 100644 --- a/fs/smb/client/namespace.c +++ b/fs/smb/client/namespace.c @@ -117,6 +117,18 @@ cifs_build_devname(char *nodename, const char *prepath) return dev; } +static bool is_dfs_mount(struct dentry *dentry) +{ + struct cifs_sb_info *cifs_sb = CIFS_SB(dentry->d_sb); + struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); + bool ret; + + spin_lock(&tcon->tc_lock); + ret = !!tcon->origin_fullpath; + spin_unlock(&tcon->tc_lock); + return ret; +} + /* Return full path out of a dentry set for automount */ static char *automount_fullpath(struct dentry *dentry, void *page) { @@ -212,8 +224,9 @@ static struct vfsmount *cifs_do_automount(struct path *path) ctx->source = NULL; goto out; } - cifs_dbg(FYI, "%s: ctx: source=%s UNC=%s prepath=%s\n", - __func__, ctx->source, ctx->UNC, ctx->prepath); + ctx->dfs_automount = is_dfs_mount(mntpt); + cifs_dbg(FYI, "%s: ctx: source=%s UNC=%s prepath=%s dfs_automount=%d\n", + __func__, ctx->source, ctx->UNC, ctx->prepath, ctx->dfs_automount); mnt = fc_mount(fc); out: diff --git a/fs/smb/client/sess.c b/fs/smb/client/sess.c index cd474cf98f30..0bb2ac929061 100644 --- a/fs/smb/client/sess.c +++ b/fs/smb/client/sess.c @@ -24,7 +24,7 @@ #include "fs_context.h" static int -cifs_ses_add_channel(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses, +cifs_ses_add_channel(struct cifs_ses *ses, struct cifs_server_iface *iface); bool @@ -69,7 +69,7 @@ bool is_ses_using_iface(struct cifs_ses *ses, struct cifs_server_iface *iface) /* channel helper functions. assumed that chan_lock is held by caller. */ -unsigned int +int cifs_ses_get_chan_index(struct cifs_ses *ses, struct TCP_Server_Info *server) { @@ -85,14 +85,17 @@ cifs_ses_get_chan_index(struct cifs_ses *ses, cifs_dbg(VFS, "unable to get chan index for server: 0x%llx", server->conn_id); WARN_ON(1); - return 0; + return CIFS_INVAL_CHAN_INDEX; } void cifs_chan_set_in_reconnect(struct cifs_ses *ses, struct TCP_Server_Info *server) { - unsigned int chan_index = cifs_ses_get_chan_index(ses, server); + int chan_index = cifs_ses_get_chan_index(ses, server); + + if (chan_index == CIFS_INVAL_CHAN_INDEX) + return; ses->chans[chan_index].in_reconnect = true; } @@ -103,6 +106,9 @@ cifs_chan_clear_in_reconnect(struct cifs_ses *ses, { unsigned int chan_index = cifs_ses_get_chan_index(ses, server); + if (chan_index == CIFS_INVAL_CHAN_INDEX) + return; + ses->chans[chan_index].in_reconnect = false; } @@ -112,6 +118,9 @@ cifs_chan_in_reconnect(struct cifs_ses *ses, { unsigned int chan_index = cifs_ses_get_chan_index(ses, server); + if (chan_index == CIFS_INVAL_CHAN_INDEX) + return true; /* err on the safer side */ + return CIFS_CHAN_IN_RECONNECT(ses, chan_index); } @@ -121,6 +130,9 @@ cifs_chan_set_need_reconnect(struct cifs_ses *ses, { unsigned int chan_index = cifs_ses_get_chan_index(ses, server); + if (chan_index == CIFS_INVAL_CHAN_INDEX) + return; + set_bit(chan_index, &ses->chans_need_reconnect); cifs_dbg(FYI, "Set reconnect bitmask for chan %u; now 0x%lx\n", chan_index, ses->chans_need_reconnect); @@ -132,6 +144,9 @@ cifs_chan_clear_need_reconnect(struct cifs_ses *ses, { unsigned int chan_index = cifs_ses_get_chan_index(ses, server); + if (chan_index == CIFS_INVAL_CHAN_INDEX) + return; + clear_bit(chan_index, &ses->chans_need_reconnect); cifs_dbg(FYI, "Cleared reconnect bitmask for chan %u; now 0x%lx\n", chan_index, ses->chans_need_reconnect); @@ -143,6 +158,9 @@ cifs_chan_needs_reconnect(struct cifs_ses *ses, { unsigned int chan_index = cifs_ses_get_chan_index(ses, server); + if (chan_index == CIFS_INVAL_CHAN_INDEX) + return true; /* err on the safer side */ + return CIFS_CHAN_NEEDS_RECONNECT(ses, chan_index); } @@ -152,19 +170,24 @@ cifs_chan_is_iface_active(struct cifs_ses *ses, { unsigned int chan_index = cifs_ses_get_chan_index(ses, server); + if (chan_index == CIFS_INVAL_CHAN_INDEX) + return true; /* err on the safer side */ + return ses->chans[chan_index].iface && ses->chans[chan_index].iface->is_active; } /* returns number of channels added */ -int cifs_try_adding_channels(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses) +int cifs_try_adding_channels(struct cifs_ses *ses) { struct TCP_Server_Info *server = ses->server; int old_chan_count, new_chan_count; int left; int rc = 0; int tries = 0; + size_t iface_weight = 0, iface_min_speed = 0; struct cifs_server_iface *iface = NULL, *niface = NULL; + struct cifs_server_iface *last_iface = NULL; spin_lock(&ses->chan_lock); @@ -192,21 +215,11 @@ int cifs_try_adding_channels(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses) } spin_unlock(&ses->chan_lock); - /* - * Keep connecting to same, fastest, iface for all channels as - * long as its RSS. Try next fastest one if not RSS or channel - * creation fails. - */ - spin_lock(&ses->iface_lock); - iface = list_first_entry(&ses->iface_list, struct cifs_server_iface, - iface_head); - spin_unlock(&ses->iface_lock); - while (left > 0) { tries++; if (tries > 3*ses->chan_max) { - cifs_dbg(FYI, "too many channel open attempts (%d channels left to open)\n", + cifs_dbg(VFS, "too many channel open attempts (%d channels left to open)\n", left); break; } @@ -214,23 +227,41 @@ int cifs_try_adding_channels(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses) spin_lock(&ses->iface_lock); if (!ses->iface_count) { spin_unlock(&ses->iface_lock); + cifs_dbg(VFS, "server %s does not advertise interfaces\n", + ses->server->hostname); break; } + if (!iface) + iface = list_first_entry(&ses->iface_list, struct cifs_server_iface, + iface_head); + last_iface = list_last_entry(&ses->iface_list, struct cifs_server_iface, + iface_head); + iface_min_speed = last_iface->speed; + list_for_each_entry_safe_from(iface, niface, &ses->iface_list, iface_head) { + /* do not mix rdma and non-rdma interfaces */ + if (iface->rdma_capable != ses->server->rdma) + continue; + /* skip ifaces that are unusable */ if (!iface->is_active || (is_ses_using_iface(ses, iface) && - !iface->rss_capable)) { + !iface->rss_capable)) + continue; + + /* check if we already allocated enough channels */ + iface_weight = iface->speed / iface_min_speed; + + if (iface->weight_fulfilled >= iface_weight) continue; - } /* take ref before unlock */ kref_get(&iface->refcount); spin_unlock(&ses->iface_lock); - rc = cifs_ses_add_channel(cifs_sb, ses, iface); + rc = cifs_ses_add_channel(ses, iface); spin_lock(&ses->iface_lock); if (rc) { @@ -241,10 +272,21 @@ int cifs_try_adding_channels(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses) continue; } - cifs_dbg(FYI, "successfully opened new channel on iface:%pIS\n", + iface->num_channels++; + iface->weight_fulfilled++; + cifs_dbg(VFS, "successfully opened new channel on iface:%pIS\n", &iface->sockaddr); break; } + + /* reached end of list. reset weight_fulfilled and start over */ + if (list_entry_is_head(iface, &ses->iface_list, iface_head)) { + list_for_each_entry(iface, &ses->iface_list, iface_head) + iface->weight_fulfilled = 0; + spin_unlock(&ses->iface_lock); + iface = NULL; + continue; + } spin_unlock(&ses->iface_lock); left--; @@ -254,6 +296,60 @@ int cifs_try_adding_channels(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses) return new_chan_count - old_chan_count; } +/* + * called when multichannel is disabled by the server. + * this always gets called from smb2_reconnect + * and cannot get called in parallel threads. + */ +void +cifs_disable_secondary_channels(struct cifs_ses *ses) +{ + int i, chan_count; + struct TCP_Server_Info *server; + struct cifs_server_iface *iface; + + spin_lock(&ses->chan_lock); + chan_count = ses->chan_count; + if (chan_count == 1) + goto done; + + ses->chan_count = 1; + + /* for all secondary channels reset the need reconnect bit */ + ses->chans_need_reconnect &= 1; + + for (i = 1; i < chan_count; i++) { + iface = ses->chans[i].iface; + server = ses->chans[i].server; + + if (iface) { + spin_lock(&ses->iface_lock); + kref_put(&iface->refcount, release_iface); + ses->chans[i].iface = NULL; + iface->num_channels--; + if (iface->weight_fulfilled) + iface->weight_fulfilled--; + spin_unlock(&ses->iface_lock); + } + + spin_unlock(&ses->chan_lock); + if (server && !server->terminate) { + server->terminate = true; + cifs_signal_cifsd_for_reconnect(server, false); + } + spin_lock(&ses->chan_lock); + + if (server) { + ses->chans[i].server = NULL; + cifs_put_tcp_session(server, false); + } + + } + +done: + spin_unlock(&ses->chan_lock); +} + /* * update the iface for the channel if necessary. * will return 0 when iface is updated, 1 if removed, 2 otherwise @@ -263,13 +359,16 @@ int cifs_chan_update_iface(struct cifs_ses *ses, struct TCP_Server_Info *server) { unsigned int chan_index; + size_t iface_weight = 0, iface_min_speed = 0; struct cifs_server_iface *iface = NULL; struct cifs_server_iface *old_iface = NULL; + struct cifs_server_iface *last_iface = NULL; + struct sockaddr_storage ss; int rc = 0; spin_lock(&ses->chan_lock); chan_index = cifs_ses_get_chan_index(ses, server); - if (!chan_index) { + if (chan_index == CIFS_INVAL_CHAN_INDEX) { spin_unlock(&ses->chan_lock); return 0; } @@ -283,14 +382,49 @@ cifs_chan_update_iface(struct cifs_ses *ses, struct TCP_Server_Info *server) } spin_unlock(&ses->chan_lock); + spin_lock(&server->srv_lock); + ss = server->dstaddr; + spin_unlock(&server->srv_lock); + spin_lock(&ses->iface_lock); + if (!ses->iface_count) { + spin_unlock(&ses->iface_lock); + cifs_dbg(VFS, "server %s does not advertise interfaces\n", ses->server->hostname); + return 0; + } + + last_iface = list_last_entry(&ses->iface_list, struct cifs_server_iface, + iface_head); + iface_min_speed = last_iface->speed; + /* then look for a new one */ list_for_each_entry(iface, &ses->iface_list, iface_head) { + if (!chan_index) { + /* if we're trying to get the updated iface for primary channel */ + if (!cifs_match_ipaddr((struct sockaddr *) &ss, + (struct sockaddr *) &iface->sockaddr)) + continue; + + kref_get(&iface->refcount); + break; + } + + /* do not mix rdma and non-rdma interfaces */ + if (iface->rdma_capable != server->rdma) + continue; + if (!iface->is_active || (is_ses_using_iface(ses, iface) && !iface->rss_capable)) { continue; } + + /* check if we already allocated enough channels */ + iface_weight = iface->speed / iface_min_speed; + + if (iface->weight_fulfilled >= iface_weight) + continue; + kref_get(&iface->refcount); break; } @@ -301,16 +435,41 @@ cifs_chan_update_iface(struct cifs_ses *ses, struct TCP_Server_Info *server) cifs_dbg(FYI, "unable to find a suitable iface\n"); } + if (!chan_index && !iface) { + cifs_dbg(FYI, "unable to get the interface matching: %pIS\n", + &ss); + spin_unlock(&ses->iface_lock); + return 0; + } + /* now drop the ref to the current iface */ if (old_iface && iface) { cifs_dbg(FYI, "replacing iface: %pIS with %pIS\n", &old_iface->sockaddr, &iface->sockaddr); + + old_iface->num_channels--; + if (old_iface->weight_fulfilled) + old_iface->weight_fulfilled--; + iface->num_channels++; + iface->weight_fulfilled++; + kref_put(&old_iface->refcount, release_iface); } else if (old_iface) { cifs_dbg(FYI, "releasing ref to iface: %pIS\n", &old_iface->sockaddr); + + old_iface->num_channels--; + if (old_iface->weight_fulfilled) + old_iface->weight_fulfilled--; + kref_put(&old_iface->refcount, release_iface); + } else if (!chan_index) { + /* special case: update interface for primary channel */ + cifs_dbg(FYI, "referencing primary channel iface: %pIS\n", + &iface->sockaddr); + iface->num_channels++; + iface->weight_fulfilled++; } else { WARN_ON(!iface); cifs_dbg(FYI, "adding new iface: %pIS\n", &iface->sockaddr); @@ -319,6 +478,11 @@ cifs_chan_update_iface(struct cifs_ses *ses, struct TCP_Server_Info *server) spin_lock(&ses->chan_lock); chan_index = cifs_ses_get_chan_index(ses, server); + if (chan_index == CIFS_INVAL_CHAN_INDEX) { + spin_unlock(&ses->chan_lock); + return 0; + } + ses->chans[chan_index].iface = iface; /* No iface is found. if secondary chan, drop connection */ @@ -354,7 +518,7 @@ cifs_ses_find_chan(struct cifs_ses *ses, struct TCP_Server_Info *server) } static int -cifs_ses_add_channel(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses, +cifs_ses_add_channel(struct cifs_ses *ses, struct cifs_server_iface *iface) { struct TCP_Server_Info *chan_server; @@ -433,7 +597,7 @@ cifs_ses_add_channel(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses, * This will be used for encoding/decoding user/domain/pw * during sess setup auth. */ - ctx->local_nls = cifs_sb->local_nls; + ctx->local_nls = ses->local_nls; /* Use RDMA if possible */ ctx->rdma = iface->rdma_capable; @@ -479,20 +643,16 @@ cifs_ses_add_channel(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses, rc = cifs_negotiate_protocol(xid, ses, chan->server); if (!rc) - rc = cifs_setup_session(xid, ses, chan->server, cifs_sb->local_nls); + rc = cifs_setup_session(xid, ses, chan->server, ses->local_nls); mutex_unlock(&ses->session_mutex); out: if (rc && chan->server) { - /* - * we should avoid race with these delayed works before we - * remove this channel - */ - cancel_delayed_work_sync(&chan->server->echo); - cancel_delayed_work_sync(&chan->server->reconnect); + cifs_put_tcp_session(chan->server, 0); spin_lock(&ses->chan_lock); + /* we rely on all bits beyond chan_count to be clear */ cifs_chan_clear_need_reconnect(ses, chan->server); ses->chan_count--; @@ -502,8 +662,6 @@ out: */ WARN_ON(ses->chan_count < 1); spin_unlock(&ses->chan_lock); - - cifs_put_tcp_session(chan->server, 0); } kfree(ctx->UNC); @@ -535,8 +693,7 @@ static __u32 cifs_ssetup_hdr(struct cifs_ses *ses, /* Now no need to set SMBFLG_CASELESS or obsolete CANONICAL PATH */ - /* BB verify whether signing required on neg or just on auth frame - (and NTLM case) */ + /* BB verify whether signing required on neg or just auth frame (and NTLM case) */ capabilities = CAP_LARGE_FILES | CAP_NT_SMBS | CAP_LEVEL_II_OPLOCKS | CAP_LARGE_WRITE_X | CAP_LARGE_READ_X; @@ -593,8 +750,10 @@ static void unicode_domain_string(char **pbcc_area, struct cifs_ses *ses, /* copy domain */ if (ses->domainName == NULL) { - /* Sending null domain better than using a bogus domain name (as - we did briefly in 2.6.18) since server will use its default */ + /* + * Sending null domain better than using a bogus domain name (as + * we did briefly in 2.6.18) since server will use its default + */ *bcc_ptr = 0; *(bcc_ptr+1) = 0; bytes_ret = 0; @@ -613,8 +772,7 @@ static void unicode_ssetup_strings(char **pbcc_area, struct cifs_ses *ses, char *bcc_ptr = *pbcc_area; int bytes_ret = 0; - /* BB FIXME add check that strings total less - than 335 or will need to send them as arrays */ + /* BB FIXME add check that strings less than 335 or will need to send as arrays */ /* copy user */ if (ses->user_name == NULL) { @@ -659,8 +817,7 @@ static void ascii_ssetup_strings(char **pbcc_area, struct cifs_ses *ses, if (WARN_ON_ONCE(len < 0)) len = CIFS_MAX_DOMAINNAME_LEN - 1; bcc_ptr += len; - } /* else we will send a null domain name - so the server will default to its own domain */ + } /* else we send a null domain name so server will default to its own domain */ *bcc_ptr = 0; bcc_ptr++; @@ -756,11 +913,14 @@ static void decode_ascii_ssetup(char **pbcc_area, __u16 bleft, if (len > bleft) return; - /* No domain field in LANMAN case. Domain is - returned by old servers in the SMB negprot response */ - /* BB For newer servers which do not support Unicode, - but thus do return domain here we could add parsing - for it later, but it is not very important */ + /* + * No domain field in LANMAN case. Domain is + * returned by old servers in the SMB negprot response + * + * BB For newer servers which do not support Unicode, + * but thus do return domain here, we could add parsing + * for it later, but it is not very important + */ cifs_dbg(FYI, "ascii: bytes left %d\n", bleft); } #endif /* CONFIG_CIFS_ALLOW_INSECURE_LEGACY */ @@ -816,9 +976,12 @@ int decode_ntlmssp_challenge(char *bcc_ptr, int blob_len, ses->ntlmssp->server_flags = server_flags; memcpy(ses->ntlmssp->cryptkey, pblob->Challenge, CIFS_CRYPTO_KEY_SIZE); - /* In particular we can examine sign flags */ - /* BB spec says that if AvId field of MsvAvTimestamp is populated then - we must set the MIC field of the AUTHENTICATE_MESSAGE */ + /* + * In particular we can examine sign flags + * + * BB spec says that if AvId field of MsvAvTimestamp is populated then + * we must set the MIC field of the AUTHENTICATE_MESSAGE + */ tioffset = le32_to_cpu(pblob->TargetInfoArray.BufferOffset); tilen = le16_to_cpu(pblob->TargetInfoArray.Length); diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c index 601e7a187f87..a959ed2c9b22 100644 --- a/fs/smb/client/smb2ops.c +++ b/fs/smb/client/smb2ops.c @@ -756,6 +756,7 @@ SMB3_request_interfaces(const unsigned int xid, struct cifs_tcon *tcon, bool in_ unsigned int ret_data_len = 0; struct network_interface_info_ioctl_rsp *out_buf = NULL; struct cifs_ses *ses = tcon->ses; + struct TCP_Server_Info *pserver; /* do not query too frequently */ if (ses->iface_last_update && @@ -780,6 +781,11 @@ SMB3_request_interfaces(const unsigned int xid, struct cifs_tcon *tcon, bool in_ if (rc) goto out; + /* check if iface is still active */ + pserver = ses->chans[0].server; + if (pserver && !cifs_chan_is_iface_active(ses, pserver)) + cifs_chan_update_iface(ses, pserver); + out: kfree(out_buf); return rc; diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c index c75a80bb6d9e..2eb29fa278c3 100644 --- a/fs/smb/client/smb2pdu.c +++ b/fs/smb/client/smb2pdu.c @@ -163,6 +163,9 @@ smb2_reconnect(__le16 smb2_command, struct cifs_tcon *tcon, int rc = 0; struct nls_table *nls_codepage = NULL; struct cifs_ses *ses; + int xid; + struct TCP_Server_Info *pserver; + unsigned int chan_index; /* * SMB2s NegProt, SessSetup, Logoff do not have tcon yet so @@ -223,6 +226,12 @@ smb2_reconnect(__le16 smb2_command, struct cifs_tcon *tcon, return -EAGAIN; } } + + /* if server is marked for termination, cifsd will cleanup */ + if (server->terminate) { + spin_unlock(&server->srv_lock); + return -EHOSTDOWN; + } spin_unlock(&server->srv_lock); again: @@ -241,12 +250,24 @@ again: tcon->need_reconnect); mutex_lock(&ses->session_mutex); + /* + * if this is called by delayed work, and the channel has been disabled + * in parallel, the delayed work can continue to execute in parallel + * there's a chance that this channel may not exist anymore + */ + spin_lock(&server->srv_lock); + if (server->tcpStatus == CifsExiting) { + spin_unlock(&server->srv_lock); + mutex_unlock(&ses->session_mutex); + rc = -EHOSTDOWN; + goto out; + } + /* * Recheck after acquire mutex. If another thread is negotiating * and the server never sends an answer the socket will be closed * and tcpStatus set to reconnect. */ - spin_lock(&server->srv_lock); if (server->tcpStatus == CifsNeedReconnect) { spin_unlock(&server->srv_lock); mutex_unlock(&ses->session_mutex); @@ -283,6 +304,53 @@ again: rc = cifs_negotiate_protocol(0, ses, server); if (!rc) { + /* + * if server stopped supporting multichannel + * and the first channel reconnected, disable all the others. + */ + if (ses->chan_count > 1 && + !(server->capabilities & SMB2_GLOBAL_CAP_MULTI_CHANNEL)) { + if (SERVER_IS_CHAN(server)) { + cifs_dbg(VFS, "server %s does not support " \ + "multichannel anymore. skipping secondary channel\n", + ses->server->hostname); + + spin_lock(&ses->chan_lock); + chan_index = cifs_ses_get_chan_index(ses, server); + if (chan_index == CIFS_INVAL_CHAN_INDEX) { + spin_unlock(&ses->chan_lock); + goto skip_terminate; + } + + ses->chans[chan_index].server = NULL; + spin_unlock(&ses->chan_lock); + + /* + * the above reference of server by channel + * needs to be dropped without holding chan_lock + * as cifs_put_tcp_session takes a higher lock + * i.e. cifs_tcp_ses_lock + */ + cifs_put_tcp_session(server, 1); + + server->terminate = true; + cifs_signal_cifsd_for_reconnect(server, false); + + /* mark primary server as needing reconnect */ + pserver = server->primary_server; + cifs_signal_cifsd_for_reconnect(pserver, false); + +skip_terminate: + mutex_unlock(&ses->session_mutex); + rc = -EHOSTDOWN; + goto out; + } else { + cifs_server_dbg(VFS, "does not support " \ + "multichannel anymore. disabling all other channels\n"); + cifs_disable_secondary_channels(ses); + } + } + rc = cifs_setup_session(0, ses, server, nls_codepage); if ((rc == -EACCES) && !tcon->retry) { mutex_unlock(&ses->session_mutex); @@ -307,17 +375,44 @@ skip_sess_setup: tcon->need_reopen_files = true; rc = cifs_tree_connect(0, tcon, nls_codepage); - mutex_unlock(&ses->session_mutex); cifs_dbg(FYI, "reconnect tcon rc = %d\n", rc); if (rc) { /* If sess reconnected but tcon didn't, something strange ... */ + mutex_unlock(&ses->session_mutex); cifs_dbg(VFS, "reconnect tcon failed rc = %d\n", rc); goto out; } + if (!rc && + (server->capabilities & SMB2_GLOBAL_CAP_MULTI_CHANNEL)) { + mutex_unlock(&ses->session_mutex); + + /* + * query server network interfaces, in case they change + */ + xid = get_xid(); + rc = SMB3_request_interfaces(xid, tcon, false); + free_xid(xid); + + if (rc) + cifs_dbg(FYI, "%s: failed to query server interfaces: %d\n", + __func__, rc); + + if (ses->chan_max > ses->chan_count && + !SERVER_IS_CHAN(server)) { + if (ses->chan_count == 1) + cifs_server_dbg(VFS, "supports multichannel now\n"); + + cifs_try_adding_channels(ses); + } + } else { + mutex_unlock(&ses->session_mutex); + } + if (smb2_command != SMB2_INTERNAL_CMD) - mod_delayed_work(cifsiod_wq, &server->reconnect, 0); + if (mod_delayed_work(cifsiod_wq, &server->reconnect, 0)) + cifs_put_tcp_session(server, false); atomic_inc(&tconInfoReconnectCount); out: @@ -3808,6 +3903,13 @@ void smb2_reconnect_server(struct work_struct *work) /* Prevent simultaneous reconnects that can corrupt tcon->rlist list */ mutex_lock(&pserver->reconnect_mutex); + /* if the server is marked for termination, drop the ref count here */ + if (server->terminate) { + cifs_put_tcp_session(server, true); + mutex_unlock(&pserver->reconnect_mutex); + return; + } + INIT_LIST_HEAD(&tmp_list); INIT_LIST_HEAD(&tmp_ses_list); cifs_dbg(FYI, "Reconnecting tcons and channels\n"); @@ -3852,12 +3954,6 @@ void smb2_reconnect_server(struct work_struct *work) } spin_unlock(&ses->chan_lock); } - /* - * Get the reference to server struct to be sure that the last call of - * cifs_put_tcon() in the loop below won't release the server pointer. - */ - if (tcon_exist || ses_exist) - server->srv_count++; spin_unlock(&cifs_tcp_ses_lock); @@ -3905,13 +4001,17 @@ void smb2_reconnect_server(struct work_struct *work) done: cifs_dbg(FYI, "Reconnecting tcons and channels finished\n"); - if (resched) + if (resched) { queue_delayed_work(cifsiod_wq, &server->reconnect, 2 * HZ); + mutex_unlock(&pserver->reconnect_mutex); + + /* no need to put tcp session as we're retrying */ + return; + } mutex_unlock(&pserver->reconnect_mutex); /* now we can safely release srv struct */ - if (tcon_exist || ses_exist) - cifs_put_tcp_session(server, 1); + cifs_put_tcp_session(server, true); } int @@ -3931,7 +4031,12 @@ SMB2_echo(struct TCP_Server_Info *server) server->ops->need_neg(server)) { spin_unlock(&server->srv_lock); /* No need to send echo on newly established connections */ - mod_delayed_work(cifsiod_wq, &server->reconnect, 0); + spin_lock(&cifs_tcp_ses_lock); + server->srv_count++; + spin_unlock(&cifs_tcp_ses_lock); + if (mod_delayed_work(cifsiod_wq, &server->reconnect, 0)) + cifs_put_tcp_session(server, false); + return rc; } spin_unlock(&server->srv_lock); diff --git a/fs/smb/client/smb2transport.c b/fs/smb/client/smb2transport.c index 23c50ed7d4b5..84ea67301303 100644 --- a/fs/smb/client/smb2transport.c +++ b/fs/smb/client/smb2transport.c @@ -413,7 +413,13 @@ generate_smb3signingkey(struct cifs_ses *ses, ses->ses_status == SES_GOOD); chan_index = cifs_ses_get_chan_index(ses, server); - /* TODO: introduce ref counting for channels when the can be freed */ + if (chan_index == CIFS_INVAL_CHAN_INDEX) { + spin_unlock(&ses->chan_lock); + spin_unlock(&ses->ses_lock); + + return -EINVAL; + } + spin_unlock(&ses->chan_lock); spin_unlock(&ses->ses_lock); diff --git a/fs/smb/client/transport.c b/fs/smb/client/transport.c index d553b7a54621..4f717ad7c21b 100644 --- a/fs/smb/client/transport.c +++ b/fs/smb/client/transport.c @@ -1023,7 +1023,7 @@ struct TCP_Server_Info *cifs_pick_channel(struct cifs_ses *ses) spin_lock(&ses->chan_lock); for (i = 0; i < ses->chan_count; i++) { server = ses->chans[i].server; - if (!server) + if (!server || server->terminate) continue; /* diff --git a/fs/smb/client/xattr.c b/fs/smb/client/xattr.c index ac199160bce6..6780aa3e98a1 100644 --- a/fs/smb/client/xattr.c +++ b/fs/smb/client/xattr.c @@ -150,10 +150,13 @@ static int cifs_xattr_set(const struct xattr_handler *handler, if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_XATTR) goto out; - if (pTcon->ses->server->ops->set_EA) + if (pTcon->ses->server->ops->set_EA) { rc = pTcon->ses->server->ops->set_EA(xid, pTcon, full_path, name, value, (__u16)size, cifs_sb->local_nls, cifs_sb); + if (rc == 0) + inode_set_ctime_current(inode); + } break; case XATTR_CIFS_ACL: